From c95346ac918c5badf51b9a7ac58a26d3bd5bb224 Mon Sep 17 00:00:00 2001 From: Andrew Price Date: Mon, 11 Mar 2024 16:40:36 +0100 Subject: gfs2: Fix invalid metadata access in punch_hole In punch_hole(), when the offset lies in the final block for a given height, there is no hole to punch, but the maximum size check fails to detect that. Consequently, punch_hole() will try to punch a hole beyond the end of the metadata and fail. Fix the maximum size check. Signed-off-by: Andrew Price Signed-off-by: Andreas Gruenbacher --- fs/gfs2/bmap.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index d9ccfd27e4f1..643175498d1c 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -1718,7 +1718,8 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length) struct buffer_head *dibh, *bh; struct gfs2_holder rd_gh; unsigned int bsize_shift = sdp->sd_sb.sb_bsize_shift; - u64 lblock = (offset + (1 << bsize_shift) - 1) >> bsize_shift; + unsigned int bsize = 1 << bsize_shift; + u64 lblock = (offset + bsize - 1) >> bsize_shift; __u16 start_list[GFS2_MAX_META_HEIGHT]; __u16 __end_list[GFS2_MAX_META_HEIGHT], *end_list = NULL; unsigned int start_aligned, end_aligned; @@ -1729,7 +1730,7 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length) u64 prev_bnr = 0; __be64 *start, *end; - if (offset >= maxsize) { + if (offset + bsize - 1 >= maxsize) { /* * The starting point lies beyond the allocated metadata; * there are no blocks to deallocate. -- cgit v1.2.3-59-g8ed1b From 74098a989b9c3370f768140b7783a7aaec2759b3 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Mon, 26 Feb 2024 16:39:13 +0100 Subject: btrfs: zoned: use zone aware sb location for scrub At the moment scrub_supers() doesn't grab the super block's location via the zoned device aware btrfs_sb_log_location() but via btrfs_sb_offset(). This leads to checksum errors on 'scrub' as we're not accessing the correct location of the super block. So use btrfs_sb_log_location() for getting the super blocks location on scrub. Reported-by: WA AM Link: http://lore.kernel.org/linux-btrfs/CANU2Z0EvUzfYxczLgGUiREoMndE9WdQnbaawV5Fv5gNXptPUKw@mail.gmail.com CC: stable@vger.kernel.org # 5.15+ Reviewed-by: Qu Wenruo Reviewed-by: Naohiro Aota Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/scrub.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index c4bd0e60db59..fa25004ab04e 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -2812,7 +2812,17 @@ static noinline_for_stack int scrub_supers(struct scrub_ctx *sctx, gen = btrfs_get_last_trans_committed(fs_info); for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) { - bytenr = btrfs_sb_offset(i); + ret = btrfs_sb_log_location(scrub_dev, i, 0, &bytenr); + if (ret == -ENOENT) + break; + + if (ret) { + spin_lock(&sctx->stat_lock); + sctx->stat.super_errors++; + spin_unlock(&sctx->stat_lock); + continue; + } + if (bytenr + BTRFS_SUPER_INFO_SIZE > scrub_dev->commit_total_bytes) break; -- cgit v1.2.3-59-g8ed1b From 2aea94ac14d1e0a8ae9e34febebe208213ba72f7 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Wed, 20 Mar 2024 11:26:07 -0700 Subject: exec: Fix NOMMU linux_binprm::exec in transfer_args_to_stack() In NOMMU kernel the value of linux_binprm::p is the offset inside the temporary program arguments array maintained in separate pages in the linux_binprm::page. linux_binprm::exec being a copy of linux_binprm::p thus must be adjusted when that array is copied to the user stack. Without that adjustment the value passed by the NOMMU kernel to the ELF program in the AT_EXECFN entry of the aux array doesn't make any sense and it may break programs that try to access memory pointed to by that entry. Adjust linux_binprm::exec before the successful return from the transfer_args_to_stack(). Cc: Fixes: b6a2fea39318 ("mm: variable length argument support") Fixes: 5edc2a5123a7 ("binfmt_elf_fdpic: wire up AT_EXECFD, AT_EXECFN, AT_SECURE") Signed-off-by: Max Filippov Link: https://lore.kernel.org/r/20240320182607.1472887-1-jcmvbkbc@gmail.com Signed-off-by: Kees Cook --- fs/exec.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index e7d9d6ad980b..f66639820580 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -895,6 +895,7 @@ int transfer_args_to_stack(struct linux_binprm *bprm, goto out; } + bprm->exec += *sp_location - MAX_ARG_PAGES * PAGE_SIZE; *sp_location = sp; out: -- cgit v1.2.3-59-g8ed1b From 5248f4097308c1cdcf163314a6ea3c8c88c98cd9 Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Thu, 21 Mar 2024 20:04:08 +0000 Subject: binfmt: replace deprecated strncpy strncpy() is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. There is a _nearly_ identical implementation of fill_psinfo present in binfmt_elf.c -- except that one uses get_task_comm over strncpy(). Let's mirror that in binfmt_elf_fdpic.c Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://github.com/KSPP/linux/issues/90 Cc: Signed-off-by: Justin Stitt Link: https://lore.kernel.org/r/20240321-strncpy-fs-binfmt_elf_fdpic-c-v2-1-0b6daec6cc56@google.com Signed-off-by: Kees Cook --- fs/binfmt_elf_fdpic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 1920ed69279b..3314249e8674 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -1359,7 +1359,7 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, SET_UID(psinfo->pr_uid, from_kuid_munged(cred->user_ns, cred->uid)); SET_GID(psinfo->pr_gid, from_kgid_munged(cred->user_ns, cred->gid)); rcu_read_unlock(); - strncpy(psinfo->pr_fname, p->comm, sizeof(psinfo->pr_fname)); + get_task_comm(psinfo->pr_fname, p); return 0; } -- cgit v1.2.3-59-g8ed1b From 9fe6e9e7b58944037714442384075c17cfde1c56 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 18 Mar 2024 17:32:09 +0100 Subject: nfsd: Fix error cleanup path in nfsd_rename() Commit a8b0026847b8 ("rename(): avoid a deadlock in the case of parents having no common ancestor") added an error bail out path. However this path does not drop the remount protection that has been acquired. Fix the cleanup path to properly drop the remount protection. Fixes: a8b0026847b8 ("rename(): avoid a deadlock in the case of parents having no common ancestor") Signed-off-by: Jan Kara Reviewed-by: Jeff Layton Acked-by: Al Viro Signed-off-by: Chuck Lever --- fs/nfsd/vfs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 6a4c506038e0..daf5221ef922 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1853,7 +1853,7 @@ retry: trap = lock_rename(tdentry, fdentry); if (IS_ERR(trap)) { err = (rqstp->rq_vers == 2) ? nfserr_acces : nfserr_xdev; - goto out; + goto out_want_write; } err = fh_fill_pre_attrs(ffhp); if (err != nfs_ok) @@ -1923,6 +1923,7 @@ retry: } out_unlock: unlock_rename(tdentry, fdentry); +out_want_write: fh_drop_write(ffhp); /* -- cgit v1.2.3-59-g8ed1b From 11763a8598f888dec631a8a903f7ada32181001f Mon Sep 17 00:00:00 2001 From: Lizhi Xu Date: Fri, 2 Feb 2024 20:15:31 +0800 Subject: fs/9p: fix uaf in in v9fs_stat2inode_dotl The incorrect logical order of accessing the st object code in v9fs_fid_iget_dotl is causing this uaf. Fixes: 724a08450f74 ("fs/9p: simplify iget to remove unnecessary paths") Reported-and-tested-by: syzbot+7a3d75905ea1a830dbe5@syzkaller.appspotmail.com Signed-off-by: Lizhi Xu Tested-by: Breno Leitao Reviewed-by: Dominique Martinet Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_inode_dotl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index ef9db3e03506..2b313fe7003e 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -78,11 +78,11 @@ struct inode *v9fs_fid_iget_dotl(struct super_block *sb, struct p9_fid *fid) retval = v9fs_init_inode(v9ses, inode, &fid->qid, st->st_mode, new_decode_dev(st->st_rdev)); + v9fs_stat2inode_dotl(st, inode, 0); kfree(st); if (retval) goto error; - v9fs_stat2inode_dotl(st, inode, 0); v9fs_set_netfs_context(inode); v9fs_cache_inode_get_cookie(inode); retval = v9fs_get_acl(inode, fid); -- cgit v1.2.3-59-g8ed1b From 10211b4a23cf4a3df5c11a10e5b3d371f16a906f Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 29 Feb 2024 22:22:50 +0000 Subject: fs/9p: remove redundant pointer v9ses Pointer v9ses is being assigned the value from the return of inlined function v9fs_inode2v9ses (which just returns inode->i_sb->s_fs_info). The pointer is not used after the assignment, so the variable is redundant and can be removed. Cleans up clang scan warnings such as: fs/9p/vfs_inode_dotl.c:300:28: warning: variable 'v9ses' set but not used [-Wunused-but-set-variable] Signed-off-by: Colin Ian King Reviewed-by: Dominique Martinet Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_inode_dotl.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'fs') diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 2b313fe7003e..55dde186041a 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -297,7 +297,6 @@ static int v9fs_vfs_mkdir_dotl(struct mnt_idmap *idmap, umode_t omode) { int err; - struct v9fs_session_info *v9ses; struct p9_fid *fid = NULL, *dfid = NULL; kgid_t gid; const unsigned char *name; @@ -307,7 +306,6 @@ static int v9fs_vfs_mkdir_dotl(struct mnt_idmap *idmap, struct posix_acl *dacl = NULL, *pacl = NULL; p9_debug(P9_DEBUG_VFS, "name %pd\n", dentry); - v9ses = v9fs_inode2v9ses(dir); omode |= S_IFDIR; if (dir->i_mode & S_ISGID) @@ -739,7 +737,6 @@ v9fs_vfs_mknod_dotl(struct mnt_idmap *idmap, struct inode *dir, kgid_t gid; const unsigned char *name; umode_t mode; - struct v9fs_session_info *v9ses; struct p9_fid *fid = NULL, *dfid = NULL; struct inode *inode; struct p9_qid qid; @@ -749,7 +746,6 @@ v9fs_vfs_mknod_dotl(struct mnt_idmap *idmap, struct inode *dir, dir->i_ino, dentry, omode, MAJOR(rdev), MINOR(rdev)); - v9ses = v9fs_inode2v9ses(dir); dfid = v9fs_parent_fid(dentry); if (IS_ERR(dfid)) { err = PTR_ERR(dfid); -- cgit v1.2.3-59-g8ed1b From a97b59ed796804612468a3fb0ac2a5567a100a7a Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Mon, 25 Mar 2024 08:51:16 +0800 Subject: erofs: drop experimental warning for FSDAX As EXT4/XFS filesystems, FSDAX functionality is considered to be stable. Let's drop this warning. Reviewed-by: Jingbo Xu Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20240325005116.106351-1-hsiangkao@linux.alibaba.com --- fs/erofs/super.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 69308fd73e4a..c0eb139adb07 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -430,7 +430,6 @@ static bool erofs_fc_set_dax_mode(struct fs_context *fc, unsigned int mode) switch (mode) { case EROFS_MOUNT_DAX_ALWAYS: - warnfc(fc, "DAX enabled. Warning: EXPERIMENTAL, use at your own risk"); set_opt(&ctx->opt, DAX_ALWAYS); clear_opt(&ctx->opt, DAX_NEVER); return true; -- cgit v1.2.3-59-g8ed1b From 15922f5dbf51dad334cde888ce6835d377678dc9 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 19 Mar 2024 11:29:03 +1100 Subject: xfs: allow sunit mount option to repair bad primary sb stripe values If a filesystem has a busted stripe alignment configuration on disk (e.g. because broken RAID firmware told mkfs that swidth was smaller than sunit), then the filesystem will refuse to mount due to the stripe validation failing. This failure is triggering during distro upgrades from old kernels lacking this check to newer kernels with this check, and currently the only way to fix it is with offline xfs_db surgery. This runtime validity checking occurs when we read the superblock for the first time and causes the mount to fail immediately. This prevents the rewrite of stripe unit/width via mount options that occurs later in the mount process. Hence there is no way to recover this situation without resorting to offline xfs_db rewrite of the values. However, we parse the mount options long before we read the superblock, and we know if the mount has been asked to re-write the stripe alignment configuration when we are reading the superblock and verifying it for the first time. Hence we can conditionally ignore stripe verification failures if the mount options specified will correct the issue. We validate that the new stripe unit/width are valid before we overwrite the superblock values, so we can ignore the invalid config at verification and fail the mount later if the new values are not valid. This, at least, gives users the chance of correcting the issue after a kernel upgrade without having to resort to xfs-db hacks. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_sb.c | 40 +++++++++++++++++++++++++++++++--------- fs/xfs/libxfs/xfs_sb.h | 5 +++-- 2 files changed, 34 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index d991eec05436..73a4b895de67 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -530,7 +530,8 @@ xfs_validate_sb_common( } if (!xfs_validate_stripe_geometry(mp, XFS_FSB_TO_B(mp, sbp->sb_unit), - XFS_FSB_TO_B(mp, sbp->sb_width), 0, false)) + XFS_FSB_TO_B(mp, sbp->sb_width), 0, + xfs_buf_daddr(bp) == XFS_SB_DADDR, false)) return -EFSCORRUPTED; /* @@ -1323,8 +1324,10 @@ xfs_sb_get_secondary( } /* - * sunit, swidth, sectorsize(optional with 0) should be all in bytes, - * so users won't be confused by values in error messages. + * sunit, swidth, sectorsize(optional with 0) should be all in bytes, so users + * won't be confused by values in error messages. This function returns false + * if the stripe geometry is invalid and the caller is unable to repair the + * stripe configuration later in the mount process. */ bool xfs_validate_stripe_geometry( @@ -1332,20 +1335,21 @@ xfs_validate_stripe_geometry( __s64 sunit, __s64 swidth, int sectorsize, + bool may_repair, bool silent) { if (swidth > INT_MAX) { if (!silent) xfs_notice(mp, "stripe width (%lld) is too large", swidth); - return false; + goto check_override; } if (sunit > swidth) { if (!silent) xfs_notice(mp, "stripe unit (%lld) is larger than the stripe width (%lld)", sunit, swidth); - return false; + goto check_override; } if (sectorsize && (int)sunit % sectorsize) { @@ -1353,21 +1357,21 @@ xfs_validate_stripe_geometry( xfs_notice(mp, "stripe unit (%lld) must be a multiple of the sector size (%d)", sunit, sectorsize); - return false; + goto check_override; } if (sunit && !swidth) { if (!silent) xfs_notice(mp, "invalid stripe unit (%lld) and stripe width of 0", sunit); - return false; + goto check_override; } if (!sunit && swidth) { if (!silent) xfs_notice(mp, "invalid stripe width (%lld) and stripe unit of 0", swidth); - return false; + goto check_override; } if (sunit && (int)swidth % (int)sunit) { @@ -1375,9 +1379,27 @@ xfs_validate_stripe_geometry( xfs_notice(mp, "stripe width (%lld) must be a multiple of the stripe unit (%lld)", swidth, sunit); - return false; + goto check_override; } return true; + +check_override: + if (!may_repair) + return false; + /* + * During mount, mp->m_dalign will not be set unless the sunit mount + * option was set. If it was set, ignore the bad stripe alignment values + * and allow the validation and overwrite later in the mount process to + * attempt to overwrite the bad stripe alignment values with the values + * supplied by mount options. + */ + if (!mp->m_dalign) + return false; + if (!silent) + xfs_notice(mp, +"Will try to correct with specified mount options sunit (%d) and swidth (%d)", + BBTOB(mp->m_dalign), BBTOB(mp->m_swidth)); + return true; } /* diff --git a/fs/xfs/libxfs/xfs_sb.h b/fs/xfs/libxfs/xfs_sb.h index 2e8e8d63d4eb..37b1ed1bc209 100644 --- a/fs/xfs/libxfs/xfs_sb.h +++ b/fs/xfs/libxfs/xfs_sb.h @@ -35,8 +35,9 @@ extern int xfs_sb_get_secondary(struct xfs_mount *mp, struct xfs_trans *tp, xfs_agnumber_t agno, struct xfs_buf **bpp); -extern bool xfs_validate_stripe_geometry(struct xfs_mount *mp, - __s64 sunit, __s64 swidth, int sectorsize, bool silent); +bool xfs_validate_stripe_geometry(struct xfs_mount *mp, + __s64 sunit, __s64 swidth, int sectorsize, bool may_repair, + bool silent); uint8_t xfs_compute_rextslog(xfs_rtbxlen_t rtextents); -- cgit v1.2.3-59-g8ed1b From f2e812c1522dab847912309b00abcc762dd696da Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 19 Mar 2024 09:36:28 +1100 Subject: xfs: don't use current->journal_info syzbot reported an ext4 panic during a page fault where found a journal handle when it didn't expect to find one. The structure it tripped over had a value of 'TRAN' in the first entry in the structure, and that indicates it tripped over a struct xfs_trans instead of a jbd2 handle. The reason for this is that the page fault was taken during a copy-out to a user buffer from an xfs bulkstat operation. XFS uses an "empty" transaction context for bulkstat to do automated metadata buffer cleanup, and so the transaction context is valid across the copyout of the bulkstat info into the user buffer. We are using empty transaction contexts like this in XFS to reduce the risk of failing to release objects we reference during the operation, especially during error handling. Hence we really need to ensure that we can take page faults from these contexts without leaving landmines for the code processing the page fault to trip over. However, this same behaviour could happen from any other filesystem that triggers a page fault or any other exception that is handled on-stack from within a task context that has current->journal_info set. Having a page fault from some other filesystem bounce into XFS where we have to run a transaction isn't a bug at all, but the usage of current->journal_info means that this could result corruption of the outer task's journal_info structure. The problem is purely that we now have two different contexts that now think they own current->journal_info. IOWs, no filesystem can allow page faults or on-stack exceptions while current->journal_info is set by the filesystem because the exception processing might use current->journal_info itself. If we end up with nested XFS transactions whilst holding an empty transaction, then it isn't an issue as the outer transaction does not hold a log reservation. If we ignore the current->journal_info usage, then the only problem that might occur is a deadlock if the exception tries to take the same locks the upper context holds. That, however, is not a problem that setting current->journal_info would solve, so it's largely an irrelevant concern here. IOWs, we really only use current->journal_info for a warning check in xfs_vm_writepages() to ensure we aren't doing writeback from a transaction context. Writeback might need to do allocation, so it can need to run transactions itself. Hence it's a debug check to warn us that we've done something silly, and largely it is not all that useful. So let's just remove all the use of current->journal_info in XFS and get rid of all the potential issues from nested contexts where current->journal_info might get misused by another filesystem context. Reported-by: syzbot+cdee56dbcdf0096ef605@syzkaller.appspotmail.com Signed-off-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Reviewed-by: Mark Tinguely Reviewed-by: Christoph Hellwig Signed-off-by: Chandan Babu R --- fs/xfs/scrub/common.c | 4 +--- fs/xfs/xfs_aops.c | 7 ------- fs/xfs/xfs_icache.c | 8 +++++--- fs/xfs/xfs_trans.h | 9 +-------- 4 files changed, 7 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index abff79a77c72..47a20cf5205f 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -1044,9 +1044,7 @@ xchk_irele( struct xfs_scrub *sc, struct xfs_inode *ip) { - if (current->journal_info != NULL) { - ASSERT(current->journal_info == sc->tp); - + if (sc->tp) { /* * If we are in a transaction, we /cannot/ drop the inode * ourselves, because the VFS will trigger writeback, which diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 1698507d1ac7..3f428620ebf2 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -503,13 +503,6 @@ xfs_vm_writepages( { struct xfs_writepage_ctx wpc = { }; - /* - * Writing back data in a transaction context can result in recursive - * transactions. This is bad, so issue a warning and get out of here. - */ - if (WARN_ON_ONCE(current->journal_info)) - return 0; - xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); return iomap_writepages(mapping, wbc, &wpc.ctx, &xfs_writeback_ops); } diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index e64265bc0b33..74f1812b03cb 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -2039,8 +2039,10 @@ xfs_inodegc_want_queue_work( * - Memory shrinkers queued the inactivation worker and it hasn't finished. * - The queue depth exceeds the maximum allowable percpu backlog. * - * Note: If the current thread is running a transaction, we don't ever want to - * wait for other transactions because that could introduce a deadlock. + * Note: If we are in a NOFS context here (e.g. current thread is running a + * transaction) the we don't want to block here as inodegc progress may require + * filesystem resources we hold to make progress and that could result in a + * deadlock. Hence we skip out of here if we are in a scoped NOFS context. */ static inline bool xfs_inodegc_want_flush_work( @@ -2048,7 +2050,7 @@ xfs_inodegc_want_flush_work( unsigned int items, unsigned int shrinker_hits) { - if (current->journal_info) + if (current->flags & PF_MEMALLOC_NOFS) return false; if (shrinker_hits > 0) diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 3f7e3a09a49f..1636663707dc 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -268,19 +268,14 @@ static inline void xfs_trans_set_context( struct xfs_trans *tp) { - ASSERT(current->journal_info == NULL); tp->t_pflags = memalloc_nofs_save(); - current->journal_info = tp; } static inline void xfs_trans_clear_context( struct xfs_trans *tp) { - if (current->journal_info == tp) { - memalloc_nofs_restore(tp->t_pflags); - current->journal_info = NULL; - } + memalloc_nofs_restore(tp->t_pflags); } static inline void @@ -288,10 +283,8 @@ xfs_trans_switch_context( struct xfs_trans *old_tp, struct xfs_trans *new_tp) { - ASSERT(current->journal_info == old_tp); new_tp->t_pflags = old_tp->t_pflags; old_tp->t_pflags = 0; - current->journal_info = new_tp; } #endif /* __XFS_TRANS_H__ */ -- cgit v1.2.3-59-g8ed1b From 6630036b7c228f57c7893ee0403e92c2db2cd21d Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Tue, 19 Mar 2024 13:50:32 +0000 Subject: fs/9p: fix uninitialized values during inode evict If an iget fails due to not being able to retrieve information from the server then the inode structure is only partially initialized. When the inode gets evicted, references to uninitialized structures (like fscache cookies) were being made. This patch checks for a bad_inode before doing anything other than clearing the inode from the cache. Since the inode is bad, it shouldn't have any state associated with it that needs to be written back (and there really isn't a way to complete those anyways). Reported-by: syzbot+eb83fe1cce5833cd66a0@syzkaller.appspotmail.com Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_inode.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 360a5304ec03..b01b1bbf2493 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -344,17 +344,21 @@ void v9fs_evict_inode(struct inode *inode) struct v9fs_inode __maybe_unused *v9inode = V9FS_I(inode); __le32 __maybe_unused version; - truncate_inode_pages_final(&inode->i_data); + if (!is_bad_inode(inode)) { + truncate_inode_pages_final(&inode->i_data); - version = cpu_to_le32(v9inode->qid.version); - netfs_clear_inode_writeback(inode, &version); + version = cpu_to_le32(v9inode->qid.version); + netfs_clear_inode_writeback(inode, &version); - clear_inode(inode); - filemap_fdatawrite(&inode->i_data); + clear_inode(inode); + filemap_fdatawrite(&inode->i_data); #ifdef CONFIG_9P_FSCACHE - fscache_relinquish_cookie(v9fs_inode_cookie(v9inode), false); + if (v9fs_inode_cookie(v9inode)) + fscache_relinquish_cookie(v9fs_inode_cookie(v9inode), false); #endif + } else + clear_inode(inode); } struct inode *v9fs_fid_iget(struct super_block *sb, struct p9_fid *fid) -- cgit v1.2.3-59-g8ed1b From 5a6cb47ef05a48cec778207fff6c189de8f27a0e Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Mon, 25 Mar 2024 12:12:40 +0800 Subject: fs: quota: use group allocation of per-cpu counters API Use group allocation of per-cpu counters api to accelerate dquot_init() and simplify code. Signed-off-by: Kefeng Wang Signed-off-by: Jan Kara Message-Id: <20240325041240.53537-1-wangkefeng.wang@huawei.com> --- fs/quota/dquot.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index dacbee455c03..808544f74e5e 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -3016,11 +3016,10 @@ static int __init dquot_init(void) if (!dquot_hash) panic("Cannot create dquot hash table"); - for (i = 0; i < _DQST_DQSTAT_LAST; i++) { - ret = percpu_counter_init(&dqstats.counter[i], 0, GFP_KERNEL); - if (ret) - panic("Cannot create dquot stat counters"); - } + ret = percpu_counter_init_many(dqstats.counter, 0, GFP_KERNEL, + _DQST_DQSTAT_LAST); + if (ret) + panic("Cannot create dquot stat counters"); /* Find power-of-two hlist_heads which can fit into allocation */ nr_hash = (1UL << order) * PAGE_SIZE / sizeof(struct hlist_head); -- cgit v1.2.3-59-g8ed1b From 1b17a46c9243e9421ee1ac6d628604bbc4ae2201 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 7 Mar 2024 12:56:33 -0600 Subject: isofs: convert isofs to use the new mount API This also renames iso9660_options to isofs_options, for consistency. Signed-off-by: Eric Sandeen cc: Jan Kara Signed-off-by: Jan Kara Message-Id: <409e28da-9c19-427a-acfb-78788f0a23b8@redhat.com> --- fs/isofs/inode.c | 473 ++++++++++++++++++++++++++++--------------------------- 1 file changed, 240 insertions(+), 233 deletions(-) (limited to 'fs') diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 2a616a9f289d..93b1077a380a 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -21,11 +21,12 @@ #include #include #include -#include #include #include #include #include +#include +#include #include "isofs.h" #include "zisofs.h" @@ -110,10 +111,10 @@ static void destroy_inodecache(void) kmem_cache_destroy(isofs_inode_cachep); } -static int isofs_remount(struct super_block *sb, int *flags, char *data) +static int isofs_reconfigure(struct fs_context *fc) { - sync_filesystem(sb); - if (!(*flags & SB_RDONLY)) + sync_filesystem(fc->root->d_sb); + if (!(fc->sb_flags & SB_RDONLY)) return -EROFS; return 0; } @@ -123,7 +124,6 @@ static const struct super_operations isofs_sops = { .free_inode = isofs_free_inode, .put_super = isofs_put_super, .statfs = isofs_statfs, - .remount_fs = isofs_remount, .show_options = isofs_show_options, }; @@ -145,7 +145,7 @@ static const struct dentry_operations isofs_dentry_ops[] = { #endif }; -struct iso9660_options{ +struct isofs_options{ unsigned int rock:1; unsigned int joliet:1; unsigned int cruft:1; @@ -289,197 +289,161 @@ isofs_dentry_cmpi_ms(const struct dentry *dentry, #endif enum { - Opt_block, Opt_check_r, Opt_check_s, Opt_cruft, Opt_gid, Opt_ignore, - Opt_iocharset, Opt_map_a, Opt_map_n, Opt_map_o, Opt_mode, Opt_nojoliet, - Opt_norock, Opt_sb, Opt_session, Opt_uid, Opt_unhide, Opt_utf8, Opt_err, - Opt_nocompress, Opt_hide, Opt_showassoc, Opt_dmode, Opt_overriderockperm, + Opt_block, Opt_check, Opt_cruft, Opt_gid, Opt_ignore, Opt_iocharset, + Opt_map, Opt_mode, Opt_nojoliet, Opt_norock, Opt_sb, Opt_session, + Opt_uid, Opt_unhide, Opt_utf8, Opt_err, Opt_nocompress, Opt_hide, + Opt_showassoc, Opt_dmode, Opt_overriderockperm, }; -static const match_table_t tokens = { - {Opt_norock, "norock"}, - {Opt_nojoliet, "nojoliet"}, - {Opt_unhide, "unhide"}, - {Opt_hide, "hide"}, - {Opt_showassoc, "showassoc"}, - {Opt_cruft, "cruft"}, - {Opt_utf8, "utf8"}, - {Opt_iocharset, "iocharset=%s"}, - {Opt_map_a, "map=acorn"}, - {Opt_map_a, "map=a"}, - {Opt_map_n, "map=normal"}, - {Opt_map_n, "map=n"}, - {Opt_map_o, "map=off"}, - {Opt_map_o, "map=o"}, - {Opt_session, "session=%u"}, - {Opt_sb, "sbsector=%u"}, - {Opt_check_r, "check=relaxed"}, - {Opt_check_r, "check=r"}, - {Opt_check_s, "check=strict"}, - {Opt_check_s, "check=s"}, - {Opt_uid, "uid=%u"}, - {Opt_gid, "gid=%u"}, - {Opt_mode, "mode=%u"}, - {Opt_dmode, "dmode=%u"}, - {Opt_overriderockperm, "overriderockperm"}, - {Opt_block, "block=%u"}, - {Opt_ignore, "conv=binary"}, - {Opt_ignore, "conv=b"}, - {Opt_ignore, "conv=text"}, - {Opt_ignore, "conv=t"}, - {Opt_ignore, "conv=mtext"}, - {Opt_ignore, "conv=m"}, - {Opt_ignore, "conv=auto"}, - {Opt_ignore, "conv=a"}, - {Opt_nocompress, "nocompress"}, - {Opt_err, NULL} +static const struct constant_table isofs_param_map[] = { + {"acorn", 'a'}, + {"a", 'a'}, + {"normal", 'n'}, + {"n", 'n'}, + {"off", 'o'}, + {"o", 'o'}, + {} }; -static int parse_options(char *options, struct iso9660_options *popt) -{ - char *p; - int option; - unsigned int uv; - - popt->map = 'n'; - popt->rock = 1; - popt->joliet = 1; - popt->cruft = 0; - popt->hide = 0; - popt->showassoc = 0; - popt->check = 'u'; /* unset */ - popt->nocompress = 0; - popt->blocksize = 1024; - popt->fmode = popt->dmode = ISOFS_INVALID_MODE; - popt->uid_set = 0; - popt->gid_set = 0; - popt->gid = GLOBAL_ROOT_GID; - popt->uid = GLOBAL_ROOT_UID; - popt->iocharset = NULL; - popt->overriderockperm = 0; - popt->session=-1; - popt->sbsector=-1; - if (!options) - return 1; - - while ((p = strsep(&options, ",")) != NULL) { - int token; - substring_t args[MAX_OPT_ARGS]; - unsigned n; - - if (!*p) - continue; +static const struct constant_table isofs_param_check[] = { + {"relaxed", 'r'}, + {"r", 'r'}, + {"strict", 's'}, + {"s", 's'}, + {} +}; - token = match_token(p, tokens, args); - switch (token) { - case Opt_norock: - popt->rock = 0; - break; - case Opt_nojoliet: - popt->joliet = 0; - break; - case Opt_hide: - popt->hide = 1; - break; - case Opt_unhide: - case Opt_showassoc: - popt->showassoc = 1; - break; - case Opt_cruft: - popt->cruft = 1; - break; +static const struct fs_parameter_spec isofs_param_spec[] = { + fsparam_flag ("norock", Opt_norock), + fsparam_flag ("nojoliet", Opt_nojoliet), + fsparam_flag ("unhide", Opt_unhide), + fsparam_flag ("hide", Opt_hide), + fsparam_flag ("showassoc", Opt_showassoc), + fsparam_flag ("cruft", Opt_cruft), + fsparam_flag ("utf8", Opt_utf8), + fsparam_string ("iocharset", Opt_iocharset), + fsparam_enum ("map", Opt_map, isofs_param_map), + fsparam_u32 ("session", Opt_session), + fsparam_u32 ("sbsector", Opt_sb), + fsparam_enum ("check", Opt_check, isofs_param_check), + fsparam_u32 ("uid", Opt_uid), + fsparam_u32 ("gid", Opt_gid), + /* Note: mode/dmode historically accepted %u not strictly %o */ + fsparam_u32 ("mode", Opt_mode), + fsparam_u32 ("dmode", Opt_dmode), + fsparam_flag ("overriderockperm", Opt_overriderockperm), + fsparam_u32 ("block", Opt_block), + fsparam_string ("conv", Opt_ignore), + fsparam_flag ("nocompress", Opt_nocompress), + {} +}; + +static int isofs_parse_param(struct fs_context *fc, + struct fs_parameter *param) +{ + struct isofs_options *popt = fc->fs_private; + struct fs_parse_result result; + int opt; + kuid_t uid; + kgid_t gid; + unsigned int n; + + /* There are no remountable options */ + if (fc->purpose == FS_CONTEXT_FOR_RECONFIGURE) + return 0; + + opt = fs_parse(fc, isofs_param_spec, param, &result); + if (opt < 0) + return opt; + + switch (opt) { + case Opt_norock: + popt->rock = 0; + break; + case Opt_nojoliet: + popt->joliet = 0; + break; + case Opt_hide: + popt->hide = 1; + break; + case Opt_unhide: + case Opt_showassoc: + popt->showassoc = 1; + break; + case Opt_cruft: + popt->cruft = 1; + break; #ifdef CONFIG_JOLIET - case Opt_utf8: - kfree(popt->iocharset); - popt->iocharset = kstrdup("utf8", GFP_KERNEL); - if (!popt->iocharset) - return 0; - break; - case Opt_iocharset: - kfree(popt->iocharset); - popt->iocharset = match_strdup(&args[0]); - if (!popt->iocharset) - return 0; - break; + case Opt_utf8: + kfree(popt->iocharset); + popt->iocharset = kstrdup("utf8", GFP_KERNEL); + if (!popt->iocharset) + return -ENOMEM; + break; + case Opt_iocharset: + kfree(popt->iocharset); + popt->iocharset = kstrdup(param->string, GFP_KERNEL); + if (!popt->iocharset) + return -ENOMEM; + break; #endif - case Opt_map_a: - popt->map = 'a'; - break; - case Opt_map_o: - popt->map = 'o'; - break; - case Opt_map_n: - popt->map = 'n'; - break; - case Opt_session: - if (match_int(&args[0], &option)) - return 0; - n = option; - /* - * Track numbers are supposed to be in range 1-99, the - * mount option starts indexing at 0. - */ - if (n >= 99) - return 0; - popt->session = n + 1; - break; - case Opt_sb: - if (match_int(&args[0], &option)) - return 0; - popt->sbsector = option; - break; - case Opt_check_r: - popt->check = 'r'; - break; - case Opt_check_s: - popt->check = 's'; - break; - case Opt_ignore: - break; - case Opt_uid: - if (match_uint(&args[0], &uv)) - return 0; - popt->uid = make_kuid(current_user_ns(), uv); - if (!uid_valid(popt->uid)) - return 0; - popt->uid_set = 1; - break; - case Opt_gid: - if (match_uint(&args[0], &uv)) - return 0; - popt->gid = make_kgid(current_user_ns(), uv); - if (!gid_valid(popt->gid)) - return 0; - popt->gid_set = 1; - break; - case Opt_mode: - if (match_int(&args[0], &option)) - return 0; - popt->fmode = option; - break; - case Opt_dmode: - if (match_int(&args[0], &option)) - return 0; - popt->dmode = option; - break; - case Opt_overriderockperm: - popt->overriderockperm = 1; - break; - case Opt_block: - if (match_int(&args[0], &option)) - return 0; - n = option; - if (n != 512 && n != 1024 && n != 2048) - return 0; - popt->blocksize = n; - break; - case Opt_nocompress: - popt->nocompress = 1; - break; - default: - return 0; - } + case Opt_map: + popt->map = result.uint_32; + break; + case Opt_session: + n = result.uint_32; + /* + * Track numbers are supposed to be in range 1-99, the + * mount option starts indexing at 0. + */ + if (n >= 99) + return -EINVAL; + popt->session = n + 1; + break; + case Opt_sb: + popt->sbsector = result.uint_32; + break; + case Opt_check: + popt->check = result.uint_32; + break; + case Opt_ignore: + break; + case Opt_uid: + uid = make_kuid(current_user_ns(), result.uint_32); + if (!uid_valid(uid)) + return -EINVAL; + popt->uid = uid; + popt->uid_set = 1; + break; + case Opt_gid: + gid = make_kgid(current_user_ns(), result.uint_32); + if (!gid_valid(gid)) + return -EINVAL; + popt->gid = gid; + popt->gid_set = 1; + break; + case Opt_mode: + popt->fmode = result.uint_32; + break; + case Opt_dmode: + popt->dmode = result.uint_32; + break; + case Opt_overriderockperm: + popt->overriderockperm = 1; + break; + case Opt_block: + n = result.uint_32; + if (n != 512 && n != 1024 && n != 2048) + return -EINVAL; + popt->blocksize = n; + break; + case Opt_nocompress: + popt->nocompress = 1; + break; + default: + return -EINVAL; } - return 1; + return 0; } /* @@ -615,7 +579,7 @@ static bool rootdir_empty(struct super_block *sb, unsigned long block) /* * Initialize the superblock and read the root inode. */ -static int isofs_fill_super(struct super_block *s, void *data, int silent) +static int isofs_fill_super(struct super_block *s, struct fs_context *fc) { struct buffer_head *bh = NULL, *pri_bh = NULL; struct hs_primary_descriptor *h_pri = NULL; @@ -623,7 +587,7 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent) struct iso_supplementary_descriptor *sec = NULL; struct iso_directory_record *rootp; struct inode *inode; - struct iso9660_options opt; + struct isofs_options *opt = fc->fs_private; struct isofs_sb_info *sbi; unsigned long first_data_zone; int joliet_level = 0; @@ -631,15 +595,13 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent) int orig_zonesize; int table, error = -EINVAL; unsigned int vol_desc_start; + int silent = fc->sb_flags & SB_SILENT; sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); if (!sbi) return -ENOMEM; s->s_fs_info = sbi; - if (!parse_options((char *)data, &opt)) - goto out_freesbi; - /* * First of all, get the hardware blocksize for this device. * If we don't know what it is, or the hardware blocksize is @@ -655,14 +617,14 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent) bdev_logical_block_size(s->s_bdev)); goto out_freesbi; } - opt.blocksize = sb_min_blocksize(s, opt.blocksize); + opt->blocksize = sb_min_blocksize(s, opt->blocksize); sbi->s_high_sierra = 0; /* default is iso9660 */ - sbi->s_session = opt.session; - sbi->s_sbsector = opt.sbsector; + sbi->s_session = opt->session; + sbi->s_sbsector = opt->sbsector; - vol_desc_start = (opt.sbsector != -1) ? - opt.sbsector : isofs_get_last_session(s,opt.session); + vol_desc_start = (opt->sbsector != -1) ? + opt->sbsector : isofs_get_last_session(s, opt->session); for (iso_blknum = vol_desc_start+16; iso_blknum < vol_desc_start+100; iso_blknum++) { @@ -696,7 +658,7 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent) else if (isonum_711(vdp->type) == ISO_VD_SUPPLEMENTARY) { sec = (struct iso_supplementary_descriptor *)vdp; if (sec->escape[0] == 0x25 && sec->escape[1] == 0x2f) { - if (opt.joliet) { + if (opt->joliet) { if (sec->escape[2] == 0x40) joliet_level = 1; else if (sec->escape[2] == 0x43) @@ -721,7 +683,7 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent) goto out_freebh; sbi->s_high_sierra = 1; - opt.rock = 0; + opt->rock = 0; h_pri = (struct hs_primary_descriptor *)vdp; goto root_found; } @@ -749,7 +711,7 @@ root_found: goto out_freebh; } - if (joliet_level && (!pri || !opt.rock)) { + if (joliet_level && (!pri || !opt->rock)) { /* This is the case of Joliet with the norock mount flag. * A disc with both Joliet and Rock Ridge is handled later */ @@ -780,7 +742,7 @@ root_found: * blocks that were 512 bytes (which should only very rarely * happen.) */ - if (orig_zonesize < opt.blocksize) + if (orig_zonesize < opt->blocksize) goto out_bad_size; /* RDE: convert log zone size to bit shift */ @@ -865,10 +827,10 @@ root_found: #ifdef CONFIG_JOLIET if (joliet_level) { - char *p = opt.iocharset ? opt.iocharset : CONFIG_NLS_DEFAULT; + char *p = opt->iocharset ? opt->iocharset : CONFIG_NLS_DEFAULT; if (strcmp(p, "utf8") != 0) { - sbi->s_nls_iocharset = opt.iocharset ? - load_nls(opt.iocharset) : load_nls_default(); + sbi->s_nls_iocharset = opt->iocharset ? + load_nls(opt->iocharset) : load_nls_default(); if (!sbi->s_nls_iocharset) goto out_freesbi; } @@ -876,29 +838,29 @@ root_found: #endif s->s_op = &isofs_sops; s->s_export_op = &isofs_export_ops; - sbi->s_mapping = opt.map; - sbi->s_rock = (opt.rock ? 2 : 0); + sbi->s_mapping = opt->map; + sbi->s_rock = (opt->rock ? 2 : 0); sbi->s_rock_offset = -1; /* initial offset, will guess until SP is found*/ - sbi->s_cruft = opt.cruft; - sbi->s_hide = opt.hide; - sbi->s_showassoc = opt.showassoc; - sbi->s_uid = opt.uid; - sbi->s_gid = opt.gid; - sbi->s_uid_set = opt.uid_set; - sbi->s_gid_set = opt.gid_set; - sbi->s_nocompress = opt.nocompress; - sbi->s_overriderockperm = opt.overriderockperm; + sbi->s_cruft = opt->cruft; + sbi->s_hide = opt->hide; + sbi->s_showassoc = opt->showassoc; + sbi->s_uid = opt->uid; + sbi->s_gid = opt->gid; + sbi->s_uid_set = opt->uid_set; + sbi->s_gid_set = opt->gid_set; + sbi->s_nocompress = opt->nocompress; + sbi->s_overriderockperm = opt->overriderockperm; /* * It would be incredibly stupid to allow people to mark every file * on the disk as suid, so we merely allow them to set the default * permissions. */ - if (opt.fmode != ISOFS_INVALID_MODE) - sbi->s_fmode = opt.fmode & 0777; + if (opt->fmode != ISOFS_INVALID_MODE) + sbi->s_fmode = opt->fmode & 0777; else sbi->s_fmode = ISOFS_INVALID_MODE; - if (opt.dmode != ISOFS_INVALID_MODE) - sbi->s_dmode = opt.dmode & 0777; + if (opt->dmode != ISOFS_INVALID_MODE) + sbi->s_dmode = opt->dmode & 0777; else sbi->s_dmode = ISOFS_INVALID_MODE; @@ -960,12 +922,12 @@ root_found: } } - if (opt.check == 'u') { + if (opt->check == 'u') { /* Only Joliet is case insensitive by default */ if (joliet_level) - opt.check = 'r'; + opt->check = 'r'; else - opt.check = 's'; + opt->check = 's'; } sbi->s_joliet_level = joliet_level; @@ -980,9 +942,9 @@ root_found: table = 0; if (joliet_level) table += 2; - if (opt.check == 'r') + if (opt->check == 'r') table++; - sbi->s_check = opt.check; + sbi->s_check = opt->check; if (table) s->s_d_op = &isofs_dentry_ops[table - 1]; @@ -994,7 +956,7 @@ root_found: goto out_no_inode; } - kfree(opt.iocharset); + kfree(opt->iocharset); return 0; @@ -1023,7 +985,7 @@ out_bad_zone_size: goto out_freebh; out_bad_size: printk(KERN_WARNING "ISOFS: Logical zone size(%d) < hardware blocksize(%u)\n", - orig_zonesize, opt.blocksize); + orig_zonesize, opt->blocksize); goto out_freebh; out_unknown_format: if (!silent) @@ -1033,7 +995,7 @@ out_freebh: brelse(bh); brelse(pri_bh); out_freesbi: - kfree(opt.iocharset); + kfree(opt->iocharset); kfree(sbi); s->s_fs_info = NULL; return error; @@ -1567,18 +1529,63 @@ struct inode *__isofs_iget(struct super_block *sb, return inode; } -static struct dentry *isofs_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data) +static int isofs_get_tree(struct fs_context *fc) { - return mount_bdev(fs_type, flags, dev_name, data, isofs_fill_super); + return get_tree_bdev(fc, isofs_fill_super); +} + +static void isofs_free_fc(struct fs_context *fc) +{ + kfree(fc->fs_private); +} + +static const struct fs_context_operations isofs_context_ops = { + .parse_param = isofs_parse_param, + .get_tree = isofs_get_tree, + .reconfigure = isofs_reconfigure, + .free = isofs_free_fc, +}; + +static int isofs_init_fs_context(struct fs_context *fc) +{ + struct isofs_options *opt; + + opt = kzalloc(sizeof(*opt), GFP_KERNEL); + if (!opt) + return -ENOMEM; + + opt->map = 'n'; + opt->rock = 1; + opt->joliet = 1; + opt->cruft = 0; + opt->hide = 0; + opt->showassoc = 0; + opt->check = 'u'; /* unset */ + opt->nocompress = 0; + opt->blocksize = 1024; + opt->fmode = opt->dmode = ISOFS_INVALID_MODE; + opt->uid_set = 0; + opt->gid_set = 0; + opt->gid = GLOBAL_ROOT_GID; + opt->uid = GLOBAL_ROOT_UID; + opt->iocharset = NULL; + opt->overriderockperm = 0; + opt->session = -1; + opt->sbsector = -1; + + fc->fs_private = opt; + fc->ops = &isofs_context_ops; + + return 0; } static struct file_system_type iso9660_fs_type = { .owner = THIS_MODULE, .name = "iso9660", - .mount = isofs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, + .init_fs_context = isofs_init_fs_context, + .parameters = isofs_param_spec, }; MODULE_ALIAS_FS("iso9660"); MODULE_ALIAS("iso9660"); -- cgit v1.2.3-59-g8ed1b From 8777446a3f2dd6dab0859e4aedef6e0240955fdb Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Fri, 15 Mar 2024 09:19:49 +0000 Subject: udf: Remove second semicolon There is a statement with two semicolons. Remove the second one, it is redundant. Signed-off-by: Colin Ian King Signed-off-by: Jan Kara Message-Id: <20240315091949.2430585-1-colin.i.king@gmail.com> --- fs/udf/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/udf/super.c b/fs/udf/super.c index 2217f7ed7a49..ba6b747a3830 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -630,7 +630,7 @@ static int udf_parse_param(struct fs_context *fc, struct fs_parameter *param) if (!uopt->nls_map) { errorf(fc, "iocharset %s not found", param->string); - return -EINVAL;; + return -EINVAL; } } break; -- cgit v1.2.3-59-g8ed1b From e6595224464b692ddae193d783402130d1625147 Mon Sep 17 00:00:00 2001 From: Nikita Kiryushin Date: Thu, 14 Mar 2024 16:36:56 +0300 Subject: fanotify: remove unneeded sub-zero check for unsigned value Unsigned size_t len in copy_fid_info_to_user is checked for negative value. This check is redundant as it is always false. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: 5e469c830fdb ("fanotify: copy event fid info to user") Signed-off-by: Nikita Kiryushin Signed-off-by: Jan Kara Message-Id: --- fs/notify/fanotify/fanotify_user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index fbdc63cc10d9..4201723357cf 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -502,7 +502,7 @@ static int copy_fid_info_to_user(__kernel_fsid_t *fsid, struct fanotify_fh *fh, } /* Pad with 0's */ - WARN_ON_ONCE(len < 0 || len >= FANOTIFY_EVENT_ALIGN); + WARN_ON_ONCE(len >= FANOTIFY_EVENT_ALIGN); if (len > 0 && clear_user(buf, len)) return -EFAULT; -- cgit v1.2.3-59-g8ed1b From ee745e4736fbf33079d0d0808e1343c2280fd59a Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Thu, 29 Feb 2024 22:38:38 +0800 Subject: f2fs: support .shutdown in f2fs_sops Support .shutdown callback in f2fs_sops, then, it can be called to shut down the file system when underlying block device is marked dead. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 2 ++ fs/f2fs/file.c | 70 +++++++++++++++++++++++++++++++++++---------------------- fs/f2fs/super.c | 6 +++++ 3 files changed, 51 insertions(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index fced2b7652f4..adfce581026b 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3497,6 +3497,8 @@ int f2fs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, struct iattr *attr); int f2fs_truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end); void f2fs_truncate_data_blocks_range(struct dnode_of_data *dn, int count); +int f2fs_do_shutdown(struct f2fs_sb_info *sbi, unsigned int flag, + bool readonly); int f2fs_precache_extents(struct inode *inode); int f2fs_fileattr_get(struct dentry *dentry, struct fileattr *fa); int f2fs_fileattr_set(struct mnt_idmap *idmap, diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 1761ad125f97..f4347db5b03e 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2224,34 +2224,13 @@ static int f2fs_ioc_abort_atomic_write(struct file *filp) return ret; } -static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg) +int f2fs_do_shutdown(struct f2fs_sb_info *sbi, unsigned int flag, + bool readonly) { - struct inode *inode = file_inode(filp); - struct f2fs_sb_info *sbi = F2FS_I_SB(inode); struct super_block *sb = sbi->sb; - __u32 in; int ret = 0; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (get_user(in, (__u32 __user *)arg)) - return -EFAULT; - - if (in != F2FS_GOING_DOWN_FULLSYNC) { - ret = mnt_want_write_file(filp); - if (ret) { - if (ret == -EROFS) { - ret = 0; - f2fs_stop_checkpoint(sbi, false, - STOP_CP_REASON_SHUTDOWN); - trace_f2fs_shutdown(sbi, in, ret); - } - return ret; - } - } - - switch (in) { + switch (flag) { case F2FS_GOING_DOWN_FULLSYNC: ret = bdev_freeze(sb->s_bdev); if (ret) @@ -2290,6 +2269,9 @@ static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg) goto out; } + if (readonly) + goto out; + f2fs_stop_gc_thread(sbi); f2fs_stop_discard_thread(sbi); @@ -2298,10 +2280,44 @@ static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg) f2fs_update_time(sbi, REQ_TIME); out: - if (in != F2FS_GOING_DOWN_FULLSYNC) - mnt_drop_write_file(filp); - trace_f2fs_shutdown(sbi, in, ret); + trace_f2fs_shutdown(sbi, flag, ret); + + return ret; +} + +static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg) +{ + struct inode *inode = file_inode(filp); + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + __u32 in; + int ret; + bool need_drop = false, readonly = false; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (get_user(in, (__u32 __user *)arg)) + return -EFAULT; + + if (in != F2FS_GOING_DOWN_FULLSYNC) { + ret = mnt_want_write_file(filp); + if (ret) { + if (ret != -EROFS) + return ret; + + /* fallback to nosync shutdown for readonly fs */ + in = F2FS_GOING_DOWN_NOSYNC; + readonly = true; + } else { + need_drop = true; + } + } + + ret = f2fs_do_shutdown(sbi, in, readonly); + + if (need_drop) + mnt_drop_write_file(filp); return ret; } diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index a6867f26f141..7c45929671ad 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -2547,6 +2547,11 @@ restore_opts: return err; } +static void f2fs_shutdown(struct super_block *sb) +{ + f2fs_do_shutdown(F2FS_SB(sb), F2FS_GOING_DOWN_NOSYNC, false); +} + #ifdef CONFIG_QUOTA static bool f2fs_need_recovery(struct f2fs_sb_info *sbi) { @@ -3146,6 +3151,7 @@ static const struct super_operations f2fs_sops = { .unfreeze_fs = f2fs_unfreeze, .statfs = f2fs_statfs, .remount_fs = f2fs_remount, + .shutdown = f2fs_shutdown, }; #ifdef CONFIG_FS_ENCRYPTION -- cgit v1.2.3-59-g8ed1b From 2a82bb02941fb53d1f8df2a360e7798ae3d9d962 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 7 Mar 2024 21:29:12 -0500 Subject: statx: stx_subvol Add a new statx field for (sub)volume identifiers, as implemented by btrfs and bcachefs. This includes bcachefs support; we'll definitely want btrfs support as well. Link: https://lore.kernel.org/linux-fsdevel/2uvhm6gweyl7iyyp2xpfryvcu2g3padagaeqcbiavjyiis6prl@yjm725bizncq/ Signed-off-by: Kent Overstreet Cc: Josef Bacik Cc: Miklos Szeredi Cc: Christian Brauner Cc: David Howells Signed-off-by: Kent Overstreet Link: https://lore.kernel.org/r/20240308022914.196982-1-kent.overstreet@linux.dev Signed-off-by: Johannes Thumshirn Signed-off-by: Christian Brauner --- fs/bcachefs/fs.c | 3 +++ fs/btrfs/inode.c | 3 +++ fs/stat.c | 1 + include/linux/stat.h | 1 + include/uapi/linux/stat.h | 4 +++- 5 files changed, 11 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 0ccee05f6887..99a0abeadbe2 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -841,6 +841,9 @@ static int bch2_getattr(struct mnt_idmap *idmap, stat->blksize = block_bytes(c); stat->blocks = inode->v.i_blocks; + stat->subvol = inode->ei_subvol; + stat->result_mask |= STATX_SUBVOL; + if (request_mask & STATX_BTIME) { stat->result_mask |= STATX_BTIME; stat->btime = bch2_time_to_timespec(c, inode->ei_inode.bi_otime); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 37701531eeb1..8cf692c708d7 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8779,6 +8779,9 @@ static int btrfs_getattr(struct mnt_idmap *idmap, generic_fillattr(idmap, request_mask, inode, stat); stat->dev = BTRFS_I(inode)->root->anon_dev; + stat->subvol = BTRFS_I(inode)->root->root_key.objectid; + stat->result_mask |= STATX_SUBVOL; + spin_lock(&BTRFS_I(inode)->lock); delalloc_bytes = BTRFS_I(inode)->new_delalloc_bytes; inode_bytes = inode_get_bytes(inode); diff --git a/fs/stat.c b/fs/stat.c index 77cdc69eb422..70bd3e888cfa 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -658,6 +658,7 @@ cp_statx(const struct kstat *stat, struct statx __user *buffer) tmp.stx_mnt_id = stat->mnt_id; tmp.stx_dio_mem_align = stat->dio_mem_align; tmp.stx_dio_offset_align = stat->dio_offset_align; + tmp.stx_subvol = stat->subvol; return copy_to_user(buffer, &tmp, sizeof(tmp)) ? -EFAULT : 0; } diff --git a/include/linux/stat.h b/include/linux/stat.h index 52150570d37a..bf92441dbad2 100644 --- a/include/linux/stat.h +++ b/include/linux/stat.h @@ -53,6 +53,7 @@ struct kstat { u32 dio_mem_align; u32 dio_offset_align; u64 change_cookie; + u64 subvol; }; /* These definitions are internal to the kernel for now. Mainly used by nfsd. */ diff --git a/include/uapi/linux/stat.h b/include/uapi/linux/stat.h index 2f2ee82d5517..67626d535316 100644 --- a/include/uapi/linux/stat.h +++ b/include/uapi/linux/stat.h @@ -126,8 +126,9 @@ struct statx { __u64 stx_mnt_id; __u32 stx_dio_mem_align; /* Memory buffer alignment for direct I/O */ __u32 stx_dio_offset_align; /* File offset alignment for direct I/O */ + __u64 stx_subvol; /* Subvolume identifier */ /* 0xa0 */ - __u64 __spare3[12]; /* Spare space for future expansion */ + __u64 __spare3[11]; /* Spare space for future expansion */ /* 0x100 */ }; @@ -155,6 +156,7 @@ struct statx { #define STATX_MNT_ID 0x00001000U /* Got stx_mnt_id */ #define STATX_DIOALIGN 0x00002000U /* Want/got direct I/O alignment info */ #define STATX_MNT_ID_UNIQUE 0x00004000U /* Want/got extended stx_mount_id */ +#define STATX_SUBVOL 0x00008000U /* Want/got stx_subvol */ #define STATX__RESERVED 0x80000000U /* Reserved for future struct statx expansion */ -- cgit v1.2.3-59-g8ed1b From 7b30851a70645400ba37a184b9a2f10e1d6987a6 Mon Sep 17 00:00:00 2001 From: "Luis Henriques (SUSE)" Date: Tue, 12 Mar 2024 10:47:57 +0000 Subject: fs_parser: move fsparam_string_empty() helper into header Since both ext4 and overlayfs define the same macro to specify string parameters that may allow empty values, define it in an header file so that this helper can be shared. Signed-off-by: Luis Henriques (SUSE) Link: https://lore.kernel.org/r/20240312104757.27333-1-luis.henriques@linux.dev Signed-off-by: Christian Brauner --- fs/ext4/super.c | 4 ---- fs/overlayfs/params.c | 4 ---- include/linux/fs_parser.h | 4 ++++ 3 files changed, 4 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index cfb8449c731f..2e906dbbe015 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1723,10 +1723,6 @@ static const struct constant_table ext4_param_dax[] = { {} }; -/* String parameter that allows empty argument */ -#define fsparam_string_empty(NAME, OPT) \ - __fsparam(fs_param_is_string, NAME, OPT, fs_param_can_be_empty, NULL) - /* * Mount option specification * We don't use fsparam_flag_no because of the way we set the diff --git a/fs/overlayfs/params.c b/fs/overlayfs/params.c index 36dcc530ac28..4860fcc4611b 100644 --- a/fs/overlayfs/params.c +++ b/fs/overlayfs/params.c @@ -139,10 +139,6 @@ static int ovl_verity_mode_def(void) return OVL_VERITY_OFF; } -#define fsparam_string_empty(NAME, OPT) \ - __fsparam(fs_param_is_string, NAME, OPT, fs_param_can_be_empty, NULL) - - const struct fs_parameter_spec ovl_parameter_spec[] = { fsparam_string_empty("lowerdir", Opt_lowerdir), fsparam_string("lowerdir+", Opt_lowerdir_add), diff --git a/include/linux/fs_parser.h b/include/linux/fs_parser.h index 01542c4b87a2..d3350979115f 100644 --- a/include/linux/fs_parser.h +++ b/include/linux/fs_parser.h @@ -132,4 +132,8 @@ static inline bool fs_validate_description(const char *name, #define fsparam_path(NAME, OPT) __fsparam(fs_param_is_path, NAME, OPT, 0, NULL) #define fsparam_fd(NAME, OPT) __fsparam(fs_param_is_fd, NAME, OPT, 0, NULL) +/* String parameter that allows empty argument */ +#define fsparam_string_empty(NAME, OPT) \ + __fsparam(fs_param_is_string, NAME, OPT, fs_param_can_be_empty, NULL) + #endif /* _LINUX_FS_PARSER_H */ -- cgit v1.2.3-59-g8ed1b From fc253215f8e8d2a5f6f057b2a9dd92c7c5134f12 Mon Sep 17 00:00:00 2001 From: Yang Li Date: Fri, 15 Mar 2024 15:38:05 +0800 Subject: fs: Add kernel-doc comments to proc_create_net_data_write() This commit adds kernel-doc style comments with complete parameter descriptions for the function proc_create_net_data_write. Signed-off-by: Yang Li Link: https://lore.kernel.org/r/20240315073805.77463-1-yang.lee@linux.alibaba.com Signed-off-by: Christian Brauner --- fs/proc/proc_net.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 2ba31b6d68c0..52f0b75cbce2 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -135,6 +135,7 @@ EXPORT_SYMBOL_GPL(proc_create_net_data); * @parent: The parent directory in which to create. * @ops: The seq_file ops with which to read the file. * @write: The write method with which to 'modify' the file. + * @state_size: The size of the per-file private state to allocate. * @data: Data for retrieval by pde_data(). * * Create a network namespaced proc file in the @parent directory with the -- cgit v1.2.3-59-g8ed1b From ac0c18f2c693f0e7a44dbbb36b14d5141e5d20e5 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 28 Feb 2024 17:19:53 +0800 Subject: fs/writeback: avoid to writeback non-expired inode in kupdate writeback In kupdate writeback, only expired inode (have been dirty for longer than dirty_expire_interval) is supposed to be written back. However, kupdate writeback will writeback non-expired inode left in b_io or b_more_io from last wb_writeback. As a result, writeback will keep being triggered unexpected when we keep dirtying pages even dirty memory is under threshold and inode is not expired. To be more specific: Assume dirty background threshold is > 1G and dirty_expire_centisecs is > 60s. When we running fio -size=1G -invalidate=0 -ioengine=libaio --time_based -runtime=60... (keep dirtying), the writeback will keep being triggered as following: wb_workfn wb_do_writeback wb_check_background_flush /* * Wb dirty background threshold starts at 0 if device was idle and * grows up when bandwidth of wb is updated. So a background * writeback is triggered. */ wb_over_bg_thresh /* * Dirtied inode will be written back and added to b_more_io list * after slice used up (because we keep dirtying the inode). */ wb_writeback Writeback is triggered per dirty_writeback_centisecs as following: wb_workfn wb_do_writeback wb_check_old_data_flush /* * Write back inode left in b_io and b_more_io from last wb_writeback * even the inode is non-expired and it will be added to b_more_io * again as slice will be used up (because we keep dirtying the * inode) */ wb_writeback Fix this by moving non-expired inode to dirty list instead of more io list for kupdate writeback in requeue_inode. Test as following: /* make it more easier to observe the issue */ echo 300000 > /proc/sys/vm/dirty_expire_centisecs echo 100 > /proc/sys/vm/dirty_writeback_centisecs /* create a idle device */ mkfs.ext4 -F /dev/vdb mount /dev/vdb /bdi1/ /* run buffer write with fio */ fio -name test -filename=/bdi1/file -size=800M -ioengine=libaio -bs=4K \ -iodepth=1 -rw=write -direct=0 --time_based -runtime=60 -invalidate=0 Fio result before fix (run three tests): 1360MB/s 1329MB/s 1455MB/s Fio result after fix (run three tests): 1737MB/s 1729MB/s 1789MB/s Writeback for non-expired inode is gone as expeted. Observe this with trace writeback_start and writeback_written as following: echo 1 > /sys/kernel/debug/tracing/events/writeback/writeback_start/enab echo 1 > /sys/kernel/debug/tracing/events/writeback/writeback_written/enable cat /sys/kernel/tracing/trace_pipe Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240228091958.288260-2-shikemeng@huaweicloud.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/fs-writeback.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index e4f17c53ddfc..fe634f00f4d9 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1561,7 +1561,8 @@ static void inode_sleep_on_writeback(struct inode *inode) * thread's back can have unexpected consequences. */ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, - struct writeback_control *wbc) + struct writeback_control *wbc, + unsigned long dirtied_before) { if (inode->i_state & I_FREEING) return; @@ -1594,7 +1595,8 @@ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, * We didn't write back all the pages. nfs_writepages() * sometimes bales out without doing anything. */ - if (wbc->nr_to_write <= 0) { + if (wbc->nr_to_write <= 0 && + !inode_dirtied_after(inode, dirtied_before)) { /* Slice used up. Queue for next turn. */ requeue_io(inode, wb); } else { @@ -1862,6 +1864,11 @@ static long writeback_sb_inodes(struct super_block *sb, unsigned long start_time = jiffies; long write_chunk; long total_wrote = 0; /* count both pages and inodes */ + unsigned long dirtied_before = jiffies; + + if (work->for_kupdate) + dirtied_before = jiffies - + msecs_to_jiffies(dirty_expire_interval * 10); while (!list_empty(&wb->b_io)) { struct inode *inode = wb_inode(wb->b_io.prev); @@ -1967,7 +1974,7 @@ static long writeback_sb_inodes(struct super_block *sb, spin_lock(&inode->i_lock); if (!(inode->i_state & I_DIRTY_ALL)) total_wrote++; - requeue_inode(inode, tmp_wb, &wbc); + requeue_inode(inode, tmp_wb, &wbc, dirtied_before); inode_sync_complete(inode); spin_unlock(&inode->i_lock); -- cgit v1.2.3-59-g8ed1b From 491681d44bf24fa9c33981d634d967cab2f59f78 Mon Sep 17 00:00:00 2001 From: Bill O'Donnell Date: Sat, 2 Mar 2024 10:48:34 -0600 Subject: qnx6: convert qnx6 to use the new mount api Convert the qnx6 filesystem to use the new mount API. Mostly untested, since there is no qnx6 fs image readily available. Testing did include parsing of the mmi_fs option. Signed-off-by: Bill O'Donnell Link: https://lore.kernel.org/r/20240302165714.859504-1-bodonnel@redhat.com Signed-off-by: Christian Brauner --- fs/qnx6/inode.c | 117 +++++++++++++++++++++++++++++++++----------------------- 1 file changed, 69 insertions(+), 48 deletions(-) (limited to 'fs') diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c index 405913f4faff..d62fbef838b6 100644 --- a/fs/qnx6/inode.c +++ b/fs/qnx6/inode.c @@ -19,11 +19,11 @@ #include #include #include -#include #include -#include #include #include +#include +#include #include "qnx6.h" static const struct super_operations qnx6_sops; @@ -31,7 +31,7 @@ static const struct super_operations qnx6_sops; static void qnx6_put_super(struct super_block *sb); static struct inode *qnx6_alloc_inode(struct super_block *sb); static void qnx6_free_inode(struct inode *inode); -static int qnx6_remount(struct super_block *sb, int *flags, char *data); +static int qnx6_reconfigure(struct fs_context *fc); static int qnx6_statfs(struct dentry *dentry, struct kstatfs *buf); static int qnx6_show_options(struct seq_file *seq, struct dentry *root); @@ -40,7 +40,6 @@ static const struct super_operations qnx6_sops = { .free_inode = qnx6_free_inode, .put_super = qnx6_put_super, .statfs = qnx6_statfs, - .remount_fs = qnx6_remount, .show_options = qnx6_show_options, }; @@ -54,10 +53,12 @@ static int qnx6_show_options(struct seq_file *seq, struct dentry *root) return 0; } -static int qnx6_remount(struct super_block *sb, int *flags, char *data) +static int qnx6_reconfigure(struct fs_context *fc) { + struct super_block *sb = fc->root->d_sb; + sync_filesystem(sb); - *flags |= SB_RDONLY; + fc->sb_flags |= SB_RDONLY; return 0; } @@ -218,39 +219,36 @@ void qnx6_superblock_debug(struct qnx6_super_block *sb, struct super_block *s) #endif enum { - Opt_mmifs, - Opt_err + Opt_mmifs +}; + +struct qnx6_context { + unsigned long s_mount_opts; }; -static const match_table_t tokens = { - {Opt_mmifs, "mmi_fs"}, - {Opt_err, NULL} +static const struct fs_parameter_spec qnx6_param_spec[] = { + fsparam_flag ("mmi_fs", Opt_mmifs), + {} }; -static int qnx6_parse_options(char *options, struct super_block *sb) +static int qnx6_parse_param(struct fs_context *fc, struct fs_parameter *param) { - char *p; - struct qnx6_sb_info *sbi = QNX6_SB(sb); - substring_t args[MAX_OPT_ARGS]; - - if (!options) - return 1; - - while ((p = strsep(&options, ",")) != NULL) { - int token; - if (!*p) - continue; - - token = match_token(p, tokens, args); - switch (token) { - case Opt_mmifs: - set_opt(sbi->s_mount_opt, MMI_FS); - break; - default: - return 0; - } + struct qnx6_context *ctx = fc->fs_private; + struct fs_parse_result result; + int opt; + + opt = fs_parse(fc, qnx6_param_spec, param, &result); + if (opt < 0) + return opt; + + switch (opt) { + case Opt_mmifs: + ctx->s_mount_opts |= QNX6_MOUNT_MMI_FS; + break; + default: + return -EINVAL; } - return 1; + return 0; } static struct buffer_head *qnx6_check_first_superblock(struct super_block *s, @@ -293,22 +291,25 @@ static struct buffer_head *qnx6_check_first_superblock(struct super_block *s, static struct inode *qnx6_private_inode(struct super_block *s, struct qnx6_root_node *p); -static int qnx6_fill_super(struct super_block *s, void *data, int silent) +static int qnx6_fill_super(struct super_block *s, struct fs_context *fc) { struct buffer_head *bh1 = NULL, *bh2 = NULL; struct qnx6_super_block *sb1 = NULL, *sb2 = NULL; struct qnx6_sb_info *sbi; + struct qnx6_context *ctx = fc->fs_private; struct inode *root; const char *errmsg; struct qnx6_sb_info *qs; int ret = -EINVAL; u64 offset; int bootblock_offset = QNX6_BOOTBLOCK_SIZE; + int silent = fc->sb_flags & SB_SILENT; qs = kzalloc(sizeof(struct qnx6_sb_info), GFP_KERNEL); if (!qs) return -ENOMEM; s->s_fs_info = qs; + qs->s_mount_opt = ctx->s_mount_opts; /* Superblock always is 512 Byte long */ if (!sb_set_blocksize(s, QNX6_SUPERBLOCK_SIZE)) { @@ -316,12 +317,7 @@ static int qnx6_fill_super(struct super_block *s, void *data, int silent) goto outnobh; } - /* parse the mount-options */ - if (!qnx6_parse_options((char *) data, s)) { - pr_err("invalid mount options.\n"); - goto outnobh; - } - if (test_opt(s, MMI_FS)) { + if (qs->s_mount_opt == QNX6_MOUNT_MMI_FS) { sb1 = qnx6_mmi_fill_super(s, silent); if (sb1) goto mmi_success; @@ -632,18 +628,43 @@ static void destroy_inodecache(void) kmem_cache_destroy(qnx6_inode_cachep); } -static struct dentry *qnx6_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data) +static int qnx6_get_tree(struct fs_context *fc) +{ + return get_tree_bdev(fc, qnx6_fill_super); +} + +static void qnx6_free_fc(struct fs_context *fc) { - return mount_bdev(fs_type, flags, dev_name, data, qnx6_fill_super); + kfree(fc->fs_private); +} + +static const struct fs_context_operations qnx6_context_ops = { + .parse_param = qnx6_parse_param, + .get_tree = qnx6_get_tree, + .reconfigure = qnx6_reconfigure, + .free = qnx6_free_fc, +}; + +static int qnx6_init_fs_context(struct fs_context *fc) +{ + struct qnx6_context *ctx; + + ctx = kzalloc(sizeof(struct qnx6_context), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + fc->ops = &qnx6_context_ops; + fc->fs_private = ctx; + + return 0; } static struct file_system_type qnx6_fs_type = { - .owner = THIS_MODULE, - .name = "qnx6", - .mount = qnx6_mount, - .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .owner = THIS_MODULE, + .name = "qnx6", + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, + .init_fs_context = qnx6_init_fs_context, + .parameters = qnx6_param_spec, }; MODULE_ALIAS_FS("qnx6"); -- cgit v1.2.3-59-g8ed1b From b01e1a6359df16d432bfdfb31b04da78236f45b8 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Fri, 1 Mar 2024 17:04:31 -0600 Subject: freevxfs: Convert freevxfs to the new mount API. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Convert the freevxfs filesystem to the new mount API. Signed-off-by: Eric Sandeen Link: https://lore.kernel.org/r/b0d1a423-4b8e-4bc1-a021-a1078aee915f@redhat.com Tested-by: Krzysztof Błaszkowski Signed-off-by: Christian Brauner --- fs/freevxfs/vxfs_super.c | 69 ++++++++++++++++++++++++++++-------------------- 1 file changed, 40 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/freevxfs/vxfs_super.c b/fs/freevxfs/vxfs_super.c index 42e03b6b1cc7..fabe60778658 100644 --- a/fs/freevxfs/vxfs_super.c +++ b/fs/freevxfs/vxfs_super.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include #include "vxfs.h" #include "vxfs_extern.h" @@ -91,10 +91,10 @@ vxfs_statfs(struct dentry *dentry, struct kstatfs *bufp) return 0; } -static int vxfs_remount(struct super_block *sb, int *flags, char *data) +static int vxfs_reconfigure(struct fs_context *fc) { - sync_filesystem(sb); - *flags |= SB_RDONLY; + sync_filesystem(fc->root->d_sb); + fc->sb_flags |= SB_RDONLY; return 0; } @@ -120,24 +120,24 @@ static const struct super_operations vxfs_super_ops = { .evict_inode = vxfs_evict_inode, .put_super = vxfs_put_super, .statfs = vxfs_statfs, - .remount_fs = vxfs_remount, }; -static int vxfs_try_sb_magic(struct super_block *sbp, int silent, +static int vxfs_try_sb_magic(struct super_block *sbp, struct fs_context *fc, unsigned blk, __fs32 magic) { struct buffer_head *bp; struct vxfs_sb *rsbp; struct vxfs_sb_info *infp = VXFS_SBI(sbp); + int silent = fc->sb_flags & SB_SILENT; int rc = -ENOMEM; bp = sb_bread(sbp, blk); do { if (!bp || !buffer_mapped(bp)) { if (!silent) { - printk(KERN_WARNING - "vxfs: unable to read disk superblock at %u\n", - blk); + warnf(fc, + "vxfs: unable to read disk superblock at %u", + blk); } break; } @@ -146,9 +146,9 @@ static int vxfs_try_sb_magic(struct super_block *sbp, int silent, rsbp = (struct vxfs_sb *)bp->b_data; if (rsbp->vs_magic != magic) { if (!silent) - printk(KERN_NOTICE - "vxfs: WRONG superblock magic %08x at %u\n", - rsbp->vs_magic, blk); + infof(fc, + "vxfs: WRONG superblock magic %08x at %u", + rsbp->vs_magic, blk); break; } @@ -169,8 +169,7 @@ static int vxfs_try_sb_magic(struct super_block *sbp, int silent, /** * vxfs_fill_super - read superblock into memory and initialize filesystem * @sbp: VFS superblock (to fill) - * @dp: fs private mount data - * @silent: do not complain loudly when sth is wrong + * @fc: filesytem context * * Description: * We are called on the first mount of a filesystem to read the @@ -182,26 +181,27 @@ static int vxfs_try_sb_magic(struct super_block *sbp, int silent, * Locking: * We are under @sbp->s_lock. */ -static int vxfs_fill_super(struct super_block *sbp, void *dp, int silent) +static int vxfs_fill_super(struct super_block *sbp, struct fs_context *fc) { struct vxfs_sb_info *infp; struct vxfs_sb *rsbp; u_long bsize; struct inode *root; int ret = -EINVAL; + int silent = fc->sb_flags & SB_SILENT; u32 j; sbp->s_flags |= SB_RDONLY; infp = kzalloc(sizeof(*infp), GFP_KERNEL); if (!infp) { - printk(KERN_WARNING "vxfs: unable to allocate incore superblock\n"); + warnf(fc, "vxfs: unable to allocate incore superblock"); return -ENOMEM; } bsize = sb_min_blocksize(sbp, BLOCK_SIZE); if (!bsize) { - printk(KERN_WARNING "vxfs: unable to set blocksize\n"); + warnf(fc, "vxfs: unable to set blocksize"); goto out; } @@ -210,24 +210,24 @@ static int vxfs_fill_super(struct super_block *sbp, void *dp, int silent) sbp->s_time_min = 0; sbp->s_time_max = U32_MAX; - if (!vxfs_try_sb_magic(sbp, silent, 1, + if (!vxfs_try_sb_magic(sbp, fc, 1, (__force __fs32)cpu_to_le32(VXFS_SUPER_MAGIC))) { /* Unixware, x86 */ infp->byte_order = VXFS_BO_LE; - } else if (!vxfs_try_sb_magic(sbp, silent, 8, + } else if (!vxfs_try_sb_magic(sbp, fc, 8, (__force __fs32)cpu_to_be32(VXFS_SUPER_MAGIC))) { /* HP-UX, parisc */ infp->byte_order = VXFS_BO_BE; } else { if (!silent) - printk(KERN_NOTICE "vxfs: can't find superblock.\n"); + infof(fc, "vxfs: can't find superblock."); goto out; } rsbp = infp->vsi_raw; j = fs32_to_cpu(infp, rsbp->vs_version); if ((j < 2 || j > 4) && !silent) { - printk(KERN_NOTICE "vxfs: unsupported VxFS version (%d)\n", j); + infof(fc, "vxfs: unsupported VxFS version (%d)", j); goto out; } @@ -244,17 +244,17 @@ static int vxfs_fill_super(struct super_block *sbp, void *dp, int silent) j = fs32_to_cpu(infp, rsbp->vs_bsize); if (!sb_set_blocksize(sbp, j)) { - printk(KERN_WARNING "vxfs: unable to set final block size\n"); + warnf(fc, "vxfs: unable to set final block size"); goto out; } if (vxfs_read_olt(sbp, bsize)) { - printk(KERN_WARNING "vxfs: unable to read olt\n"); + warnf(fc, "vxfs: unable to read olt"); goto out; } if (vxfs_read_fshead(sbp)) { - printk(KERN_WARNING "vxfs: unable to read fshead\n"); + warnf(fc, "vxfs: unable to read fshead"); goto out; } @@ -265,7 +265,7 @@ static int vxfs_fill_super(struct super_block *sbp, void *dp, int silent) } sbp->s_root = d_make_root(root); if (!sbp->s_root) { - printk(KERN_WARNING "vxfs: unable to get root dentry.\n"); + warnf(fc, "vxfs: unable to get root dentry."); goto out_free_ilist; } @@ -284,18 +284,29 @@ out: /* * The usual module blurb. */ -static struct dentry *vxfs_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data) +static int vxfs_get_tree(struct fs_context *fc) { - return mount_bdev(fs_type, flags, dev_name, data, vxfs_fill_super); + return get_tree_bdev(fc, vxfs_fill_super); +} + +static const struct fs_context_operations vxfs_context_ops = { + .get_tree = vxfs_get_tree, + .reconfigure = vxfs_reconfigure, +}; + +static int vxfs_init_fs_context(struct fs_context *fc) +{ + fc->ops = &vxfs_context_ops; + + return 0; } static struct file_system_type vxfs_fs_type = { .owner = THIS_MODULE, .name = "vxfs", - .mount = vxfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, + .init_fs_context = vxfs_init_fs_context, }; MODULE_ALIAS_FS("vxfs"); /* makes mount -t vxfs autoload the module */ MODULE_ALIAS("vxfs"); -- cgit v1.2.3-59-g8ed1b From 8f27829974b025d4df2e78894105d75e3bf349f0 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Fri, 1 Mar 2024 16:33:11 -0600 Subject: openpromfs: finish conversion to the new mount API The original mount API conversion inexplicably left out the change from ->remount_fs to ->reconfigure; do that now. Fixes: 7ab2fa7693c3 ("vfs: Convert openpromfs to use the new mount API") Signed-off-by: Eric Sandeen Link: https://lore.kernel.org/r/90b968aa-c979-420f-ba37-5acc3391b28f@redhat.com Signed-off-by: Christian Brauner --- fs/openpromfs/inode.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 4a0779e3ef79..a7b527ea50d3 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -355,10 +355,10 @@ static struct inode *openprom_iget(struct super_block *sb, ino_t ino) return inode; } -static int openprom_remount(struct super_block *sb, int *flags, char *data) +static int openpromfs_reconfigure(struct fs_context *fc) { - sync_filesystem(sb); - *flags |= SB_NOATIME; + sync_filesystem(fc->root->d_sb); + fc->sb_flags |= SB_NOATIME; return 0; } @@ -366,7 +366,6 @@ static const struct super_operations openprom_sops = { .alloc_inode = openprom_alloc_inode, .free_inode = openprom_free_inode, .statfs = simple_statfs, - .remount_fs = openprom_remount, }; static int openprom_fill_super(struct super_block *s, struct fs_context *fc) @@ -415,6 +414,7 @@ static int openpromfs_get_tree(struct fs_context *fc) static const struct fs_context_operations openpromfs_context_ops = { .get_tree = openpromfs_get_tree, + .reconfigure = openpromfs_reconfigure, }; static int openpromfs_init_fs_context(struct fs_context *fc) -- cgit v1.2.3-59-g8ed1b From a20971c187522f5a7cd8e961e7e9c88f31ea2bed Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 5 Mar 2024 17:08:39 -0600 Subject: vfs: Convert debugfs to use the new mount API Convert the debugfs filesystem to the new internal mount API as the old one will be obsoleted and removed. This allows greater flexibility in communication of mount parameters between userspace, the VFS and the filesystem. See Documentation/filesystems/mount_api.txt for more information. Signed-off-by: David Howells Co-developed-by: Eric Sandeen Signed-off-by: Eric Sandeen [sandeen: forward port to modern kernel, fix remounting] Link: https://lore.kernel.org/r/49d1f108-46e3-443f-85a3-6dd730c5d076@redhat.com cc: Greg Kroah-Hartman cc: Rafael J. Wysocki Signed-off-by: Christian Brauner --- fs/debugfs/inode.c | 198 +++++++++++++++++++++++++---------------------------- 1 file changed, 93 insertions(+), 105 deletions(-) (limited to 'fs') diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index a40da0065433..dc51df0b118d 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -14,7 +14,8 @@ #include #include -#include +#include +#include #include #include #include @@ -23,7 +24,6 @@ #include #include #include -#include #include #include #include @@ -77,7 +77,7 @@ static struct inode *debugfs_get_inode(struct super_block *sb) return inode; } -struct debugfs_mount_opts { +struct debugfs_fs_info { kuid_t uid; kgid_t gid; umode_t mode; @@ -89,68 +89,51 @@ enum { Opt_uid, Opt_gid, Opt_mode, - Opt_err }; -static const match_table_t tokens = { - {Opt_uid, "uid=%u"}, - {Opt_gid, "gid=%u"}, - {Opt_mode, "mode=%o"}, - {Opt_err, NULL} +static const struct fs_parameter_spec debugfs_param_specs[] = { + fsparam_u32 ("gid", Opt_gid), + fsparam_u32oct ("mode", Opt_mode), + fsparam_u32 ("uid", Opt_uid), + {} }; -struct debugfs_fs_info { - struct debugfs_mount_opts mount_opts; -}; - -static int debugfs_parse_options(char *data, struct debugfs_mount_opts *opts) +static int debugfs_parse_param(struct fs_context *fc, struct fs_parameter *param) { - substring_t args[MAX_OPT_ARGS]; - int option; - int token; + struct debugfs_fs_info *opts = fc->s_fs_info; + struct fs_parse_result result; kuid_t uid; kgid_t gid; - char *p; - - opts->opts = 0; - opts->mode = DEBUGFS_DEFAULT_MODE; - - while ((p = strsep(&data, ",")) != NULL) { - if (!*p) - continue; - - token = match_token(p, tokens, args); - switch (token) { - case Opt_uid: - if (match_int(&args[0], &option)) - return -EINVAL; - uid = make_kuid(current_user_ns(), option); - if (!uid_valid(uid)) - return -EINVAL; - opts->uid = uid; - break; - case Opt_gid: - if (match_int(&args[0], &option)) - return -EINVAL; - gid = make_kgid(current_user_ns(), option); - if (!gid_valid(gid)) - return -EINVAL; - opts->gid = gid; - break; - case Opt_mode: - if (match_octal(&args[0], &option)) - return -EINVAL; - opts->mode = option & S_IALLUGO; - break; - /* - * We might like to report bad mount options here; - * but traditionally debugfs has ignored all mount options - */ - } - - opts->opts |= BIT(token); + int opt; + + opt = fs_parse(fc, debugfs_param_specs, param, &result); + if (opt < 0) + return opt; + + switch (opt) { + case Opt_uid: + uid = make_kuid(current_user_ns(), result.uint_32); + if (!uid_valid(uid)) + return invalf(fc, "Unknown uid"); + opts->uid = uid; + break; + case Opt_gid: + gid = make_kgid(current_user_ns(), result.uint_32); + if (!gid_valid(gid)) + return invalf(fc, "Unknown gid"); + opts->gid = gid; + break; + case Opt_mode: + opts->mode = result.uint_32 & S_IALLUGO; + break; + /* + * We might like to report bad mount options here; + * but traditionally debugfs has ignored all mount options + */ } + opts->opts |= BIT(opt); + return 0; } @@ -158,23 +141,22 @@ static void _debugfs_apply_options(struct super_block *sb, bool remount) { struct debugfs_fs_info *fsi = sb->s_fs_info; struct inode *inode = d_inode(sb->s_root); - struct debugfs_mount_opts *opts = &fsi->mount_opts; /* * On remount, only reset mode/uid/gid if they were provided as mount * options. */ - if (!remount || opts->opts & BIT(Opt_mode)) { + if (!remount || fsi->opts & BIT(Opt_mode)) { inode->i_mode &= ~S_IALLUGO; - inode->i_mode |= opts->mode; + inode->i_mode |= fsi->mode; } - if (!remount || opts->opts & BIT(Opt_uid)) - inode->i_uid = opts->uid; + if (!remount || fsi->opts & BIT(Opt_uid)) + inode->i_uid = fsi->uid; - if (!remount || opts->opts & BIT(Opt_gid)) - inode->i_gid = opts->gid; + if (!remount || fsi->opts & BIT(Opt_gid)) + inode->i_gid = fsi->gid; } static void debugfs_apply_options(struct super_block *sb) @@ -187,35 +169,33 @@ static void debugfs_apply_options_remount(struct super_block *sb) _debugfs_apply_options(sb, true); } -static int debugfs_remount(struct super_block *sb, int *flags, char *data) +static int debugfs_reconfigure(struct fs_context *fc) { - int err; - struct debugfs_fs_info *fsi = sb->s_fs_info; + struct super_block *sb = fc->root->d_sb; + struct debugfs_fs_info *sb_opts = sb->s_fs_info; + struct debugfs_fs_info *new_opts = fc->s_fs_info; sync_filesystem(sb); - err = debugfs_parse_options(data, &fsi->mount_opts); - if (err) - goto fail; + /* structure copy of new mount options to sb */ + *sb_opts = *new_opts; debugfs_apply_options_remount(sb); -fail: - return err; + return 0; } static int debugfs_show_options(struct seq_file *m, struct dentry *root) { struct debugfs_fs_info *fsi = root->d_sb->s_fs_info; - struct debugfs_mount_opts *opts = &fsi->mount_opts; - if (!uid_eq(opts->uid, GLOBAL_ROOT_UID)) + if (!uid_eq(fsi->uid, GLOBAL_ROOT_UID)) seq_printf(m, ",uid=%u", - from_kuid_munged(&init_user_ns, opts->uid)); - if (!gid_eq(opts->gid, GLOBAL_ROOT_GID)) + from_kuid_munged(&init_user_ns, fsi->uid)); + if (!gid_eq(fsi->gid, GLOBAL_ROOT_GID)) seq_printf(m, ",gid=%u", - from_kgid_munged(&init_user_ns, opts->gid)); - if (opts->mode != DEBUGFS_DEFAULT_MODE) - seq_printf(m, ",mode=%o", opts->mode); + from_kgid_munged(&init_user_ns, fsi->gid)); + if (fsi->mode != DEBUGFS_DEFAULT_MODE) + seq_printf(m, ",mode=%o", fsi->mode); return 0; } @@ -229,7 +209,6 @@ static void debugfs_free_inode(struct inode *inode) static const struct super_operations debugfs_super_operations = { .statfs = simple_statfs, - .remount_fs = debugfs_remount, .show_options = debugfs_show_options, .free_inode = debugfs_free_inode, }; @@ -263,26 +242,14 @@ static const struct dentry_operations debugfs_dops = { .d_automount = debugfs_automount, }; -static int debug_fill_super(struct super_block *sb, void *data, int silent) +static int debugfs_fill_super(struct super_block *sb, struct fs_context *fc) { static const struct tree_descr debug_files[] = {{""}}; - struct debugfs_fs_info *fsi; int err; - fsi = kzalloc(sizeof(struct debugfs_fs_info), GFP_KERNEL); - sb->s_fs_info = fsi; - if (!fsi) { - err = -ENOMEM; - goto fail; - } - - err = debugfs_parse_options(data, &fsi->mount_opts); + err = simple_fill_super(sb, DEBUGFS_MAGIC, debug_files); if (err) - goto fail; - - err = simple_fill_super(sb, DEBUGFS_MAGIC, debug_files); - if (err) - goto fail; + return err; sb->s_op = &debugfs_super_operations; sb->s_d_op = &debugfs_dops; @@ -290,27 +257,48 @@ static int debug_fill_super(struct super_block *sb, void *data, int silent) debugfs_apply_options(sb); return 0; - -fail: - kfree(fsi); - sb->s_fs_info = NULL; - return err; } -static struct dentry *debug_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, - void *data) +static int debugfs_get_tree(struct fs_context *fc) { if (!(debugfs_allow & DEBUGFS_ALLOW_API)) - return ERR_PTR(-EPERM); + return -EPERM; + + return get_tree_single(fc, debugfs_fill_super); +} + +static void debugfs_free_fc(struct fs_context *fc) +{ + kfree(fc->s_fs_info); +} - return mount_single(fs_type, flags, data, debug_fill_super); +static const struct fs_context_operations debugfs_context_ops = { + .free = debugfs_free_fc, + .parse_param = debugfs_parse_param, + .get_tree = debugfs_get_tree, + .reconfigure = debugfs_reconfigure, +}; + +static int debugfs_init_fs_context(struct fs_context *fc) +{ + struct debugfs_fs_info *fsi; + + fsi = kzalloc(sizeof(struct debugfs_fs_info), GFP_KERNEL); + if (!fsi) + return -ENOMEM; + + fsi->mode = DEBUGFS_DEFAULT_MODE; + + fc->s_fs_info = fsi; + fc->ops = &debugfs_context_ops; + return 0; } static struct file_system_type debug_fs_type = { .owner = THIS_MODULE, .name = "debugfs", - .mount = debug_mount, + .init_fs_context = debugfs_init_fs_context, + .parameters = debugfs_param_specs, .kill_sb = kill_litter_super, }; MODULE_ALIAS_FS("debugfs"); -- cgit v1.2.3-59-g8ed1b From 78ff640819496212feea29c62174f3eb3c837134 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 5 Mar 2024 17:09:30 -0600 Subject: vfs: Convert tracefs to use the new mount API Convert the tracefs filesystem to the new internal mount API as the old one will be obsoleted and removed. This allows greater flexibility in communication of mount parameters between userspace, the VFS and the filesystem. See Documentation/filesystems/mount_api.txt for more information. Signed-off-by: David Howells Co-developed-by: Eric Sandeen Signed-off-by: Eric Sandeen [sandeen: forward port to modern kernel, fix remounting] Link: https://lore.kernel.org/r/536e99d3-345c-448b-adee-a21389d7ab4b@redhat.com cc: Steven Rostedt cc: Greg Kroah-Hartman Signed-off-by: Christian Brauner --- fs/tracefs/inode.c | 196 +++++++++++++++++++++++++---------------------------- 1 file changed, 91 insertions(+), 105 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index 5545e6bf7d26..abd0b477e287 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -11,14 +11,14 @@ #include #include -#include +#include +#include #include #include #include #include #include #include -#include #include #include #include "internal.h" @@ -231,7 +231,7 @@ struct inode *tracefs_get_inode(struct super_block *sb) return inode; } -struct tracefs_mount_opts { +struct tracefs_fs_info { kuid_t uid; kgid_t gid; umode_t mode; @@ -243,68 +243,51 @@ enum { Opt_uid, Opt_gid, Opt_mode, - Opt_err }; -static const match_table_t tokens = { - {Opt_uid, "uid=%u"}, - {Opt_gid, "gid=%u"}, - {Opt_mode, "mode=%o"}, - {Opt_err, NULL} +static const struct fs_parameter_spec tracefs_param_specs[] = { + fsparam_u32 ("gid", Opt_gid), + fsparam_u32oct ("mode", Opt_mode), + fsparam_u32 ("uid", Opt_uid), + {} }; -struct tracefs_fs_info { - struct tracefs_mount_opts mount_opts; -}; - -static int tracefs_parse_options(char *data, struct tracefs_mount_opts *opts) +static int tracefs_parse_param(struct fs_context *fc, struct fs_parameter *param) { - substring_t args[MAX_OPT_ARGS]; - int option; - int token; + struct tracefs_fs_info *opts = fc->s_fs_info; + struct fs_parse_result result; kuid_t uid; kgid_t gid; - char *p; - - opts->opts = 0; - opts->mode = TRACEFS_DEFAULT_MODE; - - while ((p = strsep(&data, ",")) != NULL) { - if (!*p) - continue; - - token = match_token(p, tokens, args); - switch (token) { - case Opt_uid: - if (match_int(&args[0], &option)) - return -EINVAL; - uid = make_kuid(current_user_ns(), option); - if (!uid_valid(uid)) - return -EINVAL; - opts->uid = uid; - break; - case Opt_gid: - if (match_int(&args[0], &option)) - return -EINVAL; - gid = make_kgid(current_user_ns(), option); - if (!gid_valid(gid)) - return -EINVAL; - opts->gid = gid; - break; - case Opt_mode: - if (match_octal(&args[0], &option)) - return -EINVAL; - opts->mode = option & S_IALLUGO; - break; - /* - * We might like to report bad mount options here; - * but traditionally tracefs has ignored all mount options - */ - } - - opts->opts |= BIT(token); + int opt; + + opt = fs_parse(fc, tracefs_param_specs, param, &result); + if (opt < 0) + return opt; + + switch (opt) { + case Opt_uid: + uid = make_kuid(current_user_ns(), result.uint_32); + if (!uid_valid(uid)) + return invalf(fc, "Unknown uid"); + opts->uid = uid; + break; + case Opt_gid: + gid = make_kgid(current_user_ns(), result.uint_32); + if (!gid_valid(gid)) + return invalf(fc, "Unknown gid"); + opts->gid = gid; + break; + case Opt_mode: + opts->mode = result.uint_32 & S_IALLUGO; + break; + /* + * We might like to report bad mount options here; + * but traditionally tracefs has ignored all mount options + */ } + opts->opts |= BIT(opt); + return 0; } @@ -312,7 +295,6 @@ static int tracefs_apply_options(struct super_block *sb, bool remount) { struct tracefs_fs_info *fsi = sb->s_fs_info; struct inode *inode = d_inode(sb->s_root); - struct tracefs_mount_opts *opts = &fsi->mount_opts; umode_t tmp_mode; /* @@ -320,50 +302,46 @@ static int tracefs_apply_options(struct super_block *sb, bool remount) * options. */ - if (!remount || opts->opts & BIT(Opt_mode)) { + if (!remount || fsi->opts & BIT(Opt_mode)) { tmp_mode = READ_ONCE(inode->i_mode) & ~S_IALLUGO; - tmp_mode |= opts->mode; + tmp_mode |= fsi->mode; WRITE_ONCE(inode->i_mode, tmp_mode); } - if (!remount || opts->opts & BIT(Opt_uid)) - inode->i_uid = opts->uid; + if (!remount || fsi->opts & BIT(Opt_uid)) + inode->i_uid = fsi->uid; - if (!remount || opts->opts & BIT(Opt_gid)) - inode->i_gid = opts->gid; + if (!remount || fsi->opts & BIT(Opt_gid)) + inode->i_gid = fsi->gid; return 0; } -static int tracefs_remount(struct super_block *sb, int *flags, char *data) +static int tracefs_reconfigure(struct fs_context *fc) { - int err; - struct tracefs_fs_info *fsi = sb->s_fs_info; + struct super_block *sb = fc->root->d_sb; + struct tracefs_fs_info *sb_opts = sb->s_fs_info; + struct tracefs_fs_info *new_opts = fc->s_fs_info; sync_filesystem(sb); - err = tracefs_parse_options(data, &fsi->mount_opts); - if (err) - goto fail; + /* structure copy of new mount options to sb */ + *sb_opts = *new_opts; - tracefs_apply_options(sb, true); - -fail: - return err; + return tracefs_apply_options(sb, true); } static int tracefs_show_options(struct seq_file *m, struct dentry *root) { struct tracefs_fs_info *fsi = root->d_sb->s_fs_info; - struct tracefs_mount_opts *opts = &fsi->mount_opts; - if (!uid_eq(opts->uid, GLOBAL_ROOT_UID)) + if (!uid_eq(fsi->uid, GLOBAL_ROOT_UID)) seq_printf(m, ",uid=%u", - from_kuid_munged(&init_user_ns, opts->uid)); - if (!gid_eq(opts->gid, GLOBAL_ROOT_GID)) + from_kuid_munged(&init_user_ns, fsi->uid)); + if (!gid_eq(fsi->gid, GLOBAL_ROOT_GID)) seq_printf(m, ",gid=%u", - from_kgid_munged(&init_user_ns, opts->gid)); - if (opts->mode != TRACEFS_DEFAULT_MODE) - seq_printf(m, ",mode=%o", opts->mode); + from_kgid_munged(&init_user_ns, fsi->gid)); + if (fsi->mode != TRACEFS_DEFAULT_MODE) + seq_printf(m, ",mode=%o", fsi->mode); return 0; } @@ -373,7 +351,6 @@ static const struct super_operations tracefs_super_operations = { .free_inode = tracefs_free_inode, .drop_inode = generic_delete_inode, .statfs = simple_statfs, - .remount_fs = tracefs_remount, .show_options = tracefs_show_options, }; @@ -403,26 +380,14 @@ static const struct dentry_operations tracefs_dentry_operations = { .d_release = tracefs_d_release, }; -static int trace_fill_super(struct super_block *sb, void *data, int silent) +static int tracefs_fill_super(struct super_block *sb, struct fs_context *fc) { static const struct tree_descr trace_files[] = {{""}}; - struct tracefs_fs_info *fsi; int err; - fsi = kzalloc(sizeof(struct tracefs_fs_info), GFP_KERNEL); - sb->s_fs_info = fsi; - if (!fsi) { - err = -ENOMEM; - goto fail; - } - - err = tracefs_parse_options(data, &fsi->mount_opts); - if (err) - goto fail; - - err = simple_fill_super(sb, TRACEFS_MAGIC, trace_files); + err = simple_fill_super(sb, TRACEFS_MAGIC, trace_files); if (err) - goto fail; + return err; sb->s_op = &tracefs_super_operations; sb->s_d_op = &tracefs_dentry_operations; @@ -430,24 +395,45 @@ static int trace_fill_super(struct super_block *sb, void *data, int silent) tracefs_apply_options(sb, false); return 0; +} -fail: - kfree(fsi); - sb->s_fs_info = NULL; - return err; +static int tracefs_get_tree(struct fs_context *fc) +{ + return get_tree_single(fc, tracefs_fill_super); } -static struct dentry *trace_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, - void *data) +static void tracefs_free_fc(struct fs_context *fc) { - return mount_single(fs_type, flags, data, trace_fill_super); + kfree(fc->s_fs_info); +} + +static const struct fs_context_operations tracefs_context_ops = { + .free = tracefs_free_fc, + .parse_param = tracefs_parse_param, + .get_tree = tracefs_get_tree, + .reconfigure = tracefs_reconfigure, +}; + +static int tracefs_init_fs_context(struct fs_context *fc) +{ + struct tracefs_fs_info *fsi; + + fsi = kzalloc(sizeof(struct tracefs_fs_info), GFP_KERNEL); + if (!fsi) + return -ENOMEM; + + fsi->mode = TRACEFS_DEFAULT_MODE; + + fc->s_fs_info = fsi; + fc->ops = &tracefs_context_ops; + return 0; } static struct file_system_type trace_fs_type = { .owner = THIS_MODULE, .name = "tracefs", - .mount = trace_mount, + .init_fs_context = tracefs_init_fs_context, + .parameters = tracefs_param_specs, .kill_sb = kill_litter_super, }; MODULE_ALIAS_FS("tracefs"); -- cgit v1.2.3-59-g8ed1b From 7cd7bfe59328741185ef6db3356489c22919e59b Mon Sep 17 00:00:00 2001 From: Bill O'Donnell Date: Thu, 7 Mar 2024 10:29:18 -0600 Subject: minix: convert minix to use the new mount api Convert the minix filesystem to use the new mount API. Tested using mount and remount on minix device. Signed-off-by: Bill O'Donnell Link: https://lore.kernel.org/r/20240307163325.998723-1-bodonnel@redhat.com Acked-by: Eric Sandeen Signed-off-by: Christian Brauner --- fs/minix/inode.c | 48 ++++++++++++++++++++++++++++++------------------ 1 file changed, 30 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 7cbd2b9f4d11..7f9a2d8aa420 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -20,11 +20,11 @@ #include #include #include +#include static int minix_write_inode(struct inode *inode, struct writeback_control *wbc); static int minix_statfs(struct dentry *dentry, struct kstatfs *buf); -static int minix_remount (struct super_block * sb, int * flags, char * data); static void minix_evict_inode(struct inode *inode) { @@ -111,19 +111,19 @@ static const struct super_operations minix_sops = { .evict_inode = minix_evict_inode, .put_super = minix_put_super, .statfs = minix_statfs, - .remount_fs = minix_remount, }; -static int minix_remount (struct super_block * sb, int * flags, char * data) +static int minix_reconfigure(struct fs_context *fc) { - struct minix_sb_info * sbi = minix_sb(sb); struct minix_super_block * ms; + struct super_block *sb = fc->root->d_sb; + struct minix_sb_info * sbi = sb->s_fs_info; sync_filesystem(sb); ms = sbi->s_ms; - if ((bool)(*flags & SB_RDONLY) == sb_rdonly(sb)) + if ((bool)(fc->sb_flags & SB_RDONLY) == sb_rdonly(sb)) return 0; - if (*flags & SB_RDONLY) { + if (fc->sb_flags & SB_RDONLY) { if (ms->s_state & MINIX_VALID_FS || !(sbi->s_mount_state & MINIX_VALID_FS)) return 0; @@ -170,7 +170,7 @@ static bool minix_check_superblock(struct super_block *sb) return true; } -static int minix_fill_super(struct super_block *s, void *data, int silent) +static int minix_fill_super(struct super_block *s, struct fs_context *fc) { struct buffer_head *bh; struct buffer_head **map; @@ -180,6 +180,7 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) struct inode *root_inode; struct minix_sb_info *sbi; int ret = -EINVAL; + int silent = fc->sb_flags & SB_SILENT; sbi = kzalloc(sizeof(struct minix_sb_info), GFP_KERNEL); if (!sbi) @@ -371,6 +372,23 @@ out: return ret; } +static int minix_get_tree(struct fs_context *fc) +{ + return get_tree_bdev(fc, minix_fill_super); +} + +static const struct fs_context_operations minix_context_ops = { + .get_tree = minix_get_tree, + .reconfigure = minix_reconfigure, +}; + +static int minix_init_fs_context(struct fs_context *fc) +{ + fc->ops = &minix_context_ops; + + return 0; +} + static int minix_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; @@ -680,18 +698,12 @@ void minix_truncate(struct inode * inode) V2_minix_truncate(inode); } -static struct dentry *minix_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data) -{ - return mount_bdev(fs_type, flags, dev_name, data, minix_fill_super); -} - static struct file_system_type minix_fs_type = { - .owner = THIS_MODULE, - .name = "minix", - .mount = minix_mount, - .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .owner = THIS_MODULE, + .name = "minix", + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, + .init_fs_context = minix_init_fs_context, }; MODULE_ALIAS_FS("minix"); -- cgit v1.2.3-59-g8ed1b From 4d69c58ef2e419550f02f988ee022fca564f7dd7 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 25 Mar 2024 19:36:02 -0600 Subject: fsnotify: Avoid -Wflex-array-member-not-at-end warning Use the `DEFINE_FLEX()` helper for an on-stack definition of a flexible structure where the size of the flexible-array member is known at compile-time, and refactor the rest of the code, accordingly. So, with these changes, fix the following warning: fs/notify/fdinfo.c:45:36: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Link: https://github.com/KSPP/linux/issues/202 Signed-off-by: Gustavo A. R. Silva Signed-off-by: Jan Kara Message-Id: --- fs/notify/fdinfo.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/notify/fdinfo.c b/fs/notify/fdinfo.c index 5c430736ec12..dec553034027 100644 --- a/fs/notify/fdinfo.c +++ b/fs/notify/fdinfo.c @@ -41,29 +41,25 @@ static void show_fdinfo(struct seq_file *m, struct file *f, #if defined(CONFIG_EXPORTFS) static void show_mark_fhandle(struct seq_file *m, struct inode *inode) { - struct { - struct file_handle handle; - u8 pad[MAX_HANDLE_SZ]; - } f; + DEFINE_FLEX(struct file_handle, f, f_handle, handle_bytes, MAX_HANDLE_SZ); int size, ret, i; - f.handle.handle_bytes = sizeof(f.pad); - size = f.handle.handle_bytes >> 2; + size = f->handle_bytes >> 2; - ret = exportfs_encode_fid(inode, (struct fid *)f.handle.f_handle, &size); + ret = exportfs_encode_fid(inode, (struct fid *)f->f_handle, &size); if ((ret == FILEID_INVALID) || (ret < 0)) { WARN_ONCE(1, "Can't encode file handler for inotify: %d\n", ret); return; } - f.handle.handle_type = ret; - f.handle.handle_bytes = size * sizeof(u32); + f->handle_type = ret; + f->handle_bytes = size * sizeof(u32); seq_printf(m, "fhandle-bytes:%x fhandle-type:%x f_handle:", - f.handle.handle_bytes, f.handle.handle_type); + f->handle_bytes, f->handle_type); - for (i = 0; i < f.handle.handle_bytes; i++) - seq_printf(m, "%02x", (int)f.handle.f_handle[i]); + for (i = 0; i < f->handle_bytes; i++) + seq_printf(m, "%02x", (int)f->f_handle[i]); } #else static void show_mark_fhandle(struct seq_file *m, struct inode *inode) -- cgit v1.2.3-59-g8ed1b From 1ec17ef59168a1a6f1105f5dc517f783839a5302 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Wed, 28 Feb 2024 12:13:27 +0100 Subject: btrfs: zoned: fix use-after-free in do_zone_finish() Shinichiro reported the following use-after-free triggered by the device replace operation in fstests btrfs/070. BTRFS info (device nullb1): scrub: finished on devid 1 with status: 0 ================================================================== BUG: KASAN: slab-use-after-free in do_zone_finish+0x91a/0xb90 [btrfs] Read of size 8 at addr ffff8881543c8060 by task btrfs-cleaner/3494007 CPU: 0 PID: 3494007 Comm: btrfs-cleaner Tainted: G W 6.8.0-rc5-kts #1 Hardware name: Supermicro Super Server/X11SPi-TF, BIOS 3.3 02/21/2020 Call Trace: dump_stack_lvl+0x5b/0x90 print_report+0xcf/0x670 ? __virt_addr_valid+0x200/0x3e0 kasan_report+0xd8/0x110 ? do_zone_finish+0x91a/0xb90 [btrfs] ? do_zone_finish+0x91a/0xb90 [btrfs] do_zone_finish+0x91a/0xb90 [btrfs] btrfs_delete_unused_bgs+0x5e1/0x1750 [btrfs] ? __pfx_btrfs_delete_unused_bgs+0x10/0x10 [btrfs] ? btrfs_put_root+0x2d/0x220 [btrfs] ? btrfs_clean_one_deleted_snapshot+0x299/0x430 [btrfs] cleaner_kthread+0x21e/0x380 [btrfs] ? __pfx_cleaner_kthread+0x10/0x10 [btrfs] kthread+0x2e3/0x3c0 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x31/0x70 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1b/0x30 Allocated by task 3493983: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 __kasan_kmalloc+0xaa/0xb0 btrfs_alloc_device+0xb3/0x4e0 [btrfs] device_list_add.constprop.0+0x993/0x1630 [btrfs] btrfs_scan_one_device+0x219/0x3d0 [btrfs] btrfs_control_ioctl+0x26e/0x310 [btrfs] __x64_sys_ioctl+0x134/0x1b0 do_syscall_64+0x99/0x190 entry_SYSCALL_64_after_hwframe+0x6e/0x76 Freed by task 3494056: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 kasan_save_free_info+0x3f/0x60 poison_slab_object+0x102/0x170 __kasan_slab_free+0x32/0x70 kfree+0x11b/0x320 btrfs_rm_dev_replace_free_srcdev+0xca/0x280 [btrfs] btrfs_dev_replace_finishing+0xd7e/0x14f0 [btrfs] btrfs_dev_replace_by_ioctl+0x1286/0x25a0 [btrfs] btrfs_ioctl+0xb27/0x57d0 [btrfs] __x64_sys_ioctl+0x134/0x1b0 do_syscall_64+0x99/0x190 entry_SYSCALL_64_after_hwframe+0x6e/0x76 The buggy address belongs to the object at ffff8881543c8000 which belongs to the cache kmalloc-1k of size 1024 The buggy address is located 96 bytes inside of freed 1024-byte region [ffff8881543c8000, ffff8881543c8400) The buggy address belongs to the physical page: page:00000000fe2c1285 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x1543c8 head:00000000fe2c1285 order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0 flags: 0x17ffffc0000840(slab|head|node=0|zone=2|lastcpupid=0x1fffff) page_type: 0xffffffff() raw: 0017ffffc0000840 ffff888100042dc0 ffffea0019e8f200 dead000000000002 raw: 0000000000000000 0000000000100010 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff8881543c7f00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffff8881543c7f80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 >ffff8881543c8000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff8881543c8080: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff8881543c8100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb This UAF happens because we're accessing stale zone information of a already removed btrfs_device in do_zone_finish(). The sequence of events is as follows: btrfs_dev_replace_start btrfs_scrub_dev btrfs_dev_replace_finishing btrfs_dev_replace_update_device_in_mapping_tree <-- devices replaced btrfs_rm_dev_replace_free_srcdev btrfs_free_device <-- device freed cleaner_kthread btrfs_delete_unused_bgs btrfs_zone_finish do_zone_finish <-- refers the freed device The reason for this is that we're using a cached pointer to the chunk_map from the block group, but on device replace this cached pointer can contain stale device entries. The staleness comes from the fact, that btrfs_block_group::physical_map is not a pointer to a btrfs_chunk_map but a memory copy of it. Also take the fs_info::dev_replace::rwsem to prevent btrfs_dev_replace_update_device_in_mapping_tree() from changing the device underneath us again. Note: btrfs_dev_replace_update_device_in_mapping_tree() is holding fs_info::mapping_tree_lock, but as this is a spinning read/write lock we cannot take it as the call to blkdev_zone_mgmt() requires a memory allocation which may not sleep. But btrfs_dev_replace_update_device_in_mapping_tree() is always called with the fs_info::dev_replace::rwsem held in write mode. Many thanks to Shinichiro for analyzing the bug. Reported-by: Shinichiro Kawasaki CC: stable@vger.kernel.org # 6.8 Reviewed-by: Filipe Manana Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/zoned.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 3317bebfca95..459d1af02c3c 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -1561,11 +1561,7 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new) if (!map) return -EINVAL; - cache->physical_map = btrfs_clone_chunk_map(map, GFP_NOFS); - if (!cache->physical_map) { - ret = -ENOMEM; - goto out; - } + cache->physical_map = map; zone_info = kcalloc(map->num_stripes, sizeof(*zone_info), GFP_NOFS); if (!zone_info) { @@ -1677,7 +1673,6 @@ out: } bitmap_free(active); kfree(zone_info); - btrfs_free_chunk_map(map); return ret; } @@ -2162,6 +2157,7 @@ static int do_zone_finish(struct btrfs_block_group *block_group, bool fully_writ struct btrfs_chunk_map *map; const bool is_metadata = (block_group->flags & (BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_SYSTEM)); + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; int ret = 0; int i; @@ -2237,6 +2233,7 @@ static int do_zone_finish(struct btrfs_block_group *block_group, bool fully_writ btrfs_clear_data_reloc_bg(block_group); spin_unlock(&block_group->lock); + down_read(&dev_replace->rwsem); map = block_group->physical_map; for (i = 0; i < map->num_stripes; i++) { struct btrfs_device *device = map->stripes[i].dev; @@ -2251,13 +2248,16 @@ static int do_zone_finish(struct btrfs_block_group *block_group, bool fully_writ zinfo->zone_size >> SECTOR_SHIFT, GFP_NOFS); - if (ret) + if (ret) { + up_read(&dev_replace->rwsem); return ret; + } if (!(block_group->flags & BTRFS_BLOCK_GROUP_DATA)) zinfo->reserved_active_zones++; btrfs_dev_clear_active_zone(device, physical); } + up_read(&dev_replace->rwsem); if (!fully_written) btrfs_dec_block_group_ro(block_group); -- cgit v1.2.3-59-g8ed1b From 9f7eb8405dcbc79c5434821e9e3e92abe187ee8e Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Fri, 1 Mar 2024 08:42:13 +0800 Subject: btrfs: validate device maj:min during open Boris managed to create a device capable of changing its maj:min without altering its device path. Only multi-devices can be scanned. A device that gets scanned and remains in the btrfs kernel cache might end up with an incorrect maj:min. Despite the temp-fsid feature patch did not introduce this bug, it could lead to issues if the above multi-device is converted to a single device with a stale maj:min. Subsequently, attempting to mount the same device with the correct maj:min might mistake it for another device with the same fsid, potentially resulting in wrongly auto-enabling the temp-fsid feature. To address this, this patch validates the device's maj:min at the time of device open and updates it if it has changed since the last scan. CC: stable@vger.kernel.org # 6.7+ Fixes: a5b8a5f9f835 ("btrfs: support cloned-device mount capability") Reported-by: Boris Burkov Co-developed-by: Boris Burkov Reviewed-by: Boris Burkov # Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e49935a54da0..c318640b4472 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -692,6 +692,16 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices, device->bdev = bdev_handle->bdev; clear_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); + if (device->devt != device->bdev->bd_dev) { + btrfs_warn(NULL, + "device %s maj:min changed from %d:%d to %d:%d", + device->name->str, MAJOR(device->devt), + MINOR(device->devt), MAJOR(device->bdev->bd_dev), + MINOR(device->bdev->bd_dev)); + + device->devt = device->bdev->bd_dev; + } + fs_devices->open_devices++; if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) && device->devid != BTRFS_DEV_REPLACE_DEVID) { -- cgit v1.2.3-59-g8ed1b From 8a565ec04d6c43f330e7401e5af3458431b29bc6 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 13 Mar 2024 11:37:31 +0000 Subject: btrfs: fix extent map leak in unexpected scenario at unpin_extent_cache() At unpin_extent_cache() if we happen to find an extent map with an unexpected start offset, we jump to the 'out' label and never release the reference we added to the extent map through the call to lookup_extent_mapping(), therefore resulting in a leak. So fix this by moving the free_extent_map() under the 'out' label. Fixes: c03c89f821e5 ("btrfs: handle errors returned from unpin_extent_cache()") Reviewed-by: Qu Wenruo Reviewed-by: Anand Jain Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 347ca13d15a9..e03953dbcd5e 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -340,9 +340,9 @@ int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen) em->mod_len = em->len; } - free_extent_map(em); out: write_unlock(&tree->lock); + free_extent_map(em); return ret; } -- cgit v1.2.3-59-g8ed1b From 4dc1d69c2b101eee0bf071187794ffed2f9c2596 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 13 Mar 2024 12:49:31 +0000 Subject: btrfs: fix warning messages not printing interval at unpin_extent_range() At unpin_extent_range() we print warning messages that are supposed to print an interval in the form "[X, Y)", with the first element being an inclusive start offset and the second element being the exclusive end offset of a range. However we end up printing the range's length instead of the range's exclusive end offset, so fix that to avoid having confusing and non-sense messages in case we hit one of these unexpected scenarios. Fixes: 00deaf04df35 ("btrfs: log messages at unpin_extent_range() during unexpected cases") Reviewed-by: Qu Wenruo Reviewed-by: Anand Jain Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index e03953dbcd5e..2cfc6e8cf76f 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -309,7 +309,7 @@ int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen) btrfs_warn(fs_info, "no extent map found for inode %llu (root %lld) when unpinning extent range [%llu, %llu), generation %llu", btrfs_ino(inode), btrfs_root_id(inode->root), - start, len, gen); + start, start + len, gen); ret = -ENOENT; goto out; } @@ -318,7 +318,7 @@ int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen) btrfs_warn(fs_info, "found extent map for inode %llu (root %lld) with unexpected start offset %llu when unpinning extent range [%llu, %llu), generation %llu", btrfs_ino(inode), btrfs_root_id(inode->root), - em->start, start, len, gen); + em->start, start, start + len, gen); ret = -EUCLEAN; goto out; } -- cgit v1.2.3-59-g8ed1b From 379c87239320a204138995e1da35ce9eca239e7a Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 13 Mar 2024 13:02:02 +0000 Subject: btrfs: fix message not properly printing interval when adding extent map At btrfs_add_extent_mapping(), if we are unable to merge the existing extent map, we print a warning message that suggests interval ranges in the form "[X, Y)", where the first element is the inclusive start offset of a range and the second element is the exclusive end offset. However we end up printing the length of the ranges instead of the exclusive end offsets. So fix this by printing the range end offsets. Reviewed-by: Qu Wenruo Reviewed-by: Anand Jain Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 2cfc6e8cf76f..16685cb8a91d 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -634,8 +634,8 @@ int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info, *em_in = NULL; WARN_ONCE(ret, "extent map merge error existing [%llu, %llu) with em [%llu, %llu) start %llu\n", - existing->start, existing->len, - orig_start, orig_len, start); + existing->start, extent_map_end(existing), + orig_start, orig_start + orig_len, start); } free_extent_map(existing); } -- cgit v1.2.3-59-g8ed1b From 2133460061e1bbecb47da73ad5ec7cf8e951006c Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 13 Mar 2024 17:14:02 +0000 Subject: btrfs: use btrfs_warn() to log message at btrfs_add_extent_mapping() At btrfs_add_extent_mapping(), if we failed to merge the extent map, which is unexpected and theoretically should never happen, we use WARN_ONCE() to log a message which is not great because we don't get information about which filesystem it relates to in case we have multiple btrfs filesystems mounted. So change this to use btrfs_warn() and surround the error check with WARN_ON() so we always get a useful stack trace and the condition is flagged as "unlikely" since it's not expected to ever happen. Reviewed-by: Qu Wenruo Reviewed-by: Anand Jain Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 16685cb8a91d..445f7716f1e2 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -629,13 +629,13 @@ int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info, */ ret = merge_extent_mapping(em_tree, existing, em, start); - if (ret) { + if (WARN_ON(ret)) { free_extent_map(em); *em_in = NULL; - WARN_ONCE(ret, -"extent map merge error existing [%llu, %llu) with em [%llu, %llu) start %llu\n", - existing->start, extent_map_end(existing), - orig_start, orig_start + orig_len, start); + btrfs_warn(fs_info, +"extent map merge error existing [%llu, %llu) with em [%llu, %llu) start %llu", + existing->start, extent_map_end(existing), + orig_start, orig_start + orig_len, start); } free_extent_map(existing); } -- cgit v1.2.3-59-g8ed1b From a8b70c7f8600bc77d03c0b032c0662259b9e615e Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Wed, 21 Feb 2024 07:35:52 -0800 Subject: btrfs: zoned: don't skip block groups with 100% zone unusable Commit f4a9f219411f ("btrfs: do not delete unused block group if it may be used soon") changed the behaviour of deleting unused block-groups on zoned filesystems. Starting with this commit, we're using btrfs_space_info_used() to calculate the number of used bytes in a space_info. But btrfs_space_info_used() also accounts btrfs_space_info::bytes_zone_unusable as used bytes. So if a block group is 100% zone_unusable it is skipped from the deletion step. In order not to skip fully zone_unusable block-groups, also check if the block-group has bytes left that can be used on a zoned filesystem. Fixes: f4a9f219411f ("btrfs: do not delete unused block group if it may be used soon") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Filipe Manana Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/block-group.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 5f7587ca1ca7..1e09aeea69c2 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -1559,7 +1559,8 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) * needing to allocate extents from the block group. */ used = btrfs_space_info_used(space_info, true); - if (space_info->total_bytes - block_group->length < used) { + if (space_info->total_bytes - block_group->length < used && + block_group->zone_unusable < block_group->length) { /* * Add a reference for the list, compensate for the ref * drop under the "next" label for the -- cgit v1.2.3-59-g8ed1b From 2f1aeab9fca1a5f583be1add175d1ee95c213cfa Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 08:28:18 +0530 Subject: btrfs: return accurate error code on open failure in open_fs_devices() When attempting to exclusive open a device which has no exclusive open permission, such as a physical device associated with the flakey dm device, the open operation will fail, resulting in a mount failure. In this particular scenario, we erroneously return -EINVAL instead of the correct error code provided by the bdev_open_by_path() function, which is -EBUSY. Fix this, by returning error code from the bdev_open_by_path() function. With this correction, the mount error message will align with that of ext4 and xfs. Reviewed-by: Boris Burkov Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index c318640b4472..dedec3d9b111 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1184,23 +1184,30 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices, struct btrfs_device *device; struct btrfs_device *latest_dev = NULL; struct btrfs_device *tmp_device; + int ret = 0; list_for_each_entry_safe(device, tmp_device, &fs_devices->devices, dev_list) { - int ret; + int ret2; - ret = btrfs_open_one_device(fs_devices, device, flags, holder); - if (ret == 0 && + ret2 = btrfs_open_one_device(fs_devices, device, flags, holder); + if (ret2 == 0 && (!latest_dev || device->generation > latest_dev->generation)) { latest_dev = device; - } else if (ret == -ENODATA) { + } else if (ret2 == -ENODATA) { fs_devices->num_devices--; list_del(&device->dev_list); btrfs_free_device(device); } + if (ret == 0 && ret2 != 0) + ret = ret2; } - if (fs_devices->open_devices == 0) + + if (fs_devices->open_devices == 0) { + if (ret) + return ret; return -EINVAL; + } fs_devices->opened = 1; fs_devices->latest_dev = latest_dev; -- cgit v1.2.3-59-g8ed1b From ef1e68236b9153c27cb7cf29ead0c532870d4215 Mon Sep 17 00:00:00 2001 From: Tavian Barnes Date: Fri, 15 Mar 2024 21:14:29 -0400 Subject: btrfs: fix race in read_extent_buffer_pages() There are reports from tree-checker that detects corrupted nodes, without any obvious pattern so possibly an overwrite in memory. After some debugging it turns out there's a race when reading an extent buffer the uptodate status can be missed. To prevent concurrent reads for the same extent buffer, read_extent_buffer_pages() performs these checks: /* (1) */ if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) return 0; /* (2) */ if (test_and_set_bit(EXTENT_BUFFER_READING, &eb->bflags)) goto done; At this point, it seems safe to start the actual read operation. Once that completes, end_bbio_meta_read() does /* (3) */ set_extent_buffer_uptodate(eb); /* (4) */ clear_bit(EXTENT_BUFFER_READING, &eb->bflags); Normally, this is enough to ensure only one read happens, and all other callers wait for it to finish before returning. Unfortunately, there is a racey interleaving: Thread A | Thread B | Thread C ---------+----------+--------- (1) | | | (1) | (2) | | (3) | | (4) | | | (2) | | | (1) When this happens, thread B kicks of an unnecessary read. Worse, thread C will see UPTODATE set and return immediately, while the read from thread B is still in progress. This race could result in tree-checker errors like this as the extent buffer is concurrently modified: BTRFS critical (device dm-0): corrupted node, root=256 block=8550954455682405139 owner mismatch, have 11858205567642294356 expect [256, 18446744073709551360] Fix it by testing UPTODATE again after setting the READING bit, and if it's been set, skip the unnecessary read. Fixes: d7172f52e993 ("btrfs: use per-buffer locking for extent_buffer reading") Link: https://lore.kernel.org/linux-btrfs/CAHk-=whNdMaN9ntZ47XRKP6DBes2E5w7fi-0U3H2+PS18p+Pzw@mail.gmail.com/ Link: https://lore.kernel.org/linux-btrfs/f51a6d5d7432455a6a858d51b49ecac183e0bbc9.1706312914.git.wqu@suse.com/ Link: https://lore.kernel.org/linux-btrfs/c7241ea4-fcc6-48d2-98c8-b5ea790d6c89@gmx.com/ CC: stable@vger.kernel.org # 6.5+ Reviewed-by: Qu Wenruo Reviewed-by: Christoph Hellwig Signed-off-by: Tavian Barnes Reviewed-by: David Sterba [ minor update of changelog ] Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 7441245b1ceb..61594eaf1f89 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4333,6 +4333,19 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num, if (test_and_set_bit(EXTENT_BUFFER_READING, &eb->bflags)) goto done; + /* + * Between the initial test_bit(EXTENT_BUFFER_UPTODATE) and the above + * test_and_set_bit(EXTENT_BUFFER_READING), someone else could have + * started and finished reading the same eb. In this case, UPTODATE + * will now be set, and we shouldn't read it in again. + */ + if (unlikely(test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags))) { + clear_bit(EXTENT_BUFFER_READING, &eb->bflags); + smp_mb__after_atomic(); + wake_up_bit(&eb->bflags, EXTENT_BUFFER_READING); + return 0; + } + clear_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags); eb->read_mirror = 0; check_buffer_tree_ref(eb); -- cgit v1.2.3-59-g8ed1b From 3127f1010c9b27d925e83081d413ea7fc361abb0 Mon Sep 17 00:00:00 2001 From: Yeongjin Gil Date: Fri, 22 Mar 2024 13:16:39 +0900 Subject: f2fs: Prevent s_writer rw_sem count mismatch in f2fs_evict_inode If f2fs_evict_inode is called between freeze_super and thaw_super, the s_writer rwsem count may become negative, resulting in hang. CPU1 CPU2 f2fs_resize_fs() f2fs_evict_inode() f2fs_freeze set SBI_IS_FREEZING skip sb_start_intwrite f2fs_unfreeze clear SBI_IS_FREEZING sb_end_intwrite To solve this problem, the call to sb_end_write is determined by whether sb_start_intwrite is called, rather than the current freezing status. Reviewed-by: Sungjong Seo Reviewed-by: Sunmin Jeong Signed-off-by: Yeongjin Gil Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/inode.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index c26effdce9aa..12b1fef31f43 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -804,6 +804,7 @@ void f2fs_evict_inode(struct inode *inode) struct f2fs_inode_info *fi = F2FS_I(inode); nid_t xnid = fi->i_xattr_nid; int err = 0; + bool freeze_protected = false; f2fs_abort_atomic_write(inode, true); @@ -843,8 +844,10 @@ void f2fs_evict_inode(struct inode *inode) f2fs_remove_ino_entry(sbi, inode->i_ino, UPDATE_INO); f2fs_remove_ino_entry(sbi, inode->i_ino, FLUSH_INO); - if (!is_sbi_flag_set(sbi, SBI_IS_FREEZING)) + if (!is_sbi_flag_set(sbi, SBI_IS_FREEZING)) { sb_start_intwrite(inode->i_sb); + freeze_protected = true; + } set_inode_flag(inode, FI_NO_ALLOC); i_size_write(inode, 0); retry: @@ -887,7 +890,7 @@ retry: if (dquot_initialize_needed(inode)) set_sbi_flag(sbi, SBI_QUOTA_NEED_REPAIR); } - if (!is_sbi_flag_set(sbi, SBI_IS_FREEZING)) + if (freeze_protected) sb_end_intwrite(inode->i_sb); no_delete: dquot_drop(inode); -- cgit v1.2.3-59-g8ed1b From 92c556ed6318e13c16746495a8d4513129eb9b0f Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 22 Mar 2024 22:59:55 +0800 Subject: f2fs: fix to detect inconsistent nat entry during truncation As Roman Smirnov reported as below: " There is a possible bug in f2fs_truncate_inode_blocks(): if (err < 0 && err != -ENOENT) goto fail; ... offset[1] = 0; offset[0]++; nofs += err; If err = -ENOENT then nofs will sum with an error code, which is strange behaviour. Also if nofs < ENOENT this will cause an overflow. err will be equal to -ENOENT with the following call stack: truncate_nodes() f2fs_get_node_page() __get_node_page() read_node_page() " If nat is corrupted, truncate_nodes() may return -ENOENT, and f2fs_truncate_inode_blocks() doesn't handle such error correctly, fix it. Reported-by: Roman Smirnov Closes: https://lore.kernel.org/linux-f2fs-devel/085b27fd2b364a3c8c3a9ca77363e246@omp.ru Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/node.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index b3de6d6cdb02..bb57bbaff7b4 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -1187,7 +1187,17 @@ skip_partial: default: BUG(); } - if (err < 0 && err != -ENOENT) + if (err == -ENOENT) { + set_sbi_flag(F2FS_P_SB(page), SBI_NEED_FSCK); + f2fs_handle_error(sbi, ERROR_INVALID_BLKADDR); + f2fs_err_ratelimited(sbi, + "truncate node fail, ino:%lu, nid:%u, " + "offset[0]:%d, offset[1]:%d, nofs:%d", + inode->i_ino, dn.nid, offset[0], + offset[1], nofs); + err = 0; + } + if (err < 0) goto fail; if (offset[1] == 0 && ri->i_nid[offset[0] - NODE_DIR1_BLOCK]) { -- cgit v1.2.3-59-g8ed1b From ac7805be6cffd5a16f86872c4e56c28a1433b222 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 25 Mar 2024 23:27:25 +0800 Subject: f2fs: introduce map_is_mergeable() for cleanup No logic changes. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index d9494b5fc7c1..08815394223a 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1507,6 +1507,22 @@ static bool f2fs_map_blocks_cached(struct inode *inode, return true; } +static bool map_is_mergeable(struct f2fs_sb_info *sbi, + struct f2fs_map_blocks *map, + block_t blkaddr, int flag, int bidx, + int ofs) +{ + if (map->m_multidev_dio && map->m_bdev != FDEV(bidx).bdev) + return false; + if (map->m_pblk != NEW_ADDR && blkaddr == (map->m_pblk + ofs)) + return true; + if (map->m_pblk == NEW_ADDR && blkaddr == NEW_ADDR) + return true; + if (flag == F2FS_GET_BLOCK_PRE_DIO) + return true; + return false; +} + /* * f2fs_map_blocks() tries to find or build mapping relationship which * maps continuous logical blocks to physical blocks, and return such @@ -1653,12 +1669,7 @@ next_block: if (map->m_multidev_dio) map->m_bdev = FDEV(bidx).bdev; - } else if ((map->m_pblk != NEW_ADDR && - blkaddr == (map->m_pblk + ofs)) || - (map->m_pblk == NEW_ADDR && blkaddr == NEW_ADDR) || - flag == F2FS_GET_BLOCK_PRE_DIO) { - if (map->m_multidev_dio && map->m_bdev != FDEV(bidx).bdev) - goto sync_out; + } else if (map_is_mergeable(sbi, map, blkaddr, flag, bidx, ofs)) { ofs++; map->m_len++; } else { -- cgit v1.2.3-59-g8ed1b From e9e9fbeb83f65d3d487e0a0838c0867292c99fb2 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 24 Mar 2024 00:01:02 -0500 Subject: smb3: add trace event for mknod Add trace points to help debug mknod and mkfifo: smb3_mknod_done smb3_mknod_enter smb3_mknod_err Example output: TASK-PID CPU# ||||| TIMESTAMP FUNCTION | | | ||||| | | mkfifo-6163 [003] ..... 960.425558: smb3_mknod_enter: xid=12 sid=0xb55130f6 tid=0x46e6241c path=\fifo1 mkfifo-6163 [003] ..... 960.432719: smb3_mknod_done: xid=12 sid=0xb55130f6 tid=0x46e6241c Reviewed-by: Bharath SM Reviewed-by: Meetakshi Setiya Signed-off-by: Steve French --- fs/smb/client/dir.c | 7 +++++++ fs/smb/client/trace.h | 4 +++- 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/dir.c b/fs/smb/client/dir.c index 89333d9bce36..d11dc3aa458b 100644 --- a/fs/smb/client/dir.c +++ b/fs/smb/client/dir.c @@ -612,11 +612,18 @@ int cifs_mknod(struct mnt_idmap *idmap, struct inode *inode, goto mknod_out; } + trace_smb3_mknod_enter(xid, tcon->ses->Suid, tcon->tid, full_path); + rc = tcon->ses->server->ops->make_node(xid, inode, direntry, tcon, full_path, mode, device_number); mknod_out: + if (rc) + trace_smb3_mknod_err(xid, tcon->ses->Suid, tcon->tid, rc); + else + trace_smb3_mknod_done(xid, tcon->ses->Suid, tcon->tid); + free_dentry_path(page); free_xid(xid); cifs_put_tlink(tlink); diff --git a/fs/smb/client/trace.h b/fs/smb/client/trace.h index f9c1fd32d0b8..5e83cb9da902 100644 --- a/fs/smb/client/trace.h +++ b/fs/smb/client/trace.h @@ -375,6 +375,7 @@ DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(get_reparse_compound_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(delete_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(mkdir_enter); DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(tdis_enter); +DEFINE_SMB3_INF_COMPOUND_ENTER_EVENT(mknod_enter); DECLARE_EVENT_CLASS(smb3_inf_compound_done_class, TP_PROTO(unsigned int xid, @@ -415,7 +416,7 @@ DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(query_wsl_ea_compound_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(delete_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(mkdir_done); DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(tdis_done); - +DEFINE_SMB3_INF_COMPOUND_DONE_EVENT(mknod_done); DECLARE_EVENT_CLASS(smb3_inf_compound_err_class, TP_PROTO(unsigned int xid, @@ -461,6 +462,7 @@ DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(query_wsl_ea_compound_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(mkdir_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(delete_err); DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(tdis_err); +DEFINE_SMB3_INF_COMPOUND_ERR_EVENT(mknod_err); /* * For logging SMB3 Status code and Command for responses which return errors -- cgit v1.2.3-59-g8ed1b From 22650a99821dda3d05f1c334ea90330b4982de56 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Tue, 26 Mar 2024 13:47:22 +0100 Subject: fs,block: yield devices early Currently a device is only really released once the umount returns to userspace due to how file closing works. That ultimately could cause an old umount assumption to be violated that concurrent umount and mount don't fail. So an exclusively held device with a temporary holder should be yielded before the filesystem is gone. Add a helper that allows callers to do that. This also allows us to remove the two holder ops that Linus wasn't excited about. Link: https://lore.kernel.org/r/20240326-vfs-bdev-end_holder-v1-1-20af85202918@kernel.org Fixes: f3a608827d1f ("bdev: open block device as files") # mainline only Reviewed-by: Christoph Hellwig Reviewed-by: Jan Kara Suggested-by: Linus Torvalds Signed-off-by: Christian Brauner --- block/bdev.c | 64 +++++++++++++++++++++++++++++++++++------ drivers/mtd/devices/block2mtd.c | 2 +- fs/bcachefs/super-io.c | 2 +- fs/cramfs/inode.c | 2 +- fs/ext4/super.c | 8 +++--- fs/f2fs/super.c | 2 +- fs/jfs/jfs_logmgr.c | 4 +-- fs/reiserfs/journal.c | 2 +- fs/romfs/super.c | 2 +- fs/super.c | 24 ++-------------- fs/xfs/xfs_buf.c | 2 +- fs/xfs/xfs_super.c | 6 ++-- include/linux/blkdev.h | 11 +------ 13 files changed, 76 insertions(+), 55 deletions(-) (limited to 'fs') diff --git a/block/bdev.c b/block/bdev.c index a1946a902df3..b8e32d933a63 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -583,9 +583,6 @@ static void bd_finish_claiming(struct block_device *bdev, void *holder, mutex_unlock(&bdev->bd_holder_lock); bd_clear_claiming(whole, holder); mutex_unlock(&bdev_lock); - - if (hops && hops->get_holder) - hops->get_holder(holder); } /** @@ -608,7 +605,6 @@ EXPORT_SYMBOL(bd_abort_claiming); static void bd_end_claim(struct block_device *bdev, void *holder) { struct block_device *whole = bdev_whole(bdev); - const struct blk_holder_ops *hops = bdev->bd_holder_ops; bool unblock = false; /* @@ -631,9 +627,6 @@ static void bd_end_claim(struct block_device *bdev, void *holder) whole->bd_holder = NULL; mutex_unlock(&bdev_lock); - if (hops && hops->put_holder) - hops->put_holder(holder); - /* * If this was the last claim, remove holder link and unblock evpoll if * it was a write holder. @@ -813,6 +806,11 @@ static void bdev_claim_write_access(struct block_device *bdev, blk_mode_t mode) bdev->bd_writers++; } +static inline bool bdev_unclaimed(const struct file *bdev_file) +{ + return bdev_file->private_data == BDEV_I(bdev_file->f_mapping->host); +} + static void bdev_yield_write_access(struct file *bdev_file) { struct block_device *bdev; @@ -820,6 +818,9 @@ static void bdev_yield_write_access(struct file *bdev_file) if (bdev_allow_write_mounted) return; + if (bdev_unclaimed(bdev_file)) + return; + bdev = file_bdev(bdev_file); if (bdev_file->f_mode & FMODE_WRITE_RESTRICTED) @@ -1012,6 +1013,20 @@ struct file *bdev_file_open_by_path(const char *path, blk_mode_t mode, } EXPORT_SYMBOL(bdev_file_open_by_path); +static inline void bd_yield_claim(struct file *bdev_file) +{ + struct block_device *bdev = file_bdev(bdev_file); + void *holder = bdev_file->private_data; + + lockdep_assert_held(&bdev->bd_disk->open_mutex); + + if (WARN_ON_ONCE(IS_ERR_OR_NULL(holder))) + return; + + if (!bdev_unclaimed(bdev_file)) + bd_end_claim(bdev, holder); +} + void bdev_release(struct file *bdev_file) { struct block_device *bdev = file_bdev(bdev_file); @@ -1036,7 +1051,7 @@ void bdev_release(struct file *bdev_file) bdev_yield_write_access(bdev_file); if (holder) - bd_end_claim(bdev, holder); + bd_yield_claim(bdev_file); /* * Trigger event checking and tell drivers to flush MEDIA_CHANGE @@ -1056,6 +1071,39 @@ put_no_open: blkdev_put_no_open(bdev); } +/** + * bdev_fput - yield claim to the block device and put the file + * @bdev_file: open block device + * + * Yield claim on the block device and put the file. Ensure that the + * block device can be reclaimed before the file is closed which is a + * deferred operation. + */ +void bdev_fput(struct file *bdev_file) +{ + if (WARN_ON_ONCE(bdev_file->f_op != &def_blk_fops)) + return; + + if (bdev_file->private_data) { + struct block_device *bdev = file_bdev(bdev_file); + struct gendisk *disk = bdev->bd_disk; + + mutex_lock(&disk->open_mutex); + bdev_yield_write_access(bdev_file); + bd_yield_claim(bdev_file); + /* + * Tell release we already gave up our hold on the + * device and if write restrictions are available that + * we already gave up write access to the device. + */ + bdev_file->private_data = BDEV_I(bdev_file->f_mapping->host); + mutex_unlock(&disk->open_mutex); + } + + fput(bdev_file); +} +EXPORT_SYMBOL(bdev_fput); + /** * lookup_bdev() - Look up a struct block_device by name. * @pathname: Name of the block device in the filesystem. diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c index 97a00ec9a4d4..caacdc0a3819 100644 --- a/drivers/mtd/devices/block2mtd.c +++ b/drivers/mtd/devices/block2mtd.c @@ -209,7 +209,7 @@ static void block2mtd_free_device(struct block2mtd_dev *dev) if (dev->bdev_file) { invalidate_mapping_pages(dev->bdev_file->f_mapping, 0, -1); - fput(dev->bdev_file); + bdev_fput(dev->bdev_file); } kfree(dev); diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index ad28e370b640..cb7b4de11a49 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -143,7 +143,7 @@ void bch2_free_super(struct bch_sb_handle *sb) { kfree(sb->bio); if (!IS_ERR_OR_NULL(sb->s_bdev_file)) - fput(sb->s_bdev_file); + bdev_fput(sb->s_bdev_file); kfree(sb->holder); kfree(sb->sb_name); diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 39e75131fd5a..9901057a15ba 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -495,7 +495,7 @@ static void cramfs_kill_sb(struct super_block *sb) sb->s_mtd = NULL; } else if (IS_ENABLED(CONFIG_CRAMFS_BLOCKDEV) && sb->s_bdev) { sync_blockdev(sb->s_bdev); - fput(sb->s_bdev_file); + bdev_fput(sb->s_bdev_file); } kfree(sbi); } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index cfb8449c731f..044135796f2b 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -5668,7 +5668,7 @@ failed_mount: brelse(sbi->s_sbh); if (sbi->s_journal_bdev_file) { invalidate_bdev(file_bdev(sbi->s_journal_bdev_file)); - fput(sbi->s_journal_bdev_file); + bdev_fput(sbi->s_journal_bdev_file); } out_fail: invalidate_bdev(sb->s_bdev); @@ -5913,7 +5913,7 @@ static struct file *ext4_get_journal_blkdev(struct super_block *sb, out_bh: brelse(bh); out_bdev: - fput(bdev_file); + bdev_fput(bdev_file); return ERR_PTR(errno); } @@ -5952,7 +5952,7 @@ static journal_t *ext4_open_dev_journal(struct super_block *sb, out_journal: jbd2_journal_destroy(journal); out_bdev: - fput(bdev_file); + bdev_fput(bdev_file); return ERR_PTR(errno); } @@ -7327,7 +7327,7 @@ static void ext4_kill_sb(struct super_block *sb) kill_block_super(sb); if (bdev_file) - fput(bdev_file); + bdev_fput(bdev_file); } static struct file_system_type ext4_fs_type = { diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index a6867f26f141..a4bc26dfdb1a 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1558,7 +1558,7 @@ static void destroy_device_list(struct f2fs_sb_info *sbi) for (i = 0; i < sbi->s_ndevs; i++) { if (i > 0) - fput(FDEV(i).bdev_file); + bdev_fput(FDEV(i).bdev_file); #ifdef CONFIG_BLK_DEV_ZONED kvfree(FDEV(i).blkz_seq); #endif diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index 73389c68e251..9609349e92e5 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -1141,7 +1141,7 @@ journal_found: lbmLogShutdown(log); close: /* close external log device */ - fput(bdev_file); + bdev_fput(bdev_file); free: /* free log descriptor */ mutex_unlock(&jfs_log_mutex); @@ -1485,7 +1485,7 @@ int lmLogClose(struct super_block *sb) bdev_file = log->bdev_file; rc = lmLogShutdown(log); - fput(bdev_file); + bdev_fput(bdev_file); kfree(log); diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 6474529c4253..e539ccd39e1e 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2589,7 +2589,7 @@ static void journal_list_init(struct super_block *sb) static void release_journal_dev(struct reiserfs_journal *journal) { if (journal->j_bdev_file) { - fput(journal->j_bdev_file); + bdev_fput(journal->j_bdev_file); journal->j_bdev_file = NULL; } } diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 2be227532f39..2cbb92462074 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -594,7 +594,7 @@ static void romfs_kill_sb(struct super_block *sb) #ifdef CONFIG_ROMFS_ON_BLOCK if (sb->s_bdev) { sync_blockdev(sb->s_bdev); - fput(sb->s_bdev_file); + bdev_fput(sb->s_bdev_file); } #endif } diff --git a/fs/super.c b/fs/super.c index 71d9779c42b1..69ce6c600968 100644 --- a/fs/super.c +++ b/fs/super.c @@ -1515,29 +1515,11 @@ static int fs_bdev_thaw(struct block_device *bdev) return error; } -static void fs_bdev_super_get(void *data) -{ - struct super_block *sb = data; - - spin_lock(&sb_lock); - sb->s_count++; - spin_unlock(&sb_lock); -} - -static void fs_bdev_super_put(void *data) -{ - struct super_block *sb = data; - - put_super(sb); -} - const struct blk_holder_ops fs_holder_ops = { .mark_dead = fs_bdev_mark_dead, .sync = fs_bdev_sync, .freeze = fs_bdev_freeze, .thaw = fs_bdev_thaw, - .get_holder = fs_bdev_super_get, - .put_holder = fs_bdev_super_put, }; EXPORT_SYMBOL_GPL(fs_holder_ops); @@ -1562,7 +1544,7 @@ int setup_bdev_super(struct super_block *sb, int sb_flags, * writable from userspace even for a read-only block device. */ if ((mode & BLK_OPEN_WRITE) && bdev_read_only(bdev)) { - fput(bdev_file); + bdev_fput(bdev_file); return -EACCES; } @@ -1573,7 +1555,7 @@ int setup_bdev_super(struct super_block *sb, int sb_flags, if (atomic_read(&bdev->bd_fsfreeze_count) > 0) { if (fc) warnf(fc, "%pg: Can't mount, blockdev is frozen", bdev); - fput(bdev_file); + bdev_fput(bdev_file); return -EBUSY; } spin_lock(&sb_lock); @@ -1693,7 +1675,7 @@ void kill_block_super(struct super_block *sb) generic_shutdown_super(sb); if (bdev) { sync_blockdev(bdev); - fput(sb->s_bdev_file); + bdev_fput(sb->s_bdev_file); } } diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 1a18c381127e..f0fa02264eda 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -2030,7 +2030,7 @@ xfs_free_buftarg( fs_put_dax(btp->bt_daxdev, btp->bt_mount); /* the main block device is closed by kill_block_super */ if (btp->bt_bdev != btp->bt_mount->m_super->s_bdev) - fput(btp->bt_bdev_file); + bdev_fput(btp->bt_bdev_file); kfree(btp); } diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index c21f10ab0f5d..bce020374c5e 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -485,7 +485,7 @@ xfs_open_devices( mp->m_logdev_targp = mp->m_ddev_targp; /* Handle won't be used, drop it */ if (logdev_file) - fput(logdev_file); + bdev_fput(logdev_file); } return 0; @@ -497,10 +497,10 @@ xfs_open_devices( xfs_free_buftarg(mp->m_ddev_targp); out_close_rtdev: if (rtdev_file) - fput(rtdev_file); + bdev_fput(rtdev_file); out_close_logdev: if (logdev_file) - fput(logdev_file); + bdev_fput(logdev_file); return error; } diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index c3e8f7cf96be..172c91879999 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1505,16 +1505,6 @@ struct blk_holder_ops { * Thaw the file system mounted on the block device. */ int (*thaw)(struct block_device *bdev); - - /* - * If needed, get a reference to the holder. - */ - void (*get_holder)(void *holder); - - /* - * Release the holder. - */ - void (*put_holder)(void *holder); }; /* @@ -1585,6 +1575,7 @@ static inline int early_lookup_bdev(const char *pathname, dev_t *dev) int bdev_freeze(struct block_device *bdev); int bdev_thaw(struct block_device *bdev); +void bdev_fput(struct file *bdev_file); struct io_comp_batch { struct request *req_list; -- cgit v1.2.3-59-g8ed1b From 8876a37277cb832e1861c35f8c661825179f73f5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Mar 2024 14:13:24 +0000 Subject: cifs: Fix duplicate fscache cookie warnings fscache emits a lot of duplicate cookie warnings with cifs because the index key for the fscache cookies does not include everything that the cifs_find_inode() function does. The latter is used with iget5_locked() to distinguish between inodes in the local inode cache. Fix this by adding the creation time and file type to the fscache cookie key. Additionally, add a couple of comments to note that if one is changed the other must be also. Signed-off-by: David Howells Fixes: 70431bfd825d ("cifs: Support fscache indexing rewrite") cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org Signed-off-by: Steve French --- fs/smb/client/fscache.c | 16 +++++++++++++++- fs/smb/client/inode.c | 2 ++ 2 files changed, 17 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/fscache.c b/fs/smb/client/fscache.c index c4a3cb736881..340efce8f052 100644 --- a/fs/smb/client/fscache.c +++ b/fs/smb/client/fscache.c @@ -12,6 +12,16 @@ #include "cifs_fs_sb.h" #include "cifsproto.h" +/* + * Key for fscache inode. [!] Contents must match comparisons in cifs_find_inode(). + */ +struct cifs_fscache_inode_key { + + __le64 uniqueid; /* server inode number */ + __le64 createtime; /* creation time on server */ + u8 type; /* S_IFMT file type */ +} __packed; + static void cifs_fscache_fill_volume_coherency( struct cifs_tcon *tcon, struct cifs_fscache_volume_coherency_data *cd) @@ -97,15 +107,19 @@ void cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) void cifs_fscache_get_inode_cookie(struct inode *inode) { struct cifs_fscache_inode_coherency_data cd; + struct cifs_fscache_inode_key key; struct cifsInodeInfo *cifsi = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); + key.uniqueid = cpu_to_le64(cifsi->uniqueid); + key.createtime = cpu_to_le64(cifsi->createtime); + key.type = (inode->i_mode & S_IFMT) >> 12; cifs_fscache_fill_coherency(&cifsi->netfs.inode, &cd); cifsi->netfs.cache = fscache_acquire_cookie(tcon->fscache, 0, - &cifsi->uniqueid, sizeof(cifsi->uniqueid), + &key, sizeof(key), &cd, sizeof(cd), i_size_read(&cifsi->netfs.inode)); if (cifsi->netfs.cache) diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index d28ab0af6049..91b07ef9e25c 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -1351,6 +1351,8 @@ cifs_find_inode(struct inode *inode, void *opaque) { struct cifs_fattr *fattr = opaque; + /* [!] The compared values must be the same in struct cifs_fscache_inode_key. */ + /* don't match inode with different uniqueid */ if (CIFS_I(inode)->uniqueid != fattr->cf_uniqueid) return 0; -- cgit v1.2.3-59-g8ed1b From 99dc2ef0397d082b63404c01cf841cf80f1418dc Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 26 Mar 2024 17:00:20 -0400 Subject: NFSD: CREATE_SESSION must never cache NFS4ERR_DELAY replies There are one or two cases where CREATE_SESSION returns NFS4ERR_DELAY in order to force the client to wait a bit and try CREATE_SESSION again. However, after commit e4469c6cc69b ("NFSD: Fix the NFSv4.1 CREATE_SESSION operation"), NFSD caches that response in the CREATE_SESSION slot. Thus, when the client resends the CREATE_SESSION, the server always returns the cached NFS4ERR_DELAY response rather than actually executing the request and properly recording its outcome. This blocks the client from making further progress. RFC 8881 Section 15.1.1.3 says: > If NFS4ERR_DELAY is returned on an operation other than SEQUENCE > that validly appears as the first operation of a request ... [t]he > request can be retried in full without modification. In this case > as well, the replier MUST avoid returning a response containing > NFS4ERR_DELAY as the response to an initial operation of a request > solely on the basis of its presence in the reply cache. Neither the original NFSD code nor the discussion in section 18.36.4 refer explicitly to this important requirement, so I missed it. Note also that not only must the server not cache NFS4ERR_DELAY, but it has to not advance the CREATE_SESSION slot sequence number so that it can properly recognize and accept the client's retry. Reported-by: Dai Ngo Fixes: e4469c6cc69b ("NFSD: Fix the NFSv4.1 CREATE_SESSION operation") Tested-by: Dai Ngo Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 36 +++++++++++++++++++++++++----------- 1 file changed, 25 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ee9aa4843443..5fcd93f7cb8c 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -3831,15 +3831,20 @@ nfsd4_create_session(struct svc_rqst *rqstp, else cs_slot = &unconf->cl_cs_slot; status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0); - if (status) { - if (status == nfserr_replay_cache) { - status = nfsd4_replay_create_session(cr_ses, cs_slot); - goto out_free_conn; - } + switch (status) { + case nfs_ok: + cs_slot->sl_seqid++; + cr_ses->seqid = cs_slot->sl_seqid; + break; + case nfserr_replay_cache: + status = nfsd4_replay_create_session(cr_ses, cs_slot); + fallthrough; + case nfserr_jukebox: + /* The server MUST NOT cache NFS4ERR_DELAY */ + goto out_free_conn; + default: goto out_cache_error; } - cs_slot->sl_seqid++; - cr_ses->seqid = cs_slot->sl_seqid; /* RFC 8881 Section 18.36.4 Phase 3: Client ID confirmation. */ if (conf) { @@ -3859,10 +3864,8 @@ nfsd4_create_session(struct svc_rqst *rqstp, old = find_confirmed_client_by_name(&unconf->cl_name, nn); if (old) { status = mark_client_expired_locked(old); - if (status) { - old = NULL; - goto out_cache_error; - } + if (status) + goto out_expired_error; trace_nfsd_clid_replaced(&old->cl_clientid); } move_to_confirmed(unconf); @@ -3894,6 +3897,17 @@ nfsd4_create_session(struct svc_rqst *rqstp, expire_client(old); return status; +out_expired_error: + old = NULL; + /* + * Revert the slot seq_nr change so the server will process + * the client's resend instead of returning a cached response. + */ + if (status == nfserr_jukebox) { + cs_slot->sl_seqid--; + cr_ses->seqid = cs_slot->sl_seqid; + goto out_free_conn; + } out_cache_error: nfsd4_cache_create_session(cr_ses, cs_slot, status); out_free_conn: -- cgit v1.2.3-59-g8ed1b From cd25e15e57e68a6b18dc9323047fe9c68b99290b Mon Sep 17 00:00:00 2001 From: Joakim Sindholt Date: Mon, 18 Mar 2024 12:22:31 +0100 Subject: fs/9p: only translate RWX permissions for plain 9P2000 Garbage in plain 9P2000's perm bits is allowed through, which causes it to be able to set (among others) the suid bit. This was presumably not the intent since the unix extended bits are handled explicitly and conditionally on .u. Signed-off-by: Joakim Sindholt Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index b01b1bbf2493..9612fdb563a3 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -83,7 +83,7 @@ static int p9mode2perm(struct v9fs_session_info *v9ses, int res; int mode = stat->mode; - res = mode & S_IALLUGO; + res = mode & 0777; /* S_IRWXUGO */ if (v9fs_proto_dotu(v9ses)) { if ((mode & P9_DMSETUID) == P9_DMSETUID) res |= S_ISUID; -- cgit v1.2.3-59-g8ed1b From 87de39e70503e04ddb58965520b15eb9efa7eef3 Mon Sep 17 00:00:00 2001 From: Joakim Sindholt Date: Mon, 18 Mar 2024 12:22:33 +0100 Subject: fs/9p: translate O_TRUNC into OTRUNC This one hits both 9P2000 and .u as it appears v9fs has never translated the O_TRUNC flag. Signed-off-by: Joakim Sindholt Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_inode.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 9612fdb563a3..c5b4d3631c47 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -178,6 +178,9 @@ int v9fs_uflags2omode(int uflags, int extended) break; } + if (uflags & O_TRUNC) + ret |= P9_OTRUNC; + if (extended) { if (uflags & O_EXCL) ret |= P9_OEXCL; -- cgit v1.2.3-59-g8ed1b From 4e5d208cc9bd5fbc95d536fa223b4b14c37b8ca8 Mon Sep 17 00:00:00 2001 From: Joakim Sindholt Date: Mon, 18 Mar 2024 12:22:34 +0100 Subject: fs/9p: fix the cache always being enabled on files with qid flags I'm not sure why this check was ever here. After updating to 6.6 I suddenly found caching had been turned on by default and neither cache=none nor the new directio would turn it off. After walking through the new code very manually I realized that it's because the caching has to be, in effect, turned off explicitly by setting P9L_DIRECT and whenever a file has a flag, in my case QTAPPEND, it doesn't get set. Setting aside QTDIR which seems to ignore the new fid->mode entirely, the rest of these either should be subject to the same cache rules as every other QTFILE or perhaps very explicitly not cached in the case of QTAUTH. Signed-off-by: Joakim Sindholt Signed-off-by: Eric Van Hensbergen --- fs/9p/fid.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/9p/fid.h b/fs/9p/fid.h index 29281b7c3887..0d6138bee2a3 100644 --- a/fs/9p/fid.h +++ b/fs/9p/fid.h @@ -49,9 +49,6 @@ static inline struct p9_fid *v9fs_fid_clone(struct dentry *dentry) static inline void v9fs_fid_add_modes(struct p9_fid *fid, unsigned int s_flags, unsigned int s_cache, unsigned int f_flags) { - if (fid->qid.type != P9_QTFILE) - return; - if ((!s_cache) || ((fid->qid.version == 0) && !(s_flags & V9FS_IGNORE_QV)) || (s_flags & V9FS_DIRECT_IO) || (f_flags & O_DIRECT)) { -- cgit v1.2.3-59-g8ed1b From 7a84602297d36617dbdadeba55a2567031e5165b Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 19 Mar 2024 12:34:45 -0400 Subject: 9p: explicitly deny setlease attempts 9p is a remote network protocol, and it doesn't support asynchronous notifications from the server. Ensure that we don't hand out any leases since we can't guarantee they'll be broken when a file's contents change. Signed-off-by: Jeff Layton Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_file.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index abdbbaee5184..348cc90bf9c5 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -520,6 +520,7 @@ const struct file_operations v9fs_file_operations = { .splice_read = v9fs_file_splice_read, .splice_write = iter_file_splice_write, .fsync = v9fs_file_fsync, + .setlease = simple_nosetlease, }; const struct file_operations v9fs_file_operations_dotl = { @@ -534,4 +535,5 @@ const struct file_operations v9fs_file_operations_dotl = { .splice_read = v9fs_file_splice_read, .splice_write = iter_file_splice_write, .fsync = v9fs_file_fsync_dotl, + .setlease = simple_nosetlease, }; -- cgit v1.2.3-59-g8ed1b From 33e62cd7b4c281cd737c62e5d8c4f0e602a8c5c5 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Wed, 27 Mar 2024 15:42:23 +0800 Subject: f2fs: multidev: fix to recognize valid zero block address As reported by Yi Zhang in mailing list [1], kernel warning was catched during zbd/010 test as below: ./check zbd/010 zbd/010 (test gap zone support with F2FS) [failed] runtime ... 3.752s something found in dmesg: [ 4378.146781] run blktests zbd/010 at 2024-02-18 11:31:13 [ 4378.192349] null_blk: module loaded [ 4378.209860] null_blk: disk nullb0 created [ 4378.413285] scsi_debug:sdebug_driver_probe: scsi_debug: trim poll_queues to 0. poll_q/nr_hw = (0/1) [ 4378.422334] scsi host15: scsi_debug: version 0191 [20210520] dev_size_mb=1024, opts=0x0, submit_queues=1, statistics=0 [ 4378.434922] scsi 15:0:0:0: Direct-Access-ZBC Linux scsi_debug 0191 PQ: 0 ANSI: 7 [ 4378.443343] scsi 15:0:0:0: Power-on or device reset occurred [ 4378.449371] sd 15:0:0:0: Attached scsi generic sg5 type 20 [ 4378.449418] sd 15:0:0:0: [sdf] Host-managed zoned block device ... (See '/mnt/tests/gitlab.com/api/v4/projects/19168116/repository/archive.zip/storage/blktests/blk/blktests/results/nodev/zbd/010.dmesg' WARNING: CPU: 22 PID: 44011 at fs/iomap/iter.c:51 CPU: 22 PID: 44011 Comm: fio Not tainted 6.8.0-rc3+ #1 RIP: 0010:iomap_iter+0x32b/0x350 Call Trace: __iomap_dio_rw+0x1df/0x830 f2fs_file_read_iter+0x156/0x3d0 [f2fs] aio_read+0x138/0x210 io_submit_one+0x188/0x8c0 __x64_sys_io_submit+0x8c/0x1a0 do_syscall_64+0x86/0x170 entry_SYSCALL_64_after_hwframe+0x6e/0x76 Shinichiro Kawasaki helps to analyse this issue and proposes a potential fixing patch in [2]. Quoted from reply of Shinichiro Kawasaki: "I confirmed that the trigger commit is dbf8e63f48af as Yi reported. I took a look in the commit, but it looks fine to me. So I thought the cause is not in the commit diff. I found the WARN is printed when the f2fs is set up with multiple devices, and read requests are mapped to the very first block of the second device in the direct read path. In this case, f2fs_map_blocks() and f2fs_map_blocks_cached() modify map->m_pblk as the physical block address from each block device. It becomes zero when it is mapped to the first block of the device. However, f2fs_iomap_begin() assumes that map->m_pblk is the physical block address of the whole f2fs, across the all block devices. It compares map->m_pblk against NULL_ADDR == 0, then go into the unexpected branch and sets the invalid iomap->length. The WARN catches the invalid iomap->length. This WARN is printed even for non-zoned block devices, by following steps. - Create two (non-zoned) null_blk devices memory backed with 128MB size each: nullb0 and nullb1. # mkfs.f2fs /dev/nullb0 -c /dev/nullb1 # mount -t f2fs /dev/nullb0 "${mount_dir}" # dd if=/dev/zero of="${mount_dir}/test.dat" bs=1M count=192 # dd if="${mount_dir}/test.dat" of=/dev/null bs=1M count=192 iflag=direct ..." So, the root cause of this issue is: when multi-devices feature is on, f2fs_map_blocks() may return zero blkaddr in non-primary device, which is a verified valid block address, however, f2fs_iomap_begin() treats it as an invalid block address, and then it triggers the warning in iomap framework code. Finally, as discussed, we decide to use a more simple and direct way that checking (map.m_flags & F2FS_MAP_MAPPED) condition instead of (map.m_pblk != NULL_ADDR) to fix this issue. Thanks a lot for the effort of Yi Zhang and Shinichiro Kawasaki on this issue. [1] https://lore.kernel.org/linux-f2fs-devel/CAHj4cs-kfojYC9i0G73PRkYzcxCTex=-vugRFeP40g_URGvnfQ@mail.gmail.com/ [2] https://lore.kernel.org/linux-f2fs-devel/gngdj77k4picagsfdtiaa7gpgnup6fsgwzsltx6milmhegmjff@iax2n4wvrqye/ Reported-by: Yi Zhang Closes: https://lore.kernel.org/linux-f2fs-devel/CAHj4cs-kfojYC9i0G73PRkYzcxCTex=-vugRFeP40g_URGvnfQ@mail.gmail.com/ Tested-by: Shin'ichiro Kawasaki Tested-by: Yi Zhang Fixes: 1517c1a7a445 ("f2fs: implement iomap operations") Fixes: 8d3c1fa3fa5e ("f2fs: don't rely on F2FS_MAP_* in f2fs_iomap_begin") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 08815394223a..fa5398ac4505 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -4196,7 +4196,7 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, if (WARN_ON_ONCE(map.m_pblk == COMPRESS_ADDR)) return -EINVAL; - if (map.m_pblk != NULL_ADDR) { + if (map.m_flags & F2FS_MAP_MAPPED) { iomap->length = blks_to_bytes(inode, map.m_len); iomap->type = IOMAP_MAPPED; iomap->flags |= IOMAP_F_MERGED; -- cgit v1.2.3-59-g8ed1b From 9f0f6bf4271488b3d3a290ba119a0e0a08df2cc6 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 25 Mar 2024 23:27:26 +0800 Subject: f2fs: support to map continuous holes or preallocated address This patch supports to map continuous holes or preallocated addresses to improve performace of lookuping mapping info during read DIO. [testcase 1] xfs_io -f /mnt/f2fs/hole -c "truncate 1m" -c "fsync" xfs_io -d /mnt/f2fs/hole -c "pread -b 1m 0 1m" [before] f2fs_direct_IO_enter: dev = (253,16), ino = 6 pos = 0 len = 1048576 ki_flags = 20000 ki_ioprio = 0 rw = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 0, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 1, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 2, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 3, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 4, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 5, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 6, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 7, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 8, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 9, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 10, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 11, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 12, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 13, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 14, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 15, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 16, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 ...... f2fs_direct_IO_exit: dev = (253,16), ino = 6 pos = 0 len = 1048576 rw = 0 ret = 1048576 [after] f2fs_direct_IO_enter: dev = (253,16), ino = 6 pos = 0 len = 1048576 ki_flags = 20000 ki_ioprio = 0 rw = 0 f2fs_map_blocks: dev = (253,16), ino = 6, file offset = 0, start blkaddr = 0x0, len = 0x100, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_direct_IO_exit: dev = (253,16), ino = 6 pos = 0 len = 1048576 rw = 0 ret = 1048576 [testcase 2] xfs_io -f /mnt/f2fs/preallocated -c "falloc 0 1m" -c "fsync" xfs_io -d /mnt/f2fs/preallocated -c "pread -b 1m 0 1m" [before] f2fs_direct_IO_enter: dev = (253,16), ino = 11 pos = 0 len = 1048576 ki_flags = 20000 ki_ioprio = 0 rw = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 0, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 1, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 2, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 3, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 4, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 5, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 6, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 7, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 8, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 9, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 10, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 11, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 12, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 13, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 14, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 15, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 16, start blkaddr = 0x0, len = 0x0, flags = 0, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 ...... f2fs_direct_IO_exit: dev = (253,16), ino = 11 pos = 0 len = 1048576 rw = 0 ret = 1048576 [after] f2fs_direct_IO_enter: dev = (253,16), ino = 11 pos = 0 len = 1048576 ki_flags = 20000 ki_ioprio = 0 rw = 0 f2fs_map_blocks: dev = (253,16), ino = 11, file offset = 0, start blkaddr = 0xffffffff, len = 0x100, flags = 4, seg_type = 1, may_create = 0, multidevice = 0, flag = 3, err = 0 f2fs_direct_IO_exit: dev = (253,16), ino = 11 pos = 0 len = 1048576 rw = 0 ret = 1048576 Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 29 +++++++++++++++++++++++------ 1 file changed, 23 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index fa5398ac4505..0d88649c60a5 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1520,6 +1520,9 @@ static bool map_is_mergeable(struct f2fs_sb_info *sbi, return true; if (flag == F2FS_GET_BLOCK_PRE_DIO) return true; + if (flag == F2FS_GET_BLOCK_DIO && + map->m_pblk == NULL_ADDR && blkaddr == NULL_ADDR) + return true; return false; } @@ -1644,6 +1647,10 @@ next_block: goto sync_out; } break; + case F2FS_GET_BLOCK_DIO: + if (map->m_next_pgofs) + *map->m_next_pgofs = pgofs + 1; + break; default: /* for defragment case */ if (map->m_next_pgofs) @@ -1662,7 +1669,8 @@ next_block: /* reserved delalloc block should be mapped for fiemap. */ if (blkaddr == NEW_ADDR) map->m_flags |= F2FS_MAP_DELALLOC; - map->m_flags |= F2FS_MAP_MAPPED; + if (flag != F2FS_GET_BLOCK_DIO || !is_hole) + map->m_flags |= F2FS_MAP_MAPPED; map->m_pblk = blkaddr; map->m_len = 1; @@ -4191,12 +4199,13 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, * We should never see delalloc or compressed extents here based on * prior flushing and checks. */ - if (WARN_ON_ONCE(map.m_pblk == NEW_ADDR)) - return -EINVAL; if (WARN_ON_ONCE(map.m_pblk == COMPRESS_ADDR)) return -EINVAL; if (map.m_flags & F2FS_MAP_MAPPED) { + if (WARN_ON_ONCE(map.m_pblk == NEW_ADDR)) + return -EINVAL; + iomap->length = blks_to_bytes(inode, map.m_len); iomap->type = IOMAP_MAPPED; iomap->flags |= IOMAP_F_MERGED; @@ -4205,9 +4214,17 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, } else { if (flags & IOMAP_WRITE) return -ENOTBLK; - iomap->length = blks_to_bytes(inode, next_pgofs) - - iomap->offset; - iomap->type = IOMAP_HOLE; + + if (map.m_pblk == NULL_ADDR) { + iomap->length = blks_to_bytes(inode, next_pgofs) - + iomap->offset; + iomap->type = IOMAP_HOLE; + } else if (map.m_pblk == NEW_ADDR) { + iomap->length = blks_to_bytes(inode, map.m_len); + iomap->type = IOMAP_UNWRITTEN; + } else { + f2fs_bug_on(F2FS_I_SB(inode), 1); + } iomap->addr = IOMAP_NULL_ADDR; } -- cgit v1.2.3-59-g8ed1b From d3876e34e7e789e2cbdd782360fef2a777391082 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 26 Mar 2024 19:28:45 +0800 Subject: f2fs: fix to wait on page writeback in __clone_blkaddrs() In below race condition, dst page may become writeback status in __clone_blkaddrs(), it needs to wait writeback before update, fix it. Thread A GC Thread - f2fs_move_file_range - filemap_write_and_wait_range(dst) - gc_data_segment - f2fs_down_write(dst) - move_data_page - set_page_writeback(dst_page) - f2fs_submit_page_write - f2fs_up_write(dst) - f2fs_down_write(dst) - __exchange_data_block - __clone_blkaddrs - f2fs_get_new_data_page - memcpy_page Fixes: 0a2aa8fbb969 ("f2fs: refactor __exchange_data_block for speed up") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index f4347db5b03e..719064730f9a 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1325,6 +1325,9 @@ static int __clone_blkaddrs(struct inode *src_inode, struct inode *dst_inode, f2fs_put_page(psrc, 1); return PTR_ERR(pdst); } + + f2fs_wait_on_page_writeback(pdst, DATA, true, true); + memcpy_page(pdst, 0, psrc, 0, PAGE_SIZE); set_page_dirty(pdst); set_page_private_gcing(pdst); -- cgit v1.2.3-59-g8ed1b From ac5eecf481c29942eb9a862e758c0c8b68090c33 Mon Sep 17 00:00:00 2001 From: Yunlei He Date: Tue, 26 Mar 2024 14:10:43 +0800 Subject: f2fs: remove clear SB_INLINECRYPT flag in default_options In f2fs_remount, SB_INLINECRYPT flag will be clear and re-set. If create new file or open file during this gap, these files will not use inlinecrypt. Worse case, it may lead to data corruption if wrappedkey_v0 is enable. Thread A: Thread B: -f2fs_remount -f2fs_file_open or f2fs_new_inode -default_options <- clear SB_INLINECRYPT flag -fscrypt_select_encryption_impl -parse_options <- set SB_INLINECRYPT again Signed-off-by: Yunlei He Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/super.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 7c45929671ad..df9765b41dac 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -2132,8 +2132,6 @@ static void default_options(struct f2fs_sb_info *sbi, bool remount) F2FS_OPTION(sbi).memory_mode = MEMORY_MODE_NORMAL; F2FS_OPTION(sbi).errors = MOUNT_ERRORS_CONTINUE; - sbi->sb->s_flags &= ~SB_INLINECRYPT; - set_opt(sbi, INLINE_XATTR); set_opt(sbi, INLINE_DATA); set_opt(sbi, INLINE_DENTRY); -- cgit v1.2.3-59-g8ed1b From c40845e3195d074b34f8f8e400e28c9403a06588 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 26 Mar 2024 15:47:16 +0100 Subject: kbuild: make -Woverride-init warnings more consistent The -Woverride-init warn about code that may be intentional or not, but the inintentional ones tend to be real bugs, so there is a bit of disagreement on whether this warning option should be enabled by default and we have multiple settings in scripts/Makefile.extrawarn as well as individual subsystems. Older versions of clang only supported -Wno-initializer-overrides with the same meaning as gcc's -Woverride-init, though all supported versions now work with both. Because of this difference, an earlier cleanup of mine accidentally turned the clang warning off for W=1 builds and only left it on for W=2, while it's still enabled for gcc with W=1. There is also one driver that only turns the warning off for newer versions of gcc but not other compilers, and some but not all the Makefiles still use a cc-disable-warning conditional that is no longer needed with supported compilers here. Address all of the above by removing the special cases for clang and always turning the warning off unconditionally where it got in the way, using the syntax that is supported by both compilers. Fixes: 2cd3271b7a31 ("kbuild: avoid duplicate warning options") Signed-off-by: Arnd Bergmann Acked-by: Hamza Mahfooz Acked-by: Jani Nikula Acked-by: Andrew Jeffery Signed-off-by: Jani Nikula Reviewed-by: Linus Walleij Signed-off-by: Masahiro Yamada --- drivers/gpu/drm/amd/display/dc/dce110/Makefile | 2 +- drivers/gpu/drm/amd/display/dc/dce112/Makefile | 2 +- drivers/gpu/drm/amd/display/dc/dce120/Makefile | 2 +- drivers/gpu/drm/amd/display/dc/dce60/Makefile | 2 +- drivers/gpu/drm/amd/display/dc/dce80/Makefile | 2 +- drivers/gpu/drm/i915/Makefile | 6 +++--- drivers/gpu/drm/xe/Makefile | 4 ++-- drivers/net/ethernet/renesas/sh_eth.c | 2 +- drivers/pinctrl/aspeed/Makefile | 2 +- fs/proc/Makefile | 2 +- kernel/bpf/Makefile | 2 +- mm/Makefile | 3 +-- scripts/Makefile.extrawarn | 10 +++------- 13 files changed, 18 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/drivers/gpu/drm/amd/display/dc/dce110/Makefile b/drivers/gpu/drm/amd/display/dc/dce110/Makefile index f0777d61c2cb..c307f040e48f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce110/Makefile @@ -23,7 +23,7 @@ # Makefile for the 'controller' sub-component of DAL. # It provides the control and status of HW CRTC block. -CFLAGS_$(AMDDALPATH)/dc/dce110/dce110_resource.o = $(call cc-disable-warning, override-init) +CFLAGS_$(AMDDALPATH)/dc/dce110/dce110_resource.o = -Wno-override-init DCE110 = dce110_timing_generator.o \ dce110_compressor.o dce110_opp_regamma_v.o \ diff --git a/drivers/gpu/drm/amd/display/dc/dce112/Makefile b/drivers/gpu/drm/amd/display/dc/dce112/Makefile index 7e92effec894..683866797709 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce112/Makefile @@ -23,7 +23,7 @@ # Makefile for the 'controller' sub-component of DAL. # It provides the control and status of HW CRTC block. -CFLAGS_$(AMDDALPATH)/dc/dce112/dce112_resource.o = $(call cc-disable-warning, override-init) +CFLAGS_$(AMDDALPATH)/dc/dce112/dce112_resource.o = -Wno-override-init DCE112 = dce112_compressor.o diff --git a/drivers/gpu/drm/amd/display/dc/dce120/Makefile b/drivers/gpu/drm/amd/display/dc/dce120/Makefile index 1e3ef68a452a..8f508e662748 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce120/Makefile @@ -24,7 +24,7 @@ # It provides the control and status of HW CRTC block. -CFLAGS_$(AMDDALPATH)/dc/dce120/dce120_resource.o = $(call cc-disable-warning, override-init) +CFLAGS_$(AMDDALPATH)/dc/dce120/dce120_resource.o = -Wno-override-init DCE120 = dce120_timing_generator.o diff --git a/drivers/gpu/drm/amd/display/dc/dce60/Makefile b/drivers/gpu/drm/amd/display/dc/dce60/Makefile index fee331accc0e..eede83ad91fa 100644 --- a/drivers/gpu/drm/amd/display/dc/dce60/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce60/Makefile @@ -23,7 +23,7 @@ # Makefile for the 'controller' sub-component of DAL. # It provides the control and status of HW CRTC block. -CFLAGS_$(AMDDALPATH)/dc/dce60/dce60_resource.o = $(call cc-disable-warning, override-init) +CFLAGS_$(AMDDALPATH)/dc/dce60/dce60_resource.o = -Wno-override-init DCE60 = dce60_timing_generator.o dce60_hw_sequencer.o \ dce60_resource.o diff --git a/drivers/gpu/drm/amd/display/dc/dce80/Makefile b/drivers/gpu/drm/amd/display/dc/dce80/Makefile index 7eefffbdc925..fba189d26652 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce80/Makefile @@ -23,7 +23,7 @@ # Makefile for the 'controller' sub-component of DAL. # It provides the control and status of HW CRTC block. -CFLAGS_$(AMDDALPATH)/dc/dce80/dce80_resource.o = $(call cc-disable-warning, override-init) +CFLAGS_$(AMDDALPATH)/dc/dce80/dce80_resource.o = -Wno-override-init DCE80 = dce80_timing_generator.o diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 3ef6ed41e62b..4c2f85632391 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -33,9 +33,9 @@ endif subdir-ccflags-$(CONFIG_DRM_I915_WERROR) += -Werror # Fine grained warnings disable -CFLAGS_i915_pci.o = $(call cc-disable-warning, override-init) -CFLAGS_display/intel_display_device.o = $(call cc-disable-warning, override-init) -CFLAGS_display/intel_fbdev.o = $(call cc-disable-warning, override-init) +CFLAGS_i915_pci.o = -Wno-override-init +CFLAGS_display/intel_display_device.o = -Wno-override-init +CFLAGS_display/intel_fbdev.o = -Wno-override-init # Support compiling the display code separately for both i915 and xe # drivers. Define I915 when building i915. diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index 5a428ca00f10..c29a850859ad 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -172,8 +172,8 @@ subdir-ccflags-$(CONFIG_DRM_XE_DISPLAY) += \ -Ddrm_i915_gem_object=xe_bo \ -Ddrm_i915_private=xe_device -CFLAGS_i915-display/intel_fbdev.o = $(call cc-disable-warning, override-init) -CFLAGS_i915-display/intel_display_device.o = $(call cc-disable-warning, override-init) +CFLAGS_i915-display/intel_fbdev.o = -Wno-override-init +CFLAGS_i915-display/intel_display_device.o = -Wno-override-init # Rule to build SOC code shared with i915 $(obj)/i915-soc/%.o: $(srctree)/drivers/gpu/drm/i915/soc/%.c FORCE diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 475e1e8c1d35..0786eb0da391 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -50,7 +50,7 @@ * the macros available to do this only define GCC 8. */ __diag_push(); -__diag_ignore(GCC, 8, "-Woverride-init", +__diag_ignore_all("-Woverride-init", "logic to initialize all and then override some is OK"); static const u16 sh_eth_offset_gigabit[SH_ETH_MAX_REGISTER_OFFSET] = { SH_ETH_OFFSET_DEFAULTS, diff --git a/drivers/pinctrl/aspeed/Makefile b/drivers/pinctrl/aspeed/Makefile index 489ea1778353..db2a7600ae2b 100644 --- a/drivers/pinctrl/aspeed/Makefile +++ b/drivers/pinctrl/aspeed/Makefile @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only # Aspeed pinctrl support -ccflags-y += $(call cc-option,-Woverride-init) +ccflags-y += -Woverride-init obj-$(CONFIG_PINCTRL_ASPEED) += pinctrl-aspeed.o pinmux-aspeed.o obj-$(CONFIG_PINCTRL_ASPEED_G4) += pinctrl-aspeed-g4.o obj-$(CONFIG_PINCTRL_ASPEED_G5) += pinctrl-aspeed-g5.o diff --git a/fs/proc/Makefile b/fs/proc/Makefile index bd08616ed8ba..7b4db9c56e6a 100644 --- a/fs/proc/Makefile +++ b/fs/proc/Makefile @@ -5,7 +5,7 @@ obj-y += proc.o -CFLAGS_task_mmu.o += $(call cc-option,-Wno-override-init,) +CFLAGS_task_mmu.o += -Wno-override-init proc-y := nommu.o task_nommu.o proc-$(CONFIG_MMU) := task_mmu.o diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile index 368c5d86b5b7..e497011261b8 100644 --- a/kernel/bpf/Makefile +++ b/kernel/bpf/Makefile @@ -4,7 +4,7 @@ ifneq ($(CONFIG_BPF_JIT_ALWAYS_ON),y) # ___bpf_prog_run() needs GCSE disabled on x86; see 3193c0836f203 for details cflags-nogcse-$(CONFIG_X86)$(CONFIG_CC_IS_GCC) := -fno-gcse endif -CFLAGS_core.o += $(call cc-disable-warning, override-init) $(cflags-nogcse-yy) +CFLAGS_core.o += -Wno-override-init $(cflags-nogcse-yy) obj-$(CONFIG_BPF_SYSCALL) += syscall.o verifier.o inode.o helpers.o tnum.o log.o token.o obj-$(CONFIG_BPF_SYSCALL) += bpf_iter.o map_iter.o task_iter.o prog_iter.o link_iter.o diff --git a/mm/Makefile b/mm/Makefile index e4b5b75aaec9..4abb40b911ec 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -29,8 +29,7 @@ KCOV_INSTRUMENT_mmzone.o := n KCOV_INSTRUMENT_vmstat.o := n KCOV_INSTRUMENT_failslab.o := n -CFLAGS_init-mm.o += $(call cc-disable-warning, override-init) -CFLAGS_init-mm.o += $(call cc-disable-warning, initializer-overrides) +CFLAGS_init-mm.o += -Wno-override-init mmu-y := nommu.o mmu-$(CONFIG_MMU) := highmem.o memory.o mincore.o \ diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn index 3ce5d503a6da..c5af566e911a 100644 --- a/scripts/Makefile.extrawarn +++ b/scripts/Makefile.extrawarn @@ -114,6 +114,8 @@ KBUILD_CFLAGS += $(call cc-disable-warning, format-overflow) KBUILD_CFLAGS += $(call cc-disable-warning, format-truncation) KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation) +KBUILD_CFLAGS += -Wno-override-init # alias for -Wno-initializer-overrides in clang + ifdef CONFIG_CC_IS_CLANG # Clang before clang-16 would warn on default argument promotions. ifneq ($(call clang-min-version, 160000),y) @@ -151,10 +153,6 @@ KBUILD_CFLAGS += -Wtype-limits KBUILD_CFLAGS += $(call cc-option, -Wmaybe-uninitialized) KBUILD_CFLAGS += $(call cc-option, -Wunused-macros) -ifdef CONFIG_CC_IS_CLANG -KBUILD_CFLAGS += -Winitializer-overrides -endif - KBUILD_CPPFLAGS += -DKBUILD_EXTRA_WARN2 else @@ -164,9 +162,7 @@ KBUILD_CFLAGS += -Wno-missing-field-initializers KBUILD_CFLAGS += -Wno-type-limits KBUILD_CFLAGS += -Wno-shift-negative-value -ifdef CONFIG_CC_IS_CLANG -KBUILD_CFLAGS += -Wno-initializer-overrides -else +ifdef CONFIG_CC_IS_GCC KBUILD_CFLAGS += -Wno-maybe-uninitialized endif -- cgit v1.2.3-59-g8ed1b From d7ed5232f0f16181506373d73e711190d5e0c868 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Sun, 31 Mar 2024 23:38:17 +0900 Subject: exfat: fix timing of synchronizing bitmap and inode Commit(f55c096f62f1 exfat: do not zero the extended part) changed the timing of synchronizing bitmap and inode in exfat_cont_expand(). The change caused xfstests generic/013 to fail if 'dirsync' or 'sync' is enabled. So this commit restores the timing. Fixes: f55c096f62f1 ("exfat: do not zero the extended part") Signed-off-by: Yuezhang Mo Signed-off-by: Namjae Jeon --- fs/exfat/file.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/exfat/file.c b/fs/exfat/file.c index cc00f1a7a1e1..9adfc38ca7da 100644 --- a/fs/exfat/file.c +++ b/fs/exfat/file.c @@ -51,7 +51,7 @@ static int exfat_cont_expand(struct inode *inode, loff_t size) clu.flags = ei->flags; ret = exfat_alloc_cluster(inode, new_num_clusters - num_clusters, - &clu, IS_DIRSYNC(inode)); + &clu, inode_needs_sync(inode)); if (ret) return ret; @@ -77,12 +77,11 @@ out: ei->i_size_aligned = round_up(size, sb->s_blocksize); ei->i_size_ondisk = ei->i_size_aligned; inode->i_blocks = round_up(size, sbi->cluster_size) >> 9; + mark_inode_dirty(inode); - if (IS_DIRSYNC(inode)) + if (IS_SYNC(inode)) return write_inode_now(inode, 1); - mark_inode_dirty(inode); - return 0; free_clu: -- cgit v1.2.3-59-g8ed1b From ebd9779683aaf089ad0173862553cdd3288ad9b4 Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Thu, 28 Mar 2024 21:44:48 +0000 Subject: smb: client: replace deprecated strncpy with strscpy strncpy() is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. In cifssmb.c: Using strncpy with a length argument equal to strlen(src) is generally dangerous because it can cause string buffers to not be NUL-terminated. In this case, however, there was extra effort made to ensure the buffer was NUL-terminated via a manual NUL-byte assignment. In an effort to rid the kernel of strncpy() use, let's swap over to using strscpy() which guarantees NUL-termination on the destination buffer. To handle the case where ea_name is NULL, let's use the ?: operator to substitute in an empty string, thereby allowing strscpy to still NUL-terminate the destintation string. Interesting note: this flex array buffer may go on to also have some value encoded after the NUL-termination: | if (ea_value_len) | memcpy(parm_data->list.name + name_len + 1, | ea_value, ea_value_len); Now for smb2ops.c and smb2transport.c: Both of these cases are simple, strncpy() is used to copy string literals which have a length less than the destination buffer's size. We can simply swap in the new 2-argument version of strscpy() introduced in Commit e6584c3964f2f ("string: Allow 2-argument strscpy()"). Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2] Link: https://github.com/KSPP/linux/issues/90 Cc: linux-hardening@vger.kernel.org Signed-off-by: Justin Stitt Reviewed-by: Kees Cook Signed-off-by: Steve French --- fs/smb/client/cifssmb.c | 6 ++---- fs/smb/client/smb2ops.c | 2 +- fs/smb/client/smb2transport.c | 2 +- 3 files changed, 4 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index 5aee55551573..23b5709ddc31 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -5854,10 +5854,8 @@ SetEARetry: parm_data->list.EA_flags = 0; /* we checked above that name len is less than 255 */ parm_data->list.name_len = (__u8)name_len; - /* EA names are always ASCII */ - if (ea_name) - strncpy(parm_data->list.name, ea_name, name_len); - parm_data->list.name[name_len] = '\0'; + /* EA names are always ASCII and NUL-terminated */ + strscpy(parm_data->list.name, ea_name ?: "", name_len + 1); parm_data->list.value_len = cpu_to_le16(ea_value_len); /* caller ensures that ea_value_len is less than 64K but we need to ensure that it fits within the smb */ diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 2ed456948f34..35bf7eb315cd 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -3913,7 +3913,7 @@ smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, strcat(message, "W"); } if (!new_oplock) - strncpy(message, "None", sizeof(message)); + strscpy(message, "None"); cinode->oplock = new_oplock; cifs_dbg(FYI, "%s Lease granted on inode %p\n", message, diff --git a/fs/smb/client/smb2transport.c b/fs/smb/client/smb2transport.c index 5a3ca62d2f07..1d6e54f7879e 100644 --- a/fs/smb/client/smb2transport.c +++ b/fs/smb/client/smb2transport.c @@ -659,7 +659,7 @@ smb2_sign_rqst(struct smb_rqst *rqst, struct TCP_Server_Info *server) } spin_unlock(&server->srv_lock); if (!is_binding && !server->session_estab) { - strncpy(shdr->Signature, "BSRSPYL", 8); + strscpy(shdr->Signature, "BSRSPYL"); return 0; } -- cgit v1.2.3-59-g8ed1b From a5e3dce493d4b12b74000b6a99b6712afa5d1a4d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 22 Mar 2024 00:15:41 -0400 Subject: bcachefs: Fix assert in bch2_backpointer_invalid() Backpointers that point to invalid devices are caught by fsck, not .key_invalid; so .key_invalid needs to check for them instead of hitting asserts. Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 8cb35ea572cb..21b3ae7df824 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -44,6 +44,11 @@ int bch2_backpointer_invalid(struct bch_fs *c, struct bkey_s_c k, struct printbuf *err) { struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); + + /* these will be caught by fsck */ + if (!bch2_dev_exists2(c, bp.k->p.inode)) + return 0; + struct bpos bucket = bp_pos_to_bucket(c, bp.k->p); int ret = 0; -- cgit v1.2.3-59-g8ed1b From 8aad8e1f659fcea1b24072e816e434e4cd12382d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 22 Mar 2024 04:01:27 -0400 Subject: bcachefs: Fix journal pins in btree write buffer btree write buffer flush has two phases - in natural key order, which is more efficient but may fail - then in journal order The journal order flush was assuming that keys were still correctly ordered by journal sequence number - but due to coalescing by the previous phase, we need an additional sort. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_write_buffer.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_write_buffer.c b/fs/bcachefs/btree_write_buffer.c index 5cbad8445782..baf63e2fddb6 100644 --- a/fs/bcachefs/btree_write_buffer.c +++ b/fs/bcachefs/btree_write_buffer.c @@ -11,6 +11,7 @@ #include "journal_reclaim.h" #include +#include static int bch2_btree_write_buffer_journal_flush(struct journal *, struct journal_entry_pin *, u64); @@ -46,6 +47,14 @@ static inline bool wb_key_ref_cmp(const struct wb_key_ref *l, const struct wb_ke #endif } +static int wb_key_seq_cmp(const void *_l, const void *_r) +{ + const struct btree_write_buffered_key *l = _l; + const struct btree_write_buffered_key *r = _r; + + return cmp_int(l->journal_seq, r->journal_seq); +} + /* Compare excluding idx, the low 24 bits: */ static inline bool wb_key_eq(const void *_l, const void *_r) { @@ -357,6 +366,11 @@ static int bch2_btree_write_buffer_flush_locked(struct btree_trans *trans) */ trace_and_count(c, write_buffer_flush_slowpath, trans, slowpath, wb->flushing.keys.nr); + sort(wb->flushing.keys.data, + wb->flushing.keys.nr, + sizeof(wb->flushing.keys.data[0]), + wb_key_seq_cmp, NULL); + darray_for_each(wb->flushing.keys, i) { if (!i->journal_seq) continue; -- cgit v1.2.3-59-g8ed1b From 688d750d10aa9c4fb71c5154521c775f94c887e0 Mon Sep 17 00:00:00 2001 From: Thomas Bertschinger Date: Wed, 20 Mar 2024 21:42:42 -0600 Subject: bcachefs: fix misplaced newline in __bch2_inode_unpacked_to_text() before: u64s 18 type inode_v3 0:1879048192:U32_MAX len 0 ver 0: mode=40700 flags= (15300000) journal_seq=4 bi_size=0 bi_sectors=0 bi_version=0bi_atime=227064388944 ... after: u64s 18 type inode_v3 0:1879048192:U32_MAX len 0 ver 0: mode=40700 flags= (15300000) journal_seq=4 bi_size=0 bi_sectors=0 bi_version=0 bi_atime=227064388944 ... Signed-off-by: Thomas Bertschinger Signed-off-by: Kent Overstreet --- fs/bcachefs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index 2b5e06770ab3..ca4a066e9a54 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -552,8 +552,8 @@ static void __bch2_inode_unpacked_to_text(struct printbuf *out, prt_printf(out, "bi_sectors=%llu", inode->bi_sectors); prt_newline(out); - prt_newline(out); prt_printf(out, "bi_version=%llu", inode->bi_version); + prt_newline(out); #define x(_name, _bits) \ prt_printf(out, #_name "=%llu", (u64) inode->_name); \ -- cgit v1.2.3-59-g8ed1b From 4bd02d3fb33d8a46e73085b8d47d21c0ccb3de9d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 28 Mar 2024 01:20:36 -0400 Subject: bcachefs: fix mount error path Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 0ccee05f6887..b5ea9fa1259d 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -1997,6 +1997,7 @@ out: return dget(sb->s_root); err_put_super: + __bch2_fs_stop(c); deactivate_locked_super(sb); return ERR_PTR(bch2_err_class(ret)); } -- cgit v1.2.3-59-g8ed1b From aa6e130e3c2965a5c26a4033ff63b5dc9549bd76 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 24 Mar 2024 19:52:03 -0400 Subject: bcachefs: Add an assertion for trying to evict btree root Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 562561a9a510..8ff21b2e1463 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -1134,6 +1134,8 @@ void bch2_btree_node_evict(struct btree_trans *trans, const struct bkey_i *k) b = btree_cache_find(bc, k); if (!b) return; + + BUG_ON(b == btree_node_root(trans->c, b)); wait_on_io: /* not allowed to wait on io with btree locks held: */ -- cgit v1.2.3-59-g8ed1b From 63332394c7e1f4f26e8e5b1387212016aaa7eae2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 21 Mar 2024 20:16:23 -0400 Subject: bcachefs: Move snapshot table size to struct snapshot_table We need to add bounds checking for snapshot table accesses - it turns out there are cases where we do need to use the snapshots table before fsck checks have completed (and indeed, fsck may not have been run). Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 1 - fs/bcachefs/snapshot.c | 25 ++++++++++++++----------- fs/bcachefs/snapshot.h | 6 +++++- fs/bcachefs/subvolume_types.h | 2 ++ 4 files changed, 21 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 799aa32b6b4d..a9ade0b1f78c 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -810,7 +810,6 @@ struct bch_fs { /* snapshot.c: */ struct snapshot_table __rcu *snapshots; - size_t snapshot_table_size; struct mutex snapshot_table_lock; struct rw_semaphore snapshot_create_lock; diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 39debe814bf3..9cd71e613dc9 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -151,36 +151,39 @@ out: static noinline struct snapshot_t *__snapshot_t_mut(struct bch_fs *c, u32 id) { size_t idx = U32_MAX - id; - size_t new_size; struct snapshot_table *new, *old; - new_size = max(16UL, roundup_pow_of_two(idx + 1)); + size_t new_bytes = kmalloc_size_roundup(struct_size(new, s, idx + 1)); + size_t new_size = (new_bytes - sizeof(*new)) / sizeof(new->s[0]); - new = kvzalloc(struct_size(new, s, new_size), GFP_KERNEL); + new = kvzalloc(new_bytes, GFP_KERNEL); if (!new) return NULL; + new->nr = new_size; + old = rcu_dereference_protected(c->snapshots, true); if (old) - memcpy(new->s, - rcu_dereference_protected(c->snapshots, true)->s, - sizeof(new->s[0]) * c->snapshot_table_size); + memcpy(new->s, old->s, sizeof(old->s[0]) * old->nr); rcu_assign_pointer(c->snapshots, new); - c->snapshot_table_size = new_size; - kvfree_rcu_mightsleep(old); + kvfree_rcu(old, rcu); - return &rcu_dereference_protected(c->snapshots, true)->s[idx]; + return &rcu_dereference_protected(c->snapshots, + lockdep_is_held(&c->snapshot_table_lock))->s[idx]; } static inline struct snapshot_t *snapshot_t_mut(struct bch_fs *c, u32 id) { size_t idx = U32_MAX - id; + struct snapshot_table *table = + rcu_dereference_protected(c->snapshots, + lockdep_is_held(&c->snapshot_table_lock)); lockdep_assert_held(&c->snapshot_table_lock); - if (likely(idx < c->snapshot_table_size)) - return &rcu_dereference_protected(c->snapshots, true)->s[idx]; + if (likely(table && idx < table->nr)) + return &table->s[idx]; return __snapshot_t_mut(c, id); } diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index 7c66ffc06385..8538b7fddfed 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -33,7 +33,11 @@ int bch2_mark_snapshot(struct btree_trans *, enum btree_id, unsigned, static inline struct snapshot_t *__snapshot_t(struct snapshot_table *t, u32 id) { - return &t->s[U32_MAX - id]; + u32 idx = U32_MAX - id; + + return likely(t && idx < t->nr) + ? &t->s[idx] + : NULL; } static inline const struct snapshot_t *snapshot_t(struct bch_fs *c, u32 id) diff --git a/fs/bcachefs/subvolume_types.h b/fs/bcachefs/subvolume_types.h index ae644adfc391..9b10c8947828 100644 --- a/fs/bcachefs/subvolume_types.h +++ b/fs/bcachefs/subvolume_types.h @@ -20,6 +20,8 @@ struct snapshot_t { }; struct snapshot_table { + struct rcu_head rcu; + size_t nr; #ifndef RUST_BINDGEN DECLARE_FLEX_ARRAY(struct snapshot_t, s); #else -- cgit v1.2.3-59-g8ed1b From ec9cc18fc2e65b08c588e01f24aaeb71551a7132 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 22 Mar 2024 16:29:23 -0400 Subject: bcachefs: Add checks for invalid snapshot IDs Previously, we assumed that keys were consistent with the snapshots btree - but that's not correct as fsck may not have been run or may not be complete. This adds checks and error handling when using the in-memory snapshots table (that mirrors the snapshots btree). Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_trans_commit.c | 2 +- fs/bcachefs/data_update.c | 9 +++++++ fs/bcachefs/errcode.h | 3 ++- fs/bcachefs/snapshot.c | 8 ++++-- fs/bcachefs/snapshot.h | 57 ++++++++++++++++------------------------ 5 files changed, 40 insertions(+), 39 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 30d69a6d133e..96669fede7d3 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -318,7 +318,7 @@ static inline void btree_insert_entry_checks(struct btree_trans *trans, !(i->flags & BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) && test_bit(JOURNAL_REPLAY_DONE, &trans->c->journal.flags) && i->k->k.p.snapshot && - bch2_snapshot_is_internal_node(trans->c, i->k->k.p.snapshot)); + bch2_snapshot_is_internal_node(trans->c, i->k->k.p.snapshot) > 0); } static __always_inline int bch2_trans_journal_res_get(struct btree_trans *trans, diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 4150feca42a2..b564404dad71 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -14,6 +14,7 @@ #include "move.h" #include "nocow_locking.h" #include "rebalance.h" +#include "snapshot.h" #include "subvolume.h" #include "trace.h" @@ -509,6 +510,14 @@ int bch2_data_update_init(struct btree_trans *trans, unsigned ptrs_locked = 0; int ret = 0; + /* + * fs is corrupt we have a key for a snapshot node that doesn't exist, + * and we have to check for this because we go rw before repairing the + * snapshots table - just skip it, we can move it later. + */ + if (unlikely(k.k->p.snapshot && !bch2_snapshot_equiv(c, k.k->p.snapshot))) + return -BCH_ERR_data_update_done; + bch2_bkey_buf_init(&m->k); bch2_bkey_buf_reassemble(&m->k, c, k); m->btree_id = btree_id; diff --git a/fs/bcachefs/errcode.h b/fs/bcachefs/errcode.h index af25d8ec60f2..01a79fa3eacb 100644 --- a/fs/bcachefs/errcode.h +++ b/fs/bcachefs/errcode.h @@ -252,7 +252,8 @@ x(BCH_ERR_nopromote, nopromote_in_flight) \ x(BCH_ERR_nopromote, nopromote_no_writes) \ x(BCH_ERR_nopromote, nopromote_enomem) \ - x(0, need_inode_lock) + x(0, need_inode_lock) \ + x(0, invalid_snapshot_node) enum bch_errcode { BCH_ERR_START = 2048, diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 9cd71e613dc9..4e074136c490 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -93,8 +93,10 @@ static int bch2_snapshot_tree_create(struct btree_trans *trans, static bool __bch2_snapshot_is_ancestor_early(struct snapshot_table *t, u32 id, u32 ancestor) { - while (id && id < ancestor) - id = __snapshot_t(t, id)->parent; + while (id && id < ancestor) { + const struct snapshot_t *s = __snapshot_t(t, id); + id = s ? s->parent : 0; + } return id == ancestor; } @@ -110,6 +112,8 @@ static bool bch2_snapshot_is_ancestor_early(struct bch_fs *c, u32 id, u32 ancest static inline u32 get_ancestor_below(struct snapshot_table *t, u32 id, u32 ancestor) { const struct snapshot_t *s = __snapshot_t(t, id); + if (!s) + return 0; if (s->skip[2] <= ancestor) return s->skip[2]; diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index 8538b7fddfed..331f20fd8d03 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -48,7 +48,8 @@ static inline const struct snapshot_t *snapshot_t(struct bch_fs *c, u32 id) static inline u32 bch2_snapshot_tree(struct bch_fs *c, u32 id) { rcu_read_lock(); - id = snapshot_t(c, id)->tree; + const struct snapshot_t *s = snapshot_t(c, id); + id = s ? s->tree : 0; rcu_read_unlock(); return id; @@ -56,7 +57,8 @@ static inline u32 bch2_snapshot_tree(struct bch_fs *c, u32 id) static inline u32 __bch2_snapshot_parent_early(struct bch_fs *c, u32 id) { - return snapshot_t(c, id)->parent; + const struct snapshot_t *s = snapshot_t(c, id); + return s ? s->parent : 0; } static inline u32 bch2_snapshot_parent_early(struct bch_fs *c, u32 id) @@ -70,19 +72,19 @@ static inline u32 bch2_snapshot_parent_early(struct bch_fs *c, u32 id) static inline u32 __bch2_snapshot_parent(struct bch_fs *c, u32 id) { -#ifdef CONFIG_BCACHEFS_DEBUG - u32 parent = snapshot_t(c, id)->parent; + const struct snapshot_t *s = snapshot_t(c, id); + if (!s) + return 0; - if (parent && - snapshot_t(c, id)->depth != snapshot_t(c, parent)->depth + 1) + u32 parent = s->parent; + if (IS_ENABLED(CONFIG_BCACHEFS_DEBU) && + parent && + s->depth != snapshot_t(c, parent)->depth + 1) panic("id %u depth=%u parent %u depth=%u\n", id, snapshot_t(c, id)->depth, parent, snapshot_t(c, parent)->depth); return parent; -#else - return snapshot_t(c, id)->parent; -#endif } static inline u32 bch2_snapshot_parent(struct bch_fs *c, u32 id) @@ -120,7 +122,8 @@ static inline u32 bch2_snapshot_root(struct bch_fs *c, u32 id) static inline u32 __bch2_snapshot_equiv(struct bch_fs *c, u32 id) { - return snapshot_t(c, id)->equiv; + const struct snapshot_t *s = snapshot_t(c, id); + return s ? s->equiv : 0; } static inline u32 bch2_snapshot_equiv(struct bch_fs *c, u32 id) @@ -137,38 +140,22 @@ static inline bool bch2_snapshot_is_equiv(struct bch_fs *c, u32 id) return id == bch2_snapshot_equiv(c, id); } -static inline bool bch2_snapshot_is_internal_node(struct bch_fs *c, u32 id) +static inline int bch2_snapshot_is_internal_node(struct bch_fs *c, u32 id) { - const struct snapshot_t *s; - bool ret; - rcu_read_lock(); - s = snapshot_t(c, id); - ret = s->children[0]; + const struct snapshot_t *s = snapshot_t(c, id); + int ret = s ? s->children[0] : -BCH_ERR_invalid_snapshot_node; rcu_read_unlock(); return ret; } -static inline u32 bch2_snapshot_is_leaf(struct bch_fs *c, u32 id) +static inline int bch2_snapshot_is_leaf(struct bch_fs *c, u32 id) { - return !bch2_snapshot_is_internal_node(c, id); -} - -static inline u32 bch2_snapshot_sibling(struct bch_fs *c, u32 id) -{ - const struct snapshot_t *s; - u32 parent = __bch2_snapshot_parent(c, id); - - if (!parent) - return 0; - - s = snapshot_t(c, __bch2_snapshot_parent(c, id)); - if (id == s->children[0]) - return s->children[1]; - if (id == s->children[1]) - return s->children[0]; - return 0; + int ret = bch2_snapshot_is_internal_node(c, id); + if (ret < 0) + return ret; + return !ret; } static inline u32 bch2_snapshot_depth(struct bch_fs *c, u32 parent) @@ -253,7 +240,7 @@ static inline int bch2_key_has_snapshot_overwrites(struct btree_trans *trans, struct bpos pos) { if (!btree_type_has_snapshots(id) || - bch2_snapshot_is_leaf(trans->c, pos.snapshot)) + bch2_snapshot_is_leaf(trans->c, pos.snapshot) > 0) return 0; return __bch2_key_has_snapshot_overwrites(trans, id, pos); -- cgit v1.2.3-59-g8ed1b From 57339b24a0eda5433751e7e0f4a8ea1e23315f60 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 23 Mar 2024 19:29:55 -0400 Subject: bcachefs: Don't do extent merging before journal replay is finished We don't normally do extent updates this early in recovery, but some of the repair paths have to and when we do, we don't want to do anything that requires the snapshots table. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_update.c b/fs/bcachefs/btree_update.c index a4b40c1656a5..8e47e260eba5 100644 --- a/fs/bcachefs/btree_update.c +++ b/fs/bcachefs/btree_update.c @@ -38,6 +38,9 @@ static noinline int extent_front_merge(struct btree_trans *trans, struct bkey_i *update; int ret; + if (unlikely(trans->journal_replay_not_finished)) + return 0; + update = bch2_bkey_make_mut_noupdate(trans, k); ret = PTR_ERR_OR_ZERO(update); if (ret) @@ -69,6 +72,9 @@ static noinline int extent_back_merge(struct btree_trans *trans, struct bch_fs *c = trans->c; int ret; + if (unlikely(trans->journal_replay_not_finished)) + return 0; + ret = bch2_key_has_snapshot_overwrites(trans, iter->btree_id, insert->k.p) ?: bch2_key_has_snapshot_overwrites(trans, iter->btree_id, k.k->p); if (ret < 0) -- cgit v1.2.3-59-g8ed1b From 36f9ef109b1c6935928d09a3e73d744291f71545 Mon Sep 17 00:00:00 2001 From: Hongbo Li Date: Mon, 25 Mar 2024 10:50:48 +0800 Subject: bcachefs: fix trans->mem realloc in __bch2_trans_kmalloc The old code doesn't consider the mem alloced from mempool when call krealloc on trans->mem. Also in bch2_trans_put, using mempool_free to free trans->mem by condition "trans->mem_bytes == BTREE_TRANS_MEM_MAX" is inaccurate when trans->mem was allocated by krealloc function. Instead, we use used_mempool stuff to record the situation, and realloc or free the trans->mem in elegant way. Also, after krealloc failed in __bch2_trans_kmalloc, the old data should be copied to the new buffer when alloc from mempool_alloc. Fixes: 31403dca5bb1 ("bcachefs: optimize __bch2_trans_get(), kill DEBUG_TRANSACTIONS") Signed-off-by: Hongbo Li Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 33 ++++++++++++++++++++++++++++++--- 1 file changed, 30 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 51bcdc6c6d1c..ae4a28a33ad1 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -2790,6 +2790,31 @@ void *__bch2_trans_kmalloc(struct btree_trans *trans, size_t size) struct btree_transaction_stats *s = btree_trans_stats(trans); s->max_mem = max(s->max_mem, new_bytes); + if (trans->used_mempool) { + if (trans->mem_bytes >= new_bytes) + goto out_change_top; + + /* No more space from mempool item, need malloc new one */ + new_mem = kmalloc(new_bytes, GFP_NOWAIT|__GFP_NOWARN); + if (unlikely(!new_mem)) { + bch2_trans_unlock(trans); + + new_mem = kmalloc(new_bytes, GFP_KERNEL); + if (!new_mem) + return ERR_PTR(-BCH_ERR_ENOMEM_trans_kmalloc); + + ret = bch2_trans_relock(trans); + if (ret) { + kfree(new_mem); + return ERR_PTR(ret); + } + } + memcpy(new_mem, trans->mem, trans->mem_top); + trans->used_mempool = false; + mempool_free(trans->mem, &c->btree_trans_mem_pool); + goto out_new_mem; + } + new_mem = krealloc(trans->mem, new_bytes, GFP_NOWAIT|__GFP_NOWARN); if (unlikely(!new_mem)) { bch2_trans_unlock(trans); @@ -2798,6 +2823,8 @@ void *__bch2_trans_kmalloc(struct btree_trans *trans, size_t size) if (!new_mem && new_bytes <= BTREE_TRANS_MEM_MAX) { new_mem = mempool_alloc(&c->btree_trans_mem_pool, GFP_KERNEL); new_bytes = BTREE_TRANS_MEM_MAX; + memcpy(new_mem, trans->mem, trans->mem_top); + trans->used_mempool = true; kfree(trans->mem); } @@ -2811,7 +2838,7 @@ void *__bch2_trans_kmalloc(struct btree_trans *trans, size_t size) if (ret) return ERR_PTR(ret); } - +out_new_mem: trans->mem = new_mem; trans->mem_bytes = new_bytes; @@ -2819,7 +2846,7 @@ void *__bch2_trans_kmalloc(struct btree_trans *trans, size_t size) trace_and_count(c, trans_restart_mem_realloced, trans, _RET_IP_, new_bytes); return ERR_PTR(btree_trans_restart(trans, BCH_ERR_transaction_restart_mem_realloced)); } - +out_change_top: p = trans->mem + trans->mem_top; trans->mem_top += size; memset(p, 0, size); @@ -3093,7 +3120,7 @@ void bch2_trans_put(struct btree_trans *trans) if (paths_allocated != trans->_paths_allocated) kvfree_rcu_mightsleep(paths_allocated); - if (trans->mem_bytes == BTREE_TRANS_MEM_MAX) + if (trans->used_mempool) mempool_free(trans->mem, &c->btree_trans_mem_pool); else kfree(trans->mem); -- cgit v1.2.3-59-g8ed1b From 048f47e83fc315499dc1943176b3ebe1a55574fb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 25 Mar 2024 13:51:29 -0400 Subject: bcachefs: btree_and_journal_iter now respects trans->journal_replay_not_finished btree_and_journal_iter is now safe to use at runtime, not just during recovery before journal keys have been freed. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_journal_iter.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index 50e04356d72c..0272ef0d7310 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -363,7 +363,7 @@ static void btree_and_journal_iter_prefetch(struct btree_and_journal_iter *_iter struct bkey_s_c bch2_btree_and_journal_iter_peek(struct btree_and_journal_iter *iter) { - struct bkey_s_c btree_k, journal_k, ret; + struct bkey_s_c btree_k, journal_k = bkey_s_c_null, ret; if (iter->prefetch && iter->journal.level) btree_and_journal_iter_prefetch(iter); @@ -375,9 +375,10 @@ again: bpos_lt(btree_k.k->p, iter->pos)) bch2_journal_iter_advance_btree(iter); - while ((journal_k = bch2_journal_iter_peek(&iter->journal)).k && - bpos_lt(journal_k.k->p, iter->pos)) - bch2_journal_iter_advance(&iter->journal); + if (iter->trans->journal_replay_not_finished) + while ((journal_k = bch2_journal_iter_peek(&iter->journal)).k && + bpos_lt(journal_k.k->p, iter->pos)) + bch2_journal_iter_advance(&iter->journal); ret = journal_k.k && (!btree_k.k || bpos_le(journal_k.k->p, btree_k.k->p)) @@ -435,7 +436,9 @@ void bch2_btree_and_journal_iter_init_node_iter(struct btree_trans *trans, bch2_btree_node_iter_init_from_start(&node_iter, b); __bch2_btree_and_journal_iter_init_node_iter(trans, iter, b, node_iter, b->data->min_key); - list_add(&iter->journal.list, &trans->c->journal_iters); + if (trans->journal_replay_not_finished && + !test_bit(BCH_FS_may_go_rw, &trans->c->flags)) + list_add(&iter->journal.list, &trans->c->journal_iters); } /* sort and dedup all keys in the journal: */ -- cgit v1.2.3-59-g8ed1b From 40cb26233a060aeb936de7ea1f6ac2659ed9951c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 26 Mar 2024 17:14:43 -0400 Subject: bcachefs: Be careful about btree node splits during journal replay Don't pick a pivot that's going to be deleted. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_journal_iter.c | 16 ++++++++++++++++ fs/bcachefs/btree_journal_iter.h | 4 ++-- fs/bcachefs/btree_update_interior.c | 9 ++++++++- 3 files changed, 26 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index 0272ef0d7310..1f588264575d 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -261,6 +261,22 @@ int bch2_journal_key_delete(struct bch_fs *c, enum btree_id id, return bch2_journal_key_insert(c, id, level, &whiteout); } +bool bch2_key_deleted_in_journal(struct btree_trans *trans, enum btree_id btree, + unsigned level, struct bpos pos) +{ + struct journal_keys *keys = &trans->c->journal_keys; + size_t idx = bch2_journal_key_search(keys, btree, level, pos); + + if (!trans->journal_replay_not_finished) + return false; + + return (idx < keys->size && + keys->data[idx].btree_id == btree && + keys->data[idx].level == level && + bpos_eq(keys->data[idx].k->k.p, pos) && + bkey_deleted(&keys->data[idx].k->k)); +} + void bch2_journal_key_overwritten(struct bch_fs *c, enum btree_id btree, unsigned level, struct bpos pos) { diff --git a/fs/bcachefs/btree_journal_iter.h b/fs/bcachefs/btree_journal_iter.h index c9d19da3ea04..09cd53091a05 100644 --- a/fs/bcachefs/btree_journal_iter.h +++ b/fs/bcachefs/btree_journal_iter.h @@ -40,8 +40,8 @@ int bch2_journal_key_insert(struct bch_fs *, enum btree_id, unsigned, struct bkey_i *); int bch2_journal_key_delete(struct bch_fs *, enum btree_id, unsigned, struct bpos); -void bch2_journal_key_overwritten(struct bch_fs *, enum btree_id, - unsigned, struct bpos); +bool bch2_key_deleted_in_journal(struct btree_trans *, enum btree_id, unsigned, struct bpos); +void bch2_journal_key_overwritten(struct bch_fs *, enum btree_id, unsigned, struct bpos); void bch2_btree_and_journal_iter_advance(struct btree_and_journal_iter *); struct bkey_s_c bch2_btree_and_journal_iter_peek(struct btree_and_journal_iter *); diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index b2f5f2e50f7e..b45c7ced15d8 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1380,9 +1380,16 @@ static void __btree_split_node(struct btree_update *as, if (bkey_deleted(k)) continue; + uk = bkey_unpack_key(b, k); + + if (b->c.level && + u64s < n1_u64s && + u64s + k->u64s >= n1_u64s && + bch2_key_deleted_in_journal(trans, b->c.btree_id, b->c.level, uk.p)) + n1_u64s += k->u64s; + i = u64s >= n1_u64s; u64s += k->u64s; - uk = bkey_unpack_key(b, k); if (!i) n1_pos = uk.p; bch2_bkey_format_add_key(&format[i], &uk); -- cgit v1.2.3-59-g8ed1b From 79032b078173f87a13f8618cdab710798be67314 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 23 Mar 2024 19:29:19 -0400 Subject: bcachefs: Improved topology repair checks Consolidate bch2_gc_check_topology() and btree_node_interior_verify(), and replace them with an improved version, bch2_btree_node_check_topology(). This checks that children of an interior node correctly span the full range of the parent node with no overlaps. Also, ensure that topology repairs at runtime are always a fatal error; in particular, this adds a check in btree_iter_down() - if we don't find a key while walking down the btree that's indicative of a topology error and should be flagged as such, not a null ptr deref. Some checks in btree_update_interior.c remaining BUG_ONS(), because we already checked the node for topology errors when starting the update, and the assertions indicate that we _just_ corrupted the btree node - i.e. the problem can't be that existing on disk corruption, they indicate an actual algorithmic bug. In the future, we'll be annotating the fsck errors list with which recovery pass corrects them; the open coded "run explicit recovery pass or fatal error" in bch2_btree_node_check_topology() will in the future be done for every fsck_err() call. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 3 +- fs/bcachefs/btree_gc.c | 133 ++++------------------------------- fs/bcachefs/btree_io.c | 2 +- fs/bcachefs/btree_iter.c | 19 ++++- fs/bcachefs/btree_update_interior.c | 134 +++++++++++++++++++++++++----------- fs/bcachefs/btree_update_interior.h | 2 + fs/bcachefs/error.h | 6 ++ fs/bcachefs/sb-errors_types.h | 3 +- 8 files changed, 137 insertions(+), 165 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 8ff21b2e1463..84474324dba9 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -808,7 +808,8 @@ static noinline void btree_bad_header(struct bch_fs *c, struct btree *b) prt_printf(&buf, "\nmax "); bch2_bpos_to_text(&buf, b->data->max_key); - bch2_fs_inconsistent(c, "%s", buf.buf); + bch2_fs_topology_error(c, "%s", buf.buf); + printbuf_exit(&buf); } diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index bdaed29f084a..6d32ac9be243 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -70,90 +70,6 @@ static inline void gc_pos_set(struct bch_fs *c, struct gc_pos new_pos) __gc_pos_set(c, new_pos); } -/* - * Missing: if an interior btree node is empty, we need to do something - - * perhaps just kill it - */ -static int bch2_gc_check_topology(struct bch_fs *c, - struct btree *b, - struct bkey_buf *prev, - struct bkey_buf cur, - bool is_last) -{ - struct bpos node_start = b->data->min_key; - struct bpos node_end = b->data->max_key; - struct bpos expected_start = bkey_deleted(&prev->k->k) - ? node_start - : bpos_successor(prev->k->k.p); - struct printbuf buf1 = PRINTBUF, buf2 = PRINTBUF; - int ret = 0; - - if (cur.k->k.type == KEY_TYPE_btree_ptr_v2) { - struct bkey_i_btree_ptr_v2 *bp = bkey_i_to_btree_ptr_v2(cur.k); - - if (!bpos_eq(expected_start, bp->v.min_key)) { - bch2_topology_error(c); - - if (bkey_deleted(&prev->k->k)) { - prt_printf(&buf1, "start of node: "); - bch2_bpos_to_text(&buf1, node_start); - } else { - bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(prev->k)); - } - bch2_bkey_val_to_text(&buf2, c, bkey_i_to_s_c(cur.k)); - - if (__fsck_err(c, - FSCK_CAN_FIX| - FSCK_CAN_IGNORE| - FSCK_NO_RATELIMIT, - btree_node_topology_bad_min_key, - "btree node with incorrect min_key at btree %s level %u:\n" - " prev %s\n" - " cur %s", - bch2_btree_id_str(b->c.btree_id), b->c.level, - buf1.buf, buf2.buf) && should_restart_for_topology_repair(c)) { - bch_info(c, "Halting mark and sweep to start topology repair pass"); - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); - goto err; - } else { - set_bit(BCH_FS_initial_gc_unfixed, &c->flags); - } - } - } - - if (is_last && !bpos_eq(cur.k->k.p, node_end)) { - bch2_topology_error(c); - - printbuf_reset(&buf1); - printbuf_reset(&buf2); - - bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(cur.k)); - bch2_bpos_to_text(&buf2, node_end); - - if (__fsck_err(c, FSCK_CAN_FIX|FSCK_CAN_IGNORE|FSCK_NO_RATELIMIT, - btree_node_topology_bad_max_key, - "btree node with incorrect max_key at btree %s level %u:\n" - " %s\n" - " expected %s", - bch2_btree_id_str(b->c.btree_id), b->c.level, - buf1.buf, buf2.buf) && - should_restart_for_topology_repair(c)) { - bch_info(c, "Halting mark and sweep to start topology repair pass"); - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); - goto err; - } else { - set_bit(BCH_FS_initial_gc_unfixed, &c->flags); - } - } - - bch2_bkey_buf_copy(prev, c, cur.k); -err: -fsck_err: - printbuf_exit(&buf2); - printbuf_exit(&buf1); - return ret; -} - static void btree_ptr_to_v2(struct btree *b, struct bkey_i_btree_ptr_v2 *dst) { switch (b->key.k.type) { @@ -445,6 +361,7 @@ again: prev = NULL; if (ret == DROP_PREV_NODE) { + bch_info(c, "dropped prev node"); bch2_btree_node_evict(trans, prev_k.k); ret = bch2_journal_key_delete(c, b->c.btree_id, b->c.level, prev_k.k->k.p); @@ -841,42 +758,30 @@ err: static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool initial) { - struct bch_fs *c = trans->c; struct btree_node_iter iter; struct bkey unpacked; struct bkey_s_c k; - struct bkey_buf prev, cur; int ret = 0; + ret = bch2_btree_node_check_topology(trans, b); + if (ret) + return ret; + if (!btree_node_type_needs_gc(btree_node_type(b))) return 0; bch2_btree_node_iter_init_from_start(&iter, b); - bch2_bkey_buf_init(&prev); - bch2_bkey_buf_init(&cur); - bkey_init(&prev.k->k); while ((k = bch2_btree_node_iter_peek_unpack(&iter, b, &unpacked)).k) { ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false, &k, initial); if (ret) - break; + return ret; bch2_btree_node_iter_advance(&iter, b); - - if (b->c.level) { - bch2_bkey_buf_reassemble(&cur, c, k); - - ret = bch2_gc_check_topology(c, b, &prev, cur, - bch2_btree_node_iter_end(&iter)); - if (ret) - break; - } } - bch2_bkey_buf_exit(&cur, c); - bch2_bkey_buf_exit(&prev, c); - return ret; + return 0; } static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id, @@ -925,14 +830,16 @@ static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b struct bch_fs *c = trans->c; struct btree_and_journal_iter iter; struct bkey_s_c k; - struct bkey_buf cur, prev; + struct bkey_buf cur; struct printbuf buf = PRINTBUF; int ret = 0; + ret = bch2_btree_node_check_topology(trans, b); + if (ret) + return ret; + bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); - bch2_bkey_buf_init(&prev); bch2_bkey_buf_init(&cur); - bkey_init(&prev.k->k); while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { BUG_ON(bpos_lt(k.k->p, b->data->min_key)); @@ -943,20 +850,7 @@ static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b if (ret) goto fsck_err; - if (b->c.level) { - bch2_bkey_buf_reassemble(&cur, c, k); - k = bkey_i_to_s_c(cur.k); - - bch2_btree_and_journal_iter_advance(&iter); - - ret = bch2_gc_check_topology(c, b, - &prev, cur, - !bch2_btree_and_journal_iter_peek(&iter).k); - if (ret) - goto fsck_err; - } else { - bch2_btree_and_journal_iter_advance(&iter); - } + bch2_btree_and_journal_iter_advance(&iter); } if (b->c.level > target_depth) { @@ -1015,7 +909,6 @@ static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b } fsck_err: bch2_bkey_buf_exit(&cur, c); - bch2_bkey_buf_exit(&prev, c); bch2_btree_and_journal_iter_exit(&iter); printbuf_exit(&buf); return ret; diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 34df8ccc5fec..b85e2cd61c38 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1657,7 +1657,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, prt_str(&buf, "btree node read error: no device to read from\n at "); bch2_btree_pos_to_text(&buf, c, b); - bch_err(c, "%s", buf.buf); + bch_err_ratelimited(c, "%s", buf.buf); if (c->recovery_passes_explicit & BIT_ULL(BCH_RECOVERY_PASS_check_topology) && c->curr_recovery_pass > BCH_RECOVERY_PASS_check_topology) diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index ae4a28a33ad1..2a211a4bebd1 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -927,8 +927,22 @@ static __always_inline int btree_path_down(struct btree_trans *trans, if (ret) goto err; } else { - bch2_bkey_buf_unpack(&tmp, c, l->b, - bch2_btree_node_iter_peek(&l->iter, l->b)); + struct bkey_packed *k = bch2_btree_node_iter_peek(&l->iter, l->b); + if (!k) { + struct printbuf buf = PRINTBUF; + + prt_str(&buf, "node not found at pos "); + bch2_bpos_to_text(&buf, path->pos); + prt_str(&buf, " within parent node "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&l->b->key)); + + bch2_fs_fatal_error(c, "%s", buf.buf); + printbuf_exit(&buf); + ret = -BCH_ERR_btree_need_topology_repair; + goto err; + } + + bch2_bkey_buf_unpack(&tmp, c, l->b, k); if ((flags & BTREE_ITER_PREFETCH) && c->opts.btree_node_prefetch) { @@ -962,7 +976,6 @@ err: return ret; } - static int bch2_btree_path_traverse_all(struct btree_trans *trans) { struct bch_fs *c = trans->c; diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index b45c7ced15d8..aae7a2687eee 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -2,6 +2,7 @@ #include "bcachefs.h" #include "alloc_foreground.h" +#include "bkey_buf.h" #include "bkey_methods.h" #include "btree_cache.h" #include "btree_gc.h" @@ -18,6 +19,7 @@ #include "journal.h" #include "journal_reclaim.h" #include "keylist.h" +#include "recovery.h" #include "replicas.h" #include "super-io.h" #include "trace.h" @@ -44,56 +46,103 @@ static btree_path_idx_t get_unlocked_mut_path(struct btree_trans *trans, return path_idx; } -/* Debug code: */ - /* * Verify that child nodes correctly span parent node's range: */ -static void btree_node_interior_verify(struct bch_fs *c, struct btree *b) +int bch2_btree_node_check_topology(struct btree_trans *trans, struct btree *b) { -#ifdef CONFIG_BCACHEFS_DEBUG - struct bpos next_node = b->data->min_key; - struct btree_node_iter iter; + struct bch_fs *c = trans->c; + struct bpos node_min = b->key.k.type == KEY_TYPE_btree_ptr_v2 + ? bkey_i_to_btree_ptr_v2(&b->key)->v.min_key + : b->data->min_key; + struct btree_and_journal_iter iter; struct bkey_s_c k; - struct bkey_s_c_btree_ptr_v2 bp; - struct bkey unpacked; - struct printbuf buf1 = PRINTBUF, buf2 = PRINTBUF; + struct printbuf buf = PRINTBUF; + struct bkey_buf prev; + int ret = 0; - BUG_ON(!b->c.level); + BUG_ON(b->key.k.type == KEY_TYPE_btree_ptr_v2 && + !bpos_eq(bkey_i_to_btree_ptr_v2(&b->key)->v.min_key, + b->data->min_key)); - if (!test_bit(JOURNAL_REPLAY_DONE, &c->journal.flags)) - return; + if (!b->c.level) + return 0; - bch2_btree_node_iter_init_from_start(&iter, b); + bch2_bkey_buf_init(&prev); + bkey_init(&prev.k->k); + bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); - while (1) { - k = bch2_btree_node_iter_peek_unpack(&iter, b, &unpacked); + while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { if (k.k->type != KEY_TYPE_btree_ptr_v2) - break; - bp = bkey_s_c_to_btree_ptr_v2(k); + goto out; - if (!bpos_eq(next_node, bp.v->min_key)) { - bch2_dump_btree_node(c, b); - bch2_bpos_to_text(&buf1, next_node); - bch2_bpos_to_text(&buf2, bp.v->min_key); - panic("expected next min_key %s got %s\n", buf1.buf, buf2.buf); - } + struct bkey_s_c_btree_ptr_v2 bp = bkey_s_c_to_btree_ptr_v2(k); - bch2_btree_node_iter_advance(&iter, b); + struct bpos expected_min = bkey_deleted(&prev.k->k) + ? node_min + : bpos_successor(prev.k->k.p); - if (bch2_btree_node_iter_end(&iter)) { - if (!bpos_eq(k.k->p, b->key.k.p)) { - bch2_dump_btree_node(c, b); - bch2_bpos_to_text(&buf1, b->key.k.p); - bch2_bpos_to_text(&buf2, k.k->p); - panic("expected end %s got %s\n", buf1.buf, buf2.buf); - } - break; + if (!bpos_eq(expected_min, bp.v->min_key)) { + bch2_topology_error(c); + + printbuf_reset(&buf); + prt_str(&buf, "end of prev node doesn't match start of next node\n"), + prt_printf(&buf, " in btree %s level %u node ", + bch2_btree_id_str(b->c.btree_id), b->c.level); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + prt_str(&buf, "\n prev "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(prev.k)); + prt_str(&buf, "\n next "); + bch2_bkey_val_to_text(&buf, c, k); + + need_fsck_err(c, btree_node_topology_bad_min_key, "%s", buf.buf); + goto topology_repair; } - next_node = bpos_successor(k.k->p); + bch2_bkey_buf_reassemble(&prev, c, k); + bch2_btree_and_journal_iter_advance(&iter); + } + + if (bkey_deleted(&prev.k->k)) { + bch2_topology_error(c); + + printbuf_reset(&buf); + prt_str(&buf, "empty interior node\n"); + prt_printf(&buf, " in btree %s level %u node ", + bch2_btree_id_str(b->c.btree_id), b->c.level); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + + need_fsck_err(c, btree_node_topology_empty_interior_node, "%s", buf.buf); + goto topology_repair; + } else if (!bpos_eq(prev.k->k.p, b->key.k.p)) { + bch2_topology_error(c); + + printbuf_reset(&buf); + prt_str(&buf, "last child node doesn't end at end of parent node\n"); + prt_printf(&buf, " in btree %s level %u node ", + bch2_btree_id_str(b->c.btree_id), b->c.level); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + prt_str(&buf, "\n last key "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(prev.k)); + + need_fsck_err(c, btree_node_topology_bad_max_key, "%s", buf.buf); + goto topology_repair; + } +out: +fsck_err: + bch2_btree_and_journal_iter_exit(&iter); + bch2_bkey_buf_exit(&prev, c); + printbuf_exit(&buf); + return ret; +topology_repair: + if ((c->recovery_passes_explicit & BIT_ULL(BCH_RECOVERY_PASS_check_topology)) && + c->curr_recovery_pass > BCH_RECOVERY_PASS_check_topology) { + bch2_inconsistent_error(c); + ret = -BCH_ERR_btree_need_topology_repair; + } else { + ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); } -#endif + goto out; } /* Calculate ideal packed bkey format for new btree nodes: */ @@ -1448,8 +1497,7 @@ static void __btree_split_node(struct btree_update *as, bch2_verify_btree_nr_keys(n[i]); - if (b->c.level) - btree_node_interior_verify(as->c, n[i]); + BUG_ON(bch2_btree_node_check_topology(trans, n[i])); } } @@ -1480,7 +1528,7 @@ static void btree_split_insert_keys(struct btree_update *as, __bch2_btree_insert_keys_interior(as, trans, path, b, node_iter, keys); - btree_node_interior_verify(as->c, b); + BUG_ON(bch2_btree_node_check_topology(trans, b)); } } @@ -1498,6 +1546,10 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans, BUG_ON(!parent && (b != btree_node_root(c, b))); BUG_ON(parent && !btree_node_intent_locked(trans->paths + path, b->c.level + 1)); + ret = bch2_btree_node_check_topology(trans, b); + if (ret) + return ret; + bch2_btree_interior_update_will_free_node(as, b); if (b->nr.live_u64s > BTREE_SPLIT_THRESHOLD(c)) { @@ -1717,7 +1769,11 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t goto split; } - btree_node_interior_verify(c, b); + ret = bch2_btree_node_check_topology(trans, b); + if (ret) { + bch2_btree_node_unlock_write(trans, path, b); + return ret; + } bch2_btree_insert_keys_interior(as, trans, path, b, keys); @@ -1735,7 +1791,7 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t bch2_btree_node_unlock_write(trans, path, b); - btree_node_interior_verify(c, b); + BUG_ON(bch2_btree_node_check_topology(trans, b)); return 0; split: /* diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index f651dd48aaa0..7ac3e17b84fd 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -10,6 +10,8 @@ #define BTREE_UPDATE_JOURNAL_RES (BTREE_UPDATE_NODES_MAX * (BKEY_BTREE_PTR_U64s_MAX + 1)) +int bch2_btree_node_check_topology(struct btree_trans *, struct btree *); + /* * Tracks an in progress split/rewrite of a btree node and the update to the * parent node: diff --git a/fs/bcachefs/error.h b/fs/bcachefs/error.h index ae1d6674c512..36caedf72d89 100644 --- a/fs/bcachefs/error.h +++ b/fs/bcachefs/error.h @@ -32,6 +32,12 @@ bool bch2_inconsistent_error(struct bch_fs *); int bch2_topology_error(struct bch_fs *); +#define bch2_fs_topology_error(c, ...) \ +({ \ + bch_err(c, "btree topology error: " __VA_ARGS__); \ + bch2_topology_error(c); \ +}) + #define bch2_fs_inconsistent(c, ...) \ ({ \ bch_err(c, __VA_ARGS__); \ diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 5178bf579f7c..8edbd9ef994c 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -265,7 +265,8 @@ x(subvol_children_bad, 257) \ x(subvol_loop, 258) \ x(subvol_unreachable, 259) \ - x(btree_node_bkey_bad_u64s, 260) + x(btree_node_bkey_bad_u64s, 260) \ + x(btree_node_topology_empty_interior_node, 261) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, -- cgit v1.2.3-59-g8ed1b From bb66009958b277a9baffaa53d835661852550151 Mon Sep 17 00:00:00 2001 From: zhuxiaohui Date: Tue, 26 Mar 2024 20:03:45 +0800 Subject: bcachefs: add REQ_SYNC and REQ_IDLE in write dio when writing file with direct_IO on bcachefs, then performance is much lower than other fs due to write back throttle in block layer: wbt_wait+1 __rq_qos_throttle+32 blk_mq_submit_bio+394 submit_bio_noacct_nocheck+649 bch2_submit_wbio_replicas+538 __bch2_write+2539 bch2_direct_write+1663 bch2_write_iter+318 aio_write+355 io_submit_one+1224 __x64_sys_io_submit+169 do_syscall_64+134 entry_SYSCALL_64_after_hwframe+110 add set REQ_SYNC and REQ_IDLE in bio->bi_opf as standard dirct-io Signed-off-by: zhuxiaohui Signed-off-by: Kent Overstreet --- fs/bcachefs/fs-io-direct.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs-io-direct.c b/fs/bcachefs/fs-io-direct.c index 33cb6da3a5ad..f49e6c0f0f68 100644 --- a/fs/bcachefs/fs-io-direct.c +++ b/fs/bcachefs/fs-io-direct.c @@ -536,7 +536,7 @@ static __always_inline long bch2_dio_write_loop(struct dio_write *dio) if (likely(!dio->iter.count) || dio->op.error) break; - bio_reset(bio, NULL, REQ_OP_WRITE); + bio_reset(bio, NULL, REQ_OP_WRITE | REQ_SYNC | REQ_IDLE); } out: return bch2_dio_write_done(dio); @@ -618,7 +618,7 @@ ssize_t bch2_direct_write(struct kiocb *req, struct iov_iter *iter) bio = bio_alloc_bioset(NULL, bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS), - REQ_OP_WRITE, + REQ_OP_WRITE | REQ_SYNC | REQ_IDLE, GFP_KERNEL, &c->dio_write_bioset); dio = container_of(bio, struct dio_write, op.wbio.bio); -- cgit v1.2.3-59-g8ed1b From 805b535a8afbcd8073a03eb25aafd82cb816bff6 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 25 Mar 2024 21:58:07 -0400 Subject: bcachefs: Check btree ptr min_key in .invalid Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 9 +++++++-- fs/bcachefs/sb-errors_types.h | 3 ++- 2 files changed, 9 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 61395b113df9..b2432d88cda6 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -189,13 +189,18 @@ int bch2_btree_ptr_v2_invalid(struct bch_fs *c, struct bkey_s_c k, enum bkey_invalid_flags flags, struct printbuf *err) { + struct bkey_s_c_btree_ptr_v2 bp = bkey_s_c_to_btree_ptr_v2(k); int ret = 0; - bkey_fsck_err_on(bkey_val_u64s(k.k) > BKEY_BTREE_PTR_VAL_U64s_MAX, c, err, - btree_ptr_v2_val_too_big, + bkey_fsck_err_on(bkey_val_u64s(k.k) > BKEY_BTREE_PTR_VAL_U64s_MAX, + c, err, btree_ptr_v2_val_too_big, "value too big (%zu > %zu)", bkey_val_u64s(k.k), BKEY_BTREE_PTR_VAL_U64s_MAX); + bkey_fsck_err_on(bpos_ge(bp.v->min_key, bp.k->p), + c, err, btree_ptr_v2_min_key_bad, + "min_key > key"); + ret = bch2_bkey_ptrs_invalid(c, k, flags, err); fsck_err: return ret; diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 8edbd9ef994c..73e9634df8ff 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -266,7 +266,8 @@ x(subvol_loop, 258) \ x(subvol_unreachable, 259) \ x(btree_node_bkey_bad_u64s, 260) \ - x(btree_node_topology_empty_interior_node, 261) + x(btree_node_topology_empty_interior_node, 261) \ + x(btree_ptr_v2_min_key_bad, 262) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, -- cgit v1.2.3-59-g8ed1b From 812a9297936a959c98a2e9e44a9a622bbe30b162 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 26 Mar 2024 16:39:54 -0400 Subject: bcachefs: Fix btree node keys accounting in topology repair path When dropping keys now outside a now because we're changing the node min/max, we need to redo the node's accounting as well. Signed-off-by: Kent Overstreet --- fs/bcachefs/bset.c | 14 ++++++++++---- fs/bcachefs/bset.h | 2 ++ fs/bcachefs/btree_io.c | 1 + fs/bcachefs/btree_update_interior.c | 1 + 4 files changed, 14 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bset.c b/fs/bcachefs/bset.c index 3fd1085b6c61..3bb477840eab 100644 --- a/fs/bcachefs/bset.c +++ b/fs/bcachefs/bset.c @@ -134,18 +134,24 @@ void bch2_dump_btree_node_iter(struct btree *b, printbuf_exit(&buf); } -#ifdef CONFIG_BCACHEFS_DEBUG - -void __bch2_verify_btree_nr_keys(struct btree *b) +struct btree_nr_keys bch2_btree_node_count_keys(struct btree *b) { struct bset_tree *t; struct bkey_packed *k; - struct btree_nr_keys nr = { 0 }; + struct btree_nr_keys nr = {}; for_each_bset(b, t) bset_tree_for_each_key(b, t, k) if (!bkey_deleted(k)) btree_keys_account_key_add(&nr, t - b->set, k); + return nr; +} + +#ifdef CONFIG_BCACHEFS_DEBUG + +void __bch2_verify_btree_nr_keys(struct btree *b) +{ + struct btree_nr_keys nr = bch2_btree_node_count_keys(b); BUG_ON(memcmp(&nr, &b->nr, sizeof(nr))); } diff --git a/fs/bcachefs/bset.h b/fs/bcachefs/bset.h index 79c77baaa383..120a79fd456b 100644 --- a/fs/bcachefs/bset.h +++ b/fs/bcachefs/bset.h @@ -458,6 +458,8 @@ struct bkey_s_c bch2_btree_node_iter_peek_unpack(struct btree_node_iter *, /* Accounting: */ +struct btree_nr_keys bch2_btree_node_count_keys(struct btree *); + static inline void btree_keys_account_key(struct btree_nr_keys *n, unsigned bset, struct bkey_packed *k, diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index b85e2cd61c38..9c71e6fb9c41 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -654,6 +654,7 @@ void bch2_btree_node_drop_keys_outside_node(struct btree *b) */ bch2_bset_set_no_aux_tree(b, b->set); bch2_btree_build_aux_trees(b); + b->nr = bch2_btree_node_count_keys(b); struct bkey_s_c k; struct bkey unpacked; diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index aae7a2687eee..826cebe4bce0 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1543,6 +1543,7 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans, u64 start_time = local_clock(); int ret = 0; + bch2_verify_btree_nr_keys(b); BUG_ON(!parent && (b != btree_node_root(c, b))); BUG_ON(parent && !btree_node_intent_locked(trans->paths + path, b->c.level + 1)); -- cgit v1.2.3-59-g8ed1b From 6f5869ffd9f111b81b95b73c6e54f07406591911 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 26 Mar 2024 17:38:22 -0400 Subject: bcachefs: Fix use after free in bch2_check_fix_ptrs() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 6d32ac9be243..26e51af9acdc 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -691,12 +691,6 @@ found: } } - ret = bch2_journal_key_insert_take(c, btree_id, level, new); - if (ret) { - kfree(new); - goto err; - } - if (level) bch2_btree_node_update_key_early(trans, btree_id, level - 1, *k, new); @@ -710,6 +704,12 @@ found: bch_info(c, "new key %s", buf.buf); } + ret = bch2_journal_key_insert_take(c, btree_id, level, new); + if (ret) { + kfree(new); + goto err; + } + *k = bkey_i_to_s_c(new); } err: -- cgit v1.2.3-59-g8ed1b From 83bb58539045b15653b61c6e8eb65f3f9c671cdf Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 26 Mar 2024 18:46:20 -0400 Subject: bcachefs: Fix repair path for missing indirect extents Signed-off-by: Kent Overstreet --- fs/bcachefs/reflink.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index c47c66c2b394..ff7864731a07 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -185,8 +185,7 @@ not_found: } else { bkey_error_init(update); update->k.p = p.k->p; - update->k.p.offset = next_idx; - update->k.size = next_idx - *idx; + update->k.size = p.k->size; set_bkey_val_u64s(&update->k, 0); } -- cgit v1.2.3-59-g8ed1b From dcc1c04587aa9bc3515153f4c89cff73f2cb45b2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 26 Mar 2024 18:46:38 -0400 Subject: bcachefs: Fix use after free in check_root_trans() Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 47d4eefaba7b..6d8367ab5ddd 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -2098,17 +2098,21 @@ static int check_root_trans(struct btree_trans *trans) if (mustfix_fsck_err_on(ret, c, root_subvol_missing, "root subvol missing")) { - struct bkey_i_subvolume root_subvol; + struct bkey_i_subvolume *root_subvol = + bch2_trans_kmalloc(trans, sizeof(*root_subvol)); + ret = PTR_ERR_OR_ZERO(root_subvol); + if (ret) + goto err; snapshot = U32_MAX; inum = BCACHEFS_ROOT_INO; - bkey_subvolume_init(&root_subvol.k_i); - root_subvol.k.p.offset = BCACHEFS_ROOT_SUBVOL; - root_subvol.v.flags = 0; - root_subvol.v.snapshot = cpu_to_le32(snapshot); - root_subvol.v.inode = cpu_to_le64(inum); - ret = bch2_btree_insert_trans(trans, BTREE_ID_subvolumes, &root_subvol.k_i, 0); + bkey_subvolume_init(&root_subvol->k_i); + root_subvol->k.p.offset = BCACHEFS_ROOT_SUBVOL; + root_subvol->v.flags = 0; + root_subvol->v.snapshot = cpu_to_le32(snapshot); + root_subvol->v.inode = cpu_to_le64(inum); + ret = bch2_btree_insert_trans(trans, BTREE_ID_subvolumes, &root_subvol->k_i, 0); bch_err_msg(c, ret, "writing root subvol"); if (ret) goto err; -- cgit v1.2.3-59-g8ed1b From 47d2080e30b0b9fc636eba4e74f9e4bdc01543d7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 25 Mar 2024 19:26:05 -0400 Subject: bcachefs: Kill bch2_bkey_ptr_data_type() Remove some duplication, and inconsistency between check_fix_ptrs and the main ptr marking paths Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 6 ++--- fs/bcachefs/backpointers.h | 32 ++++++++++++++++++++----- fs/bcachefs/btree_gc.c | 59 +++++++++++++++++++++++++--------------------- fs/bcachefs/buckets.c | 12 ++++++---- fs/bcachefs/extents.h | 24 ------------------- 5 files changed, 67 insertions(+), 66 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 21b3ae7df824..7537d78aa9db 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -29,8 +29,7 @@ static bool extent_matches_bp(struct bch_fs *c, if (p.ptr.cached) continue; - bch2_extent_ptr_to_bp(c, btree_id, level, k, p, - &bucket2, &bp2); + bch2_extent_ptr_to_bp(c, btree_id, level, k, p, entry, &bucket2, &bp2); if (bpos_eq(bucket, bucket2) && !memcmp(&bp, &bp2, sizeof(bp))) return true; @@ -507,8 +506,7 @@ static int check_extent_to_backpointers(struct btree_trans *trans, if (p.ptr.cached) continue; - bch2_extent_ptr_to_bp(c, btree, level, - k, p, &bucket_pos, &bp); + bch2_extent_ptr_to_bp(c, btree, level, k, p, entry, &bucket_pos, &bp); ret = check_bp_exists(trans, s, bucket_pos, bp, k); if (ret) diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index 327365a9feac..da012ca7daee 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -90,20 +90,40 @@ static inline int bch2_bucket_backpointer_mod(struct btree_trans *trans, return bch2_trans_update_buffered(trans, BTREE_ID_backpointers, &bp_k.k_i); } -static inline enum bch_data_type bkey_ptr_data_type(enum btree_id btree_id, unsigned level, - struct bkey_s_c k, struct extent_ptr_decoded p) +static inline enum bch_data_type bch2_bkey_ptr_data_type(struct bkey_s_c k, + struct extent_ptr_decoded p, + const union bch_extent_entry *entry) { - return level ? BCH_DATA_btree : - p.has_ec ? BCH_DATA_stripe : - BCH_DATA_user; + switch (k.k->type) { + case KEY_TYPE_btree_ptr: + case KEY_TYPE_btree_ptr_v2: + return BCH_DATA_btree; + case KEY_TYPE_extent: + case KEY_TYPE_reflink_v: + return p.has_ec ? BCH_DATA_stripe : BCH_DATA_user; + case KEY_TYPE_stripe: { + const struct bch_extent_ptr *ptr = &entry->ptr; + struct bkey_s_c_stripe s = bkey_s_c_to_stripe(k); + + BUG_ON(ptr < s.v->ptrs || + ptr >= s.v->ptrs + s.v->nr_blocks); + + return ptr >= s.v->ptrs + s.v->nr_blocks - s.v->nr_redundant + ? BCH_DATA_parity + : BCH_DATA_user; + } + default: + BUG(); + } } static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, enum btree_id btree_id, unsigned level, struct bkey_s_c k, struct extent_ptr_decoded p, + const union bch_extent_entry *entry, struct bpos *bucket_pos, struct bch_backpointer *bp) { - enum bch_data_type data_type = bkey_ptr_data_type(btree_id, level, k, p); + enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry); s64 sectors = level ? btree_sectors(c) : k.k->size; u32 bucket_offset; diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 26e51af9acdc..0afefccf4e52 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -7,6 +7,7 @@ #include "bcachefs.h" #include "alloc_background.h" #include "alloc_foreground.h" +#include "backpointers.h" #include "bkey_methods.h" #include "bkey_buf.h" #include "btree_journal_iter.h" @@ -508,7 +509,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id bkey_for_each_ptr_decode(k->k, ptrs_c, p, entry_c) { struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev); struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); - enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, &entry_c->ptr); + enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, p, entry_c); if (fsck_err_on(!g->gen_valid, c, ptr_to_missing_alloc_key, @@ -574,7 +575,8 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id continue; if (fsck_err_on(bucket_data_type(g->data_type) && - bucket_data_type(g->data_type) != data_type, c, + bucket_data_type(g->data_type) != + bucket_data_type(data_type), c, ptr_bucket_data_type_mismatch, "bucket %u:%zu different types of data in same bucket: %s, %s\n" "while marking %s", @@ -615,18 +617,13 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id } if (do_update) { - struct bkey_ptrs ptrs; - union bch_extent_entry *entry; - struct bch_extent_ptr *ptr; - struct bkey_i *new; - if (is_root) { bch_err(c, "cannot update btree roots yet"); ret = -EINVAL; goto err; } - new = kmalloc(bkey_bytes(k->k), GFP_KERNEL); + struct bkey_i *new = kmalloc(bkey_bytes(k->k), GFP_KERNEL); if (!new) { ret = -BCH_ERR_ENOMEM_gc_repair_key; bch_err_msg(c, ret, "allocating new key"); @@ -641,7 +638,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id * btree node isn't there anymore, the read path will * sort it out: */ - ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); + struct bkey_ptrs ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); bkey_for_each_ptr(ptrs, ptr) { struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); struct bucket *g = PTR_GC_BUCKET(ca, ptr); @@ -649,19 +646,26 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id ptr->gen = g->gen; } } else { - bch2_bkey_drop_ptrs(bkey_i_to_s(new), ptr, ({ - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); - struct bucket *g = PTR_GC_BUCKET(ca, ptr); - enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, ptr); - - (ptr->cached && - (!g->gen_valid || gen_cmp(ptr->gen, g->gen) > 0)) || - (!ptr->cached && - gen_cmp(ptr->gen, g->gen) < 0) || - gen_cmp(g->gen, ptr->gen) > BUCKET_GC_GEN_MAX || - (g->data_type && - g->data_type != data_type); - })); + struct bkey_ptrs ptrs; + union bch_extent_entry *entry; +restart_drop_ptrs: + ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); + bkey_for_each_ptr_decode(bkey_i_to_s(new).k, ptrs, p, entry) { + struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev); + struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); + enum bch_data_type data_type = bch2_bkey_ptr_data_type(bkey_i_to_s_c(new), p, entry); + + if ((p.ptr.cached && + (!g->gen_valid || gen_cmp(p.ptr.gen, g->gen) > 0)) || + (!p.ptr.cached && + gen_cmp(p.ptr.gen, g->gen) < 0) || + gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX || + (g->data_type && + g->data_type != data_type)) { + bch2_bkey_drop_ptr(bkey_i_to_s(new), &entry->ptr); + goto restart_drop_ptrs; + } + } again: ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); bkey_extent_entry_for_each(ptrs, entry) { @@ -736,10 +740,6 @@ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, BUG_ON(bch2_journal_seq_verify && k->k->version.lo > atomic64_read(&c->journal.seq)); - ret = bch2_check_fix_ptrs(trans, btree_id, level, is_root, k); - if (ret) - goto err; - if (fsck_err_on(k->k->version.lo > atomic64_read(&c->key_version), c, bkey_version_in_future, "key version number higher than recorded: %llu > %llu", @@ -748,8 +748,13 @@ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, atomic64_set(&c->key_version, k->k->version.lo); } + ret = bch2_check_fix_ptrs(trans, btree_id, level, is_root, k); + if (ret) + goto err; + ret = commit_do(trans, NULL, NULL, 0, - bch2_key_trigger(trans, btree_id, level, old, unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_GC)); + bch2_key_trigger(trans, btree_id, level, old, + unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_GC)); fsck_err: err: bch_err_fn(c, ret); diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 96edf2c34d43..941401a210f5 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -525,6 +525,7 @@ int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca, "different types of data in same bucket: %s, %s", bch2_data_type_str(g->data_type), bch2_data_type_str(data_type))) { + BUG(); ret = -EIO; goto err; } @@ -628,6 +629,7 @@ int bch2_check_bucket_ref(struct btree_trans *trans, bch2_data_type_str(ptr_data_type), (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf)); + BUG(); ret = -EIO; goto err; } @@ -815,14 +817,14 @@ static int __mark_pointer(struct btree_trans *trans, static int bch2_trigger_pointer(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c k, struct extent_ptr_decoded p, - s64 *sectors, - unsigned flags) + const union bch_extent_entry *entry, + s64 *sectors, unsigned flags) { bool insert = !(flags & BTREE_TRIGGER_OVERWRITE); struct bpos bucket; struct bch_backpointer bp; - bch2_extent_ptr_to_bp(trans->c, btree_id, level, k, p, &bucket, &bp); + bch2_extent_ptr_to_bp(trans->c, btree_id, level, k, p, entry, &bucket, &bp); *sectors = insert ? bp.bucket_len : -((s64) bp.bucket_len); if (flags & BTREE_TRIGGER_TRANSACTIONAL) { @@ -851,7 +853,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, if (flags & BTREE_TRIGGER_GC) { struct bch_fs *c = trans->c; struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev); - enum bch_data_type data_type = bkey_ptr_data_type(btree_id, level, k, p); + enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry); percpu_down_read(&c->mark_lock); struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); @@ -979,7 +981,7 @@ static int __trigger_extent(struct btree_trans *trans, bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { s64 disk_sectors; - ret = bch2_trigger_pointer(trans, btree_id, level, k, p, &disk_sectors, flags); + ret = bch2_trigger_pointer(trans, btree_id, level, k, p, entry, &disk_sectors, flags); if (ret < 0) return ret; diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h index fd2669cdd76f..3fd0169b98c1 100644 --- a/fs/bcachefs/extents.h +++ b/fs/bcachefs/extents.h @@ -596,30 +596,6 @@ static inline struct bch_devs_list bch2_bkey_cached_devs(struct bkey_s_c k) return ret; } -static inline unsigned bch2_bkey_ptr_data_type(struct bkey_s_c k, const struct bch_extent_ptr *ptr) -{ - switch (k.k->type) { - case KEY_TYPE_btree_ptr: - case KEY_TYPE_btree_ptr_v2: - return BCH_DATA_btree; - case KEY_TYPE_extent: - case KEY_TYPE_reflink_v: - return BCH_DATA_user; - case KEY_TYPE_stripe: { - struct bkey_s_c_stripe s = bkey_s_c_to_stripe(k); - - BUG_ON(ptr < s.v->ptrs || - ptr >= s.v->ptrs + s.v->nr_blocks); - - return ptr >= s.v->ptrs + s.v->nr_blocks - s.v->nr_redundant - ? BCH_DATA_parity - : BCH_DATA_user; - } - default: - BUG(); - } -} - unsigned bch2_bkey_nr_ptrs(struct bkey_s_c); unsigned bch2_bkey_nr_ptrs_allocated(struct bkey_s_c); unsigned bch2_bkey_nr_ptrs_fully_allocated(struct bkey_s_c); -- cgit v1.2.3-59-g8ed1b From 7f9e5080366726084eb765a5d689bdf502e7e2ed Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 14 Mar 2024 19:39:26 -0400 Subject: bcachefs: Fix bch2_btree_increase_depth() When we haven't yet allocated any btree nodes for a given btree, we first need to call the regular split path to allocate one. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 826cebe4bce0..9c4325786ba8 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1882,9 +1882,12 @@ int bch2_btree_increase_depth(struct btree_trans *trans, btree_path_idx_t path, { struct bch_fs *c = trans->c; struct btree *b = bch2_btree_id_root(c, trans->paths[path].btree_id)->b; + + if (btree_node_fake(b)) + return bch2_btree_split_leaf(trans, path, flags); + struct btree_update *as = - bch2_btree_update_start(trans, trans->paths + path, - b->c.level, true, flags); + bch2_btree_update_start(trans, trans->paths + path, b->c.level, true, flags); if (IS_ERR(as)) return PTR_ERR(as); -- cgit v1.2.3-59-g8ed1b From 11d5568d3e04a2e6734d1eccc394cfcf5ca8523c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 28 Mar 2024 01:41:03 -0400 Subject: bcachefs: fix backpointer for missing alloc key msg Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 7537d78aa9db..762c8ddfc5e7 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -382,7 +382,7 @@ static int bch2_check_btree_backpointer(struct btree_trans *trans, struct btree_ backpointer_to_missing_alloc, "backpointer for nonexistent alloc key: %llu:%llu:0\n%s", alloc_iter.pos.inode, alloc_iter.pos.offset, - (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf))) { + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { ret = bch2_btree_delete_at(trans, bp_iter, 0); goto out; } -- cgit v1.2.3-59-g8ed1b From d2554263adcb4041f3608cb7476f102fda036ccc Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 23 Mar 2024 20:07:46 -0400 Subject: bcachefs: Split out recovery_passes.c We've grown a fair amount of code for managing recovery passes; tracking which ones we're running, which ones need to be run, and flagging in the superblock which ones need to be run on the next recovery. So it's worth splitting out into its own file, this code is pretty different from the code in recovery.c. Signed-off-by: Kent Overstreet --- fs/bcachefs/Makefile | 1 + fs/bcachefs/bcachefs.h | 2 +- fs/bcachefs/btree_gc.c | 2 +- fs/bcachefs/btree_update_interior.c | 2 +- fs/bcachefs/chardev.c | 2 +- fs/bcachefs/error.c | 2 +- fs/bcachefs/fsck.c | 2 +- fs/bcachefs/recovery.c | 249 +----------------------------------- fs/bcachefs/recovery.h | 32 +---- fs/bcachefs/recovery_passes.c | 203 +++++++++++++++++++++++++++++ fs/bcachefs/recovery_passes.h | 16 +++ fs/bcachefs/recovery_passes_types.h | 69 ++++++++++ fs/bcachefs/recovery_types.h | 68 ---------- fs/bcachefs/sb-downgrade.c | 2 +- fs/bcachefs/subvolume.c | 72 +++++++++++ fs/bcachefs/subvolume.h | 3 + fs/bcachefs/super-io.c | 2 +- 17 files changed, 378 insertions(+), 351 deletions(-) create mode 100644 fs/bcachefs/recovery_passes.c create mode 100644 fs/bcachefs/recovery_passes.h create mode 100644 fs/bcachefs/recovery_passes_types.h delete mode 100644 fs/bcachefs/recovery_types.h (limited to 'fs') diff --git a/fs/bcachefs/Makefile b/fs/bcachefs/Makefile index b02796c8a595..f6d86eb4b976 100644 --- a/fs/bcachefs/Makefile +++ b/fs/bcachefs/Makefile @@ -67,6 +67,7 @@ bcachefs-y := \ quota.o \ rebalance.o \ recovery.o \ + recovery_passes.o \ reflink.o \ replicas.o \ sb-clean.o \ diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index a9ade0b1f78c..963162a627cd 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -209,7 +209,7 @@ #include "fifo.h" #include "nocow_locking_types.h" #include "opts.h" -#include "recovery_types.h" +#include "recovery_passes_types.h" #include "sb-errors_types.h" #include "seqmutex.h" #include "time_stats.h" diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 0afefccf4e52..e5d2c6daa663 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -25,7 +25,7 @@ #include "journal.h" #include "keylist.h" #include "move.h" -#include "recovery.h" +#include "recovery_passes.h" #include "reflink.h" #include "replicas.h" #include "super-io.h" diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 9c4325786ba8..983bb27298cc 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -19,7 +19,7 @@ #include "journal.h" #include "journal_reclaim.h" #include "keylist.h" -#include "recovery.h" +#include "recovery_passes.h" #include "replicas.h" #include "super-io.h" #include "trace.h" diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index 38defa19d52d..cbfa6459bdbc 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -7,7 +7,7 @@ #include "chardev.h" #include "journal.h" #include "move.h" -#include "recovery.h" +#include "recovery_passes.h" #include "replicas.h" #include "super.h" #include "super-io.h" diff --git a/fs/bcachefs/error.c b/fs/bcachefs/error.c index 043431206799..f942a3947496 100644 --- a/fs/bcachefs/error.c +++ b/fs/bcachefs/error.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" #include "error.h" -#include "recovery.h" +#include "recovery_passes.h" #include "super.h" #include "thread_with_file.h" diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 6d8367ab5ddd..aca6fae409cd 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -12,7 +12,7 @@ #include "fsck.h" #include "inode.h" #include "keylist.h" -#include "recovery.h" +#include "recovery_passes.h" #include "snapshot.h" #include "super.h" #include "xattr.h" diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 03f9d6afe467..f9e2b011f6d4 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -1,35 +1,30 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" -#include "backpointers.h" -#include "bkey_buf.h" #include "alloc_background.h" -#include "btree_gc.h" +#include "bkey_buf.h" #include "btree_journal_iter.h" #include "btree_update.h" #include "btree_update_interior.h" #include "btree_io.h" #include "buckets.h" #include "dirent.h" -#include "ec.h" #include "errcode.h" #include "error.h" #include "fs-common.h" -#include "fsck.h" #include "journal_io.h" #include "journal_reclaim.h" #include "journal_seq_blacklist.h" -#include "lru.h" #include "logged_ops.h" #include "move.h" #include "quota.h" #include "rebalance.h" #include "recovery.h" +#include "recovery_passes.h" #include "replicas.h" #include "sb-clean.h" #include "sb-downgrade.h" #include "snapshot.h" -#include "subvolume.h" #include "super-io.h" #include @@ -186,7 +181,7 @@ static int journal_sort_seq_cmp(const void *_l, const void *_r) return cmp_int(l->journal_seq, r->journal_seq); } -static int bch2_journal_replay(struct bch_fs *c) +int bch2_journal_replay(struct bch_fs *c) { struct journal_keys *keys = &c->journal_keys; DARRAY(struct journal_key *) keys_sorted = { 0 }; @@ -471,150 +466,6 @@ fsck_err: return ret; } -static int bch2_initialize_subvolumes(struct bch_fs *c) -{ - struct bkey_i_snapshot_tree root_tree; - struct bkey_i_snapshot root_snapshot; - struct bkey_i_subvolume root_volume; - int ret; - - bkey_snapshot_tree_init(&root_tree.k_i); - root_tree.k.p.offset = 1; - root_tree.v.master_subvol = cpu_to_le32(1); - root_tree.v.root_snapshot = cpu_to_le32(U32_MAX); - - bkey_snapshot_init(&root_snapshot.k_i); - root_snapshot.k.p.offset = U32_MAX; - root_snapshot.v.flags = 0; - root_snapshot.v.parent = 0; - root_snapshot.v.subvol = cpu_to_le32(BCACHEFS_ROOT_SUBVOL); - root_snapshot.v.tree = cpu_to_le32(1); - SET_BCH_SNAPSHOT_SUBVOL(&root_snapshot.v, true); - - bkey_subvolume_init(&root_volume.k_i); - root_volume.k.p.offset = BCACHEFS_ROOT_SUBVOL; - root_volume.v.flags = 0; - root_volume.v.snapshot = cpu_to_le32(U32_MAX); - root_volume.v.inode = cpu_to_le64(BCACHEFS_ROOT_INO); - - ret = bch2_btree_insert(c, BTREE_ID_snapshot_trees, &root_tree.k_i, NULL, 0) ?: - bch2_btree_insert(c, BTREE_ID_snapshots, &root_snapshot.k_i, NULL, 0) ?: - bch2_btree_insert(c, BTREE_ID_subvolumes, &root_volume.k_i, NULL, 0); - bch_err_fn(c, ret); - return ret; -} - -static int __bch2_fs_upgrade_for_subvolumes(struct btree_trans *trans) -{ - struct btree_iter iter; - struct bkey_s_c k; - struct bch_inode_unpacked inode; - int ret; - - k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, - SPOS(0, BCACHEFS_ROOT_INO, U32_MAX), 0); - ret = bkey_err(k); - if (ret) - return ret; - - if (!bkey_is_inode(k.k)) { - bch_err(trans->c, "root inode not found"); - ret = -BCH_ERR_ENOENT_inode; - goto err; - } - - ret = bch2_inode_unpack(k, &inode); - BUG_ON(ret); - - inode.bi_subvol = BCACHEFS_ROOT_SUBVOL; - - ret = bch2_inode_write(trans, &iter, &inode); -err: - bch2_trans_iter_exit(trans, &iter); - return ret; -} - -/* set bi_subvol on root inode */ -noinline_for_stack -static int bch2_fs_upgrade_for_subvolumes(struct bch_fs *c) -{ - int ret = bch2_trans_do(c, NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, - __bch2_fs_upgrade_for_subvolumes(trans)); - bch_err_fn(c, ret); - return ret; -} - -const char * const bch2_recovery_passes[] = { -#define x(_fn, ...) #_fn, - BCH_RECOVERY_PASSES() -#undef x - NULL -}; - -static int bch2_check_allocations(struct bch_fs *c) -{ - return bch2_gc(c, true, c->opts.norecovery); -} - -static int bch2_set_may_go_rw(struct bch_fs *c) -{ - struct journal_keys *keys = &c->journal_keys; - - /* - * After we go RW, the journal keys buffer can't be modified (except for - * setting journal_key->overwritten: it will be accessed by multiple - * threads - */ - move_gap(keys, keys->nr); - - set_bit(BCH_FS_may_go_rw, &c->flags); - - if (keys->nr || c->opts.fsck || !c->sb.clean) - return bch2_fs_read_write_early(c); - return 0; -} - -struct recovery_pass_fn { - int (*fn)(struct bch_fs *); - unsigned when; -}; - -static struct recovery_pass_fn recovery_pass_fns[] = { -#define x(_fn, _id, _when) { .fn = bch2_##_fn, .when = _when }, - BCH_RECOVERY_PASSES() -#undef x -}; - -u64 bch2_recovery_passes_to_stable(u64 v) -{ - static const u8 map[] = { -#define x(n, id, ...) [BCH_RECOVERY_PASS_##n] = BCH_RECOVERY_PASS_STABLE_##n, - BCH_RECOVERY_PASSES() -#undef x - }; - - u64 ret = 0; - for (unsigned i = 0; i < ARRAY_SIZE(map); i++) - if (v & BIT_ULL(i)) - ret |= BIT_ULL(map[i]); - return ret; -} - -u64 bch2_recovery_passes_from_stable(u64 v) -{ - static const u8 map[] = { -#define x(n, id, ...) [BCH_RECOVERY_PASS_STABLE_##n] = BCH_RECOVERY_PASS_##n, - BCH_RECOVERY_PASSES() -#undef x - }; - - u64 ret = 0; - for (unsigned i = 0; i < ARRAY_SIZE(map); i++) - if (v & BIT_ULL(i)) - ret |= BIT_ULL(map[i]); - return ret; -} - static bool check_version_upgrade(struct bch_fs *c) { unsigned latest_version = bcachefs_metadata_version_current; @@ -687,96 +538,6 @@ static bool check_version_upgrade(struct bch_fs *c) return false; } -u64 bch2_fsck_recovery_passes(void) -{ - u64 ret = 0; - - for (unsigned i = 0; i < ARRAY_SIZE(recovery_pass_fns); i++) - if (recovery_pass_fns[i].when & PASS_FSCK) - ret |= BIT_ULL(i); - return ret; -} - -static bool should_run_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pass) -{ - struct recovery_pass_fn *p = recovery_pass_fns + pass; - - if (c->opts.norecovery && pass > BCH_RECOVERY_PASS_snapshots_read) - return false; - if (c->recovery_passes_explicit & BIT_ULL(pass)) - return true; - if ((p->when & PASS_FSCK) && c->opts.fsck) - return true; - if ((p->when & PASS_UNCLEAN) && !c->sb.clean) - return true; - if (p->when & PASS_ALWAYS) - return true; - return false; -} - -static int bch2_run_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pass) -{ - struct recovery_pass_fn *p = recovery_pass_fns + pass; - int ret; - - if (!(p->when & PASS_SILENT)) - bch2_print(c, KERN_INFO bch2_log_msg(c, "%s..."), - bch2_recovery_passes[pass]); - ret = p->fn(c); - if (ret) - return ret; - if (!(p->when & PASS_SILENT)) - bch2_print(c, KERN_CONT " done\n"); - - return 0; -} - -static int bch2_run_recovery_passes(struct bch_fs *c) -{ - int ret = 0; - - while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns)) { - if (should_run_recovery_pass(c, c->curr_recovery_pass)) { - unsigned pass = c->curr_recovery_pass; - - ret = bch2_run_recovery_pass(c, c->curr_recovery_pass); - if (bch2_err_matches(ret, BCH_ERR_restart_recovery) || - (ret && c->curr_recovery_pass < pass)) - continue; - if (ret) - break; - - c->recovery_passes_complete |= BIT_ULL(c->curr_recovery_pass); - } - c->curr_recovery_pass++; - c->recovery_pass_done = max(c->recovery_pass_done, c->curr_recovery_pass); - } - - return ret; -} - -int bch2_run_online_recovery_passes(struct bch_fs *c) -{ - int ret = 0; - - for (unsigned i = 0; i < ARRAY_SIZE(recovery_pass_fns); i++) { - struct recovery_pass_fn *p = recovery_pass_fns + i; - - if (!(p->when & PASS_ONLINE)) - continue; - - ret = bch2_run_recovery_pass(c, i); - if (bch2_err_matches(ret, BCH_ERR_restart_recovery)) { - i = c->curr_recovery_pass; - continue; - } - if (ret) - break; - } - - return ret; -} - int bch2_fs_recovery(struct bch_fs *c) { struct bch_sb_field_clean *clean = NULL; @@ -1155,7 +916,7 @@ int bch2_fs_initialize(struct bch_fs *c) } mutex_unlock(&c->sb_lock); - c->curr_recovery_pass = ARRAY_SIZE(recovery_pass_fns); + c->curr_recovery_pass = BCH_RECOVERY_PASS_NR; set_bit(BCH_FS_may_go_rw, &c->flags); for (unsigned i = 0; i < BTREE_ID_NR; i++) @@ -1230,7 +991,7 @@ int bch2_fs_initialize(struct bch_fs *c) if (ret) goto err; - c->recovery_pass_done = ARRAY_SIZE(recovery_pass_fns) - 1; + c->recovery_pass_done = BCH_RECOVERY_PASS_NR - 1; if (enabled_qtypes(c)) { ret = bch2_fs_quota_read(c); diff --git a/fs/bcachefs/recovery.h b/fs/bcachefs/recovery.h index 4e9d24719b2e..3962fd87b50d 100644 --- a/fs/bcachefs/recovery.h +++ b/fs/bcachefs/recovery.h @@ -2,37 +2,7 @@ #ifndef _BCACHEFS_RECOVERY_H #define _BCACHEFS_RECOVERY_H -extern const char * const bch2_recovery_passes[]; - -u64 bch2_recovery_passes_to_stable(u64 v); -u64 bch2_recovery_passes_from_stable(u64 v); - -/* - * For when we need to rewind recovery passes and run a pass we skipped: - */ -static inline int bch2_run_explicit_recovery_pass(struct bch_fs *c, - enum bch_recovery_pass pass) -{ - if (c->recovery_passes_explicit & BIT_ULL(pass)) - return 0; - - bch_info(c, "running explicit recovery pass %s (%u), currently at %s (%u)", - bch2_recovery_passes[pass], pass, - bch2_recovery_passes[c->curr_recovery_pass], c->curr_recovery_pass); - - c->recovery_passes_explicit |= BIT_ULL(pass); - - if (c->curr_recovery_pass >= pass) { - c->curr_recovery_pass = pass; - c->recovery_passes_complete &= (1ULL << pass) >> 1; - return -BCH_ERR_restart_recovery; - } else { - return 0; - } -} - -int bch2_run_online_recovery_passes(struct bch_fs *); -u64 bch2_fsck_recovery_passes(void); +int bch2_journal_replay(struct bch_fs *); int bch2_fs_recovery(struct bch_fs *); int bch2_fs_initialize(struct bch_fs *); diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c new file mode 100644 index 000000000000..bab586dc395a --- /dev/null +++ b/fs/bcachefs/recovery_passes.c @@ -0,0 +1,203 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "bcachefs.h" +#include "alloc_background.h" +#include "backpointers.h" +#include "btree_gc.h" +#include "ec.h" +#include "fsck.h" +#include "inode.h" +#include "journal.h" +#include "lru.h" +#include "logged_ops.h" +#include "rebalance.h" +#include "recovery.h" +#include "recovery_passes.h" +#include "snapshot.h" +#include "subvolume.h" +#include "super.h" + +const char * const bch2_recovery_passes[] = { +#define x(_fn, ...) #_fn, + BCH_RECOVERY_PASSES() +#undef x + NULL +}; + +static int bch2_check_allocations(struct bch_fs *c) +{ + return bch2_gc(c, true, c->opts.norecovery); +} + +static int bch2_set_may_go_rw(struct bch_fs *c) +{ + struct journal_keys *keys = &c->journal_keys; + + /* + * After we go RW, the journal keys buffer can't be modified (except for + * setting journal_key->overwritten: it will be accessed by multiple + * threads + */ + move_gap(keys, keys->nr); + + set_bit(BCH_FS_may_go_rw, &c->flags); + + if (keys->nr || c->opts.fsck || !c->sb.clean) + return bch2_fs_read_write_early(c); + return 0; +} + +struct recovery_pass_fn { + int (*fn)(struct bch_fs *); + unsigned when; +}; + +static struct recovery_pass_fn recovery_pass_fns[] = { +#define x(_fn, _id, _when) { .fn = bch2_##_fn, .when = _when }, + BCH_RECOVERY_PASSES() +#undef x +}; + +u64 bch2_recovery_passes_to_stable(u64 v) +{ + static const u8 map[] = { +#define x(n, id, ...) [BCH_RECOVERY_PASS_##n] = BCH_RECOVERY_PASS_STABLE_##n, + BCH_RECOVERY_PASSES() +#undef x + }; + + u64 ret = 0; + for (unsigned i = 0; i < ARRAY_SIZE(map); i++) + if (v & BIT_ULL(i)) + ret |= BIT_ULL(map[i]); + return ret; +} + +u64 bch2_recovery_passes_from_stable(u64 v) +{ + static const u8 map[] = { +#define x(n, id, ...) [BCH_RECOVERY_PASS_STABLE_##n] = BCH_RECOVERY_PASS_##n, + BCH_RECOVERY_PASSES() +#undef x + }; + + u64 ret = 0; + for (unsigned i = 0; i < ARRAY_SIZE(map); i++) + if (v & BIT_ULL(i)) + ret |= BIT_ULL(map[i]); + return ret; +} + +/* + * For when we need to rewind recovery passes and run a pass we skipped: + */ +int bch2_run_explicit_recovery_pass(struct bch_fs *c, + enum bch_recovery_pass pass) +{ + if (c->recovery_passes_explicit & BIT_ULL(pass)) + return 0; + + bch_info(c, "running explicit recovery pass %s (%u), currently at %s (%u)", + bch2_recovery_passes[pass], pass, + bch2_recovery_passes[c->curr_recovery_pass], c->curr_recovery_pass); + + c->recovery_passes_explicit |= BIT_ULL(pass); + + if (c->curr_recovery_pass >= pass) { + c->curr_recovery_pass = pass; + c->recovery_passes_complete &= (1ULL << pass) >> 1; + return -BCH_ERR_restart_recovery; + } else { + return 0; + } +} + +u64 bch2_fsck_recovery_passes(void) +{ + u64 ret = 0; + + for (unsigned i = 0; i < ARRAY_SIZE(recovery_pass_fns); i++) + if (recovery_pass_fns[i].when & PASS_FSCK) + ret |= BIT_ULL(i); + return ret; +} + +static bool should_run_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pass) +{ + struct recovery_pass_fn *p = recovery_pass_fns + pass; + + if (c->opts.norecovery && pass > BCH_RECOVERY_PASS_snapshots_read) + return false; + if (c->recovery_passes_explicit & BIT_ULL(pass)) + return true; + if ((p->when & PASS_FSCK) && c->opts.fsck) + return true; + if ((p->when & PASS_UNCLEAN) && !c->sb.clean) + return true; + if (p->when & PASS_ALWAYS) + return true; + return false; +} + +static int bch2_run_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pass) +{ + struct recovery_pass_fn *p = recovery_pass_fns + pass; + int ret; + + if (!(p->when & PASS_SILENT)) + bch2_print(c, KERN_INFO bch2_log_msg(c, "%s..."), + bch2_recovery_passes[pass]); + ret = p->fn(c); + if (ret) + return ret; + if (!(p->when & PASS_SILENT)) + bch2_print(c, KERN_CONT " done\n"); + + return 0; +} + +int bch2_run_online_recovery_passes(struct bch_fs *c) +{ + int ret = 0; + + for (unsigned i = 0; i < ARRAY_SIZE(recovery_pass_fns); i++) { + struct recovery_pass_fn *p = recovery_pass_fns + i; + + if (!(p->when & PASS_ONLINE)) + continue; + + ret = bch2_run_recovery_pass(c, i); + if (bch2_err_matches(ret, BCH_ERR_restart_recovery)) { + i = c->curr_recovery_pass; + continue; + } + if (ret) + break; + } + + return ret; +} + +int bch2_run_recovery_passes(struct bch_fs *c) +{ + int ret = 0; + + while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns)) { + if (should_run_recovery_pass(c, c->curr_recovery_pass)) { + unsigned pass = c->curr_recovery_pass; + + ret = bch2_run_recovery_pass(c, c->curr_recovery_pass); + if (bch2_err_matches(ret, BCH_ERR_restart_recovery) || + (ret && c->curr_recovery_pass < pass)) + continue; + if (ret) + break; + + c->recovery_passes_complete |= BIT_ULL(c->curr_recovery_pass); + } + c->curr_recovery_pass++; + c->recovery_pass_done = max(c->recovery_pass_done, c->curr_recovery_pass); + } + + return ret; +} diff --git a/fs/bcachefs/recovery_passes.h b/fs/bcachefs/recovery_passes.h new file mode 100644 index 000000000000..abefa67749eb --- /dev/null +++ b/fs/bcachefs/recovery_passes.h @@ -0,0 +1,16 @@ +#ifndef _BCACHEFS_RECOVERY_PASSES_H +#define _BCACHEFS_RECOVERY_PASSES_H + +extern const char * const bch2_recovery_passes[]; + +u64 bch2_recovery_passes_to_stable(u64 v); +u64 bch2_recovery_passes_from_stable(u64 v); + +u64 bch2_fsck_recovery_passes(void); + +int bch2_run_explicit_recovery_pass(struct bch_fs *, enum bch_recovery_pass); + +int bch2_run_online_recovery_passes(struct bch_fs *); +int bch2_run_recovery_passes(struct bch_fs *); + +#endif /* _BCACHEFS_RECOVERY_PASSES_H */ diff --git a/fs/bcachefs/recovery_passes_types.h b/fs/bcachefs/recovery_passes_types.h new file mode 100644 index 000000000000..09c1ab032a67 --- /dev/null +++ b/fs/bcachefs/recovery_passes_types.h @@ -0,0 +1,69 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _BCACHEFS_RECOVERY_PASSES_TYPES_H +#define _BCACHEFS_RECOVERY_PASSES_TYPES_H + +#define PASS_SILENT BIT(0) +#define PASS_FSCK BIT(1) +#define PASS_UNCLEAN BIT(2) +#define PASS_ALWAYS BIT(3) +#define PASS_ONLINE BIT(4) + +/* + * Passes may be reordered, but the second field is a persistent identifier and + * must never change: + */ +#define BCH_RECOVERY_PASSES() \ + x(check_topology, 4, 0) \ + x(alloc_read, 0, PASS_ALWAYS) \ + x(stripes_read, 1, PASS_ALWAYS) \ + x(initialize_subvolumes, 2, 0) \ + x(snapshots_read, 3, PASS_ALWAYS) \ + x(check_allocations, 5, PASS_FSCK) \ + x(trans_mark_dev_sbs, 6, PASS_ALWAYS|PASS_SILENT) \ + x(fs_journal_alloc, 7, PASS_ALWAYS|PASS_SILENT) \ + x(set_may_go_rw, 8, PASS_ALWAYS|PASS_SILENT) \ + x(journal_replay, 9, PASS_ALWAYS) \ + x(check_alloc_info, 10, PASS_ONLINE|PASS_FSCK) \ + x(check_lrus, 11, PASS_ONLINE|PASS_FSCK) \ + x(check_btree_backpointers, 12, PASS_ONLINE|PASS_FSCK) \ + x(check_backpointers_to_extents, 13, PASS_ONLINE|PASS_FSCK) \ + x(check_extents_to_backpointers, 14, PASS_ONLINE|PASS_FSCK) \ + x(check_alloc_to_lru_refs, 15, PASS_ONLINE|PASS_FSCK) \ + x(fs_freespace_init, 16, PASS_ALWAYS|PASS_SILENT) \ + x(bucket_gens_init, 17, 0) \ + x(check_snapshot_trees, 18, PASS_ONLINE|PASS_FSCK) \ + x(check_snapshots, 19, PASS_ONLINE|PASS_FSCK) \ + x(check_subvols, 20, PASS_ONLINE|PASS_FSCK) \ + x(check_subvol_children, 35, PASS_ONLINE|PASS_FSCK) \ + x(delete_dead_snapshots, 21, PASS_ONLINE|PASS_FSCK) \ + x(fs_upgrade_for_subvolumes, 22, 0) \ + x(resume_logged_ops, 23, PASS_ALWAYS) \ + x(check_inodes, 24, PASS_FSCK) \ + x(check_extents, 25, PASS_FSCK) \ + x(check_indirect_extents, 26, PASS_FSCK) \ + x(check_dirents, 27, PASS_FSCK) \ + x(check_xattrs, 28, PASS_FSCK) \ + x(check_root, 29, PASS_ONLINE|PASS_FSCK) \ + x(check_subvolume_structure, 36, PASS_ONLINE|PASS_FSCK) \ + x(check_directory_structure, 30, PASS_ONLINE|PASS_FSCK) \ + x(check_nlinks, 31, PASS_FSCK) \ + x(delete_dead_inodes, 32, PASS_FSCK|PASS_UNCLEAN) \ + x(fix_reflink_p, 33, 0) \ + x(set_fs_needs_rebalance, 34, 0) \ + +/* We normally enumerate recovery passes in the order we run them: */ +enum bch_recovery_pass { +#define x(n, id, when) BCH_RECOVERY_PASS_##n, + BCH_RECOVERY_PASSES() +#undef x + BCH_RECOVERY_PASS_NR +}; + +/* But we also need stable identifiers that can be used in the superblock */ +enum bch_recovery_pass_stable { +#define x(n, id, when) BCH_RECOVERY_PASS_STABLE_##n = id, + BCH_RECOVERY_PASSES() +#undef x +}; + +#endif /* _BCACHEFS_RECOVERY_PASSES_TYPES_H */ diff --git a/fs/bcachefs/recovery_types.h b/fs/bcachefs/recovery_types.h deleted file mode 100644 index 4959e95e7c74..000000000000 --- a/fs/bcachefs/recovery_types.h +++ /dev/null @@ -1,68 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _BCACHEFS_RECOVERY_TYPES_H -#define _BCACHEFS_RECOVERY_TYPES_H - -#define PASS_SILENT BIT(0) -#define PASS_FSCK BIT(1) -#define PASS_UNCLEAN BIT(2) -#define PASS_ALWAYS BIT(3) -#define PASS_ONLINE BIT(4) - -/* - * Passes may be reordered, but the second field is a persistent identifier and - * must never change: - */ -#define BCH_RECOVERY_PASSES() \ - x(check_topology, 4, 0) \ - x(alloc_read, 0, PASS_ALWAYS) \ - x(stripes_read, 1, PASS_ALWAYS) \ - x(initialize_subvolumes, 2, 0) \ - x(snapshots_read, 3, PASS_ALWAYS) \ - x(check_allocations, 5, PASS_FSCK) \ - x(trans_mark_dev_sbs, 6, PASS_ALWAYS|PASS_SILENT) \ - x(fs_journal_alloc, 7, PASS_ALWAYS|PASS_SILENT) \ - x(set_may_go_rw, 8, PASS_ALWAYS|PASS_SILENT) \ - x(journal_replay, 9, PASS_ALWAYS) \ - x(check_alloc_info, 10, PASS_ONLINE|PASS_FSCK) \ - x(check_lrus, 11, PASS_ONLINE|PASS_FSCK) \ - x(check_btree_backpointers, 12, PASS_ONLINE|PASS_FSCK) \ - x(check_backpointers_to_extents, 13, PASS_ONLINE|PASS_FSCK) \ - x(check_extents_to_backpointers, 14, PASS_ONLINE|PASS_FSCK) \ - x(check_alloc_to_lru_refs, 15, PASS_ONLINE|PASS_FSCK) \ - x(fs_freespace_init, 16, PASS_ALWAYS|PASS_SILENT) \ - x(bucket_gens_init, 17, 0) \ - x(check_snapshot_trees, 18, PASS_ONLINE|PASS_FSCK) \ - x(check_snapshots, 19, PASS_ONLINE|PASS_FSCK) \ - x(check_subvols, 20, PASS_ONLINE|PASS_FSCK) \ - x(check_subvol_children, 35, PASS_ONLINE|PASS_FSCK) \ - x(delete_dead_snapshots, 21, PASS_ONLINE|PASS_FSCK) \ - x(fs_upgrade_for_subvolumes, 22, 0) \ - x(resume_logged_ops, 23, PASS_ALWAYS) \ - x(check_inodes, 24, PASS_FSCK) \ - x(check_extents, 25, PASS_FSCK) \ - x(check_indirect_extents, 26, PASS_FSCK) \ - x(check_dirents, 27, PASS_FSCK) \ - x(check_xattrs, 28, PASS_FSCK) \ - x(check_root, 29, PASS_ONLINE|PASS_FSCK) \ - x(check_subvolume_structure, 36, PASS_ONLINE|PASS_FSCK) \ - x(check_directory_structure, 30, PASS_ONLINE|PASS_FSCK) \ - x(check_nlinks, 31, PASS_FSCK) \ - x(delete_dead_inodes, 32, PASS_FSCK|PASS_UNCLEAN) \ - x(fix_reflink_p, 33, 0) \ - x(set_fs_needs_rebalance, 34, 0) \ - -/* We normally enumerate recovery passes in the order we run them: */ -enum bch_recovery_pass { -#define x(n, id, when) BCH_RECOVERY_PASS_##n, - BCH_RECOVERY_PASSES() -#undef x -}; - -/* But we also need stable identifiers that can be used in the superblock */ -enum bch_recovery_pass_stable { -#define x(n, id, when) BCH_RECOVERY_PASS_STABLE_##n = id, - BCH_RECOVERY_PASSES() -#undef x -}; - -#endif /* _BCACHEFS_RECOVERY_TYPES_H */ diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c index e4396cb0bacb..d6f81179c3a2 100644 --- a/fs/bcachefs/sb-downgrade.c +++ b/fs/bcachefs/sb-downgrade.c @@ -7,7 +7,7 @@ #include "bcachefs.h" #include "darray.h" -#include "recovery.h" +#include "recovery_passes.h" #include "sb-downgrade.h" #include "sb-errors.h" #include "super-io.h" diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index ce7aed121942..88a79c823276 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -595,6 +595,78 @@ err: return ret; } +int bch2_initialize_subvolumes(struct bch_fs *c) +{ + struct bkey_i_snapshot_tree root_tree; + struct bkey_i_snapshot root_snapshot; + struct bkey_i_subvolume root_volume; + int ret; + + bkey_snapshot_tree_init(&root_tree.k_i); + root_tree.k.p.offset = 1; + root_tree.v.master_subvol = cpu_to_le32(1); + root_tree.v.root_snapshot = cpu_to_le32(U32_MAX); + + bkey_snapshot_init(&root_snapshot.k_i); + root_snapshot.k.p.offset = U32_MAX; + root_snapshot.v.flags = 0; + root_snapshot.v.parent = 0; + root_snapshot.v.subvol = cpu_to_le32(BCACHEFS_ROOT_SUBVOL); + root_snapshot.v.tree = cpu_to_le32(1); + SET_BCH_SNAPSHOT_SUBVOL(&root_snapshot.v, true); + + bkey_subvolume_init(&root_volume.k_i); + root_volume.k.p.offset = BCACHEFS_ROOT_SUBVOL; + root_volume.v.flags = 0; + root_volume.v.snapshot = cpu_to_le32(U32_MAX); + root_volume.v.inode = cpu_to_le64(BCACHEFS_ROOT_INO); + + ret = bch2_btree_insert(c, BTREE_ID_snapshot_trees, &root_tree.k_i, NULL, 0) ?: + bch2_btree_insert(c, BTREE_ID_snapshots, &root_snapshot.k_i, NULL, 0) ?: + bch2_btree_insert(c, BTREE_ID_subvolumes, &root_volume.k_i, NULL, 0); + bch_err_fn(c, ret); + return ret; +} + +static int __bch2_fs_upgrade_for_subvolumes(struct btree_trans *trans) +{ + struct btree_iter iter; + struct bkey_s_c k; + struct bch_inode_unpacked inode; + int ret; + + k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, + SPOS(0, BCACHEFS_ROOT_INO, U32_MAX), 0); + ret = bkey_err(k); + if (ret) + return ret; + + if (!bkey_is_inode(k.k)) { + bch_err(trans->c, "root inode not found"); + ret = -BCH_ERR_ENOENT_inode; + goto err; + } + + ret = bch2_inode_unpack(k, &inode); + BUG_ON(ret); + + inode.bi_subvol = BCACHEFS_ROOT_SUBVOL; + + ret = bch2_inode_write(trans, &iter, &inode); +err: + bch2_trans_iter_exit(trans, &iter); + return ret; +} + +/* set bi_subvol on root inode */ +int bch2_fs_upgrade_for_subvolumes(struct bch_fs *c) +{ + int ret = bch2_trans_do(c, NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, + __bch2_fs_upgrade_for_subvolumes(trans)); + bch_err_fn(c, ret); + return ret; +} + int bch2_fs_subvolumes_init(struct bch_fs *c) { INIT_WORK(&c->snapshot_delete_work, bch2_delete_dead_snapshots_work); diff --git a/fs/bcachefs/subvolume.h b/fs/bcachefs/subvolume.h index 903c05162c06..d2015d549bd2 100644 --- a/fs/bcachefs/subvolume.h +++ b/fs/bcachefs/subvolume.h @@ -37,6 +37,9 @@ void bch2_delete_dead_snapshots_async(struct bch_fs *); int bch2_subvolume_unlink(struct btree_trans *, u32); int bch2_subvolume_create(struct btree_trans *, u64, u32, u32, u32 *, u32 *, bool); +int bch2_initialize_subvolumes(struct bch_fs *); +int bch2_fs_upgrade_for_subvolumes(struct bch_fs *); + int bch2_fs_subvolumes_init(struct bch_fs *); #endif /* _BCACHEFS_SUBVOLUME_H */ diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index ad28e370b640..e15f8b1f30c2 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -8,7 +8,7 @@ #include "journal.h" #include "journal_sb.h" #include "journal_seq_blacklist.h" -#include "recovery.h" +#include "recovery_passes.h" #include "replicas.h" #include "quota.h" #include "sb-clean.h" -- cgit v1.2.3-59-g8ed1b From e5aa80464155287cc309d18c1c93962357e3e393 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 23 Mar 2024 19:30:58 -0400 Subject: bcachefs: Add error messages to logged ops fns Signed-off-by: Kent Overstreet --- fs/bcachefs/io_misc.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/io_misc.c b/fs/bcachefs/io_misc.c index 1baf78594cca..82f9170dab3f 100644 --- a/fs/bcachefs/io_misc.c +++ b/fs/bcachefs/io_misc.c @@ -264,6 +264,7 @@ static int __bch2_resume_logged_op_truncate(struct btree_trans *trans, ret = 0; err: bch2_logged_op_finish(trans, op_k); + bch_err_fn(c, ret); return ret; } @@ -476,6 +477,7 @@ case LOGGED_OP_FINSERT_finish: break; } err: + bch_err_fn(c, ret); bch2_logged_op_finish(trans, op_k); bch2_trans_iter_exit(trans, &iter); return ret; -- cgit v1.2.3-59-g8ed1b From af855a5f5e74cf0ef1166759fca937ce692b4aac Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 23 Mar 2024 19:31:15 -0400 Subject: bcachefs: Resume logged ops after fsck Finishing logged ops requires the filesystem to be in a reasonably consistent state - and other fsck passes don't require it to have completed, so just run it last. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery_passes_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery_passes_types.h b/fs/bcachefs/recovery_passes_types.h index 09c1ab032a67..f30521285706 100644 --- a/fs/bcachefs/recovery_passes_types.h +++ b/fs/bcachefs/recovery_passes_types.h @@ -37,7 +37,6 @@ x(check_subvol_children, 35, PASS_ONLINE|PASS_FSCK) \ x(delete_dead_snapshots, 21, PASS_ONLINE|PASS_FSCK) \ x(fs_upgrade_for_subvolumes, 22, 0) \ - x(resume_logged_ops, 23, PASS_ALWAYS) \ x(check_inodes, 24, PASS_FSCK) \ x(check_extents, 25, PASS_FSCK) \ x(check_indirect_extents, 26, PASS_FSCK) \ @@ -47,6 +46,7 @@ x(check_subvolume_structure, 36, PASS_ONLINE|PASS_FSCK) \ x(check_directory_structure, 30, PASS_ONLINE|PASS_FSCK) \ x(check_nlinks, 31, PASS_FSCK) \ + x(resume_logged_ops, 23, PASS_ALWAYS) \ x(delete_dead_inodes, 32, PASS_FSCK|PASS_UNCLEAN) \ x(fix_reflink_p, 33, 0) \ x(set_fs_needs_rebalance, 34, 0) \ -- cgit v1.2.3-59-g8ed1b From 4fe0eeeae477328cbd26af1e6f81a94e2080ffa8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 28 Mar 2024 02:36:10 -0400 Subject: bcachefs: Flush journal immediately after replay if we did early repair Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index f9e2b011f6d4..dbedb5fd1dde 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -189,6 +189,7 @@ int bch2_journal_replay(struct bch_fs *c) u64 start_seq = c->journal_replay_seq_start; u64 end_seq = c->journal_replay_seq_start; struct btree_trans *trans = bch2_trans_get(c); + bool immediate_flush = false; int ret = 0; if (keys->nr) { @@ -210,6 +211,13 @@ int bch2_journal_replay(struct bch_fs *c) darray_for_each(*keys, k) { cond_resched(); + /* + * k->allocated means the key wasn't read in from the journal, + * rather it was from early repair code + */ + if (k->allocated) + immediate_flush = true; + /* Skip fastpath if we're low on space in the journal */ ret = c->journal.watermark ? -1 : commit_do(trans, NULL, NULL, @@ -269,6 +277,12 @@ int bch2_journal_replay(struct bch_fs *c) bch2_journal_set_replay_done(j); + /* if we did any repair, flush it immediately */ + if (immediate_flush) { + bch2_journal_flush_all_pins(&c->journal); + ret = bch2_journal_meta(&c->journal); + } + if (keys->nr) bch2_journal_log_msg(c, "journal replay finished"); err: @@ -778,6 +792,12 @@ use_clean: clear_bit(BCH_FS_fsck_running, &c->flags); + /* fsync if we fixed errors */ + if (test_bit(BCH_FS_errors_fixed, &c->flags)) { + bch2_journal_flush_all_pins(&c->journal); + bch2_journal_meta(&c->journal); + } + /* If we fixed errors, verify that fs is actually clean now: */ if (IS_ENABLED(CONFIG_BCACHEFS_DEBUG) && test_bit(BCH_FS_errors_fixed, &c->flags) && -- cgit v1.2.3-59-g8ed1b From 0a34c058fca84b10002228a1724e2e613e4dc3cc Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 30 Mar 2024 18:57:53 -0400 Subject: bcachefs: Ensure bch_sb_field_ext always exists This makes bch_sb_field_ext more consistent with the rest of -o nochanges - we don't want to be varying other codepaths based on -o nochanges, since it's used for testing in dry run mode; also fixes some potential null ptr derefs. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 25 ++++++++----------------- fs/bcachefs/super.c | 8 ++++++++ 2 files changed, 16 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index dbedb5fd1dde..e3b06430d606 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -592,16 +592,9 @@ int bch2_fs_recovery(struct bch_fs *c) if (!c->opts.nochanges) { mutex_lock(&c->sb_lock); + struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); bool write_sb = false; - struct bch_sb_field_ext *ext = - bch2_sb_field_get_minsize(&c->disk_sb, ext, sizeof(*ext) / sizeof(u64)); - if (!ext) { - ret = -BCH_ERR_ENOSPC_sb; - mutex_unlock(&c->sb_lock); - goto err; - } - if (BCH_SB_HAS_TOPOLOGY_ERRORS(c->disk_sb.sb)) { ext->recovery_passes_required[0] |= cpu_to_le64(bch2_recovery_passes_to_stable(BIT_ULL(BCH_RECOVERY_PASS_check_topology))); @@ -832,6 +825,7 @@ use_clean: } mutex_lock(&c->sb_lock); + struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); bool write_sb = false; if (BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb) != le16_to_cpu(c->disk_sb.sb->version)) { @@ -845,15 +839,12 @@ use_clean: write_sb = true; } - if (!test_bit(BCH_FS_error, &c->flags)) { - struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); - if (ext && - (!bch2_is_zero(ext->recovery_passes_required, sizeof(ext->recovery_passes_required)) || - !bch2_is_zero(ext->errors_silent, sizeof(ext->errors_silent)))) { - memset(ext->recovery_passes_required, 0, sizeof(ext->recovery_passes_required)); - memset(ext->errors_silent, 0, sizeof(ext->errors_silent)); - write_sb = true; - } + if (!test_bit(BCH_FS_error, &c->flags) && + (!bch2_is_zero(ext->recovery_passes_required, sizeof(ext->recovery_passes_required)) || + !bch2_is_zero(ext->errors_silent, sizeof(ext->errors_silent)))) { + memset(ext->recovery_passes_required, 0, sizeof(ext->recovery_passes_required)); + memset(ext->errors_silent, 0, sizeof(ext->errors_silent)); + write_sb = true; } if (c->opts.fsck && diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 1ad6e5cd9476..89ee5d3ec119 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -1015,8 +1015,16 @@ int bch2_fs_start(struct bch_fs *c) for_each_online_member(c, ca) bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx)->last_mount = cpu_to_le64(now); + struct bch_sb_field_ext *ext = + bch2_sb_field_get_minsize(&c->disk_sb, ext, sizeof(*ext) / sizeof(u64)); mutex_unlock(&c->sb_lock); + if (!ext) { + bch_err(c, "insufficient space in superblock for sb_field_ext"); + ret = -BCH_ERR_ENOSPC_sb; + goto err; + } + for_each_rw_member(c, ca) bch2_dev_allocator_add(c, ca); bch2_recalc_capacity(c); -- cgit v1.2.3-59-g8ed1b From 060ff30a8596b649a80c19935758000dde7855fe Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 29 Mar 2024 20:43:39 -0400 Subject: bcachefs: bch2_run_explicit_recovery_pass_persistent() Flag that we need to run a recovery pass and run it - persistenly, so if we crash it'll still get run. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery_passes.c | 18 ++++++++++++++++++ fs/bcachefs/recovery_passes.h | 1 + 2 files changed, 19 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index bab586dc395a..fb22cce10f66 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -16,6 +16,7 @@ #include "snapshot.h" #include "subvolume.h" #include "super.h" +#include "super-io.h" const char * const bch2_recovery_passes[] = { #define x(_fn, ...) #_fn, @@ -112,6 +113,23 @@ int bch2_run_explicit_recovery_pass(struct bch_fs *c, } } +int bch2_run_explicit_recovery_pass_persistent(struct bch_fs *c, + enum bch_recovery_pass pass) +{ + __le64 s = cpu_to_le64(bch2_recovery_passes_to_stable(BIT_ULL(pass))); + + mutex_lock(&c->sb_lock); + struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); + + if (!(ext->recovery_passes_required[0] & s)) { + ext->recovery_passes_required[0] |= s; + bch2_write_super(c); + } + mutex_unlock(&c->sb_lock); + + return bch2_run_explicit_recovery_pass(c, pass); +} + u64 bch2_fsck_recovery_passes(void) { u64 ret = 0; diff --git a/fs/bcachefs/recovery_passes.h b/fs/bcachefs/recovery_passes.h index abefa67749eb..99b464e127b8 100644 --- a/fs/bcachefs/recovery_passes.h +++ b/fs/bcachefs/recovery_passes.h @@ -9,6 +9,7 @@ u64 bch2_recovery_passes_from_stable(u64 v); u64 bch2_fsck_recovery_passes(void); int bch2_run_explicit_recovery_pass(struct bch_fs *, enum bch_recovery_pass); +int bch2_run_explicit_recovery_pass_persistent(struct bch_fs *, enum bch_recovery_pass); int bch2_run_online_recovery_passes(struct bch_fs *); int bch2_run_recovery_passes(struct bch_fs *); -- cgit v1.2.3-59-g8ed1b From 13c1e583f9179ad7953dc71ebb2f12e613b9d052 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 28 Mar 2024 21:34:14 -0400 Subject: bcachefs: Improve -o norecovery; opts.recovery_pass_limit This adds opts.recovery_pass_limit, and redoes -o norecovery to make use of it; this fixes some issues with -o norecovery so it can be safely used for data recovery. Norecovery means "don't do journal replay"; it's an important data recovery tool when we're getting stuck in journal replay. When using it this way we need to make sure we don't free journal keys after startup, so we continue to overlay them: thus it needs to imply retain_recovery_info, as well as nochanges. recovery_pass_limit is an explicit option for telling recovery to exit after a specific recovery pass; this is a much cleaner way of implementing -o norecovery, as well as being a useful debug feature in its own right. Signed-off-by: Kent Overstreet --- fs/bcachefs/opts.c | 4 ++++ fs/bcachefs/opts.h | 7 ++++++- fs/bcachefs/recovery.c | 10 ++++------ fs/bcachefs/recovery_passes.c | 12 ++++++++---- fs/bcachefs/snapshot.c | 2 +- fs/bcachefs/super.c | 5 +++-- 6 files changed, 26 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/opts.c b/fs/bcachefs/opts.c index 08ea0cfc4aef..e1800c4119b5 100644 --- a/fs/bcachefs/opts.c +++ b/fs/bcachefs/opts.c @@ -7,6 +7,7 @@ #include "disk_groups.h" #include "error.h" #include "opts.h" +#include "recovery_passes.h" #include "super-io.h" #include "util.h" @@ -205,6 +206,9 @@ const struct bch_option bch2_opt_table[] = { #define OPT_STR(_choices) .type = BCH_OPT_STR, \ .min = 0, .max = ARRAY_SIZE(_choices), \ .choices = _choices +#define OPT_STR_NOLIMIT(_choices) .type = BCH_OPT_STR, \ + .min = 0, .max = U64_MAX, \ + .choices = _choices #define OPT_FN(_fn) .type = BCH_OPT_FN, .fn = _fn #define x(_name, _bits, _flags, _type, _sb_opt, _default, _hint, _help) \ diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index 136083c11f3a..084247c13bd8 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -362,7 +362,12 @@ enum fsck_err_opts { OPT_FS|OPT_MOUNT, \ OPT_BOOL(), \ BCH2_NO_SB_OPT, false, \ - NULL, "Don't replay the journal") \ + NULL, "Exit recovery immediately prior to journal replay")\ + x(recovery_pass_last, u8, \ + OPT_FS|OPT_MOUNT, \ + OPT_STR_NOLIMIT(bch2_recovery_passes), \ + BCH2_NO_SB_OPT, 0, \ + NULL, "Exit recovery after specified pass") \ x(keep_journal, u8, \ 0, \ OPT_BOOL(), \ diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index e3b06430d606..e8b434009293 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -269,7 +269,8 @@ int bch2_journal_replay(struct bch_fs *c) bch2_trans_put(trans); trans = NULL; - if (!c->opts.keep_journal) + if (!c->opts.keep_journal && + c->recovery_pass_done >= BCH_RECOVERY_PASS_journal_replay) bch2_journal_keys_put_initial(c); replay_now_at(j, j->replay_journal_seq_end); @@ -584,11 +585,8 @@ int bch2_fs_recovery(struct bch_fs *c) goto err; } - if (c->opts.fsck && c->opts.norecovery) { - bch_err(c, "cannot select both norecovery and fsck"); - ret = -EINVAL; - goto err; - } + if (c->opts.norecovery) + c->opts.recovery_pass_last = BCH_RECOVERY_PASS_journal_replay - 1; if (!c->opts.nochanges) { mutex_lock(&c->sb_lock); diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index fb22cce10f66..0089d9456b10 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -27,7 +27,7 @@ const char * const bch2_recovery_passes[] = { static int bch2_check_allocations(struct bch_fs *c) { - return bch2_gc(c, true, c->opts.norecovery); + return bch2_gc(c, true, false); } static int bch2_set_may_go_rw(struct bch_fs *c) @@ -144,8 +144,6 @@ static bool should_run_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pa { struct recovery_pass_fn *p = recovery_pass_fns + pass; - if (c->opts.norecovery && pass > BCH_RECOVERY_PASS_snapshots_read) - return false; if (c->recovery_passes_explicit & BIT_ULL(pass)) return true; if ((p->when & PASS_FSCK) && c->opts.fsck) @@ -201,6 +199,10 @@ int bch2_run_recovery_passes(struct bch_fs *c) int ret = 0; while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns)) { + if (c->opts.recovery_pass_last && + c->curr_recovery_pass > c->opts.recovery_pass_last) + break; + if (should_run_recovery_pass(c, c->curr_recovery_pass)) { unsigned pass = c->curr_recovery_pass; @@ -213,8 +215,10 @@ int bch2_run_recovery_passes(struct bch_fs *c) c->recovery_passes_complete |= BIT_ULL(c->curr_recovery_pass); } - c->curr_recovery_pass++; + c->recovery_pass_done = max(c->recovery_pass_done, c->curr_recovery_pass); + + c->curr_recovery_pass++; } return ret; diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 4e074136c490..4577ee7939a2 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -131,7 +131,7 @@ bool __bch2_snapshot_is_ancestor(struct bch_fs *c, u32 id, u32 ancestor) rcu_read_lock(); struct snapshot_table *t = rcu_dereference(c->snapshots); - if (unlikely(c->recovery_pass_done <= BCH_RECOVERY_PASS_check_snapshots)) { + if (unlikely(c->recovery_pass_done < BCH_RECOVERY_PASS_check_snapshots)) { ret = __bch2_snapshot_is_ancestor_early(t, id, ancestor); goto out; } diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 89ee5d3ec119..bc026a77eb99 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -365,7 +365,7 @@ void bch2_fs_read_only(struct bch_fs *c) !test_bit(BCH_FS_emergency_ro, &c->flags) && test_bit(BCH_FS_started, &c->flags) && test_bit(BCH_FS_clean_shutdown, &c->flags) && - !c->opts.norecovery) { + c->recovery_pass_done >= BCH_RECOVERY_PASS_journal_replay) { BUG_ON(c->journal.last_empty_seq != journal_cur_seq(&c->journal)); BUG_ON(atomic_read(&c->btree_cache.dirty)); BUG_ON(atomic_long_read(&c->btree_key_cache.nr_dirty)); @@ -510,7 +510,8 @@ err: int bch2_fs_read_write(struct bch_fs *c) { - if (c->opts.norecovery) + if (c->opts.recovery_pass_last && + c->opts.recovery_pass_last < BCH_RECOVERY_PASS_journal_replay) return -BCH_ERR_erofs_norecovery; if (c->opts.nochanges) -- cgit v1.2.3-59-g8ed1b From cecfed9b446da5fba9d73e6448c9f0d1ff5d95ff Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 31 Mar 2024 22:34:45 -0400 Subject: bcachefs: Logged op errors should be ignored If something is wrong with a logged op, we just want to delete it - there's nothing to repair. Signed-off-by: Kent Overstreet --- fs/bcachefs/logged_ops.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/logged_ops.c b/fs/bcachefs/logged_ops.c index 9fac838d123e..b82f8209041f 100644 --- a/fs/bcachefs/logged_ops.c +++ b/fs/bcachefs/logged_ops.c @@ -37,7 +37,6 @@ static int resume_logged_op(struct btree_trans *trans, struct btree_iter *iter, const struct bch_logged_op_fn *fn = logged_op_fn(k.k->type); struct bkey_buf sk; u32 restart_count = trans->restart_count; - int ret; if (!fn) return 0; @@ -45,11 +44,11 @@ static int resume_logged_op(struct btree_trans *trans, struct btree_iter *iter, bch2_bkey_buf_init(&sk); bch2_bkey_buf_reassemble(&sk, c, k); - ret = drop_locks_do(trans, (bch2_fs_lazy_rw(c), 0)) ?: - fn->resume(trans, sk.k) ?: trans_was_restarted(trans, restart_count); + fn->resume(trans, sk.k); bch2_bkey_buf_exit(&sk, c); - return ret; + + return trans_was_restarted(trans, restart_count); } int bch2_resume_logged_ops(struct bch_fs *c) -- cgit v1.2.3-59-g8ed1b From 8ce1db8091b23f5d2a0dd1dabe8007954114cb68 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 1 Apr 2024 00:00:32 -0400 Subject: bcachefs: Fix remove_dirent() We were missing an iter_traverse(). Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index aca6fae409cd..0a47b0a473d8 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -158,9 +158,10 @@ static int __remove_dirent(struct btree_trans *trans, struct bpos pos) bch2_trans_iter_init(trans, &iter, BTREE_ID_dirents, pos, BTREE_ITER_INTENT); - ret = bch2_hash_delete_at(trans, bch2_dirent_hash_desc, - &dir_hash_info, &iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + ret = bch2_btree_iter_traverse(&iter) ?: + bch2_hash_delete_at(trans, bch2_dirent_hash_desc, + &dir_hash_info, &iter, + BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); bch2_trans_iter_exit(trans, &iter); err: bch_err_fn(c, ret); -- cgit v1.2.3-59-g8ed1b From eab3a3ce2dea1a4013a3a553722b85f55a76ac2d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 30 Mar 2024 01:00:50 -0400 Subject: bcachefs: Fix overlapping extent repair overlapping extent repair was colliding with extent past end of inode checks - don't update "extent ends at" until we know we have an extent. Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 0a47b0a473d8..cbb8b43e419f 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -1372,10 +1372,6 @@ static int check_overlapping_extents(struct btree_trans *trans, goto err; } - ret = extent_ends_at(c, extent_ends, seen, k); - if (ret) - goto err; - extent_ends->last_pos = k.k->p; err: return ret; @@ -1505,6 +1501,12 @@ static int check_extent(struct btree_trans *trans, struct btree_iter *iter, i->seen_this_pos = true; } + + if (k.k->type != KEY_TYPE_whiteout) { + ret = extent_ends_at(c, extent_ends, s, k); + if (ret) + goto err; + } out: err: fsck_err: -- cgit v1.2.3-59-g8ed1b From b3c7fd35c03c17a950737fb56a06b730a7962d28 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 30 Mar 2024 15:59:57 -0400 Subject: bcachefs: On emergency shutdown, print out current journal sequence number Signed-off-by: Kent Overstreet --- fs/bcachefs/error.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/error.c b/fs/bcachefs/error.c index f942a3947496..82a6656c941c 100644 --- a/fs/bcachefs/error.c +++ b/fs/bcachefs/error.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" #include "error.h" +#include "journal.h" #include "recovery_passes.h" #include "super.h" #include "thread_with_file.h" @@ -16,7 +17,8 @@ bool bch2_inconsistent_error(struct bch_fs *c) return false; case BCH_ON_ERROR_ro: if (bch2_fs_emergency_read_only(c)) - bch_err(c, "inconsistency detected - emergency read only"); + bch_err(c, "inconsistency detected - emergency read only at journal seq %llu", + journal_cur_seq(&c->journal)); return true; case BCH_ON_ERROR_panic: panic(bch2_fmt(c, "panic after error")); -- cgit v1.2.3-59-g8ed1b From e23d7e82b707d1d0a627e334fb46370e4f772c11 Mon Sep 17 00:00:00 2001 From: Andrey Albershteyn Date: Thu, 14 Mar 2024 18:07:02 +0100 Subject: xfs: allow cross-linking special files without project quota There's an issue that if special files is created before quota project is enabled, then it's not possible to link this file. This works fine for normal files. This happens because xfs_quota skips special files (no ioctls to set necessary flags). The check for having the same project ID for source and destination then fails as source file doesn't have any ID. mkfs.xfs -f /dev/sda mount -o prjquota /dev/sda /mnt/test mkdir /mnt/test/foo mkfifo /mnt/test/foo/fifo1 xfs_quota -xc "project -sp /mnt/test/foo 9" /mnt/test > Setting up project 9 (path /mnt/test/foo)... > xfs_quota: skipping special file /mnt/test/foo/fifo1 > Processed 1 (/etc/projects and cmdline) paths for project 9 with recursion depth infinite (-1). ln /mnt/test/foo/fifo1 /mnt/test/foo/fifo1_link > ln: failed to create hard link '/mnt/test/testdir/fifo1_link' => '/mnt/test/testdir/fifo1': Invalid cross-device link mkfifo /mnt/test/foo/fifo2 ln /mnt/test/foo/fifo2 /mnt/test/foo/fifo2_link Fix this by allowing linking of special files to the project quota if special files doesn't have any ID set (ID = 0). Signed-off-by: Andrey Albershteyn Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_inode.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index ea48774f6b76..d55b42b2480d 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1301,8 +1301,19 @@ xfs_link( */ if (unlikely((tdp->i_diflags & XFS_DIFLAG_PROJINHERIT) && tdp->i_projid != sip->i_projid)) { - error = -EXDEV; - goto error_return; + /* + * Project quota setup skips special files which can + * leave inodes in a PROJINHERIT directory without a + * project ID set. We need to allow links to be made + * to these "project-less" inodes because userspace + * expects them to succeed after project ID setup, + * but everything else should be rejected. + */ + if (!special_file(VFS_I(sip)->i_mode) || + sip->i_projid != 0) { + error = -EXDEV; + goto error_return; + } } if (!resblks) { -- cgit v1.2.3-59-g8ed1b From 3f5ef5109f6a054ce58b3bec7214ed76c9cc269f Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Wed, 27 Mar 2024 16:23:46 +0100 Subject: fs/ntfs3: add prefix to bitmap_size() and use BITS_TO_U64() bitmap_size() is a pretty generic name and one may want to use it for a generic bitmap API function. At the same time, its logic is NTFS-specific, as it aligns to the sizeof(u64), not the sizeof(long) (although it uses ideologically right ALIGN() instead of division). Add the prefix 'ntfs3_' used for that FS (not just 'ntfs_' to not mix it with the legacy module) and use generic BITS_TO_U64() while at it. Suggested-by: Yury Norov # BITS_TO_U64() Reviewed-by: Przemek Kitszel Reviewed-by: Yury Norov Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller --- fs/ntfs3/bitmap.c | 4 ++-- fs/ntfs3/fsntfs.c | 2 +- fs/ntfs3/index.c | 11 ++++++----- fs/ntfs3/ntfs_fs.h | 4 ++-- fs/ntfs3/super.c | 2 +- 5 files changed, 12 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/bitmap.c b/fs/ntfs3/bitmap.c index 845f9b22deef..c9eb01ccee51 100644 --- a/fs/ntfs3/bitmap.c +++ b/fs/ntfs3/bitmap.c @@ -654,7 +654,7 @@ int wnd_init(struct wnd_bitmap *wnd, struct super_block *sb, size_t nbits) wnd->total_zeroes = nbits; wnd->extent_max = MINUS_ONE_T; wnd->zone_bit = wnd->zone_end = 0; - wnd->nwnd = bytes_to_block(sb, bitmap_size(nbits)); + wnd->nwnd = bytes_to_block(sb, ntfs3_bitmap_size(nbits)); wnd->bits_last = nbits & (wbits - 1); if (!wnd->bits_last) wnd->bits_last = wbits; @@ -1347,7 +1347,7 @@ int wnd_extend(struct wnd_bitmap *wnd, size_t new_bits) return -EINVAL; /* Align to 8 byte boundary. */ - new_wnd = bytes_to_block(sb, bitmap_size(new_bits)); + new_wnd = bytes_to_block(sb, ntfs3_bitmap_size(new_bits)); new_last = new_bits & (wbits - 1); if (!new_last) new_last = wbits; diff --git a/fs/ntfs3/fsntfs.c b/fs/ntfs3/fsntfs.c index ae2ef5c11868..626d3f2c7e2d 100644 --- a/fs/ntfs3/fsntfs.c +++ b/fs/ntfs3/fsntfs.c @@ -522,7 +522,7 @@ static int ntfs_extend_mft(struct ntfs_sb_info *sbi) ni->mi.dirty = true; /* Step 2: Resize $MFT::BITMAP. */ - new_bitmap_bytes = bitmap_size(new_mft_total); + new_bitmap_bytes = ntfs3_bitmap_size(new_mft_total); err = attr_set_size(ni, ATTR_BITMAP, NULL, 0, &sbi->mft.bitmap.run, new_bitmap_bytes, &new_bitmap_bytes, true, NULL); diff --git a/fs/ntfs3/index.c b/fs/ntfs3/index.c index daabaad63aaf..43796aaa3d97 100644 --- a/fs/ntfs3/index.c +++ b/fs/ntfs3/index.c @@ -1456,8 +1456,8 @@ static int indx_create_allocate(struct ntfs_index *indx, struct ntfs_inode *ni, alloc->nres.valid_size = alloc->nres.data_size = cpu_to_le64(data_size); - err = ni_insert_resident(ni, bitmap_size(1), ATTR_BITMAP, in->name, - in->name_len, &bitmap, NULL, NULL); + err = ni_insert_resident(ni, ntfs3_bitmap_size(1), ATTR_BITMAP, + in->name, in->name_len, &bitmap, NULL, NULL); if (err) goto out2; @@ -1518,8 +1518,9 @@ static int indx_add_allocate(struct ntfs_index *indx, struct ntfs_inode *ni, if (bmp) { /* Increase bitmap. */ err = attr_set_size(ni, ATTR_BITMAP, in->name, in->name_len, - &indx->bitmap_run, bitmap_size(bit + 1), - NULL, true, NULL); + &indx->bitmap_run, + ntfs3_bitmap_size(bit + 1), NULL, true, + NULL); if (err) goto out1; } @@ -2092,7 +2093,7 @@ static int indx_shrink(struct ntfs_index *indx, struct ntfs_inode *ni, if (in->name == I30_NAME) i_size_write(&ni->vfs_inode, new_data); - bpb = bitmap_size(bit); + bpb = ntfs3_bitmap_size(bit); if (bpb * 8 == nbits) return 0; diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h index 79356fd29a14..ea5b5e814e63 100644 --- a/fs/ntfs3/ntfs_fs.h +++ b/fs/ntfs3/ntfs_fs.h @@ -966,9 +966,9 @@ static inline bool run_is_empty(struct runs_tree *run) } /* NTFS uses quad aligned bitmaps. */ -static inline size_t bitmap_size(size_t bits) +static inline size_t ntfs3_bitmap_size(size_t bits) { - return ALIGN((bits + 7) >> 3, 8); + return BITS_TO_U64(bits) * sizeof(u64); } #define _100ns2seconds 10000000 diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c index 9df7c20d066f..71dfeb0c4323 100644 --- a/fs/ntfs3/super.c +++ b/fs/ntfs3/super.c @@ -1341,7 +1341,7 @@ static int ntfs_fill_super(struct super_block *sb, struct fs_context *fc) /* Check bitmap boundary. */ tt = sbi->used.bitmap.nbits; - if (inode->i_size < bitmap_size(tt)) { + if (inode->i_size < ntfs3_bitmap_size(tt)) { ntfs_err(sb, "$Bitmap is corrupted."); err = -EINVAL; goto put_inode_out; -- cgit v1.2.3-59-g8ed1b From 4ca532d64648d4776d15512caed3efea05ca7195 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Wed, 27 Mar 2024 16:23:47 +0100 Subject: btrfs: rename bitmap_set_bits() -> btrfs_bitmap_set_bits() bitmap_set_bits() does not start with the FS' prefix and may collide with a new generic helper one day. It operates with the FS-specific types, so there's no change those two could do the same thing. Just add the prefix to exclude such possible conflict. Reviewed-by: Przemek Kitszel Acked-by: David Sterba Reviewed-by: Yury Norov Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller --- fs/btrfs/free-space-cache.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index c8a05d5eb9cb..3ab8dea5036b 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -1911,9 +1911,9 @@ static inline void bitmap_clear_bits(struct btrfs_free_space_ctl *ctl, ctl->free_space -= bytes; } -static void bitmap_set_bits(struct btrfs_free_space_ctl *ctl, - struct btrfs_free_space *info, u64 offset, - u64 bytes) +static void btrfs_bitmap_set_bits(struct btrfs_free_space_ctl *ctl, + struct btrfs_free_space *info, u64 offset, + u64 bytes) { unsigned long start, count, end; int extent_delta = 1; @@ -2249,7 +2249,7 @@ static u64 add_bytes_to_bitmap(struct btrfs_free_space_ctl *ctl, bytes_to_set = min(end - offset, bytes); - bitmap_set_bits(ctl, info, offset, bytes_to_set); + btrfs_bitmap_set_bits(ctl, info, offset, bytes_to_set); return bytes_to_set; -- cgit v1.2.3-59-g8ed1b From fa1af65bf8e2ded0650c2cfd9e8fcca73835485e Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 28 Mar 2024 09:41:47 +0100 Subject: btrfs use bio_list_merge_init Use bio_list_merge_init instead of open coding bio_list_merge and bio_list_init. Signed-off-by: Christoph Hellwig Acked-by: David Sterba Reviewed-by: Johannes Thumshirn Reviewed-by: Damien Le Moal Link: https://lore.kernel.org/r/20240328084147.2954434-5-hch@lst.de Signed-off-by: Jens Axboe --- fs/btrfs/raid56.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index 6f4a9cfeea44..831fac45e70f 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -331,12 +331,11 @@ static void steal_rbio(struct btrfs_raid_bio *src, struct btrfs_raid_bio *dest) static void merge_rbio(struct btrfs_raid_bio *dest, struct btrfs_raid_bio *victim) { - bio_list_merge(&dest->bio_list, &victim->bio_list); + bio_list_merge_init(&dest->bio_list, &victim->bio_list); dest->bio_list_bytes += victim->bio_list_bytes; /* Also inherit the bitmaps from @victim. */ bitmap_or(&dest->dbitmap, &victim->dbitmap, &dest->dbitmap, dest->stripe_nsectors); - bio_list_init(&victim->bio_list); } /* -- cgit v1.2.3-59-g8ed1b From ad191e0eeebf64a60ca2d16ca01a223d2b1dd25e Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 28 Mar 2024 11:48:33 -0400 Subject: dlm: fix user space lock decision to copy lvb This patch fixes the copy lvb decision for user space lock requests. Checking dlm_lvb_operations is done earlier, where granted/requested lock modes are available to use in the matrix. The decision had been moved to the wrong location, where granted mode and requested mode where the same, which causes the dlm_lvb_operations matix to produce the wrong copy decision. For PW or EX requests, the caller could get invalid lvb data. Fixes: 61bed0baa4db ("fs: dlm: use a non-static queue for callbacks") Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/ast.c | 14 ++++++++++++++ fs/dlm/dlm_internal.h | 1 + fs/dlm/user.c | 15 ++------------- 3 files changed, 17 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index 1f2f70a1b824..decedc4ee15f 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -12,6 +12,7 @@ #include #include "dlm_internal.h" +#include "lvb_table.h" #include "memory.h" #include "lock.h" #include "user.h" @@ -42,6 +43,7 @@ int dlm_enqueue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, struct dlm_ls *ls = lkb->lkb_resource->res_ls; int rv = DLM_ENQUEUE_CALLBACK_SUCCESS; struct dlm_callback *cb; + int copy_lvb = 0; int prev_mode; if (flags & DLM_CB_BAST) { @@ -73,6 +75,17 @@ int dlm_enqueue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, goto out; } } + } else if (flags & DLM_CB_CAST) { + if (test_bit(DLM_DFL_USER_BIT, &lkb->lkb_dflags)) { + if (lkb->lkb_last_cast) + prev_mode = lkb->lkb_last_cb->mode; + else + prev_mode = -1; + + if (!status && lkb->lkb_lksb->sb_lvbptr && + dlm_lvb_operations[prev_mode + 1][mode + 1]) + copy_lvb = 1; + } } cb = dlm_allocate_cb(); @@ -85,6 +98,7 @@ int dlm_enqueue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, cb->mode = mode; cb->sb_status = status; cb->sb_flags = (sbflags & 0x000000FF); + cb->copy_lvb = copy_lvb; kref_init(&cb->ref); if (!test_and_set_bit(DLM_IFL_CB_PENDING_BIT, &lkb->lkb_iflags)) rv = DLM_ENQUEUE_CALLBACK_NEED_SCHED; diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 3b4dbce849f0..a9137c90f348 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -222,6 +222,7 @@ struct dlm_callback { int sb_status; /* copy to lksb status */ uint8_t sb_flags; /* copy to lksb flags */ int8_t mode; /* rq mode of bast, gr mode of cast */ + int copy_lvb; struct list_head list; struct kref ref; diff --git a/fs/dlm/user.c b/fs/dlm/user.c index 9f9b68448830..12a483deeef5 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -21,7 +21,6 @@ #include "dlm_internal.h" #include "lockspace.h" #include "lock.h" -#include "lvb_table.h" #include "user.h" #include "ast.h" #include "config.h" @@ -806,8 +805,7 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, struct dlm_lkb *lkb; DECLARE_WAITQUEUE(wait, current); struct dlm_callback *cb; - int rv, ret, copy_lvb = 0; - int old_mode, new_mode; + int rv, ret; if (count == sizeof(struct dlm_device_version)) { rv = copy_version_to_user(buf, count); @@ -864,9 +862,6 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, lkb = list_first_entry(&proc->asts, struct dlm_lkb, lkb_cb_list); - /* rem_lkb_callback sets a new lkb_last_cast */ - old_mode = lkb->lkb_last_cast->mode; - rv = dlm_dequeue_lkb_callback(lkb, &cb); switch (rv) { case DLM_DEQUEUE_CALLBACK_EMPTY: @@ -895,12 +890,6 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, if (cb->flags & DLM_CB_BAST) { trace_dlm_bast(lkb->lkb_resource->res_ls, lkb, cb->mode); } else if (cb->flags & DLM_CB_CAST) { - new_mode = cb->mode; - - if (!cb->sb_status && lkb->lkb_lksb->sb_lvbptr && - dlm_lvb_operations[old_mode + 1][new_mode + 1]) - copy_lvb = 1; - lkb->lkb_lksb->sb_status = cb->sb_status; lkb->lkb_lksb->sb_flags = cb->sb_flags; trace_dlm_ast(lkb->lkb_resource->res_ls, lkb); @@ -908,7 +897,7 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, ret = copy_result_to_user(lkb->lkb_ua, test_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags), - cb->flags, cb->mode, copy_lvb, buf, count); + cb->flags, cb->mode, cb->copy_lvb, buf, count); kref_put(&cb->ref, dlm_release_callback); -- cgit v1.2.3-59-g8ed1b From 609ed5bde2bbe55b78142740de1451ece9896a84 Mon Sep 17 00:00:00 2001 From: Kunwu Chan Date: Thu, 28 Mar 2024 11:48:34 -0400 Subject: dlm: Simplify the allocation of slab caches in dlm_midcomms_cache_create Use the new KMEM_CACHE() macro instead of direct kmem_cache_create to simplify the creation of SLAB caches. Signed-off-by: Kunwu Chan Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/midcomms.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/dlm/midcomms.c b/fs/dlm/midcomms.c index 2247ebb61be1..8e9920f1b48b 100644 --- a/fs/dlm/midcomms.c +++ b/fs/dlm/midcomms.c @@ -226,8 +226,7 @@ static DEFINE_MUTEX(close_lock); struct kmem_cache *dlm_midcomms_cache_create(void) { - return kmem_cache_create("dlm_mhandle", sizeof(struct dlm_mhandle), - 0, 0, NULL); + return KMEM_CACHE(dlm_mhandle, 0); } static inline const char *dlm_state_str(int state) -- cgit v1.2.3-59-g8ed1b From 1131f339089bdf7ef7aa0a026bdefe1c9a22e8a1 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 28 Mar 2024 11:48:37 -0400 Subject: dlm: remove lkb from callback tracepoints Stop using lkb structs in the callback tracepoints so that lkb references are not needed. This prepares for separating lkb structs from callbacks. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/ast.c | 9 +++++++-- fs/dlm/user.c | 9 +++++++-- include/trace/events/dlm.h | 46 +++++++++++++++++++--------------------------- 3 files changed, 33 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index decedc4ee15f..dd7cca3c1472 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -171,6 +171,7 @@ void dlm_callback_work(struct work_struct *work) { struct dlm_lkb *lkb = container_of(work, struct dlm_lkb, lkb_cb_work); struct dlm_ls *ls = lkb->lkb_resource->res_ls; + struct dlm_rsb *rsb = lkb->lkb_resource; void (*castfn) (void *astparam); void (*bastfn) (void *astparam, int mode); struct dlm_callback *cb; @@ -190,14 +191,18 @@ void dlm_callback_work(struct work_struct *work) bastfn = lkb->lkb_bastfn; if (cb->flags & DLM_CB_BAST) { - trace_dlm_bast(ls, lkb, cb->mode); + trace_dlm_bast(ls->ls_global_id, lkb->lkb_id, + cb->mode, rsb->res_name, + rsb->res_length); lkb->lkb_last_bast_time = ktime_get(); lkb->lkb_last_bast_mode = cb->mode; bastfn(lkb->lkb_astparam, cb->mode); } else if (cb->flags & DLM_CB_CAST) { lkb->lkb_lksb->sb_status = cb->sb_status; lkb->lkb_lksb->sb_flags = cb->sb_flags; - trace_dlm_ast(ls, lkb); + trace_dlm_ast(ls->ls_global_id, lkb->lkb_id, + cb->sb_flags, cb->sb_status, + rsb->res_name, rsb->res_length); lkb->lkb_last_cast_time = ktime_get(); castfn(lkb->lkb_astparam); } diff --git a/fs/dlm/user.c b/fs/dlm/user.c index 12a483deeef5..6f99bbeeac9b 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -805,6 +805,7 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, struct dlm_lkb *lkb; DECLARE_WAITQUEUE(wait, current); struct dlm_callback *cb; + struct dlm_rsb *rsb; int rv, ret; if (count == sizeof(struct dlm_device_version)) { @@ -887,12 +888,16 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, } spin_unlock(&proc->asts_spin); + rsb = lkb->lkb_resource; if (cb->flags & DLM_CB_BAST) { - trace_dlm_bast(lkb->lkb_resource->res_ls, lkb, cb->mode); + trace_dlm_bast(rsb->res_ls->ls_global_id, lkb->lkb_id, + cb->mode, rsb->res_name, rsb->res_length); } else if (cb->flags & DLM_CB_CAST) { lkb->lkb_lksb->sb_status = cb->sb_status; lkb->lkb_lksb->sb_flags = cb->sb_flags; - trace_dlm_ast(lkb->lkb_resource->res_ls, lkb); + trace_dlm_ast(rsb->res_ls->ls_global_id, lkb->lkb_id, + cb->sb_flags, cb->sb_status, rsb->res_name, + rsb->res_length); } ret = copy_result_to_user(lkb->lkb_ua, diff --git a/include/trace/events/dlm.h b/include/trace/events/dlm.h index c1a146f9fc91..af160082c9e3 100644 --- a/include/trace/events/dlm.h +++ b/include/trace/events/dlm.h @@ -189,29 +189,25 @@ TRACE_EVENT(dlm_lock_end, TRACE_EVENT(dlm_bast, - TP_PROTO(struct dlm_ls *ls, struct dlm_lkb *lkb, int mode), + TP_PROTO(__u32 ls_id, __u32 lkb_id, int mode, + const char *res_name, size_t res_length), - TP_ARGS(ls, lkb, mode), + TP_ARGS(ls_id, lkb_id, mode, res_name, res_length), TP_STRUCT__entry( __field(__u32, ls_id) __field(__u32, lkb_id) __field(int, mode) - __dynamic_array(unsigned char, res_name, - lkb->lkb_resource ? lkb->lkb_resource->res_length : 0) + __dynamic_array(unsigned char, res_name, res_length) ), TP_fast_assign( - struct dlm_rsb *r; - - __entry->ls_id = ls->ls_global_id; - __entry->lkb_id = lkb->lkb_id; + __entry->ls_id = ls_id; + __entry->lkb_id = lkb_id; __entry->mode = mode; - r = lkb->lkb_resource; - if (r) - memcpy(__get_dynamic_array(res_name), r->res_name, - __get_dynamic_array_len(res_name)); + memcpy(__get_dynamic_array(res_name), res_name, + __get_dynamic_array_len(res_name)); ), TP_printk("ls_id=%u lkb_id=%x mode=%s res_name=%s", @@ -224,31 +220,27 @@ TRACE_EVENT(dlm_bast, TRACE_EVENT(dlm_ast, - TP_PROTO(struct dlm_ls *ls, struct dlm_lkb *lkb), + TP_PROTO(__u32 ls_id, __u32 lkb_id, __u8 sb_flags, int sb_status, + const char *res_name, size_t res_length), - TP_ARGS(ls, lkb), + TP_ARGS(ls_id, lkb_id, sb_flags, sb_status, res_name, res_length), TP_STRUCT__entry( __field(__u32, ls_id) __field(__u32, lkb_id) - __field(u8, sb_flags) + __field(__u8, sb_flags) __field(int, sb_status) - __dynamic_array(unsigned char, res_name, - lkb->lkb_resource ? lkb->lkb_resource->res_length : 0) + __dynamic_array(unsigned char, res_name, res_length) ), TP_fast_assign( - struct dlm_rsb *r; - - __entry->ls_id = ls->ls_global_id; - __entry->lkb_id = lkb->lkb_id; - __entry->sb_flags = lkb->lkb_lksb->sb_flags; - __entry->sb_status = lkb->lkb_lksb->sb_status; + __entry->ls_id = ls_id; + __entry->lkb_id = lkb_id; + __entry->sb_flags = sb_flags; + __entry->sb_status = sb_status; - r = lkb->lkb_resource; - if (r) - memcpy(__get_dynamic_array(res_name), r->res_name, - __get_dynamic_array_len(res_name)); + memcpy(__get_dynamic_array(res_name), res_name, + __get_dynamic_array_len(res_name)); ), TP_printk("ls_id=%u lkb_id=%x sb_flags=%s sb_status=%d res_name=%s", -- cgit v1.2.3-59-g8ed1b From 16e98462b764002b0f747dc93e5d03bd65d0b019 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 28 Mar 2024 11:48:38 -0400 Subject: dlm: remove callback queue debugfs functionality Remove the ability to dump pending lkb callbacks from debugfs. The prepares for separating lkb structs from callbacks. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/debug_fs.c | 96 ------------------------------------------------------- 1 file changed, 96 deletions(-) (limited to 'fs') diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 4fa11d9ddbb6..289d959c7700 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -366,52 +366,6 @@ static void print_format4(struct dlm_rsb *r, struct seq_file *s) unlock_rsb(r); } -static void print_format5_lock(struct seq_file *s, struct dlm_lkb *lkb) -{ - struct dlm_callback *cb; - - /* lkb_id lkb_flags mode flags sb_status sb_flags */ - - spin_lock(&lkb->lkb_cb_lock); - list_for_each_entry(cb, &lkb->lkb_callbacks, list) { - seq_printf(s, "%x %x %d %x %d %x\n", - lkb->lkb_id, - dlm_iflags_val(lkb), - cb->mode, - cb->flags, - cb->sb_status, - cb->sb_flags); - } - spin_unlock(&lkb->lkb_cb_lock); -} - -static void print_format5(struct dlm_rsb *r, struct seq_file *s) -{ - struct dlm_lkb *lkb; - - lock_rsb(r); - - list_for_each_entry(lkb, &r->res_grantqueue, lkb_statequeue) { - print_format5_lock(s, lkb); - if (seq_has_overflowed(s)) - goto out; - } - - list_for_each_entry(lkb, &r->res_convertqueue, lkb_statequeue) { - print_format5_lock(s, lkb); - if (seq_has_overflowed(s)) - goto out; - } - - list_for_each_entry(lkb, &r->res_waitqueue, lkb_statequeue) { - print_format5_lock(s, lkb); - if (seq_has_overflowed(s)) - goto out; - } - out: - unlock_rsb(r); -} - struct rsbtbl_iter { struct dlm_rsb *rsb; unsigned bucket; @@ -455,13 +409,6 @@ static int table_seq_show(struct seq_file *seq, void *iter_ptr) } print_format4(ri->rsb, seq); break; - case 5: - if (ri->header) { - seq_puts(seq, "lkb_id lkb_flags mode flags sb_status sb_flags\n"); - ri->header = 0; - } - print_format5(ri->rsb, seq); - break; } return 0; @@ -471,7 +418,6 @@ static const struct seq_operations format1_seq_ops; static const struct seq_operations format2_seq_ops; static const struct seq_operations format3_seq_ops; static const struct seq_operations format4_seq_ops; -static const struct seq_operations format5_seq_ops; static void *table_seq_start(struct seq_file *seq, loff_t *pos) { @@ -503,8 +449,6 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) ri->format = 3; if (seq->op == &format4_seq_ops) ri->format = 4; - if (seq->op == &format5_seq_ops) - ri->format = 5; tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; @@ -659,18 +603,10 @@ static const struct seq_operations format4_seq_ops = { .show = table_seq_show, }; -static const struct seq_operations format5_seq_ops = { - .start = table_seq_start, - .next = table_seq_next, - .stop = table_seq_stop, - .show = table_seq_show, -}; - static const struct file_operations format1_fops; static const struct file_operations format2_fops; static const struct file_operations format3_fops; static const struct file_operations format4_fops; -static const struct file_operations format5_fops; static int table_open1(struct inode *inode, struct file *file) { @@ -757,20 +693,6 @@ static int table_open4(struct inode *inode, struct file *file) return 0; } -static int table_open5(struct inode *inode, struct file *file) -{ - struct seq_file *seq; - int ret; - - ret = seq_open(file, &format5_seq_ops); - if (ret) - return ret; - - seq = file->private_data; - seq->private = inode->i_private; /* the dlm_ls */ - return 0; -} - static const struct file_operations format1_fops = { .owner = THIS_MODULE, .open = table_open1, @@ -804,14 +726,6 @@ static const struct file_operations format4_fops = { .release = seq_release }; -static const struct file_operations format5_fops = { - .owner = THIS_MODULE, - .open = table_open5, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release -}; - /* * dump lkb's on the ls_waiters list */ @@ -1021,16 +935,6 @@ void dlm_create_debug_file(struct dlm_ls *ls) dlm_root, ls, &waiters_fops); - - /* format 5 */ - - snprintf(name, sizeof(name), "%s_queued_asts", ls->ls_name); - - ls->ls_debug_queued_asts_dentry = debugfs_create_file(name, - 0644, - dlm_root, - ls, - &format5_fops); } void __init dlm_register_debugfs(void) -- cgit v1.2.3-59-g8ed1b From 4ed424280942f40b033d669eb9204a52e6d41639 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 28 Mar 2024 11:48:39 -0400 Subject: dlm: save callback debug info earlier Save lkb callback info when queueing the callback so that the lkb struct is not needed in the callback workqueue processing. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/ast.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index dd7cca3c1472..cadbcbe0786b 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -105,8 +105,13 @@ int dlm_enqueue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, list_add_tail(&cb->list, &lkb->lkb_callbacks); - if (flags & DLM_CB_CAST) + if (flags & DLM_CB_BAST) { + lkb->lkb_last_bast_time = ktime_get(); + lkb->lkb_last_bast_mode = cb->mode; + } else if (flags & DLM_CB_CAST) { dlm_callback_set_last_ptr(&lkb->lkb_last_cast, cb); + lkb->lkb_last_cast_time = ktime_get(); + } dlm_callback_set_last_ptr(&lkb->lkb_last_cb, cb); @@ -194,8 +199,6 @@ void dlm_callback_work(struct work_struct *work) trace_dlm_bast(ls->ls_global_id, lkb->lkb_id, cb->mode, rsb->res_name, rsb->res_length); - lkb->lkb_last_bast_time = ktime_get(); - lkb->lkb_last_bast_mode = cb->mode; bastfn(lkb->lkb_astparam, cb->mode); } else if (cb->flags & DLM_CB_CAST) { lkb->lkb_lksb->sb_status = cb->sb_status; @@ -203,7 +206,6 @@ void dlm_callback_work(struct work_struct *work) trace_dlm_ast(ls->ls_global_id, lkb->lkb_id, cb->sb_flags, cb->sb_status, rsb->res_name, rsb->res_length); - lkb->lkb_last_cast_time = ktime_get(); castfn(lkb->lkb_astparam); } -- cgit v1.2.3-59-g8ed1b From 0175e51b5134b55c89364aae68ec16271c67e472 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 28 Mar 2024 11:48:40 -0400 Subject: dlm: combine switch case fail and default statements This patch combines the failure and default cases for enqueue and dequeue a callback to the lkb callback queue that should end in both cases as it should never happen. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/ast.c | 5 ++--- fs/dlm/user.c | 9 ++++----- 2 files changed, 6 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index cadbcbe0786b..5ea0b62f276b 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -160,11 +160,10 @@ void dlm_add_cb(struct dlm_lkb *lkb, uint32_t flags, int mode, int status, } spin_unlock(&ls->ls_cb_lock); break; - case DLM_ENQUEUE_CALLBACK_FAILURE: - WARN_ON_ONCE(1); - break; case DLM_ENQUEUE_CALLBACK_SUCCESS: break; + case DLM_ENQUEUE_CALLBACK_FAILURE: + fallthrough; default: WARN_ON_ONCE(1); break; diff --git a/fs/dlm/user.c b/fs/dlm/user.c index 6f99bbeeac9b..fa99b6074e5c 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -231,10 +231,6 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode, rv = dlm_enqueue_lkb_callback(lkb, flags, mode, status, sbflags); switch (rv) { - case DLM_ENQUEUE_CALLBACK_FAILURE: - spin_unlock(&proc->asts_spin); - WARN_ON_ONCE(1); - goto out; case DLM_ENQUEUE_CALLBACK_NEED_SCHED: kref_get(&lkb->lkb_ref); list_add_tail(&lkb->lkb_cb_list, &proc->asts); @@ -242,9 +238,12 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode, break; case DLM_ENQUEUE_CALLBACK_SUCCESS: break; + case DLM_ENQUEUE_CALLBACK_FAILURE: + fallthrough; default: + spin_unlock(&proc->asts_spin); WARN_ON_ONCE(1); - break; + goto out; } spin_unlock(&proc->asts_spin); -- cgit v1.2.3-59-g8ed1b From 986ae3c2a8dfc1e229cabe9cc2e0b01b721c8980 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 28 Mar 2024 11:48:41 -0400 Subject: dlm: fix race between final callback and remove This patch fixes the following issue: node 1 is dir node 2 is master node 3 is other 1->2: unlock 2: put final lkb, rsb moved to toss 2->1: unlock_reply 1: queue lkb callback with EUNLOCK 2->1: remove 1: receive_remove ignored (rsb on keep because of queued lkb callback) 1: complete lkb callback, put_lkb, move rsb to toss 3->1: lookup 1->3: lookup_reply master=2 3->2: request 2->3: request_reply EBADR In summary: An unexpected lkb reference causes the rsb to remain on the wrong list. The rsb being on the wrong list causes receive_remove to be ignored. An ignored receive_remove causes inconsistent dir and master state. This sequence requires an unusually long delay in delivering the unlock callback, because the remove message from 2->1 usually happens after some seconds. So, it's not known exactly how frequently this sequence occurs in pratice. It's possible that the same end result could also have another unknown cause. The solution for this issue is to further separate callback state from the lkb, so that an lkb reference (and from that, an rsb ref) are not held while a callback remains queued. Then, within the unlock_reply, the lkb will be freed and the rsb moved to the toss list. So, the receive_remove will not be ignored. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/ast.c | 166 +++++++++++++++++++------------------------------- fs/dlm/ast.h | 10 +-- fs/dlm/dlm_internal.h | 60 ++++++++++-------- fs/dlm/lock.c | 20 +++--- fs/dlm/user.c | 86 +++++++------------------- 5 files changed, 129 insertions(+), 213 deletions(-) (limited to 'fs') diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index 5ea0b62f276b..e9da812352b4 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -37,12 +37,32 @@ void dlm_callback_set_last_ptr(struct dlm_callback **from, *from = to; } -int dlm_enqueue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, - int status, uint32_t sbflags) +static void dlm_callback_work(struct work_struct *work) { - struct dlm_ls *ls = lkb->lkb_resource->res_ls; + struct dlm_callback *cb = container_of(work, struct dlm_callback, work); + + if (cb->flags & DLM_CB_BAST) { + trace_dlm_bast(cb->ls_id, cb->lkb_id, cb->mode, cb->res_name, + cb->res_length); + cb->bastfn(cb->astparam, cb->mode); + } else if (cb->flags & DLM_CB_CAST) { + trace_dlm_ast(cb->ls_id, cb->lkb_id, cb->sb_status, + cb->sb_flags, cb->res_name, cb->res_length); + cb->lkb_lksb->sb_status = cb->sb_status; + cb->lkb_lksb->sb_flags = cb->sb_flags; + cb->astfn(cb->astparam); + } + + kref_put(&cb->ref, dlm_release_callback); +} + +int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, + int status, uint32_t sbflags, + struct dlm_callback **cb) +{ + struct dlm_rsb *rsb = lkb->lkb_resource; int rv = DLM_ENQUEUE_CALLBACK_SUCCESS; - struct dlm_callback *cb; + struct dlm_ls *ls = rsb->res_ls; int copy_lvb = 0; int prev_mode; @@ -88,57 +108,46 @@ int dlm_enqueue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, } } - cb = dlm_allocate_cb(); - if (!cb) { + *cb = dlm_allocate_cb(); + if (!*cb) { rv = DLM_ENQUEUE_CALLBACK_FAILURE; goto out; } - cb->flags = flags; - cb->mode = mode; - cb->sb_status = status; - cb->sb_flags = (sbflags & 0x000000FF); - cb->copy_lvb = copy_lvb; - kref_init(&cb->ref); - if (!test_and_set_bit(DLM_IFL_CB_PENDING_BIT, &lkb->lkb_iflags)) - rv = DLM_ENQUEUE_CALLBACK_NEED_SCHED; + /* for tracing */ + (*cb)->lkb_id = lkb->lkb_id; + (*cb)->ls_id = ls->ls_global_id; + memcpy((*cb)->res_name, rsb->res_name, rsb->res_length); + (*cb)->res_length = rsb->res_length; - list_add_tail(&cb->list, &lkb->lkb_callbacks); + (*cb)->flags = flags; + (*cb)->mode = mode; + (*cb)->sb_status = status; + (*cb)->sb_flags = (sbflags & 0x000000FF); + (*cb)->copy_lvb = copy_lvb; + (*cb)->lkb_lksb = lkb->lkb_lksb; + kref_init(&(*cb)->ref); if (flags & DLM_CB_BAST) { lkb->lkb_last_bast_time = ktime_get(); - lkb->lkb_last_bast_mode = cb->mode; + lkb->lkb_last_bast_mode = mode; } else if (flags & DLM_CB_CAST) { - dlm_callback_set_last_ptr(&lkb->lkb_last_cast, cb); + dlm_callback_set_last_ptr(&lkb->lkb_last_cast, *cb); lkb->lkb_last_cast_time = ktime_get(); } - dlm_callback_set_last_ptr(&lkb->lkb_last_cb, cb); + dlm_callback_set_last_ptr(&lkb->lkb_last_cb, *cb); + rv = DLM_ENQUEUE_CALLBACK_NEED_SCHED; - out: +out: return rv; } -int dlm_dequeue_lkb_callback(struct dlm_lkb *lkb, struct dlm_callback **cb) -{ - /* oldest undelivered cb is callbacks first entry */ - *cb = list_first_entry_or_null(&lkb->lkb_callbacks, - struct dlm_callback, list); - if (!*cb) - return DLM_DEQUEUE_CALLBACK_EMPTY; - - /* remove it from callbacks so shift others down */ - list_del(&(*cb)->list); - if (list_empty(&lkb->lkb_callbacks)) - return DLM_DEQUEUE_CALLBACK_LAST; - - return DLM_DEQUEUE_CALLBACK_SUCCESS; -} - void dlm_add_cb(struct dlm_lkb *lkb, uint32_t flags, int mode, int status, - uint32_t sbflags) + uint32_t sbflags) { struct dlm_ls *ls = lkb->lkb_resource->res_ls; + struct dlm_callback *cb; int rv; if (test_bit(DLM_DFL_USER_BIT, &lkb->lkb_dflags)) { @@ -146,18 +155,20 @@ void dlm_add_cb(struct dlm_lkb *lkb, uint32_t flags, int mode, int status, return; } - spin_lock(&lkb->lkb_cb_lock); - rv = dlm_enqueue_lkb_callback(lkb, flags, mode, status, sbflags); + rv = dlm_queue_lkb_callback(lkb, flags, mode, status, sbflags, + &cb); switch (rv) { case DLM_ENQUEUE_CALLBACK_NEED_SCHED: - kref_get(&lkb->lkb_ref); + cb->astfn = lkb->lkb_astfn; + cb->bastfn = lkb->lkb_bastfn; + cb->astparam = lkb->lkb_astparam; + INIT_WORK(&cb->work, dlm_callback_work); spin_lock(&ls->ls_cb_lock); - if (test_bit(LSFL_CB_DELAY, &ls->ls_flags)) { - list_add(&lkb->lkb_cb_list, &ls->ls_cb_delay); - } else { - queue_work(ls->ls_callback_wq, &lkb->lkb_cb_work); - } + if (test_bit(LSFL_CB_DELAY, &ls->ls_flags)) + list_add(&cb->list, &ls->ls_cb_delay); + else + queue_work(ls->ls_callback_wq, &cb->work); spin_unlock(&ls->ls_cb_lock); break; case DLM_ENQUEUE_CALLBACK_SUCCESS: @@ -168,67 +179,12 @@ void dlm_add_cb(struct dlm_lkb *lkb, uint32_t flags, int mode, int status, WARN_ON_ONCE(1); break; } - spin_unlock(&lkb->lkb_cb_lock); -} - -void dlm_callback_work(struct work_struct *work) -{ - struct dlm_lkb *lkb = container_of(work, struct dlm_lkb, lkb_cb_work); - struct dlm_ls *ls = lkb->lkb_resource->res_ls; - struct dlm_rsb *rsb = lkb->lkb_resource; - void (*castfn) (void *astparam); - void (*bastfn) (void *astparam, int mode); - struct dlm_callback *cb; - int rv; - - spin_lock(&lkb->lkb_cb_lock); - rv = dlm_dequeue_lkb_callback(lkb, &cb); - if (WARN_ON_ONCE(rv == DLM_DEQUEUE_CALLBACK_EMPTY)) { - clear_bit(DLM_IFL_CB_PENDING_BIT, &lkb->lkb_iflags); - spin_unlock(&lkb->lkb_cb_lock); - goto out; - } - spin_unlock(&lkb->lkb_cb_lock); - - for (;;) { - castfn = lkb->lkb_astfn; - bastfn = lkb->lkb_bastfn; - - if (cb->flags & DLM_CB_BAST) { - trace_dlm_bast(ls->ls_global_id, lkb->lkb_id, - cb->mode, rsb->res_name, - rsb->res_length); - bastfn(lkb->lkb_astparam, cb->mode); - } else if (cb->flags & DLM_CB_CAST) { - lkb->lkb_lksb->sb_status = cb->sb_status; - lkb->lkb_lksb->sb_flags = cb->sb_flags; - trace_dlm_ast(ls->ls_global_id, lkb->lkb_id, - cb->sb_flags, cb->sb_status, - rsb->res_name, rsb->res_length); - castfn(lkb->lkb_astparam); - } - - kref_put(&cb->ref, dlm_release_callback); - - spin_lock(&lkb->lkb_cb_lock); - rv = dlm_dequeue_lkb_callback(lkb, &cb); - if (rv == DLM_DEQUEUE_CALLBACK_EMPTY) { - clear_bit(DLM_IFL_CB_PENDING_BIT, &lkb->lkb_iflags); - spin_unlock(&lkb->lkb_cb_lock); - break; - } - spin_unlock(&lkb->lkb_cb_lock); - } - -out: - /* undo kref_get from dlm_add_callback, may cause lkb to be freed */ - dlm_put_lkb(lkb); } int dlm_callback_start(struct dlm_ls *ls) { - ls->ls_callback_wq = alloc_workqueue("dlm_callback", - WQ_HIGHPRI | WQ_MEM_RECLAIM, 0); + ls->ls_callback_wq = alloc_ordered_workqueue("dlm_callback", + WQ_HIGHPRI | WQ_MEM_RECLAIM); if (!ls->ls_callback_wq) { log_print("can't start dlm_callback workqueue"); return -ENOMEM; @@ -257,7 +213,7 @@ void dlm_callback_suspend(struct dlm_ls *ls) void dlm_callback_resume(struct dlm_ls *ls) { - struct dlm_lkb *lkb, *safe; + struct dlm_callback *cb, *safe; int count = 0, sum = 0; bool empty; @@ -266,9 +222,9 @@ void dlm_callback_resume(struct dlm_ls *ls) more: spin_lock(&ls->ls_cb_lock); - list_for_each_entry_safe(lkb, safe, &ls->ls_cb_delay, lkb_cb_list) { - list_del_init(&lkb->lkb_cb_list); - queue_work(ls->ls_callback_wq, &lkb->lkb_cb_work); + list_for_each_entry_safe(cb, safe, &ls->ls_cb_delay, list) { + list_del(&cb->list); + queue_work(ls->ls_callback_wq, &cb->work); count++; if (count == MAX_CB_QUEUE) break; diff --git a/fs/dlm/ast.h b/fs/dlm/ast.h index ce007892dc2d..9bd12409e1ee 100644 --- a/fs/dlm/ast.h +++ b/fs/dlm/ast.h @@ -14,19 +14,15 @@ #define DLM_ENQUEUE_CALLBACK_NEED_SCHED 1 #define DLM_ENQUEUE_CALLBACK_SUCCESS 0 #define DLM_ENQUEUE_CALLBACK_FAILURE -1 -int dlm_enqueue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, - int status, uint32_t sbflags); -#define DLM_DEQUEUE_CALLBACK_EMPTY 2 -#define DLM_DEQUEUE_CALLBACK_LAST 1 -#define DLM_DEQUEUE_CALLBACK_SUCCESS 0 -int dlm_dequeue_lkb_callback(struct dlm_lkb *lkb, struct dlm_callback **cb); +int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, + int status, uint32_t sbflags, + struct dlm_callback **cb); void dlm_add_cb(struct dlm_lkb *lkb, uint32_t flags, int mode, int status, uint32_t sbflags); void dlm_callback_set_last_ptr(struct dlm_callback **from, struct dlm_callback *to); void dlm_release_callback(struct kref *ref); -void dlm_callback_work(struct work_struct *work); int dlm_callback_start(struct dlm_ls *ls); void dlm_callback_stop(struct dlm_ls *ls); void dlm_callback_suspend(struct dlm_ls *ls); diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index a9137c90f348..cda1526fd15b 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -16,6 +16,7 @@ * This is the main header file to be included in each DLM source file. */ +#include #include #include #include @@ -204,8 +205,7 @@ struct dlm_args { #define DLM_IFL_OVERLAP_CANCEL_BIT 20 #define DLM_IFL_ENDOFLIFE_BIT 21 #define DLM_IFL_DEADLOCK_CANCEL_BIT 24 -#define DLM_IFL_CB_PENDING_BIT 25 -#define __DLM_IFL_MAX_BIT DLM_IFL_CB_PENDING_BIT +#define __DLM_IFL_MAX_BIT DLM_IFL_DEADLOCK_CANCEL_BIT /* lkb_dflags */ @@ -217,12 +217,45 @@ struct dlm_args { #define DLM_CB_CAST 0x00000001 #define DLM_CB_BAST 0x00000002 +/* much of this is just saving user space pointers associated with the + * lock that we pass back to the user lib with an ast + */ + +struct dlm_user_args { + struct dlm_user_proc *proc; /* each process that opens the lockspace + * device has private data + * (dlm_user_proc) on the struct file, + * the process's locks point back to it + */ + struct dlm_lksb lksb; + struct dlm_lksb __user *user_lksb; + void __user *castparam; + void __user *castaddr; + void __user *bastparam; + void __user *bastaddr; + uint64_t xid; +}; + struct dlm_callback { uint32_t flags; /* DLM_CBF_ */ int sb_status; /* copy to lksb status */ uint8_t sb_flags; /* copy to lksb flags */ int8_t mode; /* rq mode of bast, gr mode of cast */ - int copy_lvb; + bool copy_lvb; + struct dlm_lksb *lkb_lksb; + unsigned char lvbptr[DLM_USER_LVB_LEN]; + + union { + void *astparam; /* caller's ast arg */ + struct dlm_user_args ua; + }; + struct work_struct work; + void (*bastfn)(void *astparam, int mode); + void (*astfn)(void *astparam); + char res_name[DLM_RESNAME_MAXLEN]; + size_t res_length; + uint32_t ls_id; + uint32_t lkb_id; struct list_head list; struct kref ref; @@ -256,10 +289,6 @@ struct dlm_lkb { struct list_head lkb_ownqueue; /* list of locks for a process */ ktime_t lkb_timestamp; - spinlock_t lkb_cb_lock; - struct work_struct lkb_cb_work; - struct list_head lkb_cb_list; /* for ls_cb_delay or proc->asts */ - struct list_head lkb_callbacks; struct dlm_callback *lkb_last_cast; struct dlm_callback *lkb_last_cb; int lkb_last_bast_mode; @@ -688,23 +717,6 @@ struct dlm_ls { #define LSFL_CB_DELAY 9 #define LSFL_NODIR 10 -/* much of this is just saving user space pointers associated with the - lock that we pass back to the user lib with an ast */ - -struct dlm_user_args { - struct dlm_user_proc *proc; /* each process that opens the lockspace - device has private data - (dlm_user_proc) on the struct file, - the process's locks point back to it*/ - struct dlm_lksb lksb; - struct dlm_lksb __user *user_lksb; - void __user *castparam; - void __user *castaddr; - void __user *bastparam; - void __user *bastaddr; - uint64_t xid; -}; - #define DLM_PROC_FLAGS_CLOSING 1 #define DLM_PROC_FLAGS_COMPAT 2 diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index fd752dd03896..198672446dcd 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -1203,10 +1203,6 @@ static int _create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret, kref_init(&lkb->lkb_ref); INIT_LIST_HEAD(&lkb->lkb_ownqueue); INIT_LIST_HEAD(&lkb->lkb_rsb_lookup); - INIT_LIST_HEAD(&lkb->lkb_cb_list); - INIT_LIST_HEAD(&lkb->lkb_callbacks); - spin_lock_init(&lkb->lkb_cb_lock); - INIT_WORK(&lkb->lkb_cb_work, dlm_callback_work); idr_preload(GFP_NOFS); spin_lock(&ls->ls_lkbidr_spin); @@ -6003,6 +5999,7 @@ static struct dlm_lkb *del_proc_lock(struct dlm_ls *ls, void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) { + struct dlm_callback *cb, *cb_safe; struct dlm_lkb *lkb, *safe; dlm_lock_recovery(ls); @@ -6032,10 +6029,9 @@ void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) dlm_put_lkb(lkb); } - list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_cb_list) { - dlm_purge_lkb_callbacks(lkb); - list_del_init(&lkb->lkb_cb_list); - dlm_put_lkb(lkb); + list_for_each_entry_safe(cb, cb_safe, &proc->asts, list) { + list_del(&cb->list); + kref_put(&cb->ref, dlm_release_callback); } spin_unlock(&ls->ls_clear_proc_locks); @@ -6044,6 +6040,7 @@ void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) { + struct dlm_callback *cb, *cb_safe; struct dlm_lkb *lkb, *safe; while (1) { @@ -6073,10 +6070,9 @@ static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) spin_unlock(&proc->locks_spin); spin_lock(&proc->asts_spin); - list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_cb_list) { - dlm_purge_lkb_callbacks(lkb); - list_del_init(&lkb->lkb_cb_list); - dlm_put_lkb(lkb); + list_for_each_entry_safe(cb, cb_safe, &proc->asts, list) { + list_del(&cb->list); + kref_put(&cb->ref, dlm_release_callback); } spin_unlock(&proc->asts_spin); } diff --git a/fs/dlm/user.c b/fs/dlm/user.c index fa99b6074e5c..334a6d64d413 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -21,6 +21,7 @@ #include "dlm_internal.h" #include "lockspace.h" #include "lock.h" +#include "lvb_table.h" #include "user.h" #include "ast.h" #include "config.h" @@ -144,24 +145,6 @@ static void compat_output(struct dlm_lock_result *res, } #endif -/* should held proc->asts_spin lock */ -void dlm_purge_lkb_callbacks(struct dlm_lkb *lkb) -{ - struct dlm_callback *cb, *safe; - - list_for_each_entry_safe(cb, safe, &lkb->lkb_callbacks, list) { - list_del(&cb->list); - kref_put(&cb->ref, dlm_release_callback); - } - - clear_bit(DLM_IFL_CB_PENDING_BIT, &lkb->lkb_iflags); - - /* invalidate */ - dlm_callback_set_last_ptr(&lkb->lkb_last_cast, NULL); - dlm_callback_set_last_ptr(&lkb->lkb_last_cb, NULL); - lkb->lkb_last_bast_mode = -1; -} - /* Figure out if this lock is at the end of its life and no longer available for the application to use. The lkb still exists until the final ast is read. A lock becomes EOL in three situations: @@ -198,6 +181,7 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode, struct dlm_ls *ls; struct dlm_user_args *ua; struct dlm_user_proc *proc; + struct dlm_callback *cb; int rv; if (test_bit(DLM_DFL_ORPHAN_BIT, &lkb->lkb_dflags) || @@ -229,11 +213,18 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode, spin_lock(&proc->asts_spin); - rv = dlm_enqueue_lkb_callback(lkb, flags, mode, status, sbflags); + rv = dlm_queue_lkb_callback(lkb, flags, mode, status, sbflags, &cb); switch (rv) { case DLM_ENQUEUE_CALLBACK_NEED_SCHED: - kref_get(&lkb->lkb_ref); - list_add_tail(&lkb->lkb_cb_list, &proc->asts); + cb->ua = *ua; + cb->lkb_lksb = &cb->ua.lksb; + if (cb->copy_lvb) { + memcpy(cb->lvbptr, ua->lksb.sb_lvbptr, + DLM_USER_LVB_LEN); + cb->lkb_lksb->sb_lvbptr = cb->lvbptr; + } + + list_add_tail(&cb->list, &proc->asts); wake_up_interruptible(&proc->wait); break; case DLM_ENQUEUE_CALLBACK_SUCCESS: @@ -801,10 +792,8 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct dlm_user_proc *proc = file->private_data; - struct dlm_lkb *lkb; DECLARE_WAITQUEUE(wait, current); struct dlm_callback *cb; - struct dlm_rsb *rsb; int rv, ret; if (count == sizeof(struct dlm_device_version)) { @@ -824,8 +813,6 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, #endif return -EINVAL; - try_another: - /* do we really need this? can a read happen after a close? */ if (test_bit(DLM_PROC_FLAGS_CLOSING, &proc->flags)) return -EINVAL; @@ -860,55 +847,24 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, without removing lkb_cb_list; so empty lkb_cb_list is always consistent with empty lkb_callbacks */ - lkb = list_first_entry(&proc->asts, struct dlm_lkb, lkb_cb_list); - - rv = dlm_dequeue_lkb_callback(lkb, &cb); - switch (rv) { - case DLM_DEQUEUE_CALLBACK_EMPTY: - /* this shouldn't happen; lkb should have been removed from - * list when last item was dequeued - */ - log_print("dlm_rem_lkb_callback empty %x", lkb->lkb_id); - list_del_init(&lkb->lkb_cb_list); - spin_unlock(&proc->asts_spin); - /* removes ref for proc->asts, may cause lkb to be freed */ - dlm_put_lkb(lkb); - WARN_ON_ONCE(1); - goto try_another; - case DLM_DEQUEUE_CALLBACK_LAST: - list_del_init(&lkb->lkb_cb_list); - clear_bit(DLM_IFL_CB_PENDING_BIT, &lkb->lkb_iflags); - break; - case DLM_DEQUEUE_CALLBACK_SUCCESS: - break; - default: - WARN_ON_ONCE(1); - break; - } + cb = list_first_entry(&proc->asts, struct dlm_callback, list); + list_del(&cb->list); spin_unlock(&proc->asts_spin); - rsb = lkb->lkb_resource; if (cb->flags & DLM_CB_BAST) { - trace_dlm_bast(rsb->res_ls->ls_global_id, lkb->lkb_id, - cb->mode, rsb->res_name, rsb->res_length); + trace_dlm_bast(cb->ls_id, cb->lkb_id, cb->mode, cb->res_name, + cb->res_length); } else if (cb->flags & DLM_CB_CAST) { - lkb->lkb_lksb->sb_status = cb->sb_status; - lkb->lkb_lksb->sb_flags = cb->sb_flags; - trace_dlm_ast(rsb->res_ls->ls_global_id, lkb->lkb_id, - cb->sb_flags, cb->sb_status, rsb->res_name, - rsb->res_length); + cb->lkb_lksb->sb_status = cb->sb_status; + cb->lkb_lksb->sb_flags = cb->sb_flags; + trace_dlm_ast(cb->ls_id, cb->lkb_id, cb->sb_status, + cb->sb_flags, cb->res_name, cb->res_length); } - ret = copy_result_to_user(lkb->lkb_ua, + ret = copy_result_to_user(&cb->ua, test_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags), cb->flags, cb->mode, cb->copy_lvb, buf, count); - kref_put(&cb->ref, dlm_release_callback); - - /* removes ref for proc->asts, may cause lkb to be freed */ - if (rv == DLM_DEQUEUE_CALLBACK_LAST) - dlm_put_lkb(lkb); - return ret; } -- cgit v1.2.3-59-g8ed1b From 2bec1bbd55cf96b313566d2e1b9ff54451685e18 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Thu, 28 Mar 2024 11:48:42 -0400 Subject: dlm: remove callback reference counting Get rid of the unnecessary refcounting on callback structs. Copy interesting callback info into the lkb struct rather than maintaining pointers to callback structs from the lkb. This goes back to the way things were done prior to commit 61bed0baa4db ("fs: dlm: use a non-static queue for callbacks"). Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/ast.c | 56 ++++++++++++++++----------------------------------- fs/dlm/ast.h | 3 --- fs/dlm/debug_fs.c | 2 +- fs/dlm/dlm_internal.h | 8 ++++---- fs/dlm/lock.c | 8 +++++--- fs/dlm/memory.c | 4 ---- fs/dlm/user.c | 2 +- 7 files changed, 28 insertions(+), 55 deletions(-) (limited to 'fs') diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index e9da812352b4..03879c94fadb 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -18,25 +18,6 @@ #include "user.h" #include "ast.h" -void dlm_release_callback(struct kref *ref) -{ - struct dlm_callback *cb = container_of(ref, struct dlm_callback, ref); - - dlm_free_cb(cb); -} - -void dlm_callback_set_last_ptr(struct dlm_callback **from, - struct dlm_callback *to) -{ - if (*from) - kref_put(&(*from)->ref, dlm_release_callback); - - if (to) - kref_get(&to->ref); - - *from = to; -} - static void dlm_callback_work(struct work_struct *work) { struct dlm_callback *cb = container_of(work, struct dlm_callback, work); @@ -53,7 +34,7 @@ static void dlm_callback_work(struct work_struct *work) cb->astfn(cb->astparam); } - kref_put(&cb->ref, dlm_release_callback); + dlm_free_cb(cb); } int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, @@ -70,11 +51,11 @@ int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, /* if cb is a bast, it should be skipped if the blocking mode is * compatible with the last granted mode */ - if (lkb->lkb_last_cast) { - if (dlm_modes_compat(mode, lkb->lkb_last_cast->mode)) { + if (lkb->lkb_last_cast_cb_mode != -1) { + if (dlm_modes_compat(mode, lkb->lkb_last_cast_cb_mode)) { log_debug(ls, "skip %x bast mode %d for cast mode %d", lkb->lkb_id, mode, - lkb->lkb_last_cast->mode); + lkb->lkb_last_cast_cb_mode); goto out; } } @@ -85,8 +66,9 @@ int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, * is a bast for the same mode or a more restrictive mode. * (the addional > PR check is needed for PR/CW inversion) */ - if (lkb->lkb_last_cb && lkb->lkb_last_cb->flags & DLM_CB_BAST) { - prev_mode = lkb->lkb_last_cb->mode; + if (lkb->lkb_last_cb_mode != -1 && + lkb->lkb_last_cb_flags & DLM_CB_BAST) { + prev_mode = lkb->lkb_last_cb_mode; if ((prev_mode == mode) || (prev_mode > mode && prev_mode > DLM_LOCK_PR)) { @@ -95,19 +77,25 @@ int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, goto out; } } + + lkb->lkb_last_bast_time = ktime_get(); + lkb->lkb_last_bast_cb_mode = mode; } else if (flags & DLM_CB_CAST) { if (test_bit(DLM_DFL_USER_BIT, &lkb->lkb_dflags)) { - if (lkb->lkb_last_cast) - prev_mode = lkb->lkb_last_cb->mode; - else - prev_mode = -1; + prev_mode = lkb->lkb_last_cast_cb_mode; if (!status && lkb->lkb_lksb->sb_lvbptr && dlm_lvb_operations[prev_mode + 1][mode + 1]) copy_lvb = 1; } + + lkb->lkb_last_cast_cb_mode = mode; + lkb->lkb_last_cast_time = ktime_get(); } + lkb->lkb_last_cb_mode = mode; + lkb->lkb_last_cb_flags = flags; + *cb = dlm_allocate_cb(); if (!*cb) { rv = DLM_ENQUEUE_CALLBACK_FAILURE; @@ -126,17 +114,7 @@ int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, (*cb)->sb_flags = (sbflags & 0x000000FF); (*cb)->copy_lvb = copy_lvb; (*cb)->lkb_lksb = lkb->lkb_lksb; - kref_init(&(*cb)->ref); - - if (flags & DLM_CB_BAST) { - lkb->lkb_last_bast_time = ktime_get(); - lkb->lkb_last_bast_mode = mode; - } else if (flags & DLM_CB_CAST) { - dlm_callback_set_last_ptr(&lkb->lkb_last_cast, *cb); - lkb->lkb_last_cast_time = ktime_get(); - } - dlm_callback_set_last_ptr(&lkb->lkb_last_cb, *cb); rv = DLM_ENQUEUE_CALLBACK_NEED_SCHED; out: diff --git a/fs/dlm/ast.h b/fs/dlm/ast.h index 9bd12409e1ee..9093ff043bee 100644 --- a/fs/dlm/ast.h +++ b/fs/dlm/ast.h @@ -19,10 +19,7 @@ int dlm_queue_lkb_callback(struct dlm_lkb *lkb, uint32_t flags, int mode, struct dlm_callback **cb); void dlm_add_cb(struct dlm_lkb *lkb, uint32_t flags, int mode, int status, uint32_t sbflags); -void dlm_callback_set_last_ptr(struct dlm_callback **from, - struct dlm_callback *to); -void dlm_release_callback(struct kref *ref); int dlm_callback_start(struct dlm_ls *ls); void dlm_callback_stop(struct dlm_ls *ls); void dlm_callback_suspend(struct dlm_ls *ls); diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 289d959c7700..19cdedd56629 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -247,7 +247,7 @@ static void print_format3_lock(struct seq_file *s, struct dlm_lkb *lkb, lkb->lkb_status, lkb->lkb_grmode, lkb->lkb_rqmode, - lkb->lkb_last_bast_mode, + lkb->lkb_last_bast_cb_mode, rsb_lookup, lkb->lkb_wait_type, lkb->lkb_lvbseq, diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index cda1526fd15b..1d2ee5c2d23d 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -258,7 +258,6 @@ struct dlm_callback { uint32_t lkb_id; struct list_head list; - struct kref ref; }; struct dlm_lkb { @@ -289,9 +288,10 @@ struct dlm_lkb { struct list_head lkb_ownqueue; /* list of locks for a process */ ktime_t lkb_timestamp; - struct dlm_callback *lkb_last_cast; - struct dlm_callback *lkb_last_cb; - int lkb_last_bast_mode; + int8_t lkb_last_cast_cb_mode; + int8_t lkb_last_bast_cb_mode; + int8_t lkb_last_cb_mode; + uint8_t lkb_last_cb_flags; ktime_t lkb_last_cast_time; /* for debugging */ ktime_t lkb_last_bast_time; /* for debugging */ diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 198672446dcd..c8426f6f518c 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -1197,7 +1197,9 @@ static int _create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret, if (!lkb) return -ENOMEM; - lkb->lkb_last_bast_mode = -1; + lkb->lkb_last_bast_cb_mode = DLM_LOCK_IV; + lkb->lkb_last_cast_cb_mode = DLM_LOCK_IV; + lkb->lkb_last_cb_mode = DLM_LOCK_IV; lkb->lkb_nodeid = -1; lkb->lkb_grmode = DLM_LOCK_IV; kref_init(&lkb->lkb_ref); @@ -6031,7 +6033,7 @@ void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) list_for_each_entry_safe(cb, cb_safe, &proc->asts, list) { list_del(&cb->list); - kref_put(&cb->ref, dlm_release_callback); + dlm_free_cb(cb); } spin_unlock(&ls->ls_clear_proc_locks); @@ -6072,7 +6074,7 @@ static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) spin_lock(&proc->asts_spin); list_for_each_entry_safe(cb, cb_safe, &proc->asts, list) { list_del(&cb->list); - kref_put(&cb->ref, dlm_release_callback); + dlm_free_cb(cb); } spin_unlock(&proc->asts_spin); } diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index 64f212a066cf..be9398ddf357 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -127,10 +127,6 @@ void dlm_free_lkb(struct dlm_lkb *lkb) } } - /* drop references if they are set */ - dlm_callback_set_last_ptr(&lkb->lkb_last_cast, NULL); - dlm_callback_set_last_ptr(&lkb->lkb_last_cb, NULL); - kmem_cache_free(lkb_cache, lkb); } diff --git a/fs/dlm/user.c b/fs/dlm/user.c index 334a6d64d413..b4971ba4bdd6 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -864,7 +864,7 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, ret = copy_result_to_user(&cb->ua, test_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags), cb->flags, cb->mode, cb->copy_lvb, buf, count); - kref_put(&cb->ref, dlm_release_callback); + dlm_free_cb(cb); return ret; } -- cgit v1.2.3-59-g8ed1b From 24a9799aa8efecd0eb55a75e35f9d8e6400063aa Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Mon, 1 Apr 2024 14:13:10 -0300 Subject: smb: client: fix UAF in smb2_reconnect_server() The UAF bug is due to smb2_reconnect_server() accessing a session that is already being teared down by another thread that is executing __cifs_put_smb_ses(). This can happen when (a) the client has connection to the server but no session or (b) another thread ends up setting @ses->ses_status again to something different than SES_EXITING. To fix this, we need to make sure to unconditionally set @ses->ses_status to SES_EXITING and prevent any other threads from setting a new status while we're still tearing it down. The following can be reproduced by adding some delay to right after the ipc is freed in __cifs_put_smb_ses() - which will give smb2_reconnect_server() worker a chance to run and then accessing @ses->ipc: kinit ... mount.cifs //srv/share /mnt/1 -o sec=krb5,nohandlecache,echo_interval=10 [disconnect srv] ls /mnt/1 &>/dev/null sleep 30 kdestroy [reconnect srv] sleep 10 umount /mnt/1 ... CIFS: VFS: Verify user has a krb5 ticket and keyutils is installed CIFS: VFS: \\srv Send error in SessSetup = -126 CIFS: VFS: Verify user has a krb5 ticket and keyutils is installed CIFS: VFS: \\srv Send error in SessSetup = -126 general protection fault, probably for non-canonical address 0x6b6b6b6b6b6b6b6b: 0000 [#1] PREEMPT SMP NOPTI CPU: 3 PID: 50 Comm: kworker/3:1 Not tainted 6.9.0-rc2 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-1.fc39 04/01/2014 Workqueue: cifsiod smb2_reconnect_server [cifs] RIP: 0010:__list_del_entry_valid_or_report+0x33/0xf0 Code: 4f 08 48 85 d2 74 42 48 85 c9 74 59 48 b8 00 01 00 00 00 00 ad de 48 39 c2 74 61 48 b8 22 01 00 00 00 00 74 69 <48> 8b 01 48 39 f8 75 7b 48 8b 72 08 48 39 c6 0f 85 88 00 00 00 b8 RSP: 0018:ffffc900001bfd70 EFLAGS: 00010a83 RAX: dead000000000122 RBX: ffff88810da53838 RCX: 6b6b6b6b6b6b6b6b RDX: 6b6b6b6b6b6b6b6b RSI: ffffffffc02f6878 RDI: ffff88810da53800 RBP: ffff88810da53800 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000001 R12: ffff88810c064000 R13: 0000000000000001 R14: ffff88810c064000 R15: ffff8881039cc000 FS: 0000000000000000(0000) GS:ffff888157c00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fe3728b1000 CR3: 000000010caa4000 CR4: 0000000000750ef0 PKRU: 55555554 Call Trace: ? die_addr+0x36/0x90 ? exc_general_protection+0x1c1/0x3f0 ? asm_exc_general_protection+0x26/0x30 ? __list_del_entry_valid_or_report+0x33/0xf0 __cifs_put_smb_ses+0x1ae/0x500 [cifs] smb2_reconnect_server+0x4ed/0x710 [cifs] process_one_work+0x205/0x6b0 worker_thread+0x191/0x360 ? __pfx_worker_thread+0x10/0x10 kthread+0xe2/0x110 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x34/0x50 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30 Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/connect.c | 83 ++++++++++++++++++++----------------------------- 1 file changed, 34 insertions(+), 49 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 9b85b5341822..ee29bc57300c 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -232,7 +232,13 @@ cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, spin_lock(&cifs_tcp_ses_lock); list_for_each_entry_safe(ses, nses, &pserver->smb_ses_list, smb_ses_list) { - /* check if iface is still active */ + spin_lock(&ses->ses_lock); + if (ses->ses_status == SES_EXITING) { + spin_unlock(&ses->ses_lock); + continue; + } + spin_unlock(&ses->ses_lock); + spin_lock(&ses->chan_lock); if (cifs_ses_get_chan_index(ses, server) == CIFS_INVAL_CHAN_INDEX) { @@ -1963,31 +1969,6 @@ out: return rc; } -/** - * cifs_free_ipc - helper to release the session IPC tcon - * @ses: smb session to unmount the IPC from - * - * Needs to be called everytime a session is destroyed. - * - * On session close, the IPC is closed and the server must release all tcons of the session. - * No need to send a tree disconnect here. - * - * Besides, it will make the server to not close durable and resilient files on session close, as - * specified in MS-SMB2 3.3.5.6 Receiving an SMB2 LOGOFF Request. - */ -static int -cifs_free_ipc(struct cifs_ses *ses) -{ - struct cifs_tcon *tcon = ses->tcon_ipc; - - if (tcon == NULL) - return 0; - - tconInfoFree(tcon); - ses->tcon_ipc = NULL; - return 0; -} - static struct cifs_ses * cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) { @@ -2019,48 +2000,52 @@ cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) void __cifs_put_smb_ses(struct cifs_ses *ses) { struct TCP_Server_Info *server = ses->server; + struct cifs_tcon *tcon; unsigned int xid; size_t i; + bool do_logoff; int rc; + spin_lock(&cifs_tcp_ses_lock); spin_lock(&ses->ses_lock); - if (ses->ses_status == SES_EXITING) { + cifs_dbg(FYI, "%s: id=0x%llx ses_count=%d ses_status=%u ipc=%s\n", + __func__, ses->Suid, ses->ses_count, ses->ses_status, + ses->tcon_ipc ? ses->tcon_ipc->tree_name : "none"); + if (ses->ses_status == SES_EXITING || --ses->ses_count > 0) { spin_unlock(&ses->ses_lock); + spin_unlock(&cifs_tcp_ses_lock); return; } - spin_unlock(&ses->ses_lock); + /* ses_count can never go negative */ + WARN_ON(ses->ses_count < 0); - cifs_dbg(FYI, "%s: ses_count=%d\n", __func__, ses->ses_count); - cifs_dbg(FYI, - "%s: ses ipc: %s\n", __func__, ses->tcon_ipc ? ses->tcon_ipc->tree_name : "NONE"); + spin_lock(&ses->chan_lock); + cifs_chan_clear_need_reconnect(ses, server); + spin_unlock(&ses->chan_lock); - spin_lock(&cifs_tcp_ses_lock); - if (--ses->ses_count > 0) { - spin_unlock(&cifs_tcp_ses_lock); - return; - } - spin_lock(&ses->ses_lock); - if (ses->ses_status == SES_GOOD) - ses->ses_status = SES_EXITING; + do_logoff = ses->ses_status == SES_GOOD && server->ops->logoff; + ses->ses_status = SES_EXITING; + tcon = ses->tcon_ipc; + ses->tcon_ipc = NULL; spin_unlock(&ses->ses_lock); spin_unlock(&cifs_tcp_ses_lock); - /* ses_count can never go negative */ - WARN_ON(ses->ses_count < 0); - - spin_lock(&ses->ses_lock); - if (ses->ses_status == SES_EXITING && server->ops->logoff) { - spin_unlock(&ses->ses_lock); - cifs_free_ipc(ses); + /* + * On session close, the IPC is closed and the server must release all + * tcons of the session. No need to send a tree disconnect here. + * + * Besides, it will make the server to not close durable and resilient + * files on session close, as specified in MS-SMB2 3.3.5.6 Receiving an + * SMB2 LOGOFF Request. + */ + tconInfoFree(tcon); + if (do_logoff) { xid = get_xid(); rc = server->ops->logoff(xid, ses); if (rc) cifs_server_dbg(VFS, "%s: Session Logoff failure rc=%d\n", __func__, rc); _free_xid(xid); - } else { - spin_unlock(&ses->ses_lock); - cifs_free_ipc(ses); } spin_lock(&cifs_tcp_ses_lock); -- cgit v1.2.3-59-g8ed1b From ba947ecd39ea0e6a6f6f1101f99611fc30943bcb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 1 Apr 2024 19:16:19 -0400 Subject: bcachefs: Fix btree node reserve Sign error when checking the watermark - oops. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 983bb27298cc..29aee215384a 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -303,7 +303,7 @@ static struct btree *__bch2_btree_node_alloc(struct btree_trans *trans, struct open_buckets obs = { .nr = 0 }; struct bch_devs_list devs_have = (struct bch_devs_list) { 0 }; enum bch_watermark watermark = flags & BCH_WATERMARK_MASK; - unsigned nr_reserve = watermark > BCH_WATERMARK_reclaim + unsigned nr_reserve = watermark < BCH_WATERMARK_reclaim ? BTREE_NODE_RESERVE : 0; int ret; -- cgit v1.2.3-59-g8ed1b From e2a316b3cc45a1198f3feb18707403bb7f0cbc15 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 1 Apr 2024 19:20:36 -0400 Subject: bcachefs: BCH_WATERMARK_interior_updates This adds a new watermark, higher priority than BCH_WATERMARK_reclaim, for interior btree updates. We've seen a deadlock where journal replay triggers a ton of btree node merges, and these use up all available open buckets and then interior updates get stuck. One cause of this is that we're currently lacking btree node merging on write buffer btrees - that needs to be fixed as well. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 4 +++- fs/bcachefs/alloc_types.h | 3 ++- fs/bcachefs/btree_io.c | 2 +- fs/bcachefs/btree_trans_commit.c | 3 ++- fs/bcachefs/btree_update_interior.c | 6 +++--- fs/bcachefs/buckets.h | 1 + 6 files changed, 12 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 214b15c84d1f..a1fc30adf912 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -188,8 +188,10 @@ long bch2_bucket_alloc_new_fs(struct bch_dev *ca) static inline unsigned open_buckets_reserved(enum bch_watermark watermark) { switch (watermark) { - case BCH_WATERMARK_reclaim: + case BCH_WATERMARK_interior_updates: return 0; + case BCH_WATERMARK_reclaim: + return OPEN_BUCKETS_COUNT / 6; case BCH_WATERMARK_btree: case BCH_WATERMARK_btree_copygc: return OPEN_BUCKETS_COUNT / 4; diff --git a/fs/bcachefs/alloc_types.h b/fs/bcachefs/alloc_types.h index b91b7a461056..c2226e947c41 100644 --- a/fs/bcachefs/alloc_types.h +++ b/fs/bcachefs/alloc_types.h @@ -22,7 +22,8 @@ struct bucket_alloc_state { x(copygc) \ x(btree) \ x(btree_copygc) \ - x(reclaim) + x(reclaim) \ + x(interior_updates) enum bch_watermark { #define x(name) BCH_WATERMARK_##name, diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 9c71e6fb9c41..f3f27bb85a5b 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1861,7 +1861,7 @@ static void btree_node_write_work(struct work_struct *work) } else { ret = bch2_trans_do(c, NULL, NULL, 0, bch2_btree_node_update_key_get_iter(trans, b, &wbio->key, - BCH_WATERMARK_reclaim| + BCH_WATERMARK_interior_updates| BCH_TRANS_COMMIT_journal_reclaim| BCH_TRANS_COMMIT_no_enospc| BCH_TRANS_COMMIT_no_check_rw, diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 96669fede7d3..aa9da4970740 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -887,6 +887,7 @@ int bch2_trans_commit_error(struct btree_trans *trans, unsigned flags, int ret, unsigned long trace_ip) { struct bch_fs *c = trans->c; + enum bch_watermark watermark = flags & BCH_WATERMARK_MASK; switch (ret) { case -BCH_ERR_btree_insert_btree_node_full: @@ -905,7 +906,7 @@ int bch2_trans_commit_error(struct btree_trans *trans, unsigned flags, * flag */ if ((flags & BCH_TRANS_COMMIT_journal_reclaim) && - (flags & BCH_WATERMARK_MASK) != BCH_WATERMARK_reclaim) { + watermark < BCH_WATERMARK_reclaim) { ret = -BCH_ERR_journal_reclaim_would_deadlock; break; } diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 29aee215384a..9fd2dd0f4682 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -687,7 +687,7 @@ static void btree_update_nodes_written(struct btree_update *as) * which may require allocations as well. */ ret = commit_do(trans, &as->disk_res, &journal_seq, - BCH_WATERMARK_reclaim| + BCH_WATERMARK_interior_updates| BCH_TRANS_COMMIT_no_enospc| BCH_TRANS_COMMIT_no_check_rw| BCH_TRANS_COMMIT_journal_reclaim, @@ -1121,7 +1121,7 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, unsigned journal_flags = watermark|JOURNAL_RES_GET_CHECK; if ((flags & BCH_TRANS_COMMIT_journal_reclaim) && - watermark != BCH_WATERMARK_reclaim) + watermark < BCH_WATERMARK_reclaim) journal_flags |= JOURNAL_RES_GET_NONBLOCK; ret = drop_locks_do(trans, @@ -1217,7 +1217,7 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, */ if (bch2_err_matches(ret, ENOSPC) && (flags & BCH_TRANS_COMMIT_journal_reclaim) && - watermark != BCH_WATERMARK_reclaim) { + watermark < BCH_WATERMARK_reclaim) { ret = -BCH_ERR_journal_reclaim_would_deadlock; goto err; } diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index 6387e039f789..00aaf4bb5139 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -226,6 +226,7 @@ static inline u64 bch2_dev_buckets_reserved(struct bch_dev *ca, enum bch_waterma fallthrough; case BCH_WATERMARK_btree_copygc: case BCH_WATERMARK_reclaim: + case BCH_WATERMARK_interior_updates: break; } -- cgit v1.2.3-59-g8ed1b From c42cd606e4f004e9ba36a05b9adb9e4eead5834a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 2 Apr 2024 01:03:58 -0400 Subject: bcachefs: fix nocow lock deadlock Signed-off-by: Kent Overstreet --- fs/bcachefs/data_update.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index b564404dad71..34731ee0217f 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -580,8 +580,7 @@ int bch2_data_update_init(struct btree_trans *trans, move_ctxt_wait_event(ctxt, (locked = bch2_bucket_nocow_trylock(&c->nocow_locks, PTR_BUCKET_POS(c, &p.ptr), 0)) || - (!atomic_read(&ctxt->read_sectors) && - !atomic_read(&ctxt->write_sectors))); + list_empty(&ctxt->ios)); if (!locked) bch2_bucket_nocow_lock(&c->nocow_locks, -- cgit v1.2.3-59-g8ed1b From 94755a00a4e79236d4bfa6dc671a339828fb38ce Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Mon, 1 Apr 2024 20:01:51 +0000 Subject: udf: replace deprecated strncpy/strcpy with strscpy strncpy() is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. Also replace an instance of strcpy() which is also deprecated. s_volume_ident is a NUL-terminated string which is evident from its usage in udf_debug: | udf_debug("volIdent[] = '%s'\n", UDF_SB(sb)->s_volume_ident); s_volume_ident should also be NUL-padded as it is copied out to userspace: | if (copy_to_user((char __user *)arg, | UDF_SB(inode->i_sb)->s_volume_ident, 32)) | return -EFAULT; Considering the above, a suitable replacement is `strscpy_pad` [2] due to the fact that it guarantees both NUL-termination and NUL-padding on the destination buffer. To simplify the code, let's use the new 2-argument version of strscpy_pad() introduced in Commit e6584c3964f2f ("string: Allow 2-argument strscpy()"). Also zero-allocate @outstr so we can safely use a non-@ret length argument. This is just in case udf_dstrCS0toChar() doesn't include the NUL-byte in its return length, we won't truncate @outstr or write garbage bytes either. Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2] Link: https://github.com/KSPP/linux/issues/90 Cc: linux-hardening@vger.kernel.org Signed-off-by: Justin Stitt Signed-off-by: Jan Kara Message-Id: <20240401-strncpy-fs-udf-super-c-v1-1-80cddab7a281@google.com> --- fs/udf/super.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/udf/super.c b/fs/udf/super.c index ba6b747a3830..9381a66c6ce5 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -895,7 +895,7 @@ static int udf_load_pvoldesc(struct super_block *sb, sector_t block) int ret; struct timestamp *ts; - outstr = kmalloc(128, GFP_KERNEL); + outstr = kzalloc(128, GFP_KERNEL); if (!outstr) return -ENOMEM; @@ -921,11 +921,11 @@ static int udf_load_pvoldesc(struct super_block *sb, sector_t block) ret = udf_dstrCS0toChar(sb, outstr, 31, pvoldesc->volIdent, 32); if (ret < 0) { - strcpy(UDF_SB(sb)->s_volume_ident, "InvalidName"); + strscpy_pad(UDF_SB(sb)->s_volume_ident, "InvalidName"); pr_warn("incorrect volume identification, setting to " "'InvalidName'\n"); } else { - strncpy(UDF_SB(sb)->s_volume_ident, outstr, ret); + strscpy_pad(UDF_SB(sb)->s_volume_ident, outstr); } udf_debug("volIdent[] = '%s'\n", UDF_SB(sb)->s_volume_ident); -- cgit v1.2.3-59-g8ed1b From c1832f67035dc04fb89e6b591b64e4d515843cda Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Sun, 31 Mar 2024 21:58:26 +0900 Subject: ksmbd: don't send oplock break if rename fails Don't send oplock break if rename fails. This patch fix smb2.oplock.batch20 test. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index d478fa0c57ab..5723bbf372d7 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -5857,8 +5857,9 @@ static int smb2_rename(struct ksmbd_work *work, if (!file_info->ReplaceIfExists) flags = RENAME_NOREPLACE; - smb_break_all_levII_oplock(work, fp, 0); rc = ksmbd_vfs_rename(work, &fp->filp->f_path, new_name, flags); + if (!rc) + smb_break_all_levII_oplock(work, fp, 0); out: kfree(new_name); return rc; -- cgit v1.2.3-59-g8ed1b From a677ebd8ca2f2632ccdecbad7b87641274e15aac Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Sun, 31 Mar 2024 21:59:10 +0900 Subject: ksmbd: validate payload size in ipc response If installing malicious ksmbd-tools, ksmbd.mountd can return invalid ipc response to ksmbd kernel server. ksmbd should validate payload size of ipc response from ksmbd.mountd to avoid memory overrun or slab-out-of-bounds. This patch validate 3 ipc response that has payload. Cc: stable@vger.kernel.org Reported-by: Chao Ma Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/ksmbd_netlink.h | 3 ++- fs/smb/server/mgmt/share_config.c | 7 ++++++- fs/smb/server/transport_ipc.c | 37 +++++++++++++++++++++++++++++++++++++ 3 files changed, 45 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/ksmbd_netlink.h b/fs/smb/server/ksmbd_netlink.h index 8ca8a45c4c62..686b321c5a8b 100644 --- a/fs/smb/server/ksmbd_netlink.h +++ b/fs/smb/server/ksmbd_netlink.h @@ -167,7 +167,8 @@ struct ksmbd_share_config_response { __u16 force_uid; __u16 force_gid; __s8 share_name[KSMBD_REQ_MAX_SHARE_NAME]; - __u32 reserved[112]; /* Reserved room */ + __u32 reserved[111]; /* Reserved room */ + __u32 payload_sz; __u32 veto_list_sz; __s8 ____payload[]; }; diff --git a/fs/smb/server/mgmt/share_config.c b/fs/smb/server/mgmt/share_config.c index 328a412259dc..a2f0a2edceb8 100644 --- a/fs/smb/server/mgmt/share_config.c +++ b/fs/smb/server/mgmt/share_config.c @@ -158,7 +158,12 @@ static struct ksmbd_share_config *share_config_request(struct unicode_map *um, share->name = kstrdup(name, GFP_KERNEL); if (!test_share_config_flag(share, KSMBD_SHARE_FLAG_PIPE)) { - share->path = kstrdup(ksmbd_share_config_path(resp), + int path_len = PATH_MAX; + + if (resp->payload_sz) + path_len = resp->payload_sz - resp->veto_list_sz; + + share->path = kstrndup(ksmbd_share_config_path(resp), path_len, GFP_KERNEL); if (share->path) share->path_sz = strlen(share->path); diff --git a/fs/smb/server/transport_ipc.c b/fs/smb/server/transport_ipc.c index f29bb03f0dc4..8752ac82c557 100644 --- a/fs/smb/server/transport_ipc.c +++ b/fs/smb/server/transport_ipc.c @@ -65,6 +65,7 @@ struct ipc_msg_table_entry { struct hlist_node ipc_table_hlist; void *response; + unsigned int msg_sz; }; static struct delayed_work ipc_timer_work; @@ -275,6 +276,7 @@ static int handle_response(int type, void *payload, size_t sz) } memcpy(entry->response, payload, sz); + entry->msg_sz = sz; wake_up_interruptible(&entry->wait); ret = 0; break; @@ -453,6 +455,34 @@ out: return ret; } +static int ipc_validate_msg(struct ipc_msg_table_entry *entry) +{ + unsigned int msg_sz = entry->msg_sz; + + if (entry->type == KSMBD_EVENT_RPC_REQUEST) { + struct ksmbd_rpc_command *resp = entry->response; + + msg_sz = sizeof(struct ksmbd_rpc_command) + resp->payload_sz; + } else if (entry->type == KSMBD_EVENT_SPNEGO_AUTHEN_REQUEST) { + struct ksmbd_spnego_authen_response *resp = entry->response; + + msg_sz = sizeof(struct ksmbd_spnego_authen_response) + + resp->session_key_len + resp->spnego_blob_len; + } else if (entry->type == KSMBD_EVENT_SHARE_CONFIG_REQUEST) { + struct ksmbd_share_config_response *resp = entry->response; + + if (resp->payload_sz) { + if (resp->payload_sz < resp->veto_list_sz) + return -EINVAL; + + msg_sz = sizeof(struct ksmbd_share_config_response) + + resp->payload_sz; + } + } + + return entry->msg_sz != msg_sz ? -EINVAL : 0; +} + static void *ipc_msg_send_request(struct ksmbd_ipc_msg *msg, unsigned int handle) { struct ipc_msg_table_entry entry; @@ -477,6 +507,13 @@ static void *ipc_msg_send_request(struct ksmbd_ipc_msg *msg, unsigned int handle ret = wait_event_interruptible_timeout(entry.wait, entry.response != NULL, IPC_WAIT_TIMEOUT); + if (entry.response) { + ret = ipc_validate_msg(&entry); + if (ret) { + kvfree(entry.response); + entry.response = NULL; + } + } out: down_write(&ipc_msg_table_lock); hash_del(&entry.ipc_table_hlist); -- cgit v1.2.3-59-g8ed1b From 5ed11af19e56f0434ce0959376d136005745a936 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Tue, 2 Apr 2024 09:31:22 +0900 Subject: ksmbd: do not set SMB2_GLOBAL_CAP_ENCRYPTION for SMB 3.1.1 SMB2_GLOBAL_CAP_ENCRYPTION flag should be used only for 3.0 and 3.0.2 dialects. This flags set cause compatibility problems with other SMB clients. Reported-by: James Christopher Adduono Tested-by: James Christopher Adduono Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2ops.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/smb2ops.c b/fs/smb/server/smb2ops.c index a45f7dca482e..606aa3c5189a 100644 --- a/fs/smb/server/smb2ops.c +++ b/fs/smb/server/smb2ops.c @@ -228,6 +228,11 @@ void init_smb3_0_server(struct ksmbd_conn *conn) conn->cli_cap & SMB2_GLOBAL_CAP_ENCRYPTION) conn->vals->capabilities |= SMB2_GLOBAL_CAP_ENCRYPTION; + if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_ENCRYPTION || + (!(server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_ENCRYPTION_OFF) && + conn->cli_cap & SMB2_GLOBAL_CAP_ENCRYPTION)) + conn->vals->capabilities |= SMB2_GLOBAL_CAP_ENCRYPTION; + if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB3_MULTICHANNEL) conn->vals->capabilities |= SMB2_GLOBAL_CAP_MULTI_CHANNEL; } @@ -278,11 +283,6 @@ int init_smb3_11_server(struct ksmbd_conn *conn) conn->vals->capabilities |= SMB2_GLOBAL_CAP_LEASING | SMB2_GLOBAL_CAP_DIRECTORY_LEASING; - if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_ENCRYPTION || - (!(server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_ENCRYPTION_OFF) && - conn->cli_cap & SMB2_GLOBAL_CAP_ENCRYPTION)) - conn->vals->capabilities |= SMB2_GLOBAL_CAP_ENCRYPTION; - if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB3_MULTICHANNEL) conn->vals->capabilities |= SMB2_GLOBAL_CAP_MULTI_CHANNEL; -- cgit v1.2.3-59-g8ed1b From e9e62243a3e2322cf639f653a0b0a88a76446ce7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 2 Apr 2024 10:11:35 +0100 Subject: cifs: Fix caching to try to do open O_WRONLY as rdwr on server When we're engaged in local caching of a cifs filesystem, we cannot perform caching of a partially written cache granule unless we can read the rest of the granule. This can result in unexpected access errors being reported to the user. Fix this by the following: if a file is opened O_WRONLY locally, but the mount was given the "-o fsc" flag, try first opening the remote file with GENERIC_READ|GENERIC_WRITE and if that returns -EACCES, try dropping the GENERIC_READ and doing the open again. If that last succeeds, invalidate the cache for that file as for O_DIRECT. Fixes: 70431bfd825d ("cifs: Support fscache indexing rewrite") Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org Signed-off-by: Steve French --- fs/smb/client/dir.c | 15 +++++++++++++++ fs/smb/client/file.c | 48 ++++++++++++++++++++++++++++++++++++++---------- fs/smb/client/fscache.h | 6 ++++++ 3 files changed, 59 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/dir.c b/fs/smb/client/dir.c index d11dc3aa458b..864b194dbaa0 100644 --- a/fs/smb/client/dir.c +++ b/fs/smb/client/dir.c @@ -189,6 +189,7 @@ static int cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned int disposition; struct TCP_Server_Info *server = tcon->ses->server; struct cifs_open_parms oparms; + int rdwr_for_fscache = 0; *oplock = 0; if (tcon->ses->server->oplocks) @@ -200,6 +201,10 @@ static int cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned return PTR_ERR(full_path); } + /* If we're caching, we need to be able to fill in around partial writes. */ + if (cifs_fscache_enabled(inode) && (oflags & O_ACCMODE) == O_WRONLY) + rdwr_for_fscache = 1; + #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY if (tcon->unix_ext && cap_unix(tcon->ses) && !tcon->broken_posix_open && (CIFS_UNIX_POSIX_PATH_OPS_CAP & @@ -276,6 +281,8 @@ static int cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned desired_access |= GENERIC_READ; /* is this too little? */ if (OPEN_FMODE(oflags) & FMODE_WRITE) desired_access |= GENERIC_WRITE; + if (rdwr_for_fscache == 1) + desired_access |= GENERIC_READ; disposition = FILE_OVERWRITE_IF; if ((oflags & (O_CREAT | O_EXCL)) == (O_CREAT | O_EXCL)) @@ -304,6 +311,7 @@ static int cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned if (!tcon->unix_ext && (mode & S_IWUGO) == 0) create_options |= CREATE_OPTION_READONLY; +retry_open: oparms = (struct cifs_open_parms) { .tcon = tcon, .cifs_sb = cifs_sb, @@ -317,8 +325,15 @@ static int cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned rc = server->ops->open(xid, &oparms, oplock, buf); if (rc) { cifs_dbg(FYI, "cifs_create returned 0x%x\n", rc); + if (rc == -EACCES && rdwr_for_fscache == 1) { + desired_access &= ~GENERIC_READ; + rdwr_for_fscache = 2; + goto retry_open; + } goto out; } + if (rdwr_for_fscache == 2) + cifs_invalidate_cache(inode, FSCACHE_INVAL_DIO_WRITE); #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY /* diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 16aadce492b2..ab536ce8a04a 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -206,12 +206,12 @@ cifs_mark_open_files_invalid(struct cifs_tcon *tcon) */ } -static inline int cifs_convert_flags(unsigned int flags) +static inline int cifs_convert_flags(unsigned int flags, int rdwr_for_fscache) { if ((flags & O_ACCMODE) == O_RDONLY) return GENERIC_READ; else if ((flags & O_ACCMODE) == O_WRONLY) - return GENERIC_WRITE; + return rdwr_for_fscache == 1 ? (GENERIC_READ | GENERIC_WRITE) : GENERIC_WRITE; else if ((flags & O_ACCMODE) == O_RDWR) { /* GENERIC_ALL is too much permission to request can cause unnecessary access denied on create */ @@ -348,11 +348,16 @@ static int cifs_nt_open(const char *full_path, struct inode *inode, struct cifs_ int create_options = CREATE_NOT_DIR; struct TCP_Server_Info *server = tcon->ses->server; struct cifs_open_parms oparms; + int rdwr_for_fscache = 0; if (!server->ops->open) return -ENOSYS; - desired_access = cifs_convert_flags(f_flags); + /* If we're caching, we need to be able to fill in around partial writes. */ + if (cifs_fscache_enabled(inode) && (f_flags & O_ACCMODE) == O_WRONLY) + rdwr_for_fscache = 1; + + desired_access = cifs_convert_flags(f_flags, rdwr_for_fscache); /********************************************************************* * open flag mapping table: @@ -389,6 +394,7 @@ static int cifs_nt_open(const char *full_path, struct inode *inode, struct cifs_ if (f_flags & O_DIRECT) create_options |= CREATE_NO_BUFFER; +retry_open: oparms = (struct cifs_open_parms) { .tcon = tcon, .cifs_sb = cifs_sb, @@ -400,8 +406,16 @@ static int cifs_nt_open(const char *full_path, struct inode *inode, struct cifs_ }; rc = server->ops->open(xid, &oparms, oplock, buf); - if (rc) + if (rc) { + if (rc == -EACCES && rdwr_for_fscache == 1) { + desired_access = cifs_convert_flags(f_flags, 0); + rdwr_for_fscache = 2; + goto retry_open; + } return rc; + } + if (rdwr_for_fscache == 2) + cifs_invalidate_cache(inode, FSCACHE_INVAL_DIO_WRITE); /* TODO: Add support for calling posix query info but with passing in fid */ if (tcon->unix_ext) @@ -834,11 +848,11 @@ int cifs_open(struct inode *inode, struct file *file) use_cache: fscache_use_cookie(cifs_inode_cookie(file_inode(file)), file->f_mode & FMODE_WRITE); - if (file->f_flags & O_DIRECT && - (!((file->f_flags & O_ACCMODE) != O_RDONLY) || - file->f_flags & O_APPEND)) - cifs_invalidate_cache(file_inode(file), - FSCACHE_INVAL_DIO_WRITE); + if (!(file->f_flags & O_DIRECT)) + goto out; + if ((file->f_flags & (O_ACCMODE | O_APPEND)) == O_RDONLY) + goto out; + cifs_invalidate_cache(file_inode(file), FSCACHE_INVAL_DIO_WRITE); out: free_dentry_path(page); @@ -903,6 +917,7 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush) int disposition = FILE_OPEN; int create_options = CREATE_NOT_DIR; struct cifs_open_parms oparms; + int rdwr_for_fscache = 0; xid = get_xid(); mutex_lock(&cfile->fh_mutex); @@ -966,7 +981,11 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush) } #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ - desired_access = cifs_convert_flags(cfile->f_flags); + /* If we're caching, we need to be able to fill in around partial writes. */ + if (cifs_fscache_enabled(inode) && (cfile->f_flags & O_ACCMODE) == O_WRONLY) + rdwr_for_fscache = 1; + + desired_access = cifs_convert_flags(cfile->f_flags, rdwr_for_fscache); /* O_SYNC also has bit for O_DSYNC so following check picks up either */ if (cfile->f_flags & O_SYNC) @@ -978,6 +997,7 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush) if (server->ops->get_lease_key) server->ops->get_lease_key(inode, &cfile->fid); +retry_open: oparms = (struct cifs_open_parms) { .tcon = tcon, .cifs_sb = cifs_sb, @@ -1003,6 +1023,11 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush) /* indicate that we need to relock the file */ oparms.reconnect = true; } + if (rc == -EACCES && rdwr_for_fscache == 1) { + desired_access = cifs_convert_flags(cfile->f_flags, 0); + rdwr_for_fscache = 2; + goto retry_open; + } if (rc) { mutex_unlock(&cfile->fh_mutex); @@ -1011,6 +1036,9 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush) goto reopen_error_exit; } + if (rdwr_for_fscache == 2) + cifs_invalidate_cache(inode, FSCACHE_INVAL_DIO_WRITE); + #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY reopen_success: #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ diff --git a/fs/smb/client/fscache.h b/fs/smb/client/fscache.h index a3d73720914f..1f2ea9f5cc9a 100644 --- a/fs/smb/client/fscache.h +++ b/fs/smb/client/fscache.h @@ -109,6 +109,11 @@ static inline void cifs_readahead_to_fscache(struct inode *inode, __cifs_readahead_to_fscache(inode, pos, len); } +static inline bool cifs_fscache_enabled(struct inode *inode) +{ + return fscache_cookie_enabled(cifs_inode_cookie(inode)); +} + #else /* CONFIG_CIFS_FSCACHE */ static inline void cifs_fscache_fill_coherency(struct inode *inode, @@ -124,6 +129,7 @@ static inline void cifs_fscache_release_inode_cookie(struct inode *inode) {} static inline void cifs_fscache_unuse_inode_cookie(struct inode *inode, bool update) {} static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { return NULL; } static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) {} +static inline bool cifs_fscache_enabled(struct inode *inode) { return false; } static inline int cifs_fscache_query_occupancy(struct inode *inode, pgoff_t first, unsigned int nr_pages, -- cgit v1.2.3-59-g8ed1b From 062a7f0ff46eb57aff526897bd2bebfdb1d3046a Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Mon, 1 Apr 2024 22:37:42 -0500 Subject: smb: client: guarantee refcounted children from parent session Avoid potential use-after-free bugs when walking DFS referrals, mounting and performing DFS failover by ensuring that all children from parent @tcon->ses are also refcounted. They're all needed across the entire DFS mount. Get rid of @tcon->dfs_ses_list while we're at it, too. Cc: stable@vger.kernel.org # 6.4+ Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202404021527.ZlRkIxgv-lkp@intel.com/ Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/cifsglob.h | 2 -- fs/smb/client/cifsproto.h | 20 +++++++++---------- fs/smb/client/connect.c | 25 ++++++++++++++++++----- fs/smb/client/dfs.c | 51 ++++++++++++++++++++++------------------------- fs/smb/client/dfs.h | 33 +++++++++++++++++++----------- fs/smb/client/dfs_cache.c | 11 +--------- fs/smb/client/misc.c | 6 ------ 7 files changed, 76 insertions(+), 72 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 7ed9d05f6890..286afbe346be 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1281,7 +1281,6 @@ struct cifs_tcon { struct cached_fids *cfids; /* BB add field for back pointer to sb struct(s)? */ #ifdef CONFIG_CIFS_DFS_UPCALL - struct list_head dfs_ses_list; struct delayed_work dfs_cache_work; #endif struct delayed_work query_interfaces; /* query interfaces workqueue job */ @@ -1804,7 +1803,6 @@ struct cifs_mount_ctx { struct TCP_Server_Info *server; struct cifs_ses *ses; struct cifs_tcon *tcon; - struct list_head dfs_ses_list; }; static inline void __free_dfs_info_param(struct dfs_info3_param *param) diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index 0723e1b57256..8e0a348f1f66 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -725,31 +725,31 @@ struct super_block *cifs_get_tcon_super(struct cifs_tcon *tcon); void cifs_put_tcon_super(struct super_block *sb); int cifs_wait_for_server_reconnect(struct TCP_Server_Info *server, bool retry); -/* Put references of @ses and @ses->dfs_root_ses */ +/* Put references of @ses and its children */ static inline void cifs_put_smb_ses(struct cifs_ses *ses) { - struct cifs_ses *rses = ses->dfs_root_ses; + struct cifs_ses *next; - __cifs_put_smb_ses(ses); - if (rses) - __cifs_put_smb_ses(rses); + do { + next = ses->dfs_root_ses; + __cifs_put_smb_ses(ses); + } while ((ses = next)); } -/* Get an active reference of @ses and @ses->dfs_root_ses. +/* Get an active reference of @ses and its children. * * NOTE: make sure to call this function when incrementing reference count of * @ses to ensure that any DFS root session attached to it (@ses->dfs_root_ses) * will also get its reference count incremented. * - * cifs_put_smb_ses() will put both references, so call it when you're done. + * cifs_put_smb_ses() will put all references, so call it when you're done. */ static inline void cifs_smb_ses_inc_refcount(struct cifs_ses *ses) { lockdep_assert_held(&cifs_tcp_ses_lock); - ses->ses_count++; - if (ses->dfs_root_ses) - ses->dfs_root_ses->ses_count++; + for (; ses; ses = ses->dfs_root_ses) + ses->ses_count++; } static inline bool dfs_src_pathname_equal(const char *s1, const char *s2) diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index ee29bc57300c..38b75cfa06e3 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -1866,6 +1866,9 @@ static int match_session(struct cifs_ses *ses, struct smb3_fs_context *ctx) ctx->sectype != ses->sectype) return 0; + if (ctx->dfs_root_ses != ses->dfs_root_ses) + return 0; + /* * If an existing session is limited to less channels than * requested, it should not be reused @@ -2358,9 +2361,9 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) * need to lock before changing something in the session. */ spin_lock(&cifs_tcp_ses_lock); + if (ctx->dfs_root_ses) + cifs_smb_ses_inc_refcount(ctx->dfs_root_ses); ses->dfs_root_ses = ctx->dfs_root_ses; - if (ses->dfs_root_ses) - ses->dfs_root_ses->ses_count++; list_add(&ses->smb_ses_list, &server->smb_ses_list); spin_unlock(&cifs_tcp_ses_lock); @@ -3311,6 +3314,9 @@ void cifs_mount_put_conns(struct cifs_mount_ctx *mnt_ctx) cifs_put_smb_ses(mnt_ctx->ses); else if (mnt_ctx->server) cifs_put_tcp_session(mnt_ctx->server, 0); + mnt_ctx->ses = NULL; + mnt_ctx->tcon = NULL; + mnt_ctx->server = NULL; mnt_ctx->cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_POSIX_PATHS; free_xid(mnt_ctx->xid); } @@ -3589,8 +3595,6 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx) bool isdfs; int rc; - INIT_LIST_HEAD(&mnt_ctx.dfs_ses_list); - rc = dfs_mount_share(&mnt_ctx, &isdfs); if (rc) goto error; @@ -3621,7 +3625,6 @@ out: return rc; error: - dfs_put_root_smb_sessions(&mnt_ctx.dfs_ses_list); cifs_mount_put_conns(&mnt_ctx); return rc; } @@ -3636,6 +3639,18 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx) goto error; rc = cifs_mount_get_tcon(&mnt_ctx); + if (!rc) { + /* + * Prevent superblock from being created with any missing + * connections. + */ + if (WARN_ON(!mnt_ctx.server)) + rc = -EHOSTDOWN; + else if (WARN_ON(!mnt_ctx.ses)) + rc = -EACCES; + else if (WARN_ON(!mnt_ctx.tcon)) + rc = -ENOENT; + } if (rc) goto error; diff --git a/fs/smb/client/dfs.c b/fs/smb/client/dfs.c index 449c59830039..3ec965547e3d 100644 --- a/fs/smb/client/dfs.c +++ b/fs/smb/client/dfs.c @@ -66,33 +66,20 @@ static int get_session(struct cifs_mount_ctx *mnt_ctx, const char *full_path) } /* - * Track individual DFS referral servers used by new DFS mount. - * - * On success, their lifetime will be shared by final tcon (dfs_ses_list). - * Otherwise, they will be put by dfs_put_root_smb_sessions() in cifs_mount(). + * Get an active reference of @ses so that next call to cifs_put_tcon() won't + * release it as any new DFS referrals must go through its IPC tcon. */ -static int add_root_smb_session(struct cifs_mount_ctx *mnt_ctx) +static void add_root_smb_session(struct cifs_mount_ctx *mnt_ctx) { struct smb3_fs_context *ctx = mnt_ctx->fs_ctx; - struct dfs_root_ses *root_ses; struct cifs_ses *ses = mnt_ctx->ses; if (ses) { - root_ses = kmalloc(sizeof(*root_ses), GFP_KERNEL); - if (!root_ses) - return -ENOMEM; - - INIT_LIST_HEAD(&root_ses->list); - spin_lock(&cifs_tcp_ses_lock); cifs_smb_ses_inc_refcount(ses); spin_unlock(&cifs_tcp_ses_lock); - root_ses->ses = ses; - list_add_tail(&root_ses->list, &mnt_ctx->dfs_ses_list); } - /* Select new DFS referral server so that new referrals go through it */ ctx->dfs_root_ses = ses; - return 0; } static inline int parse_dfs_target(struct smb3_fs_context *ctx, @@ -185,11 +172,8 @@ again: continue; } - if (is_refsrv) { - rc = add_root_smb_session(mnt_ctx); - if (rc) - goto out; - } + if (is_refsrv) + add_root_smb_session(mnt_ctx); rc = ref_walk_advance(rw); if (!rc) { @@ -232,6 +216,7 @@ static int __dfs_mount_share(struct cifs_mount_ctx *mnt_ctx) struct smb3_fs_context *ctx = mnt_ctx->fs_ctx; struct cifs_tcon *tcon; char *origin_fullpath; + bool new_tcon = true; int rc; origin_fullpath = dfs_get_path(cifs_sb, ctx->source); @@ -239,6 +224,18 @@ static int __dfs_mount_share(struct cifs_mount_ctx *mnt_ctx) return PTR_ERR(origin_fullpath); rc = dfs_referral_walk(mnt_ctx); + if (!rc) { + /* + * Prevent superblock from being created with any missing + * connections. + */ + if (WARN_ON(!mnt_ctx->server)) + rc = -EHOSTDOWN; + else if (WARN_ON(!mnt_ctx->ses)) + rc = -EACCES; + else if (WARN_ON(!mnt_ctx->tcon)) + rc = -ENOENT; + } if (rc) goto out; @@ -247,15 +244,14 @@ static int __dfs_mount_share(struct cifs_mount_ctx *mnt_ctx) if (!tcon->origin_fullpath) { tcon->origin_fullpath = origin_fullpath; origin_fullpath = NULL; + } else { + new_tcon = false; } spin_unlock(&tcon->tc_lock); - if (list_empty(&tcon->dfs_ses_list)) { - list_replace_init(&mnt_ctx->dfs_ses_list, &tcon->dfs_ses_list); + if (new_tcon) { queue_delayed_work(dfscache_wq, &tcon->dfs_cache_work, dfs_cache_get_ttl() * HZ); - } else { - dfs_put_root_smb_sessions(&mnt_ctx->dfs_ses_list); } out: @@ -298,7 +294,6 @@ int dfs_mount_share(struct cifs_mount_ctx *mnt_ctx, bool *isdfs) if (rc) return rc; - ctx->dfs_root_ses = mnt_ctx->ses; /* * If called with 'nodfs' mount option, then skip DFS resolving. Otherwise unconditionally * try to get an DFS referral (even cached) to determine whether it is an DFS mount. @@ -324,7 +319,9 @@ int dfs_mount_share(struct cifs_mount_ctx *mnt_ctx, bool *isdfs) *isdfs = true; add_root_smb_session(mnt_ctx); - return __dfs_mount_share(mnt_ctx); + rc = __dfs_mount_share(mnt_ctx); + dfs_put_root_smb_sessions(mnt_ctx); + return rc; } /* Update dfs referral path of superblock */ diff --git a/fs/smb/client/dfs.h b/fs/smb/client/dfs.h index 875ab7ae57fc..e5c4dcf83750 100644 --- a/fs/smb/client/dfs.h +++ b/fs/smb/client/dfs.h @@ -7,7 +7,9 @@ #define _CIFS_DFS_H #include "cifsglob.h" +#include "cifsproto.h" #include "fs_context.h" +#include "dfs_cache.h" #include "cifs_unicode.h" #include @@ -114,11 +116,6 @@ static inline void ref_walk_set_tgt_hint(struct dfs_ref_walk *rw) ref_walk_tit(rw)); } -struct dfs_root_ses { - struct list_head list; - struct cifs_ses *ses; -}; - int dfs_parse_target_referral(const char *full_path, const struct dfs_info3_param *ref, struct smb3_fs_context *ctx); int dfs_mount_share(struct cifs_mount_ctx *mnt_ctx, bool *isdfs); @@ -133,20 +130,32 @@ static inline int dfs_get_referral(struct cifs_mount_ctx *mnt_ctx, const char *p { struct smb3_fs_context *ctx = mnt_ctx->fs_ctx; struct cifs_sb_info *cifs_sb = mnt_ctx->cifs_sb; + struct cifs_ses *rses = ctx->dfs_root_ses ?: mnt_ctx->ses; - return dfs_cache_find(mnt_ctx->xid, ctx->dfs_root_ses, cifs_sb->local_nls, + return dfs_cache_find(mnt_ctx->xid, rses, cifs_sb->local_nls, cifs_remap(cifs_sb), path, ref, tl); } -static inline void dfs_put_root_smb_sessions(struct list_head *head) +/* + * cifs_get_smb_ses() already guarantees an active reference of + * @ses->dfs_root_ses when a new session is created, so we need to put extra + * references of all DFS root sessions that were used across the mount process + * in dfs_mount_share(). + */ +static inline void dfs_put_root_smb_sessions(struct cifs_mount_ctx *mnt_ctx) { - struct dfs_root_ses *root, *tmp; + const struct smb3_fs_context *ctx = mnt_ctx->fs_ctx; + struct cifs_ses *ses = ctx->dfs_root_ses; + struct cifs_ses *cur; + + if (!ses) + return; - list_for_each_entry_safe(root, tmp, head, list) { - list_del_init(&root->list); - cifs_put_smb_ses(root->ses); - kfree(root); + for (cur = ses; cur; cur = cur->dfs_root_ses) { + if (cur->dfs_root_ses) + cifs_put_smb_ses(cur->dfs_root_ses); } + cifs_put_smb_ses(ses); } #endif /* _CIFS_DFS_H */ diff --git a/fs/smb/client/dfs_cache.c b/fs/smb/client/dfs_cache.c index 508d831fabe3..0552a864ff08 100644 --- a/fs/smb/client/dfs_cache.c +++ b/fs/smb/client/dfs_cache.c @@ -1278,21 +1278,12 @@ int dfs_cache_remount_fs(struct cifs_sb_info *cifs_sb) void dfs_cache_refresh(struct work_struct *work) { struct TCP_Server_Info *server; - struct dfs_root_ses *rses; struct cifs_tcon *tcon; struct cifs_ses *ses; tcon = container_of(work, struct cifs_tcon, dfs_cache_work.work); - ses = tcon->ses; - server = ses->server; - mutex_lock(&server->refpath_lock); - if (server->leaf_fullpath) - __refresh_tcon(server->leaf_fullpath + 1, ses, false); - mutex_unlock(&server->refpath_lock); - - list_for_each_entry(rses, &tcon->dfs_ses_list, list) { - ses = rses->ses; + for (ses = tcon->ses; ses; ses = ses->dfs_root_ses) { server = ses->server; mutex_lock(&server->refpath_lock); if (server->leaf_fullpath) diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c index c3771fc81328..1ea22b3955a2 100644 --- a/fs/smb/client/misc.c +++ b/fs/smb/client/misc.c @@ -138,9 +138,6 @@ tcon_info_alloc(bool dir_leases_enabled) atomic_set(&ret_buf->num_local_opens, 0); atomic_set(&ret_buf->num_remote_opens, 0); ret_buf->stats_from_time = ktime_get_real_seconds(); -#ifdef CONFIG_CIFS_DFS_UPCALL - INIT_LIST_HEAD(&ret_buf->dfs_ses_list); -#endif return ret_buf; } @@ -156,9 +153,6 @@ tconInfoFree(struct cifs_tcon *tcon) atomic_dec(&tconInfoAllocCount); kfree(tcon->nativeFileSystem); kfree_sensitive(tcon->password); -#ifdef CONFIG_CIFS_DFS_UPCALL - dfs_put_root_smb_sessions(&tcon->dfs_ses_list); -#endif kfree(tcon->origin_fullpath); kfree(tcon); } -- cgit v1.2.3-59-g8ed1b From 0a05ad21d77a188d06481c36d6016805a881bcc0 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Mon, 1 Apr 2024 22:44:07 -0300 Subject: smb: client: refresh referral without acquiring refpath_lock Avoid refreshing DFS referral with refpath_lock acquired as the I/O could block for a while due to a potentially disconnected or slow DFS root server and then making other threads - that use same @server and don't require a DFS root server - unable to make any progress. Cc: stable@vger.kernel.org # 6.4+ Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/dfs_cache.c | 44 ++++++++++++++++++++++++-------------------- 1 file changed, 24 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/dfs_cache.c b/fs/smb/client/dfs_cache.c index 0552a864ff08..11c8efecf7aa 100644 --- a/fs/smb/client/dfs_cache.c +++ b/fs/smb/client/dfs_cache.c @@ -1172,8 +1172,8 @@ static bool is_ses_good(struct cifs_ses *ses) return ret; } -/* Refresh dfs referral of tcon and mark it for reconnect if needed */ -static int __refresh_tcon(const char *path, struct cifs_ses *ses, bool force_refresh) +/* Refresh dfs referral of @ses and mark it for reconnect if needed */ +static void __refresh_ses_referral(struct cifs_ses *ses, bool force_refresh) { struct TCP_Server_Info *server = ses->server; DFS_CACHE_TGT_LIST(old_tl); @@ -1181,10 +1181,21 @@ static int __refresh_tcon(const char *path, struct cifs_ses *ses, bool force_ref bool needs_refresh = false; struct cache_entry *ce; unsigned int xid; + char *path = NULL; int rc = 0; xid = get_xid(); + mutex_lock(&server->refpath_lock); + if (server->leaf_fullpath) { + path = kstrdup(server->leaf_fullpath + 1, GFP_ATOMIC); + if (!path) + rc = -ENOMEM; + } + mutex_unlock(&server->refpath_lock); + if (!path) + goto out; + down_read(&htable_rw_lock); ce = lookup_cache_entry(path); needs_refresh = force_refresh || IS_ERR(ce) || cache_entry_expired(ce); @@ -1218,19 +1229,17 @@ out: free_xid(xid); dfs_cache_free_tgts(&old_tl); dfs_cache_free_tgts(&new_tl); - return rc; + kfree(path); } -static int refresh_tcon(struct cifs_tcon *tcon, bool force_refresh) +static inline void refresh_ses_referral(struct cifs_ses *ses) { - struct TCP_Server_Info *server = tcon->ses->server; - struct cifs_ses *ses = tcon->ses; + __refresh_ses_referral(ses, false); +} - mutex_lock(&server->refpath_lock); - if (server->leaf_fullpath) - __refresh_tcon(server->leaf_fullpath + 1, ses, force_refresh); - mutex_unlock(&server->refpath_lock); - return 0; +static inline void force_refresh_ses_referral(struct cifs_ses *ses) +{ + __refresh_ses_referral(ses, true); } /** @@ -1271,25 +1280,20 @@ int dfs_cache_remount_fs(struct cifs_sb_info *cifs_sb) */ cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; - return refresh_tcon(tcon, true); + force_refresh_ses_referral(tcon->ses); + return 0; } /* Refresh all DFS referrals related to DFS tcon */ void dfs_cache_refresh(struct work_struct *work) { - struct TCP_Server_Info *server; struct cifs_tcon *tcon; struct cifs_ses *ses; tcon = container_of(work, struct cifs_tcon, dfs_cache_work.work); - for (ses = tcon->ses; ses; ses = ses->dfs_root_ses) { - server = ses->server; - mutex_lock(&server->refpath_lock); - if (server->leaf_fullpath) - __refresh_tcon(server->leaf_fullpath + 1, ses, false); - mutex_unlock(&server->refpath_lock); - } + for (ses = tcon->ses; ses; ses = ses->dfs_root_ses) + refresh_ses_referral(ses); queue_delayed_work(dfscache_wq, &tcon->dfs_cache_work, atomic_read(&dfs_cache_ttl) * HZ); -- cgit v1.2.3-59-g8ed1b From 4a5ba0e0bfe552ac7451f57e304f6343c3d87f89 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Mon, 1 Apr 2024 22:44:08 -0300 Subject: smb: client: handle DFS tcons in cifs_construct_tcon() The tcons created by cifs_construct_tcon() on multiuser mounts must also be able to failover and refresh DFS referrals, so set the appropriate fields in order to get a full DFS tcon. They could be shared among different superblocks later, too. Cc: stable@vger.kernel.org # 6.4+ Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202404021518.3Xu2VU4s-lkp@intel.com/ Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/connect.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 38b75cfa06e3..9a3578881496 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -3995,6 +3995,7 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) struct cifs_ses *ses; struct cifs_tcon *tcon = NULL; struct smb3_fs_context *ctx; + char *origin_fullpath = NULL; ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); if (ctx == NULL) @@ -4018,6 +4019,7 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) ctx->sign = master_tcon->ses->sign; ctx->seal = master_tcon->seal; ctx->witness = master_tcon->use_witness; + ctx->dfs_root_ses = master_tcon->ses->dfs_root_ses; rc = cifs_set_vol_auth(ctx, master_tcon->ses); if (rc) { @@ -4037,12 +4039,39 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) goto out; } +#ifdef CONFIG_CIFS_DFS_UPCALL + spin_lock(&master_tcon->tc_lock); + if (master_tcon->origin_fullpath) { + spin_unlock(&master_tcon->tc_lock); + origin_fullpath = dfs_get_path(cifs_sb, cifs_sb->ctx->source); + if (IS_ERR(origin_fullpath)) { + tcon = ERR_CAST(origin_fullpath); + origin_fullpath = NULL; + cifs_put_smb_ses(ses); + goto out; + } + } else { + spin_unlock(&master_tcon->tc_lock); + } +#endif + tcon = cifs_get_tcon(ses, ctx); if (IS_ERR(tcon)) { cifs_put_smb_ses(ses); goto out; } +#ifdef CONFIG_CIFS_DFS_UPCALL + if (origin_fullpath) { + spin_lock(&tcon->tc_lock); + tcon->origin_fullpath = origin_fullpath; + spin_unlock(&tcon->tc_lock); + origin_fullpath = NULL; + queue_delayed_work(dfscache_wq, &tcon->dfs_cache_work, + dfs_cache_get_ttl() * HZ); + } +#endif + #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY if (cap_unix(ses)) reset_cifs_unix_caps(0, tcon, NULL, ctx); @@ -4051,6 +4080,7 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) out: kfree(ctx->username); kfree_sensitive(ctx->password); + kfree(origin_fullpath); kfree(ctx); return tcon; -- cgit v1.2.3-59-g8ed1b From 93cee45ccfebc62a3bb4cd622b89e00c8c7d8493 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Mon, 1 Apr 2024 22:44:09 -0300 Subject: smb: client: serialise cifs_construct_tcon() with cifs_mount_mutex Serialise cifs_construct_tcon() with cifs_mount_mutex to handle parallel mounts that may end up reusing the session and tcon created by it. Cc: stable@vger.kernel.org # 6.4+ Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/connect.c | 13 ++++++++++++- fs/smb/client/fs_context.c | 6 +++--- fs/smb/client/fs_context.h | 12 ++++++++++++ 3 files changed, 27 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 9a3578881496..95e4bda4fd51 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -3988,7 +3988,7 @@ cifs_set_vol_auth(struct smb3_fs_context *ctx, struct cifs_ses *ses) } static struct cifs_tcon * -cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) +__cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) { int rc; struct cifs_tcon *master_tcon = cifs_sb_master_tcon(cifs_sb); @@ -4086,6 +4086,17 @@ out: return tcon; } +static struct cifs_tcon * +cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) +{ + struct cifs_tcon *ret; + + cifs_mount_lock(); + ret = __cifs_construct_tcon(cifs_sb, fsuid); + cifs_mount_unlock(); + return ret; +} + struct cifs_tcon * cifs_sb_master_tcon(struct cifs_sb_info *cifs_sb) { diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c index bdcbe6ff2739..b7bfe705b2c4 100644 --- a/fs/smb/client/fs_context.c +++ b/fs/smb/client/fs_context.c @@ -37,7 +37,7 @@ #include "rfc1002pdu.h" #include "fs_context.h" -static DEFINE_MUTEX(cifs_mount_mutex); +DEFINE_MUTEX(cifs_mount_mutex); static const match_table_t cifs_smb_version_tokens = { { Smb_1, SMB1_VERSION_STRING }, @@ -783,9 +783,9 @@ static int smb3_get_tree(struct fs_context *fc) if (err) return err; - mutex_lock(&cifs_mount_mutex); + cifs_mount_lock(); ret = smb3_get_tree_common(fc); - mutex_unlock(&cifs_mount_mutex); + cifs_mount_unlock(); return ret; } diff --git a/fs/smb/client/fs_context.h b/fs/smb/client/fs_context.h index 7863f2248c4d..8a35645e0b65 100644 --- a/fs/smb/client/fs_context.h +++ b/fs/smb/client/fs_context.h @@ -304,4 +304,16 @@ extern void smb3_update_mnt_flags(struct cifs_sb_info *cifs_sb); #define MAX_CACHED_FIDS 16 extern char *cifs_sanitize_prepath(char *prepath, gfp_t gfp); +extern struct mutex cifs_mount_mutex; + +static inline void cifs_mount_lock(void) +{ + mutex_lock(&cifs_mount_mutex); +} + +static inline void cifs_mount_unlock(void) +{ + mutex_unlock(&cifs_mount_mutex); +} + #endif -- cgit v1.2.3-59-g8ed1b From 141fb8cd206ace23c02cd2791c6da52c1d77d42a Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Tue, 19 Mar 2024 10:54:22 -0700 Subject: btrfs: qgroup: correctly model root qgroup rsv in convert We use add_root_meta_rsv and sub_root_meta_rsv to track prealloc and pertrans reservations for subvolumes when quotas are enabled. The convert function does not properly increment pertrans after decrementing prealloc, so the count is not accurate. Note: we check that the fs is not read-only to mirror the logic in qgroup_convert_meta, which checks that before adding to the pertrans rsv. Fixes: 8287475a2055 ("btrfs: qgroup: Use root::qgroup_meta_rsv_* to record qgroup meta reserved space") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 5f90f0605b12..cf8820ce7aa2 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -4495,6 +4495,8 @@ void btrfs_qgroup_convert_reserved_meta(struct btrfs_root *root, int num_bytes) BTRFS_QGROUP_RSV_META_PREALLOC); trace_qgroup_meta_convert(root, num_bytes); qgroup_convert_meta(fs_info, root->root_key.objectid, num_bytes); + if (!sb_rdonly(fs_info->sb)) + add_root_meta_rsv(root, num_bytes, BTRFS_QGROUP_RSV_META_PERTRANS); } /* -- cgit v1.2.3-59-g8ed1b From 74e97958121aa1f5854da6effba70143f051b0cd Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Thu, 21 Mar 2024 10:02:04 -0700 Subject: btrfs: qgroup: fix qgroup prealloc rsv leak in subvolume operations Create subvolume, create snapshot and delete subvolume all use btrfs_subvolume_reserve_metadata() to reserve metadata for the changes done to the parent subvolume's fs tree, which cannot be mediated in the normal way via start_transaction. When quota groups (squota or qgroups) are enabled, this reserves qgroup metadata of type PREALLOC. Once the operation is associated to a transaction, we convert PREALLOC to PERTRANS, which gets cleared in bulk at the end of the transaction. However, the error paths of these three operations were not implementing this lifecycle correctly. They unconditionally converted the PREALLOC to PERTRANS in a generic cleanup step regardless of errors or whether the operation was fully associated to a transaction or not. This resulted in error paths occasionally converting this rsv to PERTRANS without calling record_root_in_trans successfully, which meant that unless that root got recorded in the transaction by some other thread, the end of the transaction would not free that root's PERTRANS, leaking it. Ultimately, this resulted in hitting a WARN in CONFIG_BTRFS_DEBUG builds at unmount for the leaked reservation. The fix is to ensure that every qgroup PREALLOC reservation observes the following properties: 1. any failure before record_root_in_trans is called successfully results in freeing the PREALLOC reservation. 2. after record_root_in_trans, we convert to PERTRANS, and now the transaction owns freeing the reservation. This patch enforces those properties on the three operations. Without it, generic/269 with squotas enabled at mkfs time would fail in ~5-10 runs on my system. With this patch, it ran successfully 1000 times in a row. Fixes: e85fde5162bf ("btrfs: qgroup: fix qgroup meta rsv leak for subvolume operations") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/inode.c | 13 ++++++++++++- fs/btrfs/ioctl.c | 37 ++++++++++++++++++++++++++++--------- fs/btrfs/root-tree.c | 10 ---------- fs/btrfs/root-tree.h | 2 -- 4 files changed, 40 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 37701531eeb1..1f9e74c4161d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4503,6 +4503,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) struct btrfs_trans_handle *trans; struct btrfs_block_rsv block_rsv; u64 root_flags; + u64 qgroup_reserved = 0; int ret; down_write(&fs_info->subvol_sem); @@ -4547,12 +4548,20 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, 5, true); if (ret) goto out_undead; + qgroup_reserved = block_rsv.qgroup_rsv_reserved; trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out_release; } + ret = btrfs_record_root_in_trans(trans, root); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_end_trans; + } + btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved); + qgroup_reserved = 0; trans->block_rsv = &block_rsv; trans->bytes_reserved = block_rsv.size; @@ -4611,7 +4620,9 @@ out_end_trans: ret = btrfs_end_transaction(trans); inode->i_flags |= S_DEAD; out_release: - btrfs_subvolume_release_metadata(root, &block_rsv); + btrfs_block_rsv_release(fs_info, &block_rsv, (u64)-1, NULL); + if (qgroup_reserved) + btrfs_qgroup_free_meta_prealloc(root, qgroup_reserved); out_undead: if (ret) { spin_lock(&dest->root_item_lock); diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 38459a89b27c..5a507237c4fa 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -613,6 +613,7 @@ static noinline int create_subvol(struct mnt_idmap *idmap, int ret; dev_t anon_dev; u64 objectid; + u64 qgroup_reserved = 0; root_item = kzalloc(sizeof(*root_item), GFP_KERNEL); if (!root_item) @@ -650,13 +651,18 @@ static noinline int create_subvol(struct mnt_idmap *idmap, trans_num_items, false); if (ret) goto out_new_inode_args; + qgroup_reserved = block_rsv.qgroup_rsv_reserved; trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { ret = PTR_ERR(trans); - btrfs_subvolume_release_metadata(root, &block_rsv); - goto out_new_inode_args; + goto out_release_rsv; } + ret = btrfs_record_root_in_trans(trans, BTRFS_I(dir)->root); + if (ret) + goto out; + btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved); + qgroup_reserved = 0; trans->block_rsv = &block_rsv; trans->bytes_reserved = block_rsv.size; /* Tree log can't currently deal with an inode which is a new root. */ @@ -767,9 +773,11 @@ static noinline int create_subvol(struct mnt_idmap *idmap, out: trans->block_rsv = NULL; trans->bytes_reserved = 0; - btrfs_subvolume_release_metadata(root, &block_rsv); - btrfs_end_transaction(trans); +out_release_rsv: + btrfs_block_rsv_release(fs_info, &block_rsv, (u64)-1, NULL); + if (qgroup_reserved) + btrfs_qgroup_free_meta_prealloc(root, qgroup_reserved); out_new_inode_args: btrfs_new_inode_args_destroy(&new_inode_args); out_inode: @@ -791,6 +799,8 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, struct btrfs_pending_snapshot *pending_snapshot; unsigned int trans_num_items; struct btrfs_trans_handle *trans; + struct btrfs_block_rsv *block_rsv; + u64 qgroup_reserved = 0; int ret; /* We do not support snapshotting right now. */ @@ -827,19 +837,19 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, goto free_pending; } - btrfs_init_block_rsv(&pending_snapshot->block_rsv, - BTRFS_BLOCK_RSV_TEMP); + block_rsv = &pending_snapshot->block_rsv; + btrfs_init_block_rsv(block_rsv, BTRFS_BLOCK_RSV_TEMP); /* * 1 to add dir item * 1 to add dir index * 1 to update parent inode item */ trans_num_items = create_subvol_num_items(inherit) + 3; - ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root, - &pending_snapshot->block_rsv, + ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root, block_rsv, trans_num_items, false); if (ret) goto free_pending; + qgroup_reserved = block_rsv->qgroup_rsv_reserved; pending_snapshot->dentry = dentry; pending_snapshot->root = root; @@ -852,6 +862,13 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, ret = PTR_ERR(trans); goto fail; } + ret = btrfs_record_root_in_trans(trans, BTRFS_I(dir)->root); + if (ret) { + btrfs_end_transaction(trans); + goto fail; + } + btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved); + qgroup_reserved = 0; trans->pending_snapshot = pending_snapshot; @@ -881,7 +898,9 @@ fail: if (ret && pending_snapshot->snap) pending_snapshot->snap->anon_dev = 0; btrfs_put_root(pending_snapshot->snap); - btrfs_subvolume_release_metadata(root, &pending_snapshot->block_rsv); + btrfs_block_rsv_release(fs_info, block_rsv, (u64)-1, NULL); + if (qgroup_reserved) + btrfs_qgroup_free_meta_prealloc(root, qgroup_reserved); free_pending: if (pending_snapshot->anon_dev) free_anon_bdev(pending_snapshot->anon_dev); diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 4bb538a372ce..7007f9e0c972 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -548,13 +548,3 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, } return ret; } - -void btrfs_subvolume_release_metadata(struct btrfs_root *root, - struct btrfs_block_rsv *rsv) -{ - struct btrfs_fs_info *fs_info = root->fs_info; - u64 qgroup_to_release; - - btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); - btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release); -} diff --git a/fs/btrfs/root-tree.h b/fs/btrfs/root-tree.h index 6f929cf3bd49..8f5739e732b9 100644 --- a/fs/btrfs/root-tree.h +++ b/fs/btrfs/root-tree.h @@ -18,8 +18,6 @@ struct btrfs_trans_handle; int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, struct btrfs_block_rsv *rsv, int nitems, bool use_global_rsv); -void btrfs_subvolume_release_metadata(struct btrfs_root *root, - struct btrfs_block_rsv *rsv); int btrfs_add_root_ref(struct btrfs_trans_handle *trans, u64 root_id, u64 ref_id, u64 dirid, u64 sequence, const struct fscrypt_str *name); -- cgit v1.2.3-59-g8ed1b From 71537e35c324ea6fbd68377a4f26bb93a831ae35 Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Thu, 21 Mar 2024 10:14:24 -0700 Subject: btrfs: record delayed inode root in transaction When running delayed inode updates, we do not record the inode's root in the transaction, but we do allocate PREALLOC and thus converted PERTRANS space for it. To be sure we free that PERTRANS meta rsv, we must ensure that we record the root in the transaction. Fixes: 4f5427ccce5d ("btrfs: delayed-inode: Use new qgroup meta rsv for delayed inode and item") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/delayed-inode.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index dd6f566a383f..121ab890bd05 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1133,6 +1133,9 @@ __btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans, if (ret) return ret; + ret = btrfs_record_root_in_trans(trans, node->root); + if (ret) + return ret; ret = btrfs_update_delayed_inode(trans, node->root, path, node); return ret; } -- cgit v1.2.3-59-g8ed1b From 211de93367304ab395357f8cb12568a4d1e20701 Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Thu, 21 Mar 2024 10:18:39 -0700 Subject: btrfs: qgroup: convert PREALLOC to PERTRANS after record_root_in_trans The transaction is only able to free PERTRANS reservations for a root once that root has been recorded with the TRANS tag on the roots radix tree. Therefore, until we are sure that this root will get tagged, it isn't safe to convert. Generally, this is not an issue as *some* transaction will likely tag the root before long and this reservation will get freed in that transaction, but technically it could stick around until unmount and result in a warning about leaked metadata reservation space. This path is most exercised by running the generic/269 fstest with CONFIG_BTRFS_DEBUG. Fixes: a6496849671a ("btrfs: fix start transaction qgroup rsv double free") CC: stable@vger.kernel.org # 6.6+ Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 46e8426adf4f..1f10082d35d9 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -745,14 +745,6 @@ again: h->reloc_reserved = reloc_reserved; } - /* - * Now that we have found a transaction to be a part of, convert the - * qgroup reservation from prealloc to pertrans. A different transaction - * can't race in and free our pertrans out from under us. - */ - if (qgroup_reserved) - btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved); - got_it: if (!current->journal_info) current->journal_info = h; @@ -786,8 +778,15 @@ got_it: * not just freed. */ btrfs_end_transaction(h); - return ERR_PTR(ret); + goto reserve_fail; } + /* + * Now that we have found a transaction to be a part of, convert the + * qgroup reservation from prealloc to pertrans. A different transaction + * can't race in and free our pertrans out from under us. + */ + if (qgroup_reserved) + btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved); return h; -- cgit v1.2.3-59-g8ed1b From 3c6f0c5ecc8910d4ffb0dfe85609ebc0c91c8f34 Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Tue, 26 Mar 2024 11:55:22 -0700 Subject: btrfs: make btrfs_clear_delalloc_extent() free delalloc reserve Currently, this call site in btrfs_clear_delalloc_extent() only converts the reservation. We are marking it not delalloc, so I don't think it makes sense to keep the rsv around. This is a path where we are not sure to join a transaction, so it leads to incorrect free-ing during umount. Helps with the pass rate of generic/269 and generic/475. Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1f9e74c4161d..c65fe5de4022 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2533,7 +2533,7 @@ void btrfs_clear_delalloc_extent(struct btrfs_inode *inode, */ if (bits & EXTENT_CLEAR_META_RESV && root != fs_info->tree_root) - btrfs_delalloc_release_metadata(inode, len, false); + btrfs_delalloc_release_metadata(inode, len, true); /* For sanity tests. */ if (btrfs_is_testing(fs_info)) -- cgit v1.2.3-59-g8ed1b From 6e68de0bb0ed59e0554a0c15ede7308c47351e2d Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Tue, 26 Mar 2024 12:01:28 -0700 Subject: btrfs: always clear PERTRANS metadata during commit It is possible to clear a root's IN_TRANS tag from the radix tree, but not clear its PERTRANS, if there is some error in between. Eliminate that possibility by moving the free up to where we clear the tag. Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 1f10082d35d9..85f359e0e0a7 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1494,6 +1494,7 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans) radix_tree_tag_clear(&fs_info->fs_roots_radix, (unsigned long)root->root_key.objectid, BTRFS_ROOT_TRANS_TAG); + btrfs_qgroup_free_meta_all_pertrans(root); spin_unlock(&fs_info->fs_roots_radix_lock); btrfs_free_log(trans, root); @@ -1518,7 +1519,6 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans) if (ret2) return ret2; spin_lock(&fs_info->fs_roots_radix_lock); - btrfs_qgroup_free_meta_all_pertrans(root); } } spin_unlock(&fs_info->fs_roots_radix_lock); -- cgit v1.2.3-59-g8ed1b From 97ca7c1f93bbac6982717a7055cd727813c45e61 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sun, 25 Feb 2024 08:29:25 -0800 Subject: mean_and_variance: Drop always failing tests mean_and_variance_test_2 and mean_and_variance_test_4 always fail. The input parameters to those tests are identical to the input parameters to tests 1 and 3, yet the expected result for tests 2 and 4 is different for the mean and stddev tests. That will always fail. Expected mean_and_variance_get_mean(mv) == mean[i], but mean_and_variance_get_mean(mv) == 22 (0x16) mean[i] == 10 (0xa) Drop the bad tests. Fixes: 65bc41090720 ("mean and variance: More tests") Closes: https://lore.kernel.org/lkml/065b94eb-6a24-4248-b7d7-d3212efb4787@roeck-us.net/ Cc: Kent Overstreet Signed-off-by: Guenter Roeck Signed-off-by: Kent Overstreet --- fs/bcachefs/mean_and_variance_test.c | 28 +--------------------------- 1 file changed, 1 insertion(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/mean_and_variance_test.c b/fs/bcachefs/mean_and_variance_test.c index db63b3f3b338..4c298e74723d 100644 --- a/fs/bcachefs/mean_and_variance_test.c +++ b/fs/bcachefs/mean_and_variance_test.c @@ -136,20 +136,8 @@ static void mean_and_variance_test_1(struct kunit *test) d, mean, stddev, weighted_mean, weighted_stddev); } -static void mean_and_variance_test_2(struct kunit *test) -{ - s64 d[] = { 100, 10, 10, 10, 10, 10, 10 }; - s64 mean[] = { 10, 10, 10, 10, 10, 10, 10 }; - s64 stddev[] = { 9, 9, 9, 9, 9, 9, 9 }; - s64 weighted_mean[] = { 32, 27, 22, 19, 17, 15, 14 }; - s64 weighted_stddev[] = { 38, 35, 31, 27, 24, 21, 18 }; - - do_mean_and_variance_test(test, 10, 6, ARRAY_SIZE(d), 2, - d, mean, stddev, weighted_mean, weighted_stddev); -} - /* Test behaviour where we switch from one steady state to another: */ -static void mean_and_variance_test_3(struct kunit *test) +static void mean_and_variance_test_2(struct kunit *test) { s64 d[] = { 100, 100, 100, 100, 100 }; s64 mean[] = { 22, 32, 40, 46, 50 }; @@ -161,18 +149,6 @@ static void mean_and_variance_test_3(struct kunit *test) d, mean, stddev, weighted_mean, weighted_stddev); } -static void mean_and_variance_test_4(struct kunit *test) -{ - s64 d[] = { 100, 100, 100, 100, 100 }; - s64 mean[] = { 10, 11, 12, 13, 14 }; - s64 stddev[] = { 9, 13, 15, 17, 19 }; - s64 weighted_mean[] = { 32, 49, 61, 71, 78 }; - s64 weighted_stddev[] = { 38, 44, 44, 41, 38 }; - - do_mean_and_variance_test(test, 10, 6, ARRAY_SIZE(d), 2, - d, mean, stddev, weighted_mean, weighted_stddev); -} - static void mean_and_variance_fast_divpow2(struct kunit *test) { s64 i; @@ -230,8 +206,6 @@ static struct kunit_case mean_and_variance_test_cases[] = { KUNIT_CASE(mean_and_variance_weighted_advanced_test), KUNIT_CASE(mean_and_variance_test_1), KUNIT_CASE(mean_and_variance_test_2), - KUNIT_CASE(mean_and_variance_test_3), - KUNIT_CASE(mean_and_variance_test_4), {} }; -- cgit v1.2.3-59-g8ed1b From ca0dcef7cf6c2f746403eac4ac427bd2115e07a8 Mon Sep 17 00:00:00 2001 From: Kunwu Chan Date: Tue, 2 Apr 2024 15:17:56 -0400 Subject: dlm: Simplify the allocation of slab caches in dlm_lowcomms_msg_cache_create Use the new KMEM_CACHE() macro instead of direct kmem_cache_create to simplify the creation of SLAB caches. Signed-off-by: Kunwu Chan Acked-by: Alexander Aring Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lowcomms.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 6296c62c10fa..712165a1e567 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -248,7 +248,7 @@ struct kmem_cache *dlm_lowcomms_writequeue_cache_create(void) struct kmem_cache *dlm_lowcomms_msg_cache_create(void) { - return kmem_cache_create("dlm_msg", sizeof(struct dlm_msg), 0, 0, NULL); + return KMEM_CACHE(dlm_msg, 0); } /* need to held writequeue_lock */ -- cgit v1.2.3-59-g8ed1b From e0319af2b6cdfa7c39edf73dcb813b7ff1261fa5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 2 Apr 2024 16:42:27 -0400 Subject: bcachefs: Improve bch2_btree_update_to_text() Print out the mode as a string, and also print out the btree and watermark. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 41 +++++++++++++++++++++++++------------ fs/bcachefs/btree_update_interior.h | 24 ++++++++++++++-------- 2 files changed, 43 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 9fd2dd0f4682..2b46e6ad248a 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -26,6 +26,13 @@ #include +const char * const bch2_btree_update_modes[] = { +#define x(t) #t, + BCH_WATERMARKS() +#undef x + NULL +}; + static int bch2_btree_insert_node(struct btree_update *, struct btree_trans *, btree_path_idx_t, struct btree *, struct keylist *); static void bch2_btree_update_add_new_node(struct btree_update *, struct btree *); @@ -846,11 +853,11 @@ static void btree_update_updated_node(struct btree_update *as, struct btree *b) mutex_lock(&c->btree_interior_update_lock); list_add_tail(&as->unwritten_list, &c->btree_interior_updates_unwritten); - BUG_ON(as->mode != BTREE_INTERIOR_NO_UPDATE); + BUG_ON(as->mode != BTREE_UPDATE_none); BUG_ON(!btree_node_dirty(b)); BUG_ON(!b->c.level); - as->mode = BTREE_INTERIOR_UPDATING_NODE; + as->mode = BTREE_UPDATE_node; as->b = b; set_btree_node_write_blocked(b); @@ -873,7 +880,7 @@ static void btree_update_reparent(struct btree_update *as, lockdep_assert_held(&c->btree_interior_update_lock); child->b = NULL; - child->mode = BTREE_INTERIOR_UPDATING_AS; + child->mode = BTREE_UPDATE_update; bch2_journal_pin_copy(&c->journal, &as->journal, &child->journal, bch2_update_reparent_journal_pin_flush); @@ -884,7 +891,7 @@ static void btree_update_updated_root(struct btree_update *as, struct btree *b) struct bkey_i *insert = &b->key; struct bch_fs *c = as->c; - BUG_ON(as->mode != BTREE_INTERIOR_NO_UPDATE); + BUG_ON(as->mode != BTREE_UPDATE_none); BUG_ON(as->journal_u64s + jset_u64s(insert->k.u64s) > ARRAY_SIZE(as->journal_entries)); @@ -898,7 +905,7 @@ static void btree_update_updated_root(struct btree_update *as, struct btree *b) mutex_lock(&c->btree_interior_update_lock); list_add_tail(&as->unwritten_list, &c->btree_interior_updates_unwritten); - as->mode = BTREE_INTERIOR_UPDATING_ROOT; + as->mode = BTREE_UPDATE_root; mutex_unlock(&c->btree_interior_update_lock); } @@ -1076,7 +1083,7 @@ static void bch2_btree_update_done(struct btree_update *as, struct btree_trans * struct bch_fs *c = as->c; u64 start_time = as->start_time; - BUG_ON(as->mode == BTREE_INTERIOR_NO_UPDATE); + BUG_ON(as->mode == BTREE_UPDATE_none); if (as->took_gc_lock) up_read(&as->c->gc_lock); @@ -1172,7 +1179,8 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, as->c = c; as->start_time = start_time; as->ip_started = _RET_IP_; - as->mode = BTREE_INTERIOR_NO_UPDATE; + as->mode = BTREE_UPDATE_none; + as->watermark = watermark; as->took_gc_lock = true; as->btree_id = path->btree_id; as->update_level = update_level; @@ -2509,18 +2517,25 @@ void bch2_btree_root_alloc(struct bch_fs *c, enum btree_id id) bch2_trans_run(c, __bch2_btree_root_alloc(trans, id)); } +static void bch2_btree_update_to_text(struct printbuf *out, struct btree_update *as) +{ + prt_printf(out, "%ps: btree=%s watermark=%s mode=%s nodes_written=%u cl.remaining=%u journal_seq=%llu\n", + (void *) as->ip_started, + bch2_btree_id_str(as->btree_id), + bch2_watermarks[as->watermark], + bch2_btree_update_modes[as->mode], + as->nodes_written, + closure_nr_remaining(&as->cl), + as->journal.seq); +} + void bch2_btree_updates_to_text(struct printbuf *out, struct bch_fs *c) { struct btree_update *as; mutex_lock(&c->btree_interior_update_lock); list_for_each_entry(as, &c->btree_interior_update_list, list) - prt_printf(out, "%ps: mode=%u nodes_written=%u cl.remaining=%u journal_seq=%llu\n", - (void *) as->ip_started, - as->mode, - as->nodes_written, - closure_nr_remaining(&as->cl), - as->journal.seq); + bch2_btree_update_to_text(out, as); mutex_unlock(&c->btree_interior_update_lock); } diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index 7ac3e17b84fd..1d41d3f4b107 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -12,6 +12,18 @@ int bch2_btree_node_check_topology(struct btree_trans *, struct btree *); +#define BTREE_UPDATE_MODES() \ + x(none) \ + x(node) \ + x(root) \ + x(update) + +enum btree_update_mode { +#define x(n) BTREE_UPDATE_##n, + BTREE_UPDATE_MODES() +#undef x +}; + /* * Tracks an in progress split/rewrite of a btree node and the update to the * parent node: @@ -39,14 +51,8 @@ struct btree_update { struct list_head list; struct list_head unwritten_list; - /* What kind of update are we doing? */ - enum { - BTREE_INTERIOR_NO_UPDATE, - BTREE_INTERIOR_UPDATING_NODE, - BTREE_INTERIOR_UPDATING_ROOT, - BTREE_INTERIOR_UPDATING_AS, - } mode; - + enum btree_update_mode mode; + enum bch_watermark watermark; unsigned nodes_written:1; unsigned took_gc_lock:1; @@ -56,7 +62,7 @@ struct btree_update { struct disk_reservation disk_res; /* - * BTREE_INTERIOR_UPDATING_NODE: + * BTREE_UPDATE_node: * The update that made the new nodes visible was a regular update to an * existing interior node - @b. We can't write out the update to @b * until the new nodes we created are finished writing, so we block @b -- cgit v1.2.3-59-g8ed1b From 197080156f27b6bf8cf198e9313dfbb94769c736 Mon Sep 17 00:00:00 2001 From: Zhiguo Niu Date: Wed, 27 Mar 2024 16:53:40 +0800 Subject: f2fs: fix to adjust appropirate defragment pg_end A length that exceeds the real size of the inode may be specified from user, although these out-of-range areas are not mapped, but they still need to be check in while loop, which is unnecessary. Signed-off-by: Zhiguo Niu Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 719064730f9a..bd55bbb4fa67 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2626,12 +2626,13 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi, bool fragmented = false; int err; - pg_start = range->start >> PAGE_SHIFT; - pg_end = (range->start + range->len) >> PAGE_SHIFT; - f2fs_balance_fs(sbi, true); inode_lock(inode); + pg_start = range->start >> PAGE_SHIFT; + pg_end = min_t(pgoff_t, + (range->start + range->len) >> PAGE_SHIFT, + DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE)); if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { err = -EINVAL; @@ -2646,8 +2647,9 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi, } /* writeback all dirty pages in the range */ - err = filemap_write_and_wait_range(inode->i_mapping, range->start, - range->start + range->len - 1); + err = filemap_write_and_wait_range(inode->i_mapping, + pg_start << PAGE_SHIFT, + (pg_end << PAGE_SHIFT) - 1); if (err) goto out; -- cgit v1.2.3-59-g8ed1b From 7ee88737ab802ac832f978d6e6258571fe08d870 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 2 Apr 2024 18:30:14 -0400 Subject: bcachefs: Check for bad needs_discard before doing discard In the discard worker, we were failing to validate the bucket state - meaning a corrupt needs_discard btree could cause us to discard a bucket that we shouldn't. If check_alloc_info hasn't run yet we just want to bail out, otherwise it's a filesystem inconsistent error. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 47 +++++++++++++++++++++++------------------- 1 file changed, 26 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 893e38f9db80..4ff56fa4d539 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1713,34 +1713,37 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, if (ret) goto out; - if (BCH_ALLOC_V4_NEED_INC_GEN(&a->v)) { - a->v.gen++; - SET_BCH_ALLOC_V4_NEED_INC_GEN(&a->v, false); - goto write; - } - - if (a->v.journal_seq > c->journal.flushed_seq_ondisk) { - if (c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info) { - bch2_trans_inconsistent(trans, - "clearing need_discard but journal_seq %llu > flushed_seq %llu\n" - "%s", - a->v.journal_seq, - c->journal.flushed_seq_ondisk, - (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); + if (a->v.dirty_sectors) { + if (bch2_trans_inconsistent_on(c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info, + trans, "attempting to discard bucket with dirty data\n%s", + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) ret = -EIO; - } goto out; } if (a->v.data_type != BCH_DATA_need_discard) { - if (c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info) { - bch2_trans_inconsistent(trans, - "bucket incorrectly set in need_discard btree\n" - "%s", - (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); - ret = -EIO; + if (data_type_is_empty(a->v.data_type) && + BCH_ALLOC_V4_NEED_INC_GEN(&a->v)) { + a->v.gen++; + SET_BCH_ALLOC_V4_NEED_INC_GEN(&a->v, false); + goto write; } + if (bch2_trans_inconsistent_on(c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info, + trans, "bucket incorrectly set in need_discard btree\n" + "%s", + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + ret = -EIO; + goto out; + } + + if (a->v.journal_seq > c->journal.flushed_seq_ondisk) { + if (bch2_trans_inconsistent_on(c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info, + trans, "clearing need_discard but journal_seq %llu > flushed_seq %llu\n%s", + a->v.journal_seq, + c->journal.flushed_seq_ondisk, + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + ret = -EIO; goto out; } @@ -1835,6 +1838,7 @@ static int bch2_clear_bucket_needs_discard(struct btree_trans *trans, struct bpo if (ret) goto err; + BUG_ON(a->v.dirty_sectors); SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); a->v.data_type = alloc_data_type(a->v, a->v.data_type); @@ -1942,6 +1946,7 @@ static int invalidate_one_bucket(struct btree_trans *trans, goto out; BUG_ON(a->v.data_type != BCH_DATA_cached); + BUG_ON(a->v.dirty_sectors); if (!a->v.cached_sectors) bch_err(c, "invalidating empty bucket, confused"); -- cgit v1.2.3-59-g8ed1b From fa14b50460baba40c8b1138c517fb8cf04464292 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 2 Apr 2024 18:57:05 -0400 Subject: bcachefs: ratelimit informational fsck errors Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index cbb8b43e419f..b704fb0dda95 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -1131,8 +1131,8 @@ static int check_i_sectors_notnested(struct btree_trans *trans, struct inode_wal i->count = count2; if (i->count != count2) { - bch_err(c, "fsck counted i_sectors wrong for inode %llu:%u: got %llu should be %llu", - w->last_pos.inode, i->snapshot, i->count, count2); + bch_err_ratelimited(c, "fsck counted i_sectors wrong for inode %llu:%u: got %llu should be %llu", + w->last_pos.inode, i->snapshot, i->count, count2); return -BCH_ERR_internal_fsck_err; } @@ -1587,8 +1587,8 @@ static int check_subdir_count_notnested(struct btree_trans *trans, struct inode_ return count2; if (i->count != count2) { - bch_err(c, "fsck counted subdirectories wrong: got %llu should be %llu", - i->count, count2); + bch_err_ratelimited(c, "fsck counted subdirectories wrong for inum %llu:%u: got %llu should be %llu", + w->last_pos.inode, i->snapshot, i->count, count2); i->count = count2; if (i->inode.bi_nlink == i->count) continue; -- cgit v1.2.3-59-g8ed1b From 0200ceed3042222a1f78b3e79ec71a5a52977e3a Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 29 Feb 2024 22:51:38 +0000 Subject: vboxsf: remove redundant variable out_len The variable out_len is being used to accumulate the number of bytes but it is not being used for any other purpose. The variable is redundant and can be removed. Cleans up clang scan build warning: fs/vboxsf/utils.c:443:9: warning: variable 'out_len' set but not used [-Wunused-but-set-variable] Signed-off-by: Colin Ian King Link: https://lore.kernel.org/r/20240229225138.351909-1-colin.i.king@gmail.com Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- fs/vboxsf/utils.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/vboxsf/utils.c b/fs/vboxsf/utils.c index 72ac9320e6a3..9515bbf0b54c 100644 --- a/fs/vboxsf/utils.c +++ b/fs/vboxsf/utils.c @@ -440,7 +440,6 @@ int vboxsf_nlscpy(struct vboxsf_sbi *sbi, char *name, size_t name_bound_len, { const char *in; char *out; - size_t out_len; size_t out_bound_len; size_t in_bound_len; @@ -448,7 +447,6 @@ int vboxsf_nlscpy(struct vboxsf_sbi *sbi, char *name, size_t name_bound_len, in_bound_len = utf8_len; out = name; - out_len = 0; /* Reserve space for terminating 0 */ out_bound_len = name_bound_len - 1; @@ -469,7 +467,6 @@ int vboxsf_nlscpy(struct vboxsf_sbi *sbi, char *name, size_t name_bound_len, out += nb; out_bound_len -= nb; - out_len += nb; } *out = 0; -- cgit v1.2.3-59-g8ed1b From de3f64b738af57e2732b91a0774facc675b75b54 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Wed, 1 Nov 2023 11:49:48 +0100 Subject: vboxsf: Avoid an spurious warning if load_nls_xxx() fails If an load_nls_xxx() function fails a few lines above, the 'sbi->bdi_id' is still 0. So, in the error handling path, we will call ida_simple_remove(..., 0) which is not allocated yet. In order to prevent a spurious "ida_free called for id=0 which is not allocated." message, tweak the error handling path and add a new label. Fixes: 0fd169576648 ("fs: Add VirtualBox guest shared folder (vboxsf) support") Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/d09eaaa4e2e08206c58a1a27ca9b3e81dc168773.1698835730.git.christophe.jaillet@wanadoo.fr Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- fs/vboxsf/super.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c index cabe8ac4fefc..1ed6b1a76bf9 100644 --- a/fs/vboxsf/super.c +++ b/fs/vboxsf/super.c @@ -151,7 +151,7 @@ static int vboxsf_fill_super(struct super_block *sb, struct fs_context *fc) if (!sbi->nls) { vbg_err("vboxsf: Count not load '%s' nls\n", nls_name); err = -EINVAL; - goto fail_free; + goto fail_destroy_idr; } } @@ -224,6 +224,7 @@ fail_free: ida_simple_remove(&vboxsf_bdi_ida, sbi->bdi_id); if (sbi->nls) unload_nls(sbi->nls); +fail_destroy_idr: idr_destroy(&sbi->ino_idr); kfree(sbi); return err; -- cgit v1.2.3-59-g8ed1b From 0141d68f86d78953fb4c3983d666e92f7df4a43d Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Wed, 1 Nov 2023 11:49:49 +0100 Subject: vboxsf: Remove usage of the deprecated ida_simple_xx() API ida_alloc() and ida_free() should be preferred to the deprecated ida_simple_get() and ida_simple_remove(). This is less verbose. Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/b3c057c86b73f0309a6362031d21f4d7ebb60587.1698835730.git.christophe.jaillet@wanadoo.fr Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- fs/vboxsf/super.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c index 1ed6b1a76bf9..ffb1d565da39 100644 --- a/fs/vboxsf/super.c +++ b/fs/vboxsf/super.c @@ -155,7 +155,7 @@ static int vboxsf_fill_super(struct super_block *sb, struct fs_context *fc) } } - sbi->bdi_id = ida_simple_get(&vboxsf_bdi_ida, 0, 0, GFP_KERNEL); + sbi->bdi_id = ida_alloc(&vboxsf_bdi_ida, GFP_KERNEL); if (sbi->bdi_id < 0) { err = sbi->bdi_id; goto fail_free; @@ -221,7 +221,7 @@ fail_unmap: vboxsf_unmap_folder(sbi->root); fail_free: if (sbi->bdi_id >= 0) - ida_simple_remove(&vboxsf_bdi_ida, sbi->bdi_id); + ida_free(&vboxsf_bdi_ida, sbi->bdi_id); if (sbi->nls) unload_nls(sbi->nls); fail_destroy_idr: @@ -269,7 +269,7 @@ static void vboxsf_put_super(struct super_block *sb) vboxsf_unmap_folder(sbi->root); if (sbi->bdi_id >= 0) - ida_simple_remove(&vboxsf_bdi_ida, sbi->bdi_id); + ida_free(&vboxsf_bdi_ida, sbi->bdi_id); if (sbi->nls) unload_nls(sbi->nls); -- cgit v1.2.3-59-g8ed1b From 1ece2c43b88660ddbdf8ecb772e9c41ed9cda3dd Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 19 Mar 2024 12:32:04 -0400 Subject: vboxsf: explicitly deny setlease attempts vboxsf does not break leases on its own, so it can't properly handle the case where the hypervisor changes the data. Don't allow file leases on vboxsf. Signed-off-by: Jeff Layton Link: https://lore.kernel.org/r/20240319-setlease-v1-1-5997d67e04b3@kernel.org Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- fs/vboxsf/file.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/vboxsf/file.c b/fs/vboxsf/file.c index 2307f8037efc..118dedef8ebe 100644 --- a/fs/vboxsf/file.c +++ b/fs/vboxsf/file.c @@ -218,6 +218,7 @@ const struct file_operations vboxsf_reg_fops = { .release = vboxsf_file_release, .fsync = noop_fsync, .splice_read = filemap_splice_read, + .setlease = simple_nosetlease, }; const struct inode_operations vboxsf_reg_iops = { -- cgit v1.2.3-59-g8ed1b From 701b38995e5bdd2a293936c55782140423827fb1 Mon Sep 17 00:00:00 2001 From: Roberto Sassu Date: Wed, 3 Apr 2024 09:57:29 +0200 Subject: security: Place security_path_post_mknod() where the original IMA call was Commit 08abce60d63f ("security: Introduce path_post_mknod hook") introduced security_path_post_mknod(), to replace the IMA-specific call to ima_post_path_mknod(). For symmetry with security_path_mknod(), security_path_post_mknod() was called after a successful mknod operation, for any file type, rather than only for regular files at the time there was the IMA call. However, as reported by VFS maintainers, successful mknod operation does not mean that the dentry always has an inode attached to it (for example, not for FIFOs on a SAMBA mount). If that condition happens, the kernel crashes when security_path_post_mknod() attempts to verify if the inode associated to the dentry is private. Move security_path_post_mknod() where the ima_post_path_mknod() call was, which is obviously correct from IMA/EVM perspective. IMA/EVM are the only in-kernel users, and only need to inspect regular files. Reported-by: Steve French Closes: https://lore.kernel.org/linux-kernel/CAH2r5msAVzxCUHHG8VKrMPUKQHmBpE6K9_vjhgDa1uAvwx4ppw@mail.gmail.com/ Suggested-by: Al Viro Fixes: 08abce60d63f ("security: Introduce path_post_mknod hook") Signed-off-by: Roberto Sassu Reviewed-by: Christian Brauner Reviewed-by: Mimi Zohar Acked-by: Paul Moore Signed-off-by: Linus Torvalds --- fs/namei.c | 7 ++----- security/security.c | 4 ++-- 2 files changed, 4 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index ceb9ddf8dfdd..c5b2a25be7d0 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -4050,6 +4050,8 @@ retry: case 0: case S_IFREG: error = vfs_create(idmap, path.dentry->d_inode, dentry, mode, true); + if (!error) + security_path_post_mknod(idmap, dentry); break; case S_IFCHR: case S_IFBLK: error = vfs_mknod(idmap, path.dentry->d_inode, @@ -4060,11 +4062,6 @@ retry: dentry, mode, 0); break; } - - if (error) - goto out2; - - security_path_post_mknod(idmap, dentry); out2: done_path_create(&path, dentry); if (retry_estale(error, lookup_flags)) { diff --git a/security/security.c b/security/security.c index 7e118858b545..0a9a0ac3f266 100644 --- a/security/security.c +++ b/security/security.c @@ -1793,11 +1793,11 @@ int security_path_mknod(const struct path *dir, struct dentry *dentry, EXPORT_SYMBOL(security_path_mknod); /** - * security_path_post_mknod() - Update inode security field after file creation + * security_path_post_mknod() - Update inode security after reg file creation * @idmap: idmap of the mount * @dentry: new file * - * Update inode security field after a file has been created. + * Update inode security field after a regular file has been created. */ void security_path_post_mknod(struct mnt_idmap *idmap, struct dentry *dentry) { -- cgit v1.2.3-59-g8ed1b From 27fcec6c27caec05a512d2f1be7f855d8899cb8d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 30 Mar 2024 22:25:45 -0400 Subject: bcachefs: Clear recovery_passes_required as they complete without errors Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 4 +--- fs/bcachefs/recovery_passes.c | 41 ++++++++++++++++++++++++++++++++--------- fs/bcachefs/util.h | 10 ++++++++++ 3 files changed, 43 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index e8b434009293..88b794d5dbd0 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -838,9 +838,7 @@ use_clean: } if (!test_bit(BCH_FS_error, &c->flags) && - (!bch2_is_zero(ext->recovery_passes_required, sizeof(ext->recovery_passes_required)) || - !bch2_is_zero(ext->errors_silent, sizeof(ext->errors_silent)))) { - memset(ext->recovery_passes_required, 0, sizeof(ext->recovery_passes_required)); + !bch2_is_zero(ext->errors_silent, sizeof(ext->errors_silent))) { memset(ext->errors_silent, 0, sizeof(ext->errors_silent)); write_sb = true; } diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index 0089d9456b10..b066c5155b74 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -59,18 +59,23 @@ static struct recovery_pass_fn recovery_pass_fns[] = { #undef x }; -u64 bch2_recovery_passes_to_stable(u64 v) -{ - static const u8 map[] = { +static const u8 passes_to_stable_map[] = { #define x(n, id, ...) [BCH_RECOVERY_PASS_##n] = BCH_RECOVERY_PASS_STABLE_##n, BCH_RECOVERY_PASSES() #undef x - }; +}; + +static enum bch_recovery_pass_stable bch2_recovery_pass_to_stable(enum bch_recovery_pass pass) +{ + return passes_to_stable_map[pass]; +} +u64 bch2_recovery_passes_to_stable(u64 v) +{ u64 ret = 0; - for (unsigned i = 0; i < ARRAY_SIZE(map); i++) + for (unsigned i = 0; i < ARRAY_SIZE(passes_to_stable_map); i++) if (v & BIT_ULL(i)) - ret |= BIT_ULL(map[i]); + ret |= BIT_ULL(passes_to_stable_map[i]); return ret; } @@ -116,13 +121,13 @@ int bch2_run_explicit_recovery_pass(struct bch_fs *c, int bch2_run_explicit_recovery_pass_persistent(struct bch_fs *c, enum bch_recovery_pass pass) { - __le64 s = cpu_to_le64(bch2_recovery_passes_to_stable(BIT_ULL(pass))); + enum bch_recovery_pass_stable s = bch2_recovery_pass_to_stable(pass); mutex_lock(&c->sb_lock); struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); - if (!(ext->recovery_passes_required[0] & s)) { - ext->recovery_passes_required[0] |= s; + if (!test_bit_le64(s, ext->recovery_passes_required)) { + __set_bit_le64(s, ext->recovery_passes_required); bch2_write_super(c); } mutex_unlock(&c->sb_lock); @@ -130,6 +135,21 @@ int bch2_run_explicit_recovery_pass_persistent(struct bch_fs *c, return bch2_run_explicit_recovery_pass(c, pass); } +static void bch2_clear_recovery_pass_required(struct bch_fs *c, + enum bch_recovery_pass pass) +{ + enum bch_recovery_pass_stable s = bch2_recovery_pass_to_stable(pass); + + mutex_lock(&c->sb_lock); + struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); + + if (test_bit_le64(s, ext->recovery_passes_required)) { + __clear_bit_le64(s, ext->recovery_passes_required); + bch2_write_super(c); + } + mutex_unlock(&c->sb_lock); +} + u64 bch2_fsck_recovery_passes(void) { u64 ret = 0; @@ -218,6 +238,9 @@ int bch2_run_recovery_passes(struct bch_fs *c) c->recovery_pass_done = max(c->recovery_pass_done, c->curr_recovery_pass); + if (!test_bit(BCH_FS_error, &c->flags)) + bch2_clear_recovery_pass_required(c, c->curr_recovery_pass); + c->curr_recovery_pass++; } diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index 175aee3074c7..4577c0789a3a 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -797,4 +797,14 @@ static inline void __set_bit_le64(size_t bit, __le64 *addr) addr[bit / 64] |= cpu_to_le64(BIT_ULL(bit % 64)); } +static inline void __clear_bit_le64(size_t bit, __le64 *addr) +{ + addr[bit / 64] &= !cpu_to_le64(BIT_ULL(bit % 64)); +} + +static inline bool test_bit_le64(size_t bit, __le64 *addr) +{ + return (addr[bit / 64] & cpu_to_le64(BIT_ULL(bit % 64))) != 0; +} + #endif /* _BCACHEFS_UTIL_H */ -- cgit v1.2.3-59-g8ed1b From bdbf953b3c9036e284e28c9484dda5c81b2a45fa Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 19 Mar 2024 18:56:26 -0400 Subject: bcachefs: bch2_shoot_down_journal_keys() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_journal_iter.c | 19 +++++++++++++++++++ fs/bcachefs/btree_journal_iter.h | 4 ++++ fs/bcachefs/recovery.c | 22 ++++++++++++---------- 3 files changed, 35 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index 1f588264575d..5cbcbfe85235 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -567,3 +567,22 @@ int bch2_journal_keys_sort(struct bch_fs *c) bch_verbose(c, "Journal keys: %zu read, %zu after sorting and compacting", nr_read, keys->nr); return 0; } + +void bch2_shoot_down_journal_keys(struct bch_fs *c, enum btree_id btree, + unsigned level_min, unsigned level_max, + struct bpos start, struct bpos end) +{ + struct journal_keys *keys = &c->journal_keys; + size_t dst = 0; + + move_gap(keys, keys->nr); + + darray_for_each(*keys, i) + if (!(i->btree_id == btree && + i->level >= level_min && + i->level <= level_max && + bpos_ge(i->k->k.p, start) && + bpos_le(i->k->k.p, end))) + keys->data[dst++] = *i; + keys->nr = keys->gap = dst; +} diff --git a/fs/bcachefs/btree_journal_iter.h b/fs/bcachefs/btree_journal_iter.h index 09cd53091a05..af25046ebcaa 100644 --- a/fs/bcachefs/btree_journal_iter.h +++ b/fs/bcachefs/btree_journal_iter.h @@ -66,4 +66,8 @@ void bch2_journal_entries_free(struct bch_fs *); int bch2_journal_keys_sort(struct bch_fs *); +void bch2_shoot_down_journal_keys(struct bch_fs *, enum btree_id, + unsigned, unsigned, + struct bpos, struct bpos); + #endif /* _BCACHEFS_BTREE_JOURNAL_ITER_H */ diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 88b794d5dbd0..7cbf30e3ef73 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -47,7 +47,7 @@ static bool btree_id_is_alloc(enum btree_id id) } /* for -o reconstruct_alloc: */ -static void do_reconstruct_alloc(struct bch_fs *c) +static void bch2_reconstruct_alloc(struct bch_fs *c) { bch2_journal_log_msg(c, "dropping alloc info"); bch_info(c, "dropping and reconstructing all alloc info"); @@ -82,15 +82,17 @@ static void do_reconstruct_alloc(struct bch_fs *c) c->recovery_passes_explicit |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); - struct journal_keys *keys = &c->journal_keys; - size_t src, dst; - - move_gap(keys, keys->nr); - for (src = 0, dst = 0; src < keys->nr; src++) - if (!btree_id_is_alloc(keys->data[src].btree_id)) - keys->data[dst++] = keys->data[src]; - keys->nr = keys->gap = dst; + bch2_shoot_down_journal_keys(c, BTREE_ID_alloc, + 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); + bch2_shoot_down_journal_keys(c, BTREE_ID_backpointers, + 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); + bch2_shoot_down_journal_keys(c, BTREE_ID_need_discard, + 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); + bch2_shoot_down_journal_keys(c, BTREE_ID_freespace, + 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); + bch2_shoot_down_journal_keys(c, BTREE_ID_bucket_gens, + 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); } /* @@ -731,7 +733,7 @@ use_clean: c->journal_replay_seq_end = blacklist_seq - 1; if (c->opts.reconstruct_alloc) - do_reconstruct_alloc(c); + bch2_reconstruct_alloc(c); zero_out_btree_mem_ptr(&c->journal_keys); -- cgit v1.2.3-59-g8ed1b From ca1e02f7e9a1352b3f7b04de821ae74c9e07df74 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 22 Mar 2024 19:26:33 -0400 Subject: bcachefs: Etyzinger cleanups Pull out eytzinger.c and kill eytzinger_cmp_fn. We now provide eytzinger0_sort and eytzinger0_sort_r, which use the standard cmp_func_t and cmp_r_func_t callbacks. Signed-off-by: Kent Overstreet --- fs/bcachefs/Makefile | 1 + fs/bcachefs/eytzinger.c | 234 ++++++++++++++++++++++++++++++++++++ fs/bcachefs/eytzinger.h | 63 ++++++---- fs/bcachefs/journal_seq_blacklist.c | 3 +- fs/bcachefs/replicas.c | 19 +-- fs/bcachefs/util.c | 143 ---------------------- fs/bcachefs/util.h | 4 - 7 files changed, 285 insertions(+), 182 deletions(-) create mode 100644 fs/bcachefs/eytzinger.c (limited to 'fs') diff --git a/fs/bcachefs/Makefile b/fs/bcachefs/Makefile index f6d86eb4b976..0933493a322c 100644 --- a/fs/bcachefs/Makefile +++ b/fs/bcachefs/Makefile @@ -37,6 +37,7 @@ bcachefs-y := \ error.o \ extents.o \ extent_update.o \ + eytzinger.o \ fs.o \ fs-common.o \ fs-ioctl.o \ diff --git a/fs/bcachefs/eytzinger.c b/fs/bcachefs/eytzinger.c new file mode 100644 index 000000000000..4ce5e957a6e9 --- /dev/null +++ b/fs/bcachefs/eytzinger.c @@ -0,0 +1,234 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "eytzinger.h" + +/** + * is_aligned - is this pointer & size okay for word-wide copying? + * @base: pointer to data + * @size: size of each element + * @align: required alignment (typically 4 or 8) + * + * Returns true if elements can be copied using word loads and stores. + * The size must be a multiple of the alignment, and the base address must + * be if we do not have CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS. + * + * For some reason, gcc doesn't know to optimize "if (a & mask || b & mask)" + * to "if ((a | b) & mask)", so we do that by hand. + */ +__attribute_const__ __always_inline +static bool is_aligned(const void *base, size_t size, unsigned char align) +{ + unsigned char lsbits = (unsigned char)size; + + (void)base; +#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS + lsbits |= (unsigned char)(uintptr_t)base; +#endif + return (lsbits & (align - 1)) == 0; +} + +/** + * swap_words_32 - swap two elements in 32-bit chunks + * @a: pointer to the first element to swap + * @b: pointer to the second element to swap + * @n: element size (must be a multiple of 4) + * + * Exchange the two objects in memory. This exploits base+index addressing, + * which basically all CPUs have, to minimize loop overhead computations. + * + * For some reason, on x86 gcc 7.3.0 adds a redundant test of n at the + * bottom of the loop, even though the zero flag is still valid from the + * subtract (since the intervening mov instructions don't alter the flags). + * Gcc 8.1.0 doesn't have that problem. + */ +static void swap_words_32(void *a, void *b, size_t n) +{ + do { + u32 t = *(u32 *)(a + (n -= 4)); + *(u32 *)(a + n) = *(u32 *)(b + n); + *(u32 *)(b + n) = t; + } while (n); +} + +/** + * swap_words_64 - swap two elements in 64-bit chunks + * @a: pointer to the first element to swap + * @b: pointer to the second element to swap + * @n: element size (must be a multiple of 8) + * + * Exchange the two objects in memory. This exploits base+index + * addressing, which basically all CPUs have, to minimize loop overhead + * computations. + * + * We'd like to use 64-bit loads if possible. If they're not, emulating + * one requires base+index+4 addressing which x86 has but most other + * processors do not. If CONFIG_64BIT, we definitely have 64-bit loads, + * but it's possible to have 64-bit loads without 64-bit pointers (e.g. + * x32 ABI). Are there any cases the kernel needs to worry about? + */ +static void swap_words_64(void *a, void *b, size_t n) +{ + do { +#ifdef CONFIG_64BIT + u64 t = *(u64 *)(a + (n -= 8)); + *(u64 *)(a + n) = *(u64 *)(b + n); + *(u64 *)(b + n) = t; +#else + /* Use two 32-bit transfers to avoid base+index+4 addressing */ + u32 t = *(u32 *)(a + (n -= 4)); + *(u32 *)(a + n) = *(u32 *)(b + n); + *(u32 *)(b + n) = t; + + t = *(u32 *)(a + (n -= 4)); + *(u32 *)(a + n) = *(u32 *)(b + n); + *(u32 *)(b + n) = t; +#endif + } while (n); +} + +/** + * swap_bytes - swap two elements a byte at a time + * @a: pointer to the first element to swap + * @b: pointer to the second element to swap + * @n: element size + * + * This is the fallback if alignment doesn't allow using larger chunks. + */ +static void swap_bytes(void *a, void *b, size_t n) +{ + do { + char t = ((char *)a)[--n]; + ((char *)a)[n] = ((char *)b)[n]; + ((char *)b)[n] = t; + } while (n); +} + +/* + * The values are arbitrary as long as they can't be confused with + * a pointer, but small integers make for the smallest compare + * instructions. + */ +#define SWAP_WORDS_64 (swap_r_func_t)0 +#define SWAP_WORDS_32 (swap_r_func_t)1 +#define SWAP_BYTES (swap_r_func_t)2 +#define SWAP_WRAPPER (swap_r_func_t)3 + +struct wrapper { + cmp_func_t cmp; + swap_func_t swap; +}; + +/* + * The function pointer is last to make tail calls most efficient if the + * compiler decides not to inline this function. + */ +static void do_swap(void *a, void *b, size_t size, swap_r_func_t swap_func, const void *priv) +{ + if (swap_func == SWAP_WRAPPER) { + ((const struct wrapper *)priv)->swap(a, b, (int)size); + return; + } + + if (swap_func == SWAP_WORDS_64) + swap_words_64(a, b, size); + else if (swap_func == SWAP_WORDS_32) + swap_words_32(a, b, size); + else if (swap_func == SWAP_BYTES) + swap_bytes(a, b, size); + else + swap_func(a, b, (int)size, priv); +} + +#define _CMP_WRAPPER ((cmp_r_func_t)0L) + +static int do_cmp(const void *a, const void *b, cmp_r_func_t cmp, const void *priv) +{ + if (cmp == _CMP_WRAPPER) + return ((const struct wrapper *)priv)->cmp(a, b); + return cmp(a, b, priv); +} + +static inline int eytzinger0_do_cmp(void *base, size_t n, size_t size, + cmp_r_func_t cmp_func, const void *priv, + size_t l, size_t r) +{ + return do_cmp(base + inorder_to_eytzinger0(l, n) * size, + base + inorder_to_eytzinger0(r, n) * size, + cmp_func, priv); +} + +static inline void eytzinger0_do_swap(void *base, size_t n, size_t size, + swap_r_func_t swap_func, const void *priv, + size_t l, size_t r) +{ + do_swap(base + inorder_to_eytzinger0(l, n) * size, + base + inorder_to_eytzinger0(r, n) * size, + size, swap_func, priv); +} + +void eytzinger0_sort_r(void *base, size_t n, size_t size, + cmp_r_func_t cmp_func, + swap_r_func_t swap_func, + const void *priv) +{ + int i, c, r; + + /* called from 'sort' without swap function, let's pick the default */ + if (swap_func == SWAP_WRAPPER && !((struct wrapper *)priv)->swap) + swap_func = NULL; + + if (!swap_func) { + if (is_aligned(base, size, 8)) + swap_func = SWAP_WORDS_64; + else if (is_aligned(base, size, 4)) + swap_func = SWAP_WORDS_32; + else + swap_func = SWAP_BYTES; + } + + /* heapify */ + for (i = n / 2 - 1; i >= 0; --i) { + for (r = i; r * 2 + 1 < n; r = c) { + c = r * 2 + 1; + + if (c + 1 < n && + eytzinger0_do_cmp(base, n, size, cmp_func, priv, c, c + 1) < 0) + c++; + + if (eytzinger0_do_cmp(base, n, size, cmp_func, priv, r, c) >= 0) + break; + + eytzinger0_do_swap(base, n, size, swap_func, priv, r, c); + } + } + + /* sort */ + for (i = n - 1; i > 0; --i) { + eytzinger0_do_swap(base, n, size, swap_func, priv, 0, i); + + for (r = 0; r * 2 + 1 < i; r = c) { + c = r * 2 + 1; + + if (c + 1 < i && + eytzinger0_do_cmp(base, n, size, cmp_func, priv, c, c + 1) < 0) + c++; + + if (eytzinger0_do_cmp(base, n, size, cmp_func, priv, r, c) >= 0) + break; + + eytzinger0_do_swap(base, n, size, swap_func, priv, r, c); + } + } +} + +void eytzinger0_sort(void *base, size_t n, size_t size, + cmp_func_t cmp_func, + swap_func_t swap_func) +{ + struct wrapper w = { + .cmp = cmp_func, + .swap = swap_func, + }; + + return eytzinger0_sort_r(base, n, size, _CMP_WRAPPER, SWAP_WRAPPER, &w); +} diff --git a/fs/bcachefs/eytzinger.h b/fs/bcachefs/eytzinger.h index b04750dbf870..ee0e2df33322 100644 --- a/fs/bcachefs/eytzinger.h +++ b/fs/bcachefs/eytzinger.h @@ -5,23 +5,33 @@ #include #include -#include "util.h" +#ifdef EYTZINGER_DEBUG +#define EYTZINGER_BUG_ON(cond) BUG_ON(cond) +#else +#define EYTZINGER_BUG_ON(cond) +#endif /* * Traversal for trees in eytzinger layout - a full binary tree layed out in an - * array - */ - -/* - * One based indexing version: + * array. * - * With one based indexing each level of the tree starts at a power of two - - * good for cacheline alignment: + * Consider using an eytzinger tree any time you would otherwise be doing binary + * search over an array. Binary search is a worst case scenario for branch + * prediction and prefetching, but in an eytzinger tree every node's children + * are adjacent in memory, thus we can prefetch children before knowing the + * result of the comparison, assuming multiple nodes fit on a cacheline. + * + * Two variants are provided, for one based indexing and zero based indexing. + * + * Zero based indexing is more convenient, but one based indexing has better + * alignment and thus better performance because each new level of the tree + * starts at a power of two, and thus if element 0 was cacheline aligned, each + * new level will be as well. */ static inline unsigned eytzinger1_child(unsigned i, unsigned child) { - EBUG_ON(child > 1); + EYTZINGER_BUG_ON(child > 1); return (i << 1) + child; } @@ -58,7 +68,7 @@ static inline unsigned eytzinger1_last(unsigned size) static inline unsigned eytzinger1_next(unsigned i, unsigned size) { - EBUG_ON(i > size); + EYTZINGER_BUG_ON(i > size); if (eytzinger1_right_child(i) <= size) { i = eytzinger1_right_child(i); @@ -74,7 +84,7 @@ static inline unsigned eytzinger1_next(unsigned i, unsigned size) static inline unsigned eytzinger1_prev(unsigned i, unsigned size) { - EBUG_ON(i > size); + EYTZINGER_BUG_ON(i > size); if (eytzinger1_left_child(i) <= size) { i = eytzinger1_left_child(i) + 1; @@ -101,7 +111,7 @@ static inline unsigned __eytzinger1_to_inorder(unsigned i, unsigned size, unsigned shift = __fls(size) - b; int s; - EBUG_ON(!i || i > size); + EYTZINGER_BUG_ON(!i || i > size); i ^= 1U << b; i <<= 1; @@ -126,7 +136,7 @@ static inline unsigned __inorder_to_eytzinger1(unsigned i, unsigned size, unsigned shift; int s; - EBUG_ON(!i || i > size); + EYTZINGER_BUG_ON(!i || i > size); /* * sign bit trick: @@ -164,7 +174,7 @@ static inline unsigned inorder_to_eytzinger1(unsigned i, unsigned size) static inline unsigned eytzinger0_child(unsigned i, unsigned child) { - EBUG_ON(child > 1); + EYTZINGER_BUG_ON(child > 1); return (i << 1) + 1 + child; } @@ -231,11 +241,9 @@ static inline unsigned inorder_to_eytzinger0(unsigned i, unsigned size) (_i) != -1; \ (_i) = eytzinger0_next((_i), (_size))) -typedef int (*eytzinger_cmp_fn)(const void *l, const void *r, size_t size); - /* return greatest node <= @search, or -1 if not found */ static inline ssize_t eytzinger0_find_le(void *base, size_t nr, size_t size, - eytzinger_cmp_fn cmp, const void *search) + cmp_func_t cmp, const void *search) { unsigned i, n = 0; @@ -244,21 +252,24 @@ static inline ssize_t eytzinger0_find_le(void *base, size_t nr, size_t size, do { i = n; - n = eytzinger0_child(i, cmp(search, base + i * size, size) >= 0); + n = eytzinger0_child(i, cmp(base + i * size, search) <= 0); } while (n < nr); if (n & 1) { /* @i was greater than @search, return previous node: */ - - if (i == eytzinger0_first(nr)) - return -1; - return eytzinger0_prev(i, nr); } else { return i; } } +static inline ssize_t eytzinger0_find_gt(void *base, size_t nr, size_t size, + cmp_func_t cmp, const void *search) +{ + ssize_t idx = eytzinger0_find_le(base, nr, size, cmp, search); + return eytzinger0_next(idx, size); +} + #define eytzinger0_find(base, nr, size, _cmp, search) \ ({ \ void *_base = (base); \ @@ -269,13 +280,13 @@ static inline ssize_t eytzinger0_find_le(void *base, size_t nr, size_t size, int _res; \ \ while (_i < _nr && \ - (_res = _cmp(_search, _base + _i * _size, _size))) \ + (_res = _cmp(_search, _base + _i * _size))) \ _i = eytzinger0_child(_i, _res > 0); \ _i; \ }) -void eytzinger0_sort(void *, size_t, size_t, - int (*cmp_func)(const void *, const void *, size_t), - void (*swap_func)(void *, void *, size_t)); +void eytzinger0_sort_r(void *, size_t, size_t, + cmp_r_func_t, swap_r_func_t, const void *); +void eytzinger0_sort(void *, size_t, size_t, cmp_func_t, swap_func_t); #endif /* _EYTZINGER_H */ diff --git a/fs/bcachefs/journal_seq_blacklist.c b/fs/bcachefs/journal_seq_blacklist.c index b5303874fc35..37a024e034d4 100644 --- a/fs/bcachefs/journal_seq_blacklist.c +++ b/fs/bcachefs/journal_seq_blacklist.c @@ -95,8 +95,7 @@ out: return ret ?: bch2_blacklist_table_initialize(c); } -static int journal_seq_blacklist_table_cmp(const void *_l, - const void *_r, size_t size) +static int journal_seq_blacklist_table_cmp(const void *_l, const void *_r) { const struct journal_seq_blacklist_table_entry *l = _l; const struct journal_seq_blacklist_table_entry *r = _r; diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c index cc2672c12031..678b9c20e251 100644 --- a/fs/bcachefs/replicas.c +++ b/fs/bcachefs/replicas.c @@ -6,12 +6,15 @@ #include "replicas.h" #include "super-io.h" +#include + static int bch2_cpu_replicas_to_sb_replicas(struct bch_fs *, struct bch_replicas_cpu *); /* Some (buggy!) compilers don't allow memcmp to be passed as a pointer */ -static int bch2_memcmp(const void *l, const void *r, size_t size) +static int bch2_memcmp(const void *l, const void *r, const void *priv) { + size_t size = (size_t) priv; return memcmp(l, r, size); } @@ -39,7 +42,8 @@ void bch2_replicas_entry_sort(struct bch_replicas_entry_v1 *e) static void bch2_cpu_replicas_sort(struct bch_replicas_cpu *r) { - eytzinger0_sort(r->entries, r->nr, r->entry_size, bch2_memcmp, NULL); + eytzinger0_sort_r(r->entries, r->nr, r->entry_size, + bch2_memcmp, NULL, (void *)(size_t)r->entry_size); } static void bch2_replicas_entry_v0_to_text(struct printbuf *out, @@ -228,7 +232,7 @@ static inline int __replicas_entry_idx(struct bch_replicas_cpu *r, verify_replicas_entry(search); -#define entry_cmp(_l, _r, size) memcmp(_l, _r, entry_size) +#define entry_cmp(_l, _r) memcmp(_l, _r, entry_size) idx = eytzinger0_find(r->entries, r->nr, r->entry_size, entry_cmp, search); #undef entry_cmp @@ -824,10 +828,11 @@ static int bch2_cpu_replicas_validate(struct bch_replicas_cpu *cpu_r, { unsigned i; - sort_cmp_size(cpu_r->entries, - cpu_r->nr, - cpu_r->entry_size, - bch2_memcmp, NULL); + sort_r(cpu_r->entries, + cpu_r->nr, + cpu_r->entry_size, + bch2_memcmp, NULL, + (void *)(size_t)cpu_r->entry_size); for (i = 0; i < cpu_r->nr; i++) { struct bch_replicas_entry_v1 *e = diff --git a/fs/bcachefs/util.c b/fs/bcachefs/util.c index 216fadf16928..92c6ad75e702 100644 --- a/fs/bcachefs/util.c +++ b/fs/bcachefs/util.c @@ -707,149 +707,6 @@ void memcpy_from_bio(void *dst, struct bio *src, struct bvec_iter src_iter) } } -static int alignment_ok(const void *base, size_t align) -{ - return IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) || - ((unsigned long)base & (align - 1)) == 0; -} - -static void u32_swap(void *a, void *b, size_t size) -{ - u32 t = *(u32 *)a; - *(u32 *)a = *(u32 *)b; - *(u32 *)b = t; -} - -static void u64_swap(void *a, void *b, size_t size) -{ - u64 t = *(u64 *)a; - *(u64 *)a = *(u64 *)b; - *(u64 *)b = t; -} - -static void generic_swap(void *a, void *b, size_t size) -{ - char t; - - do { - t = *(char *)a; - *(char *)a++ = *(char *)b; - *(char *)b++ = t; - } while (--size > 0); -} - -static inline int do_cmp(void *base, size_t n, size_t size, - int (*cmp_func)(const void *, const void *, size_t), - size_t l, size_t r) -{ - return cmp_func(base + inorder_to_eytzinger0(l, n) * size, - base + inorder_to_eytzinger0(r, n) * size, - size); -} - -static inline void do_swap(void *base, size_t n, size_t size, - void (*swap_func)(void *, void *, size_t), - size_t l, size_t r) -{ - swap_func(base + inorder_to_eytzinger0(l, n) * size, - base + inorder_to_eytzinger0(r, n) * size, - size); -} - -void eytzinger0_sort(void *base, size_t n, size_t size, - int (*cmp_func)(const void *, const void *, size_t), - void (*swap_func)(void *, void *, size_t)) -{ - int i, c, r; - - if (!swap_func) { - if (size == 4 && alignment_ok(base, 4)) - swap_func = u32_swap; - else if (size == 8 && alignment_ok(base, 8)) - swap_func = u64_swap; - else - swap_func = generic_swap; - } - - /* heapify */ - for (i = n / 2 - 1; i >= 0; --i) { - for (r = i; r * 2 + 1 < n; r = c) { - c = r * 2 + 1; - - if (c + 1 < n && - do_cmp(base, n, size, cmp_func, c, c + 1) < 0) - c++; - - if (do_cmp(base, n, size, cmp_func, r, c) >= 0) - break; - - do_swap(base, n, size, swap_func, r, c); - } - } - - /* sort */ - for (i = n - 1; i > 0; --i) { - do_swap(base, n, size, swap_func, 0, i); - - for (r = 0; r * 2 + 1 < i; r = c) { - c = r * 2 + 1; - - if (c + 1 < i && - do_cmp(base, n, size, cmp_func, c, c + 1) < 0) - c++; - - if (do_cmp(base, n, size, cmp_func, r, c) >= 0) - break; - - do_swap(base, n, size, swap_func, r, c); - } - } -} - -void sort_cmp_size(void *base, size_t num, size_t size, - int (*cmp_func)(const void *, const void *, size_t), - void (*swap_func)(void *, void *, size_t size)) -{ - /* pre-scale counters for performance */ - int i = (num/2 - 1) * size, n = num * size, c, r; - - if (!swap_func) { - if (size == 4 && alignment_ok(base, 4)) - swap_func = u32_swap; - else if (size == 8 && alignment_ok(base, 8)) - swap_func = u64_swap; - else - swap_func = generic_swap; - } - - /* heapify */ - for ( ; i >= 0; i -= size) { - for (r = i; r * 2 + size < n; r = c) { - c = r * 2 + size; - if (c < n - size && - cmp_func(base + c, base + c + size, size) < 0) - c += size; - if (cmp_func(base + r, base + c, size) >= 0) - break; - swap_func(base + r, base + c, size); - } - } - - /* sort */ - for (i = n - size; i > 0; i -= size) { - swap_func(base, base + i, size); - for (r = 0; r * 2 + size < i; r = c) { - c = r * 2 + size; - if (c < i - size && - cmp_func(base + c, base + c + size, size) < 0) - c += size; - if (cmp_func(base + r, base + c, size) >= 0) - break; - swap_func(base + r, base + c, size); - } - } -} - #if 0 void eytzinger1_test(void) { diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index 4577c0789a3a..b7e7c29278fc 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -631,10 +631,6 @@ static inline void memset_u64s_tail(void *s, int c, unsigned bytes) memset(s + bytes, c, rem); } -void sort_cmp_size(void *base, size_t num, size_t size, - int (*cmp_func)(const void *, const void *, size_t), - void (*swap_func)(void *, void *, size_t)); - /* just the memmove, doesn't update @_nr */ #define __array_insert_item(_array, _nr, _pos) \ memmove(&(_array)[(_pos) + 1], \ -- cgit v1.2.3-59-g8ed1b From f2f61f4192de536fa36bebe49dc2c241213b53ee Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 14 Mar 2024 22:17:40 -0400 Subject: bcachefs: bch2_btree_root_alloc() -> bch2_btree_root_alloc_fake() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 8 ++++---- fs/bcachefs/btree_update_interior.h | 2 +- fs/bcachefs/recovery.c | 4 ++-- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 2b46e6ad248a..32397b99752f 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -2466,7 +2466,7 @@ void bch2_btree_set_root_for_read(struct bch_fs *c, struct btree *b) bch2_btree_set_root_inmem(c, b); } -static int __bch2_btree_root_alloc(struct btree_trans *trans, enum btree_id id) +static int __bch2_btree_root_alloc_fake(struct btree_trans *trans, enum btree_id id, unsigned level) { struct bch_fs *c = trans->c; struct closure cl; @@ -2485,7 +2485,7 @@ static int __bch2_btree_root_alloc(struct btree_trans *trans, enum btree_id id) set_btree_node_fake(b); set_btree_node_need_rewrite(b); - b->c.level = 0; + b->c.level = level; b->c.btree_id = id; bkey_btree_ptr_init(&b->key); @@ -2512,9 +2512,9 @@ static int __bch2_btree_root_alloc(struct btree_trans *trans, enum btree_id id) return 0; } -void bch2_btree_root_alloc(struct bch_fs *c, enum btree_id id) +void bch2_btree_root_alloc_fake(struct bch_fs *c, enum btree_id id, unsigned level) { - bch2_trans_run(c, __bch2_btree_root_alloc(trans, id)); + bch2_trans_run(c, __bch2_btree_root_alloc_fake(trans, id, level)); } static void bch2_btree_update_to_text(struct printbuf *out, struct btree_update *as) diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index 1d41d3f4b107..88dcf5a22a3b 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -171,7 +171,7 @@ int bch2_btree_node_update_key_get_iter(struct btree_trans *, struct btree *, struct bkey_i *, unsigned, bool); void bch2_btree_set_root_for_read(struct bch_fs *, struct btree *); -void bch2_btree_root_alloc(struct bch_fs *, enum btree_id); +void bch2_btree_root_alloc_fake(struct bch_fs *, enum btree_id, unsigned); static inline unsigned btree_update_reserve_required(struct bch_fs *c, struct btree *b) diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 7cbf30e3ef73..f234de0ac834 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -476,7 +476,7 @@ static int read_btree_roots(struct bch_fs *c) if (!r->b) { r->alive = false; r->level = 0; - bch2_btree_root_alloc(c, i); + bch2_btree_root_alloc_fake(c, i, 0); } } fsck_err: @@ -929,7 +929,7 @@ int bch2_fs_initialize(struct bch_fs *c) set_bit(BCH_FS_may_go_rw, &c->flags); for (unsigned i = 0; i < BTREE_ID_NR; i++) - bch2_btree_root_alloc(c, i); + bch2_btree_root_alloc_fake(c, i, 0); for_each_member_device(c, ca) bch2_dev_usage_init(ca); -- cgit v1.2.3-59-g8ed1b From b268aa4e7fb8be3c50e25a09008fb2feed2cd345 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 10 Mar 2024 16:18:41 -0400 Subject: bcachefs: Don't skip fake btree roots in fsck When a btree root is unreadable, we might still have keys fro the journal to walk and mark. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index e5d2c6daa663..9db9c8a5beaa 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -931,9 +931,6 @@ static int bch2_gc_btree_init(struct btree_trans *trans, b = bch2_btree_id_root(c, btree_id)->b; - if (btree_node_fake(b)) - return 0; - six_lock_read(&b->c.lock, NULL, NULL); printbuf_reset(&buf); bch2_bpos_to_text(&buf, b->data->min_key); -- cgit v1.2.3-59-g8ed1b From 4409b8081d1624af814a9cda781ad9cdda3973cb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 11 Mar 2024 23:11:46 -0400 Subject: bcachefs: Repair pass for scanning for btree nodes If a btree root or interior btree node goes bad, we're going to lose a lot of data, unless we can recover the nodes that it pointed to by scanning. Fortunately btree node headers are fully self describing, and additionally the magic number is xored with the filesytem UUID, so we can do so safely. This implements the scanning - next patch will rework topology repair to make use of the found nodes. Signed-off-by: Kent Overstreet --- fs/bcachefs/Makefile | 1 + fs/bcachefs/bcachefs.h | 3 + fs/bcachefs/btree_node_scan.c | 495 ++++++++++++++++++++++++++++++++++++ fs/bcachefs/btree_node_scan.h | 11 + fs/bcachefs/btree_node_scan_types.h | 30 +++ fs/bcachefs/extents.c | 52 ++-- fs/bcachefs/extents.h | 1 + fs/bcachefs/opts.h | 4 +- fs/bcachefs/recovery.c | 54 ++-- fs/bcachefs/recovery_passes.c | 1 + fs/bcachefs/recovery_passes_types.h | 1 + fs/bcachefs/super.c | 3 + 12 files changed, 605 insertions(+), 51 deletions(-) create mode 100644 fs/bcachefs/btree_node_scan.c create mode 100644 fs/bcachefs/btree_node_scan.h create mode 100644 fs/bcachefs/btree_node_scan_types.h (limited to 'fs') diff --git a/fs/bcachefs/Makefile b/fs/bcachefs/Makefile index 0933493a322c..66ca0bbee639 100644 --- a/fs/bcachefs/Makefile +++ b/fs/bcachefs/Makefile @@ -17,6 +17,7 @@ bcachefs-y := \ btree_journal_iter.o \ btree_key_cache.o \ btree_locking.o \ + btree_node_scan.o \ btree_trans_commit.o \ btree_update.o \ btree_update_interior.o \ diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 963162a627cd..93a61dbaa3d8 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -456,6 +456,7 @@ enum bch_time_stats { #include "alloc_types.h" #include "btree_types.h" +#include "btree_node_scan_types.h" #include "btree_write_buffer_types.h" #include "buckets_types.h" #include "buckets_waiting_for_journal_types.h" @@ -1103,6 +1104,8 @@ struct bch_fs { struct journal_keys journal_keys; struct list_head journal_iters; + struct find_btree_nodes found_btree_nodes; + u64 last_bucket_seq_cleanup; u64 counters_on_mount[BCH_COUNTER_NR]; diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c new file mode 100644 index 000000000000..3f33be7e5e5c --- /dev/null +++ b/fs/bcachefs/btree_node_scan.c @@ -0,0 +1,495 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "bcachefs.h" +#include "btree_cache.h" +#include "btree_io.h" +#include "btree_journal_iter.h" +#include "btree_node_scan.h" +#include "btree_update_interior.h" +#include "buckets.h" +#include "error.h" +#include "journal_io.h" +#include "recovery_passes.h" + +#include +#include + +struct find_btree_nodes_worker { + struct closure *cl; + struct find_btree_nodes *f; + struct bch_dev *ca; +}; + +static void found_btree_node_to_text(struct printbuf *out, struct bch_fs *c, const struct found_btree_node *n) +{ + prt_printf(out, "%s l=%u seq=%u cookie=%llx ", bch2_btree_id_str(n->btree_id), n->level, n->seq, n->cookie); + bch2_bpos_to_text(out, n->min_key); + prt_str(out, "-"); + bch2_bpos_to_text(out, n->max_key); + + if (n->range_updated) + prt_str(out, " range updated"); + if (n->overwritten) + prt_str(out, " overwritten"); + + for (unsigned i = 0; i < n->nr_ptrs; i++) { + prt_char(out, ' '); + bch2_extent_ptr_to_text(out, c, n->ptrs + i); + } +} + +static void found_btree_nodes_to_text(struct printbuf *out, struct bch_fs *c, found_btree_nodes nodes) +{ + printbuf_indent_add(out, 2); + darray_for_each(nodes, i) { + found_btree_node_to_text(out, c, i); + prt_newline(out); + } + printbuf_indent_sub(out, 2); +} + +static void found_btree_node_to_key(struct bkey_i *k, const struct found_btree_node *f) +{ + struct bkey_i_btree_ptr_v2 *bp = bkey_btree_ptr_v2_init(k); + + set_bkey_val_u64s(&bp->k, sizeof(struct bch_btree_ptr_v2) / sizeof(u64) + f->nr_ptrs); + bp->k.p = f->max_key; + bp->v.seq = cpu_to_le64(f->cookie); + bp->v.sectors_written = 0; + bp->v.flags = 0; + bp->v.min_key = f->min_key; + SET_BTREE_PTR_RANGE_UPDATED(&bp->v, f->range_updated); + memcpy(bp->v.start, f->ptrs, sizeof(struct bch_extent_ptr) * f->nr_ptrs); +} + +static bool found_btree_node_is_readable(struct btree_trans *trans, + const struct found_btree_node *f) +{ + struct { __BKEY_PADDED(k, BKEY_BTREE_PTR_VAL_U64s_MAX); } k; + + found_btree_node_to_key(&k.k, f); + + struct btree *b = bch2_btree_node_get_noiter(trans, &k.k, f->btree_id, f->level, false); + bool ret = !IS_ERR_OR_NULL(b); + if (ret) + six_unlock_read(&b->c.lock); + + /* + * We might update this node's range; if that happens, we need the node + * to be re-read so the read path can trim keys that are no longer in + * this node + */ + if (b != btree_node_root(trans->c, b)) + bch2_btree_node_evict(trans, &k.k); + return ret; +} + +static int found_btree_node_cmp_cookie(const void *_l, const void *_r) +{ + const struct found_btree_node *l = _l; + const struct found_btree_node *r = _r; + + return cmp_int(l->btree_id, r->btree_id) ?: + cmp_int(l->level, r->level) ?: + cmp_int(l->cookie, r->cookie); +} + +/* + * Given two found btree nodes, if their sequence numbers are equal, take the + * one that's readable: + */ +static int found_btree_node_cmp_time(const struct found_btree_node *l, + const struct found_btree_node *r) +{ + return cmp_int(l->seq, r->seq); +} + +static int found_btree_node_cmp_pos(const void *_l, const void *_r) +{ + const struct found_btree_node *l = _l; + const struct found_btree_node *r = _r; + + return cmp_int(l->btree_id, r->btree_id) ?: + -cmp_int(l->level, r->level) ?: + bpos_cmp(l->min_key, r->min_key) ?: + -found_btree_node_cmp_time(l, r); +} + +static void try_read_btree_node(struct find_btree_nodes *f, struct bch_dev *ca, + struct bio *bio, struct btree_node *bn, u64 offset) +{ + struct bch_fs *c = container_of(f, struct bch_fs, found_btree_nodes); + + bio_reset(bio, ca->disk_sb.bdev, REQ_OP_READ); + bio->bi_iter.bi_sector = offset; + bch2_bio_map(bio, bn, PAGE_SIZE); + + submit_bio_wait(bio); + if (bch2_dev_io_err_on(bio->bi_status, ca, BCH_MEMBER_ERROR_read, + "IO error in try_read_btree_node() at %llu: %s", + offset, bch2_blk_status_to_str(bio->bi_status))) + return; + + if (le64_to_cpu(bn->magic) != bset_magic(c)) + return; + + rcu_read_lock(); + struct found_btree_node n = { + .btree_id = BTREE_NODE_ID(bn), + .level = BTREE_NODE_LEVEL(bn), + .seq = BTREE_NODE_SEQ(bn), + .cookie = le64_to_cpu(bn->keys.seq), + .min_key = bn->min_key, + .max_key = bn->max_key, + .nr_ptrs = 1, + .ptrs[0].type = 1 << BCH_EXTENT_ENTRY_ptr, + .ptrs[0].offset = offset, + .ptrs[0].dev = ca->dev_idx, + .ptrs[0].gen = *bucket_gen(ca, sector_to_bucket(ca, offset)), + }; + rcu_read_unlock(); + + if (bch2_trans_run(c, found_btree_node_is_readable(trans, &n))) { + mutex_lock(&f->lock); + if (BSET_BIG_ENDIAN(&bn->keys) != CPU_BIG_ENDIAN) { + bch_err(c, "try_read_btree_node() can't handle endian conversion"); + f->ret = -EINVAL; + goto unlock; + } + + if (darray_push(&f->nodes, n)) + f->ret = -ENOMEM; +unlock: + mutex_unlock(&f->lock); + } +} + +static int read_btree_nodes_worker(void *p) +{ + struct find_btree_nodes_worker *w = p; + struct bch_fs *c = container_of(w->f, struct bch_fs, found_btree_nodes); + struct bch_dev *ca = w->ca; + void *buf = (void *) __get_free_page(GFP_KERNEL); + struct bio *bio = bio_alloc(NULL, 1, 0, GFP_KERNEL); + unsigned long last_print = jiffies; + + if (!buf || !bio) { + bch_err(c, "read_btree_nodes_worker: error allocating bio/buf"); + w->f->ret = -ENOMEM; + goto err; + } + + for (u64 bucket = ca->mi.first_bucket; bucket < ca->mi.nbuckets; bucket++) + for (unsigned bucket_offset = 0; + bucket_offset + btree_sectors(c) <= ca->mi.bucket_size; + bucket_offset += btree_sectors(c)) { + if (time_after(jiffies, last_print + HZ * 30)) { + u64 cur_sector = bucket * ca->mi.bucket_size + bucket_offset; + u64 end_sector = ca->mi.nbuckets * ca->mi.bucket_size; + + bch_info(ca, "%s: %2u%% done", __func__, + (unsigned) div64_u64(cur_sector * 100, end_sector)); + last_print = jiffies; + } + + try_read_btree_node(w->f, ca, bio, buf, + bucket * ca->mi.bucket_size + bucket_offset); + } +err: + bio_put(bio); + free_page((unsigned long) buf); + percpu_ref_get(&ca->io_ref); + closure_put(w->cl); + kfree(w); + return 0; +} + +static int read_btree_nodes(struct find_btree_nodes *f) +{ + struct bch_fs *c = container_of(f, struct bch_fs, found_btree_nodes); + struct closure cl; + int ret = 0; + + closure_init_stack(&cl); + + for_each_online_member(c, ca) { + struct find_btree_nodes_worker *w = kmalloc(sizeof(*w), GFP_KERNEL); + struct task_struct *t; + + if (!w) { + percpu_ref_put(&ca->io_ref); + ret = -ENOMEM; + goto err; + } + + percpu_ref_get(&ca->io_ref); + closure_get(&cl); + w->cl = &cl; + w->f = f; + w->ca = ca; + + t = kthread_run(read_btree_nodes_worker, w, "read_btree_nodes/%s", ca->name); + ret = IS_ERR_OR_NULL(t); + if (ret) { + percpu_ref_put(&ca->io_ref); + closure_put(&cl); + f->ret = ret; + bch_err(c, "error starting kthread: %i", ret); + break; + } + } +err: + closure_sync(&cl); + return f->ret ?: ret; +} + +static void bubble_up(struct found_btree_node *n, struct found_btree_node *end) +{ + while (n + 1 < end && + found_btree_node_cmp_pos(n, n + 1) > 0) { + swap(n[0], n[1]); + n++; + } +} + +static int handle_overwrites(struct bch_fs *c, + struct found_btree_node *start, + struct found_btree_node *end) +{ + struct found_btree_node *n; +again: + for (n = start + 1; + n < end && + n->btree_id == start->btree_id && + n->level == start->level && + bpos_lt(n->min_key, start->max_key); + n++) { + int cmp = found_btree_node_cmp_time(start, n); + + if (cmp > 0) { + if (bpos_cmp(start->max_key, n->max_key) >= 0) + n->overwritten = true; + else { + n->range_updated = true; + n->min_key = bpos_successor(start->max_key); + n->range_updated = true; + bubble_up(n, end); + goto again; + } + } else if (cmp < 0) { + BUG_ON(bpos_cmp(n->min_key, start->min_key) <= 0); + + start->max_key = bpos_predecessor(n->min_key); + start->range_updated = true; + } else { + struct printbuf buf = PRINTBUF; + + prt_str(&buf, "overlapping btree nodes with same seq! halting\n "); + found_btree_node_to_text(&buf, c, start); + prt_str(&buf, "\n "); + found_btree_node_to_text(&buf, c, n); + bch_err(c, "%s", buf.buf); + printbuf_exit(&buf); + return -1; + } + } + + return 0; +} + +int bch2_scan_for_btree_nodes(struct bch_fs *c) +{ + struct find_btree_nodes *f = &c->found_btree_nodes; + struct printbuf buf = PRINTBUF; + size_t dst; + int ret = 0; + + if (f->nodes.nr) + return 0; + + mutex_init(&f->lock); + + ret = read_btree_nodes(f); + if (ret) + return ret; + + if (!f->nodes.nr) { + bch_err(c, "%s: no btree nodes found", __func__); + ret = -EINVAL; + goto err; + } + + if (0 && c->opts.verbose) { + printbuf_reset(&buf); + prt_printf(&buf, "%s: nodes found:\n", __func__); + found_btree_nodes_to_text(&buf, c, f->nodes); + bch2_print_string_as_lines(KERN_INFO, buf.buf); + } + + sort(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_cookie, NULL); + + dst = 0; + darray_for_each(f->nodes, i) { + struct found_btree_node *prev = dst ? f->nodes.data + dst - 1 : NULL; + + if (prev && + prev->cookie == i->cookie) { + if (prev->nr_ptrs == ARRAY_SIZE(prev->ptrs)) { + bch_err(c, "%s: found too many replicas for btree node", __func__); + ret = -EINVAL; + goto err; + } + prev->ptrs[prev->nr_ptrs++] = i->ptrs[0]; + } else { + f->nodes.data[dst++] = *i; + } + } + f->nodes.nr = dst; + + sort(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_pos, NULL); + + if (0 && c->opts.verbose) { + printbuf_reset(&buf); + prt_printf(&buf, "%s: nodes after merging replicas:\n", __func__); + found_btree_nodes_to_text(&buf, c, f->nodes); + bch2_print_string_as_lines(KERN_INFO, buf.buf); + } + + dst = 0; + darray_for_each(f->nodes, i) { + if (i->overwritten) + continue; + + ret = handle_overwrites(c, i, &darray_top(f->nodes)); + if (ret) + goto err; + + BUG_ON(i->overwritten); + f->nodes.data[dst++] = *i; + } + f->nodes.nr = dst; + + if (c->opts.verbose) { + printbuf_reset(&buf); + prt_printf(&buf, "%s: nodes found after overwrites:\n", __func__); + found_btree_nodes_to_text(&buf, c, f->nodes); + bch2_print_string_as_lines(KERN_INFO, buf.buf); + } + + eytzinger0_sort(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_pos, NULL); +err: + printbuf_exit(&buf); + return ret; +} + +static int found_btree_node_range_start_cmp(const void *_l, const void *_r) +{ + const struct found_btree_node *l = _l; + const struct found_btree_node *r = _r; + + return cmp_int(l->btree_id, r->btree_id) ?: + -cmp_int(l->level, r->level) ?: + bpos_cmp(l->max_key, r->min_key); +} + +#define for_each_found_btree_node_in_range(_f, _search, _idx) \ + for (size_t _idx = eytzinger0_find_gt((_f)->nodes.data, (_f)->nodes.nr, \ + sizeof((_f)->nodes.data[0]), \ + found_btree_node_range_start_cmp, &search); \ + _idx < (_f)->nodes.nr && \ + (_f)->nodes.data[_idx].btree_id == _search.btree_id && \ + (_f)->nodes.data[_idx].level == _search.level && \ + bpos_lt((_f)->nodes.data[_idx].min_key, _search.max_key); \ + _idx = eytzinger0_next(_idx, (_f)->nodes.nr)) + +bool bch2_btree_node_is_stale(struct bch_fs *c, struct btree *b) +{ + struct find_btree_nodes *f = &c->found_btree_nodes; + + struct found_btree_node search = { + .btree_id = b->c.btree_id, + .level = b->c.level, + .min_key = b->data->min_key, + .max_key = b->key.k.p, + }; + + for_each_found_btree_node_in_range(f, search, idx) + if (f->nodes.data[idx].seq > BTREE_NODE_SEQ(b->data)) + return true; + return false; +} + +bool bch2_btree_has_scanned_nodes(struct bch_fs *c, enum btree_id btree) +{ + struct found_btree_node search = { + .btree_id = btree, + .level = 0, + .min_key = POS_MIN, + .max_key = SPOS_MAX, + }; + + for_each_found_btree_node_in_range(&c->found_btree_nodes, search, idx) + return true; + return false; +} + +int bch2_get_scanned_nodes(struct bch_fs *c, enum btree_id btree, + unsigned level, struct bpos node_min, struct bpos node_max) +{ + struct find_btree_nodes *f = &c->found_btree_nodes; + + int ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes); + if (ret) + return ret; + + if (c->opts.verbose) { + struct printbuf buf = PRINTBUF; + + prt_printf(&buf, "recovering %s l=%u ", bch2_btree_id_str(btree), level); + bch2_bpos_to_text(&buf, node_min); + prt_str(&buf, " - "); + bch2_bpos_to_text(&buf, node_max); + + bch_info(c, "%s(): %s", __func__, buf.buf); + printbuf_exit(&buf); + } + + struct found_btree_node search = { + .btree_id = btree, + .level = level, + .min_key = node_min, + .max_key = node_max, + }; + + for_each_found_btree_node_in_range(f, search, idx) { + struct found_btree_node n = f->nodes.data[idx]; + + n.range_updated |= bpos_lt(n.min_key, node_min); + n.min_key = bpos_max(n.min_key, node_min); + + n.range_updated |= bpos_gt(n.max_key, node_max); + n.max_key = bpos_min(n.max_key, node_max); + + struct { __BKEY_PADDED(k, BKEY_BTREE_PTR_VAL_U64s_MAX); } tmp; + + found_btree_node_to_key(&tmp.k, &n); + + struct printbuf buf = PRINTBUF; + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&tmp.k)); + bch_verbose(c, "%s(): recovering %s", __func__, buf.buf); + printbuf_exit(&buf); + + BUG_ON(bch2_bkey_invalid(c, bkey_i_to_s_c(&tmp.k), BKEY_TYPE_btree, 0, NULL)); + + ret = bch2_journal_key_insert(c, btree, level + 1, &tmp.k); + if (ret) + return ret; + } + + return 0; +} + +void bch2_find_btree_nodes_exit(struct find_btree_nodes *f) +{ + darray_exit(&f->nodes); +} diff --git a/fs/bcachefs/btree_node_scan.h b/fs/bcachefs/btree_node_scan.h new file mode 100644 index 000000000000..08687b209787 --- /dev/null +++ b/fs/bcachefs/btree_node_scan.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _BCACHEFS_BTREE_NODE_SCAN_H +#define _BCACHEFS_BTREE_NODE_SCAN_H + +int bch2_scan_for_btree_nodes(struct bch_fs *); +bool bch2_btree_node_is_stale(struct bch_fs *, struct btree *); +bool bch2_btree_has_scanned_nodes(struct bch_fs *, enum btree_id); +int bch2_get_scanned_nodes(struct bch_fs *, enum btree_id, unsigned, struct bpos, struct bpos); +void bch2_find_btree_nodes_exit(struct find_btree_nodes *); + +#endif /* _BCACHEFS_BTREE_NODE_SCAN_H */ diff --git a/fs/bcachefs/btree_node_scan_types.h b/fs/bcachefs/btree_node_scan_types.h new file mode 100644 index 000000000000..abb7b27d556a --- /dev/null +++ b/fs/bcachefs/btree_node_scan_types.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _BCACHEFS_BTREE_NODE_SCAN_TYPES_H +#define _BCACHEFS_BTREE_NODE_SCAN_TYPES_H + +#include "darray.h" + +struct found_btree_node { + bool range_updated:1; + bool overwritten:1; + u8 btree_id; + u8 level; + u32 seq; + u64 cookie; + + struct bpos min_key; + struct bpos max_key; + + unsigned nr_ptrs; + struct bch_extent_ptr ptrs[BCH_REPLICAS_MAX]; +}; + +typedef DARRAY(struct found_btree_node) found_btree_nodes; + +struct find_btree_nodes { + int ret; + struct mutex lock; + found_btree_nodes nodes; +}; + +#endif /* _BCACHEFS_BTREE_NODE_SCAN_TYPES_H */ diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index b2432d88cda6..0e3ca99fbd2d 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -978,6 +978,31 @@ bool bch2_extent_normalize(struct bch_fs *c, struct bkey_s k) return bkey_deleted(k.k); } +void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struct bch_extent_ptr *ptr) +{ + struct bch_dev *ca = c && ptr->dev < c->sb.nr_devices && c->devs[ptr->dev] + ? bch_dev_bkey_exists(c, ptr->dev) + : NULL; + + if (!ca) { + prt_printf(out, "ptr: %u:%llu gen %u%s", ptr->dev, + (u64) ptr->offset, ptr->gen, + ptr->cached ? " cached" : ""); + } else { + u32 offset; + u64 b = sector_to_bucket_and_offset(ca, ptr->offset, &offset); + + prt_printf(out, "ptr: %u:%llu:%u gen %u", + ptr->dev, b, offset, ptr->gen); + if (ptr->cached) + prt_str(out, " cached"); + if (ptr->unwritten) + prt_str(out, " unwritten"); + if (ca && ptr_stale(ca, ptr)) + prt_printf(out, " stale"); + } +} + void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) { @@ -993,31 +1018,10 @@ void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, prt_printf(out, " "); switch (__extent_entry_type(entry)) { - case BCH_EXTENT_ENTRY_ptr: { - const struct bch_extent_ptr *ptr = entry_to_ptr(entry); - struct bch_dev *ca = c && ptr->dev < c->sb.nr_devices && c->devs[ptr->dev] - ? bch_dev_bkey_exists(c, ptr->dev) - : NULL; - - if (!ca) { - prt_printf(out, "ptr: %u:%llu gen %u%s", ptr->dev, - (u64) ptr->offset, ptr->gen, - ptr->cached ? " cached" : ""); - } else { - u32 offset; - u64 b = sector_to_bucket_and_offset(ca, ptr->offset, &offset); - - prt_printf(out, "ptr: %u:%llu:%u gen %u", - ptr->dev, b, offset, ptr->gen); - if (ptr->cached) - prt_str(out, " cached"); - if (ptr->unwritten) - prt_str(out, " unwritten"); - if (ca && ptr_stale(ca, ptr)) - prt_printf(out, " stale"); - } + case BCH_EXTENT_ENTRY_ptr: + bch2_extent_ptr_to_text(out, c, entry_to_ptr(entry)); break; - } + case BCH_EXTENT_ENTRY_crc32: case BCH_EXTENT_ENTRY_crc64: case BCH_EXTENT_ENTRY_crc128: { diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h index 3fd0169b98c1..528e817eacbd 100644 --- a/fs/bcachefs/extents.h +++ b/fs/bcachefs/extents.h @@ -676,6 +676,7 @@ bch2_extent_has_ptr(struct bkey_s_c, struct extent_ptr_decoded, struct bkey_s); void bch2_extent_ptr_set_cached(struct bkey_s, struct bch_extent_ptr *); bool bch2_extent_normalize(struct bch_fs *, struct bkey_s); +void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *, const struct bch_extent_ptr *); void bch2_bkey_ptrs_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_bkey_ptrs_invalid(struct bch_fs *, struct bkey_s_c, diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index 084247c13bd8..1ac4135cca1c 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -368,11 +368,11 @@ enum fsck_err_opts { OPT_STR_NOLIMIT(bch2_recovery_passes), \ BCH2_NO_SB_OPT, 0, \ NULL, "Exit recovery after specified pass") \ - x(keep_journal, u8, \ + x(retain_recovery_info, u8, \ 0, \ OPT_BOOL(), \ BCH2_NO_SB_OPT, false, \ - NULL, "Don't free journal entries/keys after startup")\ + NULL, "Don't free journal entries/keys, scanned btree nodes after startup")\ x(read_entire_journal, u8, \ 0, \ OPT_BOOL(), \ diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index f234de0ac834..24671020f22b 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -4,6 +4,7 @@ #include "alloc_background.h" #include "bkey_buf.h" #include "btree_journal_iter.h" +#include "btree_node_scan.h" #include "btree_update.h" #include "btree_update_interior.h" #include "btree_io.h" @@ -271,7 +272,7 @@ int bch2_journal_replay(struct bch_fs *c) bch2_trans_put(trans); trans = NULL; - if (!c->opts.keep_journal && + if (!c->opts.retain_recovery_info && c->recovery_pass_done >= BCH_RECOVERY_PASS_journal_replay) bch2_journal_keys_put_initial(c); @@ -435,10 +436,9 @@ static int journal_replay_early(struct bch_fs *c, static int read_btree_roots(struct bch_fs *c) { - unsigned i; int ret = 0; - for (i = 0; i < btree_id_nr_alive(c); i++) { + for (unsigned i = 0; i < btree_id_nr_alive(c); i++) { struct btree_root *r = bch2_btree_id_root(c, i); if (!r->alive) @@ -447,33 +447,36 @@ static int read_btree_roots(struct bch_fs *c) if (btree_id_is_alloc(i) && c->opts.reconstruct_alloc) continue; - if (r->error) { - __fsck_err(c, - btree_id_is_alloc(i) - ? FSCK_CAN_IGNORE : 0, - btree_root_bkey_invalid, - "invalid btree root %s", - bch2_btree_id_str(i)); - if (i == BTREE_ID_alloc) + if (mustfix_fsck_err_on((ret = r->error), + c, btree_root_bkey_invalid, + "invalid btree root %s", + bch2_btree_id_str(i)) || + mustfix_fsck_err_on((ret = r->error = bch2_btree_root_read(c, i, &r->key, r->level)), + c, btree_root_read_error, + "error reading btree root %s l=%u: %s", + bch2_btree_id_str(i), r->level, bch2_err_str(ret))) { + if (btree_id_is_alloc(i)) { + c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_allocations); + c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_info); + c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_lrus); + c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_extents_to_backpointers); + c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_to_lru_refs); c->sb.compat &= ~(1ULL << BCH_COMPAT_alloc_info); - } + r->error = 0; + } else if (!(c->recovery_passes_explicit & BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes))) { + bch_info(c, "will run btree node scan"); + c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes); + c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); + } - ret = bch2_btree_root_read(c, i, &r->key, r->level); - if (ret) { - fsck_err(c, - btree_root_read_error, - "error reading btree root %s", - bch2_btree_id_str(i)); - if (btree_id_is_alloc(i)) - c->sb.compat &= ~(1ULL << BCH_COMPAT_alloc_info); ret = 0; } } - for (i = 0; i < BTREE_ID_NR; i++) { + for (unsigned i = 0; i < BTREE_ID_NR; i++) { struct btree_root *r = bch2_btree_id_root(c, i); - if (!r->b) { + if (!r->b && !r->error) { r->alive = false; r->level = 0; bch2_btree_root_alloc_fake(c, i, 0); @@ -653,7 +656,7 @@ int bch2_fs_recovery(struct bch_fs *c) goto err; } - if (!c->sb.clean || c->opts.fsck || c->opts.keep_journal) { + if (!c->sb.clean || c->opts.fsck || c->opts.retain_recovery_info) { struct genradix_iter iter; struct journal_replay **i; @@ -883,9 +886,10 @@ use_clean: out: bch2_flush_fsck_errs(c); - if (!c->opts.keep_journal && - test_bit(JOURNAL_REPLAY_DONE, &c->journal.flags)) + if (!c->opts.retain_recovery_info) { bch2_journal_keys_put_initial(c); + bch2_find_btree_nodes_exit(&c->found_btree_nodes); + } kfree(clean); if (!ret && diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index b066c5155b74..cb501460d615 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -4,6 +4,7 @@ #include "alloc_background.h" #include "backpointers.h" #include "btree_gc.h" +#include "btree_node_scan.h" #include "ec.h" #include "fsck.h" #include "inode.h" diff --git a/fs/bcachefs/recovery_passes_types.h b/fs/bcachefs/recovery_passes_types.h index f30521285706..840542cfd65b 100644 --- a/fs/bcachefs/recovery_passes_types.h +++ b/fs/bcachefs/recovery_passes_types.h @@ -13,6 +13,7 @@ * must never change: */ #define BCH_RECOVERY_PASSES() \ + x(scan_for_btree_nodes, 37, 0) \ x(check_topology, 4, 0) \ x(alloc_read, 0, PASS_ALWAYS) \ x(stripes_read, 1, PASS_ALWAYS) \ diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index bc026a77eb99..ed63018f21be 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -15,6 +15,7 @@ #include "btree_gc.h" #include "btree_journal_iter.h" #include "btree_key_cache.h" +#include "btree_node_scan.h" #include "btree_update_interior.h" #include "btree_io.h" #include "btree_write_buffer.h" @@ -536,6 +537,7 @@ static void __bch2_fs_free(struct bch_fs *c) for (i = 0; i < BCH_TIME_STAT_NR; i++) bch2_time_stats_exit(&c->times[i]); + bch2_find_btree_nodes_exit(&c->found_btree_nodes); bch2_free_pending_node_rewrites(c); bch2_fs_sb_errors_exit(c); bch2_fs_counters_exit(c); @@ -560,6 +562,7 @@ static void __bch2_fs_free(struct bch_fs *c) bch2_io_clock_exit(&c->io_clock[READ]); bch2_fs_compress_exit(c); bch2_journal_keys_put_initial(c); + bch2_find_btree_nodes_exit(&c->found_btree_nodes); BUG_ON(atomic_read(&c->journal_keys.ref)); bch2_fs_btree_write_buffer_exit(c); percpu_free_rwsem(&c->mark_lock); -- cgit v1.2.3-59-g8ed1b From 43f5ea4646b2271a9a5af3729dfdf644d69b3282 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 16 Mar 2024 22:45:30 -0400 Subject: bcachefs: Topology repair now uses nodes found by scanning to fill holes With the new btree node scan code, we can now recover from corrupt btree roots - simply create a new fake root at depth 1, and then insert all the leaves we found. If the root wasn't corrupt but there's corruption elsewhere in the btree, we can fill in holes as needed with the newest version of a given node(s) from the scan; we also check if a given btree node is older than what we found from the scan. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 303 +++++++++++++++++++++++++++--------------- fs/bcachefs/sb-errors_types.h | 3 +- 2 files changed, 199 insertions(+), 107 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 9db9c8a5beaa..6280da1244b5 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -13,6 +13,7 @@ #include "btree_journal_iter.h" #include "btree_key_cache.h" #include "btree_locking.h" +#include "btree_node_scan.h" #include "btree_update_interior.h" #include "btree_io.h" #include "btree_gc.h" @@ -41,6 +42,7 @@ #define DROP_THIS_NODE 10 #define DROP_PREV_NODE 11 +#define DID_FILL_FROM_SCAN 12 static struct bkey_s unsafe_bkey_s_c_to_s(struct bkey_s_c k) { @@ -129,6 +131,17 @@ static int set_node_min(struct bch_fs *c, struct btree *b, struct bpos new_min) struct bkey_i_btree_ptr_v2 *new; int ret; + if (c->opts.verbose) { + struct printbuf buf = PRINTBUF; + + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + prt_str(&buf, " -> "); + bch2_bpos_to_text(&buf, new_min); + + bch_info(c, "%s(): %s", __func__, buf.buf); + printbuf_exit(&buf); + } + new = kmalloc_array(BKEY_BTREE_PTR_U64s_MAX, sizeof(u64), GFP_KERNEL); if (!new) return -BCH_ERR_ENOMEM_gc_repair_key; @@ -154,6 +167,17 @@ static int set_node_max(struct bch_fs *c, struct btree *b, struct bpos new_max) struct bkey_i_btree_ptr_v2 *new; int ret; + if (c->opts.verbose) { + struct printbuf buf = PRINTBUF; + + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + prt_str(&buf, " -> "); + bch2_bpos_to_text(&buf, new_max); + + bch_info(c, "%s(): %s", __func__, buf.buf); + printbuf_exit(&buf); + } + ret = bch2_journal_key_delete(c, b->c.btree_id, b->c.level + 1, b->key.k.p); if (ret) return ret; @@ -185,127 +209,138 @@ static int set_node_max(struct bch_fs *c, struct btree *b, struct bpos new_max) return 0; } -static int btree_repair_node_boundaries(struct bch_fs *c, struct btree *b, - struct btree *prev, struct btree *cur) +static int btree_check_node_boundaries(struct bch_fs *c, struct btree *b, + struct btree *prev, struct btree *cur, + struct bpos *pulled_from_scan) { struct bpos expected_start = !prev ? b->data->min_key : bpos_successor(prev->key.k.p); - struct printbuf buf1 = PRINTBUF, buf2 = PRINTBUF; + struct printbuf buf = PRINTBUF; int ret = 0; - if (!prev) { - prt_printf(&buf1, "start of node: "); - bch2_bpos_to_text(&buf1, b->data->min_key); - } else { - bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(&prev->key)); + BUG_ON(b->key.k.type == KEY_TYPE_btree_ptr_v2 && + !bpos_eq(bkey_i_to_btree_ptr_v2(&b->key)->v.min_key, + b->data->min_key)); + + if (bpos_eq(expected_start, cur->data->min_key)) + return 0; + + prt_printf(&buf, " at btree %s level %u:\n parent: ", + bch2_btree_id_str(b->c.btree_id), b->c.level); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + + if (prev) { + prt_printf(&buf, "\n prev: "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&prev->key)); } - bch2_bkey_val_to_text(&buf2, c, bkey_i_to_s_c(&cur->key)); - - if (prev && - bpos_gt(expected_start, cur->data->min_key) && - BTREE_NODE_SEQ(cur->data) > BTREE_NODE_SEQ(prev->data)) { - /* cur overwrites prev: */ - - if (mustfix_fsck_err_on(bpos_ge(prev->data->min_key, - cur->data->min_key), c, - btree_node_topology_overwritten_by_next_node, - "btree node overwritten by next node at btree %s level %u:\n" - " node %s\n" - " next %s", - bch2_btree_id_str(b->c.btree_id), b->c.level, - buf1.buf, buf2.buf)) { - ret = DROP_PREV_NODE; - goto out; - } + prt_str(&buf, "\n next: "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&cur->key)); - if (mustfix_fsck_err_on(!bpos_eq(prev->key.k.p, - bpos_predecessor(cur->data->min_key)), c, - btree_node_topology_bad_max_key, - "btree node with incorrect max_key at btree %s level %u:\n" - " node %s\n" - " next %s", - bch2_btree_id_str(b->c.btree_id), b->c.level, - buf1.buf, buf2.buf)) - ret = set_node_max(c, prev, - bpos_predecessor(cur->data->min_key)); - } else { - /* prev overwrites cur: */ - - if (mustfix_fsck_err_on(bpos_ge(expected_start, - cur->data->max_key), c, - btree_node_topology_overwritten_by_prev_node, - "btree node overwritten by prev node at btree %s level %u:\n" - " prev %s\n" - " node %s", - bch2_btree_id_str(b->c.btree_id), b->c.level, - buf1.buf, buf2.buf)) { - ret = DROP_THIS_NODE; - goto out; - } + if (bpos_lt(expected_start, cur->data->min_key)) { /* gap */ + if (b->c.level == 1 && + bpos_lt(*pulled_from_scan, cur->data->min_key)) { + ret = bch2_get_scanned_nodes(c, b->c.btree_id, 0, + expected_start, + bpos_predecessor(cur->data->min_key)); + if (ret) + goto err; - if (mustfix_fsck_err_on(!bpos_eq(expected_start, cur->data->min_key), c, - btree_node_topology_bad_min_key, - "btree node with incorrect min_key at btree %s level %u:\n" - " prev %s\n" - " node %s", - bch2_btree_id_str(b->c.btree_id), b->c.level, - buf1.buf, buf2.buf)) - ret = set_node_min(c, cur, expected_start); + *pulled_from_scan = cur->data->min_key; + ret = DID_FILL_FROM_SCAN; + } else { + if (mustfix_fsck_err(c, btree_node_topology_bad_min_key, + "btree node with incorrect min_key%s", buf.buf)) + ret = set_node_min(c, cur, expected_start); + } + } else { /* overlap */ + if (prev && BTREE_NODE_SEQ(cur->data) > BTREE_NODE_SEQ(prev->data)) { /* cur overwrites prev */ + if (bpos_ge(prev->data->min_key, cur->data->min_key)) { /* fully? */ + if (mustfix_fsck_err(c, btree_node_topology_overwritten_by_next_node, + "btree node overwritten by next node%s", buf.buf)) + ret = DROP_PREV_NODE; + } else { + if (mustfix_fsck_err(c, btree_node_topology_bad_max_key, + "btree node with incorrect max_key%s", buf.buf)) + ret = set_node_max(c, prev, + bpos_predecessor(cur->data->min_key)); + } + } else { + if (bpos_ge(expected_start, cur->data->max_key)) { /* fully? */ + if (mustfix_fsck_err(c, btree_node_topology_overwritten_by_prev_node, + "btree node overwritten by prev node%s", buf.buf)) + ret = DROP_THIS_NODE; + } else { + if (mustfix_fsck_err(c, btree_node_topology_bad_min_key, + "btree node with incorrect min_key%s", buf.buf)) + ret = set_node_min(c, cur, expected_start); + } + } } -out: +err: fsck_err: - printbuf_exit(&buf2); - printbuf_exit(&buf1); + printbuf_exit(&buf); return ret; } static int btree_repair_node_end(struct bch_fs *c, struct btree *b, - struct btree *child) + struct btree *child, struct bpos *pulled_from_scan) { - struct printbuf buf1 = PRINTBUF, buf2 = PRINTBUF; + struct printbuf buf = PRINTBUF; int ret = 0; - bch2_bkey_val_to_text(&buf1, c, bkey_i_to_s_c(&child->key)); - bch2_bpos_to_text(&buf2, b->key.k.p); + if (bpos_eq(child->key.k.p, b->key.k.p)) + return 0; - if (mustfix_fsck_err_on(!bpos_eq(child->key.k.p, b->key.k.p), c, - btree_node_topology_bad_max_key, - "btree node with incorrect max_key at btree %s level %u:\n" - " %s\n" - " expected %s", - bch2_btree_id_str(b->c.btree_id), b->c.level, - buf1.buf, buf2.buf)) { - ret = set_node_max(c, child, b->key.k.p); - if (ret) - goto err; + prt_printf(&buf, "at btree %s level %u:\n parent: ", + bch2_btree_id_str(b->c.btree_id), b->c.level); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + + prt_str(&buf, "\n child: "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&child->key)); + + if (mustfix_fsck_err(c, btree_node_topology_bad_max_key, + "btree node with incorrect max_key%s", buf.buf)) { + if (b->c.level == 1 && + bpos_lt(*pulled_from_scan, b->key.k.p)) { + ret = bch2_get_scanned_nodes(c, b->c.btree_id, 0, + bpos_successor(child->key.k.p), b->key.k.p); + if (ret) + goto err; + + *pulled_from_scan = b->key.k.p; + ret = DID_FILL_FROM_SCAN; + } else { + ret = set_node_max(c, child, b->key.k.p); + } } err: fsck_err: - printbuf_exit(&buf2); - printbuf_exit(&buf1); + printbuf_exit(&buf); return ret; } -static int bch2_btree_repair_topology_recurse(struct btree_trans *trans, struct btree *b) +static int bch2_btree_repair_topology_recurse(struct btree_trans *trans, struct btree *b, + struct bpos *pulled_from_scan) { struct bch_fs *c = trans->c; struct btree_and_journal_iter iter; struct bkey_s_c k; struct bkey_buf prev_k, cur_k; struct btree *prev = NULL, *cur = NULL; - bool have_child, dropped_children = false; + bool have_child, new_pass = false; struct printbuf buf = PRINTBUF; int ret = 0; if (!b->c.level) return 0; -again: - prev = NULL; - have_child = dropped_children = false; + bch2_bkey_buf_init(&prev_k); bch2_bkey_buf_init(&cur_k); +again: + cur = prev = NULL; + have_child = new_pass = false; bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); iter.prefetch = true; @@ -332,9 +367,10 @@ again: b->c.level - 1, buf.buf)) { bch2_btree_node_evict(trans, cur_k.k); - ret = bch2_journal_key_delete(c, b->c.btree_id, - b->c.level, cur_k.k->k.p); cur = NULL; + ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes) ?: + bch2_journal_key_delete(c, b->c.btree_id, + b->c.level, cur_k.k->k.p); if (ret) break; continue; @@ -344,7 +380,23 @@ again: if (ret) break; - ret = btree_repair_node_boundaries(c, b, prev, cur); + if (bch2_btree_node_is_stale(c, cur)) { + bch_info(c, "btree node %s older than nodes found by scanning", buf.buf); + six_unlock_read(&cur->c.lock); + bch2_btree_node_evict(trans, cur_k.k); + ret = bch2_journal_key_delete(c, b->c.btree_id, + b->c.level, cur_k.k->k.p); + cur = NULL; + if (ret) + break; + continue; + } + + ret = btree_check_node_boundaries(c, b, prev, cur, pulled_from_scan); + if (ret == DID_FILL_FROM_SCAN) { + new_pass = true; + ret = 0; + } if (ret == DROP_THIS_NODE) { six_unlock_read(&cur->c.lock); @@ -370,8 +422,6 @@ again: break; bch2_btree_and_journal_iter_exit(&iter); - bch2_bkey_buf_exit(&prev_k, c); - bch2_bkey_buf_exit(&cur_k, c); goto again; } else if (ret) break; @@ -383,7 +433,11 @@ again: if (!ret && !IS_ERR_OR_NULL(prev)) { BUG_ON(cur); - ret = btree_repair_node_end(c, b, prev); + ret = btree_repair_node_end(c, b, prev, pulled_from_scan); + if (ret == DID_FILL_FROM_SCAN) { + new_pass = true; + ret = 0; + } } if (!IS_ERR_OR_NULL(prev)) @@ -397,6 +451,10 @@ again: goto err; bch2_btree_and_journal_iter_exit(&iter); + + if (new_pass) + goto again; + bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); iter.prefetch = true; @@ -413,7 +471,7 @@ again: if (ret) goto err; - ret = bch2_btree_repair_topology_recurse(trans, cur); + ret = bch2_btree_repair_topology_recurse(trans, cur, pulled_from_scan); six_unlock_read(&cur->c.lock); cur = NULL; @@ -421,7 +479,7 @@ again: bch2_btree_node_evict(trans, cur_k.k); ret = bch2_journal_key_delete(c, b->c.btree_id, b->c.level, cur_k.k->k.p); - dropped_children = true; + new_pass = true; } if (ret) @@ -448,12 +506,14 @@ fsck_err: six_unlock_read(&cur->c.lock); bch2_btree_and_journal_iter_exit(&iter); - bch2_bkey_buf_exit(&prev_k, c); - bch2_bkey_buf_exit(&cur_k, c); - if (!ret && dropped_children) + if (!ret && new_pass) goto again; + BUG_ON(!ret && bch2_btree_node_check_topology(trans, b)); + + bch2_bkey_buf_exit(&prev_k, c); + bch2_bkey_buf_exit(&cur_k, c); printbuf_exit(&buf); return ret; } @@ -461,32 +521,63 @@ fsck_err: int bch2_check_topology(struct bch_fs *c) { struct btree_trans *trans = bch2_trans_get(c); - struct btree *b; - unsigned i; + struct bpos pulled_from_scan = POS_MIN; int ret = 0; - for (i = 0; i < btree_id_nr_alive(c) && !ret; i++) { + for (unsigned i = 0; i < btree_id_nr_alive(c) && !ret; i++) { struct btree_root *r = bch2_btree_id_root(c, i); + bool reconstructed_root = false; - if (!r->alive) - continue; + if (r->error) { + ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes); + if (ret) + break; +reconstruct_root: + bch_info(c, "btree root %s unreadable, must recover from scan", bch2_btree_id_str(i)); - b = r->b; - if (btree_node_fake(b)) - continue; + r->alive = false; + r->error = 0; + + if (!bch2_btree_has_scanned_nodes(c, i)) { + mustfix_fsck_err(c, btree_root_unreadable_and_scan_found_nothing, + "no nodes found for btree %s, continue?", bch2_btree_id_str(i)); + bch2_btree_root_alloc_fake(c, i, 0); + } else { + bch2_btree_root_alloc_fake(c, i, 1); + ret = bch2_get_scanned_nodes(c, i, 0, POS_MIN, SPOS_MAX); + if (ret) + break; + } + + bch2_shoot_down_journal_keys(c, i, 1, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); + reconstructed_root = true; + } + + struct btree *b = r->b; btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_read); - ret = bch2_btree_repair_topology_recurse(trans, b); + ret = bch2_btree_repair_topology_recurse(trans, b, &pulled_from_scan); six_unlock_read(&b->c.lock); if (ret == DROP_THIS_NODE) { - bch_err(c, "empty btree root - repair unimplemented"); - ret = -BCH_ERR_fsck_repair_unimplemented; + bch2_btree_node_hash_remove(&c->btree_cache, b); + mutex_lock(&c->btree_cache.lock); + list_move(&b->list, &c->btree_cache.freeable); + mutex_unlock(&c->btree_cache.lock); + + r->b = NULL; + + if (!reconstructed_root) + goto reconstruct_root; + + bch_err(c, "empty btree root %s", bch2_btree_id_str(i)); + bch2_btree_root_alloc_fake(c, i, 0); + r->alive = false; + ret = 0; } } - +fsck_err: bch2_trans_put(trans); - return ret; } diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 73e9634df8ff..2fec03a24c95 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -267,7 +267,8 @@ x(subvol_unreachable, 259) \ x(btree_node_bkey_bad_u64s, 260) \ x(btree_node_topology_empty_interior_node, 261) \ - x(btree_ptr_v2_min_key_bad, 262) + x(btree_ptr_v2_min_key_bad, 262) \ + x(btree_root_unreadable_and_scan_found_nothing, 263) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, -- cgit v1.2.3-59-g8ed1b From 55936afe11077a84d9e1c5068169af328bbf2811 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 15 Mar 2024 23:03:42 -0400 Subject: bcachefs: Flag btrees with missing data We need this to know when we should attempt to reconstruct the snapshots btree Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 1 + fs/bcachefs/bcachefs_format.h | 1 + fs/bcachefs/btree_io.c | 13 +++++++++---- fs/bcachefs/recovery.c | 23 +++++++++++++++++++++++ fs/bcachefs/recovery.h | 2 ++ fs/bcachefs/super-io.c | 9 ++++++++- 6 files changed, 44 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 93a61dbaa3d8..d1a0e5478503 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -797,6 +797,7 @@ struct bch_fs { u64 features; u64 compat; unsigned long errors_silent[BITS_TO_LONGS(BCH_SB_ERR_MAX)]; + u64 btrees_lost_data; } sb; diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index bff8750ac0d7..63102992d955 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -818,6 +818,7 @@ struct bch_sb_field_ext { struct bch_sb_field field; __le64 recovery_passes_required[2]; __le64 errors_silent[8]; + __le64 btrees_lost_data; }; struct bch_sb_field_downgrade_entry { diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index f3f27bb85a5b..d7de82ac3893 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1264,10 +1264,12 @@ out: return retry_read; fsck_err: if (ret == -BCH_ERR_btree_node_read_err_want_retry || - ret == -BCH_ERR_btree_node_read_err_must_retry) + ret == -BCH_ERR_btree_node_read_err_must_retry) { retry_read = 1; - else + } else { set_btree_node_read_error(b); + bch2_btree_lost_data(c, b->c.btree_id); + } goto out; } @@ -1328,6 +1330,7 @@ start: if (!can_retry) { set_btree_node_read_error(b); + bch2_btree_lost_data(c, b->c.btree_id); break; } } @@ -1527,9 +1530,10 @@ fsck_err: ret = -1; } - if (ret) + if (ret) { set_btree_node_read_error(b); - else if (*saw_error) + bch2_btree_lost_data(c, b->c.btree_id); + } else if (*saw_error) bch2_btree_node_rewrite_async(c, b); for (i = 0; i < ra->nr; i++) { @@ -1665,6 +1669,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, bch2_fatal_error(c); set_btree_node_read_error(b); + bch2_btree_lost_data(c, b->c.btree_id); clear_btree_node_read_in_flight(b); wake_up_bit(&b->flags, BTREE_NODE_read_in_flight); printbuf_exit(&buf); diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 24671020f22b..b3c67ae3d3b2 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -33,6 +33,20 @@ #define QSTR(n) { { { .len = strlen(n) } }, .name = n } +void bch2_btree_lost_data(struct bch_fs *c, enum btree_id btree) +{ + u64 b = BIT_ULL(btree); + + if (!(c->sb.btrees_lost_data & b)) { + bch_err(c, "flagging btree %s lost data", bch2_btree_id_str(btree)); + + mutex_lock(&c->sb_lock); + bch2_sb_field_get(c->disk_sb.sb, ext)->btrees_lost_data |= cpu_to_le64(b); + bch2_write_super(c); + mutex_unlock(&c->sb_lock); + } +} + static bool btree_id_is_alloc(enum btree_id id) { switch (id) { @@ -470,6 +484,7 @@ static int read_btree_roots(struct bch_fs *c) } ret = 0; + bch2_btree_lost_data(c, i); } } @@ -848,6 +863,14 @@ use_clean: write_sb = true; } + if (c->opts.fsck && + !test_bit(BCH_FS_error, &c->flags) && + c->recovery_pass_done == BCH_RECOVERY_PASS_NR - 1 && + ext->btrees_lost_data) { + ext->btrees_lost_data = 0; + write_sb = true; + } + if (c->opts.fsck && !test_bit(BCH_FS_error, &c->flags) && !test_bit(BCH_FS_errors_not_fixed, &c->flags)) { diff --git a/fs/bcachefs/recovery.h b/fs/bcachefs/recovery.h index 3962fd87b50d..4bf818de1f2f 100644 --- a/fs/bcachefs/recovery.h +++ b/fs/bcachefs/recovery.h @@ -2,6 +2,8 @@ #ifndef _BCACHEFS_RECOVERY_H #define _BCACHEFS_RECOVERY_H +void bch2_btree_lost_data(struct bch_fs *, enum btree_id); + int bch2_journal_replay(struct bch_fs *); int bch2_fs_recovery(struct bch_fs *); diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index e15f8b1f30c2..e0aa3655b63b 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -527,9 +527,11 @@ static void bch2_sb_update(struct bch_fs *c) memset(c->sb.errors_silent, 0, sizeof(c->sb.errors_silent)); struct bch_sb_field_ext *ext = bch2_sb_field_get(src, ext); - if (ext) + if (ext) { le_bitvector_to_cpu(c->sb.errors_silent, (void *) ext->errors_silent, sizeof(c->sb.errors_silent) * 8); + c->sb.btrees_lost_data = le64_to_cpu(ext->btrees_lost_data); + } for_each_member_device(c, ca) { struct bch_member m = bch2_sb_member_get(src, ca->dev_idx); @@ -1162,6 +1164,11 @@ static void bch2_sb_ext_to_text(struct printbuf *out, struct bch_sb *sb, kfree(errors_silent); } + + prt_printf(out, "Btrees with missing data:"); + prt_tab(out); + prt_bitflags(out, __bch2_btree_ids, le64_to_cpu(e->btrees_lost_data)); + prt_newline(out); } static const struct bch_sb_field_ops bch_sb_field_ops_ext = { -- cgit v1.2.3-59-g8ed1b From a292be3b68f3fdad6cff50c716174f51b119efd1 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 27 Mar 2024 22:50:19 -0400 Subject: bcachefs: Reconstruct missing snapshot nodes When the snapshots btree is going, we'll have to delete huge amounts of data - unless we can reconstruct it by looking at the keys that refer to it. Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 1 + fs/bcachefs/recovery.c | 1 + fs/bcachefs/recovery_passes_types.h | 1 + fs/bcachefs/sb-errors_types.h | 3 +- fs/bcachefs/snapshot.c | 173 +++++++++++++++++++++++++++++++++++- fs/bcachefs/snapshot.h | 26 +++++- 6 files changed, 199 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index d1a0e5478503..a31a5f706929 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -615,6 +615,7 @@ struct bch_dev { */ #define BCH_FS_FLAGS() \ + x(new_fs) \ x(started) \ x(may_go_rw) \ x(rw) \ diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index b3c67ae3d3b2..b76c16152579 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -938,6 +938,7 @@ int bch2_fs_initialize(struct bch_fs *c) int ret; bch_notice(c, "initializing new filesystem"); + set_bit(BCH_FS_new_fs, &c->flags); mutex_lock(&c->sb_lock); c->disk_sb.sb->compat[0] |= cpu_to_le64(1ULL << BCH_COMPAT_extents_above_btree_updates_done); diff --git a/fs/bcachefs/recovery_passes_types.h b/fs/bcachefs/recovery_passes_types.h index 840542cfd65b..773aea9a0080 100644 --- a/fs/bcachefs/recovery_passes_types.h +++ b/fs/bcachefs/recovery_passes_types.h @@ -32,6 +32,7 @@ x(check_alloc_to_lru_refs, 15, PASS_ONLINE|PASS_FSCK) \ x(fs_freespace_init, 16, PASS_ALWAYS|PASS_SILENT) \ x(bucket_gens_init, 17, 0) \ + x(reconstruct_snapshots, 38, 0) \ x(check_snapshot_trees, 18, PASS_ONLINE|PASS_FSCK) \ x(check_snapshots, 19, PASS_ONLINE|PASS_FSCK) \ x(check_subvols, 20, PASS_ONLINE|PASS_FSCK) \ diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 2fec03a24c95..2f8f4d2388b0 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -268,7 +268,8 @@ x(btree_node_bkey_bad_u64s, 260) \ x(btree_node_topology_empty_interior_node, 261) \ x(btree_ptr_v2_min_key_bad, 262) \ - x(btree_root_unreadable_and_scan_found_nothing, 263) + x(btree_root_unreadable_and_scan_found_nothing, 263) \ + x(snapshot_node_missing, 264) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 4577ee7939a2..0e806f04f3d7 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -8,6 +8,7 @@ #include "errcode.h" #include "error.h" #include "fs.h" +#include "recovery_passes.h" #include "snapshot.h" #include @@ -574,6 +575,13 @@ static int check_snapshot_tree(struct btree_trans *trans, u32 subvol_id; ret = bch2_snapshot_tree_master_subvol(trans, root_id, &subvol_id); + bch_err_fn(c, ret); + + if (bch2_err_matches(ret, ENOENT)) { /* nothing to be done here */ + ret = 0; + goto err; + } + if (ret) goto err; @@ -731,7 +739,6 @@ static int check_snapshot(struct btree_trans *trans, u32 parent_id = bch2_snapshot_parent_early(c, k.k->p.offset); u32 real_depth; struct printbuf buf = PRINTBUF; - bool should_have_subvol; u32 i, id; int ret = 0; @@ -777,7 +784,7 @@ static int check_snapshot(struct btree_trans *trans, } } - should_have_subvol = BCH_SNAPSHOT_SUBVOL(&s) && + bool should_have_subvol = BCH_SNAPSHOT_SUBVOL(&s) && !BCH_SNAPSHOT_DELETED(&s); if (should_have_subvol) { @@ -879,6 +886,154 @@ int bch2_check_snapshots(struct bch_fs *c) return ret; } +static int check_snapshot_exists(struct btree_trans *trans, u32 id) +{ + struct bch_fs *c = trans->c; + + if (bch2_snapshot_equiv(c, id)) + return 0; + + u32 tree_id; + int ret = bch2_snapshot_tree_create(trans, id, 0, &tree_id); + if (ret) + return ret; + + struct bkey_i_snapshot *snapshot = bch2_trans_kmalloc(trans, sizeof(*snapshot)); + ret = PTR_ERR_OR_ZERO(snapshot); + if (ret) + return ret; + + bkey_snapshot_init(&snapshot->k_i); + snapshot->k.p = POS(0, id); + snapshot->v.tree = cpu_to_le32(tree_id); + snapshot->v.btime.lo = cpu_to_le64(bch2_current_time(c)); + + return bch2_btree_insert_trans(trans, BTREE_ID_snapshots, &snapshot->k_i, 0) ?: + bch2_mark_snapshot(trans, BTREE_ID_snapshots, 0, + bkey_s_c_null, bkey_i_to_s(&snapshot->k_i), 0) ?: + bch2_snapshot_set_equiv(trans, bkey_i_to_s_c(&snapshot->k_i)); +} + +/* Figure out which snapshot nodes belong in the same tree: */ +struct snapshot_tree_reconstruct { + enum btree_id btree; + struct bpos cur_pos; + snapshot_id_list cur_ids; + DARRAY(snapshot_id_list) trees; +}; + +static void snapshot_tree_reconstruct_exit(struct snapshot_tree_reconstruct *r) +{ + darray_for_each(r->trees, i) + darray_exit(i); + darray_exit(&r->trees); + darray_exit(&r->cur_ids); +} + +static inline bool same_snapshot(struct snapshot_tree_reconstruct *r, struct bpos pos) +{ + return r->btree == BTREE_ID_inodes + ? r->cur_pos.offset == pos.offset + : r->cur_pos.inode == pos.inode; +} + +static inline bool snapshot_id_lists_have_common(snapshot_id_list *l, snapshot_id_list *r) +{ + darray_for_each(*l, i) + if (snapshot_list_has_id(r, *i)) + return true; + return false; +} + +static void snapshot_id_list_to_text(struct printbuf *out, snapshot_id_list *s) +{ + bool first = true; + darray_for_each(*s, i) { + if (!first) + prt_char(out, ' '); + first = false; + prt_printf(out, "%u", *i); + } +} + +static int snapshot_tree_reconstruct_next(struct bch_fs *c, struct snapshot_tree_reconstruct *r) +{ + if (r->cur_ids.nr) { + darray_for_each(r->trees, i) + if (snapshot_id_lists_have_common(i, &r->cur_ids)) { + int ret = snapshot_list_merge(c, i, &r->cur_ids); + if (ret) + return ret; + goto out; + } + darray_push(&r->trees, r->cur_ids); + darray_init(&r->cur_ids); + } +out: + r->cur_ids.nr = 0; + return 0; +} + +static int get_snapshot_trees(struct bch_fs *c, struct snapshot_tree_reconstruct *r, struct bpos pos) +{ + if (!same_snapshot(r, pos)) + snapshot_tree_reconstruct_next(c, r); + r->cur_pos = pos; + return snapshot_list_add_nodup(c, &r->cur_ids, pos.snapshot); +} + +int bch2_reconstruct_snapshots(struct bch_fs *c) +{ + struct btree_trans *trans = bch2_trans_get(c); + struct printbuf buf = PRINTBUF; + struct snapshot_tree_reconstruct r = {}; + int ret = 0; + + for (unsigned btree = 0; btree < BTREE_ID_NR; btree++) { + if (btree_type_has_snapshots(btree)) { + r.btree = btree; + + ret = for_each_btree_key(trans, iter, btree, POS_MIN, + BTREE_ITER_ALL_SNAPSHOTS|BTREE_ITER_PREFETCH, k, ({ + get_snapshot_trees(c, &r, k.k->p); + })); + if (ret) + goto err; + + snapshot_tree_reconstruct_next(c, &r); + } + } + + darray_for_each(r.trees, t) { + printbuf_reset(&buf); + snapshot_id_list_to_text(&buf, t); + + darray_for_each(*t, id) { + if (fsck_err_on(!bch2_snapshot_equiv(c, *id), + c, snapshot_node_missing, + "snapshot node %u from tree %s missing", *id, buf.buf)) { + if (t->nr > 1) { + bch_err(c, "cannot reconstruct snapshot trees with multiple nodes"); + ret = -BCH_ERR_fsck_repair_unimplemented; + goto err; + } + + ret = commit_do(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, + check_snapshot_exists(trans, *id)); + if (ret) + goto err; + } + } + } +fsck_err: +err: + bch2_trans_put(trans); + snapshot_tree_reconstruct_exit(&r); + printbuf_exit(&buf); + bch_err_fn(c, ret); + return ret; +} + /* * Mark a snapshot as deleted, for future cleanup: */ @@ -1689,6 +1844,20 @@ int bch2_snapshots_read(struct bch_fs *c) POS_MIN, 0, k, (set_is_ancestor_bitmap(c, k.k->p.offset), 0))); bch_err_fn(c, ret); + + /* + * It's important that we check if we need to reconstruct snapshots + * before going RW, so we mark that pass as required in the superblock - + * otherwise, we could end up deleting keys with missing snapshot nodes + * instead + */ + BUG_ON(!test_bit(BCH_FS_new_fs, &c->flags) && + test_bit(BCH_FS_may_go_rw, &c->flags)); + + if (bch2_err_matches(ret, EIO) || + (c->sb.btrees_lost_data & BIT_ULL(BTREE_ID_snapshots))) + ret = bch2_run_explicit_recovery_pass_persistent(c, BCH_RECOVERY_PASS_reconstruct_snapshots); + return ret; } diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index 331f20fd8d03..b7d2fed37c4f 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -209,15 +209,34 @@ static inline bool snapshot_list_has_ancestor(struct bch_fs *c, snapshot_id_list static inline int snapshot_list_add(struct bch_fs *c, snapshot_id_list *s, u32 id) { - int ret; - BUG_ON(snapshot_list_has_id(s, id)); - ret = darray_push(s, id); + int ret = darray_push(s, id); if (ret) bch_err(c, "error reallocating snapshot_id_list (size %zu)", s->size); return ret; } +static inline int snapshot_list_add_nodup(struct bch_fs *c, snapshot_id_list *s, u32 id) +{ + int ret = snapshot_list_has_id(s, id) + ? 0 + : darray_push(s, id); + if (ret) + bch_err(c, "error reallocating snapshot_id_list (size %zu)", s->size); + return ret; +} + +static inline int snapshot_list_merge(struct bch_fs *c, snapshot_id_list *dst, snapshot_id_list *src) +{ + darray_for_each(*src, i) { + int ret = snapshot_list_add_nodup(c, dst, *i); + if (ret) + return ret; + } + + return 0; +} + int bch2_snapshot_lookup(struct btree_trans *trans, u32 id, struct bch_snapshot *s); int bch2_snapshot_get_subvol(struct btree_trans *, u32, @@ -229,6 +248,7 @@ int bch2_snapshot_node_create(struct btree_trans *, u32, int bch2_check_snapshot_trees(struct bch_fs *); int bch2_check_snapshots(struct bch_fs *); +int bch2_reconstruct_snapshots(struct bch_fs *); int bch2_snapshot_node_set_deleted(struct btree_trans *, u32); void bch2_delete_dead_snapshots_work(struct work_struct *); -- cgit v1.2.3-59-g8ed1b From 4c02e63dadc7f5a92732ce3e267d0511749f60fb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 30 Mar 2024 18:43:00 -0400 Subject: bcachefs: Check for extents that point to same space In backpointer repair, if we get a missing backpointer - but there's already a backpointer that points to an existing extent - we've got multiple extents that point to the same space and need to decide which to keep. Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 173 ++++++++++++++++++++++++++++++++++++++++-- fs/bcachefs/sb-errors_types.h | 3 +- 2 files changed, 168 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 762c8ddfc5e7..114328acde72 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -8,6 +8,7 @@ #include "btree_update.h" #include "btree_update_interior.h" #include "btree_write_buffer.h" +#include "checksum.h" #include "error.h" #include @@ -418,6 +419,84 @@ struct extents_to_bp_state { struct bkey_buf last_flushed; }; +static int drop_dev_and_update(struct btree_trans *trans, enum btree_id btree, + struct bkey_s_c extent, unsigned dev) +{ + struct bkey_i *n = bch2_bkey_make_mut_noupdate(trans, extent); + int ret = PTR_ERR_OR_ZERO(n); + if (ret) + return ret; + + bch2_bkey_drop_device(bkey_i_to_s(n), dev); + return bch2_btree_insert_trans(trans, btree, n, 0); +} + +static int check_extent_checksum(struct btree_trans *trans, + enum btree_id btree, struct bkey_s_c extent, + enum btree_id o_btree, struct bkey_s_c extent2, unsigned dev) +{ + struct bch_fs *c = trans->c; + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(extent); + const union bch_extent_entry *entry; + struct extent_ptr_decoded p; + struct printbuf buf = PRINTBUF; + void *data_buf = NULL; + struct bio *bio = NULL; + size_t bytes; + int ret = 0; + + if (bkey_is_btree_ptr(extent.k)) + return false; + + bkey_for_each_ptr_decode(extent.k, ptrs, p, entry) + if (p.ptr.dev == dev) + goto found; + BUG(); +found: + if (!p.crc.csum_type) + return false; + + bytes = p.crc.compressed_size << 9; + + struct bch_dev *ca = bch_dev_bkey_exists(c, dev); + if (!bch2_dev_get_ioref(ca, READ)) + return false; + + data_buf = kvmalloc(bytes, GFP_KERNEL); + if (!data_buf) { + ret = -ENOMEM; + goto err; + } + + bio = bio_alloc(ca->disk_sb.bdev, 1, REQ_OP_READ, GFP_KERNEL); + bio->bi_iter.bi_sector = p.ptr.offset; + bch2_bio_map(bio, data_buf, bytes); + ret = submit_bio_wait(bio); + if (ret) + goto err; + + prt_str(&buf, "extents pointing to same space, but first extent checksum bad:"); + prt_printf(&buf, "\n %s ", bch2_btree_id_str(btree)); + bch2_bkey_val_to_text(&buf, c, extent); + prt_printf(&buf, "\n %s ", bch2_btree_id_str(o_btree)); + bch2_bkey_val_to_text(&buf, c, extent2); + + struct nonce nonce = extent_nonce(extent.k->version, p.crc); + struct bch_csum csum = bch2_checksum(c, p.crc.csum_type, nonce, data_buf, bytes); + if (fsck_err_on(bch2_crc_cmp(csum, p.crc.csum), + c, dup_backpointer_to_bad_csum_extent, + "%s", buf.buf)) + ret = drop_dev_and_update(trans, btree, extent, dev) ?: 1; +fsck_err: +err: + if (bio) + bio_put(bio); + kvfree(data_buf); + percpu_ref_put(&ca->io_ref); + printbuf_exit(&buf); + return ret; +} + static int check_bp_exists(struct btree_trans *trans, struct extents_to_bp_state *s, struct bpos bucket, @@ -425,7 +504,8 @@ static int check_bp_exists(struct btree_trans *trans, struct bkey_s_c orig_k) { struct bch_fs *c = trans->c; - struct btree_iter bp_iter = { NULL }; + struct btree_iter bp_iter = {}; + struct btree_iter other_extent_iter = {}; struct printbuf buf = PRINTBUF; struct bkey_s_c bp_k; struct bkey_buf tmp; @@ -433,13 +513,19 @@ static int check_bp_exists(struct btree_trans *trans, bch2_bkey_buf_init(&tmp); + if (!bch2_dev_bucket_exists(c, bucket)) { + prt_str(&buf, "extent for nonexistent device:bucket "); + bch2_bpos_to_text(&buf, bucket); + prt_str(&buf, "\n "); + bch2_bkey_val_to_text(&buf, c, orig_k); + bch_err(c, "%s", buf.buf); + return -BCH_ERR_fsck_repair_unimplemented; + } + if (bpos_lt(bucket, s->bucket_start) || bpos_gt(bucket, s->bucket_end)) return 0; - if (!bch2_dev_bucket_exists(c, bucket)) - goto missing; - bp_k = bch2_bkey_get_iter(trans, &bp_iter, BTREE_ID_backpointers, bucket_pos_to_bp(c, bucket, bp.bucket_offset), 0); @@ -465,21 +551,94 @@ static int check_bp_exists(struct btree_trans *trans, ret = -BCH_ERR_transaction_restart_write_buffer_flush; goto out; } - goto missing; + + goto check_existing_bp; } out: err: fsck_err: + bch2_trans_iter_exit(trans, &other_extent_iter); bch2_trans_iter_exit(trans, &bp_iter); bch2_bkey_buf_exit(&tmp, c); printbuf_exit(&buf); return ret; +check_existing_bp: + /* Do we have a backpointer for a different extent? */ + if (bp_k.k->type != KEY_TYPE_backpointer) + goto missing; + + struct bch_backpointer other_bp = *bkey_s_c_to_backpointer(bp_k).v; + + struct bkey_s_c other_extent = + bch2_backpointer_get_key(trans, &other_extent_iter, bp_k.k->p, other_bp, 0); + ret = bkey_err(other_extent); + if (ret == -BCH_ERR_backpointer_to_overwritten_btree_node) + ret = 0; + if (ret) + goto err; + + if (!other_extent.k) + goto missing; + + if (bch2_extents_match(orig_k, other_extent)) { + printbuf_reset(&buf); + prt_printf(&buf, "duplicate versions of same extent, deleting smaller\n "); + bch2_bkey_val_to_text(&buf, c, orig_k); + prt_str(&buf, "\n "); + bch2_bkey_val_to_text(&buf, c, other_extent); + bch_err(c, "%s", buf.buf); + + if (other_extent.k->size <= orig_k.k->size) { + ret = drop_dev_and_update(trans, other_bp.btree_id, other_extent, bucket.inode); + if (ret) + goto err; + goto out; + } else { + ret = drop_dev_and_update(trans, bp.btree_id, orig_k, bucket.inode); + if (ret) + goto err; + goto missing; + } + } + + ret = check_extent_checksum(trans, other_bp.btree_id, other_extent, bp.btree_id, orig_k, bucket.inode); + if (ret < 0) + goto err; + if (ret) { + ret = 0; + goto missing; + } + + ret = check_extent_checksum(trans, bp.btree_id, orig_k, other_bp.btree_id, other_extent, bucket.inode); + if (ret < 0) + goto err; + if (ret) { + ret = 0; + goto out; + } + + printbuf_reset(&buf); + prt_printf(&buf, "duplicate extents pointing to same space on dev %llu\n ", bucket.inode); + bch2_bkey_val_to_text(&buf, c, orig_k); + prt_str(&buf, "\n "); + bch2_bkey_val_to_text(&buf, c, other_extent); + bch_err(c, "%s", buf.buf); + ret = -BCH_ERR_fsck_repair_unimplemented; + goto err; missing: + printbuf_reset(&buf); prt_printf(&buf, "missing backpointer for btree=%s l=%u ", bch2_btree_id_str(bp.btree_id), bp.level); bch2_bkey_val_to_text(&buf, c, orig_k); - prt_printf(&buf, "\nbp pos "); - bch2_bpos_to_text(&buf, bp_iter.pos); + prt_printf(&buf, "\n got: "); + bch2_bkey_val_to_text(&buf, c, bp_k); + + struct bkey_i_backpointer n_bp_k; + bkey_backpointer_init(&n_bp_k.k_i); + n_bp_k.k.p = bucket_pos_to_bp(trans->c, bucket, bp.bucket_offset); + n_bp_k.v = bp; + prt_printf(&buf, "\n want: "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&n_bp_k.k_i)); if (fsck_err(c, ptr_to_missing_backpointer, "%s", buf.buf)) ret = bch2_bucket_backpointer_mod(trans, bucket, bp, orig_k, true); diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 2f8f4d2388b0..d7d609131030 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -269,7 +269,8 @@ x(btree_node_topology_empty_interior_node, 261) \ x(btree_ptr_v2_min_key_bad, 262) \ x(btree_root_unreadable_and_scan_found_nothing, 263) \ - x(snapshot_node_missing, 264) + x(snapshot_node_missing, 264) \ + x(dup_backpointer_to_bad_csum_extent, 265) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, -- cgit v1.2.3-59-g8ed1b From cc0532900bcf1896a81dcdd30873ffa6c4f6926b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 31 Mar 2024 02:03:03 -0400 Subject: bcachefs: Subvolume reconstruction We can now recreate missing subvolumes from dirents and/or inodes. Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 167 +++++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 148 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index b704fb0dda95..c1edc5647ba0 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -63,9 +63,7 @@ static int subvol_lookup(struct btree_trans *trans, u32 subvol, u32 *snapshot, u64 *inum) { struct bch_subvolume s; - int ret; - - ret = bch2_subvolume_get(trans, subvol, false, 0, &s); + int ret = bch2_subvolume_get(trans, subvol, false, 0, &s); *snapshot = le32_to_cpu(s.snapshot); *inum = le64_to_cpu(s.inode); @@ -170,7 +168,8 @@ err: /* Get lost+found, create if it doesn't exist: */ static int lookup_lostfound(struct btree_trans *trans, u32 snapshot, - struct bch_inode_unpacked *lostfound) + struct bch_inode_unpacked *lostfound, + u64 reattaching_inum) { struct bch_fs *c = trans->c; struct qstr lostfound_str = QSTR("lost+found"); @@ -185,19 +184,36 @@ static int lookup_lostfound(struct btree_trans *trans, u32 snapshot, return ret; subvol_inum root_inum = { .subvol = le32_to_cpu(st.master_subvol) }; - u32 subvol_snapshot; - ret = subvol_lookup(trans, le32_to_cpu(st.master_subvol), - &subvol_snapshot, &root_inum.inum); - bch_err_msg(c, ret, "looking up root subvol"); + struct bch_subvolume subvol; + ret = bch2_subvolume_get(trans, le32_to_cpu(st.master_subvol), + false, 0, &subvol); + bch_err_msg(c, ret, "looking up root subvol %u for snapshot %u", + le32_to_cpu(st.master_subvol), snapshot); if (ret) return ret; + if (!subvol.inode) { + struct btree_iter iter; + struct bkey_i_subvolume *subvol = bch2_bkey_get_mut_typed(trans, &iter, + BTREE_ID_subvolumes, POS(0, le32_to_cpu(st.master_subvol)), + 0, subvolume); + ret = PTR_ERR_OR_ZERO(subvol); + if (ret) + return ret; + + subvol->v.inode = cpu_to_le64(reattaching_inum); + bch2_trans_iter_exit(trans, &iter); + } + + root_inum.inum = le64_to_cpu(subvol.inode); + struct bch_inode_unpacked root_inode; struct bch_hash_info root_hash_info; u32 root_inode_snapshot = snapshot; ret = lookup_inode(trans, root_inum.inum, &root_inode, &root_inode_snapshot); - bch_err_msg(c, ret, "looking up root inode"); + bch_err_msg(c, ret, "looking up root inode %llu for subvol %u", + root_inum.inum, le32_to_cpu(st.master_subvol)); if (ret) return ret; @@ -293,7 +309,7 @@ static int reattach_inode(struct btree_trans *trans, snprintf(name_buf, sizeof(name_buf), "%llu", inode->bi_inum); } - ret = lookup_lostfound(trans, dirent_snapshot, &lostfound); + ret = lookup_lostfound(trans, dirent_snapshot, &lostfound, inode->bi_inum); if (ret) return ret; @@ -364,6 +380,85 @@ static int reattach_subvol(struct btree_trans *trans, struct bkey_s_c_subvolume return ret; } +static int reconstruct_subvol(struct btree_trans *trans, u32 snapshotid, u32 subvolid, u64 inum) +{ + struct bch_fs *c = trans->c; + + if (!bch2_snapshot_is_leaf(c, snapshotid)) { + bch_err(c, "need to reconstruct subvol, but have interior node snapshot"); + return -BCH_ERR_fsck_repair_unimplemented; + } + + /* + * If inum isn't set, that means we're being called from check_dirents, + * not check_inodes - the root of this subvolume doesn't exist or we + * would have found it there: + */ + if (!inum) { + struct btree_iter inode_iter = {}; + struct bch_inode_unpacked new_inode; + u64 cpu = raw_smp_processor_id(); + + bch2_inode_init_early(c, &new_inode); + bch2_inode_init_late(&new_inode, bch2_current_time(c), 0, 0, S_IFDIR|0755, 0, NULL); + + new_inode.bi_subvol = subvolid; + + int ret = bch2_inode_create(trans, &inode_iter, &new_inode, snapshotid, cpu) ?: + bch2_btree_iter_traverse(&inode_iter) ?: + bch2_inode_write(trans, &inode_iter, &new_inode); + bch2_trans_iter_exit(trans, &inode_iter); + if (ret) + return ret; + + inum = new_inode.bi_inum; + } + + bch_info(c, "reconstructing subvol %u with root inode %llu", subvolid, inum); + + struct bkey_i_subvolume *new_subvol = bch2_trans_kmalloc(trans, sizeof(*new_subvol)); + int ret = PTR_ERR_OR_ZERO(new_subvol); + if (ret) + return ret; + + bkey_subvolume_init(&new_subvol->k_i); + new_subvol->k.p.offset = subvolid; + new_subvol->v.snapshot = cpu_to_le32(snapshotid); + new_subvol->v.inode = cpu_to_le64(inum); + ret = bch2_btree_insert_trans(trans, BTREE_ID_subvolumes, &new_subvol->k_i, 0); + if (ret) + return ret; + + struct btree_iter iter; + struct bkey_i_snapshot *s = bch2_bkey_get_mut_typed(trans, &iter, + BTREE_ID_snapshots, POS(0, snapshotid), + 0, snapshot); + ret = PTR_ERR_OR_ZERO(s); + bch_err_msg(c, ret, "getting snapshot %u", snapshotid); + if (ret) + return ret; + + u32 snapshot_tree = le32_to_cpu(s->v.tree); + + s->v.subvol = cpu_to_le32(subvolid); + SET_BCH_SNAPSHOT_SUBVOL(&s->v, true); + bch2_trans_iter_exit(trans, &iter); + + struct bkey_i_snapshot_tree *st = bch2_bkey_get_mut_typed(trans, &iter, + BTREE_ID_snapshot_trees, POS(0, snapshot_tree), + 0, snapshot_tree); + ret = PTR_ERR_OR_ZERO(st); + bch_err_msg(c, ret, "getting snapshot tree %u", snapshot_tree); + if (ret) + return ret; + + if (!st->v.master_subvol) + st->v.master_subvol = cpu_to_le32(subvolid); + + bch2_trans_iter_exit(trans, &iter); + return 0; +} + struct snapshots_seen_entry { u32 id; u32 equiv; @@ -1065,6 +1160,11 @@ static int check_inode(struct btree_trans *trans, if (ret && !bch2_err_matches(ret, ENOENT)) goto err; + if (ret && (c->sb.btrees_lost_data & BIT_ULL(BTREE_ID_subvolumes))) { + ret = reconstruct_subvol(trans, k.k->p.snapshot, u.bi_subvol, u.bi_inum); + goto do_update; + } + if (fsck_err_on(ret, c, inode_bi_subvol_missing, "inode %llu:%u bi_subvol points to missing subvolume %u", @@ -1082,7 +1182,7 @@ static int check_inode(struct btree_trans *trans, do_update = true; } } - +do_update: if (do_update) { ret = __bch2_fsck_write_inode(trans, &u, iter->pos.snapshot); bch_err_msg(c, ret, "in fsck updating inode"); @@ -1785,6 +1885,7 @@ static int check_dirent_to_subvol(struct btree_trans *trans, struct btree_iter * u32 parent_subvol = le32_to_cpu(d.v->d_parent_subvol); u32 target_subvol = le32_to_cpu(d.v->d_child_subvol); u32 parent_snapshot; + u32 new_parent_subvol = 0; u64 parent_inum; struct printbuf buf = PRINTBUF; int ret = 0; @@ -1793,6 +1894,27 @@ static int check_dirent_to_subvol(struct btree_trans *trans, struct btree_iter * if (ret && !bch2_err_matches(ret, ENOENT)) return ret; + if (ret || + (!ret && !bch2_snapshot_is_ancestor(c, parent_snapshot, d.k->p.snapshot))) { + int ret2 = find_snapshot_subvol(trans, d.k->p.snapshot, &new_parent_subvol); + if (ret2 && !bch2_err_matches(ret, ENOENT)) + return ret2; + } + + if (ret && + !new_parent_subvol && + (c->sb.btrees_lost_data & BIT_ULL(BTREE_ID_subvolumes))) { + /* + * Couldn't find a subvol for dirent's snapshot - but we lost + * subvols, so we need to reconstruct: + */ + ret = reconstruct_subvol(trans, d.k->p.snapshot, parent_subvol, 0); + if (ret) + return ret; + + parent_snapshot = d.k->p.snapshot; + } + if (fsck_err_on(ret, c, dirent_to_missing_parent_subvol, "dirent parent_subvol points to missing subvolume\n%s", (bch2_bkey_val_to_text(&buf, c, d.s_c), buf.buf)) || @@ -1801,10 +1923,10 @@ static int check_dirent_to_subvol(struct btree_trans *trans, struct btree_iter * "dirent not visible in parent_subvol (not an ancestor of subvol snap %u)\n%s", parent_snapshot, (bch2_bkey_val_to_text(&buf, c, d.s_c), buf.buf))) { - u32 new_parent_subvol; - ret = find_snapshot_subvol(trans, d.k->p.snapshot, &new_parent_subvol); - if (ret) - goto err; + if (!new_parent_subvol) { + bch_err(c, "could not find a subvol for snapshot %u", d.k->p.snapshot); + return -BCH_ERR_fsck_repair_unimplemented; + } struct bkey_i_dirent *new_dirent = bch2_bkey_make_mut_typed(trans, iter, &d.s_c, 0, dirent); ret = PTR_ERR_OR_ZERO(new_dirent); @@ -1850,9 +1972,16 @@ static int check_dirent_to_subvol(struct btree_trans *trans, struct btree_iter * ret = lookup_inode(trans, target_inum, &subvol_root, &target_snapshot); if (ret && !bch2_err_matches(ret, ENOENT)) - return ret; + goto err; - if (fsck_err_on(parent_subvol != subvol_root.bi_parent_subvol, + if (ret) { + bch_err(c, "subvol %u points to missing inode root %llu", target_subvol, target_inum); + ret = -BCH_ERR_fsck_repair_unimplemented; + ret = 0; + goto err; + } + + if (fsck_err_on(!ret && parent_subvol != subvol_root.bi_parent_subvol, c, inode_bi_parent_wrong, "subvol root %llu has wrong bi_parent_subvol: got %u, should be %u", target_inum, @@ -1860,13 +1989,13 @@ static int check_dirent_to_subvol(struct btree_trans *trans, struct btree_iter * subvol_root.bi_parent_subvol = parent_subvol; ret = __bch2_fsck_write_inode(trans, &subvol_root, target_snapshot); if (ret) - return ret; + goto err; } ret = check_dirent_target(trans, iter, d, &subvol_root, target_snapshot); if (ret) - return ret; + goto err; out: err: fsck_err: -- cgit v1.2.3-59-g8ed1b From 09d4c2acbf4c864fef0f520bbcba256c9a19102e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 1 Apr 2024 00:00:56 -0400 Subject: bcachefs: reconstruct_inode() If an inode is missing, but corresponding extents and dirent still exist, it's well worth recreating it - this does so. Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 50 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index c1edc5647ba0..8e2010212cc3 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -459,6 +459,33 @@ static int reconstruct_subvol(struct btree_trans *trans, u32 snapshotid, u32 sub return 0; } +static int reconstruct_inode(struct btree_trans *trans, u32 snapshot, u64 inum, u64 size, unsigned mode) +{ + struct bch_fs *c = trans->c; + struct bch_inode_unpacked new_inode; + + bch2_inode_init_early(c, &new_inode); + bch2_inode_init_late(&new_inode, bch2_current_time(c), 0, 0, mode|0755, 0, NULL); + new_inode.bi_size = size; + new_inode.bi_inum = inum; + + return __bch2_fsck_write_inode(trans, &new_inode, snapshot); +} + +static int reconstruct_reg_inode(struct btree_trans *trans, u32 snapshot, u64 inum) +{ + struct btree_iter iter = {}; + + bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, SPOS(inum, U64_MAX, snapshot), 0); + struct bkey_s_c k = bch2_btree_iter_peek_prev(&iter); + bch2_trans_iter_exit(trans, &iter); + int ret = bkey_err(k); + if (ret) + return ret; + + return reconstruct_inode(trans, snapshot, inum, k.k->p.offset << 9, S_IFREG); +} + struct snapshots_seen_entry { u32 id; u32 equiv; @@ -1535,6 +1562,17 @@ static int check_extent(struct btree_trans *trans, struct btree_iter *iter, goto err; if (k.k->type != KEY_TYPE_whiteout) { + if (!i && (c->sb.btrees_lost_data & BIT_ULL(BTREE_ID_inodes))) { + ret = reconstruct_reg_inode(trans, k.k->p.snapshot, k.k->p.inode) ?: + bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); + if (ret) + goto err; + + inode->last_pos.inode--; + ret = -BCH_ERR_transaction_restart_nested; + goto err; + } + if (fsck_err_on(!i, c, extent_in_missing_inode, "extent in missing inode:\n %s", (printbuf_reset(&buf), @@ -2012,7 +2050,6 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, struct snapshots_seen *s) { struct bch_fs *c = trans->c; - struct bkey_s_c_dirent d; struct inode_walker_entry *i; struct printbuf buf = PRINTBUF; struct bpos equiv; @@ -2051,6 +2088,17 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, *hash_info = bch2_hash_info_init(c, &dir->inodes.data[0].inode); dir->first_this_inode = false; + if (!i && (c->sb.btrees_lost_data & BIT_ULL(BTREE_ID_inodes))) { + ret = reconstruct_inode(trans, k.k->p.snapshot, k.k->p.inode, 0, S_IFDIR) ?: + bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); + if (ret) + goto err; + + dir->last_pos.inode--; + ret = -BCH_ERR_transaction_restart_nested; + goto err; + } + if (fsck_err_on(!i, c, dirent_in_missing_dir_inode, "dirent in nonexisting directory:\n%s", (printbuf_reset(&buf), @@ -2085,7 +2133,7 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, if (k.k->type != KEY_TYPE_dirent) goto out; - d = bkey_s_c_to_dirent(k); + struct bkey_s_c_dirent d = bkey_s_c_to_dirent(k); if (d.v->d_type == DT_SUBVOL) { ret = check_dirent_to_subvol(trans, iter, d); -- cgit v1.2.3-59-g8ed1b From 173217bd73365867378b5e75a86f0049e1069ee8 Mon Sep 17 00:00:00 2001 From: Ritvik Budhiraja Date: Tue, 2 Apr 2024 14:01:28 -0500 Subject: smb3: retrying on failed server close In the current implementation, CIFS close sends a close to the server and does not check for the success of the server close. This patch adds functionality to check for server close return status and retries in case of an EBUSY or EAGAIN error. This can help avoid handle leaks Cc: stable@vger.kernel.org Signed-off-by: Ritvik Budhiraja Signed-off-by: Steve French --- fs/smb/client/cached_dir.c | 6 +++-- fs/smb/client/cifsfs.c | 11 ++++++++ fs/smb/client/cifsglob.h | 7 ++++-- fs/smb/client/file.c | 63 +++++++++++++++++++++++++++++++++++++++++----- fs/smb/client/smb1ops.c | 4 +-- fs/smb/client/smb2ops.c | 9 ++++--- fs/smb/client/smb2pdu.c | 2 +- 7 files changed, 85 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cached_dir.c b/fs/smb/client/cached_dir.c index a0017724d523..13a9d7acf8f8 100644 --- a/fs/smb/client/cached_dir.c +++ b/fs/smb/client/cached_dir.c @@ -417,6 +417,7 @@ smb2_close_cached_fid(struct kref *ref) { struct cached_fid *cfid = container_of(ref, struct cached_fid, refcount); + int rc; spin_lock(&cfid->cfids->cfid_list_lock); if (cfid->on_list) { @@ -430,9 +431,10 @@ smb2_close_cached_fid(struct kref *ref) cfid->dentry = NULL; if (cfid->is_open) { - SMB2_close(0, cfid->tcon, cfid->fid.persistent_fid, + rc = SMB2_close(0, cfid->tcon, cfid->fid.persistent_fid, cfid->fid.volatile_fid); - atomic_dec(&cfid->tcon->num_remote_opens); + if (rc != -EBUSY && rc != -EAGAIN) + atomic_dec(&cfid->tcon->num_remote_opens); } free_cached_dir(cfid); diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index aa6f1ecb7c0e..d41eedbff674 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -156,6 +156,7 @@ struct workqueue_struct *decrypt_wq; struct workqueue_struct *fileinfo_put_wq; struct workqueue_struct *cifsoplockd_wq; struct workqueue_struct *deferredclose_wq; +struct workqueue_struct *serverclose_wq; __u32 cifs_lock_secret; /* @@ -1888,6 +1889,13 @@ init_cifs(void) goto out_destroy_cifsoplockd_wq; } + serverclose_wq = alloc_workqueue("serverclose", + WQ_FREEZABLE|WQ_MEM_RECLAIM, 0); + if (!serverclose_wq) { + rc = -ENOMEM; + goto out_destroy_serverclose_wq; + } + rc = cifs_init_inodecache(); if (rc) goto out_destroy_deferredclose_wq; @@ -1962,6 +1970,8 @@ out_destroy_decrypt_wq: destroy_workqueue(decrypt_wq); out_destroy_cifsiod_wq: destroy_workqueue(cifsiod_wq); +out_destroy_serverclose_wq: + destroy_workqueue(serverclose_wq); out_clean_proc: cifs_proc_clean(); return rc; @@ -1991,6 +2001,7 @@ exit_cifs(void) destroy_workqueue(cifsoplockd_wq); destroy_workqueue(decrypt_wq); destroy_workqueue(fileinfo_put_wq); + destroy_workqueue(serverclose_wq); destroy_workqueue(cifsiod_wq); cifs_proc_clean(); } diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 286afbe346be..77ca7861a2cc 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -442,10 +442,10 @@ struct smb_version_operations { /* set fid protocol-specific info */ void (*set_fid)(struct cifsFileInfo *, struct cifs_fid *, __u32); /* close a file */ - void (*close)(const unsigned int, struct cifs_tcon *, + int (*close)(const unsigned int, struct cifs_tcon *, struct cifs_fid *); /* close a file, returning file attributes and timestamps */ - void (*close_getattr)(const unsigned int xid, struct cifs_tcon *tcon, + int (*close_getattr)(const unsigned int xid, struct cifs_tcon *tcon, struct cifsFileInfo *pfile_info); /* send a flush request to the server */ int (*flush)(const unsigned int, struct cifs_tcon *, struct cifs_fid *); @@ -1439,6 +1439,7 @@ struct cifsFileInfo { bool swapfile:1; bool oplock_break_cancelled:1; bool status_file_deleted:1; /* file has been deleted */ + bool offload:1; /* offload final part of _put to a wq */ unsigned int oplock_epoch; /* epoch from the lease break */ __u32 oplock_level; /* oplock/lease level from the lease break */ int count; @@ -1447,6 +1448,7 @@ struct cifsFileInfo { struct cifs_search_info srch_inf; struct work_struct oplock_break; /* work for oplock breaks */ struct work_struct put; /* work for the final part of _put */ + struct work_struct serverclose; /* work for serverclose */ struct delayed_work deferred; bool deferred_close_scheduled; /* Flag to indicate close is scheduled */ char *symlink_target; @@ -2103,6 +2105,7 @@ extern struct workqueue_struct *decrypt_wq; extern struct workqueue_struct *fileinfo_put_wq; extern struct workqueue_struct *cifsoplockd_wq; extern struct workqueue_struct *deferredclose_wq; +extern struct workqueue_struct *serverclose_wq; extern __u32 cifs_lock_secret; extern mempool_t *cifs_sm_req_poolp; diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index ab536ce8a04a..9be37d0fe724 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -459,6 +459,7 @@ cifs_down_write(struct rw_semaphore *sem) } static void cifsFileInfo_put_work(struct work_struct *work); +void serverclose_work(struct work_struct *work); struct cifsFileInfo *cifs_new_fileinfo(struct cifs_fid *fid, struct file *file, struct tcon_link *tlink, __u32 oplock, @@ -505,6 +506,7 @@ struct cifsFileInfo *cifs_new_fileinfo(struct cifs_fid *fid, struct file *file, cfile->tlink = cifs_get_tlink(tlink); INIT_WORK(&cfile->oplock_break, cifs_oplock_break); INIT_WORK(&cfile->put, cifsFileInfo_put_work); + INIT_WORK(&cfile->serverclose, serverclose_work); INIT_DELAYED_WORK(&cfile->deferred, smb2_deferred_work_close); mutex_init(&cfile->fh_mutex); spin_lock_init(&cfile->file_info_lock); @@ -596,6 +598,40 @@ static void cifsFileInfo_put_work(struct work_struct *work) cifsFileInfo_put_final(cifs_file); } +void serverclose_work(struct work_struct *work) +{ + struct cifsFileInfo *cifs_file = container_of(work, + struct cifsFileInfo, serverclose); + + struct cifs_tcon *tcon = tlink_tcon(cifs_file->tlink); + + struct TCP_Server_Info *server = tcon->ses->server; + int rc = 0; + int retries = 0; + int MAX_RETRIES = 4; + + do { + if (server->ops->close_getattr) + rc = server->ops->close_getattr(0, tcon, cifs_file); + else if (server->ops->close) + rc = server->ops->close(0, tcon, &cifs_file->fid); + + if (rc == -EBUSY || rc == -EAGAIN) { + retries++; + msleep(250); + } + } while ((rc == -EBUSY || rc == -EAGAIN) && (retries < MAX_RETRIES) + ); + + if (retries == MAX_RETRIES) + pr_warn("Serverclose failed %d times, giving up\n", MAX_RETRIES); + + if (cifs_file->offload) + queue_work(fileinfo_put_wq, &cifs_file->put); + else + cifsFileInfo_put_final(cifs_file); +} + /** * cifsFileInfo_put - release a reference of file priv data * @@ -636,10 +672,13 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, struct cifs_fid fid = {}; struct cifs_pending_open open; bool oplock_break_cancelled; + bool serverclose_offloaded = false; spin_lock(&tcon->open_file_lock); spin_lock(&cifsi->open_file_lock); spin_lock(&cifs_file->file_info_lock); + + cifs_file->offload = offload; if (--cifs_file->count > 0) { spin_unlock(&cifs_file->file_info_lock); spin_unlock(&cifsi->open_file_lock); @@ -681,13 +720,20 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, if (!tcon->need_reconnect && !cifs_file->invalidHandle) { struct TCP_Server_Info *server = tcon->ses->server; unsigned int xid; + int rc = 0; xid = get_xid(); if (server->ops->close_getattr) - server->ops->close_getattr(xid, tcon, cifs_file); + rc = server->ops->close_getattr(xid, tcon, cifs_file); else if (server->ops->close) - server->ops->close(xid, tcon, &cifs_file->fid); + rc = server->ops->close(xid, tcon, &cifs_file->fid); _free_xid(xid); + + if (rc == -EBUSY || rc == -EAGAIN) { + // Server close failed, hence offloading it as an async op + queue_work(serverclose_wq, &cifs_file->serverclose); + serverclose_offloaded = true; + } } if (oplock_break_cancelled) @@ -695,10 +741,15 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, cifs_del_pending_open(&open); - if (offload) - queue_work(fileinfo_put_wq, &cifs_file->put); - else - cifsFileInfo_put_final(cifs_file); + // if serverclose has been offloaded to wq (on failure), it will + // handle offloading put as well. If serverclose not offloaded, + // we need to handle offloading put here. + if (!serverclose_offloaded) { + if (offload) + queue_work(fileinfo_put_wq, &cifs_file->put); + else + cifsFileInfo_put_final(cifs_file); + } } int cifs_open(struct inode *inode, struct file *file) diff --git a/fs/smb/client/smb1ops.c b/fs/smb/client/smb1ops.c index a9eaba8083b0..212ec6f66ec6 100644 --- a/fs/smb/client/smb1ops.c +++ b/fs/smb/client/smb1ops.c @@ -753,11 +753,11 @@ cifs_set_fid(struct cifsFileInfo *cfile, struct cifs_fid *fid, __u32 oplock) cinode->can_cache_brlcks = CIFS_CACHE_WRITE(cinode); } -static void +static int cifs_close_file(const unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *fid) { - CIFSSMBClose(xid, tcon, fid->netfid); + return CIFSSMBClose(xid, tcon, fid->netfid); } static int diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 35bf7eb315cd..87b63f6ad2e2 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -1412,14 +1412,14 @@ smb2_set_fid(struct cifsFileInfo *cfile, struct cifs_fid *fid, __u32 oplock) memcpy(cfile->fid.create_guid, fid->create_guid, 16); } -static void +static int smb2_close_file(const unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *fid) { - SMB2_close(xid, tcon, fid->persistent_fid, fid->volatile_fid); + return SMB2_close(xid, tcon, fid->persistent_fid, fid->volatile_fid); } -static void +static int smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon, struct cifsFileInfo *cfile) { @@ -1430,7 +1430,7 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon, rc = __SMB2_close(xid, tcon, cfile->fid.persistent_fid, cfile->fid.volatile_fid, &file_inf); if (rc) - return; + return rc; inode = d_inode(cfile->dentry); @@ -1459,6 +1459,7 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon, /* End of file and Attributes should not have to be updated on close */ spin_unlock(&inode->i_lock); + return rc; } static int diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 3ea688558e6c..c0c4933af5fc 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -3628,9 +3628,9 @@ replay_again: memcpy(&pbuf->network_open_info, &rsp->network_open_info, sizeof(pbuf->network_open_info)); + atomic_dec(&tcon->num_remote_opens); } - atomic_dec(&tcon->num_remote_opens); close_exit: SMB2_close_free(&rqst); free_rsp_buf(resp_buftype, rsp); -- cgit v1.2.3-59-g8ed1b From ca545b7f0823f19db0f1148d59bc5e1a56634502 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:33:53 -0300 Subject: smb: client: fix potential UAF in cifs_debug_files_proc_show() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/cifs_debug.c | 2 ++ fs/smb/client/cifsglob.h | 10 ++++++++++ 2 files changed, 12 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/cifs_debug.c b/fs/smb/client/cifs_debug.c index 226d4835c92d..c9aec9a38ad3 100644 --- a/fs/smb/client/cifs_debug.c +++ b/fs/smb/client/cifs_debug.c @@ -250,6 +250,8 @@ static int cifs_debug_files_proc_show(struct seq_file *m, void *v) spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(server, &cifs_tcp_ses_list, tcp_ses_list) { list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { spin_lock(&tcon->open_file_lock); list_for_each_entry(cfile, &tcon->openFileList, tlist) { diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 77ca7861a2cc..f6a302205f89 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -2325,4 +2325,14 @@ struct smb2_compound_vars { struct kvec ea_iov; }; +static inline bool cifs_ses_exiting(struct cifs_ses *ses) +{ + bool ret; + + spin_lock(&ses->ses_lock); + ret = ses->ses_status == SES_EXITING; + spin_unlock(&ses->ses_lock); + return ret; +} + #endif /* _CIFS_GLOB_H */ -- cgit v1.2.3-59-g8ed1b From 58acd1f497162e7d282077f816faa519487be045 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:33:54 -0300 Subject: smb: client: fix potential UAF in cifs_dump_full_key() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/ioctl.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/ioctl.c b/fs/smb/client/ioctl.c index c012dfdba80d..855ac5a62edf 100644 --- a/fs/smb/client/ioctl.c +++ b/fs/smb/client/ioctl.c @@ -247,7 +247,9 @@ static int cifs_dump_full_key(struct cifs_tcon *tcon, struct smb3_full_key_debug spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(server_it, &cifs_tcp_ses_list, tcp_ses_list) { list_for_each_entry(ses_it, &server_it->smb_ses_list, smb_ses_list) { - if (ses_it->Suid == out.session_id) { + spin_lock(&ses_it->ses_lock); + if (ses_it->ses_status != SES_EXITING && + ses_it->Suid == out.session_id) { ses = ses_it; /* * since we are using the session outside the crit @@ -255,9 +257,11 @@ static int cifs_dump_full_key(struct cifs_tcon *tcon, struct smb3_full_key_debug * so increment its refcount */ cifs_smb_ses_inc_refcount(ses); + spin_unlock(&ses_it->ses_lock); found = true; goto search_end; } + spin_unlock(&ses_it->ses_lock); } } search_end: -- cgit v1.2.3-59-g8ed1b From d3da25c5ac84430f89875ca7485a3828150a7e0a Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:33:55 -0300 Subject: smb: client: fix potential UAF in cifs_stats_proc_write() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/cifs_debug.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/cifs_debug.c b/fs/smb/client/cifs_debug.c index c9aec9a38ad3..8535c9907462 100644 --- a/fs/smb/client/cifs_debug.c +++ b/fs/smb/client/cifs_debug.c @@ -678,6 +678,8 @@ static ssize_t cifs_stats_proc_write(struct file *file, } #endif /* CONFIG_CIFS_STATS2 */ list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { atomic_set(&tcon->num_smbs_sent, 0); spin_lock(&tcon->stat_lock); -- cgit v1.2.3-59-g8ed1b From 0865ffefea197b437ba78b5dd8d8e256253efd65 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:33:56 -0300 Subject: smb: client: fix potential UAF in cifs_stats_proc_show() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/cifs_debug.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/cifs_debug.c b/fs/smb/client/cifs_debug.c index 8535c9907462..c71ae5c04306 100644 --- a/fs/smb/client/cifs_debug.c +++ b/fs/smb/client/cifs_debug.c @@ -759,6 +759,8 @@ static int cifs_stats_proc_show(struct seq_file *m, void *v) } #endif /* STATS2 */ list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { i++; seq_printf(m, "\n%d) %s", i, tcon->tree_name); -- cgit v1.2.3-59-g8ed1b From 705c76fbf726c7a2f6ff9143d4013b18daaaebf1 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:33:58 -0300 Subject: smb: client: fix potential UAF in smb2_is_valid_lease_break() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/smb2misc.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/smb2misc.c b/fs/smb/client/smb2misc.c index 82b84a4941dd..ed6d033dce96 100644 --- a/fs/smb/client/smb2misc.c +++ b/fs/smb/client/smb2misc.c @@ -622,6 +622,8 @@ smb2_is_valid_lease_break(char *buffer, struct TCP_Server_Info *server) /* look up tcon based on tid & uid */ spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { spin_lock(&tcon->open_file_lock); cifs_stats_inc( -- cgit v1.2.3-59-g8ed1b From 22863485a4626ec6ecf297f4cc0aef709bc862e4 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:33:59 -0300 Subject: smb: client: fix potential UAF in smb2_is_valid_oplock_break() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/smb2misc.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/smb2misc.c b/fs/smb/client/smb2misc.c index ed6d033dce96..cc72be5a93a9 100644 --- a/fs/smb/client/smb2misc.c +++ b/fs/smb/client/smb2misc.c @@ -699,6 +699,8 @@ smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server) /* look up tcon based on tid & uid */ spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { spin_lock(&tcon->open_file_lock); -- cgit v1.2.3-59-g8ed1b From 69ccf040acddf33a3a85ec0f6b45ef84b0f7ec29 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:34:00 -0300 Subject: smb: client: fix potential UAF in is_valid_oplock_break() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/misc.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c index 1ea22b3955a2..33ac4f8f5050 100644 --- a/fs/smb/client/misc.c +++ b/fs/smb/client/misc.c @@ -481,6 +481,8 @@ is_valid_oplock_break(char *buffer, struct TCP_Server_Info *srv) /* look up tcon based on tid & uid */ spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { if (tcon->tid != buf->Tid) continue; -- cgit v1.2.3-59-g8ed1b From 63981561ffd2d4987807df4126f96a11e18b0c1d Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:34:02 -0300 Subject: smb: client: fix potential UAF in smb2_is_network_name_deleted() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/smb2ops.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 87b63f6ad2e2..b156eefa75d7 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -2481,6 +2481,8 @@ smb2_is_network_name_deleted(char *buf, struct TCP_Server_Info *server) spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { if (tcon->tid == le32_to_cpu(shdr->Id.SyncId.TreeId)) { spin_lock(&tcon->tc_lock); -- cgit v1.2.3-59-g8ed1b From e0e50401cc3921c9eaf1b0e667db174519ea939f Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 2 Apr 2024 16:34:04 -0300 Subject: smb: client: fix potential UAF in cifs_signal_cifsd_for_reconnect() Skip sessions that are being teared down (status == SES_EXITING) to avoid UAF. Cc: stable@vger.kernel.org Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/connect.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 95e4bda4fd51..85679ae106fd 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -175,6 +175,8 @@ cifs_signal_cifsd_for_reconnect(struct TCP_Server_Info *server, spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + if (cifs_ses_exiting(ses)) + continue; spin_lock(&ses->chan_lock); for (i = 0; i < ses->chan_count; i++) { if (!ses->chans[i].server) -- cgit v1.2.3-59-g8ed1b From e60aa472400b1ff8d0e6c563a2eb05916927f10a Mon Sep 17 00:00:00 2001 From: Thomas Bertschinger Date: Thu, 14 Mar 2024 10:02:18 -0600 Subject: bcachefs: create debugfs dir for each btree MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This creates a subdirectory for each individual btree under the btrees/ debugfs directory. Directory structure, before: /sys/kernel/debug/bcachefs/$FS_ID/btrees/ ├── alloc ├── alloc-bfloat-failed ├── alloc-formats ├── backpointers ├── backpointers-bfloat-failed ├── backpointers-formats ... Directory structure, after: /sys/kernel/debug/bcachefs/$FS_ID/btrees/ ├── alloc │   ├── bfloat-failed │   ├── formats │   └── keys ├── backpointers │   ├── bfloat-failed │   ├── formats │   └── keys ... Signed-off-by: Thomas Bertschinger Signed-off-by: Kent Overstreet --- fs/bcachefs/debug.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index 208ce6f0fc43..7b681ae91510 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -866,6 +866,20 @@ void bch2_fs_debug_exit(struct bch_fs *c) debugfs_remove_recursive(c->fs_debug_dir); } +static void bch2_fs_debug_btree_init(struct bch_fs *c, struct btree_debug *bd) +{ + struct dentry *d; + + d = debugfs_create_dir(bch2_btree_id_str(bd->id), c->btree_debug_dir); + + debugfs_create_file("keys", 0400, d, bd, &btree_debug_ops); + + debugfs_create_file("formats", 0400, d, bd, &btree_format_debug_ops); + + debugfs_create_file("bfloat-failed", 0400, d, bd, + &bfloat_failed_debug_ops); +} + void bch2_fs_debug_init(struct bch_fs *c) { struct btree_debug *bd; @@ -902,21 +916,7 @@ void bch2_fs_debug_init(struct bch_fs *c) bd < c->btree_debug + ARRAY_SIZE(c->btree_debug); bd++) { bd->id = bd - c->btree_debug; - debugfs_create_file(bch2_btree_id_str(bd->id), - 0400, c->btree_debug_dir, bd, - &btree_debug_ops); - - snprintf(name, sizeof(name), "%s-formats", - bch2_btree_id_str(bd->id)); - - debugfs_create_file(name, 0400, c->btree_debug_dir, bd, - &btree_format_debug_ops); - - snprintf(name, sizeof(name), "%s-bfloat-failed", - bch2_btree_id_str(bd->id)); - - debugfs_create_file(name, 0400, c->btree_debug_dir, bd, - &bfloat_failed_debug_ops); + bch2_fs_debug_btree_init(c, bd); } } -- cgit v1.2.3-59-g8ed1b From d2f277e26f521ccf6fb438463b41dba6123caabe Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:45 +0200 Subject: fsnotify: rename fsnotify_{get,put}_sb_connectors() Instead of counting the number of connectors in an sb, we would like to count the number of watched objects per priority group. As a start, create an accessor fsnotify_sb_watched_objects() to s_fsnotify_connectors and rename the fsnotify_{get,put}_sb_connectors() helpers to fsnotify_{get,put}_sb_watchers() to better describes the counter. Increment the counter at the end of fsnotify_attach_connector_to_object() if connector was attached instead of decrementing it on race to connect. This is fine, because fsnotify_delete_sb() cannot be running in parallel to fsnotify_attach_connector_to_object() which requires a reference to a filesystem object. Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-2-amir73il@gmail.com> --- fs/notify/fsnotify.c | 4 +-- fs/notify/mark.c | 67 ++++++++++++++++++++++------------------ include/linux/fsnotify.h | 2 +- include/linux/fsnotify_backend.h | 5 +++ 4 files changed, 45 insertions(+), 33 deletions(-) (limited to 'fs') diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 2fc105a72a8f..503e7c75e777 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -92,8 +92,8 @@ void fsnotify_sb_delete(struct super_block *sb) fsnotify_unmount_inodes(sb); fsnotify_clear_marks_by_sb(sb); /* Wait for outstanding object references from connectors */ - wait_var_event(&sb->s_fsnotify_connectors, - !atomic_long_read(&sb->s_fsnotify_connectors)); + wait_var_event(fsnotify_sb_watched_objects(sb), + !atomic_long_read(fsnotify_sb_watched_objects(sb))); } /* diff --git a/fs/notify/mark.c b/fs/notify/mark.c index d6944ff86ffa..3bcce2671bd5 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -116,10 +116,43 @@ __u32 fsnotify_conn_mask(struct fsnotify_mark_connector *conn) return *fsnotify_conn_mask_p(conn); } +static void fsnotify_get_sb_watched_objects(struct super_block *sb) +{ + atomic_long_inc(fsnotify_sb_watched_objects(sb)); +} + +static void fsnotify_put_sb_watched_objects(struct super_block *sb) +{ + if (atomic_long_dec_and_test(fsnotify_sb_watched_objects(sb))) + wake_up_var(fsnotify_sb_watched_objects(sb)); +} + static void fsnotify_get_inode_ref(struct inode *inode) { ihold(inode); - atomic_long_inc(&inode->i_sb->s_fsnotify_connectors); + fsnotify_get_sb_watched_objects(inode->i_sb); +} + +static void fsnotify_put_inode_ref(struct inode *inode) +{ + fsnotify_put_sb_watched_objects(inode->i_sb); + iput(inode); +} + +static void fsnotify_get_sb_watchers(struct fsnotify_mark_connector *conn) +{ + struct super_block *sb = fsnotify_connector_sb(conn); + + if (sb) + fsnotify_get_sb_watched_objects(sb); +} + +static void fsnotify_put_sb_watchers(struct fsnotify_mark_connector *conn) +{ + struct super_block *sb = fsnotify_connector_sb(conn); + + if (sb) + fsnotify_put_sb_watched_objects(sb); } /* @@ -213,31 +246,6 @@ static void fsnotify_connector_destroy_workfn(struct work_struct *work) } } -static void fsnotify_put_inode_ref(struct inode *inode) -{ - struct super_block *sb = inode->i_sb; - - iput(inode); - if (atomic_long_dec_and_test(&sb->s_fsnotify_connectors)) - wake_up_var(&sb->s_fsnotify_connectors); -} - -static void fsnotify_get_sb_connectors(struct fsnotify_mark_connector *conn) -{ - struct super_block *sb = fsnotify_connector_sb(conn); - - if (sb) - atomic_long_inc(&sb->s_fsnotify_connectors); -} - -static void fsnotify_put_sb_connectors(struct fsnotify_mark_connector *conn) -{ - struct super_block *sb = fsnotify_connector_sb(conn); - - if (sb && atomic_long_dec_and_test(&sb->s_fsnotify_connectors)) - wake_up_var(&sb->s_fsnotify_connectors); -} - static void *fsnotify_detach_connector_from_object( struct fsnotify_mark_connector *conn, unsigned int *type) @@ -261,7 +269,7 @@ static void *fsnotify_detach_connector_from_object( fsnotify_conn_sb(conn)->s_fsnotify_mask = 0; } - fsnotify_put_sb_connectors(conn); + fsnotify_put_sb_watchers(conn); rcu_assign_pointer(*(conn->obj), NULL); conn->obj = NULL; conn->type = FSNOTIFY_OBJ_TYPE_DETACHED; @@ -549,8 +557,6 @@ static int fsnotify_attach_connector_to_object(fsnotify_connp_t *connp, conn->flags = 0; conn->type = obj_type; conn->obj = connp; - conn->flags = 0; - fsnotify_get_sb_connectors(conn); /* * cmpxchg() provides the barrier so that readers of *connp can see @@ -558,10 +564,11 @@ static int fsnotify_attach_connector_to_object(fsnotify_connp_t *connp, */ if (cmpxchg(connp, NULL, conn)) { /* Someone else created list structure for us */ - fsnotify_put_sb_connectors(conn); kmem_cache_free(fsnotify_mark_connector_cachep, conn); + return 0; } + fsnotify_get_sb_watchers(conn); return 0; } diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 1a9de119a0f7..e470bb67c9a3 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -20,7 +20,7 @@ /* Are there any inode/mount/sb objects that are being watched at all? */ static inline bool fsnotify_sb_has_watchers(struct super_block *sb) { - return atomic_long_read(&sb->s_fsnotify_connectors); + return atomic_long_read(fsnotify_sb_watched_objects(sb)); } /* diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index 8f40c349b228..d4e3bc55d174 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -483,6 +483,11 @@ struct fsnotify_mark_connector { struct hlist_head list; }; +static inline atomic_long_t *fsnotify_sb_watched_objects(struct super_block *sb) +{ + return &sb->s_fsnotify_connectors; +} + /* * A mark is simply an object attached to an in core inode which allows an * fsnotify listener to indicate they are either no longer interested in events -- cgit v1.2.3-59-g8ed1b From f115815d75332f9dabb0d7c29c8f67b0f26889c5 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:46 +0200 Subject: fsnotify: create helpers to get sb and connp from object In preparation to passing an object pointer to add/remove/find mark helpers, create helpers to get sb and connp by object type. Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-3-amir73il@gmail.com> --- fs/notify/fsnotify.h | 15 +++++++++++++++ fs/notify/mark.c | 15 +++++++++++++++ 2 files changed, 30 insertions(+) (limited to 'fs') diff --git a/fs/notify/fsnotify.h b/fs/notify/fsnotify.h index fde74eb333cc..6d9951aac449 100644 --- a/fs/notify/fsnotify.h +++ b/fs/notify/fsnotify.h @@ -27,6 +27,21 @@ static inline struct super_block *fsnotify_conn_sb( return container_of(conn->obj, struct super_block, s_fsnotify_marks); } +static inline struct super_block *fsnotify_object_sb(void *obj, + enum fsnotify_obj_type obj_type) +{ + switch (obj_type) { + case FSNOTIFY_OBJ_TYPE_INODE: + return ((struct inode *)obj)->i_sb; + case FSNOTIFY_OBJ_TYPE_VFSMOUNT: + return ((struct vfsmount *)obj)->mnt_sb; + case FSNOTIFY_OBJ_TYPE_SB: + return (struct super_block *)obj; + default: + return NULL; + } +} + static inline struct super_block *fsnotify_connector_sb( struct fsnotify_mark_connector *conn) { diff --git a/fs/notify/mark.c b/fs/notify/mark.c index 3bcce2671bd5..d9977af25832 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -97,6 +97,21 @@ void fsnotify_get_mark(struct fsnotify_mark *mark) refcount_inc(&mark->refcnt); } +static fsnotify_connp_t *fsnotify_object_connp(void *obj, + enum fsnotify_obj_type obj_type) +{ + switch (obj_type) { + case FSNOTIFY_OBJ_TYPE_INODE: + return &((struct inode *)obj)->i_fsnotify_marks; + case FSNOTIFY_OBJ_TYPE_VFSMOUNT: + return &real_mount(obj)->mnt_fsnotify_marks; + case FSNOTIFY_OBJ_TYPE_SB: + return &((struct super_block *)obj)->s_fsnotify_marks; + default: + return NULL; + } +} + static __u32 *fsnotify_conn_mask_p(struct fsnotify_mark_connector *conn) { if (conn->type == FSNOTIFY_OBJ_TYPE_INODE) -- cgit v1.2.3-59-g8ed1b From 230d97d39ee2eb9030309f04f98615aaeb420dac Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:47 +0200 Subject: fsnotify: create a wrapper fsnotify_find_inode_mark() In preparation to passing an object pointer to fsnotify_find_mark(), add a wrapper fsnotify_find_inode_mark() and use it where possible. Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-4-amir73il@gmail.com> --- fs/nfsd/filecache.c | 4 ++-- fs/notify/dnotify/dnotify.c | 4 ++-- fs/notify/inotify/inotify_user.c | 2 +- include/linux/fsnotify_backend.h | 7 +++++++ kernel/audit_tree.c | 2 +- kernel/audit_watch.c | 2 +- 6 files changed, 14 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index ddd3e0d9cfa6..ad9083ca144b 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -159,8 +159,8 @@ nfsd_file_mark_find_or_create(struct nfsd_file *nf, struct inode *inode) do { fsnotify_group_lock(nfsd_file_fsnotify_group); - mark = fsnotify_find_mark(&inode->i_fsnotify_marks, - nfsd_file_fsnotify_group); + mark = fsnotify_find_inode_mark(inode, + nfsd_file_fsnotify_group); if (mark) { nfm = nfsd_file_mark_get(container_of(mark, struct nfsd_file_mark, diff --git a/fs/notify/dnotify/dnotify.c b/fs/notify/dnotify/dnotify.c index 3464fa7e8538..f3669403fabf 100644 --- a/fs/notify/dnotify/dnotify.c +++ b/fs/notify/dnotify/dnotify.c @@ -162,7 +162,7 @@ void dnotify_flush(struct file *filp, fl_owner_t id) if (!S_ISDIR(inode->i_mode)) return; - fsn_mark = fsnotify_find_mark(&inode->i_fsnotify_marks, dnotify_group); + fsn_mark = fsnotify_find_inode_mark(inode, dnotify_group); if (!fsn_mark) return; dn_mark = container_of(fsn_mark, struct dnotify_mark, fsn_mark); @@ -326,7 +326,7 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned int arg) fsnotify_group_lock(dnotify_group); /* add the new_fsn_mark or find an old one. */ - fsn_mark = fsnotify_find_mark(&inode->i_fsnotify_marks, dnotify_group); + fsn_mark = fsnotify_find_inode_mark(inode, dnotify_group); if (fsn_mark) { dn_mark = container_of(fsn_mark, struct dnotify_mark, fsn_mark); spin_lock(&fsn_mark->lock); diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index 85d8fdd55329..4ffc30606e0b 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -544,7 +544,7 @@ static int inotify_update_existing_watch(struct fsnotify_group *group, int create = (arg & IN_MASK_CREATE); int ret; - fsn_mark = fsnotify_find_mark(&inode->i_fsnotify_marks, group); + fsn_mark = fsnotify_find_inode_mark(inode, group); if (!fsn_mark) return -ENOENT; else if (create) { diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index d4e3bc55d174..992b57a7e95f 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -789,6 +789,13 @@ static inline int fsnotify_add_inode_mark_locked(struct fsnotify_mark *mark, FSNOTIFY_OBJ_TYPE_INODE, add_flags); } +static inline struct fsnotify_mark *fsnotify_find_inode_mark( + struct inode *inode, + struct fsnotify_group *group) +{ + return fsnotify_find_mark(&inode->i_fsnotify_marks, group); +} + /* given a group and a mark, flag mark to be freed when all references are dropped */ extern void fsnotify_destroy_mark(struct fsnotify_mark *mark, struct fsnotify_group *group); diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 1b07e6f12a07..f2f38903b2fe 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -463,7 +463,7 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) int n; fsnotify_group_lock(audit_tree_group); - mark = fsnotify_find_mark(&inode->i_fsnotify_marks, audit_tree_group); + mark = fsnotify_find_inode_mark(inode, audit_tree_group); if (!mark) return create_chunk(inode, tree); diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c index 7a98cd176a12..7f358740e958 100644 --- a/kernel/audit_watch.c +++ b/kernel/audit_watch.c @@ -90,7 +90,7 @@ static inline struct audit_parent *audit_find_parent(struct inode *inode) struct audit_parent *parent = NULL; struct fsnotify_mark *entry; - entry = fsnotify_find_mark(&inode->i_fsnotify_marks, audit_watch_group); + entry = fsnotify_find_inode_mark(inode, audit_watch_group); if (entry) parent = container_of(entry, struct audit_parent, mark); -- cgit v1.2.3-59-g8ed1b From b5cae086cc2fde629495d988106d70e44c90cb20 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:48 +0200 Subject: fanotify: merge two checks regarding add of ignore mark There are two similar checks for adding an ignore mark without FAN_MARK_IGNORED_SURV_MODIFY, one for the old FAN_MARK_IGNORED flag and one for the new FAN_MARK_IGNORE flag. Merge the two checks into a single location. Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-5-amir73il@gmail.com> --- fs/notify/fanotify/fanotify_user.c | 35 +++++++++++++++++------------------ 1 file changed, 17 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 4201723357cf..04a0b31f325d 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -1412,18 +1412,6 @@ static int fanotify_add_inode_mark(struct fsnotify_group *group, struct inode *inode, __u32 mask, unsigned int flags, struct fan_fsid *fsid) { - pr_debug("%s: group=%p inode=%p\n", __func__, group, inode); - - /* - * If some other task has this inode open for write we should not add - * an ignore mask, unless that ignore mask is supposed to survive - * modification changes anyway. - */ - if ((flags & FANOTIFY_MARK_IGNORE_BITS) && - !(flags & FAN_MARK_IGNORED_SURV_MODIFY) && - inode_is_open_for_write(inode)) - return 0; - return fanotify_add_mark(group, &inode->i_fsnotify_marks, FSNOTIFY_OBJ_TYPE_INODE, mask, flags, fsid); } @@ -1913,12 +1901,23 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, else mnt = path.mnt; - ret = mnt ? -EINVAL : -EISDIR; - /* FAN_MARK_IGNORE requires SURV_MODIFY for sb/mount/dir marks */ - if (mark_cmd == FAN_MARK_ADD && ignore == FAN_MARK_IGNORE && - (mnt || S_ISDIR(inode->i_mode)) && - !(flags & FAN_MARK_IGNORED_SURV_MODIFY)) - goto path_put_and_out; + /* + * If some other task has this inode open for write we should not add + * an ignore mask, unless that ignore mask is supposed to survive + * modification changes anyway. + */ + if (mark_cmd == FAN_MARK_ADD && (flags & FANOTIFY_MARK_IGNORE_BITS) && + !(flags & FAN_MARK_IGNORED_SURV_MODIFY)) { + ret = mnt ? -EINVAL : -EISDIR; + /* FAN_MARK_IGNORE requires SURV_MODIFY for sb/mount/dir marks */ + if (ignore == FAN_MARK_IGNORE && + (mnt || S_ISDIR(inode->i_mode))) + goto path_put_and_out; + + ret = 0; + if (inode && inode_is_open_for_write(inode)) + goto path_put_and_out; + } /* Mask out FAN_EVENT_ON_CHILD flag for sb/mount/non-dir marks */ if (mnt || !S_ISDIR(inode->i_mode)) { -- cgit v1.2.3-59-g8ed1b From 687c217c6aa2c24e6ddf98cc7f86a5b986e5918d Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:49 +0200 Subject: fsnotify: pass object pointer and type to fsnotify mark helpers Instead of passing fsnotify_connp_t, pass the pointer to the marked object. Store the object pointer in the connector and move the definition of fsnotify_connp_t to internal fsnotify subsystem API, so it is no longer used by fsnotify backends. Suggested-by: Jan Kara Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-6-amir73il@gmail.com> --- fs/notify/fanotify/fanotify_user.c | 95 ++++++++------------------------------ fs/notify/fsnotify.h | 23 ++++----- fs/notify/mark.c | 28 ++++++----- include/linux/fsnotify_backend.h | 33 +++++-------- 4 files changed, 59 insertions(+), 120 deletions(-) (limited to 'fs') diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 04a0b31f325d..925015ee7fbd 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -1076,7 +1076,7 @@ static __u32 fanotify_mark_remove_from_mask(struct fsnotify_mark *fsn_mark, } static int fanotify_remove_mark(struct fsnotify_group *group, - fsnotify_connp_t *connp, __u32 mask, + void *obj, unsigned int obj_type, __u32 mask, unsigned int flags, __u32 umask) { struct fsnotify_mark *fsn_mark = NULL; @@ -1084,7 +1084,7 @@ static int fanotify_remove_mark(struct fsnotify_group *group, int destroy_mark; fsnotify_group_lock(group); - fsn_mark = fsnotify_find_mark(connp, group); + fsn_mark = fsnotify_find_mark(obj, obj_type, group); if (!fsn_mark) { fsnotify_group_unlock(group); return -ENOENT; @@ -1105,30 +1105,6 @@ static int fanotify_remove_mark(struct fsnotify_group *group, return 0; } -static int fanotify_remove_vfsmount_mark(struct fsnotify_group *group, - struct vfsmount *mnt, __u32 mask, - unsigned int flags, __u32 umask) -{ - return fanotify_remove_mark(group, &real_mount(mnt)->mnt_fsnotify_marks, - mask, flags, umask); -} - -static int fanotify_remove_sb_mark(struct fsnotify_group *group, - struct super_block *sb, __u32 mask, - unsigned int flags, __u32 umask) -{ - return fanotify_remove_mark(group, &sb->s_fsnotify_marks, mask, - flags, umask); -} - -static int fanotify_remove_inode_mark(struct fsnotify_group *group, - struct inode *inode, __u32 mask, - unsigned int flags, __u32 umask) -{ - return fanotify_remove_mark(group, &inode->i_fsnotify_marks, mask, - flags, umask); -} - static bool fanotify_mark_update_flags(struct fsnotify_mark *fsn_mark, unsigned int fan_flags) { @@ -1249,7 +1225,7 @@ static int fanotify_set_mark_fsid(struct fsnotify_group *group, } static struct fsnotify_mark *fanotify_add_new_mark(struct fsnotify_group *group, - fsnotify_connp_t *connp, + void *obj, unsigned int obj_type, unsigned int fan_flags, struct fan_fsid *fsid) @@ -1288,7 +1264,7 @@ static struct fsnotify_mark *fanotify_add_new_mark(struct fsnotify_group *group, fan_mark->fsid.val[0] = fan_mark->fsid.val[1] = 0; } - ret = fsnotify_add_mark_locked(mark, connp, obj_type, 0); + ret = fsnotify_add_mark_locked(mark, obj, obj_type, 0); if (ret) goto out_put_mark; @@ -1344,7 +1320,7 @@ static int fanotify_may_update_existing_mark(struct fsnotify_mark *fsn_mark, } static int fanotify_add_mark(struct fsnotify_group *group, - fsnotify_connp_t *connp, unsigned int obj_type, + void *obj, unsigned int obj_type, __u32 mask, unsigned int fan_flags, struct fan_fsid *fsid) { @@ -1353,9 +1329,9 @@ static int fanotify_add_mark(struct fsnotify_group *group, int ret = 0; fsnotify_group_lock(group); - fsn_mark = fsnotify_find_mark(connp, group); + fsn_mark = fsnotify_find_mark(obj, obj_type, group); if (!fsn_mark) { - fsn_mark = fanotify_add_new_mark(group, connp, obj_type, + fsn_mark = fanotify_add_new_mark(group, obj, obj_type, fan_flags, fsid); if (IS_ERR(fsn_mark)) { fsnotify_group_unlock(group); @@ -1392,30 +1368,6 @@ out: return ret; } -static int fanotify_add_vfsmount_mark(struct fsnotify_group *group, - struct vfsmount *mnt, __u32 mask, - unsigned int flags, struct fan_fsid *fsid) -{ - return fanotify_add_mark(group, &real_mount(mnt)->mnt_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_VFSMOUNT, mask, flags, fsid); -} - -static int fanotify_add_sb_mark(struct fsnotify_group *group, - struct super_block *sb, __u32 mask, - unsigned int flags, struct fan_fsid *fsid) -{ - return fanotify_add_mark(group, &sb->s_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_SB, mask, flags, fsid); -} - -static int fanotify_add_inode_mark(struct fsnotify_group *group, - struct inode *inode, __u32 mask, - unsigned int flags, struct fan_fsid *fsid) -{ - return fanotify_add_mark(group, &inode->i_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_INODE, mask, flags, fsid); -} - static struct fsnotify_event *fanotify_alloc_overflow_event(void) { struct fanotify_event *oevent; @@ -1738,6 +1690,7 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, unsigned int mark_cmd = flags & FANOTIFY_MARK_CMD_BITS; unsigned int ignore = flags & FANOTIFY_MARK_IGNORE_BITS; unsigned int obj_type, fid_mode; + void *obj; u32 umask = 0; int ret; @@ -1896,10 +1849,16 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, } /* inode held in place by reference to path; group by fget on fd */ - if (mark_type == FAN_MARK_INODE) + if (mark_type == FAN_MARK_INODE) { inode = path.dentry->d_inode; - else + obj = inode; + } else { mnt = path.mnt; + if (mark_type == FAN_MARK_MOUNT) + obj = mnt; + else + obj = mnt->mnt_sb; + } /* * If some other task has this inode open for write we should not add @@ -1935,26 +1894,12 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, /* create/update an inode mark */ switch (mark_cmd) { case FAN_MARK_ADD: - if (mark_type == FAN_MARK_MOUNT) - ret = fanotify_add_vfsmount_mark(group, mnt, mask, - flags, fsid); - else if (mark_type == FAN_MARK_FILESYSTEM) - ret = fanotify_add_sb_mark(group, mnt->mnt_sb, mask, - flags, fsid); - else - ret = fanotify_add_inode_mark(group, inode, mask, - flags, fsid); + ret = fanotify_add_mark(group, obj, obj_type, mask, flags, + fsid); break; case FAN_MARK_REMOVE: - if (mark_type == FAN_MARK_MOUNT) - ret = fanotify_remove_vfsmount_mark(group, mnt, mask, - flags, umask); - else if (mark_type == FAN_MARK_FILESYSTEM) - ret = fanotify_remove_sb_mark(group, mnt->mnt_sb, mask, - flags, umask); - else - ret = fanotify_remove_inode_mark(group, inode, mask, - flags, umask); + ret = fanotify_remove_mark(group, obj, obj_type, mask, flags, + umask); break; default: ret = -EINVAL; diff --git a/fs/notify/fsnotify.h b/fs/notify/fsnotify.h index 6d9951aac449..b1fec7cd8ee4 100644 --- a/fs/notify/fsnotify.h +++ b/fs/notify/fsnotify.h @@ -9,22 +9,28 @@ #include "../mount.h" +/* + * fsnotify_connp_t is what we embed in objects which connector can be attached + * to. + */ +typedef struct fsnotify_mark_connector __rcu *fsnotify_connp_t; + static inline struct inode *fsnotify_conn_inode( struct fsnotify_mark_connector *conn) { - return container_of(conn->obj, struct inode, i_fsnotify_marks); + return conn->obj; } static inline struct mount *fsnotify_conn_mount( struct fsnotify_mark_connector *conn) { - return container_of(conn->obj, struct mount, mnt_fsnotify_marks); + return real_mount(conn->obj); } static inline struct super_block *fsnotify_conn_sb( struct fsnotify_mark_connector *conn) { - return container_of(conn->obj, struct super_block, s_fsnotify_marks); + return conn->obj; } static inline struct super_block *fsnotify_object_sb(void *obj, @@ -45,16 +51,7 @@ static inline struct super_block *fsnotify_object_sb(void *obj, static inline struct super_block *fsnotify_connector_sb( struct fsnotify_mark_connector *conn) { - switch (conn->type) { - case FSNOTIFY_OBJ_TYPE_INODE: - return fsnotify_conn_inode(conn)->i_sb; - case FSNOTIFY_OBJ_TYPE_VFSMOUNT: - return fsnotify_conn_mount(conn)->mnt.mnt_sb; - case FSNOTIFY_OBJ_TYPE_SB: - return fsnotify_conn_sb(conn); - default: - return NULL; - } + return fsnotify_object_sb(conn->obj, conn->type); } /* destroy all events sitting in this groups notification queue */ diff --git a/fs/notify/mark.c b/fs/notify/mark.c index d9977af25832..1f611e4eab53 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -265,6 +265,7 @@ static void *fsnotify_detach_connector_from_object( struct fsnotify_mark_connector *conn, unsigned int *type) { + fsnotify_connp_t *connp = fsnotify_object_connp(conn->obj, conn->type); struct inode *inode = NULL; *type = conn->type; @@ -285,7 +286,7 @@ static void *fsnotify_detach_connector_from_object( } fsnotify_put_sb_watchers(conn); - rcu_assign_pointer(*(conn->obj), NULL); + rcu_assign_pointer(*connp, NULL); conn->obj = NULL; conn->type = FSNOTIFY_OBJ_TYPE_DETACHED; @@ -560,7 +561,7 @@ int fsnotify_compare_groups(struct fsnotify_group *a, struct fsnotify_group *b) } static int fsnotify_attach_connector_to_object(fsnotify_connp_t *connp, - unsigned int obj_type) + void *obj, unsigned int obj_type) { struct fsnotify_mark_connector *conn; @@ -571,7 +572,7 @@ static int fsnotify_attach_connector_to_object(fsnotify_connp_t *connp, INIT_HLIST_HEAD(&conn->list); conn->flags = 0; conn->type = obj_type; - conn->obj = connp; + conn->obj = obj; /* * cmpxchg() provides the barrier so that readers of *connp can see @@ -620,24 +621,25 @@ out: * to which group and for which inodes. These marks are ordered according to * priority, highest number first, and then by the group's location in memory. */ -static int fsnotify_add_mark_list(struct fsnotify_mark *mark, - fsnotify_connp_t *connp, +static int fsnotify_add_mark_list(struct fsnotify_mark *mark, void *obj, unsigned int obj_type, int add_flags) { struct fsnotify_mark *lmark, *last = NULL; struct fsnotify_mark_connector *conn; + fsnotify_connp_t *connp; int cmp; int err = 0; if (WARN_ON(!fsnotify_valid_obj_type(obj_type))) return -EINVAL; + connp = fsnotify_object_connp(obj, obj_type); restart: spin_lock(&mark->lock); conn = fsnotify_grab_connector(connp); if (!conn) { spin_unlock(&mark->lock); - err = fsnotify_attach_connector_to_object(connp, obj_type); + err = fsnotify_attach_connector_to_object(connp, obj, obj_type); if (err) return err; goto restart; @@ -689,7 +691,7 @@ out_err: * event types should be delivered to which group. */ int fsnotify_add_mark_locked(struct fsnotify_mark *mark, - fsnotify_connp_t *connp, unsigned int obj_type, + void *obj, unsigned int obj_type, int add_flags) { struct fsnotify_group *group = mark->group; @@ -710,7 +712,7 @@ int fsnotify_add_mark_locked(struct fsnotify_mark *mark, fsnotify_get_mark(mark); /* for g_list */ spin_unlock(&mark->lock); - ret = fsnotify_add_mark_list(mark, connp, obj_type, add_flags); + ret = fsnotify_add_mark_list(mark, obj, obj_type, add_flags); if (ret) goto err; @@ -728,14 +730,14 @@ err: return ret; } -int fsnotify_add_mark(struct fsnotify_mark *mark, fsnotify_connp_t *connp, +int fsnotify_add_mark(struct fsnotify_mark *mark, void *obj, unsigned int obj_type, int add_flags) { int ret; struct fsnotify_group *group = mark->group; fsnotify_group_lock(group); - ret = fsnotify_add_mark_locked(mark, connp, obj_type, add_flags); + ret = fsnotify_add_mark_locked(mark, obj, obj_type, add_flags); fsnotify_group_unlock(group); return ret; } @@ -745,12 +747,16 @@ EXPORT_SYMBOL_GPL(fsnotify_add_mark); * Given a list of marks, find the mark associated with given group. If found * take a reference to that mark and return it, else return NULL. */ -struct fsnotify_mark *fsnotify_find_mark(fsnotify_connp_t *connp, +struct fsnotify_mark *fsnotify_find_mark(void *obj, unsigned int obj_type, struct fsnotify_group *group) { + fsnotify_connp_t *connp = fsnotify_object_connp(obj, obj_type); struct fsnotify_mark_connector *conn; struct fsnotify_mark *mark; + if (!connp) + return NULL; + conn = fsnotify_grab_connector(connp); if (!conn) return NULL; diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index 992b57a7e95f..face68fcf850 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -456,13 +456,6 @@ FSNOTIFY_ITER_FUNCS(sb, SB) type < FSNOTIFY_ITER_TYPE_COUNT; \ type++) -/* - * fsnotify_connp_t is what we embed in objects which connector can be attached - * to. fsnotify_connp_t * is how we refer from connector back to object. - */ -struct fsnotify_mark_connector; -typedef struct fsnotify_mark_connector __rcu *fsnotify_connp_t; - /* * Inode/vfsmount/sb point to this structure which tracks all marks attached to * the inode/vfsmount/sb. The reference to inode/vfsmount/sb is held by this @@ -476,7 +469,7 @@ struct fsnotify_mark_connector { unsigned short flags; /* flags [lock] */ union { /* Object pointer [lock] */ - fsnotify_connp_t *obj; + void *obj; /* Used listing heads to free after srcu period expires */ struct fsnotify_mark_connector *destroy_next; }; @@ -763,37 +756,35 @@ extern void fsnotify_recalc_mask(struct fsnotify_mark_connector *conn); extern void fsnotify_init_mark(struct fsnotify_mark *mark, struct fsnotify_group *group); /* Find mark belonging to given group in the list of marks */ -extern struct fsnotify_mark *fsnotify_find_mark(fsnotify_connp_t *connp, - struct fsnotify_group *group); +struct fsnotify_mark *fsnotify_find_mark(void *obj, unsigned int obj_type, + struct fsnotify_group *group); /* attach the mark to the object */ -extern int fsnotify_add_mark(struct fsnotify_mark *mark, - fsnotify_connp_t *connp, unsigned int obj_type, - int add_flags); -extern int fsnotify_add_mark_locked(struct fsnotify_mark *mark, - fsnotify_connp_t *connp, - unsigned int obj_type, int add_flags); +int fsnotify_add_mark(struct fsnotify_mark *mark, void *obj, + unsigned int obj_type, int add_flags); +int fsnotify_add_mark_locked(struct fsnotify_mark *mark, void *obj, + unsigned int obj_type, int add_flags); /* attach the mark to the inode */ static inline int fsnotify_add_inode_mark(struct fsnotify_mark *mark, struct inode *inode, int add_flags) { - return fsnotify_add_mark(mark, &inode->i_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_INODE, add_flags); + return fsnotify_add_mark(mark, inode, FSNOTIFY_OBJ_TYPE_INODE, + add_flags); } static inline int fsnotify_add_inode_mark_locked(struct fsnotify_mark *mark, struct inode *inode, int add_flags) { - return fsnotify_add_mark_locked(mark, &inode->i_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_INODE, add_flags); + return fsnotify_add_mark_locked(mark, inode, FSNOTIFY_OBJ_TYPE_INODE, + add_flags); } static inline struct fsnotify_mark *fsnotify_find_inode_mark( struct inode *inode, struct fsnotify_group *group) { - return fsnotify_find_mark(&inode->i_fsnotify_marks, group); + return fsnotify_find_mark(inode, FSNOTIFY_OBJ_TYPE_INODE, group); } /* given a group and a mark, flag mark to be freed when all references are dropped */ -- cgit v1.2.3-59-g8ed1b From c9d4603b054f9a63c07c7012040af4c80adb2c60 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:50 +0200 Subject: fsnotify: create helper fsnotify_update_sb_watchers() We would like to count watched objects by priority group, so we will need to update the watched object counter after adding/removing marks. Create a helper fsnotify_update_sb_watchers() and call it after attaching/detaching a mark, instead of fsnotify_{get,put}_sb_watchers() only after attaching/detaching a connector. Soon, we will use this helper to count watched objects by the highest watching priority group. Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-7-amir73il@gmail.com> --- fs/notify/mark.c | 36 ++++++++++++++++++++++-------------- include/linux/fsnotify_backend.h | 1 + 2 files changed, 23 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/notify/mark.c b/fs/notify/mark.c index 1f611e4eab53..647c49c20467 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -154,20 +154,23 @@ static void fsnotify_put_inode_ref(struct inode *inode) iput(inode); } -static void fsnotify_get_sb_watchers(struct fsnotify_mark_connector *conn) +/* + * Grab or drop watched objects reference depending on whether the connector + * is attached and has any marks attached. + */ +static void fsnotify_update_sb_watchers(struct super_block *sb, + struct fsnotify_mark_connector *conn) { - struct super_block *sb = fsnotify_connector_sb(conn); + bool is_watched = conn->flags & FSNOTIFY_CONN_FLAG_IS_WATCHED; + bool has_marks = conn->obj && !hlist_empty(&conn->list); - if (sb) + if (has_marks && !is_watched) { + conn->flags |= FSNOTIFY_CONN_FLAG_IS_WATCHED; fsnotify_get_sb_watched_objects(sb); -} - -static void fsnotify_put_sb_watchers(struct fsnotify_mark_connector *conn) -{ - struct super_block *sb = fsnotify_connector_sb(conn); - - if (sb) + } else if (!has_marks && is_watched) { + conn->flags &= ~FSNOTIFY_CONN_FLAG_IS_WATCHED; fsnotify_put_sb_watched_objects(sb); + } } /* @@ -266,6 +269,7 @@ static void *fsnotify_detach_connector_from_object( unsigned int *type) { fsnotify_connp_t *connp = fsnotify_object_connp(conn->obj, conn->type); + struct super_block *sb = fsnotify_connector_sb(conn); struct inode *inode = NULL; *type = conn->type; @@ -285,10 +289,10 @@ static void *fsnotify_detach_connector_from_object( fsnotify_conn_sb(conn)->s_fsnotify_mask = 0; } - fsnotify_put_sb_watchers(conn); rcu_assign_pointer(*connp, NULL); conn->obj = NULL; conn->type = FSNOTIFY_OBJ_TYPE_DETACHED; + fsnotify_update_sb_watchers(sb, conn); return inode; } @@ -340,6 +344,11 @@ void fsnotify_put_mark(struct fsnotify_mark *mark) objp = fsnotify_detach_connector_from_object(conn, &type); free_conn = true; } else { + struct super_block *sb = fsnotify_connector_sb(conn); + + /* Update watched objects after detaching mark */ + if (sb) + fsnotify_update_sb_watchers(sb, conn); objp = __fsnotify_recalc_mask(conn); type = conn->type; } @@ -581,10 +590,7 @@ static int fsnotify_attach_connector_to_object(fsnotify_connp_t *connp, if (cmpxchg(connp, NULL, conn)) { /* Someone else created list structure for us */ kmem_cache_free(fsnotify_mark_connector_cachep, conn); - return 0; } - - fsnotify_get_sb_watchers(conn); return 0; } @@ -624,6 +630,7 @@ out: static int fsnotify_add_mark_list(struct fsnotify_mark *mark, void *obj, unsigned int obj_type, int add_flags) { + struct super_block *sb = fsnotify_object_sb(obj, obj_type); struct fsnotify_mark *lmark, *last = NULL; struct fsnotify_mark_connector *conn; fsnotify_connp_t *connp; @@ -673,6 +680,7 @@ restart: /* mark should be the last entry. last is the current last entry */ hlist_add_behind_rcu(&mark->obj_list, &last->obj_list); added: + fsnotify_update_sb_watchers(sb, conn); /* * Since connector is attached to object using cmpxchg() we are * guaranteed that connector initialization is fully visible by anyone diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index face68fcf850..83004d9e07a3 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -465,6 +465,7 @@ FSNOTIFY_ITER_FUNCS(sb, SB) struct fsnotify_mark_connector { spinlock_t lock; unsigned short type; /* Type of object [lock] */ +#define FSNOTIFY_CONN_FLAG_IS_WATCHED 0x01 #define FSNOTIFY_CONN_FLAG_HAS_IREF 0x02 unsigned short flags; /* flags [lock] */ union { -- cgit v1.2.3-59-g8ed1b From 07a3b8d0bf726a1e49b050bbc6bd72f031e505fe Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:51 +0200 Subject: fsnotify: lazy attach fsnotify_sb_info state to sb Define a container struct fsnotify_sb_info to hold per-sb state, including the reference to sb marks connector. Allocate the fsnotify_sb_info state before attaching connector to any object on the sb and free it only when killing sb. This state is going to be used for storing per priority watched objects counters. Suggested-by: Jan Kara Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-8-amir73il@gmail.com> --- fs/notify/fsnotify.c | 16 +++++++++++++--- fs/notify/fsnotify.h | 9 ++++++++- fs/notify/mark.c | 32 +++++++++++++++++++++++++++++++- include/linux/fs.h | 8 ++++---- include/linux/fsnotify_backend.h | 17 +++++++++++++++++ 5 files changed, 73 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 503e7c75e777..fb3f36bc6ea9 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -89,11 +89,18 @@ static void fsnotify_unmount_inodes(struct super_block *sb) void fsnotify_sb_delete(struct super_block *sb) { + struct fsnotify_sb_info *sbinfo = fsnotify_sb_info(sb); + + /* Were any marks ever added to any object on this sb? */ + if (!sbinfo) + return; + fsnotify_unmount_inodes(sb); fsnotify_clear_marks_by_sb(sb); /* Wait for outstanding object references from connectors */ wait_var_event(fsnotify_sb_watched_objects(sb), !atomic_long_read(fsnotify_sb_watched_objects(sb))); + kfree(sbinfo); } /* @@ -489,6 +496,7 @@ int fsnotify(__u32 mask, const void *data, int data_type, struct inode *dir, { const struct path *path = fsnotify_data_path(data, data_type); struct super_block *sb = fsnotify_data_sb(data, data_type); + struct fsnotify_sb_info *sbinfo = fsnotify_sb_info(sb); struct fsnotify_iter_info iter_info = {}; struct mount *mnt = NULL; struct inode *inode2 = NULL; @@ -525,7 +533,7 @@ int fsnotify(__u32 mask, const void *data, int data_type, struct inode *dir, * SRCU because we have no references to any objects and do not * need SRCU to keep them "alive". */ - if (!sb->s_fsnotify_marks && + if ((!sbinfo || !sbinfo->sb_marks) && (!mnt || !mnt->mnt_fsnotify_marks) && (!inode || !inode->i_fsnotify_marks) && (!inode2 || !inode2->i_fsnotify_marks)) @@ -552,8 +560,10 @@ int fsnotify(__u32 mask, const void *data, int data_type, struct inode *dir, iter_info.srcu_idx = srcu_read_lock(&fsnotify_mark_srcu); - iter_info.marks[FSNOTIFY_ITER_TYPE_SB] = - fsnotify_first_mark(&sb->s_fsnotify_marks); + if (sbinfo) { + iter_info.marks[FSNOTIFY_ITER_TYPE_SB] = + fsnotify_first_mark(&sbinfo->sb_marks); + } if (mnt) { iter_info.marks[FSNOTIFY_ITER_TYPE_VFSMOUNT] = fsnotify_first_mark(&mnt->mnt_fsnotify_marks); diff --git a/fs/notify/fsnotify.h b/fs/notify/fsnotify.h index b1fec7cd8ee4..2d059f789ee3 100644 --- a/fs/notify/fsnotify.h +++ b/fs/notify/fsnotify.h @@ -54,6 +54,13 @@ static inline struct super_block *fsnotify_connector_sb( return fsnotify_object_sb(conn->obj, conn->type); } +static inline fsnotify_connp_t *fsnotify_sb_marks(struct super_block *sb) +{ + struct fsnotify_sb_info *sbinfo = fsnotify_sb_info(sb); + + return sbinfo ? &sbinfo->sb_marks : NULL; +} + /* destroy all events sitting in this groups notification queue */ extern void fsnotify_flush_notify(struct fsnotify_group *group); @@ -79,7 +86,7 @@ static inline void fsnotify_clear_marks_by_mount(struct vfsmount *mnt) /* run the list of all marks associated with sb and destroy them */ static inline void fsnotify_clear_marks_by_sb(struct super_block *sb) { - fsnotify_destroy_marks(&sb->s_fsnotify_marks); + fsnotify_destroy_marks(fsnotify_sb_marks(sb)); } /* diff --git a/fs/notify/mark.c b/fs/notify/mark.c index 647c49c20467..b2f5d8c9cce1 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -106,7 +106,7 @@ static fsnotify_connp_t *fsnotify_object_connp(void *obj, case FSNOTIFY_OBJ_TYPE_VFSMOUNT: return &real_mount(obj)->mnt_fsnotify_marks; case FSNOTIFY_OBJ_TYPE_SB: - return &((struct super_block *)obj)->s_fsnotify_marks; + return fsnotify_sb_marks(obj); default: return NULL; } @@ -569,6 +569,26 @@ int fsnotify_compare_groups(struct fsnotify_group *a, struct fsnotify_group *b) return -1; } +static int fsnotify_attach_info_to_sb(struct super_block *sb) +{ + struct fsnotify_sb_info *sbinfo; + + /* sb info is freed on fsnotify_sb_delete() */ + sbinfo = kzalloc(sizeof(*sbinfo), GFP_KERNEL); + if (!sbinfo) + return -ENOMEM; + + /* + * cmpxchg() provides the barrier so that callers of fsnotify_sb_info() + * will observe an initialized structure + */ + if (cmpxchg(&sb->s_fsnotify_info, NULL, sbinfo)) { + /* Someone else created sbinfo for us */ + kfree(sbinfo); + } + return 0; +} + static int fsnotify_attach_connector_to_object(fsnotify_connp_t *connp, void *obj, unsigned int obj_type) { @@ -640,6 +660,16 @@ static int fsnotify_add_mark_list(struct fsnotify_mark *mark, void *obj, if (WARN_ON(!fsnotify_valid_obj_type(obj_type))) return -EINVAL; + /* + * Attach the sb info before attaching a connector to any object on sb. + * The sb info will remain attached as long as sb lives. + */ + if (!fsnotify_sb_info(sb)) { + err = fsnotify_attach_info_to_sb(sb); + if (err) + return err; + } + connp = fsnotify_object_connp(obj, obj_type); restart: spin_lock(&mark->lock); diff --git a/include/linux/fs.h b/include/linux/fs.h index 00fc429b0af0..7f40b592f711 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -73,6 +73,8 @@ struct fscrypt_inode_info; struct fscrypt_operations; struct fsverity_info; struct fsverity_operations; +struct fsnotify_mark_connector; +struct fsnotify_sb_info; struct fs_context; struct fs_parameter_spec; struct fileattr; @@ -620,8 +622,6 @@ is_uncached_acl(struct posix_acl *acl) #define IOP_XATTR 0x0008 #define IOP_DEFAULT_READLINK 0x0010 -struct fsnotify_mark_connector; - /* * Keep mostly read-only and often accessed (especially for * the RCU path lookup and 'stat' data) fields at the beginning @@ -1249,7 +1249,7 @@ struct super_block { /* * Keep s_fs_info, s_time_gran, s_fsnotify_mask, and - * s_fsnotify_marks together for cache efficiency. They are frequently + * s_fsnotify_info together for cache efficiency. They are frequently * accessed and rarely modified. */ void *s_fs_info; /* Filesystem private info */ @@ -1261,7 +1261,7 @@ struct super_block { time64_t s_time_max; #ifdef CONFIG_FSNOTIFY __u32 s_fsnotify_mask; - struct fsnotify_mark_connector __rcu *s_fsnotify_marks; + struct fsnotify_sb_info *s_fsnotify_info; #endif /* diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index 83004d9e07a3..c9f2b2f6b493 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -477,6 +477,23 @@ struct fsnotify_mark_connector { struct hlist_head list; }; +/* + * Container for per-sb fsnotify state (sb marks and more). + * Attached lazily on first marked object on the sb and freed when killing sb. + */ +struct fsnotify_sb_info { + struct fsnotify_mark_connector __rcu *sb_marks; +}; + +static inline struct fsnotify_sb_info *fsnotify_sb_info(struct super_block *sb) +{ +#ifdef CONFIG_FSNOTIFY + return READ_ONCE(sb->s_fsnotify_info); +#else + return NULL; +#endif +} + static inline atomic_long_t *fsnotify_sb_watched_objects(struct super_block *sb) { return &sb->s_fsnotify_connectors; -- cgit v1.2.3-59-g8ed1b From 477cf917dd02853ba78a73cdeb6548889e5f8cd7 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:53 +0200 Subject: fsnotify: use an enum for group priority constants And use meaningfull names for the constants. Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-10-amir73il@gmail.com> --- fs/notify/fanotify/fanotify_user.c | 11 +++++------ include/linux/fsnotify_backend.h | 20 ++++++++++++-------- 2 files changed, 17 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 925015ee7fbd..483a6a1255fb 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -1516,13 +1516,13 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) INIT_LIST_HEAD(&group->fanotify_data.access_list); switch (class) { case FAN_CLASS_NOTIF: - group->priority = FS_PRIO_0; + group->priority = FSNOTIFY_PRIO_NORMAL; break; case FAN_CLASS_CONTENT: - group->priority = FS_PRIO_1; + group->priority = FSNOTIFY_PRIO_CONTENT; break; case FAN_CLASS_PRE_CONTENT: - group->priority = FS_PRIO_2; + group->priority = FSNOTIFY_PRIO_PRE_CONTENT; break; default: fd = -EINVAL; @@ -1774,12 +1774,11 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, goto fput_and_out; /* - * group->priority == FS_PRIO_0 == FAN_CLASS_NOTIF. These are not - * allowed to set permissions events. + * Permission events require minimum priority FAN_CLASS_CONTENT. */ ret = -EINVAL; if (mask & FANOTIFY_PERM_EVENTS && - group->priority == FS_PRIO_0) + group->priority < FSNOTIFY_PRIO_CONTENT) goto fput_and_out; if (mask & FAN_FS_ERROR && diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index ec592aeadfa3..fc38587d8564 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -176,6 +176,17 @@ struct fsnotify_event { struct list_head list; }; +/* + * fsnotify group priorities. + * Events are sent in order from highest priority to lowest priority. + */ +enum fsnotify_group_prio { + FSNOTIFY_PRIO_NORMAL = 0, /* normal notifiers, no permissions */ + FSNOTIFY_PRIO_CONTENT, /* fanotify permission events */ + FSNOTIFY_PRIO_PRE_CONTENT, /* fanotify pre-content events */ + __FSNOTIFY_PRIO_NUM +}; + /* * A group is a "thing" that wants to receive notification about filesystem * events. The mask holds the subset of event types this group cares about. @@ -201,14 +212,7 @@ struct fsnotify_group { wait_queue_head_t notification_waitq; /* read() on the notification file blocks on this waitq */ unsigned int q_len; /* events on the queue */ unsigned int max_events; /* maximum events allowed on the list */ - /* - * Valid fsnotify group priorities. Events are send in order from highest - * priority to lowest priority. We default to the lowest priority. - */ - #define FS_PRIO_0 0 /* normal notifiers, no permissions */ - #define FS_PRIO_1 1 /* fanotify content based access control */ - #define FS_PRIO_2 2 /* fanotify pre-content access */ - unsigned int priority; + enum fsnotify_group_prio priority; /* priority for sending events */ bool shutdown; /* group is being shut down, don't queue more events */ #define FSNOTIFY_GROUP_USER 0x01 /* user allocated group */ -- cgit v1.2.3-59-g8ed1b From a5e57b4d370c6d320e5bfb0c919fe00aee29e039 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 17 Mar 2024 20:41:54 +0200 Subject: fsnotify: optimize the case of no permission event watchers Commit e43de7f0862b ("fsnotify: optimize the case of no marks of any type") optimized the case where there are no fsnotify watchers on any of the filesystem's objects. It is quite common for a system to have a single local filesystem and it is quite common for the system to have some inotify watches on some config files or directories, so the optimization of no marks at all is often not in effect. Permission event watchers, which require high priority group are more rare, so optimizing the case of no marks og high priority groups can improve performance for more systems, especially for performance sensitive io workloads. Count per-sb watched objects by high priority groups and use that the optimize out the call to __fsnotify_parent() and fsnotify() in fsnotify permission hooks. Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240317184154.1200192-11-amir73il@gmail.com> --- fs/notify/fsnotify.c | 3 +++ fs/notify/mark.c | 30 +++++++++++++++++++++++++++--- include/linux/fsnotify.h | 19 ++++++++++++++++--- include/linux/fsnotify_backend.h | 11 ++++++++--- 4 files changed, 54 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index fb3f36bc6ea9..2ae965ef37e8 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -100,6 +100,9 @@ void fsnotify_sb_delete(struct super_block *sb) /* Wait for outstanding object references from connectors */ wait_var_event(fsnotify_sb_watched_objects(sb), !atomic_long_read(fsnotify_sb_watched_objects(sb))); + WARN_ON(fsnotify_sb_has_priority_watchers(sb, FSNOTIFY_PRIO_CONTENT)); + WARN_ON(fsnotify_sb_has_priority_watchers(sb, + FSNOTIFY_PRIO_PRE_CONTENT)); kfree(sbinfo); } diff --git a/fs/notify/mark.c b/fs/notify/mark.c index b2f5d8c9cce1..c3eefa70633c 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -161,13 +161,36 @@ static void fsnotify_put_inode_ref(struct inode *inode) static void fsnotify_update_sb_watchers(struct super_block *sb, struct fsnotify_mark_connector *conn) { + struct fsnotify_sb_info *sbinfo = fsnotify_sb_info(sb); bool is_watched = conn->flags & FSNOTIFY_CONN_FLAG_IS_WATCHED; - bool has_marks = conn->obj && !hlist_empty(&conn->list); + struct fsnotify_mark *first_mark = NULL; + unsigned int highest_prio = 0; - if (has_marks && !is_watched) { + if (conn->obj) + first_mark = hlist_entry_safe(conn->list.first, + struct fsnotify_mark, obj_list); + if (first_mark) + highest_prio = first_mark->group->priority; + if (WARN_ON(highest_prio >= __FSNOTIFY_PRIO_NUM)) + highest_prio = 0; + + /* + * If the highest priority of group watching this object is prio, + * then watched object has a reference on counters [0..prio]. + * Update priority >= 1 watched objects counters. + */ + for (unsigned int p = conn->prio + 1; p <= highest_prio; p++) + atomic_long_inc(&sbinfo->watched_objects[p]); + for (unsigned int p = conn->prio; p > highest_prio; p--) + atomic_long_dec(&sbinfo->watched_objects[p]); + conn->prio = highest_prio; + + /* Update priority >= 0 (a.k.a total) watched objects counter */ + BUILD_BUG_ON(FSNOTIFY_PRIO_NORMAL != 0); + if (first_mark && !is_watched) { conn->flags |= FSNOTIFY_CONN_FLAG_IS_WATCHED; fsnotify_get_sb_watched_objects(sb); - } else if (!has_marks && is_watched) { + } else if (!first_mark && is_watched) { conn->flags &= ~FSNOTIFY_CONN_FLAG_IS_WATCHED; fsnotify_put_sb_watched_objects(sb); } @@ -600,6 +623,7 @@ static int fsnotify_attach_connector_to_object(fsnotify_connp_t *connp, spin_lock_init(&conn->lock); INIT_HLIST_HEAD(&conn->list); conn->flags = 0; + conn->prio = 0; conn->type = obj_type; conn->obj = obj; diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 48dc65702415..4da80e92f804 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -17,8 +17,9 @@ #include #include -/* Are there any inode/mount/sb objects that are being watched at all? */ -static inline bool fsnotify_sb_has_watchers(struct super_block *sb) +/* Are there any inode/mount/sb objects watched with priority prio or above? */ +static inline bool fsnotify_sb_has_priority_watchers(struct super_block *sb, + int prio) { struct fsnotify_sb_info *sbinfo = fsnotify_sb_info(sb); @@ -26,7 +27,13 @@ static inline bool fsnotify_sb_has_watchers(struct super_block *sb) if (!sbinfo) return false; - return atomic_long_read(&sbinfo->watched_objects); + return atomic_long_read(&sbinfo->watched_objects[prio]); +} + +/* Are there any inode/mount/sb objects that are being watched at all? */ +static inline bool fsnotify_sb_has_watchers(struct super_block *sb) +{ + return fsnotify_sb_has_priority_watchers(sb, 0); } /* @@ -109,6 +116,12 @@ static inline int fsnotify_file(struct file *file, __u32 mask) return 0; path = &file->f_path; + /* Permission events require group prio >= FSNOTIFY_PRIO_CONTENT */ + if (mask & ALL_FSNOTIFY_PERM_EVENTS && + !fsnotify_sb_has_priority_watchers(path->dentry->d_sb, + FSNOTIFY_PRIO_CONTENT)) + return 0; + return fsnotify_parent(path->dentry, mask, path, FSNOTIFY_EVENT_PATH); } diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index fc38587d8564..7f1ab8264e41 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -468,7 +468,8 @@ FSNOTIFY_ITER_FUNCS(sb, SB) */ struct fsnotify_mark_connector { spinlock_t lock; - unsigned short type; /* Type of object [lock] */ + unsigned char type; /* Type of object [lock] */ + unsigned char prio; /* Highest priority group */ #define FSNOTIFY_CONN_FLAG_IS_WATCHED 0x01 #define FSNOTIFY_CONN_FLAG_HAS_IREF 0x02 unsigned short flags; /* flags [lock] */ @@ -490,8 +491,12 @@ struct fsnotify_sb_info { /* * Number of inode/mount/sb objects that are being watched in this sb. * Note that inodes objects are currently double-accounted. + * + * The value in watched_objects[prio] is the number of objects that are + * watched by groups of priority >= prio, so watched_objects[0] is the + * total number of watched objects in this sb. */ - atomic_long_t watched_objects; + atomic_long_t watched_objects[__FSNOTIFY_PRIO_NUM]; }; static inline struct fsnotify_sb_info *fsnotify_sb_info(struct super_block *sb) @@ -505,7 +510,7 @@ static inline struct fsnotify_sb_info *fsnotify_sb_info(struct super_block *sb) static inline atomic_long_t *fsnotify_sb_watched_objects(struct super_block *sb) { - return &fsnotify_sb_info(sb)->watched_objects; + return &fsnotify_sb_info(sb)->watched_objects[0]; } /* -- cgit v1.2.3-59-g8ed1b From 3e90417f41f41a7d4a6d5ee0ca216698e9ab4381 Mon Sep 17 00:00:00 2001 From: "Ritesh Harjani (IBM)" Date: Thu, 29 Feb 2024 19:54:13 +0530 Subject: ext2: set FMODE_CAN_ODIRECT instead of a dummy direct_IO method Since commit a2ad63daa88b ("VFS: add FMODE_CAN_ODIRECT file flag") file systems can just set the FMODE_CAN_ODIRECT flag at open time instead of wiring up a dummy direct_IO method to indicate support for direct I/O. Signed-off-by: Ritesh Harjani (IBM) Signed-off-by: Jan Kara Message-Id: <94f78492f55c3f421359fb6e0d8fab6e79ea17b2.1709215665.git.ritesh.list@gmail.com> --- fs/ext2/file.c | 8 +++++++- fs/ext2/inode.c | 2 -- 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/ext2/file.c b/fs/ext2/file.c index 4ddc36f4dbd4..10b061ac5bc0 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -302,6 +302,12 @@ static ssize_t ext2_file_write_iter(struct kiocb *iocb, struct iov_iter *from) return generic_file_write_iter(iocb, from); } +static int ext2_file_open(struct inode *inode, struct file *filp) +{ + filp->f_mode |= FMODE_CAN_ODIRECT; + return dquot_file_open(inode, filp); +} + const struct file_operations ext2_file_operations = { .llseek = generic_file_llseek, .read_iter = ext2_file_read_iter, @@ -311,7 +317,7 @@ const struct file_operations ext2_file_operations = { .compat_ioctl = ext2_compat_ioctl, #endif .mmap = ext2_file_mmap, - .open = dquot_file_open, + .open = ext2_file_open, .release = ext2_release_file, .fsync = ext2_fsync, .get_unmapped_area = thp_get_unmapped_area, diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index f3d570a9302b..0caa1650cee8 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -965,7 +965,6 @@ const struct address_space_operations ext2_aops = { .write_begin = ext2_write_begin, .write_end = ext2_write_end, .bmap = ext2_bmap, - .direct_IO = noop_direct_IO, .writepages = ext2_writepages, .migrate_folio = buffer_migrate_folio, .is_partially_uptodate = block_is_partially_uptodate, @@ -974,7 +973,6 @@ const struct address_space_operations ext2_aops = { static const struct address_space_operations ext2_dax_aops = { .writepages = ext2_dax_writepages, - .direct_IO = noop_direct_IO, .dirty_folio = noop_dirty_folio, }; -- cgit v1.2.3-59-g8ed1b From 01e5f4fc0fead3ef19000e1d2fc748e87aac3f02 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 4 Apr 2024 15:50:26 -0400 Subject: bcachefs: Make snapshot_is_ancestor() safe Snapshot table accesses generally need to be checking for invalid snapshot ID now, fix one that was missed. Signed-off-by: Kent Overstreet --- fs/bcachefs/snapshot.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 0e806f04f3d7..544322d5c251 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -125,6 +125,15 @@ static inline u32 get_ancestor_below(struct snapshot_table *t, u32 id, u32 ances return s->parent; } +static bool test_ancestor_bitmap(struct snapshot_table *t, u32 id, u32 ancestor) +{ + const struct snapshot_t *s = __snapshot_t(t, id); + if (!s) + return false; + + return test_bit(ancestor - id - 1, s->is_ancestor); +} + bool __bch2_snapshot_is_ancestor(struct bch_fs *c, u32 id, u32 ancestor) { bool ret; @@ -140,13 +149,11 @@ bool __bch2_snapshot_is_ancestor(struct bch_fs *c, u32 id, u32 ancestor) while (id && id < ancestor - IS_ANCESTOR_BITMAP) id = get_ancestor_below(t, id, ancestor); - if (id && id < ancestor) { - ret = test_bit(ancestor - id - 1, __snapshot_t(t, id)->is_ancestor); + ret = id && id < ancestor + ? test_ancestor_bitmap(t, id, ancestor) + : id == ancestor; - EBUG_ON(ret != __bch2_snapshot_is_ancestor_early(t, id, ancestor)); - } else { - ret = id == ancestor; - } + EBUG_ON(ret != __bch2_snapshot_is_ancestor_early(t, id, ancestor)); out: rcu_read_unlock(); -- cgit v1.2.3-59-g8ed1b From be42e4a621fee05e3299169fbb1068b473e779c2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 4 Apr 2024 16:51:40 -0400 Subject: bcachefs: Bump limit in btree_trans_too_many_iters() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.h | 2 +- fs/bcachefs/btree_types.h | 14 ++++++++++++++ 2 files changed, 15 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 24772538e4cc..1d58d447b386 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -642,7 +642,7 @@ int __bch2_btree_trans_too_many_iters(struct btree_trans *); static inline int btree_trans_too_many_iters(struct btree_trans *trans) { - if (bitmap_weight(trans->paths_allocated, trans->nr_paths) > BTREE_ITER_INITIAL - 8) + if (bitmap_weight(trans->paths_allocated, trans->nr_paths) > BTREE_ITER_NORMAL_LIMIT - 8) return __bch2_btree_trans_too_many_iters(trans); return 0; diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index 9404d96c38f3..e0c982a4195c 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -364,7 +364,21 @@ struct btree_insert_entry { unsigned long ip_allocated; }; +/* Number of btree paths we preallocate, usually enough */ #define BTREE_ITER_INITIAL 64 +/* + * Lmiit for btree_trans_too_many_iters(); this is enough that almost all code + * paths should run inside this limit, and if they don't it usually indicates a + * bug (leaking/duplicated btree paths). + * + * exception: some fsck paths + * + * bugs with excessive path usage seem to have possibly been eliminated now, so + * we might consider eliminating this (and btree_trans_too_many_iter()) at some + * point. + */ +#define BTREE_ITER_NORMAL_LIMIT 256 +/* never exceed limit */ #define BTREE_ITER_MAX (1U << 10) struct btree_trans_commit_hook; -- cgit v1.2.3-59-g8ed1b From 9fb3036fe3d9414ae32a97d01d7ccf7550e168a7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 3 Apr 2024 19:15:53 -0400 Subject: bcachefs: Move btree_updates to debugfs sysfs is limited to PAGE_SIZE, and when we're debugging strange deadlocks/priority inversions we need to see the full list. Signed-off-by: Kent Overstreet --- fs/bcachefs/debug.c | 45 ++++++++++++++++++++++++++++++++++++++++++--- fs/bcachefs/sysfs.c | 6 ------ 2 files changed, 42 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index 7b681ae91510..cd99b7399414 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -13,6 +13,7 @@ #include "btree_iter.h" #include "btree_locking.h" #include "btree_update.h" +#include "btree_update_interior.h" #include "buckets.h" #include "debug.h" #include "error.h" @@ -668,7 +669,7 @@ static ssize_t bch2_journal_pins_read(struct file *file, char __user *buf, i->size = size; i->ret = 0; - do { + while (1) { err = flush_buf(i); if (err) return err; @@ -676,9 +677,12 @@ static ssize_t bch2_journal_pins_read(struct file *file, char __user *buf, if (!i->size) break; + if (done) + break; + done = bch2_journal_seq_pins_to_text(&i->buf, &c->journal, &i->iter); i->iter++; - } while (!done); + } if (i->buf.allocation_failure) return -ENOMEM; @@ -693,13 +697,45 @@ static const struct file_operations journal_pins_ops = { .read = bch2_journal_pins_read, }; +static ssize_t bch2_btree_updates_read(struct file *file, char __user *buf, + size_t size, loff_t *ppos) +{ + struct dump_iter *i = file->private_data; + struct bch_fs *c = i->c; + int err; + + i->ubuf = buf; + i->size = size; + i->ret = 0; + + if (!i->iter) { + bch2_btree_updates_to_text(&i->buf, c); + i->iter++; + } + + err = flush_buf(i); + if (err) + return err; + + if (i->buf.allocation_failure) + return -ENOMEM; + + return i->ret; +} + +static const struct file_operations btree_updates_ops = { + .owner = THIS_MODULE, + .open = bch2_dump_open, + .release = bch2_dump_release, + .read = bch2_btree_updates_read, +}; + static int btree_transaction_stats_open(struct inode *inode, struct file *file) { struct bch_fs *c = inode->i_private; struct dump_iter *i; i = kzalloc(sizeof(struct dump_iter), GFP_KERNEL); - if (!i) return -ENOMEM; @@ -902,6 +938,9 @@ void bch2_fs_debug_init(struct bch_fs *c) debugfs_create_file("journal_pins", 0400, c->fs_debug_dir, c->btree_debug, &journal_pins_ops); + debugfs_create_file("btree_updates", 0400, c->fs_debug_dir, + c->btree_debug, &btree_updates_ops); + debugfs_create_file("btree_transaction_stats", 0400, c->fs_debug_dir, c, &btree_transaction_stats_op); diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index c86a93a8d8fc..b18b0cc81b59 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -17,7 +17,6 @@ #include "btree_iter.h" #include "btree_key_cache.h" #include "btree_update.h" -#include "btree_update_interior.h" #include "btree_gc.h" #include "buckets.h" #include "clock.h" @@ -166,7 +165,6 @@ read_attribute(btree_write_stats); read_attribute(btree_cache_size); read_attribute(compression_stats); read_attribute(journal_debug); -read_attribute(btree_updates); read_attribute(btree_cache); read_attribute(btree_key_cache); read_attribute(stripes_heap); @@ -415,9 +413,6 @@ SHOW(bch2_fs) if (attr == &sysfs_journal_debug) bch2_journal_debug_to_text(out, &c->journal); - if (attr == &sysfs_btree_updates) - bch2_btree_updates_to_text(out, c); - if (attr == &sysfs_btree_cache) bch2_btree_cache_to_text(out, c); @@ -639,7 +634,6 @@ SYSFS_OPS(bch2_fs_internal); struct attribute *bch2_fs_internal_files[] = { &sysfs_flags, &sysfs_journal_debug, - &sysfs_btree_updates, &sysfs_btree_cache, &sysfs_btree_key_cache, &sysfs_new_stripes, -- cgit v1.2.3-59-g8ed1b From d880a43836d5e2ba951b10471104cdacc2eefbed Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 3 Apr 2024 19:52:10 -0400 Subject: bcachefs: Further improve btree_update_to_text() Print start and end level of the btree update; also a bit of cleanup. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 100 +++++++++++++++++------------------- fs/bcachefs/btree_update_interior.h | 3 +- 2 files changed, 48 insertions(+), 55 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 32397b99752f..24fbd5be70a2 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -26,9 +26,9 @@ #include -const char * const bch2_btree_update_modes[] = { +static const char * const bch2_btree_update_modes[] = { #define x(t) #t, - BCH_WATERMARKS() + BTREE_UPDATE_MODES() #undef x NULL }; @@ -850,15 +850,17 @@ static void btree_update_updated_node(struct btree_update *as, struct btree *b) { struct bch_fs *c = as->c; - mutex_lock(&c->btree_interior_update_lock); - list_add_tail(&as->unwritten_list, &c->btree_interior_updates_unwritten); - BUG_ON(as->mode != BTREE_UPDATE_none); + BUG_ON(as->update_level_end < b->c.level); BUG_ON(!btree_node_dirty(b)); BUG_ON(!b->c.level); + mutex_lock(&c->btree_interior_update_lock); + list_add_tail(&as->unwritten_list, &c->btree_interior_updates_unwritten); + as->mode = BTREE_UPDATE_node; as->b = b; + as->update_level_end = b->c.level; set_btree_node_write_blocked(b); list_add(&as->write_blocked_list, &b->write_blocked); @@ -1100,7 +1102,7 @@ static void bch2_btree_update_done(struct btree_update *as, struct btree_trans * static struct btree_update * bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, - unsigned level, bool split, unsigned flags) + unsigned level_start, bool split, unsigned flags) { struct bch_fs *c = trans->c; struct btree_update *as; @@ -1108,7 +1110,7 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, int disk_res_flags = (flags & BCH_TRANS_COMMIT_no_enospc) ? BCH_DISK_RESERVATION_NOFAIL : 0; unsigned nr_nodes[2] = { 0, 0 }; - unsigned update_level = level; + unsigned level_end = level_start; enum bch_watermark watermark = flags & BCH_WATERMARK_MASK; int ret = 0; u32 restart_count = trans->restart_count; @@ -1140,17 +1142,17 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, } while (1) { - nr_nodes[!!update_level] += 1 + split; - update_level++; + nr_nodes[!!level_end] += 1 + split; + level_end++; - ret = bch2_btree_path_upgrade(trans, path, update_level + 1); + ret = bch2_btree_path_upgrade(trans, path, level_end + 1); if (ret) return ERR_PTR(ret); - if (!btree_path_node(path, update_level)) { + if (!btree_path_node(path, level_end)) { /* Allocating new root? */ nr_nodes[1] += split; - update_level = BTREE_MAX_DEPTH; + level_end = BTREE_MAX_DEPTH; break; } @@ -1158,11 +1160,11 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, * Always check for space for two keys, even if we won't have to * split at prior level - it might have been a merge instead: */ - if (bch2_btree_node_insert_fits(path->l[update_level].b, + if (bch2_btree_node_insert_fits(path->l[level_end].b, BKEY_BTREE_PTR_U64s_MAX * 2)) break; - split = path->l[update_level].b->nr.live_u64s > BTREE_SPLIT_THRESHOLD(c); + split = path->l[level_end].b->nr.live_u64s > BTREE_SPLIT_THRESHOLD(c); } if (!down_read_trylock(&c->gc_lock)) { @@ -1176,14 +1178,15 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, as = mempool_alloc(&c->btree_interior_update_pool, GFP_NOFS); memset(as, 0, sizeof(*as)); closure_init(&as->cl, NULL); - as->c = c; - as->start_time = start_time; - as->ip_started = _RET_IP_; - as->mode = BTREE_UPDATE_none; - as->watermark = watermark; - as->took_gc_lock = true; - as->btree_id = path->btree_id; - as->update_level = update_level; + as->c = c; + as->start_time = start_time; + as->ip_started = _RET_IP_; + as->mode = BTREE_UPDATE_none; + as->watermark = watermark; + as->took_gc_lock = true; + as->btree_id = path->btree_id; + as->update_level_start = level_start; + as->update_level_end = level_end; INIT_LIST_HEAD(&as->list); INIT_LIST_HEAD(&as->unwritten_list); INIT_LIST_HEAD(&as->write_blocked_list); @@ -1373,12 +1376,12 @@ static void bch2_insert_fixup_btree_ptr(struct btree_update *as, } static void -__bch2_btree_insert_keys_interior(struct btree_update *as, - struct btree_trans *trans, - struct btree_path *path, - struct btree *b, - struct btree_node_iter node_iter, - struct keylist *keys) +bch2_btree_insert_keys_interior(struct btree_update *as, + struct btree_trans *trans, + struct btree_path *path, + struct btree *b, + struct btree_node_iter node_iter, + struct keylist *keys) { struct bkey_i *insert = bch2_keylist_front(keys); struct bkey_packed *k; @@ -1534,7 +1537,7 @@ static void btree_split_insert_keys(struct btree_update *as, bch2_btree_node_iter_init(&node_iter, b, &bch2_keylist_front(keys)->k.p); - __bch2_btree_insert_keys_interior(as, trans, path, b, node_iter, keys); + bch2_btree_insert_keys_interior(as, trans, path, b, node_iter, keys); BUG_ON(bch2_btree_node_check_topology(trans, b)); } @@ -1714,27 +1717,6 @@ err: goto out; } -static void -bch2_btree_insert_keys_interior(struct btree_update *as, - struct btree_trans *trans, - struct btree_path *path, - struct btree *b, - struct keylist *keys) -{ - struct btree_path *linked; - unsigned i; - - __bch2_btree_insert_keys_interior(as, trans, path, b, - path->l[b->c.level].iter, keys); - - btree_update_updated_node(as, b); - - trans_for_each_path_with_node(trans, b, linked, i) - bch2_btree_node_iter_peek(&linked->l[b->c.level].iter, b); - - bch2_trans_verify_paths(trans); -} - /** * bch2_btree_insert_node - insert bkeys into a given btree node * @@ -1755,7 +1737,8 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t struct keylist *keys) { struct bch_fs *c = as->c; - struct btree_path *path = trans->paths + path_idx; + struct btree_path *path = trans->paths + path_idx, *linked; + unsigned i; int old_u64s = le16_to_cpu(btree_bset_last(b)->u64s); int old_live_u64s = b->nr.live_u64s; int live_u64s_added, u64s_added; @@ -1784,7 +1767,13 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t return ret; } - bch2_btree_insert_keys_interior(as, trans, path, b, keys); + bch2_btree_insert_keys_interior(as, trans, path, b, + path->l[b->c.level].iter, keys); + + trans_for_each_path_with_node(trans, b, linked, i) + bch2_btree_node_iter_peek(&linked->l[b->c.level].iter, b); + + bch2_trans_verify_paths(trans); live_u64s_added = (int) b->nr.live_u64s - old_live_u64s; u64s_added = (int) le16_to_cpu(btree_bset_last(b)->u64s) - old_u64s; @@ -1798,6 +1787,7 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t bch2_maybe_compact_whiteouts(c, b)) bch2_trans_node_reinit_iter(trans, b); + btree_update_updated_node(as, b); bch2_btree_node_unlock_write(trans, path, b); BUG_ON(bch2_btree_node_check_topology(trans, b)); @@ -1807,7 +1797,7 @@ split: * We could attempt to avoid the transaction restart, by calling * bch2_btree_path_upgrade() and allocating more nodes: */ - if (b->c.level >= as->update_level) { + if (b->c.level >= as->update_level_end) { trace_and_count(c, trans_restart_split_race, trans, _THIS_IP_, b); return btree_trans_restart(trans, BCH_ERR_transaction_restart_split_race); } @@ -2519,9 +2509,11 @@ void bch2_btree_root_alloc_fake(struct bch_fs *c, enum btree_id id, unsigned lev static void bch2_btree_update_to_text(struct printbuf *out, struct btree_update *as) { - prt_printf(out, "%ps: btree=%s watermark=%s mode=%s nodes_written=%u cl.remaining=%u journal_seq=%llu\n", + prt_printf(out, "%ps: btree=%s l=%u-%u watermark=%s mode=%s nodes_written=%u cl.remaining=%u journal_seq=%llu\n", (void *) as->ip_started, bch2_btree_id_str(as->btree_id), + as->update_level_start, + as->update_level_end, bch2_watermarks[as->watermark], bch2_btree_update_modes[as->mode], as->nodes_written, diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index 88dcf5a22a3b..c1a479ebaad1 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -57,7 +57,8 @@ struct btree_update { unsigned took_gc_lock:1; enum btree_id btree_id; - unsigned update_level; + unsigned update_level_start; + unsigned update_level_end; struct disk_reservation disk_res; -- cgit v1.2.3-59-g8ed1b From 9802ff48f3fd8ae5d6699c5a32afc76769920c98 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 20 Feb 2024 21:08:24 -0500 Subject: bcachefs: Print shutdown journal sequence number Signed-off-by: Kent Overstreet --- fs/bcachefs/super.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index ed63018f21be..8daf80a38d60 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -288,8 +288,13 @@ static void __bch2_fs_read_only(struct bch_fs *c) if (test_bit(JOURNAL_REPLAY_DONE, &c->journal.flags) && !test_bit(BCH_FS_emergency_ro, &c->flags)) set_bit(BCH_FS_clean_shutdown, &c->flags); + bch2_fs_journal_stop(&c->journal); + bch_info(c, "%sshutdown complete, journal seq %llu", + test_bit(BCH_FS_clean_shutdown, &c->flags) ? "" : "un", + c->journal.seq_ondisk); + /* * After stopping journal: */ -- cgit v1.2.3-59-g8ed1b From 24457f1be29f1e7042e50a7749f5c2dde8c433c8 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 4 Apr 2024 15:12:00 -0700 Subject: nfs: Handle error of rpc_proc_register() in nfs_net_init(). syzkaller reported a warning [0] triggered while destroying immature netns. rpc_proc_register() was called in init_nfs_fs(), but its error has been ignored since at least the initial commit 1da177e4c3f4 ("Linux-2.6.12-rc2"). Recently, commit d47151b79e32 ("nfs: expose /proc/net/sunrpc/nfs in net namespaces") converted the procfs to per-netns and made the problem more visible. Even when rpc_proc_register() fails, nfs_net_init() could succeed, and thus nfs_net_exit() will be called while destroying the netns. Then, remove_proc_entry() will be called for non-existing proc directory and trigger the warning below. Let's handle the error of rpc_proc_register() properly in nfs_net_init(). [0]: name 'nfs' WARNING: CPU: 1 PID: 1710 at fs/proc/generic.c:711 remove_proc_entry+0x1bb/0x2d0 fs/proc/generic.c:711 Modules linked in: CPU: 1 PID: 1710 Comm: syz-executor.2 Not tainted 6.8.0-12822-gcd51db110a7e #12 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 RIP: 0010:remove_proc_entry+0x1bb/0x2d0 fs/proc/generic.c:711 Code: 41 5d 41 5e c3 e8 85 09 b5 ff 48 c7 c7 88 58 64 86 e8 09 0e 71 02 e8 74 09 b5 ff 4c 89 e6 48 c7 c7 de 1b 80 84 e8 c5 ad 97 ff <0f> 0b eb b1 e8 5c 09 b5 ff 48 c7 c7 88 58 64 86 e8 e0 0d 71 02 eb RSP: 0018:ffffc9000c6d7ce0 EFLAGS: 00010286 RAX: 0000000000000000 RBX: ffff8880422b8b00 RCX: ffffffff8110503c RDX: ffff888030652f00 RSI: ffffffff81105045 RDI: 0000000000000001 RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000001 R11: ffffffff81bb62cb R12: ffffffff84807ffc R13: ffff88804ad6fcc0 R14: ffffffff84807ffc R15: ffffffff85741ff8 FS: 00007f30cfba8640(0000) GS:ffff88807dd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007ff51afe8000 CR3: 000000005a60a005 CR4: 0000000000770ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: rpc_proc_unregister+0x64/0x70 net/sunrpc/stats.c:310 nfs_net_exit+0x1c/0x30 fs/nfs/inode.c:2438 ops_exit_list+0x62/0xb0 net/core/net_namespace.c:170 setup_net+0x46c/0x660 net/core/net_namespace.c:372 copy_net_ns+0x244/0x590 net/core/net_namespace.c:505 create_new_namespaces+0x2ed/0x770 kernel/nsproxy.c:110 unshare_nsproxy_namespaces+0xae/0x160 kernel/nsproxy.c:228 ksys_unshare+0x342/0x760 kernel/fork.c:3322 __do_sys_unshare kernel/fork.c:3393 [inline] __se_sys_unshare kernel/fork.c:3391 [inline] __x64_sys_unshare+0x1f/0x30 kernel/fork.c:3391 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0x4f/0x110 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x46/0x4e RIP: 0033:0x7f30d0febe5d Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 73 9f 1b 00 f7 d8 64 89 01 48 RSP: 002b:00007f30cfba7cc8 EFLAGS: 00000246 ORIG_RAX: 0000000000000110 RAX: ffffffffffffffda RBX: 00000000004bbf80 RCX: 00007f30d0febe5d RDX: 0000000000000000 RSI: 0000000000000000 RDI: 000000006c020600 RBP: 00000000004bbf80 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000002 R13: 000000000000000b R14: 00007f30d104c530 R15: 0000000000000000 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzkaller Signed-off-by: Kuniyuki Iwashima Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 91b4d811958a..6fe4b47c3928 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -2429,7 +2429,12 @@ static int nfs_net_init(struct net *net) struct nfs_net *nn = net_generic(net, nfs_net_id); nfs_clients_init(net); - rpc_proc_register(net, &nn->rpcstats); + + if (!rpc_proc_register(net, &nn->rpcstats)) { + nfs_clients_exit(net); + return -ENOMEM; + } + return nfs_fs_proc_net_init(net); } -- cgit v1.2.3-59-g8ed1b From 5957e0a28b5177849f7666d041b32f5dc7d27427 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 5 Apr 2024 02:43:08 -0400 Subject: bcachefs: Fix rebalance from durability=0 device Signed-off-by: Kent Overstreet --- fs/bcachefs/data_update.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 34731ee0217f..0022b51ce3c0 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -598,6 +598,8 @@ int bch2_data_update_init(struct btree_trans *trans, i++; } + unsigned durability_required = max(0, (int) (io_opts.data_replicas - durability_have)); + /* * If current extent durability is less than io_opts.data_replicas, * we're not trying to rereplicate the extent up to data_replicas here - @@ -607,7 +609,7 @@ int bch2_data_update_init(struct btree_trans *trans, * rereplicate, currently, so that users don't get an unexpected -ENOSPC */ if (!(m->data_opts.write_flags & BCH_WRITE_CACHED) && - durability_have >= io_opts.data_replicas) { + !durability_required) { m->data_opts.kill_ptrs |= m->data_opts.rewrite_ptrs; m->data_opts.rewrite_ptrs = 0; /* if iter == NULL, it's just a promote */ @@ -616,11 +618,18 @@ int bch2_data_update_init(struct btree_trans *trans, goto done; } - m->op.nr_replicas = min(durability_removing, io_opts.data_replicas - durability_have) + + m->op.nr_replicas = min(durability_removing, durability_required) + m->data_opts.extra_replicas; - m->op.nr_replicas_required = m->op.nr_replicas; - BUG_ON(!m->op.nr_replicas); + /* + * If device(s) were set to durability=0 after data was written to them + * we can end up with a duribilty=0 extent, and the normal algorithm + * that tries not to increase durability doesn't work: + */ + if (!(durability_have + durability_removing)) + m->op.nr_replicas = max((unsigned) m->op.nr_replicas, 1); + + m->op.nr_replicas_required = m->op.nr_replicas; if (reserve_sectors) { ret = bch2_disk_reservation_add(c, &m->op.res, reserve_sectors, -- cgit v1.2.3-59-g8ed1b From caeb4b0a11b3393e43f7fa8e0a5a18462acc66bd Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 31 Mar 2024 17:52:12 -0400 Subject: aio: Fix null ptr deref in aio_complete() wakeup list_del_init_careful() needs to be the last access to the wait queue entry - it effectively unlocks access. Previously, finish_wait() would see the empty list head and skip taking the lock, and then we'd return - but the completion path would still attempt to do the wakeup after the task_struct pointer had been overwritten. Fixes: 71eb6b6b0ba9 ("fs/aio: obey min_nr when doing wakeups") Cc: stable@vger.kernel.org Link: https://lore.kernel.org/linux-fsdevel/CAHTA-ubfwwB51A5Wg5M6H_rPEQK9pNf8FkAGH=vr=FEkyRrtqw@mail.gmail.com/ Signed-off-by: Kent Overstreet Link: https://lore.kernel.org/stable/20240331215212.522544-1-kent.overstreet%40linux.dev Link: https://lore.kernel.org/r/20240331215212.522544-1-kent.overstreet@linux.dev Signed-off-by: Christian Brauner --- fs/aio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index 9cdaa2faa536..0f4f531c9780 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1202,8 +1202,8 @@ static void aio_complete(struct aio_kiocb *iocb) spin_lock_irqsave(&ctx->wait.lock, flags); list_for_each_entry_safe(curr, next, &ctx->wait.head, w.entry) if (avail >= curr->min_nr) { - list_del_init_careful(&curr->w.entry); wake_up_process(curr->w.private); + list_del_init_careful(&curr->w.entry); } spin_unlock_irqrestore(&ctx->wait.lock, flags); } -- cgit v1.2.3-59-g8ed1b From d92109891f21cf367caa2cc6dff11a4411d917f4 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 28 Feb 2024 17:19:54 +0800 Subject: fs/writeback: bail out if there is no more inodes for IO and queued once For case there is no more inodes for IO in io list from last wb_writeback, We may bail out early even there is inode in dirty list should be written back. Only bail out when we queued once to avoid missing dirtied inode. This is from code reading... Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240228091958.288260-3-shikemeng@huaweicloud.com Reviewed-by: Jan Kara [brauner@kernel.org: fold in memory corruption fix from Jan in [1]] Link: https://lore.kernel.org/r/20240405132346.bid7gibby3lxxhez@quack3 [1] Signed-off-by: Christian Brauner --- fs/fs-writeback.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index fe634f00f4d9..f864c7d6ef92 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -2076,6 +2076,7 @@ static long wb_writeback(struct bdi_writeback *wb, struct inode *inode; long progress; struct blk_plug plug; + bool queued = false; blk_start_plug(&plug); for (;;) { @@ -2118,8 +2119,10 @@ static long wb_writeback(struct bdi_writeback *wb, dirtied_before = jiffies; trace_writeback_start(wb, work); - if (list_empty(&wb->b_io)) + if (list_empty(&wb->b_io)) { queue_io(wb, work, dirtied_before); + queued = true; + } if (work->sb) progress = writeback_sb_inodes(work->sb, wb, work); else @@ -2134,7 +2137,7 @@ static long wb_writeback(struct bdi_writeback *wb, * mean the overall work is done. So we keep looping as long * as made some progress on cleaning pages or inodes. */ - if (progress) { + if (progress || !queued) { spin_unlock(&wb->list_lock); continue; } -- cgit v1.2.3-59-g8ed1b From 2ddc93461214507b8e50ba7218d6260be8d623d1 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 28 Feb 2024 17:19:55 +0800 Subject: fs/writeback: remove unused parameter wb of finish_writeback_work Remove unused parameter wb of finish_writeback_work. Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240228091958.288260-4-shikemeng@huaweicloud.com Reviewed-by: Tim Chen Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/fs-writeback.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index f864c7d6ef92..a754eb254535 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -166,8 +166,7 @@ static void wb_wakeup_delayed(struct bdi_writeback *wb) spin_unlock_irq(&wb->work_lock); } -static void finish_writeback_work(struct bdi_writeback *wb, - struct wb_writeback_work *work) +static void finish_writeback_work(struct wb_writeback_work *work) { struct wb_completion *done = work->done; @@ -196,7 +195,7 @@ static void wb_queue_work(struct bdi_writeback *wb, list_add_tail(&work->list, &wb->work_list); mod_delayed_work(bdi_wq, &wb->dwork, 0); } else - finish_writeback_work(wb, work); + finish_writeback_work(work); spin_unlock_irq(&wb->work_lock); } @@ -2272,7 +2271,7 @@ static long wb_do_writeback(struct bdi_writeback *wb) while ((work = get_next_work_item(wb)) != NULL) { trace_writeback_exec(wb, work); wrote += wb_writeback(wb, work); - finish_writeback_work(wb, work); + finish_writeback_work(work); } /* -- cgit v1.2.3-59-g8ed1b From 639924abc1ae28eb05893a402081e8e6cff23b8a Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 28 Feb 2024 17:19:56 +0800 Subject: fs/writeback: only calculate dirtied_before when b_io is empty The dirtied_before is only used when b_io is not empty, so only calculate when b_io is not empty. Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240228091958.288260-5-shikemeng@huaweicloud.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/fs-writeback.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index a754eb254535..cbfb3e05e120 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -2105,20 +2105,21 @@ static long wb_writeback(struct bdi_writeback *wb, spin_lock(&wb->list_lock); - /* - * Kupdate and background works are special and we want to - * include all inodes that need writing. Livelock avoidance is - * handled by these works yielding to any other work so we are - * safe. - */ - if (work->for_kupdate) { - dirtied_before = jiffies - - msecs_to_jiffies(dirty_expire_interval * 10); - } else if (work->for_background) - dirtied_before = jiffies; - trace_writeback_start(wb, work); if (list_empty(&wb->b_io)) { + /* + * Kupdate and background works are special and we want + * to include all inodes that need writing. Livelock + * avoidance is handled by these works yielding to any + * other work so we are safe. + */ + if (work->for_kupdate) { + dirtied_before = jiffies - + msecs_to_jiffies(dirty_expire_interval * + 10); + } else if (work->for_background) + dirtied_before = jiffies; + queue_io(wb, work, dirtied_before); queued = true; } -- cgit v1.2.3-59-g8ed1b From ba679de9c3fc511f457ea0ad8f5a22e9152fa07b Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 28 Feb 2024 17:19:57 +0800 Subject: fs/writeback: correct comment of __wakeup_flusher_threads_bdi Commit e8e8a0c6c9bfc ("writeback: move nr_pages == 0 logic to one location") removed parameter nr_pages of __wakeup_flusher_threads_bdi and we try to writeback all dirty pages in __wakeup_flusher_threads_bdi now. Just correct stale comment. Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240228091958.288260-6-shikemeng@huaweicloud.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/fs-writeback.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index cbfb3e05e120..e853da9a7fbb 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -2332,8 +2332,7 @@ void wb_workfn(struct work_struct *work) } /* - * Start writeback of `nr_pages' pages on this bdi. If `nr_pages' is zero, - * write back the whole world. + * Start writeback of all dirty pages on this bdi. */ static void __wakeup_flusher_threads_bdi(struct backing_dev_info *bdi, enum wb_reason reason) -- cgit v1.2.3-59-g8ed1b From 6a1ee87176ffed1d9e749bc66a2ad85be2d7dbb7 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 28 Feb 2024 17:19:58 +0800 Subject: fs/writeback: remove unnecessary return in writeback_inodes_sb writeback_inodes_sb doesn't have return value, just remove unnecessary return in it. Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240228091958.288260-7-shikemeng@huaweicloud.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/fs-writeback.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index e853da9a7fbb..92a5b8283528 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -2735,7 +2735,7 @@ EXPORT_SYMBOL(writeback_inodes_sb_nr); */ void writeback_inodes_sb(struct super_block *sb, enum wb_reason reason) { - return writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason); + writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason); } EXPORT_SYMBOL(writeback_inodes_sb); -- cgit v1.2.3-59-g8ed1b From 85a6a1aff08ec9f5b929d345d066e2830e8818e5 Mon Sep 17 00:00:00 2001 From: Brian Kubisiak Date: Sun, 17 Mar 2024 07:46:00 -0700 Subject: ecryptfs: Fix buffer size for tag 66 packet The 'TAG 66 Packet Format' description is missing the cipher code and checksum fields that are packed into the message packet. As a result, the buffer allocated for the packet is 3 bytes too small and write_tag_66_packet() will write up to 3 bytes past the end of the buffer. Fix this by increasing the size of the allocation so the whole packet will always fit in the buffer. This fixes the below kasan slab-out-of-bounds bug: BUG: KASAN: slab-out-of-bounds in ecryptfs_generate_key_packet_set+0x7d6/0xde0 Write of size 1 at addr ffff88800afbb2a5 by task touch/181 CPU: 0 PID: 181 Comm: touch Not tainted 6.6.13-gnu #1 4c9534092be820851bb687b82d1f92a426598dc6 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2/GNU Guix 04/01/2014 Call Trace: dump_stack_lvl+0x4c/0x70 print_report+0xc5/0x610 ? ecryptfs_generate_key_packet_set+0x7d6/0xde0 ? kasan_complete_mode_report_info+0x44/0x210 ? ecryptfs_generate_key_packet_set+0x7d6/0xde0 kasan_report+0xc2/0x110 ? ecryptfs_generate_key_packet_set+0x7d6/0xde0 __asan_store1+0x62/0x80 ecryptfs_generate_key_packet_set+0x7d6/0xde0 ? __pfx_ecryptfs_generate_key_packet_set+0x10/0x10 ? __alloc_pages+0x2e2/0x540 ? __pfx_ovl_open+0x10/0x10 [overlay 30837f11141636a8e1793533a02e6e2e885dad1d] ? dentry_open+0x8f/0xd0 ecryptfs_write_metadata+0x30a/0x550 ? __pfx_ecryptfs_write_metadata+0x10/0x10 ? ecryptfs_get_lower_file+0x6b/0x190 ecryptfs_initialize_file+0x77/0x150 ecryptfs_create+0x1c2/0x2f0 path_openat+0x17cf/0x1ba0 ? __pfx_path_openat+0x10/0x10 do_filp_open+0x15e/0x290 ? __pfx_do_filp_open+0x10/0x10 ? __kasan_check_write+0x18/0x30 ? _raw_spin_lock+0x86/0xf0 ? __pfx__raw_spin_lock+0x10/0x10 ? __kasan_check_write+0x18/0x30 ? alloc_fd+0xf4/0x330 do_sys_openat2+0x122/0x160 ? __pfx_do_sys_openat2+0x10/0x10 __x64_sys_openat+0xef/0x170 ? __pfx___x64_sys_openat+0x10/0x10 do_syscall_64+0x60/0xd0 entry_SYSCALL_64_after_hwframe+0x6e/0xd8 RIP: 0033:0x7f00a703fd67 Code: 25 00 00 41 00 3d 00 00 41 00 74 37 64 8b 04 25 18 00 00 00 85 c0 75 5b 44 89 e2 48 89 ee bf 9c ff ff ff b8 01 01 00 00 0f 05 <48> 3d 00 f0 ff ff 0f 87 85 00 00 00 48 83 c4 68 5d 41 5c c3 0f 1f RSP: 002b:00007ffc088e30b0 EFLAGS: 00000246 ORIG_RAX: 0000000000000101 RAX: ffffffffffffffda RBX: 00007ffc088e3368 RCX: 00007f00a703fd67 RDX: 0000000000000941 RSI: 00007ffc088e48d7 RDI: 00000000ffffff9c RBP: 00007ffc088e48d7 R08: 0000000000000001 R09: 0000000000000000 R10: 00000000000001b6 R11: 0000000000000246 R12: 0000000000000941 R13: 0000000000000000 R14: 00007ffc088e48d7 R15: 00007f00a7180040 Allocated by task 181: kasan_save_stack+0x2f/0x60 kasan_set_track+0x29/0x40 kasan_save_alloc_info+0x25/0x40 __kasan_kmalloc+0xc5/0xd0 __kmalloc+0x66/0x160 ecryptfs_generate_key_packet_set+0x6d2/0xde0 ecryptfs_write_metadata+0x30a/0x550 ecryptfs_initialize_file+0x77/0x150 ecryptfs_create+0x1c2/0x2f0 path_openat+0x17cf/0x1ba0 do_filp_open+0x15e/0x290 do_sys_openat2+0x122/0x160 __x64_sys_openat+0xef/0x170 do_syscall_64+0x60/0xd0 entry_SYSCALL_64_after_hwframe+0x6e/0xd8 Fixes: dddfa461fc89 ("[PATCH] eCryptfs: Public key; packet management") Signed-off-by: Brian Kubisiak Link: https://lore.kernel.org/r/5j2q56p6qkhezva6b2yuqfrsurmvrrqtxxzrnp3wqu7xrz22i7@hoecdztoplbl Signed-off-by: Christian Brauner --- fs/ecryptfs/keystore.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index 3fe41964c0d8..7f9f68c00ef6 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -300,9 +300,11 @@ write_tag_66_packet(char *signature, u8 cipher_code, * | Key Identifier Size | 1 or 2 bytes | * | Key Identifier | arbitrary | * | File Encryption Key Size | 1 or 2 bytes | + * | Cipher Code | 1 byte | * | File Encryption Key | arbitrary | + * | Checksum | 2 bytes | */ - data_len = (5 + ECRYPTFS_SIG_SIZE_HEX + crypt_stat->key_size); + data_len = (8 + ECRYPTFS_SIG_SIZE_HEX + crypt_stat->key_size); *packet = kmalloc(data_len, GFP_KERNEL); message = *packet; if (!message) { -- cgit v1.2.3-59-g8ed1b From 75a07b557a11a713c85f5fa8f9ed99b8c9b6f5d8 Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Thu, 21 Mar 2024 21:16:38 +0800 Subject: fs: aio: use a folio in aio_setup_ring() Use a folio throughout aio_setup_ring() to remove calls to compound_head(), also use folio_end_read() to simultaneously mark the folio uptodate and unlock it. Signed-off-by: Kefeng Wang Link: https://lore.kernel.org/r/20240321131640.948634-2-wangkefeng.wang@huawei.com Reviewed-by: Matthew Wilcox (Oracle) Signed-off-by: Christian Brauner --- fs/aio.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index 9cdaa2faa536..60da236ad575 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -527,17 +527,19 @@ static int aio_setup_ring(struct kioctx *ctx, unsigned int nr_events) } for (i = 0; i < nr_pages; i++) { - struct page *page; - page = find_or_create_page(file->f_mapping, - i, GFP_USER | __GFP_ZERO); - if (!page) + struct folio *folio; + + folio = __filemap_get_folio(file->f_mapping, i, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, + GFP_USER | __GFP_ZERO); + if (IS_ERR(folio)) break; - pr_debug("pid(%d) page[%d]->count=%d\n", - current->pid, i, page_count(page)); - SetPageUptodate(page); - unlock_page(page); - ctx->ring_pages[i] = page; + pr_debug("pid(%d) [%d] folio->count=%d\n", current->pid, i, + folio_ref_count(folio)); + folio_end_read(folio, true); + + ctx->ring_pages[i] = &folio->page; } ctx->nr_pages = i; -- cgit v1.2.3-59-g8ed1b From 5ea0a35511895f10b965e24ed6f11adae47668d2 Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Thu, 21 Mar 2024 21:16:39 +0800 Subject: fs: aio: use a folio in aio_free_ring() Use a folio throughout aio_free_ring() to remove calls to compound_head(), also move pr_debug after folio check to remove unnecessary print. Signed-off-by: Kefeng Wang Link: https://lore.kernel.org/r/20240321131640.948634-3-wangkefeng.wang@huawei.com Reviewed-by: Matthew Wilcox (Oracle) Signed-off-by: Christian Brauner --- fs/aio.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index 60da236ad575..738654b58bfb 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -334,14 +334,15 @@ static void aio_free_ring(struct kioctx *ctx) put_aio_ring_file(ctx); for (i = 0; i < ctx->nr_pages; i++) { - struct page *page; - pr_debug("pid(%d) [%d] page->count=%d\n", current->pid, i, - page_count(ctx->ring_pages[i])); - page = ctx->ring_pages[i]; - if (!page) + struct folio *folio = page_folio(ctx->ring_pages[i]); + + if (!folio) continue; + + pr_debug("pid(%d) [%d] folio->count=%d\n", current->pid, i, + folio_ref_count(folio)); ctx->ring_pages[i] = NULL; - put_page(page); + folio_put(folio); } if (ctx->ring_pages && ctx->ring_pages != ctx->internal_pages) { -- cgit v1.2.3-59-g8ed1b From 16594e60cd8e6d4ce600b96b343ec11854a0f8d8 Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Thu, 21 Mar 2024 21:16:40 +0800 Subject: fs: aio: convert to ring_folios and internal_folios Since aio use folios in most functions, convert ring/internal_pages to ring/internal_folios, let's directly use folio instead of page throughout aio to remove hidden calls to compound_head(), eg, flush_dcache_page(). Signed-off-by: Kefeng Wang Link: https://lore.kernel.org/r/20240321131640.948634-4-wangkefeng.wang@huawei.com Reviewed-by: Matthew Wilcox (Oracle) Signed-off-by: Christian Brauner --- fs/aio.c | 62 +++++++++++++++++++++++++++++++------------------------------- 1 file changed, 31 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index 738654b58bfb..fb22a17859c6 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -122,7 +122,7 @@ struct kioctx { unsigned long mmap_base; unsigned long mmap_size; - struct page **ring_pages; + struct folio **ring_folios; long nr_pages; struct rcu_work free_rwork; /* see free_ioctx() */ @@ -160,7 +160,7 @@ struct kioctx { spinlock_t completion_lock; } ____cacheline_aligned_in_smp; - struct page *internal_pages[AIO_RING_PAGES]; + struct folio *internal_folios[AIO_RING_PAGES]; struct file *aio_ring_file; unsigned id; @@ -334,20 +334,20 @@ static void aio_free_ring(struct kioctx *ctx) put_aio_ring_file(ctx); for (i = 0; i < ctx->nr_pages; i++) { - struct folio *folio = page_folio(ctx->ring_pages[i]); + struct folio *folio = ctx->ring_folios[i]; if (!folio) continue; pr_debug("pid(%d) [%d] folio->count=%d\n", current->pid, i, folio_ref_count(folio)); - ctx->ring_pages[i] = NULL; + ctx->ring_folios[i] = NULL; folio_put(folio); } - if (ctx->ring_pages && ctx->ring_pages != ctx->internal_pages) { - kfree(ctx->ring_pages); - ctx->ring_pages = NULL; + if (ctx->ring_folios && ctx->ring_folios != ctx->internal_folios) { + kfree(ctx->ring_folios); + ctx->ring_folios = NULL; } } @@ -442,7 +442,7 @@ static int aio_migrate_folio(struct address_space *mapping, struct folio *dst, idx = src->index; if (idx < (pgoff_t)ctx->nr_pages) { /* Make sure the old folio hasn't already been changed */ - if (ctx->ring_pages[idx] != &src->page) + if (ctx->ring_folios[idx] != src) rc = -EAGAIN; } else rc = -EINVAL; @@ -466,8 +466,8 @@ static int aio_migrate_folio(struct address_space *mapping, struct folio *dst, */ spin_lock_irqsave(&ctx->completion_lock, flags); folio_migrate_copy(dst, src); - BUG_ON(ctx->ring_pages[idx] != &src->page); - ctx->ring_pages[idx] = &dst->page; + BUG_ON(ctx->ring_folios[idx] != src); + ctx->ring_folios[idx] = dst; spin_unlock_irqrestore(&ctx->completion_lock, flags); /* The old folio is no longer accessible. */ @@ -517,11 +517,11 @@ static int aio_setup_ring(struct kioctx *ctx, unsigned int nr_events) nr_events = (PAGE_SIZE * nr_pages - sizeof(struct aio_ring)) / sizeof(struct io_event); - ctx->ring_pages = ctx->internal_pages; + ctx->ring_folios = ctx->internal_folios; if (nr_pages > AIO_RING_PAGES) { - ctx->ring_pages = kcalloc(nr_pages, sizeof(struct page *), - GFP_KERNEL); - if (!ctx->ring_pages) { + ctx->ring_folios = kcalloc(nr_pages, sizeof(struct folio *), + GFP_KERNEL); + if (!ctx->ring_folios) { put_aio_ring_file(ctx); return -ENOMEM; } @@ -540,7 +540,7 @@ static int aio_setup_ring(struct kioctx *ctx, unsigned int nr_events) folio_ref_count(folio)); folio_end_read(folio, true); - ctx->ring_pages[i] = &folio->page; + ctx->ring_folios[i] = folio; } ctx->nr_pages = i; @@ -573,7 +573,7 @@ static int aio_setup_ring(struct kioctx *ctx, unsigned int nr_events) ctx->user_id = ctx->mmap_base; ctx->nr_events = nr_events; /* trusted copy */ - ring = page_address(ctx->ring_pages[0]); + ring = folio_address(ctx->ring_folios[0]); ring->nr = nr_events; /* user copy */ ring->id = ~0U; ring->head = ring->tail = 0; @@ -581,7 +581,7 @@ static int aio_setup_ring(struct kioctx *ctx, unsigned int nr_events) ring->compat_features = AIO_RING_COMPAT_FEATURES; ring->incompat_features = AIO_RING_INCOMPAT_FEATURES; ring->header_length = sizeof(struct aio_ring); - flush_dcache_page(ctx->ring_pages[0]); + flush_dcache_folio(ctx->ring_folios[0]); return 0; } @@ -692,9 +692,9 @@ static int ioctx_add_table(struct kioctx *ctx, struct mm_struct *mm) /* While kioctx setup is in progress, * we are protected from page migration - * changes ring_pages by ->ring_lock. + * changes ring_folios by ->ring_lock. */ - ring = page_address(ctx->ring_pages[0]); + ring = folio_address(ctx->ring_folios[0]); ring->id = ctx->id; return 0; } @@ -1036,7 +1036,7 @@ static void user_refill_reqs_available(struct kioctx *ctx) * against ctx->completed_events below will make sure we do the * safe/right thing. */ - ring = page_address(ctx->ring_pages[0]); + ring = folio_address(ctx->ring_folios[0]); head = ring->head; refill_reqs_available(ctx, head, ctx->tail); @@ -1148,12 +1148,12 @@ static void aio_complete(struct aio_kiocb *iocb) if (++tail >= ctx->nr_events) tail = 0; - ev_page = page_address(ctx->ring_pages[pos / AIO_EVENTS_PER_PAGE]); + ev_page = folio_address(ctx->ring_folios[pos / AIO_EVENTS_PER_PAGE]); event = ev_page + pos % AIO_EVENTS_PER_PAGE; *event = iocb->ki_res; - flush_dcache_page(ctx->ring_pages[pos / AIO_EVENTS_PER_PAGE]); + flush_dcache_folio(ctx->ring_folios[pos / AIO_EVENTS_PER_PAGE]); pr_debug("%p[%u]: %p: %p %Lx %Lx %Lx\n", ctx, tail, iocb, (void __user *)(unsigned long)iocb->ki_res.obj, @@ -1166,10 +1166,10 @@ static void aio_complete(struct aio_kiocb *iocb) ctx->tail = tail; - ring = page_address(ctx->ring_pages[0]); + ring = folio_address(ctx->ring_folios[0]); head = ring->head; ring->tail = tail; - flush_dcache_page(ctx->ring_pages[0]); + flush_dcache_folio(ctx->ring_folios[0]); ctx->completed_events++; if (ctx->completed_events > 1) @@ -1241,8 +1241,8 @@ static long aio_read_events_ring(struct kioctx *ctx, sched_annotate_sleep(); mutex_lock(&ctx->ring_lock); - /* Access to ->ring_pages here is protected by ctx->ring_lock. */ - ring = page_address(ctx->ring_pages[0]); + /* Access to ->ring_folios here is protected by ctx->ring_lock. */ + ring = folio_address(ctx->ring_folios[0]); head = ring->head; tail = ring->tail; @@ -1263,20 +1263,20 @@ static long aio_read_events_ring(struct kioctx *ctx, while (ret < nr) { long avail; struct io_event *ev; - struct page *page; + struct folio *folio; avail = (head <= tail ? tail : ctx->nr_events) - head; if (head == tail) break; pos = head + AIO_EVENTS_OFFSET; - page = ctx->ring_pages[pos / AIO_EVENTS_PER_PAGE]; + folio = ctx->ring_folios[pos / AIO_EVENTS_PER_PAGE]; pos %= AIO_EVENTS_PER_PAGE; avail = min(avail, nr - ret); avail = min_t(long, avail, AIO_EVENTS_PER_PAGE - pos); - ev = page_address(page); + ev = folio_address(folio); copy_ret = copy_to_user(event + ret, ev + pos, sizeof(*ev) * avail); @@ -1290,9 +1290,9 @@ static long aio_read_events_ring(struct kioctx *ctx, head %= ctx->nr_events; } - ring = page_address(ctx->ring_pages[0]); + ring = folio_address(ctx->ring_folios[0]); ring->head = head; - flush_dcache_page(ctx->ring_pages[0]); + flush_dcache_folio(ctx->ring_folios[0]); pr_debug("%li h%u t%u\n", ret, head, tail); out: -- cgit v1.2.3-59-g8ed1b From 68d6f4f3fbd9b1baae53e7cf33fb3362b5a21494 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 25 Mar 2024 19:34:01 -0600 Subject: fs: Annotate struct file_handle with __counted_by() and use struct_size() Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). While there, use struct_size() helper, instead of the open-coded version. [brauner@kernel.org: contains a fix by Edward for an OOB access] Reported-by: syzbot+4139435cb1b34cf759c2@syzkaller.appspotmail.com Signed-off-by: Edward Adam Davis Link: https://lore.kernel.org/r/tencent_A7845DD769577306D813742365E976E3A205@qq.com Signed-off-by: Gustavo A. R. Silva Link: https://lore.kernel.org/r/ZgImCXTdGDTeBvSS@neat Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/fhandle.c | 6 +++--- include/linux/fs.h | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/fhandle.c b/fs/fhandle.c index 57a12614addf..8a7f86c2139a 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -36,7 +36,7 @@ static long do_sys_name_to_handle(const struct path *path, if (f_handle.handle_bytes > MAX_HANDLE_SZ) return -EINVAL; - handle = kzalloc(sizeof(struct file_handle) + f_handle.handle_bytes, + handle = kzalloc(struct_size(handle, f_handle, f_handle.handle_bytes), GFP_KERNEL); if (!handle) return -ENOMEM; @@ -71,7 +71,7 @@ static long do_sys_name_to_handle(const struct path *path, /* copy the mount id */ if (put_user(real_mount(path->mnt)->mnt_id, mnt_id) || copy_to_user(ufh, handle, - sizeof(struct file_handle) + handle_bytes)) + struct_size(handle, f_handle, handle_bytes))) retval = -EFAULT; kfree(handle); return retval; @@ -192,7 +192,7 @@ static int handle_to_path(int mountdirfd, struct file_handle __user *ufh, retval = -EINVAL; goto out_err; } - handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_bytes, + handle = kmalloc(struct_size(handle, f_handle, f_handle.handle_bytes), GFP_KERNEL); if (!handle) { retval = -ENOMEM; diff --git a/include/linux/fs.h b/include/linux/fs.h index 00fc429b0af0..914245085597 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1033,7 +1033,7 @@ struct file_handle { __u32 handle_bytes; int handle_type; /* file identifier */ - unsigned char f_handle[]; + unsigned char f_handle[] __counted_by(handle_bytes); }; static inline struct file *get_file(struct file *f) -- cgit v1.2.3-59-g8ed1b From 10396f4df8b75ff6ab0aa2cd74296565466f2c8d Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 5 Apr 2024 13:56:18 -0400 Subject: nfsd: hold a lighter-weight client reference over CB_RECALL_ANY Currently the CB_RECALL_ANY job takes a cl_rpc_users reference to the client. While a callback job is technically an RPC that counter is really more for client-driven RPCs, and this has the effect of preventing the client from being unhashed until the callback completes. If nfsd decides to send a CB_RECALL_ANY just as the client reboots, we can end up in a situation where the callback can't complete on the (now dead) callback channel, but the new client can't connect because the old client can't be unhashed. This usually manifests as a NFS4ERR_DELAY return on the CREATE_SESSION operation. The job is only holding a reference to the client so it can clear a flag after the RPC completes. Fix this by having CB_RECALL_ANY instead hold a reference to the cl_nfsdfs.cl_ref. Typically we only take that sort of reference when dealing with the nfsdfs info files, but it should work appropriately here to ensure that the nfs4_client doesn't disappear. Fixes: 44df6f439a17 ("NFSD: add delegation reaper to react to low memory condition") Reported-by: Vladimir Benes Signed-off-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 5fcd93f7cb8c..3cef81e196c6 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -3042,12 +3042,9 @@ static void nfsd4_cb_recall_any_release(struct nfsd4_callback *cb) { struct nfs4_client *clp = cb->cb_clp; - struct nfsd_net *nn = net_generic(clp->net, nfsd_net_id); - spin_lock(&nn->client_lock); clear_bit(NFSD4_CLIENT_CB_RECALL_ANY, &clp->cl_flags); - put_client_renew_locked(clp); - spin_unlock(&nn->client_lock); + drop_client(clp); } static int @@ -6616,7 +6613,7 @@ deleg_reaper(struct nfsd_net *nn) list_add(&clp->cl_ra_cblist, &cblist); /* release in nfsd4_cb_recall_any_release */ - atomic_inc(&clp->cl_rpc_users); + kref_get(&clp->cl_nfsdfs.cl_ref); set_bit(NFSD4_CLIENT_CB_RECALL_ANY, &clp->cl_flags); clp->cl_ra_time = ktime_get_boottime_seconds(); } -- cgit v1.2.3-59-g8ed1b From a6c4162d844dae4dbfea1bf9ecffcb852d3ed615 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 5 Apr 2024 18:01:02 +0300 Subject: bcachefs: fix ! vs ~ typo in __clear_bit_le64() The ! was obviously intended to be ~. As it is, this function does the equivalent to: "addr[bit / 64] = 0;". Fixes: 27fcec6c27ca ("bcachefs: Clear recovery_passes_required as they complete without errors") Signed-off-by: Dan Carpenter Signed-off-by: Kent Overstreet --- fs/bcachefs/util.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index b7e7c29278fc..de639e8a3ab5 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -795,7 +795,7 @@ static inline void __set_bit_le64(size_t bit, __le64 *addr) static inline void __clear_bit_le64(size_t bit, __le64 *addr) { - addr[bit / 64] &= !cpu_to_le64(BIT_ULL(bit % 64)); + addr[bit / 64] &= ~cpu_to_le64(BIT_ULL(bit % 64)); } static inline bool test_bit_le64(size_t bit, __le64 *addr) -- cgit v1.2.3-59-g8ed1b From cf979fca9a05d7d0b116257e5c4dc12b6bb7eb3a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 5 Apr 2024 16:21:18 -0400 Subject: bcachefs: fix rand_delete unit test Signed-off-by: Kent Overstreet --- fs/bcachefs/tests.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/tests.c b/fs/bcachefs/tests.c index b3fe9fc57747..bfec656f94c0 100644 --- a/fs/bcachefs/tests.c +++ b/fs/bcachefs/tests.c @@ -672,7 +672,7 @@ static int __do_delete(struct btree_trans *trans, struct bpos pos) bch2_trans_iter_init(trans, &iter, BTREE_ID_xattrs, pos, BTREE_ITER_INTENT); - k = bch2_btree_iter_peek(&iter); + k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)); ret = bkey_err(k); if (ret) goto err; -- cgit v1.2.3-59-g8ed1b From 374b3d38feff4a1cb4ecadace9bd915ffd91fe4b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 5 Apr 2024 22:23:29 -0400 Subject: bcachefs: Fix BCH_IOCTL_FSCK_OFFLINE for encrypted filesystems To open an encrypted filesystem, we use request_key() to get the encryption key from the user's keyring - but request_key() needs to happen in the context of the process that invoked the ioctl. This easily fixed by using bch2_fs_open() in nostart mode. Signed-off-by: Kent Overstreet --- fs/bcachefs/chardev.c | 94 +++++++++++++++++++++++++++------------------------ 1 file changed, 50 insertions(+), 44 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index cbfa6459bdbc..1e033df330b0 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -134,42 +134,38 @@ static long bch2_ioctl_incremental(struct bch_ioctl_incremental __user *user_arg struct fsck_thread { struct thread_with_stdio thr; struct bch_fs *c; - char **devs; - size_t nr_devs; struct bch_opts opts; }; static void bch2_fsck_thread_exit(struct thread_with_stdio *_thr) { struct fsck_thread *thr = container_of(_thr, struct fsck_thread, thr); - if (thr->devs) - for (size_t i = 0; i < thr->nr_devs; i++) - kfree(thr->devs[i]); - kfree(thr->devs); kfree(thr); } static int bch2_fsck_offline_thread_fn(struct thread_with_stdio *stdio) { struct fsck_thread *thr = container_of(stdio, struct fsck_thread, thr); - struct bch_fs *c = bch2_fs_open(thr->devs, thr->nr_devs, thr->opts); - - if (IS_ERR(c)) - return PTR_ERR(c); + struct bch_fs *c = thr->c; - int ret = 0; - if (test_bit(BCH_FS_errors_fixed, &c->flags)) - ret |= 1; - if (test_bit(BCH_FS_error, &c->flags)) - ret |= 4; + int ret = PTR_ERR_OR_ZERO(c); + if (ret) + return ret; - bch2_fs_stop(c); + ret = bch2_fs_start(thr->c); + if (ret) + goto err; - if (ret & 1) + if (test_bit(BCH_FS_errors_fixed, &c->flags)) { bch2_stdio_redirect_printf(&stdio->stdio, false, "%s: errors fixed\n", c->name); - if (ret & 4) + ret |= 1; + } + if (test_bit(BCH_FS_error, &c->flags)) { bch2_stdio_redirect_printf(&stdio->stdio, false, "%s: still has errors\n", c->name); - + ret |= 4; + } +err: + bch2_fs_stop(c); return ret; } @@ -182,7 +178,7 @@ static long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_a { struct bch_ioctl_fsck_offline arg; struct fsck_thread *thr = NULL; - u64 *devs = NULL; + darray_str(devs) = {}; long ret = 0; if (copy_from_user(&arg, user_arg, sizeof(arg))) @@ -194,29 +190,32 @@ static long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_a if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (!(devs = kcalloc(arg.nr_devs, sizeof(*devs), GFP_KERNEL)) || - !(thr = kzalloc(sizeof(*thr), GFP_KERNEL)) || - !(thr->devs = kcalloc(arg.nr_devs, sizeof(*thr->devs), GFP_KERNEL))) { - ret = -ENOMEM; - goto err; - } + for (size_t i = 0; i < arg.nr_devs; i++) { + u64 dev_u64; + ret = copy_from_user_errcode(&dev_u64, &user_arg->devs[i], sizeof(u64)); + if (ret) + goto err; - thr->opts = bch2_opts_empty(); - thr->nr_devs = arg.nr_devs; + char *dev_str = strndup_user((char __user *)(unsigned long) dev_u64, PATH_MAX); + ret = PTR_ERR_OR_ZERO(dev_str); + if (ret) + goto err; - if (copy_from_user(devs, &user_arg->devs[0], - array_size(sizeof(user_arg->devs[0]), arg.nr_devs))) { - ret = -EINVAL; - goto err; + ret = darray_push(&devs, dev_str); + if (ret) { + kfree(dev_str); + goto err; + } } - for (size_t i = 0; i < arg.nr_devs; i++) { - thr->devs[i] = strndup_user((char __user *)(unsigned long) devs[i], PATH_MAX); - ret = PTR_ERR_OR_ZERO(thr->devs[i]); - if (ret) - goto err; + thr = kzalloc(sizeof(*thr), GFP_KERNEL); + if (!thr) { + ret = -ENOMEM; + goto err; } + thr->opts = bch2_opts_empty(); + if (arg.opts) { char *optstr = strndup_user((char __user *)(unsigned long) arg.opts, 1 << 16); @@ -230,15 +229,22 @@ static long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_a opt_set(thr->opts, stdio, (u64)(unsigned long)&thr->thr.stdio); + /* We need request_key() to be called before we punt to kthread: */ + opt_set(thr->opts, nostart, true); + + thr->c = bch2_fs_open(devs.data, arg.nr_devs, thr->opts); + ret = bch2_run_thread_with_stdio(&thr->thr, &bch2_offline_fsck_ops); -err: - if (ret < 0) { - if (thr) - bch2_fsck_thread_exit(&thr->thr); - pr_err("ret %s", bch2_err_str(ret)); - } - kfree(devs); +out: + darray_for_each(devs, i) + kfree(*i); + darray_exit(&devs); return ret; +err: + if (thr) + bch2_fsck_thread_exit(&thr->thr); + pr_err("ret %s", bch2_err_str(ret)); + goto out; } static long bch2_global_ioctl(unsigned cmd, void __user *arg) -- cgit v1.2.3-59-g8ed1b From 05801b6526156aefe55c0440fab877109c9a89c5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 5 Apr 2024 22:30:30 -0400 Subject: bcachefs: Disable errors=panic for BCH_IOCTL_FSCK_OFFLINE BCH_IOCTL_FSCK_OFFLINE allows the userspace fsck tool to use the kernel implementation of fsck - primarily when the kernel version is a better version match. It should look and act exactly like the normal userspace fsck that the user expected to be invoking, so errors should never result in a kernel panic. We may want to consider further restricting errors=panic - it's only intended for debugging in controlled test environments, it should have no purpose it normal usage. Signed-off-by: Kent Overstreet --- fs/bcachefs/chardev.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index 1e033df330b0..72781aad6ba7 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -234,6 +234,10 @@ static long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_a thr->c = bch2_fs_open(devs.data, arg.nr_devs, thr->opts); + if (!IS_ERR(thr->c) && + thr->c->opts.errors == BCH_ON_ERROR_panic) + thr->c->opts.errors = BCH_ON_ERROR_ro; + ret = bch2_run_thread_with_stdio(&thr->thr, &bch2_offline_fsck_ops); out: darray_for_each(devs, i) -- cgit v1.2.3-59-g8ed1b From 6088234ce83acec4aaf56ecc0e9525bac18b4295 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 5 Apr 2024 23:27:27 -0400 Subject: bcachefs: JOURNAL_SPACE_LOW "bcachefs; Fix deadlock in bch2_btree_update_start()" was a significant performance regression (nearly 50%) on multithreaded random writes with fio. The reason is that the journal watermark checks multiple things, including the state of the btree write buffer, and on multithreaded update heavy workloads we're bottleneked on write buffer flushing - we don't want kicknig off btree updates to depend on the state of the write buffer. This isn't strictly correct; the interior btree update path does do write buffer updates, but it's a tiny fraction of total accounting updates and we're more concerned with space in the journal itself. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 2 +- fs/bcachefs/btree_update_interior.c | 18 +++++++----------- fs/bcachefs/journal_reclaim.c | 2 ++ fs/bcachefs/journal_types.h | 1 + fs/bcachefs/util.h | 8 ++++++++ 5 files changed, 19 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 581edcb0911b..4ae7890f07c6 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -659,7 +659,7 @@ static int btree_key_cache_flush_pos(struct btree_trans *trans, commit_flags |= BCH_WATERMARK_reclaim; if (ck->journal.seq != journal_last_seq(j) || - j->watermark == BCH_WATERMARK_stripe) + !test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)) commit_flags |= BCH_TRANS_COMMIT_no_journal_res; ret = bch2_btree_iter_traverse(&b_iter) ?: diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 24fbd5be70a2..a4a63e363047 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1125,18 +1125,14 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, flags &= ~BCH_WATERMARK_MASK; flags |= watermark; - if (watermark < c->journal.watermark) { - struct journal_res res = { 0 }; - unsigned journal_flags = watermark|JOURNAL_RES_GET_CHECK; + if (watermark < BCH_WATERMARK_reclaim && + test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)) { + if (flags & BCH_TRANS_COMMIT_journal_reclaim) + return ERR_PTR(-BCH_ERR_journal_reclaim_would_deadlock); - if ((flags & BCH_TRANS_COMMIT_journal_reclaim) && - watermark < BCH_WATERMARK_reclaim) - journal_flags |= JOURNAL_RES_GET_NONBLOCK; - - ret = drop_locks_do(trans, - bch2_journal_res_get(&c->journal, &res, 1, journal_flags)); - if (bch2_err_matches(ret, BCH_ERR_operation_blocked)) - ret = -BCH_ERR_journal_reclaim_would_deadlock; + bch2_trans_unlock(trans); + wait_event(c->journal.wait, !test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)); + ret = bch2_trans_relock(trans); if (ret) return ERR_PTR(ret); } diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index ab811c0dad26..04a577848b01 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -67,6 +67,8 @@ void bch2_journal_set_watermark(struct journal *j) track_event_change(&c->times[BCH_TIME_blocked_write_buffer_full], low_on_wb)) trace_and_count(c, journal_full, c); + mod_bit(JOURNAL_SPACE_LOW, &j->flags, low_on_space || low_on_pin); + swap(watermark, j->watermark); if (watermark > j->watermark) journal_wake(j); diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index 8c053cb64ca5..b5161b5d76a0 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -134,6 +134,7 @@ enum journal_flags { JOURNAL_STARTED, JOURNAL_MAY_SKIP_FLUSH, JOURNAL_NEED_FLUSH_WRITE, + JOURNAL_SPACE_LOW, }; /* Reasons we may fail to get a journal reservation: */ diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index de639e8a3ab5..5cf885b09986 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -788,6 +788,14 @@ static inline int copy_from_user_errcode(void *to, const void __user *from, unsi #endif +static inline void mod_bit(long nr, volatile unsigned long *addr, bool v) +{ + if (v) + set_bit(nr, addr); + else + clear_bit(nr, addr); +} + static inline void __set_bit_le64(size_t bit, __le64 *addr) { addr[bit / 64] |= cpu_to_le64(BIT_ULL(bit % 64)); -- cgit v1.2.3-59-g8ed1b From 2d793e9315e335ef299024fe8bf7742c9c974b33 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Sat, 6 Apr 2024 16:19:20 +0200 Subject: bcachefs: Rename struct field swap to prevent macro naming collision The struct field swap can collide with the swap() macro defined in linux/minmax.h. Rename the struct field to prevent such collisions. Signed-off-by: Thorsten Blum Signed-off-by: Kent Overstreet --- fs/bcachefs/eytzinger.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/eytzinger.c b/fs/bcachefs/eytzinger.c index 4ce5e957a6e9..0f955c3c76a7 100644 --- a/fs/bcachefs/eytzinger.c +++ b/fs/bcachefs/eytzinger.c @@ -115,7 +115,7 @@ static void swap_bytes(void *a, void *b, size_t n) struct wrapper { cmp_func_t cmp; - swap_func_t swap; + swap_func_t swap_func; }; /* @@ -125,7 +125,7 @@ struct wrapper { static void do_swap(void *a, void *b, size_t size, swap_r_func_t swap_func, const void *priv) { if (swap_func == SWAP_WRAPPER) { - ((const struct wrapper *)priv)->swap(a, b, (int)size); + ((const struct wrapper *)priv)->swap_func(a, b, (int)size); return; } @@ -174,7 +174,7 @@ void eytzinger0_sort_r(void *base, size_t n, size_t size, int i, c, r; /* called from 'sort' without swap function, let's pick the default */ - if (swap_func == SWAP_WRAPPER && !((struct wrapper *)priv)->swap) + if (swap_func == SWAP_WRAPPER && !((struct wrapper *)priv)->swap_func) swap_func = NULL; if (!swap_func) { @@ -227,7 +227,7 @@ void eytzinger0_sort(void *base, size_t n, size_t size, { struct wrapper w = { .cmp = cmp_func, - .swap = swap_func, + .swap_func = swap_func, }; return eytzinger0_sort_r(base, n, size, _CMP_WRAPPER, SWAP_WRAPPER, &w); -- cgit v1.2.3-59-g8ed1b From 30e615a2ce6601d85729caefd8ac15634f848e59 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 21:45:46 -0400 Subject: bcachefs: Fix gap buffer bug in bch2_journal_key_insert_take() Multiple bug fixes for journal iters: - When the journal keys gap buffer is resized, we have to adjust the iterators for moving the gap to the end - We don't want to rewind iterators to point to the key we just inserted if it's not for the correct btree/level Also, add some new assertions. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_journal_iter.c | 55 ++++++++++++++++++++++++++++++++-------- 1 file changed, 45 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index 5cbcbfe85235..a6413a8747d3 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -130,12 +130,30 @@ struct bkey_i *bch2_journal_keys_peek_slot(struct bch_fs *c, enum btree_id btree return bch2_journal_keys_peek_upto(c, btree_id, level, pos, pos, &idx); } +static void journal_iter_verify(struct journal_iter *iter) +{ + struct journal_keys *keys = iter->keys; + size_t gap_size = keys->size - keys->nr; + + BUG_ON(iter->idx >= keys->gap && + iter->idx < keys->gap + gap_size); + + if (iter->idx < keys->size) { + struct journal_key *k = keys->data + iter->idx; + + int cmp = cmp_int(k->btree_id, iter->btree_id) ?: + cmp_int(k->level, iter->level); + BUG_ON(cmp < 0); + } +} + static void journal_iters_fix(struct bch_fs *c) { struct journal_keys *keys = &c->journal_keys; /* The key we just inserted is immediately before the gap: */ size_t gap_end = keys->gap + (keys->size - keys->nr); - struct btree_and_journal_iter *iter; + struct journal_key *new_key = &keys->data[keys->gap - 1]; + struct journal_iter *iter; /* * If an iterator points one after the key we just inserted, decrement @@ -143,9 +161,14 @@ static void journal_iters_fix(struct bch_fs *c) * decrement was unnecessary, bch2_btree_and_journal_iter_peek() will * handle that: */ - list_for_each_entry(iter, &c->journal_iters, journal.list) - if (iter->journal.idx == gap_end) - iter->journal.idx = keys->gap - 1; + list_for_each_entry(iter, &c->journal_iters, list) { + journal_iter_verify(iter); + if (iter->idx == gap_end && + new_key->btree_id == iter->btree_id && + new_key->level == iter->level) + iter->idx = keys->gap - 1; + journal_iter_verify(iter); + } } static void journal_iters_move_gap(struct bch_fs *c, size_t old_gap, size_t new_gap) @@ -192,7 +215,12 @@ int bch2_journal_key_insert_take(struct bch_fs *c, enum btree_id id, if (idx > keys->gap) idx -= keys->size - keys->nr; + size_t old_gap = keys->gap; + if (keys->nr == keys->size) { + journal_iters_move_gap(c, old_gap, keys->size); + old_gap = keys->size; + struct journal_keys new_keys = { .nr = keys->nr, .size = max_t(size_t, keys->size, 8) * 2, @@ -216,7 +244,7 @@ int bch2_journal_key_insert_take(struct bch_fs *c, enum btree_id id, keys->gap = keys->nr; } - journal_iters_move_gap(c, keys->gap, idx); + journal_iters_move_gap(c, old_gap, idx); move_gap(keys, idx); @@ -301,16 +329,21 @@ static void bch2_journal_iter_advance(struct journal_iter *iter) static struct bkey_s_c bch2_journal_iter_peek(struct journal_iter *iter) { - struct journal_key *k = iter->keys->data + iter->idx; + journal_iter_verify(iter); + + while (iter->idx < iter->keys->size) { + struct journal_key *k = iter->keys->data + iter->idx; + + int cmp = cmp_int(k->btree_id, iter->btree_id) ?: + cmp_int(k->level, iter->level); + if (cmp > 0) + break; + BUG_ON(cmp); - while (k < iter->keys->data + iter->keys->size && - k->btree_id == iter->btree_id && - k->level == iter->level) { if (!k->overwritten) return bkey_i_to_s_c(k->k); bch2_journal_iter_advance(iter); - k = iter->keys->data + iter->idx; } return bkey_s_c_null; @@ -330,6 +363,8 @@ static void bch2_journal_iter_init(struct bch_fs *c, iter->level = level; iter->keys = &c->journal_keys; iter->idx = bch2_journal_key_search(&c->journal_keys, id, level, pos); + + journal_iter_verify(iter); } static struct bkey_s_c bch2_journal_iter_peek_btree(struct btree_and_journal_iter *iter) -- cgit v1.2.3-59-g8ed1b From 469ad583c1293f5d9f45183050b3beeb4a8c3475 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 7 Apr 2024 03:04:50 -0400 Subject: erofs: switch erofs_bread() to passing offset instead of block number Callers are happier that way, especially since we no longer need to play with splitting offset into block number and offset within block, passing the former to erofs_bread(), then adding the latter... erofs_bread() always reads entire pages, anyway. Signed-off-by: Al Viro --- fs/erofs/data.c | 5 ++--- fs/erofs/dir.c | 2 +- fs/erofs/internal.h | 2 +- fs/erofs/namei.c | 2 +- fs/erofs/super.c | 8 ++++---- fs/erofs/xattr.c | 35 +++++++++++++---------------------- fs/erofs/zdata.c | 4 ++-- 7 files changed, 24 insertions(+), 34 deletions(-) (limited to 'fs') diff --git a/fs/erofs/data.c b/fs/erofs/data.c index 52524bd9698b..d3c446dda2ff 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -29,11 +29,10 @@ void erofs_put_metabuf(struct erofs_buf *buf) * Derive the block size from inode->i_blkbits to make compatible with * anonymous inode in fscache mode. */ -void *erofs_bread(struct erofs_buf *buf, erofs_blk_t blkaddr, +void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset, enum erofs_kmap_type type) { struct inode *inode = buf->inode; - erofs_off_t offset = (erofs_off_t)blkaddr << inode->i_blkbits; pgoff_t index = offset >> PAGE_SHIFT; struct page *page = buf->page; struct folio *folio; @@ -77,7 +76,7 @@ void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb, erofs_blk_t blkaddr, enum erofs_kmap_type type) { erofs_init_metabuf(buf, sb); - return erofs_bread(buf, blkaddr, type); + return erofs_bread(buf, erofs_pos(sb, blkaddr), type); } static int erofs_map_blocks_flatmode(struct inode *inode, diff --git a/fs/erofs/dir.c b/fs/erofs/dir.c index b80abec0531a..9d38f39bb4f7 100644 --- a/fs/erofs/dir.c +++ b/fs/erofs/dir.c @@ -63,7 +63,7 @@ static int erofs_readdir(struct file *f, struct dir_context *ctx) struct erofs_dirent *de; unsigned int nameoff, maxsize; - de = erofs_bread(&buf, i, EROFS_KMAP); + de = erofs_bread(&buf, erofs_pos(sb, i), EROFS_KMAP); if (IS_ERR(de)) { erofs_err(sb, "fail to readdir of logical block %u of nid %llu", i, EROFS_I(dir)->nid); diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 39c67119f43b..9e30c67c135c 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -409,7 +409,7 @@ void *erofs_read_metadata(struct super_block *sb, struct erofs_buf *buf, erofs_off_t *offset, int *lengthp); void erofs_unmap_metabuf(struct erofs_buf *buf); void erofs_put_metabuf(struct erofs_buf *buf); -void *erofs_bread(struct erofs_buf *buf, erofs_blk_t blkaddr, +void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset, enum erofs_kmap_type type); void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb); void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb, diff --git a/fs/erofs/namei.c b/fs/erofs/namei.c index f0110a78acb2..11afa48996a3 100644 --- a/fs/erofs/namei.c +++ b/fs/erofs/namei.c @@ -100,7 +100,7 @@ static void *erofs_find_target_block(struct erofs_buf *target, struct erofs_dirent *de; buf.inode = dir; - de = erofs_bread(&buf, mid, EROFS_KMAP); + de = erofs_bread(&buf, erofs_pos(dir->i_sb, mid), EROFS_KMAP); if (!IS_ERR(de)) { const int nameoff = nameoff_from_disk(de->nameoff, bsz); const int ndirents = nameoff / sizeof(*de); diff --git a/fs/erofs/super.c b/fs/erofs/super.c index c0eb139adb07..fdefc3772620 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -132,11 +132,11 @@ void *erofs_read_metadata(struct super_block *sb, struct erofs_buf *buf, int len, i, cnt; *offset = round_up(*offset, 4); - ptr = erofs_bread(buf, erofs_blknr(sb, *offset), EROFS_KMAP); + ptr = erofs_bread(buf, *offset, EROFS_KMAP); if (IS_ERR(ptr)) return ptr; - len = le16_to_cpu(*(__le16 *)&ptr[erofs_blkoff(sb, *offset)]); + len = le16_to_cpu(*(__le16 *)ptr); if (!len) len = U16_MAX + 1; buffer = kmalloc(len, GFP_KERNEL); @@ -148,12 +148,12 @@ void *erofs_read_metadata(struct super_block *sb, struct erofs_buf *buf, for (i = 0; i < len; i += cnt) { cnt = min_t(int, sb->s_blocksize - erofs_blkoff(sb, *offset), len - i); - ptr = erofs_bread(buf, erofs_blknr(sb, *offset), EROFS_KMAP); + ptr = erofs_bread(buf, *offset, EROFS_KMAP); if (IS_ERR(ptr)) { kfree(buffer); return ptr; } - memcpy(buffer + i, ptr + erofs_blkoff(sb, *offset), cnt); + memcpy(buffer + i, ptr, cnt); *offset += cnt; } return buffer; diff --git a/fs/erofs/xattr.c b/fs/erofs/xattr.c index b58316b49a43..ec233917830a 100644 --- a/fs/erofs/xattr.c +++ b/fs/erofs/xattr.c @@ -81,13 +81,13 @@ static int erofs_init_inode_xattrs(struct inode *inode) it.pos = erofs_iloc(inode) + vi->inode_isize; /* read in shared xattr array (non-atomic, see kmalloc below) */ - it.kaddr = erofs_bread(&it.buf, erofs_blknr(sb, it.pos), EROFS_KMAP); + it.kaddr = erofs_bread(&it.buf, it.pos, EROFS_KMAP); if (IS_ERR(it.kaddr)) { ret = PTR_ERR(it.kaddr); goto out_unlock; } - ih = it.kaddr + erofs_blkoff(sb, it.pos); + ih = it.kaddr; vi->xattr_name_filter = le32_to_cpu(ih->h_name_filter); vi->xattr_shared_count = ih->h_shared_count; vi->xattr_shared_xattrs = kmalloc_array(vi->xattr_shared_count, @@ -102,16 +102,14 @@ static int erofs_init_inode_xattrs(struct inode *inode) it.pos += sizeof(struct erofs_xattr_ibody_header); for (i = 0; i < vi->xattr_shared_count; ++i) { - it.kaddr = erofs_bread(&it.buf, erofs_blknr(sb, it.pos), - EROFS_KMAP); + it.kaddr = erofs_bread(&it.buf, it.pos, EROFS_KMAP); if (IS_ERR(it.kaddr)) { kfree(vi->xattr_shared_xattrs); vi->xattr_shared_xattrs = NULL; ret = PTR_ERR(it.kaddr); goto out_unlock; } - vi->xattr_shared_xattrs[i] = le32_to_cpu(*(__le32 *) - (it.kaddr + erofs_blkoff(sb, it.pos))); + vi->xattr_shared_xattrs[i] = le32_to_cpu(*(__le32 *)it.kaddr); it.pos += sizeof(__le32); } erofs_put_metabuf(&it.buf); @@ -185,12 +183,11 @@ static int erofs_xattr_copy_to_buffer(struct erofs_xattr_iter *it, void *src; for (processed = 0; processed < len; processed += slice) { - it->kaddr = erofs_bread(&it->buf, erofs_blknr(sb, it->pos), - EROFS_KMAP); + it->kaddr = erofs_bread(&it->buf, it->pos, EROFS_KMAP); if (IS_ERR(it->kaddr)) return PTR_ERR(it->kaddr); - src = it->kaddr + erofs_blkoff(sb, it->pos); + src = it->kaddr; slice = min_t(unsigned int, sb->s_blocksize - erofs_blkoff(sb, it->pos), len - processed); memcpy(it->buffer + it->buffer_ofs, src, slice); @@ -208,8 +205,7 @@ static int erofs_listxattr_foreach(struct erofs_xattr_iter *it) int err; /* 1. handle xattr entry */ - entry = *(struct erofs_xattr_entry *) - (it->kaddr + erofs_blkoff(it->sb, it->pos)); + entry = *(struct erofs_xattr_entry *)it->kaddr; it->pos += sizeof(struct erofs_xattr_entry); base_index = entry.e_name_index; @@ -259,8 +255,7 @@ static int erofs_getxattr_foreach(struct erofs_xattr_iter *it) unsigned int slice, processed, value_sz; /* 1. handle xattr entry */ - entry = *(struct erofs_xattr_entry *) - (it->kaddr + erofs_blkoff(sb, it->pos)); + entry = *(struct erofs_xattr_entry *)it->kaddr; it->pos += sizeof(struct erofs_xattr_entry); value_sz = le16_to_cpu(entry.e_value_size); @@ -291,8 +286,7 @@ static int erofs_getxattr_foreach(struct erofs_xattr_iter *it) /* 2. handle xattr name */ for (processed = 0; processed < entry.e_name_len; processed += slice) { - it->kaddr = erofs_bread(&it->buf, erofs_blknr(sb, it->pos), - EROFS_KMAP); + it->kaddr = erofs_bread(&it->buf, it->pos, EROFS_KMAP); if (IS_ERR(it->kaddr)) return PTR_ERR(it->kaddr); @@ -300,7 +294,7 @@ static int erofs_getxattr_foreach(struct erofs_xattr_iter *it) sb->s_blocksize - erofs_blkoff(sb, it->pos), entry.e_name_len - processed); if (memcmp(it->name.name + it->infix_len + processed, - it->kaddr + erofs_blkoff(sb, it->pos), slice)) + it->kaddr, slice)) return -ENOATTR; it->pos += slice; } @@ -336,13 +330,11 @@ static int erofs_xattr_iter_inline(struct erofs_xattr_iter *it, it->pos = erofs_iloc(inode) + vi->inode_isize + xattr_header_sz; while (remaining) { - it->kaddr = erofs_bread(&it->buf, erofs_blknr(it->sb, it->pos), - EROFS_KMAP); + it->kaddr = erofs_bread(&it->buf, it->pos, EROFS_KMAP); if (IS_ERR(it->kaddr)) return PTR_ERR(it->kaddr); - entry_sz = erofs_xattr_entry_size(it->kaddr + - erofs_blkoff(it->sb, it->pos)); + entry_sz = erofs_xattr_entry_size(it->kaddr); /* xattr on-disk corruption: xattr entry beyond xattr_isize */ if (remaining < entry_sz) { DBG_BUGON(1); @@ -375,8 +367,7 @@ static int erofs_xattr_iter_shared(struct erofs_xattr_iter *it, for (i = 0; i < vi->xattr_shared_count; ++i) { it->pos = erofs_pos(sb, sbi->xattr_blkaddr) + vi->xattr_shared_xattrs[i] * sizeof(__le32); - it->kaddr = erofs_bread(&it->buf, erofs_blknr(sb, it->pos), - EROFS_KMAP); + it->kaddr = erofs_bread(&it->buf, it->pos, EROFS_KMAP); if (IS_ERR(it->kaddr)) return PTR_ERR(it->kaddr); diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 3216b920d369..9ffdae7fcd5b 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -940,12 +940,12 @@ static int z_erofs_read_fragment(struct super_block *sb, struct page *page, for (; cur < end; cur += cnt, pos += cnt) { cnt = min_t(unsigned int, end - cur, sb->s_blocksize - erofs_blkoff(sb, pos)); - src = erofs_bread(&buf, erofs_blknr(sb, pos), EROFS_KMAP); + src = erofs_bread(&buf, pos, EROFS_KMAP); if (IS_ERR(src)) { erofs_put_metabuf(&buf); return PTR_ERR(src); } - memcpy_to_page(page, cur, src + erofs_blkoff(sb, pos), cnt); + memcpy_to_page(page, cur, src, cnt); } erofs_put_metabuf(&buf); return 0; -- cgit v1.2.3-59-g8ed1b From 210a03c9d51aa0e6e6f06980116e3256da8d4c48 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Thu, 28 Mar 2024 13:27:24 +0100 Subject: fs: claw back a few FMODE_* bits There's a bunch of flags that are purely based on what the file operations support while also never being conditionally set or unset. IOW, they're not subject to change for individual files. Imho, such flags don't need to live in f_mode they might as well live in the fops structs itself. And the fops struct already has that lonely mmap_supported_flags member. We might as well turn that into a generic fop_flags member and move a few flags from FMODE_* space into FOP_* space. That gets us four FMODE_* bits back and the ability for new static flags that are about file ops to not have to live in FMODE_* space but in their own FOP_* space. It's not the most beautiful thing ever but it gets the job done. Yes, there'll be an additional pointer chase but hopefully that won't matter for these flags. I suspect there's a few more we can move into there and that we can also redirect a bunch of new flag suggestions that follow this pattern into the fop_flags field instead of f_mode. Link: https://lore.kernel.org/r/20240328-gewendet-spargel-aa60a030ef74@brauner Reviewed-by: Christoph Hellwig Reviewed-by: Jan Kara Reviewed-by: Jens Axboe Signed-off-by: Christian Brauner --- block/bdev.c | 2 +- block/fops.c | 1 + drivers/dax/device.c | 2 +- fs/btrfs/file.c | 4 ++-- fs/ext4/file.c | 6 +++--- fs/f2fs/file.c | 3 ++- fs/read_write.c | 2 +- fs/xfs/xfs_file.c | 8 +++++--- include/linux/fs.h | 22 ++++++++++++---------- io_uring/io_uring.c | 2 +- io_uring/rw.c | 9 +++++---- mm/mmap.c | 4 +++- 12 files changed, 37 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/block/bdev.c b/block/bdev.c index 7a5f611c3d2e..1322dfe32c5d 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -904,7 +904,7 @@ int bdev_open(struct block_device *bdev, blk_mode_t mode, void *holder, disk_unblock_events(disk); bdev_file->f_flags |= O_LARGEFILE; - bdev_file->f_mode |= FMODE_BUF_RASYNC | FMODE_CAN_ODIRECT; + bdev_file->f_mode |= FMODE_CAN_ODIRECT; if (bdev_nowait(bdev)) bdev_file->f_mode |= FMODE_NOWAIT; bdev_file->f_mapping = bdev->bd_inode->i_mapping; diff --git a/block/fops.c b/block/fops.c index 679d9b752fe8..af6c244314af 100644 --- a/block/fops.c +++ b/block/fops.c @@ -863,6 +863,7 @@ const struct file_operations def_blk_fops = { .splice_read = filemap_splice_read, .splice_write = iter_file_splice_write, .fallocate = blkdev_fallocate, + .fop_flags = FOP_BUFFER_RASYNC, }; static __init int blkdev_init(void) diff --git a/drivers/dax/device.c b/drivers/dax/device.c index 93ebedc5ec8c..c24ef4d3cf31 100644 --- a/drivers/dax/device.c +++ b/drivers/dax/device.c @@ -377,7 +377,7 @@ static const struct file_operations dax_fops = { .release = dax_release, .get_unmapped_area = dax_get_unmapped_area, .mmap = dax_mmap, - .mmap_supported_flags = MAP_SYNC, + .fop_flags = FOP_MMAP_SYNC, }; static void dev_dax_cdev_del(void *cdev) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index f9d76072398d..1640c46f2153 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -3719,8 +3719,7 @@ static int btrfs_file_open(struct inode *inode, struct file *filp) { int ret; - filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC | FMODE_BUF_WASYNC | - FMODE_CAN_ODIRECT; + filp->f_mode |= FMODE_NOWAIT | FMODE_CAN_ODIRECT; ret = fsverity_file_open(inode, filp); if (ret) @@ -3850,6 +3849,7 @@ const struct file_operations btrfs_file_operations = { .compat_ioctl = btrfs_compat_ioctl, #endif .remap_file_range = btrfs_remap_file_range, + .fop_flags = FOP_BUFFER_RASYNC | FOP_BUFFER_WASYNC, }; int btrfs_fdatawrite_range(struct inode *inode, loff_t start, loff_t end) diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 54d6ff22585c..28c51b0cc4db 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -885,8 +885,7 @@ static int ext4_file_open(struct inode *inode, struct file *filp) return ret; } - filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC | - FMODE_DIO_PARALLEL_WRITE; + filp->f_mode |= FMODE_NOWAIT; return dquot_file_open(inode, filp); } @@ -938,7 +937,6 @@ const struct file_operations ext4_file_operations = { .compat_ioctl = ext4_compat_ioctl, #endif .mmap = ext4_file_mmap, - .mmap_supported_flags = MAP_SYNC, .open = ext4_file_open, .release = ext4_release_file, .fsync = ext4_sync_file, @@ -946,6 +944,8 @@ const struct file_operations ext4_file_operations = { .splice_read = ext4_file_splice_read, .splice_write = iter_file_splice_write, .fallocate = ext4_fallocate, + .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | + FOP_DIO_PARALLEL_WRITE, }; const struct inode_operations ext4_file_inode_operations = { diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 1761ad125f97..2b65e09822d4 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -569,7 +569,7 @@ static int f2fs_file_open(struct inode *inode, struct file *filp) if (err) return err; - filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC; + filp->f_mode |= FMODE_NOWAIT; filp->f_mode |= FMODE_CAN_ODIRECT; return dquot_file_open(inode, filp); @@ -5045,4 +5045,5 @@ const struct file_operations f2fs_file_operations = { .splice_read = f2fs_file_splice_read, .splice_write = iter_file_splice_write, .fadvise = f2fs_file_fadvise, + .fop_flags = FOP_BUFFER_RASYNC, }; diff --git a/fs/read_write.c b/fs/read_write.c index d4c036e82b6c..2115d1f40bd5 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -1685,7 +1685,7 @@ int generic_write_checks_count(struct kiocb *iocb, loff_t *count) if ((iocb->ki_flags & IOCB_NOWAIT) && !((iocb->ki_flags & IOCB_DIRECT) || - (file->f_mode & FMODE_BUF_WASYNC))) + (file->f_op->fop_flags & FOP_BUFFER_WASYNC))) return -EINVAL; return generic_write_check_limits(iocb->ki_filp, iocb->ki_pos, count); diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 632653e00906..147439ad3581 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1230,8 +1230,7 @@ xfs_file_open( { if (xfs_is_shutdown(XFS_M(inode->i_sb))) return -EIO; - file->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC | FMODE_BUF_WASYNC | - FMODE_DIO_PARALLEL_WRITE | FMODE_CAN_ODIRECT; + file->f_mode |= FMODE_NOWAIT | FMODE_CAN_ODIRECT; return generic_file_open(inode, file); } @@ -1490,7 +1489,6 @@ const struct file_operations xfs_file_operations = { .compat_ioctl = xfs_file_compat_ioctl, #endif .mmap = xfs_file_mmap, - .mmap_supported_flags = MAP_SYNC, .open = xfs_file_open, .release = xfs_file_release, .fsync = xfs_file_fsync, @@ -1498,6 +1496,8 @@ const struct file_operations xfs_file_operations = { .fallocate = xfs_file_fallocate, .fadvise = xfs_file_fadvise, .remap_file_range = xfs_file_remap_range, + .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | FOP_BUFFER_WASYNC | + FOP_DIO_PARALLEL_WRITE, }; const struct file_operations xfs_dir_file_operations = { @@ -1510,4 +1510,6 @@ const struct file_operations xfs_dir_file_operations = { .compat_ioctl = xfs_file_compat_ioctl, #endif .fsync = xfs_dir_fsync, + .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | FOP_BUFFER_WASYNC | + FOP_DIO_PARALLEL_WRITE, }; diff --git a/include/linux/fs.h b/include/linux/fs.h index 914245085597..53cddf0131ca 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -163,9 +163,6 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset, #define FMODE_NOREUSE ((__force fmode_t)0x800000) -/* File supports non-exclusive O_DIRECT writes from multiple threads */ -#define FMODE_DIO_PARALLEL_WRITE ((__force fmode_t)0x1000000) - /* File is embedded in backing_file object */ #define FMODE_BACKING ((__force fmode_t)0x2000000) @@ -181,12 +178,6 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset, /* File does not contribute to nr_files count */ #define FMODE_NOACCOUNT ((__force fmode_t)0x20000000) -/* File supports async buffered reads */ -#define FMODE_BUF_RASYNC ((__force fmode_t)0x40000000) - -/* File supports async nowait buffered writes */ -#define FMODE_BUF_WASYNC ((__force fmode_t)0x80000000) - /* * Attribute flags. These should be or-ed together to figure out what * has been changed! @@ -2001,8 +1992,11 @@ struct iov_iter; struct io_uring_cmd; struct offset_ctx; +typedef unsigned int __bitwise fop_flags_t; + struct file_operations { struct module *owner; + fop_flags_t fop_flags; loff_t (*llseek) (struct file *, loff_t, int); ssize_t (*read) (struct file *, char __user *, size_t, loff_t *); ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *); @@ -2015,7 +2009,6 @@ struct file_operations { long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long); long (*compat_ioctl) (struct file *, unsigned int, unsigned long); int (*mmap) (struct file *, struct vm_area_struct *); - unsigned long mmap_supported_flags; int (*open) (struct inode *, struct file *); int (*flush) (struct file *, fl_owner_t id); int (*release) (struct inode *, struct file *); @@ -2046,6 +2039,15 @@ struct file_operations { unsigned int poll_flags); } __randomize_layout; +/* Supports async buffered reads */ +#define FOP_BUFFER_RASYNC ((__force fop_flags_t)(1 << 0)) +/* Supports async buffered writes */ +#define FOP_BUFFER_WASYNC ((__force fop_flags_t)(1 << 1)) +/* Supports synchronous page faults for mappings */ +#define FOP_MMAP_SYNC ((__force fop_flags_t)(1 << 2)) +/* Supports non-exclusive O_DIRECT writes from multiple threads */ +#define FOP_DIO_PARALLEL_WRITE ((__force fop_flags_t)(1 << 3)) + /* Wrap a directory iterator that needs exclusive inode access */ int wrap_directory_iterator(struct file *, struct dir_context *, int (*) (struct file *, struct dir_context *)); diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 5d4b448fdc50..d73c9ad2d2f8 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -471,7 +471,7 @@ static void io_prep_async_work(struct io_kiocb *req) /* don't serialize this request if the fs doesn't need it */ if (should_hash && (req->file->f_flags & O_DIRECT) && - (req->file->f_mode & FMODE_DIO_PARALLEL_WRITE)) + (req->file->f_op->fop_flags & FOP_DIO_PARALLEL_WRITE)) should_hash = false; if (should_hash || (ctx->flags & IORING_SETUP_IOPOLL)) io_wq_hash_work(&req->work, file_inode(req->file)); diff --git a/io_uring/rw.c b/io_uring/rw.c index 0585ebcc9773..d9dfde1142a1 100644 --- a/io_uring/rw.c +++ b/io_uring/rw.c @@ -683,7 +683,8 @@ static bool io_rw_should_retry(struct io_kiocb *req) * just use poll if we can, and don't attempt if the fs doesn't * support callback based unlocks */ - if (io_file_can_poll(req) || !(req->file->f_mode & FMODE_BUF_RASYNC)) + if (io_file_can_poll(req) || + !(req->file->f_op->fop_flags & FOP_BUFFER_RASYNC)) return false; wait->wait.func = io_async_buf_func; @@ -1022,10 +1023,10 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags) if (unlikely(!io_file_supports_nowait(req))) goto copy_iov; - /* File path supports NOWAIT for non-direct_IO only for block devices. */ + /* Check if we can support NOWAIT. */ if (!(kiocb->ki_flags & IOCB_DIRECT) && - !(kiocb->ki_filp->f_mode & FMODE_BUF_WASYNC) && - (req->flags & REQ_F_ISREG)) + !(req->file->f_op->fop_flags & FOP_BUFFER_WASYNC) && + (req->flags & REQ_F_ISREG)) goto copy_iov; kiocb->ki_flags |= IOCB_NOWAIT; diff --git a/mm/mmap.c b/mm/mmap.c index 6dbda99a47da..3490af70f259 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1294,7 +1294,9 @@ unsigned long do_mmap(struct file *file, unsigned long addr, if (!file_mmap_ok(file, inode, pgoff, len)) return -EOVERFLOW; - flags_mask = LEGACY_MAP_MASK | file->f_op->mmap_supported_flags; + flags_mask = LEGACY_MAP_MASK; + if (file->f_op->fop_flags & FOP_MMAP_SYNC) + flags_mask |= MAP_SYNC; switch (flags & MAP_TYPE) { case MAP_SHARED: -- cgit v1.2.3-59-g8ed1b From 09e913f5826936c0f6632d6d0d35a36295fac7cc Mon Sep 17 00:00:00 2001 From: Hongbo Li Date: Tue, 26 Mar 2024 12:04:56 +0800 Subject: bcachefs: fix the count of nr_freed_pcpu after changing bc->freed_nonpcpu list When allocating bkey_cached from bc->freed_pcpu list, it missed decreasing the count of nr_freed_pcpu which would cause the mismatch between the value of nr_freed_pcpu and the list items. This problem also exists in moving new bkey_cached to bc->freed_pcpu list. If these happened, the bug info may appear in bch2_fs_btree_key_cache_exit by the follow code: BUG_ON(list_count_nodes(&bc->freed_pcpu) != bc->nr_freed_pcpu); BUG_ON(list_count_nodes(&bc->freed_nonpcpu) != bc->nr_freed_nonpcpu); Fixes: c65c13f0eac6 ("bcachefs: Run btree key cache shrinker less aggressively") Signed-off-by: Hongbo Li Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 4ae7890f07c6..88a3582a3275 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -169,6 +169,7 @@ static void bkey_cached_move_to_freelist(struct btree_key_cache *bc, } else { mutex_lock(&bc->lock); list_move_tail(&ck->list, &bc->freed_pcpu); + bc->nr_freed_pcpu++; mutex_unlock(&bc->lock); } } @@ -245,6 +246,7 @@ bkey_cached_alloc(struct btree_trans *trans, struct btree_path *path, if (!list_empty(&bc->freed_pcpu)) { ck = list_last_entry(&bc->freed_pcpu, struct bkey_cached, list); list_del_init(&ck->list); + bc->nr_freed_pcpu--; } mutex_unlock(&bc->lock); } -- cgit v1.2.3-59-g8ed1b From b897b148ee30c7fca995e6d15cf791f52993920b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 16:20:17 -0400 Subject: bcachefs: fix bch2_get_acl() transaction restart handling bch2_acl_from_disk() uses allocate_dropping_locks, and can thus return a transaction restart - this wasn't handled. Signed-off-by: Kent Overstreet --- fs/bcachefs/acl.c | 30 ++++++++++++++---------------- 1 file changed, 14 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/acl.c b/fs/bcachefs/acl.c index 3640f417cce1..5c180fdc3efb 100644 --- a/fs/bcachefs/acl.c +++ b/fs/bcachefs/acl.c @@ -281,7 +281,6 @@ struct posix_acl *bch2_get_acl(struct mnt_idmap *idmap, struct xattr_search_key search = X_SEARCH(acl_to_xattr_type(type), "", 0); struct btree_trans *trans = bch2_trans_get(c); struct btree_iter iter = { NULL }; - struct bkey_s_c_xattr xattr; struct posix_acl *acl = NULL; struct bkey_s_c k; int ret; @@ -290,28 +289,27 @@ retry: ret = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, &hash, inode_inum(inode), &search, 0); - if (ret) { - if (!bch2_err_matches(ret, ENOENT)) - acl = ERR_PTR(ret); - goto out; - } + if (ret) + goto err; k = bch2_btree_iter_peek_slot(&iter); ret = bkey_err(k); - if (ret) { - acl = ERR_PTR(ret); - goto out; - } + if (ret) + goto err; - xattr = bkey_s_c_to_xattr(k); + struct bkey_s_c_xattr xattr = bkey_s_c_to_xattr(k); acl = bch2_acl_from_disk(trans, xattr_val(xattr.v), - le16_to_cpu(xattr.v->x_val_len)); + le16_to_cpu(xattr.v->x_val_len)); + ret = PTR_ERR_OR_ZERO(acl); +err: + if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) + goto retry; - if (!IS_ERR(acl)) + if (ret) + acl = !bch2_err_matches(ret, ENOENT) ? ERR_PTR(ret) : NULL; + + if (!IS_ERR_OR_NULL(acl)) set_cached_acl(&inode->v, type, acl); -out: - if (bch2_err_matches(PTR_ERR_OR_ZERO(acl), BCH_ERR_transaction_restart)) - goto retry; bch2_trans_iter_exit(trans, &iter); bch2_trans_put(trans); -- cgit v1.2.3-59-g8ed1b From 9c432404b9555c9444cbf6c8feaf52c0d8cad486 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 8 Apr 2024 22:32:08 -0400 Subject: bcachefs: fix eytzinger0_find_gt() - fix return types: promoting from unsigned to ssize_t does not do what we want here, and was pointless since the rest of the eytzinger code is u32 - nr, not size Signed-off-by: Kent Overstreet --- fs/bcachefs/eytzinger.h | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/eytzinger.h b/fs/bcachefs/eytzinger.h index ee0e2df33322..24840aee335c 100644 --- a/fs/bcachefs/eytzinger.h +++ b/fs/bcachefs/eytzinger.h @@ -242,8 +242,8 @@ static inline unsigned inorder_to_eytzinger0(unsigned i, unsigned size) (_i) = eytzinger0_next((_i), (_size))) /* return greatest node <= @search, or -1 if not found */ -static inline ssize_t eytzinger0_find_le(void *base, size_t nr, size_t size, - cmp_func_t cmp, const void *search) +static inline int eytzinger0_find_le(void *base, size_t nr, size_t size, + cmp_func_t cmp, const void *search) { unsigned i, n = 0; @@ -256,18 +256,32 @@ static inline ssize_t eytzinger0_find_le(void *base, size_t nr, size_t size, } while (n < nr); if (n & 1) { - /* @i was greater than @search, return previous node: */ + /* + * @i was greater than @search, return previous node: + * + * if @i was leftmost/smallest element, + * eytzinger0_prev(eytzinger0_first())) returns -1, as expected + */ return eytzinger0_prev(i, nr); } else { return i; } } -static inline ssize_t eytzinger0_find_gt(void *base, size_t nr, size_t size, - cmp_func_t cmp, const void *search) +static inline int eytzinger0_find_gt(void *base, size_t nr, size_t size, + cmp_func_t cmp, const void *search) { ssize_t idx = eytzinger0_find_le(base, nr, size, cmp, search); - return eytzinger0_next(idx, size); + + /* + * if eytitzinger0_find_le() returned -1 - no element was <= search - we + * want to return the first element; next/prev identities mean this work + * as expected + * + * similarly if find_le() returns last element, we should return -1; + * identities mean this all works out: + */ + return eytzinger0_next(idx, nr); } #define eytzinger0_find(base, nr, size, _cmp, search) \ -- cgit v1.2.3-59-g8ed1b From 359571c327a726d622786aef3833637dacfd5d38 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 00:02:47 -0400 Subject: bcachefs: Fix check_topology() when using node scan shoot down journal keys _before_ populating journal keys with pointers to scanned nodes Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 6280da1244b5..a1c2c0ffb0cc 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -544,12 +544,12 @@ reconstruct_root: bch2_btree_root_alloc_fake(c, i, 0); } else { bch2_btree_root_alloc_fake(c, i, 1); + bch2_shoot_down_journal_keys(c, i, 1, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); ret = bch2_get_scanned_nodes(c, i, 0, POS_MIN, SPOS_MAX); if (ret) break; } - bch2_shoot_down_journal_keys(c, i, 1, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); reconstructed_root = true; } -- cgit v1.2.3-59-g8ed1b From 5ab4beb759c05c74fb385ac5ca0ade5d3db67975 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 00:49:39 -0400 Subject: bcachefs: Don't scan for btree nodes when we can reconstruct Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 14 ++++++++++++++ fs/bcachefs/btree_gc.c | 11 ++++++++--- fs/bcachefs/btree_node_scan.c | 8 +++++++- fs/bcachefs/recovery.c | 14 -------------- 4 files changed, 29 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 63102992d955..364ae42022af 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -1535,6 +1535,20 @@ enum btree_id { BTREE_ID_NR }; +static inline bool btree_id_is_alloc(enum btree_id id) +{ + switch (id) { + case BTREE_ID_alloc: + case BTREE_ID_backpointers: + case BTREE_ID_need_discard: + case BTREE_ID_freespace: + case BTREE_ID_bucket_gens: + return true; + default: + return false; + } +} + #define BTREE_MAX_DEPTH 4U /* Btree nodes */ diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index a1c2c0ffb0cc..d2555da55c6d 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -368,11 +368,16 @@ again: buf.buf)) { bch2_btree_node_evict(trans, cur_k.k); cur = NULL; - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes) ?: - bch2_journal_key_delete(c, b->c.btree_id, - b->c.level, cur_k.k->k.p); + ret = bch2_journal_key_delete(c, b->c.btree_id, + b->c.level, cur_k.k->k.p); if (ret) break; + + if (!btree_id_is_alloc(b->c.btree_id)) { + ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes); + if (ret) + break; + } continue; } diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index 3f33be7e5e5c..a7d0593b3871 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -133,6 +133,9 @@ static void try_read_btree_node(struct find_btree_nodes *f, struct bch_dev *ca, if (le64_to_cpu(bn->magic) != bset_magic(c)) return; + if (btree_id_is_alloc(BTREE_NODE_ID(bn))) + return; + rcu_read_lock(); struct found_btree_node n = { .btree_id = BTREE_NODE_ID(bn), @@ -290,7 +293,7 @@ again: found_btree_node_to_text(&buf, c, n); bch_err(c, "%s", buf.buf); printbuf_exit(&buf); - return -1; + return -BCH_ERR_fsck_repair_unimplemented; } } @@ -436,6 +439,9 @@ bool bch2_btree_has_scanned_nodes(struct bch_fs *c, enum btree_id btree) int bch2_get_scanned_nodes(struct bch_fs *c, enum btree_id btree, unsigned level, struct bpos node_min, struct bpos node_max) { + if (btree_id_is_alloc(btree)) + return 0; + struct find_btree_nodes *f = &c->found_btree_nodes; int ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes); diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index b76c16152579..0f328aba9760 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -47,20 +47,6 @@ void bch2_btree_lost_data(struct bch_fs *c, enum btree_id btree) } } -static bool btree_id_is_alloc(enum btree_id id) -{ - switch (id) { - case BTREE_ID_alloc: - case BTREE_ID_backpointers: - case BTREE_ID_need_discard: - case BTREE_ID_freespace: - case BTREE_ID_bucket_gens: - return true; - default: - return false; - } -} - /* for -o reconstruct_alloc: */ static void bch2_reconstruct_alloc(struct bch_fs *c) { -- cgit v1.2.3-59-g8ed1b From c473bcdd80d4ab2ae79a7a509a6712818366e32a Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 28 Mar 2024 15:30:44 +0100 Subject: nilfs2: fix out-of-range warning clang-14 points out that v_size is always smaller than a 64KB page size if that is configured by the CPU architecture: fs/nilfs2/ioctl.c:63:19: error: result of comparison of constant 65536 with expression of type '__u16' (aka 'unsigned short') is always false [-Werror,-Wtautological-constant-out-of-range-compare] if (argv->v_size > PAGE_SIZE) ~~~~~~~~~~~~ ^ ~~~~~~~~~ This is ok, so just shut up that warning with a cast. Signed-off-by: Arnd Bergmann Link: https://lore.kernel.org/r/20240328143051.1069575-7-arnd@kernel.org Fixes: 3358b4aaa84f ("nilfs2: fix problems of memory allocation in ioctl") Acked-by: Ryusuke Konishi Reviewed-by: Justin Stitt Signed-off-by: Christian Brauner --- fs/nilfs2/ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index f1a01c191cf5..8be471ce4f19 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -60,7 +60,7 @@ static int nilfs_ioctl_wrap_copy(struct the_nilfs *nilfs, if (argv->v_nmembs == 0) return 0; - if (argv->v_size > PAGE_SIZE) + if ((size_t)argv->v_size > PAGE_SIZE) return -EINVAL; /* -- cgit v1.2.3-59-g8ed1b From 629171657a2864d819a3bbecabe0a5e001d05c7a Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Fri, 22 Mar 2024 21:41:18 +0000 Subject: orangefs: cleanup uses of strncpy strncpy() is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. There is some care taken to ensure these destination buffers are NUL-terminated by bounding the strncpy()'s by ORANGEFS_NAME_MAX - 1 or ORANGEFS_MAX_SERVER_ADDR_LEN - 1. Instead, we can use the new 2-argument version of strscpy() to guarantee NUL-termination on the destination buffers while simplifying the code. Based on usage with printf-likes, we can see these buffers are expected to be NUL-terminated: | gossip_debug(GOSSIP_NAME_DEBUG, | "%s: doing lookup on %s under %pU,%d\n", | __func__, | new_op->upcall.req.lookup.d_name, | &new_op->upcall.req.lookup.parent_refn.khandle, | new_op->upcall.req.lookup.parent_refn.fs_id); ... | gossip_debug(GOSSIP_SUPER_DEBUG, | "Attempting ORANGEFS Remount via host %s\n", | new_op->upcall.req.fs_mount.orangefs_config_server); NUL-padding isn't required for any of these destination buffers as they've all been zero-allocated with op_alloc() or kzalloc(). Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2] Link: https://github.com/KSPP/linux/issues/90 Cc: linux-hardening@vger.kernel.org Signed-off-by: Justin Stitt Link: https://lore.kernel.org/r/20240322-strncpy-fs-orangefs-dcache-c-v1-1-15d12debbf38@google.com Reviewed-by: Kees Cook Signed-off-by: Christian Brauner --- fs/orangefs/dcache.c | 4 +--- fs/orangefs/namei.c | 26 ++++++++------------------ fs/orangefs/super.c | 17 ++++++----------- 3 files changed, 15 insertions(+), 32 deletions(-) (limited to 'fs') diff --git a/fs/orangefs/dcache.c b/fs/orangefs/dcache.c index 8bbe9486e3a6..395a00ed8ac7 100644 --- a/fs/orangefs/dcache.c +++ b/fs/orangefs/dcache.c @@ -33,9 +33,7 @@ static int orangefs_revalidate_lookup(struct dentry *dentry) new_op->upcall.req.lookup.sym_follow = ORANGEFS_LOOKUP_LINK_NO_FOLLOW; new_op->upcall.req.lookup.parent_refn = parent->refn; - strncpy(new_op->upcall.req.lookup.d_name, - dentry->d_name.name, - ORANGEFS_NAME_MAX - 1); + strscpy(new_op->upcall.req.lookup.d_name, dentry->d_name.name); gossip_debug(GOSSIP_DCACHE_DEBUG, "%s:%s:%d interrupt flag [%d]\n", diff --git a/fs/orangefs/namei.c b/fs/orangefs/namei.c index c9dfd5c6a097..200558ec72f0 100644 --- a/fs/orangefs/namei.c +++ b/fs/orangefs/namei.c @@ -41,8 +41,7 @@ static int orangefs_create(struct mnt_idmap *idmap, fill_default_sys_attrs(new_op->upcall.req.create.attributes, ORANGEFS_TYPE_METAFILE, mode); - strncpy(new_op->upcall.req.create.d_name, - dentry->d_name.name, ORANGEFS_NAME_MAX - 1); + strscpy(new_op->upcall.req.create.d_name, dentry->d_name.name); ret = service_operation(new_op, __func__, get_interruptible_flag(dir)); @@ -137,8 +136,7 @@ static struct dentry *orangefs_lookup(struct inode *dir, struct dentry *dentry, &parent->refn.khandle); new_op->upcall.req.lookup.parent_refn = parent->refn; - strncpy(new_op->upcall.req.lookup.d_name, dentry->d_name.name, - ORANGEFS_NAME_MAX - 1); + strscpy(new_op->upcall.req.lookup.d_name, dentry->d_name.name); gossip_debug(GOSSIP_NAME_DEBUG, "%s: doing lookup on %s under %pU,%d\n", @@ -192,8 +190,7 @@ static int orangefs_unlink(struct inode *dir, struct dentry *dentry) return -ENOMEM; new_op->upcall.req.remove.parent_refn = parent->refn; - strncpy(new_op->upcall.req.remove.d_name, dentry->d_name.name, - ORANGEFS_NAME_MAX - 1); + strscpy(new_op->upcall.req.remove.d_name, dentry->d_name.name); ret = service_operation(new_op, "orangefs_unlink", get_interruptible_flag(inode)); @@ -247,10 +244,8 @@ static int orangefs_symlink(struct mnt_idmap *idmap, ORANGEFS_TYPE_SYMLINK, mode); - strncpy(new_op->upcall.req.sym.entry_name, - dentry->d_name.name, - ORANGEFS_NAME_MAX - 1); - strncpy(new_op->upcall.req.sym.target, symname, ORANGEFS_NAME_MAX - 1); + strscpy(new_op->upcall.req.sym.entry_name, dentry->d_name.name); + strscpy(new_op->upcall.req.sym.target, symname); ret = service_operation(new_op, __func__, get_interruptible_flag(dir)); @@ -324,8 +319,7 @@ static int orangefs_mkdir(struct mnt_idmap *idmap, struct inode *dir, fill_default_sys_attrs(new_op->upcall.req.mkdir.attributes, ORANGEFS_TYPE_DIRECTORY, mode); - strncpy(new_op->upcall.req.mkdir.d_name, - dentry->d_name.name, ORANGEFS_NAME_MAX - 1); + strscpy(new_op->upcall.req.mkdir.d_name, dentry->d_name.name); ret = service_operation(new_op, __func__, get_interruptible_flag(dir)); @@ -405,12 +399,8 @@ static int orangefs_rename(struct mnt_idmap *idmap, new_op->upcall.req.rename.old_parent_refn = ORANGEFS_I(old_dir)->refn; new_op->upcall.req.rename.new_parent_refn = ORANGEFS_I(new_dir)->refn; - strncpy(new_op->upcall.req.rename.d_old_name, - old_dentry->d_name.name, - ORANGEFS_NAME_MAX - 1); - strncpy(new_op->upcall.req.rename.d_new_name, - new_dentry->d_name.name, - ORANGEFS_NAME_MAX - 1); + strscpy(new_op->upcall.req.rename.d_old_name, old_dentry->d_name.name); + strscpy(new_op->upcall.req.rename.d_new_name, new_dentry->d_name.name); ret = service_operation(new_op, "orangefs_rename", diff --git a/fs/orangefs/super.c b/fs/orangefs/super.c index 34849b4a3243..fb4d09c2f531 100644 --- a/fs/orangefs/super.c +++ b/fs/orangefs/super.c @@ -253,9 +253,8 @@ int orangefs_remount(struct orangefs_sb_info_s *orangefs_sb) new_op = op_alloc(ORANGEFS_VFS_OP_FS_MOUNT); if (!new_op) return -ENOMEM; - strncpy(new_op->upcall.req.fs_mount.orangefs_config_server, - orangefs_sb->devname, - ORANGEFS_MAX_SERVER_ADDR_LEN); + strscpy(new_op->upcall.req.fs_mount.orangefs_config_server, + orangefs_sb->devname); gossip_debug(GOSSIP_SUPER_DEBUG, "Attempting ORANGEFS Remount via host %s\n", @@ -400,8 +399,7 @@ static int orangefs_unmount(int id, __s32 fs_id, const char *devname) return -ENOMEM; op->upcall.req.fs_umount.id = id; op->upcall.req.fs_umount.fs_id = fs_id; - strncpy(op->upcall.req.fs_umount.orangefs_config_server, - devname, ORANGEFS_MAX_SERVER_ADDR_LEN - 1); + strscpy(op->upcall.req.fs_umount.orangefs_config_server, devname); r = service_operation(op, "orangefs_fs_umount", 0); /* Not much to do about an error here. */ if (r) @@ -494,9 +492,7 @@ struct dentry *orangefs_mount(struct file_system_type *fst, if (!new_op) return ERR_PTR(-ENOMEM); - strncpy(new_op->upcall.req.fs_mount.orangefs_config_server, - devname, - ORANGEFS_MAX_SERVER_ADDR_LEN - 1); + strscpy(new_op->upcall.req.fs_mount.orangefs_config_server, devname); gossip_debug(GOSSIP_SUPER_DEBUG, "Attempting ORANGEFS Mount via host %s\n", @@ -543,9 +539,8 @@ struct dentry *orangefs_mount(struct file_system_type *fst, * on successful mount, store the devname and data * used */ - strncpy(ORANGEFS_SB(sb)->devname, - devname, - ORANGEFS_MAX_SERVER_ADDR_LEN - 1); + strscpy(ORANGEFS_SB(sb)->devname, devname); + /* mount_pending must be cleared */ ORANGEFS_SB(sb)->mount_pending = 0; -- cgit v1.2.3-59-g8ed1b From 886b94d25a8eba4c42634dddc3cbfd6391a24d25 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 9 Apr 2024 10:41:33 +0200 Subject: fs: Add FOP_HUGE_PAGES Instead of checking for specific file_operations, add a bit to file_operations which denotes a file that only contain hugetlb pages. This lets us make hugetlbfs_file_operations static, and removes is_file_shm_hugepages() completely. Signed-off-by: Matthew Wilcox (Oracle) Link: https://lore.kernel.org/r/20240407201122.3783877-1-willy@infradead.org Signed-off-by: Christian Brauner --- fs/hugetlbfs/inode.c | 5 +++-- include/linux/fs.h | 2 ++ include/linux/hugetlb.h | 8 ++------ include/linux/shm.h | 5 ----- ipc/shm.c | 10 +++------- 5 files changed, 10 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6502c7e776d1..34ac73cc36b1 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -40,7 +40,7 @@ #include static const struct address_space_operations hugetlbfs_aops; -const struct file_operations hugetlbfs_file_operations; +static const struct file_operations hugetlbfs_file_operations; static const struct inode_operations hugetlbfs_dir_inode_operations; static const struct inode_operations hugetlbfs_inode_operations; @@ -1301,13 +1301,14 @@ static void init_once(void *foo) inode_init_once(&ei->vfs_inode); } -const struct file_operations hugetlbfs_file_operations = { +static const struct file_operations hugetlbfs_file_operations = { .read_iter = hugetlbfs_read_iter, .mmap = hugetlbfs_file_mmap, .fsync = noop_fsync, .get_unmapped_area = hugetlb_get_unmapped_area, .llseek = default_llseek, .fallocate = hugetlbfs_fallocate, + .fop_flags = FOP_HUGE_PAGES, }; static const struct inode_operations hugetlbfs_dir_inode_operations = { diff --git a/include/linux/fs.h b/include/linux/fs.h index 2ba98930c4c4..b053f98393e6 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2052,6 +2052,8 @@ struct file_operations { #define FOP_MMAP_SYNC ((__force fop_flags_t)(1 << 2)) /* Supports non-exclusive O_DIRECT writes from multiple threads */ #define FOP_DIO_PARALLEL_WRITE ((__force fop_flags_t)(1 << 3)) +/* Contains huge pages */ +#define FOP_HUGE_PAGES ((__force fop_flags_t)(1 << 4)) /* Wrap a directory iterator that needs exclusive inode access */ int wrap_directory_iterator(struct file *, struct dir_context *, diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 77b30a8c6076..b06f7c426d38 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -554,17 +554,13 @@ static inline struct hugetlbfs_inode_info *HUGETLBFS_I(struct inode *inode) return container_of(inode, struct hugetlbfs_inode_info, vfs_inode); } -extern const struct file_operations hugetlbfs_file_operations; extern const struct vm_operations_struct hugetlb_vm_ops; struct file *hugetlb_file_setup(const char *name, size_t size, vm_flags_t acct, int creat_flags, int page_size_log); -static inline bool is_file_hugepages(struct file *file) +static inline bool is_file_hugepages(const struct file *file) { - if (file->f_op == &hugetlbfs_file_operations) - return true; - - return is_file_shm_hugepages(file); + return file->f_op->fop_flags & FOP_HUGE_PAGES; } static inline struct hstate *hstate_inode(struct inode *i) diff --git a/include/linux/shm.h b/include/linux/shm.h index c55bef0538e5..1d3d3ae958fb 100644 --- a/include/linux/shm.h +++ b/include/linux/shm.h @@ -16,7 +16,6 @@ struct sysv_shm { long do_shmat(int shmid, char __user *shmaddr, int shmflg, unsigned long *addr, unsigned long shmlba); -bool is_file_shm_hugepages(struct file *file); void exit_shm(struct task_struct *task); #define shm_init_task(task) INIT_LIST_HEAD(&(task)->sysvshm.shm_clist) #else @@ -30,10 +29,6 @@ static inline long do_shmat(int shmid, char __user *shmaddr, { return -ENOSYS; } -static inline bool is_file_shm_hugepages(struct file *file) -{ - return false; -} static inline void exit_shm(struct task_struct *task) { } diff --git a/ipc/shm.c b/ipc/shm.c index a89f001a8bf0..3e3071252dac 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -662,8 +662,8 @@ static const struct file_operations shm_file_operations = { }; /* - * shm_file_operations_huge is now identical to shm_file_operations, - * but we keep it distinct for the sake of is_file_shm_hugepages(). + * shm_file_operations_huge is now identical to shm_file_operations + * except for fop_flags */ static const struct file_operations shm_file_operations_huge = { .mmap = shm_mmap, @@ -672,13 +672,9 @@ static const struct file_operations shm_file_operations_huge = { .get_unmapped_area = shm_get_unmapped_area, .llseek = noop_llseek, .fallocate = shm_fallocate, + .fop_flags = FOP_HUGE_PAGES, }; -bool is_file_shm_hugepages(struct file *file) -{ - return file->f_op == &shm_file_operations_huge; -} - static const struct vm_operations_struct shm_vm_ops = { .open = shm_open, /* callback for a new vm-area open */ .close = shm_close, /* callback for when the vm-area is released */ -- cgit v1.2.3-59-g8ed1b From abe6acfa7d7b666d785eae706bd34b63f3c2b11f Mon Sep 17 00:00:00 2001 From: Günther Noack Date: Fri, 5 Apr 2024 21:40:29 +0000 Subject: fs: Return ENOTTY directly if FS_IOC_GETUUID or FS_IOC_GETFSSYSFSPATH fail MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit These IOCTL commands should be implemented by setting attributes on the superblock, rather than in the IOCTL hooks in struct file_operations. By returning -ENOTTY instead of -ENOIOCTLCMD, we instruct the fs/ioctl.c logic to return -ENOTTY immediately, rather than attempting to call f_op->unlocked_ioctl() or f_op->compat_ioctl() as a fallback. Why this is safe: Before this change, fs/ioctl.c would unsuccessfully attempt calling the IOCTL hooks, and then return -ENOTTY. By returning -ENOTTY directly, we return the same error code immediately, but save ourselves the fallback attempt. Motivation: This simplifies the logic for these IOCTL commands and lets us reason about the side effects of these IOCTLs more easily. It will be possible to permit these IOCTLs under LSM IOCTL policies, without having to worry about them getting dispatched to problematic device drivers (which sometimes do work before looking at the IOCTL command number). Link: https://lore.kernel.org/all/cnwpkeovzbumhprco7q2c2y6zxzmxfpwpwe3tyy6c3gg2szgqd@vfzjaw5v5imr/ Cc: Kent Overstreet Cc: Christian Brauner Cc: Jan Kara Cc: Dave Chinner Cc: Darrick J. Wong Cc: Theodore Ts'o Cc: Josef Bacik Signed-off-by: Günther Noack Link: https://lore.kernel.org/r/20240405214040.101396-2-gnoack@google.com Acked-by: Kent Overstreet Signed-off-by: Christian Brauner --- fs/ioctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ioctl.c b/fs/ioctl.c index 1d5abfdf0f22..fb0628e680c4 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -769,7 +769,7 @@ static int ioctl_getfsuuid(struct file *file, void __user *argp) struct fsuuid2 u = { .len = sb->s_uuid_len, }; if (!sb->s_uuid_len) - return -ENOIOCTLCMD; + return -ENOTTY; memcpy(&u.uuid[0], &sb->s_uuid, sb->s_uuid_len); @@ -781,7 +781,7 @@ static int ioctl_get_fs_sysfs_path(struct file *file, void __user *argp) struct super_block *sb = file_inode(file)->i_sb; if (!strlen(sb->s_sysfs_name)) - return -ENOIOCTLCMD; + return -ENOTTY; struct fs_sysfs_path u = {}; -- cgit v1.2.3-59-g8ed1b From 8bfb40be31ddea0cb4664b352e1797cfe6c91976 Mon Sep 17 00:00:00 2001 From: linke li Date: Wed, 3 Apr 2024 10:10:08 +0800 Subject: fs/dcache: Re-use value stored to dentry->d_flags instead of re-reading Currently, the __d_clear_type_and_inode() writes the value flags to dentry->d_flags, then immediately re-reads it in order to use it in a if statement. This re-read is useless because no other update to dentry->d_flags can occur at this point. This commit therefore re-use flags in the if statement instead of re-reading dentry->d_flags. Signed-off-by: linke li Link: https://lore.kernel.org/r/tencent_5E187BD0A61BA28605E85405F15228254D0A@qq.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/dcache.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/dcache.c b/fs/dcache.c index 71a8e943a0fa..407095188f83 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -355,7 +355,7 @@ static inline void __d_clear_type_and_inode(struct dentry *dentry) flags &= ~DCACHE_ENTRY_TYPE; WRITE_ONCE(dentry->d_flags, flags); dentry->d_inode = NULL; - if (dentry->d_flags & DCACHE_LRU_LIST) + if (flags & DCACHE_LRU_LIST) this_cpu_inc(nr_dentry_negative); } -- cgit v1.2.3-59-g8ed1b From fbbdc255fbee59b4207a5398fdb4f04590681a79 Mon Sep 17 00:00:00 2001 From: Zhenhua Huang Date: Mon, 8 Apr 2024 21:43:57 -0700 Subject: fs/proc: remove redundant comments from /proc/bootconfig commit 717c7c894d4b ("fs/proc: Add boot loader arguments as comment to /proc/bootconfig") adds bootloader argument comments into /proc/bootconfig. /proc/bootconfig shows boot_command_line[] multiple times following every xbc key value pair, that's duplicated and not necessary. Remove redundant ones. Output before and after the fix is like: key1 = value1 *bootloader argument comments* key2 = value2 *bootloader argument comments* key3 = value3 *bootloader argument comments* ... key1 = value1 key2 = value2 key3 = value3 *bootloader argument comments* ... Link: https://lore.kernel.org/all/20240409044358.1156477-1-paulmck@kernel.org/ Fixes: 717c7c894d4b ("fs/proc: Add boot loader arguments as comment to /proc/bootconfig") Signed-off-by: Zhenhua Huang Signed-off-by: Paul E. McKenney Cc: Cc: Cc: stable@vger.kernel.org Acked-by: Masami Hiramatsu (Google) Signed-off-by: Masami Hiramatsu (Google) --- fs/proc/bootconfig.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/proc/bootconfig.c b/fs/proc/bootconfig.c index 902b326e1e56..e5635a6b127b 100644 --- a/fs/proc/bootconfig.c +++ b/fs/proc/bootconfig.c @@ -62,12 +62,12 @@ static int __init copy_xbc_key_value_list(char *dst, size_t size) break; dst += ret; } - if (ret >= 0 && boot_command_line[0]) { - ret = snprintf(dst, rest(dst, end), "# Parameters from bootloader:\n# %s\n", - boot_command_line); - if (ret > 0) - dst += ret; - } + } + if (ret >= 0 && boot_command_line[0]) { + ret = snprintf(dst, rest(dst, end), "# Parameters from bootloader:\n# %s\n", + boot_command_line); + if (ret > 0) + dst += ret; } out: kfree(key); -- cgit v1.2.3-59-g8ed1b From c722cea208789d9e2660992bcd05fb9fac3adb56 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 8 Apr 2024 21:43:58 -0700 Subject: fs/proc: Skip bootloader comment if no embedded kernel parameters If the "bootconfig" kernel command-line argument was specified or if the kernel was built with CONFIG_BOOT_CONFIG_FORCE, but if there are no embedded kernel parameter, omit the "# Parameters from bootloader:" comment from the /proc/bootconfig file. This will cause automation to fall back to the /proc/cmdline file, which will be identical to the comment in this no-embedded-kernel-parameters case. Link: https://lore.kernel.org/all/20240409044358.1156477-2-paulmck@kernel.org/ Fixes: 8b8ce6c75430 ("fs/proc: remove redundant comments from /proc/bootconfig") Signed-off-by: Masami Hiramatsu Signed-off-by: Paul E. McKenney Cc: stable@vger.kernel.org Acked-by: Masami Hiramatsu (Google) Signed-off-by: Masami Hiramatsu (Google) --- fs/proc/bootconfig.c | 2 +- include/linux/bootconfig.h | 1 + init/main.c | 5 +++++ 3 files changed, 7 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/proc/bootconfig.c b/fs/proc/bootconfig.c index e5635a6b127b..87dcaae32ff8 100644 --- a/fs/proc/bootconfig.c +++ b/fs/proc/bootconfig.c @@ -63,7 +63,7 @@ static int __init copy_xbc_key_value_list(char *dst, size_t size) dst += ret; } } - if (ret >= 0 && boot_command_line[0]) { + if (cmdline_has_extra_options() && ret >= 0 && boot_command_line[0]) { ret = snprintf(dst, rest(dst, end), "# Parameters from bootloader:\n# %s\n", boot_command_line); if (ret > 0) diff --git a/include/linux/bootconfig.h b/include/linux/bootconfig.h index ca73940e26df..e5ee2c694401 100644 --- a/include/linux/bootconfig.h +++ b/include/linux/bootconfig.h @@ -10,6 +10,7 @@ #ifdef __KERNEL__ #include #include +bool __init cmdline_has_extra_options(void); #else /* !__KERNEL__ */ /* * NOTE: This is only for tools/bootconfig, because tools/bootconfig will diff --git a/init/main.c b/init/main.c index 2ca52474d0c3..881f6230ee59 100644 --- a/init/main.c +++ b/init/main.c @@ -487,6 +487,11 @@ static int __init warn_bootconfig(char *str) early_param("bootconfig", warn_bootconfig); +bool __init cmdline_has_extra_options(void) +{ + return extra_command_line || extra_init_args; +} + /* Change NUL term back to "=", to make "param" the whole string. */ static void __init repair_env_string(char *param, char *val) { -- cgit v1.2.3-59-g8ed1b From 98808644b920ed7bb33fe7b33d8f09d4e392e6c2 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:17:57 -0400 Subject: dlm: remove allocation parameter in msg allocation Remove the context parameter for message allocations and always use GFP_ATOMIC. This prepares for softirq message processing. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lock.c | 31 ++++++++++++------------------- fs/dlm/lowcomms.c | 16 +++++++--------- fs/dlm/lowcomms.h | 5 ++--- fs/dlm/memory.c | 8 ++++---- fs/dlm/memory.h | 4 ++-- fs/dlm/midcomms.c | 24 ++++++++++-------------- fs/dlm/midcomms.h | 3 +-- fs/dlm/rcom.c | 7 +++---- 8 files changed, 41 insertions(+), 57 deletions(-) (limited to 'fs') diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index c8426f6f518c..2f94ffc3cf82 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -3330,8 +3330,7 @@ int dlm_unlock(dlm_lockspace_t *lockspace, static int _create_message(struct dlm_ls *ls, int mb_len, int to_nodeid, int mstype, struct dlm_message **ms_ret, - struct dlm_mhandle **mh_ret, - gfp_t allocation) + struct dlm_mhandle **mh_ret) { struct dlm_message *ms; struct dlm_mhandle *mh; @@ -3341,7 +3340,7 @@ static int _create_message(struct dlm_ls *ls, int mb_len, pass into midcomms_commit and a message buffer (mb) that we write our data into */ - mh = dlm_midcomms_get_mhandle(to_nodeid, mb_len, allocation, &mb); + mh = dlm_midcomms_get_mhandle(to_nodeid, mb_len, &mb); if (!mh) return -ENOBUFS; @@ -3363,8 +3362,7 @@ static int _create_message(struct dlm_ls *ls, int mb_len, static int create_message(struct dlm_rsb *r, struct dlm_lkb *lkb, int to_nodeid, int mstype, struct dlm_message **ms_ret, - struct dlm_mhandle **mh_ret, - gfp_t allocation) + struct dlm_mhandle **mh_ret) { int mb_len = sizeof(struct dlm_message); @@ -3385,7 +3383,7 @@ static int create_message(struct dlm_rsb *r, struct dlm_lkb *lkb, } return _create_message(r->res_ls, mb_len, to_nodeid, mstype, - ms_ret, mh_ret, allocation); + ms_ret, mh_ret); } /* further lowcomms enhancements or alternate implementations may make @@ -3454,7 +3452,7 @@ static int send_common(struct dlm_rsb *r, struct dlm_lkb *lkb, int mstype) if (error) return error; - error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh, GFP_NOFS); + error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh); if (error) goto fail; @@ -3514,8 +3512,7 @@ static int send_grant(struct dlm_rsb *r, struct dlm_lkb *lkb) to_nodeid = lkb->lkb_nodeid; - error = create_message(r, lkb, to_nodeid, DLM_MSG_GRANT, &ms, &mh, - GFP_NOFS); + error = create_message(r, lkb, to_nodeid, DLM_MSG_GRANT, &ms, &mh); if (error) goto out; @@ -3536,8 +3533,7 @@ static int send_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int mode) to_nodeid = lkb->lkb_nodeid; - error = create_message(r, NULL, to_nodeid, DLM_MSG_BAST, &ms, &mh, - GFP_NOFS); + error = create_message(r, NULL, to_nodeid, DLM_MSG_BAST, &ms, &mh); if (error) goto out; @@ -3562,8 +3558,7 @@ static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb) if (error) return error; - error = create_message(r, NULL, to_nodeid, DLM_MSG_LOOKUP, &ms, &mh, - GFP_NOFS); + error = create_message(r, NULL, to_nodeid, DLM_MSG_LOOKUP, &ms, &mh); if (error) goto fail; @@ -3587,8 +3582,7 @@ static int send_remove(struct dlm_rsb *r) to_nodeid = dlm_dir_nodeid(r); - error = create_message(r, NULL, to_nodeid, DLM_MSG_REMOVE, &ms, &mh, - GFP_ATOMIC); + error = create_message(r, NULL, to_nodeid, DLM_MSG_REMOVE, &ms, &mh); if (error) goto out; @@ -3609,7 +3603,7 @@ static int send_common_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, to_nodeid = lkb->lkb_nodeid; - error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh, GFP_NOFS); + error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh); if (error) goto out; @@ -3651,8 +3645,7 @@ static int send_lookup_reply(struct dlm_ls *ls, struct dlm_mhandle *mh; int error, nodeid = le32_to_cpu(ms_in->m_header.h_nodeid); - error = create_message(r, NULL, nodeid, DLM_MSG_LOOKUP_REPLY, &ms, &mh, - GFP_NOFS); + error = create_message(r, NULL, nodeid, DLM_MSG_LOOKUP_REPLY, &ms, &mh); if (error) goto out; @@ -6103,7 +6096,7 @@ static int send_purge(struct dlm_ls *ls, int nodeid, int pid) int error; error = _create_message(ls, sizeof(struct dlm_message), nodeid, - DLM_MSG_PURGE, &ms, &mh, GFP_NOFS); + DLM_MSG_PURGE, &ms, &mh); if (error) return error; ms->m_nodeid = cpu_to_le32(nodeid); diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 712165a1e567..ab2cfbd2ea77 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -1229,14 +1229,13 @@ out: }; static struct dlm_msg *dlm_lowcomms_new_msg_con(struct connection *con, int len, - gfp_t allocation, char **ppc, - void (*cb)(void *data), + char **ppc, void (*cb)(void *data), void *data) { struct writequeue_entry *e; struct dlm_msg *msg; - msg = dlm_allocate_msg(allocation); + msg = dlm_allocate_msg(); if (!msg) return NULL; @@ -1261,9 +1260,8 @@ static struct dlm_msg *dlm_lowcomms_new_msg_con(struct connection *con, int len, * dlm_lowcomms_commit_msg which is a must call if success */ #ifndef __CHECKER__ -struct dlm_msg *dlm_lowcomms_new_msg(int nodeid, int len, gfp_t allocation, - char **ppc, void (*cb)(void *data), - void *data) +struct dlm_msg *dlm_lowcomms_new_msg(int nodeid, int len, char **ppc, + void (*cb)(void *data), void *data) { struct connection *con; struct dlm_msg *msg; @@ -1284,7 +1282,7 @@ struct dlm_msg *dlm_lowcomms_new_msg(int nodeid, int len, gfp_t allocation, return NULL; } - msg = dlm_lowcomms_new_msg_con(con, len, allocation, ppc, cb, data); + msg = dlm_lowcomms_new_msg_con(con, len, ppc, cb, data); if (!msg) { srcu_read_unlock(&connections_srcu, idx); return NULL; @@ -1348,8 +1346,8 @@ int dlm_lowcomms_resend_msg(struct dlm_msg *msg) if (msg->retransmit) return 1; - msg_resend = dlm_lowcomms_new_msg_con(msg->entry->con, msg->len, - GFP_ATOMIC, &ppc, NULL, NULL); + msg_resend = dlm_lowcomms_new_msg_con(msg->entry->con, msg->len, &ppc, + NULL, NULL); if (!msg_resend) return -ENOMEM; diff --git a/fs/dlm/lowcomms.h b/fs/dlm/lowcomms.h index 3e8dca66183b..8deb16f8f620 100644 --- a/fs/dlm/lowcomms.h +++ b/fs/dlm/lowcomms.h @@ -39,9 +39,8 @@ void dlm_lowcomms_stop(void); void dlm_lowcomms_init(void); void dlm_lowcomms_exit(void); int dlm_lowcomms_close(int nodeid); -struct dlm_msg *dlm_lowcomms_new_msg(int nodeid, int len, gfp_t allocation, - char **ppc, void (*cb)(void *data), - void *data); +struct dlm_msg *dlm_lowcomms_new_msg(int nodeid, int len, char **ppc, + void (*cb)(void *data), void *data); void dlm_lowcomms_commit_msg(struct dlm_msg *msg); void dlm_lowcomms_put_msg(struct dlm_msg *msg); int dlm_lowcomms_resend_msg(struct dlm_msg *msg); diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index be9398ddf357..ab663ca66aca 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -130,9 +130,9 @@ void dlm_free_lkb(struct dlm_lkb *lkb) kmem_cache_free(lkb_cache, lkb); } -struct dlm_mhandle *dlm_allocate_mhandle(gfp_t allocation) +struct dlm_mhandle *dlm_allocate_mhandle(void) { - return kmem_cache_alloc(mhandle_cache, allocation); + return kmem_cache_alloc(mhandle_cache, GFP_ATOMIC); } void dlm_free_mhandle(struct dlm_mhandle *mhandle) @@ -150,9 +150,9 @@ void dlm_free_writequeue(struct writequeue_entry *writequeue) kmem_cache_free(writequeue_cache, writequeue); } -struct dlm_msg *dlm_allocate_msg(gfp_t allocation) +struct dlm_msg *dlm_allocate_msg(void) { - return kmem_cache_alloc(msg_cache, allocation); + return kmem_cache_alloc(msg_cache, GFP_ATOMIC); } void dlm_free_msg(struct dlm_msg *msg) diff --git a/fs/dlm/memory.h b/fs/dlm/memory.h index 6b29563d24f7..15198d46b42a 100644 --- a/fs/dlm/memory.h +++ b/fs/dlm/memory.h @@ -20,11 +20,11 @@ struct dlm_lkb *dlm_allocate_lkb(struct dlm_ls *ls); void dlm_free_lkb(struct dlm_lkb *l); char *dlm_allocate_lvb(struct dlm_ls *ls); void dlm_free_lvb(char *l); -struct dlm_mhandle *dlm_allocate_mhandle(gfp_t allocation); +struct dlm_mhandle *dlm_allocate_mhandle(void); void dlm_free_mhandle(struct dlm_mhandle *mhandle); struct writequeue_entry *dlm_allocate_writequeue(void); void dlm_free_writequeue(struct writequeue_entry *writequeue); -struct dlm_msg *dlm_allocate_msg(gfp_t allocation); +struct dlm_msg *dlm_allocate_msg(void); void dlm_free_msg(struct dlm_msg *msg); struct dlm_callback *dlm_allocate_cb(void); void dlm_free_cb(struct dlm_callback *cb); diff --git a/fs/dlm/midcomms.c b/fs/dlm/midcomms.c index 8e9920f1b48b..ed6fb9b9a582 100644 --- a/fs/dlm/midcomms.c +++ b/fs/dlm/midcomms.c @@ -379,8 +379,7 @@ static int dlm_send_ack(int nodeid, uint32_t seq) struct dlm_msg *msg; char *ppc; - msg = dlm_lowcomms_new_msg(nodeid, mb_len, GFP_ATOMIC, &ppc, - NULL, NULL); + msg = dlm_lowcomms_new_msg(nodeid, mb_len, &ppc, NULL, NULL); if (!msg) return -ENOMEM; @@ -428,7 +427,7 @@ static int dlm_send_fin(struct midcomms_node *node, struct dlm_mhandle *mh; char *ppc; - mh = dlm_midcomms_get_mhandle(node->nodeid, mb_len, GFP_ATOMIC, &ppc); + mh = dlm_midcomms_get_mhandle(node->nodeid, mb_len, &ppc); if (!mh) return -ENOMEM; @@ -976,13 +975,13 @@ static void midcomms_new_msg_cb(void *data) } static struct dlm_msg *dlm_midcomms_get_msg_3_2(struct dlm_mhandle *mh, int nodeid, - int len, gfp_t allocation, char **ppc) + int len, char **ppc) { struct dlm_opts *opts; struct dlm_msg *msg; msg = dlm_lowcomms_new_msg(nodeid, len + DLM_MIDCOMMS_OPT_LEN, - allocation, ppc, midcomms_new_msg_cb, mh); + ppc, midcomms_new_msg_cb, mh); if (!msg) return NULL; @@ -1001,8 +1000,7 @@ static struct dlm_msg *dlm_midcomms_get_msg_3_2(struct dlm_mhandle *mh, int node * dlm_midcomms_commit_mhandle which is a must call if success */ #ifndef __CHECKER__ -struct dlm_mhandle *dlm_midcomms_get_mhandle(int nodeid, int len, - gfp_t allocation, char **ppc) +struct dlm_mhandle *dlm_midcomms_get_mhandle(int nodeid, int len, char **ppc) { struct midcomms_node *node; struct dlm_mhandle *mh; @@ -1017,7 +1015,7 @@ struct dlm_mhandle *dlm_midcomms_get_mhandle(int nodeid, int len, /* this is a bug, however we going on and hope it will be resolved */ WARN_ON_ONCE(test_bit(DLM_NODE_FLAG_STOP_TX, &node->flags)); - mh = dlm_allocate_mhandle(allocation); + mh = dlm_allocate_mhandle(); if (!mh) goto err; @@ -1028,8 +1026,7 @@ struct dlm_mhandle *dlm_midcomms_get_mhandle(int nodeid, int len, switch (node->version) { case DLM_VERSION_3_1: - msg = dlm_lowcomms_new_msg(nodeid, len, allocation, ppc, - NULL, NULL); + msg = dlm_lowcomms_new_msg(nodeid, len, ppc, NULL, NULL); if (!msg) { dlm_free_mhandle(mh); goto err; @@ -1040,8 +1037,7 @@ struct dlm_mhandle *dlm_midcomms_get_mhandle(int nodeid, int len, /* send ack back if necessary */ dlm_send_ack_threshold(node, DLM_SEND_ACK_BACK_MSG_THRESHOLD); - msg = dlm_midcomms_get_msg_3_2(mh, nodeid, len, allocation, - ppc); + msg = dlm_midcomms_get_msg_3_2(mh, nodeid, len, ppc); if (!msg) { dlm_free_mhandle(mh); goto err; @@ -1501,8 +1497,8 @@ int dlm_midcomms_rawmsg_send(struct midcomms_node *node, void *buf, rd.node = node; rd.buf = buf; - msg = dlm_lowcomms_new_msg(node->nodeid, buflen, GFP_NOFS, - &msgbuf, midcomms_new_rawmsg_cb, &rd); + msg = dlm_lowcomms_new_msg(node->nodeid, buflen, &msgbuf, + midcomms_new_rawmsg_cb, &rd); if (!msg) return -ENOMEM; diff --git a/fs/dlm/midcomms.h b/fs/dlm/midcomms.h index e7246fb3ef57..278d26fdeb2c 100644 --- a/fs/dlm/midcomms.h +++ b/fs/dlm/midcomms.h @@ -16,8 +16,7 @@ struct midcomms_node; int dlm_validate_incoming_buffer(int nodeid, unsigned char *buf, int len); int dlm_process_incoming_buffer(int nodeid, unsigned char *buf, int buflen); -struct dlm_mhandle *dlm_midcomms_get_mhandle(int nodeid, int len, - gfp_t allocation, char **ppc); +struct dlm_mhandle *dlm_midcomms_get_mhandle(int nodeid, int len, char **ppc); void dlm_midcomms_commit_mhandle(struct dlm_mhandle *mh, const void *name, int namelen); int dlm_midcomms_addr(int nodeid, struct sockaddr_storage *addr, int len); diff --git a/fs/dlm/rcom.c b/fs/dlm/rcom.c index 3b734aed26b5..2e3f529f3ff2 100644 --- a/fs/dlm/rcom.c +++ b/fs/dlm/rcom.c @@ -55,7 +55,7 @@ static int create_rcom(struct dlm_ls *ls, int to_nodeid, int type, int len, struct dlm_mhandle *mh; char *mb; - mh = dlm_midcomms_get_mhandle(to_nodeid, mb_len, GFP_NOFS, &mb); + mh = dlm_midcomms_get_mhandle(to_nodeid, mb_len, &mb); if (!mh) { log_print("%s to %d type %d len %d ENOBUFS", __func__, to_nodeid, type, len); @@ -75,8 +75,7 @@ static int create_rcom_stateless(struct dlm_ls *ls, int to_nodeid, int type, struct dlm_msg *msg; char *mb; - msg = dlm_lowcomms_new_msg(to_nodeid, mb_len, GFP_NOFS, &mb, - NULL, NULL); + msg = dlm_lowcomms_new_msg(to_nodeid, mb_len, &mb, NULL, NULL); if (!msg) { log_print("create_rcom to %d type %d len %d ENOBUFS", to_nodeid, type, len); @@ -510,7 +509,7 @@ int dlm_send_ls_not_ready(int nodeid, const struct dlm_rcom *rc_in) char *mb; int mb_len = sizeof(struct dlm_rcom) + sizeof(struct rcom_config); - mh = dlm_midcomms_get_mhandle(nodeid, mb_len, GFP_NOFS, &mb); + mh = dlm_midcomms_get_mhandle(nodeid, mb_len, &mb); if (!mh) return -ENOBUFS; -- cgit v1.2.3-59-g8ed1b From 0f9b12142be1af8555cfe53c6fbecb8e60a40dac Mon Sep 17 00:00:00 2001 From: Wenjie Qi Date: Sun, 7 Apr 2024 15:21:23 +0800 Subject: f2fs: fix zoned block device information initialization If the max open zones of zoned devices are less than the active logs of F2FS, the device may error due to insufficient zone resources when multiple active logs are being written at the same time. Signed-off-by: Wenjie Qi Signed-off-by: Chao Yu Reviewed-by: Daeho Jeong Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 1 + fs/f2fs/super.c | 27 +++++++++++++++++++++++++++ 2 files changed, 28 insertions(+) (limited to 'fs') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index adfce581026b..e9ef971f4dba 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1557,6 +1557,7 @@ struct f2fs_sb_info { #ifdef CONFIG_BLK_DEV_ZONED unsigned int blocks_per_blkz; /* F2FS blocks per zone */ + unsigned int max_open_zones; /* max open zone resources of the zoned device */ #endif /* for node-related operations */ diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index df9765b41dac..8ac4734c2df6 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -2324,6 +2324,17 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) if (err) goto restore_opts; +#ifdef CONFIG_BLK_DEV_ZONED + if (f2fs_sb_has_blkzoned(sbi) && + sbi->max_open_zones < F2FS_OPTION(sbi).active_logs) { + f2fs_err(sbi, + "zoned: max open zones %u is too small, need at least %u open zones", + sbi->max_open_zones, F2FS_OPTION(sbi).active_logs); + err = -EINVAL; + goto restore_opts; + } +#endif + /* flush outstanding errors before changing fs state */ flush_work(&sbi->s_error_work); @@ -3866,11 +3877,24 @@ static int init_blkz_info(struct f2fs_sb_info *sbi, int devi) sector_t nr_sectors = bdev_nr_sectors(bdev); struct f2fs_report_zones_args rep_zone_arg; u64 zone_sectors; + unsigned int max_open_zones; int ret; if (!f2fs_sb_has_blkzoned(sbi)) return 0; + if (bdev_is_zoned(FDEV(devi).bdev)) { + max_open_zones = bdev_max_open_zones(bdev); + if (max_open_zones && (max_open_zones < sbi->max_open_zones)) + sbi->max_open_zones = max_open_zones; + if (sbi->max_open_zones < F2FS_OPTION(sbi).active_logs) { + f2fs_err(sbi, + "zoned: max open zones %u is too small, need at least %u open zones", + sbi->max_open_zones, F2FS_OPTION(sbi).active_logs); + return -EINVAL; + } + } + zone_sectors = bdev_zone_sectors(bdev); if (sbi->blocks_per_blkz && sbi->blocks_per_blkz != SECTOR_TO_BLOCK(zone_sectors)) @@ -4184,6 +4208,9 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi) logical_blksize = bdev_logical_block_size(sbi->sb->s_bdev); sbi->aligned_blksize = true; +#ifdef CONFIG_BLK_DEV_ZONED + sbi->max_open_zones = UINT_MAX; +#endif for (i = 0; i < max_devices; i++) { if (i == 0) -- cgit v1.2.3-59-g8ed1b From 10398ef57aa189153406c110f5957145030f08fe Mon Sep 17 00:00:00 2001 From: Andrew Price Date: Thu, 11 Jan 2024 17:42:58 +0000 Subject: gfs2: Improve gfs2_consist_inode() usage gfs2_consist_inode() logs an error message with the source file and line number. When we jump before calling it, the line number becomes less useful as it no longer relates to the source of the error. To aid troubleshooting, replace the gotos with the gfs2_consist_inode() calls so that the error messages are more informative. Signed-off-by: Andrew Price Signed-off-by: Andreas Gruenbacher --- fs/gfs2/dir.c | 31 +++++++++++++++++-------------- fs/gfs2/glops.c | 34 ++++++++++++++++++++-------------- fs/gfs2/xattr.c | 28 ++++++++++++++++------------ 3 files changed, 53 insertions(+), 40 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 560e4624c09f..dbf1aede744c 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -562,15 +562,18 @@ static struct gfs2_dirent *gfs2_dirent_scan(struct inode *inode, void *buf, int ret = 0; ret = gfs2_dirent_offset(GFS2_SB(inode), buf); - if (ret < 0) - goto consist_inode; - + if (ret < 0) { + gfs2_consist_inode(GFS2_I(inode)); + return ERR_PTR(-EIO); + } offset = ret; prev = NULL; dent = buf + offset; size = be16_to_cpu(dent->de_rec_len); - if (gfs2_check_dirent(GFS2_SB(inode), dent, offset, size, len, 1)) - goto consist_inode; + if (gfs2_check_dirent(GFS2_SB(inode), dent, offset, size, len, 1)) { + gfs2_consist_inode(GFS2_I(inode)); + return ERR_PTR(-EIO); + } do { ret = scan(dent, name, opaque); if (ret) @@ -582,8 +585,10 @@ static struct gfs2_dirent *gfs2_dirent_scan(struct inode *inode, void *buf, dent = buf + offset; size = be16_to_cpu(dent->de_rec_len); if (gfs2_check_dirent(GFS2_SB(inode), dent, offset, size, - len, 0)) - goto consist_inode; + len, 0)) { + gfs2_consist_inode(GFS2_I(inode)); + return ERR_PTR(-EIO); + } } while(1); switch(ret) { @@ -597,10 +602,6 @@ static struct gfs2_dirent *gfs2_dirent_scan(struct inode *inode, void *buf, BUG_ON(ret > 0); return ERR_PTR(ret); } - -consist_inode: - gfs2_consist_inode(GFS2_I(inode)); - return ERR_PTR(-EIO); } static int dirent_check_reclen(struct gfs2_inode *dip, @@ -609,14 +610,16 @@ static int dirent_check_reclen(struct gfs2_inode *dip, const void *ptr = d; u16 rec_len = be16_to_cpu(d->de_rec_len); - if (unlikely(rec_len < sizeof(struct gfs2_dirent))) - goto broken; + if (unlikely(rec_len < sizeof(struct gfs2_dirent))) { + gfs2_consist_inode(dip); + return -EIO; + } ptr += rec_len; if (ptr < end_p) return rec_len; if (ptr == end_p) return -ENOENT; -broken: + gfs2_consist_inode(dip); return -EIO; } diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 45653cbc8a87..ae8d4731907d 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -409,10 +409,14 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) struct inode *inode = &ip->i_inode; bool is_new = inode->i_state & I_NEW; - if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr))) - goto corrupt; - if (unlikely(!is_new && inode_wrong_type(inode, mode))) - goto corrupt; + if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr))) { + gfs2_consist_inode(ip); + return -EIO; + } + if (unlikely(!is_new && inode_wrong_type(inode, mode))) { + gfs2_consist_inode(ip); + return -EIO; + } ip->i_no_formal_ino = be64_to_cpu(str->di_num.no_formal_ino); inode->i_mode = mode; if (is_new) { @@ -449,26 +453,28 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) /* i_diskflags and i_eattr must be set before gfs2_set_inode_flags() */ gfs2_set_inode_flags(inode); height = be16_to_cpu(str->di_height); - if (unlikely(height > sdp->sd_max_height)) - goto corrupt; + if (unlikely(height > sdp->sd_max_height)) { + gfs2_consist_inode(ip); + return -EIO; + } ip->i_height = (u8)height; depth = be16_to_cpu(str->di_depth); - if (unlikely(depth > GFS2_DIR_MAX_DEPTH)) - goto corrupt; + if (unlikely(depth > GFS2_DIR_MAX_DEPTH)) { + gfs2_consist_inode(ip); + return -EIO; + } ip->i_depth = (u8)depth; ip->i_entries = be32_to_cpu(str->di_entries); - if (gfs2_is_stuffed(ip) && inode->i_size > gfs2_max_stuffed_size(ip)) - goto corrupt; - + if (gfs2_is_stuffed(ip) && inode->i_size > gfs2_max_stuffed_size(ip)) { + gfs2_consist_inode(ip); + return -EIO; + } if (S_ISREG(inode->i_mode)) gfs2_set_aops(inode); return 0; -corrupt: - gfs2_consist_inode(ip); - return -EIO; } /** diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c index 8c96ba6230d1..17ae5070a90e 100644 --- a/fs/gfs2/xattr.c +++ b/fs/gfs2/xattr.c @@ -96,30 +96,34 @@ static int ea_foreach_i(struct gfs2_inode *ip, struct buffer_head *bh, return -EIO; for (ea = GFS2_EA_BH2FIRST(bh);; prev = ea, ea = GFS2_EA2NEXT(ea)) { - if (!GFS2_EA_REC_LEN(ea)) - goto fail; + if (!GFS2_EA_REC_LEN(ea)) { + gfs2_consist_inode(ip); + return -EIO; + } if (!(bh->b_data <= (char *)ea && (char *)GFS2_EA2NEXT(ea) <= - bh->b_data + bh->b_size)) - goto fail; - if (!gfs2_eatype_valid(sdp, ea->ea_type)) - goto fail; + bh->b_data + bh->b_size)) { + gfs2_consist_inode(ip); + return -EIO; + } + if (!gfs2_eatype_valid(sdp, ea->ea_type)) { + gfs2_consist_inode(ip); + return -EIO; + } error = ea_call(ip, bh, ea, prev, data); if (error) return error; if (GFS2_EA_IS_LAST(ea)) { if ((char *)GFS2_EA2NEXT(ea) != - bh->b_data + bh->b_size) - goto fail; + bh->b_data + bh->b_size) { + gfs2_consist_inode(ip); + return -EIO; + } break; } } return error; - -fail: - gfs2_consist_inode(ip); - return -EIO; } static int ea_foreach(struct gfs2_inode *ip, ea_call_t ea_call, void *data) -- cgit v1.2.3-59-g8ed1b From b204b1b61eff8d5ac1aefcb00547a5f4002be47d Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Tue, 27 Feb 2024 20:42:42 +0100 Subject: gfs2: Get rid of newlines in log messages Get rid of attempts to create multi-line syslog entries; this only makes the messages harder to read. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/bmap.c | 2 +- fs/gfs2/rgrp.c | 10 +++++----- fs/gfs2/util.c | 50 ++++++++++++++++++++++++++------------------------ 3 files changed, 32 insertions(+), 30 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 643175498d1c..b8d64f2259d6 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -1827,7 +1827,7 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length) gfs2_assert_withdraw(sdp, bh); if (gfs2_assert_withdraw(sdp, prev_bnr != bh->b_blocknr)) { - fs_emerg(sdp, "inode %llu, block:%llu, i_h:%u," + fs_emerg(sdp, "inode %llu, block:%llu, i_h:%u, " "s_h:%u, mp_h:%u\n", (unsigned long long)ip->i_no_addr, prev_bnr, ip->i_height, strip_h, mp_h); diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 26d6c1eea559..29c772816765 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -814,11 +814,11 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd) bi = rgd->rd_bits + (length - 1); if ((bi->bi_start + bi->bi_bytes) * GFS2_NBBY != rgd->rd_data) { gfs2_lm(sdp, - "ri_addr = %llu\n" - "ri_length = %u\n" - "ri_data0 = %llu\n" - "ri_data = %u\n" - "ri_bitbytes = %u\n" + "ri_addr=%llu " + "ri_length=%u " + "ri_data0=%llu " + "ri_data=%u " + "ri_bitbytes=%u " "start=%u len=%u offset=%u\n", (unsigned long long)rgd->rd_addr, rgd->rd_length, diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index f52141ce9485..fd65a2e34102 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -376,8 +376,8 @@ void gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion, return; fs_err(sdp, - "fatal: assertion \"%s\" failed\n" - " function = %s, file = %s, line = %u\n", + "fatal: assertion \"%s\" failed - " + "function = %s, file = %s, line = %u\n", assertion, function, file, line); /* @@ -407,7 +407,8 @@ void gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, return; if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW) - fs_warn(sdp, "warning: assertion \"%s\" failed at function = %s, file = %s, line = %u\n", + fs_warn(sdp, "warning: assertion \"%s\" failed - " + "function = %s, file = %s, line = %u\n", assertion, function, file, line); if (sdp->sd_args.ar_debug) @@ -416,10 +417,10 @@ void gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, dump_stack(); if (sdp->sd_args.ar_errors == GFS2_ERRORS_PANIC) - panic("GFS2: fsid=%s: warning: assertion \"%s\" failed\n" - "GFS2: fsid=%s: function = %s, file = %s, line = %u\n", + panic("GFS2: fsid=%s: warning: assertion \"%s\" failed - " + "function = %s, file = %s, line = %u\n", sdp->sd_fsname, assertion, - sdp->sd_fsname, function, file, line); + function, file, line); sdp->sd_last_warning = jiffies; } @@ -432,7 +433,8 @@ void gfs2_consist_i(struct gfs2_sbd *sdp, const char *function, char *file, unsigned int line) { gfs2_lm(sdp, - "fatal: filesystem consistency error - function = %s, file = %s, line = %u\n", + "fatal: filesystem consistency error - " + "function = %s, file = %s, line = %u\n", function, file, line); gfs2_withdraw(sdp); } @@ -447,9 +449,9 @@ void gfs2_consist_inode_i(struct gfs2_inode *ip, struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); gfs2_lm(sdp, - "fatal: filesystem consistency error\n" - " inode = %llu %llu\n" - " function = %s, file = %s, line = %u\n", + "fatal: filesystem consistency error - " + "inode = %llu %llu, " + "function = %s, file = %s, line = %u\n", (unsigned long long)ip->i_no_formal_ino, (unsigned long long)ip->i_no_addr, function, file, line); @@ -470,9 +472,9 @@ void gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, sprintf(fs_id_buf, "fsid=%s: ", sdp->sd_fsname); gfs2_rgrp_dump(NULL, rgd, fs_id_buf); gfs2_lm(sdp, - "fatal: filesystem consistency error\n" - " RG = %llu\n" - " function = %s, file = %s, line = %u\n", + "fatal: filesystem consistency error - " + "RG = %llu, " + "function = %s, file = %s, line = %u\n", (unsigned long long)rgd->rd_addr, function, file, line); gfs2_dump_glock(NULL, rgd->rd_gl, 1); @@ -492,9 +494,9 @@ int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, int me; gfs2_lm(sdp, - "fatal: invalid metadata block\n" - " bh = %llu (%s)\n" - " function = %s, file = %s, line = %u\n", + "fatal: invalid metadata block - " + "bh = %llu (%s), " + "function = %s, file = %s, line = %u\n", (unsigned long long)bh->b_blocknr, type, function, file, line); me = gfs2_withdraw(sdp); @@ -514,9 +516,9 @@ int gfs2_metatype_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, int me; gfs2_lm(sdp, - "fatal: invalid metadata block\n" - " bh = %llu (type: exp=%u, found=%u)\n" - " function = %s, file = %s, line = %u\n", + "fatal: invalid metadata block - " + "bh = %llu (type: exp=%u, found=%u), " + "function = %s, file = %s, line = %u\n", (unsigned long long)bh->b_blocknr, type, t, function, file, line); me = gfs2_withdraw(sdp); @@ -533,8 +535,8 @@ int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function, char *file, unsigned int line) { gfs2_lm(sdp, - "fatal: I/O error\n" - " function = %s, file = %s, line = %u\n", + "fatal: I/O error - " + "function = %s, file = %s, line = %u\n", function, file, line); return gfs2_withdraw(sdp); } @@ -551,9 +553,9 @@ void gfs2_io_error_bh_i(struct gfs2_sbd *sdp, struct buffer_head *bh, if (gfs2_withdrawing_or_withdrawn(sdp)) return; - fs_err(sdp, "fatal: I/O error\n" - " block = %llu\n" - " function = %s, file = %s, line = %u\n", + fs_err(sdp, "fatal: I/O error - " + "block = %llu, " + "function = %s, file = %s, line = %u\n", (unsigned long long)bh->b_blocknr, function, file, line); if (withdraw) gfs2_withdraw(sdp); -- cgit v1.2.3-59-g8ed1b From 795405c4b995efc4f6b6b561200c384bdda07c02 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Tue, 27 Feb 2024 22:57:08 +0100 Subject: gfs2: Remove unnecessary gfs2_meta_check_ii argument The type argument of gfs2_meta_check_ii() is always set to "magic number", so remove that argument and hardcode the string in gfs2_meta_check_ii(). Change the string to "bad magic number" to emphasize that the problem is the incorrect magic number. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/util.c | 6 +++--- fs/gfs2/util.h | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index fd65a2e34102..963d77c4ca21 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -488,16 +488,16 @@ void gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, */ int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, - const char *type, const char *function, char *file, + const char *function, char *file, unsigned int line) { int me; gfs2_lm(sdp, "fatal: invalid metadata block - " - "bh = %llu (%s), " + "bh = %llu (bad magic number), " "function = %s, file = %s, line = %u\n", - (unsigned long long)bh->b_blocknr, type, + (unsigned long long)bh->b_blocknr, function, file, line); me = gfs2_withdraw(sdp); return (me) ? -1 : -2; diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index ba071998461f..d8d62bf0424d 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -92,7 +92,7 @@ gfs2_consist_rgrpd_i((rgd), __func__, __FILE__, __LINE__) int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, - const char *type, const char *function, + const char *function, char *file, unsigned int line); static inline int gfs2_meta_check(struct gfs2_sbd *sdp, @@ -123,7 +123,7 @@ static inline int gfs2_metatype_check_i(struct gfs2_sbd *sdp, u32 magic = be32_to_cpu(mh->mh_magic); u16 t = be32_to_cpu(mh->mh_type); if (unlikely(magic != GFS2_MAGIC)) - return gfs2_meta_check_ii(sdp, bh, "magic number", function, + return gfs2_meta_check_ii(sdp, bh, function, file, line); if (unlikely(t != type)) return gfs2_metatype_check_ii(sdp, bh, type, t, function, -- cgit v1.2.3-59-g8ed1b From 52c2d389fe0a5cbfd88bc68e5b31d8811d2e5933 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 13 Mar 2024 13:24:35 +0100 Subject: gfs2: Follow-up to flag rename in sysfs status file Follow up to commit e7beb8b6de1a ("gfs2: Rename SDF_DEACTIVATING to SDF_KILL") and change the description of the SDF_KILL flag from "Deactivating" to "Killing". Signed-off-by: Andreas Gruenbacher --- fs/gfs2/sys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 250f340cb44d..32d3484270f3 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -88,7 +88,7 @@ static ssize_t status_show(struct gfs2_sbd *sdp, char *buf) "Withdraw In Prog: %d\n" "Remote Withdraw: %d\n" "Withdraw Recovery: %d\n" - "Deactivating: %d\n" + "Killing: %d\n" "sd_log_error: %d\n" "sd_log_flush_lock: %d\n" "sd_log_num_revoke: %u\n" -- cgit v1.2.3-59-g8ed1b From 3592bfaf746a4d0fefe6fe11527aa335073674c1 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Mon, 15 Jan 2024 22:01:12 +0100 Subject: gfs2: Use [NO_]CREATE consistently for gfs2_glock_get When calling gfs2_glock_get(), consistently pass the CREATE or NO_CREATE flag for the create argument. This is the only place that passes 0 instead. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/sys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 32d3484270f3..ecc699f8d9fc 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -336,7 +336,7 @@ static ssize_t demote_rq_store(struct gfs2_sbd *sdp, const char *buf, size_t len return -EINVAL; if (!test_and_set_bit(SDF_DEMOTE, &sdp->sd_flags)) fs_info(sdp, "demote interface used\n"); - rv = gfs2_glock_get(sdp, glnum, glops, 0, &gl); + rv = gfs2_glock_get(sdp, glnum, glops, NO_CREATE, &gl); if (rv) return rv; gfs2_glock_cb(gl, glmode); -- cgit v1.2.3-59-g8ed1b From b01189333ee91c1ae6cd96dfd1e3a3c2e69202f0 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 26 Jan 2024 11:49:44 +0100 Subject: gfs2: Don't forget to complete delayed withdraw Commit fffe9bee14b0 ("gfs2: Delay withdraw from atomic context") switched from gfs2_withdraw() to gfs2_withdraw_delayed() in gfs2_ail_error(), but failed to then check if a delayed withdraw had occurred. Fix that by adding the missing check in __gfs2_ail_flush(), where the spin locks are already dropped and a withdraw is possible. Fixes: fffe9bee14b0 ("gfs2: Delay withdraw from atomic context") Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glops.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index ae8d4731907d..68677fb69a73 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -82,6 +82,9 @@ static void __gfs2_ail_flush(struct gfs2_glock *gl, bool fsync, GLOCK_BUG_ON(gl, !fsync && atomic_read(&gl->gl_ail_count)); spin_unlock(&sdp->sd_ail_lock); gfs2_log_unlock(sdp); + + if (gfs2_withdrawing(sdp)) + gfs2_withdraw(sdp); } -- cgit v1.2.3-59-g8ed1b From 35264909e9d1973ab9aaa2a1b07cda70f12bb828 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Mon, 11 Mar 2024 15:51:59 +0100 Subject: gfs2: Fix NULL pointer dereference in gfs2_log_flush In gfs2_jindex_free(), set sdp->sd_jdesc to NULL under the log flush lock to provide exclusion against gfs2_log_flush(). In gfs2_log_flush(), check if sdp->sd_jdesc is non-NULL before dereferencing it. Otherwise, we could run into a NULL pointer dereference when outstanding glock work races with an unmount (glock_work_func -> run_queue -> do_xmote -> inode_go_sync -> gfs2_log_flush). Signed-off-by: Andreas Gruenbacher --- fs/gfs2/log.c | 3 ++- fs/gfs2/super.c | 4 ++++ 2 files changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 8cddf955ebc0..a6dd68b458ce 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -1108,7 +1108,8 @@ repeat: lops_before_commit(sdp, tr); if (gfs2_withdrawing_or_withdrawn(sdp)) goto out_withdraw; - gfs2_log_submit_bio(&sdp->sd_jdesc->jd_log_bio, REQ_OP_WRITE); + if (sdp->sd_jdesc) + gfs2_log_submit_bio(&sdp->sd_jdesc->jd_log_bio, REQ_OP_WRITE); if (gfs2_withdrawing_or_withdrawn(sdp)) goto out_withdraw; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index e5f79466340d..9de789b78bc5 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -67,9 +67,13 @@ void gfs2_jindex_free(struct gfs2_sbd *sdp) sdp->sd_journals = 0; spin_unlock(&sdp->sd_jindex_spin); + down_write(&sdp->sd_log_flush_lock); sdp->sd_jdesc = NULL; + up_write(&sdp->sd_log_flush_lock); + while (!list_empty(&list)) { jd = list_first_entry(&list, struct gfs2_jdesc, jd_list); + BUG_ON(jd->jd_log_bio); gfs2_free_journal_extents(jd); list_del(&jd->jd_list); iput(jd->jd_inode); -- cgit v1.2.3-59-g8ed1b From f80d882edcf242d0256d9e51b09d5fb7a3a0d3b4 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 15 Mar 2024 16:56:09 +0100 Subject: gfs2: Get rid of gfs2_glock_queue_put in signal_our_withdraw In function signal_our_withdraw(), we are calling gfs2_glock_queue_put() in a context in which we are actually allowed to sleep, so replace that with a simple call to gfs2_glock_put(). Signed-off-by: Andreas Gruenbacher --- fs/gfs2/util.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index 963d77c4ca21..0535a0206512 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -255,7 +255,7 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp) gfs2_glock_nq(&sdp->sd_live_gh); } - gfs2_glock_queue_put(live_gl); /* drop extra reference we acquired */ + gfs2_glock_put(live_gl); /* drop extra reference we acquired */ clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags); /* -- cgit v1.2.3-59-g8ed1b From ee2be7d7c7f32783f60ee5fe59b91548a4571f10 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 15 Mar 2024 16:45:39 +0100 Subject: gfs2: Replace gfs2_glock_queue_put with gfs2_glock_put_async Function gfs2_glock_queue_put() puts a glock reference by enqueuing glock work instead of putting the reference directly. This ensures that the operation won't sleep, but it is costly and really only necessary when putting the final glock reference. Replace it with a new gfs2_glock_put_async() function that only queues glock work when putting the last glock reference. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 27 +++++++++++++++++---------- fs/gfs2/glock.h | 2 +- fs/gfs2/log.c | 2 +- fs/gfs2/super.c | 4 ++-- 4 files changed, 21 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 34540f9d011c..ed90033b9c72 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -285,14 +285,6 @@ static void __gfs2_glock_put(struct gfs2_glock *gl) sdp->sd_lockstruct.ls_ops->lm_put_lock(gl); } -/* - * Cause the glock to be put in work queue context. - */ -void gfs2_glock_queue_put(struct gfs2_glock *gl) -{ - gfs2_glock_queue_work(gl, 0); -} - /** * gfs2_glock_put() - Decrement reference count on glock * @gl: The glock to put @@ -307,6 +299,22 @@ void gfs2_glock_put(struct gfs2_glock *gl) __gfs2_glock_put(gl); } +/* + * gfs2_glock_put_async - Decrement reference count without sleeping + * @gl: The glock to put + * + * Decrement the reference count on glock immediately unless it is the last + * reference. Defer putting the last reference to work queue context. + */ +void gfs2_glock_put_async(struct gfs2_glock *gl) +{ + if (lockref_put_or_lock(&gl->gl_lockref)) + return; + + __gfs2_glock_queue_work(gl, 0); + spin_unlock(&gl->gl_lockref.lock); +} + /** * may_grant - check if it's ok to grant a new lock * @gl: The glock @@ -2529,8 +2537,7 @@ static void gfs2_glock_iter_next(struct gfs2_glock_iter *gi, loff_t n) if (gl) { if (n == 0) return; - if (!lockref_put_not_zero(&gl->gl_lockref)) - gfs2_glock_queue_put(gl); + gfs2_glock_put_async(gl); } for (;;) { gl = rhashtable_walk_next(&gi->hti); diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index 0114f3e0ebe0..2c697674a86f 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -172,7 +172,7 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, int create, struct gfs2_glock **glp); struct gfs2_glock *gfs2_glock_hold(struct gfs2_glock *gl); void gfs2_glock_put(struct gfs2_glock *gl); -void gfs2_glock_queue_put(struct gfs2_glock *gl); +void gfs2_glock_put_async(struct gfs2_glock *gl); void __gfs2_holder_init(struct gfs2_glock *gl, unsigned int state, u16 flags, struct gfs2_holder *gh, diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index a6dd68b458ce..6ee6013fb825 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -786,7 +786,7 @@ void gfs2_glock_remove_revoke(struct gfs2_glock *gl) { if (atomic_dec_return(&gl->gl_revokes) == 0) { clear_bit(GLF_LFLUSH, &gl->gl_flags); - gfs2_glock_queue_put(gl); + gfs2_glock_put_async(gl); } } diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 9de789b78bc5..d481db9510ac 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -1049,7 +1049,7 @@ static int gfs2_drop_inode(struct inode *inode) gfs2_glock_hold(gl); if (!gfs2_queue_try_to_evict(gl)) - gfs2_glock_queue_put(gl); + gfs2_glock_put_async(gl); return 0; } @@ -1255,7 +1255,7 @@ out_qs: static void gfs2_glock_put_eventually(struct gfs2_glock *gl) { if (current->flags & PF_MEMALLOC) - gfs2_glock_queue_put(gl); + gfs2_glock_put_async(gl); else gfs2_glock_put(gl); } -- cgit v1.2.3-59-g8ed1b From 927cfc90d27cb7732a62464f95fd9aa7edfa9b70 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Tue, 9 Apr 2024 07:11:48 +0200 Subject: gfs2: Don't set GLF_LOCK in gfs2_dispose_glock_lru In gfs2_dispose_glock_lru(), we want to skip glocks which are in the process of transitioning state (as indicated by the set GLF_LOCK flag), but we we don't need to set that flag for requesting a state transition. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index ed90033b9c72..a700e424f790 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -2017,14 +2017,13 @@ add_back_to_lru: atomic_inc(&lru_count); continue; } - if (test_and_set_bit(GLF_LOCK, &gl->gl_flags)) { + if (test_bit(GLF_LOCK, &gl->gl_flags)) { spin_unlock(&gl->gl_lockref.lock); goto add_back_to_lru; } gl->gl_lockref.count++; if (demote_ok(gl)) handle_callback(gl, LM_ST_UNLOCKED, 0, false); - WARN_ON(!test_and_clear_bit(GLF_LOCK, &gl->gl_flags)); __gfs2_glock_queue_work(gl, 0); spin_unlock(&gl->gl_lockref.lock); cond_resched_lock(&lru_lock); -- cgit v1.2.3-59-g8ed1b From 262ee3a07e98e8e56230547d4fe75d485348a55a Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 29 Mar 2024 15:46:32 +0100 Subject: gfs2: Get rid of unnecessary test_and_set_bit The GLF_LOCK flag is protected by the gl->gl_lockref.lock spin lock which is held when entering run_queue(), so we can use test_bit() and set_bit() here. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index a700e424f790..fa047331ea3c 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -842,8 +842,9 @@ __acquires(&gl->gl_lockref.lock) { struct gfs2_holder *gh = NULL; - if (test_and_set_bit(GLF_LOCK, &gl->gl_flags)) + if (test_bit(GLF_LOCK, &gl->gl_flags)) return; + set_bit(GLF_LOCK, &gl->gl_flags); GLOCK_BUG_ON(gl, test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags)); -- cgit v1.2.3-59-g8ed1b From 5d9231111966b6c5a65016d58dcbeab91055bc91 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 5 Apr 2024 13:47:51 +0200 Subject: gfs2: Fix "ignore unlock failures after withdraw" Commit 3e11e53041502 tries to suppress dlm_lock() lock conversion errors that occur when the lockspace has already been released. It does that by setting and checking the SDF_SKIP_DLM_UNLOCK flag. This conflicts with the intended meaning of the SDF_SKIP_DLM_UNLOCK flag, so check whether the lockspace is still allocated instead. (Given the current DLM API, checking for this kind of error after the fact seems easier that than to make sure that the lockspace is still allocated before calling dlm_lock(). Changing the DLM API so that users maintain the lockspace references themselves would be an option.) Fixes: 3e11e53041502 ("GFS2: ignore unlock failures after withdraw") Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 4 +++- fs/gfs2/util.c | 1 - 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index fa047331ea3c..ec1f8eb28950 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -810,11 +810,13 @@ skip_inval: } if (sdp->sd_lockstruct.ls_ops->lm_lock) { + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + /* lock_dlm */ ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags); if (ret == -EINVAL && gl->gl_target == LM_ST_UNLOCKED && target == LM_ST_UNLOCKED && - test_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags)) { + test_bit(DFL_UNMOUNT, &ls->ls_recover_flags)) { finish_xmote(gl, target); gfs2_glock_queue_work(gl, 0); } else if (ret) { diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index 0535a0206512..09238604d741 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -350,7 +350,6 @@ int gfs2_withdraw(struct gfs2_sbd *sdp) fs_err(sdp, "telling LM to unmount\n"); lm->lm_unmount(sdp); } - set_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags); fs_err(sdp, "File system withdrawn\n"); dump_stack(); clear_bit(SDF_WITHDRAW_IN_PROG, &sdp->sd_flags); -- cgit v1.2.3-59-g8ed1b From c9a0a4b028e4fe16c79e685b2135ce2f097d0d0e Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Thu, 28 Mar 2024 19:14:07 +0100 Subject: Revert "gfs2: fix glock shrinker ref issues" This reverts commit 62862485a4c3a52029fc30f4bdde9af04afdafc9. Commit 62862485a4c3 tried to fix issues introduced by commit 228804a35caa ("gfs2: Make glock lru list scanning safer"), but like that commit, it failed to account for the bias state_change() adds to the glock reference count for locked glocks. Revert commit 62862485a4c3 so that we can fix commit 228804a35caa properly. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index ec1f8eb28950..2882a42e88aa 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -2056,9 +2056,7 @@ static long gfs2_scan_glock_lru(int nr) if (!test_bit(GLF_LOCK, &gl->gl_flags)) { if (!spin_trylock(&gl->gl_lockref.lock)) continue; - if (gl->gl_lockref.count <= 1 && - (gl->gl_state == LM_ST_UNLOCKED || - demote_ok(gl))) { + if (!gl->gl_lockref.count) { list_move(&gl->gl_lru, &dispose); atomic_dec(&lru_count); freed++; -- cgit v1.2.3-59-g8ed1b From acf1f42faf5a27913eb7b8140cc255f1b3d14e48 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Thu, 28 Mar 2024 19:59:45 +0100 Subject: gfs2: Fix "Make glock lru list scanning safer" Commit 228804a35caa tried to add a refcount check to gfs2_scan_glock_lru() to make sure that glocks that are still referenced cannot be freed. It failed to account for the bias state_change() adds to the refcount for held glocks, so held glocks are no longer removed from the glock cache, which can lead to out-of-memory problems. Fix that. (The inodes those glocks are associated with do get shrunk and do get pushed out of memory.) In addition, use the same eligibility check in gfs2_scan_glock_lru() and gfs2_dispose_glock_lru(). Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 2882a42e88aa..04e0a8ac61d7 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1987,6 +1987,14 @@ static int glock_cmp(void *priv, const struct list_head *a, return 0; } +static bool can_free_glock(struct gfs2_glock *gl) +{ + bool held = gl->gl_state != LM_ST_UNLOCKED; + + return !test_bit(GLF_LOCK, &gl->gl_flags) && + gl->gl_lockref.count == held; +} + /** * gfs2_dispose_glock_lru - Demote a list of glocks * @list: The list to dispose of @@ -2020,7 +2028,7 @@ add_back_to_lru: atomic_inc(&lru_count); continue; } - if (test_bit(GLF_LOCK, &gl->gl_flags)) { + if (!can_free_glock(gl)) { spin_unlock(&gl->gl_lockref.lock); goto add_back_to_lru; } @@ -2052,16 +2060,10 @@ static long gfs2_scan_glock_lru(int nr) list_for_each_entry_safe(gl, next, &lru_list, gl_lru) { if (nr-- <= 0) break; - /* Test for being demotable */ - if (!test_bit(GLF_LOCK, &gl->gl_flags)) { - if (!spin_trylock(&gl->gl_lockref.lock)) - continue; - if (!gl->gl_lockref.count) { - list_move(&gl->gl_lru, &dispose); - atomic_dec(&lru_count); - freed++; - } - spin_unlock(&gl->gl_lockref.lock); + if (can_free_glock(gl)) { + list_move(&gl->gl_lru, &dispose); + atomic_dec(&lru_count); + freed++; } } if (!list_empty(&dispose)) -- cgit v1.2.3-59-g8ed1b From 455597a55f402e52e1c577c921bf5fe3aa4d2281 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:17:58 -0400 Subject: dlm: switch to GFP_ATOMIC in dlm allocations Replace GFP_NOFS with GFP_ATOMIC. Also stop using idr_preload which uses a non-bh spin_lock. This is further preparation for softirq message processing. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lock.c | 2 -- fs/dlm/memory.c | 6 +++--- fs/dlm/recover.c | 2 -- fs/dlm/requestqueue.c | 2 +- 4 files changed, 4 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 2f94ffc3cf82..d87464614bc5 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -1206,13 +1206,11 @@ static int _create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret, INIT_LIST_HEAD(&lkb->lkb_ownqueue); INIT_LIST_HEAD(&lkb->lkb_rsb_lookup); - idr_preload(GFP_NOFS); spin_lock(&ls->ls_lkbidr_spin); rv = idr_alloc(&ls->ls_lkbidr, lkb, start, end, GFP_NOWAIT); if (rv >= 0) lkb->lkb_id = rv; spin_unlock(&ls->ls_lkbidr_spin); - idr_preload_end(); if (rv < 0) { log_error(ls, "create_lkb idr error %d", rv); diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index ab663ca66aca..15a8b1cee433 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -84,7 +84,7 @@ char *dlm_allocate_lvb(struct dlm_ls *ls) { char *p; - p = kzalloc(ls->ls_lvblen, GFP_NOFS); + p = kzalloc(ls->ls_lvblen, GFP_ATOMIC); return p; } @@ -97,7 +97,7 @@ struct dlm_rsb *dlm_allocate_rsb(struct dlm_ls *ls) { struct dlm_rsb *r; - r = kmem_cache_zalloc(rsb_cache, GFP_NOFS); + r = kmem_cache_zalloc(rsb_cache, GFP_ATOMIC); return r; } @@ -112,7 +112,7 @@ struct dlm_lkb *dlm_allocate_lkb(struct dlm_ls *ls) { struct dlm_lkb *lkb; - lkb = kmem_cache_zalloc(lkb_cache, GFP_NOFS); + lkb = kmem_cache_zalloc(lkb_cache, GFP_ATOMIC); return lkb; } diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 53917c0aa3c0..ce6dc914cb86 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -310,7 +310,6 @@ static int recover_idr_add(struct dlm_rsb *r) struct dlm_ls *ls = r->res_ls; int rv; - idr_preload(GFP_NOFS); spin_lock(&ls->ls_recover_idr_lock); if (r->res_id) { rv = -1; @@ -326,7 +325,6 @@ static int recover_idr_add(struct dlm_rsb *r) rv = 0; out_unlock: spin_unlock(&ls->ls_recover_idr_lock); - idr_preload_end(); return rv; } diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c index 892d6ca21e74..c05940afd063 100644 --- a/fs/dlm/requestqueue.c +++ b/fs/dlm/requestqueue.c @@ -37,7 +37,7 @@ void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, int length = le16_to_cpu(ms->m_header.h_length) - sizeof(struct dlm_message); - e = kmalloc(sizeof(struct rq_entry) + length, GFP_NOFS); + e = kmalloc(sizeof(struct rq_entry) + length, GFP_ATOMIC); if (!e) { log_print("dlm_add_requestqueue: out of memory len %d", length); return; -- cgit v1.2.3-59-g8ed1b From 29e345f3c68e2bcf094162fc36394d348ccfb9ff Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:17:59 -0400 Subject: dlm: move root_list functionality to recover.c Move dlm_create_root_list() and dlm_release_root_list() to recover.c and declare them static because they are only used there. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/recover.c | 42 ------------------------------------------ fs/dlm/recover.h | 2 -- fs/dlm/recoverd.c | 39 +++++++++++++++++++++++++++++++++++++++ 3 files changed, 39 insertions(+), 44 deletions(-) (limited to 'fs') diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index ce6dc914cb86..6abc283f8f36 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -889,48 +889,6 @@ void dlm_recover_rsbs(struct dlm_ls *ls) /* Create a single list of all root rsb's to be used during recovery */ -int dlm_create_root_list(struct dlm_ls *ls) -{ - struct rb_node *n; - struct dlm_rsb *r; - int i, error = 0; - - down_write(&ls->ls_root_sem); - if (!list_empty(&ls->ls_root_list)) { - log_error(ls, "root list not empty"); - error = -EINVAL; - goto out; - } - - for (i = 0; i < ls->ls_rsbtbl_size; i++) { - spin_lock(&ls->ls_rsbtbl[i].lock); - for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { - r = rb_entry(n, struct dlm_rsb, res_hashnode); - list_add(&r->res_root_list, &ls->ls_root_list); - dlm_hold_rsb(r); - } - - if (!RB_EMPTY_ROOT(&ls->ls_rsbtbl[i].toss)) - log_error(ls, "dlm_create_root_list toss not empty"); - spin_unlock(&ls->ls_rsbtbl[i].lock); - } - out: - up_write(&ls->ls_root_sem); - return error; -} - -void dlm_release_root_list(struct dlm_ls *ls) -{ - struct dlm_rsb *r, *safe; - - down_write(&ls->ls_root_sem); - list_for_each_entry_safe(r, safe, &ls->ls_root_list, res_root_list) { - list_del_init(&r->res_root_list); - dlm_put_rsb(r); - } - up_write(&ls->ls_root_sem); -} - void dlm_clear_toss(struct dlm_ls *ls) { struct rb_node *n, *next; diff --git a/fs/dlm/recover.h b/fs/dlm/recover.h index dbc51013ecad..0b54550ee055 100644 --- a/fs/dlm/recover.h +++ b/fs/dlm/recover.h @@ -23,8 +23,6 @@ int dlm_recover_masters(struct dlm_ls *ls, uint64_t seq); int dlm_recover_master_reply(struct dlm_ls *ls, const struct dlm_rcom *rc); int dlm_recover_locks(struct dlm_ls *ls, uint64_t seq); void dlm_recovered_lock(struct dlm_rsb *r); -int dlm_create_root_list(struct dlm_ls *ls); -void dlm_release_root_list(struct dlm_ls *ls); void dlm_clear_toss(struct dlm_ls *ls); void dlm_recover_rsbs(struct dlm_ls *ls); diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index 4d17491dea2f..8eb42554ccb0 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -20,6 +20,45 @@ #include "requestqueue.h" #include "recoverd.h" +static void dlm_create_root_list(struct dlm_ls *ls) +{ + struct rb_node *n; + struct dlm_rsb *r; + int i; + + down_write(&ls->ls_root_sem); + if (!list_empty(&ls->ls_root_list)) { + log_error(ls, "root list not empty"); + goto out; + } + + for (i = 0; i < ls->ls_rsbtbl_size; i++) { + spin_lock_bh(&ls->ls_rsbtbl[i].lock); + for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { + r = rb_entry(n, struct dlm_rsb, res_hashnode); + list_add(&r->res_root_list, &ls->ls_root_list); + dlm_hold_rsb(r); + } + + if (!RB_EMPTY_ROOT(&ls->ls_rsbtbl[i].toss)) + log_error(ls, "%s toss not empty", __func__); + spin_unlock_bh(&ls->ls_rsbtbl[i].lock); + } + out: + up_write(&ls->ls_root_sem); +} + +static void dlm_release_root_list(struct dlm_ls *ls) +{ + struct dlm_rsb *r, *safe; + + down_write(&ls->ls_root_sem); + list_for_each_entry_safe(r, safe, &ls->ls_root_list, res_root_list) { + list_del_init(&r->res_root_list); + dlm_put_rsb(r); + } + up_write(&ls->ls_root_sem); +} /* If the start for which we're re-enabling locking (seq) has been superseded by a newer stop (ls_recover_seq), we need to leave locking disabled. -- cgit v1.2.3-59-g8ed1b From aff46e0f24cd3adc54ec83f4cf834ff9ccb69307 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:00 -0400 Subject: dlm: use a new list for recovery of master rsb names Add a new "masters_list" for master rsb structs, with a new rwlock. The new list is created and used during the recovery process to send the master rsb names to new nodes. With this change, the current "root_list" can be used without locking. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dir.c | 22 +++++++----------- fs/dlm/dlm_internal.h | 3 +++ fs/dlm/lock.c | 2 ++ fs/dlm/lockspace.c | 2 ++ fs/dlm/recoverd.c | 64 +++++++++++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 79 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index f6acba4310a7..10753486049a 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -216,16 +216,13 @@ static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, const char *name, if (!rv) return r; - down_read(&ls->ls_root_sem); - list_for_each_entry(r, &ls->ls_root_list, res_root_list) { + list_for_each_entry(r, &ls->ls_masters_list, res_masters_list) { if (len == r->res_length && !memcmp(name, r->res_name, len)) { - up_read(&ls->ls_root_sem); log_debug(ls, "find_rsb_root revert to root_list %s", r->res_name); return r; } } - up_read(&ls->ls_root_sem); return NULL; } @@ -241,7 +238,7 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, int offset = 0, dir_nodeid; __be16 be_namelen; - down_read(&ls->ls_root_sem); + read_lock(&ls->ls_masters_lock); if (inlen > 1) { r = find_rsb_root(ls, inbuf, inlen); @@ -250,16 +247,13 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, nodeid, inlen, inlen, inbuf); goto out; } - list = r->res_root_list.next; + list = r->res_masters_list.next; } else { - list = ls->ls_root_list.next; + list = ls->ls_masters_list.next; } - for (offset = 0; list != &ls->ls_root_list; list = list->next) { - r = list_entry(list, struct dlm_rsb, res_root_list); - if (r->res_nodeid) - continue; - + for (offset = 0; list != &ls->ls_masters_list; list = list->next) { + r = list_entry(list, struct dlm_rsb, res_masters_list); dir_nodeid = dlm_dir_nodeid(r); if (dir_nodeid != nodeid) continue; @@ -294,7 +288,7 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, * terminating record. */ - if ((list == &ls->ls_root_list) && + if ((list == &ls->ls_masters_list) && (offset + sizeof(uint16_t) <= outlen)) { be_namelen = cpu_to_be16(0xFFFF); memcpy(outbuf + offset, &be_namelen, sizeof(__be16)); @@ -302,6 +296,6 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, ls->ls_recover_dir_sent_msg++; } out: - up_read(&ls->ls_root_sem); + read_unlock(&ls->ls_masters_lock); } diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 1d2ee5c2d23d..3524f2b33f2c 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -342,6 +342,7 @@ struct dlm_rsb { struct list_head res_waitqueue; struct list_head res_root_list; /* used for recovery */ + struct list_head res_masters_list; /* used for recovery */ struct list_head res_recover_list; /* used for recovery */ int res_recover_locks_count; @@ -675,6 +676,8 @@ struct dlm_ls { struct list_head ls_root_list; /* root resources */ struct rw_semaphore ls_root_sem; /* protect root_list */ + struct list_head ls_masters_list; /* root resources */ + rwlock_t ls_masters_lock; /* protect root_list */ const struct dlm_lockspace_ops *ls_ops; void *ls_ops_arg; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index d87464614bc5..e0ab7432ca4d 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -423,6 +423,7 @@ static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len, INIT_LIST_HEAD(&r->res_waitqueue); INIT_LIST_HEAD(&r->res_root_list); INIT_LIST_HEAD(&r->res_recover_list); + INIT_LIST_HEAD(&r->res_masters_list); *r_ret = r; return 0; @@ -1168,6 +1169,7 @@ static void kill_rsb(struct kref *kref) DLM_ASSERT(list_empty(&r->res_waitqueue), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_root_list), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_recover_list), dlm_dump_rsb(r);); + DLM_ASSERT(list_empty(&r->res_masters_list), dlm_dump_rsb(r);); } /* Attaching/detaching lkb's from rsb's is for rsb reference counting. diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 0455dddb0797..c427c76b5f07 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -582,6 +582,8 @@ static int new_lockspace(const char *name, const char *cluster, init_waitqueue_head(&ls->ls_wait_general); INIT_LIST_HEAD(&ls->ls_root_list); init_rwsem(&ls->ls_root_sem); + INIT_LIST_HEAD(&ls->ls_masters_list); + rwlock_init(&ls->ls_masters_lock); spin_lock(&lslist_lock); ls->ls_create_count = 1; diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index 8eb42554ccb0..dfce8fc6a783 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -20,6 +20,48 @@ #include "requestqueue.h" #include "recoverd.h" +static int dlm_create_masters_list(struct dlm_ls *ls) +{ + struct rb_node *n; + struct dlm_rsb *r; + int i, error = 0; + + write_lock(&ls->ls_masters_lock); + if (!list_empty(&ls->ls_masters_list)) { + log_error(ls, "root list not empty"); + error = -EINVAL; + goto out; + } + + for (i = 0; i < ls->ls_rsbtbl_size; i++) { + spin_lock_bh(&ls->ls_rsbtbl[i].lock); + for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { + r = rb_entry(n, struct dlm_rsb, res_hashnode); + if (r->res_nodeid) + continue; + + list_add(&r->res_masters_list, &ls->ls_masters_list); + dlm_hold_rsb(r); + } + spin_unlock_bh(&ls->ls_rsbtbl[i].lock); + } + out: + write_unlock(&ls->ls_masters_lock); + return error; +} + +static void dlm_release_masters_list(struct dlm_ls *ls) +{ + struct dlm_rsb *r, *safe; + + write_lock(&ls->ls_masters_lock); + list_for_each_entry_safe(r, safe, &ls->ls_masters_list, res_masters_list) { + list_del_init(&r->res_masters_list); + dlm_put_rsb(r); + } + write_unlock(&ls->ls_masters_lock); +} + static void dlm_create_root_list(struct dlm_ls *ls) { struct rb_node *n; @@ -123,6 +165,23 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) dlm_recover_dir_nodeid(ls); + /* Create a snapshot of all active rsbs were we are the master of. + * During the barrier between dlm_recover_members_wait() and + * dlm_recover_directory() other nodes can dump their necessary + * directory dlm_rsb (r->res_dir_nodeid == nodeid) in rcom + * communication dlm_copy_master_names() handling. + * + * TODO We should create a per lockspace list that contains rsbs + * that we are the master of. Instead of creating this list while + * recovery we keep track of those rsbs while locking handling and + * recovery can use it when necessary. + */ + error = dlm_create_masters_list(ls); + if (error) { + log_rinfo(ls, "dlm_create_masters_list error %d", error); + goto fail; + } + ls->ls_recover_dir_sent_res = 0; ls->ls_recover_dir_sent_msg = 0; ls->ls_recover_locks_in = 0; @@ -132,6 +191,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_members_wait(ls, rv->seq); if (error) { log_rinfo(ls, "dlm_recover_members_wait error %d", error); + dlm_release_masters_list(ls); goto fail; } @@ -145,6 +205,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_directory(ls, rv->seq); if (error) { log_rinfo(ls, "dlm_recover_directory error %d", error); + dlm_release_masters_list(ls); goto fail; } @@ -153,9 +214,12 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_directory_wait(ls, rv->seq); if (error) { log_rinfo(ls, "dlm_recover_directory_wait error %d", error); + dlm_release_masters_list(ls); goto fail; } + dlm_release_masters_list(ls); + log_rinfo(ls, "dlm_recover_directory %u out %u messages", ls->ls_recover_dir_sent_res, ls->ls_recover_dir_sent_msg); -- cgit v1.2.3-59-g8ed1b From 3a747f4a2ee85d51b905e2df940de4a924f8060a Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:01 -0400 Subject: dlm: move rsb root_list to ls_recover() stack Move the rsb root_list from the lockspace to a stack variable since it is now only used by the ls_recover() function. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dir.c | 6 ++---- fs/dlm/dir.h | 3 ++- fs/dlm/dlm_internal.h | 6 ++---- fs/dlm/lock.c | 6 ++---- fs/dlm/lock.h | 2 +- fs/dlm/lockspace.c | 2 -- fs/dlm/recover.c | 30 ++++++++++------------------ fs/dlm/recover.h | 8 +++++--- fs/dlm/recoverd.c | 54 ++++++++++++++++++++++----------------------------- 9 files changed, 47 insertions(+), 70 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index 10753486049a..3da00c46cbb3 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -47,15 +47,13 @@ int dlm_dir_nodeid(struct dlm_rsb *r) return r->res_dir_nodeid; } -void dlm_recover_dir_nodeid(struct dlm_ls *ls) +void dlm_recover_dir_nodeid(struct dlm_ls *ls, const struct list_head *root_list) { struct dlm_rsb *r; - down_read(&ls->ls_root_sem); - list_for_each_entry(r, &ls->ls_root_list, res_root_list) { + list_for_each_entry(r, root_list, res_root_list) { r->res_dir_nodeid = dlm_hash2nodeid(ls, r->res_hash); } - up_read(&ls->ls_root_sem); } int dlm_recover_directory(struct dlm_ls *ls, uint64_t seq) diff --git a/fs/dlm/dir.h b/fs/dlm/dir.h index 39ecb69d7ef3..5b2a7ee3762d 100644 --- a/fs/dlm/dir.h +++ b/fs/dlm/dir.h @@ -14,7 +14,8 @@ int dlm_dir_nodeid(struct dlm_rsb *rsb); int dlm_hash2nodeid(struct dlm_ls *ls, uint32_t hash); -void dlm_recover_dir_nodeid(struct dlm_ls *ls); +void dlm_recover_dir_nodeid(struct dlm_ls *ls, + const struct list_head *root_list); int dlm_recover_directory(struct dlm_ls *ls, uint64_t seq); void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, char *outbuf, int outlen, int nodeid); diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 3524f2b33f2c..f434325d5bc8 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -674,10 +674,8 @@ struct dlm_ls { wait_queue_head_t ls_recover_lock_wait; spinlock_t ls_clear_proc_locks; - struct list_head ls_root_list; /* root resources */ - struct rw_semaphore ls_root_sem; /* protect root_list */ - struct list_head ls_masters_list; /* root resources */ - rwlock_t ls_masters_lock; /* protect root_list */ + struct list_head ls_masters_list; /* root resources */ + rwlock_t ls_masters_lock; /* protect root_list */ const struct dlm_lockspace_ops *ls_ops; void *ls_ops_arg; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index e0ab7432ca4d..43a2f4d0af53 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -5227,7 +5227,7 @@ static void purge_dead_list(struct dlm_ls *ls, struct dlm_rsb *r, /* Get rid of locks held by nodes that are gone. */ -void dlm_recover_purge(struct dlm_ls *ls) +void dlm_recover_purge(struct dlm_ls *ls, const struct list_head *root_list) { struct dlm_rsb *r; struct dlm_member *memb; @@ -5246,8 +5246,7 @@ void dlm_recover_purge(struct dlm_ls *ls) if (!nodes_count) return; - down_write(&ls->ls_root_sem); - list_for_each_entry(r, &ls->ls_root_list, res_root_list) { + list_for_each_entry(r, root_list, res_root_list) { hold_rsb(r); lock_rsb(r); if (is_master(r)) { @@ -5262,7 +5261,6 @@ void dlm_recover_purge(struct dlm_ls *ls) unhold_rsb(r); cond_resched(); } - up_write(&ls->ls_root_sem); if (lkb_count) log_rinfo(ls, "dlm_recover_purge %u locks for %u nodes", diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index b54e2cbbe6e2..c8ff7780d3cc 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -31,7 +31,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, int dlm_search_rsb_tree(struct rb_root *tree, const void *name, int len, struct dlm_rsb **r_ret); -void dlm_recover_purge(struct dlm_ls *ls); +void dlm_recover_purge(struct dlm_ls *ls, const struct list_head *root_list); void dlm_purge_mstcpy_locks(struct dlm_rsb *r); void dlm_recover_grant(struct dlm_ls *ls); int dlm_recover_waiters_post(struct dlm_ls *ls); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index c427c76b5f07..da756e5c0f6c 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -580,8 +580,6 @@ static int new_lockspace(const char *name, const char *cluster, ls->ls_recover_list_count = 0; ls->ls_local_handle = ls; init_waitqueue_head(&ls->ls_wait_general); - INIT_LIST_HEAD(&ls->ls_root_list); - init_rwsem(&ls->ls_root_sem); INIT_LIST_HEAD(&ls->ls_masters_list); rwlock_init(&ls->ls_masters_lock); diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 6abc283f8f36..172c6b73f37a 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -519,7 +519,8 @@ static int recover_master_static(struct dlm_rsb *r, unsigned int *count) * the correct dir node. */ -int dlm_recover_masters(struct dlm_ls *ls, uint64_t seq) +int dlm_recover_masters(struct dlm_ls *ls, uint64_t seq, + const struct list_head *root_list) { struct dlm_rsb *r; unsigned int total = 0; @@ -529,10 +530,8 @@ int dlm_recover_masters(struct dlm_ls *ls, uint64_t seq) log_rinfo(ls, "dlm_recover_masters"); - down_read(&ls->ls_root_sem); - list_for_each_entry(r, &ls->ls_root_list, res_root_list) { + list_for_each_entry(r, root_list, res_root_list) { if (dlm_recovery_stopped(ls)) { - up_read(&ls->ls_root_sem); error = -EINTR; goto out; } @@ -546,12 +545,9 @@ int dlm_recover_masters(struct dlm_ls *ls, uint64_t seq) cond_resched(); total++; - if (error) { - up_read(&ls->ls_root_sem); + if (error) goto out; - } } - up_read(&ls->ls_root_sem); log_rinfo(ls, "dlm_recover_masters %u of %u", count, total); @@ -656,13 +652,13 @@ static int recover_locks(struct dlm_rsb *r, uint64_t seq) return error; } -int dlm_recover_locks(struct dlm_ls *ls, uint64_t seq) +int dlm_recover_locks(struct dlm_ls *ls, uint64_t seq, + const struct list_head *root_list) { struct dlm_rsb *r; int error, count = 0; - down_read(&ls->ls_root_sem); - list_for_each_entry(r, &ls->ls_root_list, res_root_list) { + list_for_each_entry(r, root_list, res_root_list) { if (is_master(r)) { rsb_clear_flag(r, RSB_NEW_MASTER); continue; @@ -673,19 +669,15 @@ int dlm_recover_locks(struct dlm_ls *ls, uint64_t seq) if (dlm_recovery_stopped(ls)) { error = -EINTR; - up_read(&ls->ls_root_sem); goto out; } error = recover_locks(r, seq); - if (error) { - up_read(&ls->ls_root_sem); + if (error) goto out; - } count += r->res_recover_locks_count; } - up_read(&ls->ls_root_sem); log_rinfo(ls, "dlm_recover_locks %d out", count); @@ -854,13 +846,12 @@ static void recover_grant(struct dlm_rsb *r) rsb_set_flag(r, RSB_RECOVER_GRANT); } -void dlm_recover_rsbs(struct dlm_ls *ls) +void dlm_recover_rsbs(struct dlm_ls *ls, const struct list_head *root_list) { struct dlm_rsb *r; unsigned int count = 0; - down_read(&ls->ls_root_sem); - list_for_each_entry(r, &ls->ls_root_list, res_root_list) { + list_for_each_entry(r, root_list, res_root_list) { lock_rsb(r); if (is_master(r)) { if (rsb_flag(r, RSB_RECOVER_CONVERT)) @@ -881,7 +872,6 @@ void dlm_recover_rsbs(struct dlm_ls *ls) rsb_clear_flag(r, RSB_NEW_MASTER2); unlock_rsb(r); } - up_read(&ls->ls_root_sem); if (count) log_rinfo(ls, "dlm_recover_rsbs %d done", count); diff --git a/fs/dlm/recover.h b/fs/dlm/recover.h index 0b54550ee055..efc79a6e577d 100644 --- a/fs/dlm/recover.h +++ b/fs/dlm/recover.h @@ -19,12 +19,14 @@ int dlm_recover_members_wait(struct dlm_ls *ls, uint64_t seq); int dlm_recover_directory_wait(struct dlm_ls *ls, uint64_t seq); int dlm_recover_locks_wait(struct dlm_ls *ls, uint64_t seq); int dlm_recover_done_wait(struct dlm_ls *ls, uint64_t seq); -int dlm_recover_masters(struct dlm_ls *ls, uint64_t seq); +int dlm_recover_masters(struct dlm_ls *ls, uint64_t seq, + const struct list_head *root_list); int dlm_recover_master_reply(struct dlm_ls *ls, const struct dlm_rcom *rc); -int dlm_recover_locks(struct dlm_ls *ls, uint64_t seq); +int dlm_recover_locks(struct dlm_ls *ls, uint64_t seq, + const struct list_head *root_list); void dlm_recovered_lock(struct dlm_rsb *r); void dlm_clear_toss(struct dlm_ls *ls); -void dlm_recover_rsbs(struct dlm_ls *ls); +void dlm_recover_rsbs(struct dlm_ls *ls, const struct list_head *root_list); #endif /* __RECOVER_DOT_H__ */ diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index dfce8fc6a783..f6acc7351625 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -62,23 +62,17 @@ static void dlm_release_masters_list(struct dlm_ls *ls) write_unlock(&ls->ls_masters_lock); } -static void dlm_create_root_list(struct dlm_ls *ls) +static void dlm_create_root_list(struct dlm_ls *ls, struct list_head *root_list) { struct rb_node *n; struct dlm_rsb *r; int i; - down_write(&ls->ls_root_sem); - if (!list_empty(&ls->ls_root_list)) { - log_error(ls, "root list not empty"); - goto out; - } - for (i = 0; i < ls->ls_rsbtbl_size; i++) { spin_lock_bh(&ls->ls_rsbtbl[i].lock); for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); - list_add(&r->res_root_list, &ls->ls_root_list); + list_add(&r->res_root_list, root_list); dlm_hold_rsb(r); } @@ -86,20 +80,16 @@ static void dlm_create_root_list(struct dlm_ls *ls) log_error(ls, "%s toss not empty", __func__); spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } - out: - up_write(&ls->ls_root_sem); } -static void dlm_release_root_list(struct dlm_ls *ls) +static void dlm_release_root_list(struct list_head *root_list) { struct dlm_rsb *r, *safe; - down_write(&ls->ls_root_sem); - list_for_each_entry_safe(r, safe, &ls->ls_root_list, res_root_list) { + list_for_each_entry_safe(r, safe, root_list, res_root_list) { list_del_init(&r->res_root_list); dlm_put_rsb(r); } - up_write(&ls->ls_root_sem); } /* If the start for which we're re-enabling locking (seq) has been superseded @@ -131,6 +121,7 @@ static int enable_locking(struct dlm_ls *ls, uint64_t seq) static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) { + LIST_HEAD(root_list); unsigned long start; int error, neg = 0; @@ -147,7 +138,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) * routines. */ - dlm_create_root_list(ls); + dlm_create_root_list(ls, &root_list); /* * Add or remove nodes from the lockspace's ls_nodes list. @@ -163,7 +154,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) goto fail; } - dlm_recover_dir_nodeid(ls); + dlm_recover_dir_nodeid(ls, &root_list); /* Create a snapshot of all active rsbs were we are the master of. * During the barrier between dlm_recover_members_wait() and @@ -179,7 +170,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_create_masters_list(ls); if (error) { log_rinfo(ls, "dlm_create_masters_list error %d", error); - goto fail; + goto fail_root_list; } ls->ls_recover_dir_sent_res = 0; @@ -192,7 +183,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) if (error) { log_rinfo(ls, "dlm_recover_members_wait error %d", error); dlm_release_masters_list(ls); - goto fail; + goto fail_root_list; } start = jiffies; @@ -206,7 +197,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) if (error) { log_rinfo(ls, "dlm_recover_directory error %d", error); dlm_release_masters_list(ls); - goto fail; + goto fail_root_list; } dlm_set_recover_status(ls, DLM_RS_DIR); @@ -215,7 +206,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) if (error) { log_rinfo(ls, "dlm_recover_directory_wait error %d", error); dlm_release_masters_list(ls); - goto fail; + goto fail_root_list; } dlm_release_masters_list(ls); @@ -233,7 +224,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) if (dlm_recovery_stopped(ls)) { error = -EINTR; - goto fail; + goto fail_root_list; } if (neg || dlm_no_directory(ls)) { @@ -241,27 +232,27 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) * Clear lkb's for departed nodes. */ - dlm_recover_purge(ls); + dlm_recover_purge(ls, &root_list); /* * Get new master nodeid's for rsb's that were mastered on * departed nodes. */ - error = dlm_recover_masters(ls, rv->seq); + error = dlm_recover_masters(ls, rv->seq, &root_list); if (error) { log_rinfo(ls, "dlm_recover_masters error %d", error); - goto fail; + goto fail_root_list; } /* * Send our locks on remastered rsb's to the new masters. */ - error = dlm_recover_locks(ls, rv->seq); + error = dlm_recover_locks(ls, rv->seq, &root_list); if (error) { log_rinfo(ls, "dlm_recover_locks error %d", error); - goto fail; + goto fail_root_list; } dlm_set_recover_status(ls, DLM_RS_LOCKS); @@ -269,7 +260,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_locks_wait(ls, rv->seq); if (error) { log_rinfo(ls, "dlm_recover_locks_wait error %d", error); - goto fail; + goto fail_root_list; } log_rinfo(ls, "dlm_recover_locks %u in", @@ -281,7 +272,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) * settings. */ - dlm_recover_rsbs(ls); + dlm_recover_rsbs(ls, &root_list); } else { /* * Other lockspace members may be going through the "neg" steps @@ -293,11 +284,11 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) error = dlm_recover_locks_wait(ls, rv->seq); if (error) { log_rinfo(ls, "dlm_recover_locks_wait error %d", error); - goto fail; + goto fail_root_list; } } - dlm_release_root_list(ls); + dlm_release_root_list(&root_list); /* * Purge directory-related requests that are saved in requestqueue. @@ -346,8 +337,9 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) return 0; + fail_root_list: + dlm_release_root_list(&root_list); fail: - dlm_release_root_list(ls); mutex_unlock(&ls->ls_recoverd_active); return error; -- cgit v1.2.3-59-g8ed1b From 3ae67760567438ff857e79bd799154b1f7da3b2a Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:02 -0400 Subject: dlm: add new struct to save position in dlm_copy_master_names Add a new struct to save the current position in the rsb masters_list while sending the rsb names to other nodes. The rsb names are sent in multiple chunks, and for each new chunk, the new "dlm_dir_dump" struct saves the last position in the masters_list. The new struct is also used to save more information to sanity check the recovery process. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dir.c | 115 ++++++++++++++++++++++++++++++++++++++++++++++++-- fs/dlm/dlm_internal.h | 4 +- fs/dlm/lockspace.c | 2 + fs/dlm/recoverd.c | 5 --- 4 files changed, 116 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index 3da00c46cbb3..0dc8a1d9e411 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -224,6 +224,80 @@ static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, const char *name, return NULL; } +struct dlm_dir_dump { + /* init values to match if whole + * dump fits to one seq. Sanity check only. + */ + uint64_t seq_init; + uint64_t nodeid_init; + /* compare local pointer with last lookup, + * just a sanity check. + */ + struct list_head *last; + + unsigned int sent_res; /* for log info */ + unsigned int sent_msg; /* for log info */ + + struct list_head list; +}; + +static void drop_dir_ctx(struct dlm_ls *ls, int nodeid) +{ + struct dlm_dir_dump *dd, *safe; + + write_lock(&ls->ls_dir_dump_lock); + list_for_each_entry_safe(dd, safe, &ls->ls_dir_dump_list, list) { + if (dd->nodeid_init == nodeid) { + log_error(ls, "drop dump seq %llu", + (unsigned long long)dd->seq_init); + list_del(&dd->list); + kfree(dd); + } + } + write_unlock(&ls->ls_dir_dump_lock); +} + +static struct dlm_dir_dump *lookup_dir_dump(struct dlm_ls *ls, int nodeid) +{ + struct dlm_dir_dump *iter, *dd = NULL; + + read_lock(&ls->ls_dir_dump_lock); + list_for_each_entry(iter, &ls->ls_dir_dump_list, list) { + if (iter->nodeid_init == nodeid) { + dd = iter; + break; + } + } + read_unlock(&ls->ls_dir_dump_lock); + + return dd; +} + +static struct dlm_dir_dump *init_dir_dump(struct dlm_ls *ls, int nodeid) +{ + struct dlm_dir_dump *dd; + + dd = lookup_dir_dump(ls, nodeid); + if (dd) { + log_error(ls, "found ongoing dir dump for node %d, will drop it", + nodeid); + drop_dir_ctx(ls, nodeid); + } + + dd = kzalloc(sizeof(*dd), GFP_ATOMIC); + if (!dd) + return NULL; + + dd->seq_init = ls->ls_recover_seq; + dd->nodeid_init = nodeid; + + write_lock(&ls->ls_dir_dump_lock); + list_add(&dd->list, &ls->ls_dir_dump_list); + write_unlock(&ls->ls_dir_dump_lock); + + return dd; +} + /* Find the rsb where we left off (or start again), then send rsb names for rsb's we're master of and whose directory node matches the requesting node. inbuf is the rsb name last sent, inlen is the name's length */ @@ -234,11 +308,20 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, struct list_head *list; struct dlm_rsb *r; int offset = 0, dir_nodeid; + struct dlm_dir_dump *dd; __be16 be_namelen; read_lock(&ls->ls_masters_lock); if (inlen > 1) { + dd = lookup_dir_dump(ls, nodeid); + if (!dd) { + log_error(ls, "failed to lookup dir dump context nodeid: %d", + nodeid); + goto out; + } + + /* next chunk in dump */ r = find_rsb_root(ls, inbuf, inlen); if (!r) { log_error(ls, "copy_master_names from %d start %d %.*s", @@ -246,8 +329,25 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, goto out; } list = r->res_masters_list.next; + + /* sanity checks */ + if (dd->last != &r->res_masters_list || + dd->seq_init != ls->ls_recover_seq) { + log_error(ls, "failed dir dump sanity check seq_init: %llu seq: %llu", + (unsigned long long)dd->seq_init, + (unsigned long long)ls->ls_recover_seq); + goto out; + } } else { + dd = init_dir_dump(ls, nodeid); + if (!dd) { + log_error(ls, "failed to allocate dir dump context"); + goto out; + } + + /* start dump */ list = ls->ls_masters_list.next; + dd->last = list; } for (offset = 0; list != &ls->ls_masters_list; list = list->next) { @@ -269,7 +369,7 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, be_namelen = cpu_to_be16(0); memcpy(outbuf + offset, &be_namelen, sizeof(__be16)); offset += sizeof(__be16); - ls->ls_recover_dir_sent_msg++; + dd->sent_msg++; goto out; } @@ -278,7 +378,8 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, offset += sizeof(__be16); memcpy(outbuf + offset, r->res_name, r->res_length); offset += r->res_length; - ls->ls_recover_dir_sent_res++; + dd->sent_res++; + dd->last = list; } /* @@ -288,10 +389,18 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, if ((list == &ls->ls_masters_list) && (offset + sizeof(uint16_t) <= outlen)) { + /* end dump */ be_namelen = cpu_to_be16(0xFFFF); memcpy(outbuf + offset, &be_namelen, sizeof(__be16)); offset += sizeof(__be16); - ls->ls_recover_dir_sent_msg++; + dd->sent_msg++; + log_rinfo(ls, "dlm_recover_directory nodeid %d sent %u res out %u messages", + nodeid, dd->sent_res, dd->sent_msg); + + write_lock(&ls->ls_dir_dump_lock); + list_del_init(&dd->list); + write_unlock(&ls->ls_dir_dump_lock); + kfree(dd); } out: read_unlock(&ls->ls_masters_lock); diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index f434325d5bc8..e03a379832d5 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -660,8 +660,6 @@ struct dlm_ls { struct mutex ls_requestqueue_mutex; struct dlm_rcom *ls_recover_buf; int ls_recover_nodeid; /* for debugging */ - unsigned int ls_recover_dir_sent_res; /* for log info */ - unsigned int ls_recover_dir_sent_msg; /* for log info */ unsigned int ls_recover_locks_in; /* for log info */ uint64_t ls_rcom_seq; spinlock_t ls_rcom_spin; @@ -676,6 +674,8 @@ struct dlm_ls { struct list_head ls_masters_list; /* root resources */ rwlock_t ls_masters_lock; /* protect root_list */ + struct list_head ls_dir_dump_list; /* root resources */ + rwlock_t ls_dir_dump_lock; /* protect root_list */ const struct dlm_lockspace_ops *ls_ops; void *ls_ops_arg; diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index da756e5c0f6c..af7769f8e38c 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -582,6 +582,8 @@ static int new_lockspace(const char *name, const char *cluster, init_waitqueue_head(&ls->ls_wait_general); INIT_LIST_HEAD(&ls->ls_masters_list); rwlock_init(&ls->ls_masters_lock); + INIT_LIST_HEAD(&ls->ls_dir_dump_list); + rwlock_init(&ls->ls_dir_dump_lock); spin_lock(&lslist_lock); ls->ls_create_count = 1; diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index f6acc7351625..0b1a62167798 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -173,8 +173,6 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) goto fail_root_list; } - ls->ls_recover_dir_sent_res = 0; - ls->ls_recover_dir_sent_msg = 0; ls->ls_recover_locks_in = 0; dlm_set_recover_status(ls, DLM_RS_NODES); @@ -211,9 +209,6 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) dlm_release_masters_list(ls); - log_rinfo(ls, "dlm_recover_directory %u out %u messages", - ls->ls_recover_dir_sent_res, ls->ls_recover_dir_sent_msg); - /* * We may have outstanding operations that are waiting for a reply from * a failed node. Mark these to be resent after recovery. Unlock and -- cgit v1.2.3-59-g8ed1b From 6b52ea7916036cb01e7e37153f5e0ad342add765 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:03 -0400 Subject: dlm: drop mutex use in waiters recovery The waiters_mutex no longer needs to be used in the waiters recovery functions dlm_recover_waiters_pre() and dlm_recover_waiters_pre(). During recovery, ordinary locking operations are paused, and the recovery thread is the only context accessing the waiters list, so the lock is not needed. Access to the waiters list from debugfs functions is avoided by taking the top level recovery lock in the debugfs dump function. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/debug_fs.c | 13 +++++++++++++ fs/dlm/lock.c | 17 +++++++++-------- fs/dlm/lock.h | 1 + 3 files changed, 23 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 19cdedd56629..c238a9308323 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -737,6 +737,12 @@ static ssize_t waiters_read(struct file *file, char __user *userbuf, size_t len = DLM_DEBUG_BUF_LEN, pos = 0, ret, rv; mutex_lock(&debug_buf_lock); + ret = dlm_lock_recovery_try(ls); + if (!ret) { + rv = -EAGAIN; + goto out; + } + mutex_lock(&ls->ls_waiters_mutex); memset(debug_buf, 0, sizeof(debug_buf)); @@ -749,8 +755,10 @@ static ssize_t waiters_read(struct file *file, char __user *userbuf, pos += ret; } mutex_unlock(&ls->ls_waiters_mutex); + dlm_unlock_recovery(ls); rv = simple_read_from_buffer(userbuf, count, ppos, debug_buf, pos); +out: mutex_unlock(&debug_buf_lock); return rv; } @@ -772,7 +780,12 @@ static ssize_t waiters_write(struct file *file, const char __user *user_buf, if (n != 3) return -EINVAL; + error = dlm_lock_recovery_try(ls); + if (!error) + return -EAGAIN; + error = dlm_debug_add_lkb_to_waiters(ls, lkb_id, mstype, to_nodeid); + dlm_unlock_recovery(ls); if (error) return error; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 43a2f4d0af53..395b904a82f4 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -201,7 +201,7 @@ void dlm_dump_rsb(struct dlm_rsb *r) /* Threads cannot use the lockspace while it's being recovered */ -static inline void dlm_lock_recovery(struct dlm_ls *ls) +void dlm_lock_recovery(struct dlm_ls *ls) { down_read(&ls->ls_in_recovery); } @@ -1556,7 +1556,11 @@ static int remove_from_waiters(struct dlm_lkb *lkb, int mstype) } /* Handles situations where we might be processing a "fake" or "local" reply in - which we can't try to take waiters_mutex again. */ + * the recovery context which stops any locking activity. Only debugfs might + * change the lockspace waiters but they will held the recovery lock to ensure + * remove_from_waiters_ms() in local case will be the only user manipulating the + * lockspace waiters in recovery context. + */ static int remove_from_waiters_ms(struct dlm_lkb *lkb, const struct dlm_message *ms, bool local) @@ -1566,6 +1570,9 @@ static int remove_from_waiters_ms(struct dlm_lkb *lkb, if (!local) mutex_lock(&ls->ls_waiters_mutex); + else + WARN_ON_ONCE(!rwsem_is_locked(&ls->ls_in_recovery) || + !dlm_locking_stopped(ls)); error = _remove_from_waiters(lkb, le32_to_cpu(ms->m_type), ms); if (!local) mutex_unlock(&ls->ls_waiters_mutex); @@ -4398,7 +4405,6 @@ static void _receive_convert_reply(struct dlm_lkb *lkb, if (error) goto out; - /* local reply can happen with waiters_mutex held */ error = remove_from_waiters_ms(lkb, ms, local); if (error) goto out; @@ -4437,7 +4443,6 @@ static void _receive_unlock_reply(struct dlm_lkb *lkb, if (error) goto out; - /* local reply can happen with waiters_mutex held */ error = remove_from_waiters_ms(lkb, ms, local); if (error) goto out; @@ -4489,7 +4494,6 @@ static void _receive_cancel_reply(struct dlm_lkb *lkb, if (error) goto out; - /* local reply can happen with waiters_mutex held */ error = remove_from_waiters_ms(lkb, ms, local); if (error) goto out; @@ -4890,8 +4894,6 @@ void dlm_recover_waiters_pre(struct dlm_ls *ls) if (!ms_local) return; - mutex_lock(&ls->ls_waiters_mutex); - list_for_each_entry_safe(lkb, safe, &ls->ls_waiters, lkb_wait_reply) { dir_nodeid = dlm_dir_nodeid(lkb->lkb_resource); @@ -4984,7 +4986,6 @@ void dlm_recover_waiters_pre(struct dlm_ls *ls) } schedule(); } - mutex_unlock(&ls->ls_waiters_mutex); kfree(ms_local); } diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index c8ff7780d3cc..b2fd74a2f8eb 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -23,6 +23,7 @@ void dlm_hold_rsb(struct dlm_rsb *r); int dlm_put_lkb(struct dlm_lkb *lkb); void dlm_scan_rsbs(struct dlm_ls *ls); int dlm_lock_recovery_try(struct dlm_ls *ls); +void dlm_lock_recovery(struct dlm_ls *ls); void dlm_unlock_recovery(struct dlm_ls *ls); int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, -- cgit v1.2.3-59-g8ed1b From 097691dbadcdcaa6428adf0417cd227602eb60a4 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:04 -0400 Subject: dlm: convert ls_waiters_mutex to spinlock Convert the waiters mutex to a spinlock in prepration for processing messages in softirq context. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/debug_fs.c | 4 ++-- fs/dlm/dlm_internal.h | 2 +- fs/dlm/lock.c | 20 ++++++++++---------- fs/dlm/lockspace.c | 2 +- 4 files changed, 14 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index c238a9308323..487dcf05d076 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -743,7 +743,7 @@ static ssize_t waiters_read(struct file *file, char __user *userbuf, goto out; } - mutex_lock(&ls->ls_waiters_mutex); + spin_lock(&ls->ls_waiters_lock); memset(debug_buf, 0, sizeof(debug_buf)); list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) { @@ -754,7 +754,7 @@ static ssize_t waiters_read(struct file *file, char __user *userbuf, break; pos += ret; } - mutex_unlock(&ls->ls_waiters_mutex); + spin_unlock(&ls->ls_waiters_lock); dlm_unlock_recovery(ls); rv = simple_read_from_buffer(userbuf, count, ppos, debug_buf, pos); diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index e03a379832d5..98029fd5cd2b 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -595,7 +595,7 @@ struct dlm_ls { struct dlm_rsbtable *ls_rsbtbl; uint32_t ls_rsbtbl_size; - struct mutex ls_waiters_mutex; + spinlock_t ls_waiters_lock; struct list_head ls_waiters; /* lkbs needing a reply */ struct mutex ls_orphans_mutex; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 395b904a82f4..af677add4f5f 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -1406,7 +1406,7 @@ static int add_to_waiters(struct dlm_lkb *lkb, int mstype, int to_nodeid) struct dlm_ls *ls = lkb->lkb_resource->res_ls; int error = 0; - mutex_lock(&ls->ls_waiters_mutex); + spin_lock(&ls->ls_waiters_lock); if (is_overlap_unlock(lkb) || (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL))) { @@ -1449,7 +1449,7 @@ static int add_to_waiters(struct dlm_lkb *lkb, int mstype, int to_nodeid) log_error(ls, "addwait error %x %d flags %x %d %d %s", lkb->lkb_id, error, dlm_iflags_val(lkb), mstype, lkb->lkb_wait_type, lkb->lkb_resource->res_name); - mutex_unlock(&ls->ls_waiters_mutex); + spin_unlock(&ls->ls_waiters_lock); return error; } @@ -1549,9 +1549,9 @@ static int remove_from_waiters(struct dlm_lkb *lkb, int mstype) struct dlm_ls *ls = lkb->lkb_resource->res_ls; int error; - mutex_lock(&ls->ls_waiters_mutex); + spin_lock(&ls->ls_waiters_lock); error = _remove_from_waiters(lkb, mstype, NULL); - mutex_unlock(&ls->ls_waiters_mutex); + spin_unlock(&ls->ls_waiters_lock); return error; } @@ -1569,13 +1569,13 @@ static int remove_from_waiters_ms(struct dlm_lkb *lkb, int error; if (!local) - mutex_lock(&ls->ls_waiters_mutex); + spin_lock(&ls->ls_waiters_lock); else WARN_ON_ONCE(!rwsem_is_locked(&ls->ls_in_recovery) || !dlm_locking_stopped(ls)); error = _remove_from_waiters(lkb, le32_to_cpu(ms->m_type), ms); if (!local) - mutex_unlock(&ls->ls_waiters_mutex); + spin_unlock(&ls->ls_waiters_lock); return error; } @@ -4993,7 +4993,7 @@ static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls) { struct dlm_lkb *lkb = NULL, *iter; - mutex_lock(&ls->ls_waiters_mutex); + spin_lock(&ls->ls_waiters_lock); list_for_each_entry(iter, &ls->ls_waiters, lkb_wait_reply) { if (test_bit(DLM_IFL_RESEND_BIT, &iter->lkb_iflags)) { hold_lkb(iter); @@ -5001,7 +5001,7 @@ static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls) break; } } - mutex_unlock(&ls->ls_waiters_mutex); + spin_unlock(&ls->ls_waiters_lock); return lkb; } @@ -5101,9 +5101,9 @@ int dlm_recover_waiters_post(struct dlm_ls *ls) } /* Forcibly remove from waiters list */ - mutex_lock(&ls->ls_waiters_mutex); + spin_lock(&ls->ls_waiters_lock); list_del_init(&lkb->lkb_wait_reply); - mutex_unlock(&ls->ls_waiters_mutex); + spin_unlock(&ls->ls_waiters_lock); /* * The lkb is now clear of all prior waiters state and can be diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index af7769f8e38c..945139805605 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -515,7 +515,7 @@ static int new_lockspace(const char *name, const char *cluster, spin_lock_init(&ls->ls_lkbidr_spin); INIT_LIST_HEAD(&ls->ls_waiters); - mutex_init(&ls->ls_waiters_mutex); + spin_lock_init(&ls->ls_waiters_lock); INIT_LIST_HEAD(&ls->ls_orphans); mutex_init(&ls->ls_orphans_mutex); -- cgit v1.2.3-59-g8ed1b From cc396e2355b5ca6e1aee005f3ce99bab8f37f5ff Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:05 -0400 Subject: dlm: convert res_lock to spinlock Convert the rsb struct res_lock from a mutex to a spinlock in preparation for processing messages in softirq context. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dlm_internal.h | 2 +- fs/dlm/lock.c | 2 +- fs/dlm/lock.h | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 98029fd5cd2b..9f98a815f935 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -320,7 +320,7 @@ struct dlm_lkb { struct dlm_rsb { struct dlm_ls *res_ls; /* the lockspace */ struct kref res_ref; - struct mutex res_mutex; + spinlock_t res_lock; unsigned long res_flags; int res_length; /* length of rsb name */ int res_nodeid; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index af677add4f5f..7b309231eebd 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -415,7 +415,7 @@ static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len, r->res_ls = ls; r->res_length = len; memcpy(r->res_name, name, len); - mutex_init(&r->res_mutex); + spin_lock_init(&r->res_lock); INIT_LIST_HEAD(&r->res_lookup); INIT_LIST_HEAD(&r->res_grantqueue); diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index b2fd74a2f8eb..0f6b2700c0da 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -69,12 +69,12 @@ static inline int is_master(struct dlm_rsb *r) static inline void lock_rsb(struct dlm_rsb *r) { - mutex_lock(&r->res_mutex); + spin_lock(&r->res_lock); } static inline void unlock_rsb(struct dlm_rsb *r) { - mutex_unlock(&r->res_mutex); + spin_unlock(&r->res_lock); } #endif -- cgit v1.2.3-59-g8ed1b From c288745f1d4a2ead903e81d2f4716e9d40b0ad85 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:06 -0400 Subject: dlm: avoid blocking receive at the end of recovery The end of the recovery process transitioned to normal message processing by temporarily blocking the receiving context, processing saved messages, then unblocking the receiving context. To avoid blocking the receiving context, the old wait_queue and mutex are replaced by a new rwlock and the new RECV_MSG_BLOCKED flag. Received messages are added to the list of saved messages, protected by the rwlock, until the flag is cleared, which happens when all saved messages have been processed. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dlm_internal.h | 5 ++--- fs/dlm/lock.c | 16 ++++++++++++++-- fs/dlm/lockspace.c | 4 +--- fs/dlm/member.c | 5 +++++ fs/dlm/requestqueue.c | 41 ++++++++--------------------------------- 5 files changed, 30 insertions(+), 41 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 9f98a815f935..61820b8c47a7 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -655,9 +655,7 @@ struct dlm_ls { struct rw_semaphore ls_in_recovery; /* block local requests */ struct rw_semaphore ls_recv_active; /* block dlm_recv */ struct list_head ls_requestqueue;/* queue remote requests */ - atomic_t ls_requestqueue_cnt; - wait_queue_head_t ls_requestqueue_wait; - struct mutex ls_requestqueue_mutex; + rwlock_t ls_requestqueue_lock; struct dlm_rcom *ls_recover_buf; int ls_recover_nodeid; /* for debugging */ unsigned int ls_recover_locks_in; /* for log info */ @@ -717,6 +715,7 @@ struct dlm_ls { #define LSFL_UEVENT_WAIT 7 #define LSFL_CB_DELAY 9 #define LSFL_NODIR 10 +#define LSFL_RECV_MSG_BLOCKED 11 #define DLM_PROC_FLAGS_CLOSING 1 #define DLM_PROC_FLAGS_COMPAT 2 diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 7b309231eebd..98d9c5a4be00 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -4752,20 +4752,32 @@ static void _receive_message(struct dlm_ls *ls, const struct dlm_message *ms, static void dlm_receive_message(struct dlm_ls *ls, const struct dlm_message *ms, int nodeid) { - if (dlm_locking_stopped(ls)) { +try_again: + read_lock(&ls->ls_requestqueue_lock); + if (test_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags)) { /* If we were a member of this lockspace, left, and rejoined, other nodes may still be sending us messages from the lockspace generation before we left. */ if (WARN_ON_ONCE(!ls->ls_generation)) { + read_unlock(&ls->ls_requestqueue_lock); log_limit(ls, "receive %d from %d ignore old gen", le32_to_cpu(ms->m_type), nodeid); return; } + read_unlock(&ls->ls_requestqueue_lock); + write_lock(&ls->ls_requestqueue_lock); + /* recheck because we hold writelock now */ + if (!test_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags)) { + write_unlock_bh(&ls->ls_requestqueue_lock); + goto try_again; + } + dlm_add_requestqueue(ls, nodeid, ms); + write_unlock(&ls->ls_requestqueue_lock); } else { - dlm_wait_requestqueue(ls); _receive_message(ls, ms, 0); + read_unlock(&ls->ls_requestqueue_lock); } } diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 945139805605..757e473bc619 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -554,9 +554,7 @@ static int new_lockspace(const char *name, const char *cluster, init_rwsem(&ls->ls_in_recovery); init_rwsem(&ls->ls_recv_active); INIT_LIST_HEAD(&ls->ls_requestqueue); - atomic_set(&ls->ls_requestqueue_cnt, 0); - init_waitqueue_head(&ls->ls_requestqueue_wait); - mutex_init(&ls->ls_requestqueue_mutex); + rwlock_init(&ls->ls_requestqueue_lock); spin_lock_init(&ls->ls_clear_proc_locks); /* Due backwards compatibility with 3.1 we need to use maximum diff --git a/fs/dlm/member.c b/fs/dlm/member.c index be7909ead71b..707cebcdc533 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -642,6 +642,11 @@ int dlm_ls_stop(struct dlm_ls *ls) set_bit(LSFL_RECOVER_STOP, &ls->ls_flags); new = test_and_clear_bit(LSFL_RUNNING, &ls->ls_flags); ls->ls_recover_seq++; + + /* activate requestqueue and stop processing */ + write_lock(&ls->ls_requestqueue_lock); + set_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags); + write_unlock(&ls->ls_requestqueue_lock); spin_unlock(&ls->ls_recover_lock); /* diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c index c05940afd063..9b646026df46 100644 --- a/fs/dlm/requestqueue.c +++ b/fs/dlm/requestqueue.c @@ -48,10 +48,7 @@ void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, memcpy(&e->request, ms, sizeof(*ms)); memcpy(&e->request.m_extra, ms->m_extra, length); - atomic_inc(&ls->ls_requestqueue_cnt); - mutex_lock(&ls->ls_requestqueue_mutex); list_add_tail(&e->list, &ls->ls_requestqueue); - mutex_unlock(&ls->ls_requestqueue_mutex); } /* @@ -71,16 +68,14 @@ int dlm_process_requestqueue(struct dlm_ls *ls) struct dlm_message *ms; int error = 0; - mutex_lock(&ls->ls_requestqueue_mutex); - + write_lock(&ls->ls_requestqueue_lock); for (;;) { if (list_empty(&ls->ls_requestqueue)) { - mutex_unlock(&ls->ls_requestqueue_mutex); + clear_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags); error = 0; break; } - e = list_entry(ls->ls_requestqueue.next, struct rq_entry, list); - mutex_unlock(&ls->ls_requestqueue_mutex); + e = list_first_entry(&ls->ls_requestqueue, struct rq_entry, list); ms = &e->request; @@ -93,41 +88,23 @@ int dlm_process_requestqueue(struct dlm_ls *ls) e->recover_seq); dlm_receive_message_saved(ls, &e->request, e->recover_seq); - - mutex_lock(&ls->ls_requestqueue_mutex); list_del(&e->list); - if (atomic_dec_and_test(&ls->ls_requestqueue_cnt)) - wake_up(&ls->ls_requestqueue_wait); kfree(e); if (dlm_locking_stopped(ls)) { log_debug(ls, "process_requestqueue abort running"); - mutex_unlock(&ls->ls_requestqueue_mutex); error = -EINTR; break; } + write_unlock(&ls->ls_requestqueue_lock); schedule(); + write_lock(&ls->ls_requestqueue_lock); } + write_unlock(&ls->ls_requestqueue_lock); return error; } -/* - * After recovery is done, locking is resumed and dlm_recoverd takes all the - * saved requests and processes them as they would have been by dlm_recv. At - * the same time, dlm_recv will start receiving new requests from remote nodes. - * We want to delay dlm_recv processing new requests until dlm_recoverd has - * finished processing the old saved requests. We don't check for locking - * stopped here because dlm_ls_stop won't stop locking until it's suspended us - * (dlm_recv). - */ - -void dlm_wait_requestqueue(struct dlm_ls *ls) -{ - wait_event(ls->ls_requestqueue_wait, - atomic_read(&ls->ls_requestqueue_cnt) == 0); -} - static int purge_request(struct dlm_ls *ls, struct dlm_message *ms, int nodeid) { __le32 type = ms->m_type; @@ -158,17 +135,15 @@ void dlm_purge_requestqueue(struct dlm_ls *ls) struct dlm_message *ms; struct rq_entry *e, *safe; - mutex_lock(&ls->ls_requestqueue_mutex); + write_lock(&ls->ls_requestqueue_lock); list_for_each_entry_safe(e, safe, &ls->ls_requestqueue, list) { ms = &e->request; if (purge_request(ls, ms, e->nodeid)) { list_del(&e->list); - if (atomic_dec_and_test(&ls->ls_requestqueue_cnt)) - wake_up(&ls->ls_requestqueue_wait); kfree(e); } } - mutex_unlock(&ls->ls_requestqueue_mutex); + write_unlock(&ls->ls_requestqueue_lock); } -- cgit v1.2.3-59-g8ed1b From d52c9b8fefa3ed4f1893eea8c5f38748a83356fc Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:07 -0400 Subject: dlm: convert ls_recv_active from rw_semaphore to rwlock Convert ls_recv_active rw_semaphore to an rwlock to avoid sleeping, in preparation for softirq message processing. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dlm_internal.h | 2 +- fs/dlm/lock.c | 4 ++-- fs/dlm/lockspace.c | 2 +- fs/dlm/member.c | 4 ++-- fs/dlm/recoverd.c | 4 ++-- 5 files changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 61820b8c47a7..269c12e0824f 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -653,7 +653,7 @@ struct dlm_ls { uint64_t ls_recover_seq; struct dlm_recover *ls_recover_args; struct rw_semaphore ls_in_recovery; /* block local requests */ - struct rw_semaphore ls_recv_active; /* block dlm_recv */ + rwlock_t ls_recv_active; /* block dlm_recv */ struct list_head ls_requestqueue;/* queue remote requests */ rwlock_t ls_requestqueue_lock; struct dlm_rcom *ls_recover_buf; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 98d9c5a4be00..2f53fdfe262a 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -4837,7 +4837,7 @@ void dlm_receive_buffer(const union dlm_packet *p, int nodeid) /* this rwsem allows dlm_ls_stop() to wait for all dlm_recv threads to be inactive (in this ls) before transitioning to recovery mode */ - down_read(&ls->ls_recv_active); + read_lock(&ls->ls_recv_active); if (hd->h_cmd == DLM_MSG) dlm_receive_message(ls, &p->message, nodeid); else if (hd->h_cmd == DLM_RCOM) @@ -4845,7 +4845,7 @@ void dlm_receive_buffer(const union dlm_packet *p, int nodeid) else log_error(ls, "invalid h_cmd %d from %d lockspace %x", hd->h_cmd, nodeid, le32_to_cpu(hd->u.h_lockspace)); - up_read(&ls->ls_recv_active); + read_unlock(&ls->ls_recv_active); dlm_put_lockspace(ls); } diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 757e473bc619..c021bf684fbc 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -552,7 +552,7 @@ static int new_lockspace(const char *name, const char *cluster, ls->ls_recover_seq = get_random_u64(); ls->ls_recover_args = NULL; init_rwsem(&ls->ls_in_recovery); - init_rwsem(&ls->ls_recv_active); + rwlock_init(&ls->ls_recv_active); INIT_LIST_HEAD(&ls->ls_requestqueue); rwlock_init(&ls->ls_requestqueue_lock); spin_lock_init(&ls->ls_clear_proc_locks); diff --git a/fs/dlm/member.c b/fs/dlm/member.c index 707cebcdc533..ac1b555af9d6 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -630,7 +630,7 @@ int dlm_ls_stop(struct dlm_ls *ls) * message to the requestqueue without races. */ - down_write(&ls->ls_recv_active); + write_lock(&ls->ls_recv_active); /* * Abort any recovery that's in progress (see RECOVER_STOP, @@ -654,7 +654,7 @@ int dlm_ls_stop(struct dlm_ls *ls) * requestqueue for later. */ - up_write(&ls->ls_recv_active); + write_unlock(&ls->ls_recv_active); /* * This in_recovery lock does two things: diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index 0b1a62167798..a11ae1da2f60 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -103,7 +103,7 @@ static int enable_locking(struct dlm_ls *ls, uint64_t seq) { int error = -EINTR; - down_write(&ls->ls_recv_active); + write_lock(&ls->ls_recv_active); spin_lock(&ls->ls_recover_lock); if (ls->ls_recover_seq == seq) { @@ -115,7 +115,7 @@ static int enable_locking(struct dlm_ls *ls, uint64_t seq) } spin_unlock(&ls->ls_recover_lock); - up_write(&ls->ls_recv_active); + write_unlock(&ls->ls_recv_active); return error; } -- cgit v1.2.3-59-g8ed1b From 308533b4b1d55892d939286313fb73c1527444ce Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:08 -0400 Subject: dlm: remove schedule in receive path Remove an explicit schedule() call in the message processing path, in preparation for softirq message processing. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lock.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 2f53fdfe262a..e4cec14f9973 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -2543,7 +2543,6 @@ static void process_lookup_list(struct dlm_rsb *r) list_for_each_entry_safe(lkb, safe, &r->res_lookup, lkb_rsb_lookup) { list_del_init(&lkb->lkb_rsb_lookup); _request_lock(r, lkb); - schedule(); } } -- cgit v1.2.3-59-g8ed1b From 578acf9a87a87531df5b59b3799ccc1256a4bbcc Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:09 -0400 Subject: dlm: use spin_lock_bh for message processing Use spin_lock_bh for all spinlocks involved in message processing, in preparation for softirq message processing. DLM lock requests from user space involve dlm processing in user context, in addition to the standard kernel context, necessitating bh variants. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/ast.c | 12 +-- fs/dlm/debug_fs.c | 28 +++---- fs/dlm/dir.c | 24 +++--- fs/dlm/lock.c | 206 +++++++++++++++++++++++++++++--------------------- fs/dlm/lock.h | 4 +- fs/dlm/lockspace.c | 51 ++++++------- fs/dlm/lowcomms.c | 16 ++-- fs/dlm/member.c | 22 +++--- fs/dlm/midcomms.c | 40 +++++----- fs/dlm/rcom.c | 26 +++---- fs/dlm/recover.c | 52 ++++++------- fs/dlm/recoverd.c | 20 ++--- fs/dlm/requestqueue.c | 12 +-- fs/dlm/user.c | 32 ++++---- 14 files changed, 287 insertions(+), 258 deletions(-) (limited to 'fs') diff --git a/fs/dlm/ast.c b/fs/dlm/ast.c index 03879c94fadb..59711486d801 100644 --- a/fs/dlm/ast.c +++ b/fs/dlm/ast.c @@ -142,12 +142,12 @@ void dlm_add_cb(struct dlm_lkb *lkb, uint32_t flags, int mode, int status, cb->astparam = lkb->lkb_astparam; INIT_WORK(&cb->work, dlm_callback_work); - spin_lock(&ls->ls_cb_lock); + spin_lock_bh(&ls->ls_cb_lock); if (test_bit(LSFL_CB_DELAY, &ls->ls_flags)) list_add(&cb->list, &ls->ls_cb_delay); else queue_work(ls->ls_callback_wq, &cb->work); - spin_unlock(&ls->ls_cb_lock); + spin_unlock_bh(&ls->ls_cb_lock); break; case DLM_ENQUEUE_CALLBACK_SUCCESS: break; @@ -179,9 +179,9 @@ void dlm_callback_stop(struct dlm_ls *ls) void dlm_callback_suspend(struct dlm_ls *ls) { if (ls->ls_callback_wq) { - spin_lock(&ls->ls_cb_lock); + spin_lock_bh(&ls->ls_cb_lock); set_bit(LSFL_CB_DELAY, &ls->ls_flags); - spin_unlock(&ls->ls_cb_lock); + spin_unlock_bh(&ls->ls_cb_lock); flush_workqueue(ls->ls_callback_wq); } @@ -199,7 +199,7 @@ void dlm_callback_resume(struct dlm_ls *ls) return; more: - spin_lock(&ls->ls_cb_lock); + spin_lock_bh(&ls->ls_cb_lock); list_for_each_entry_safe(cb, safe, &ls->ls_cb_delay, list) { list_del(&cb->list); queue_work(ls->ls_callback_wq, &cb->work); @@ -210,7 +210,7 @@ more: empty = list_empty(&ls->ls_cb_delay); if (empty) clear_bit(LSFL_CB_DELAY, &ls->ls_flags); - spin_unlock(&ls->ls_cb_lock); + spin_unlock_bh(&ls->ls_cb_lock); sum += count; if (!empty) { diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 487dcf05d076..cba5514688ee 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -452,7 +452,7 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; - spin_lock(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); if (!RB_EMPTY_ROOT(tree)) { for (node = rb_first(tree); node; node = rb_next(node)) { r = rb_entry(node, struct dlm_rsb, res_hashnode); @@ -460,12 +460,12 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); return ri; } } } - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); /* * move to the first rsb in the next non-empty bucket @@ -484,18 +484,18 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) } tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; - spin_lock(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); if (!RB_EMPTY_ROOT(tree)) { node = rb_first(tree); r = rb_entry(node, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); *pos = n; return ri; } - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); } } @@ -516,7 +516,7 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) * move to the next rsb in the same bucket */ - spin_lock(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); rp = ri->rsb; next = rb_next(&rp->res_hashnode); @@ -524,12 +524,12 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) r = rb_entry(next, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); dlm_put_rsb(rp); ++*pos; return ri; } - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); dlm_put_rsb(rp); /* @@ -550,18 +550,18 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) } tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; - spin_lock(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); if (!RB_EMPTY_ROOT(tree)) { next = rb_first(tree); r = rb_entry(next, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); *pos = n; return ri; } - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); } } @@ -743,7 +743,7 @@ static ssize_t waiters_read(struct file *file, char __user *userbuf, goto out; } - spin_lock(&ls->ls_waiters_lock); + spin_lock_bh(&ls->ls_waiters_lock); memset(debug_buf, 0, sizeof(debug_buf)); list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) { @@ -754,7 +754,7 @@ static ssize_t waiters_read(struct file *file, char __user *userbuf, break; pos += ret; } - spin_unlock(&ls->ls_waiters_lock); + spin_unlock_bh(&ls->ls_waiters_lock); dlm_unlock_recovery(ls); rv = simple_read_from_buffer(userbuf, count, ppos, debug_buf, pos); diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index 0dc8a1d9e411..ff3a51c759b5 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -204,12 +204,12 @@ static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, const char *name, hash = jhash(name, len, 0); bucket = hash & (ls->ls_rsbtbl_size - 1); - spin_lock(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].keep, name, len, &r); if (rv) rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].toss, name, len, &r); - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); if (!rv) return r; @@ -245,7 +245,7 @@ static void drop_dir_ctx(struct dlm_ls *ls, int nodeid) { struct dlm_dir_dump *dd, *safe; - write_lock(&ls->ls_dir_dump_lock); + write_lock_bh(&ls->ls_dir_dump_lock); list_for_each_entry_safe(dd, safe, &ls->ls_dir_dump_list, list) { if (dd->nodeid_init == nodeid) { log_error(ls, "drop dump seq %llu", @@ -254,21 +254,21 @@ static void drop_dir_ctx(struct dlm_ls *ls, int nodeid) kfree(dd); } } - write_unlock(&ls->ls_dir_dump_lock); + write_unlock_bh(&ls->ls_dir_dump_lock); } static struct dlm_dir_dump *lookup_dir_dump(struct dlm_ls *ls, int nodeid) { struct dlm_dir_dump *iter, *dd = NULL; - read_lock(&ls->ls_dir_dump_lock); + read_lock_bh(&ls->ls_dir_dump_lock); list_for_each_entry(iter, &ls->ls_dir_dump_list, list) { if (iter->nodeid_init == nodeid) { dd = iter; break; } } - read_unlock(&ls->ls_dir_dump_lock); + read_unlock_bh(&ls->ls_dir_dump_lock); return dd; } @@ -291,9 +291,9 @@ static struct dlm_dir_dump *init_dir_dump(struct dlm_ls *ls, int nodeid) dd->seq_init = ls->ls_recover_seq; dd->nodeid_init = nodeid; - write_lock(&ls->ls_dir_dump_lock); + write_lock_bh(&ls->ls_dir_dump_lock); list_add(&dd->list, &ls->ls_dir_dump_list); - write_unlock(&ls->ls_dir_dump_lock); + write_unlock_bh(&ls->ls_dir_dump_lock); return dd; } @@ -311,7 +311,7 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, struct dlm_dir_dump *dd; __be16 be_namelen; - read_lock(&ls->ls_masters_lock); + read_lock_bh(&ls->ls_masters_lock); if (inlen > 1) { dd = lookup_dir_dump(ls, nodeid); @@ -397,12 +397,12 @@ void dlm_copy_master_names(struct dlm_ls *ls, const char *inbuf, int inlen, log_rinfo(ls, "dlm_recover_directory nodeid %d sent %u res out %u messages", nodeid, dd->sent_res, dd->sent_msg); - write_lock(&ls->ls_dir_dump_lock); + write_lock_bh(&ls->ls_dir_dump_lock); list_del_init(&dd->list); - write_unlock(&ls->ls_dir_dump_lock); + write_unlock_bh(&ls->ls_dir_dump_lock); kfree(dd); } out: - read_unlock(&ls->ls_masters_lock); + read_unlock_bh(&ls->ls_masters_lock); } diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index e4cec14f9973..4ff4ef2a5f87 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -333,6 +333,36 @@ void dlm_hold_rsb(struct dlm_rsb *r) hold_rsb(r); } +/* TODO move this to lib/refcount.c */ +static __must_check bool +dlm_refcount_dec_and_lock_bh(refcount_t *r, spinlock_t *lock) +__cond_acquires(lock) +{ + if (refcount_dec_not_one(r)) + return false; + + spin_lock_bh(lock); + if (!refcount_dec_and_test(r)) { + spin_unlock_bh(lock); + return false; + } + + return true; +} + +/* TODO move this to include/linux/kref.h */ +static inline int dlm_kref_put_lock_bh(struct kref *kref, + void (*release)(struct kref *kref), + spinlock_t *lock) +{ + if (dlm_refcount_dec_and_lock_bh(&kref->refcount, lock)) { + release(kref); + return 1; + } + + return 0; +} + /* When all references to the rsb are gone it's transferred to the tossed list for later disposal. */ @@ -342,10 +372,10 @@ static void put_rsb(struct dlm_rsb *r) uint32_t bucket = r->res_bucket; int rv; - rv = kref_put_lock(&r->res_ref, toss_rsb, - &ls->ls_rsbtbl[bucket].lock); + rv = dlm_kref_put_lock_bh(&r->res_ref, toss_rsb, + &ls->ls_rsbtbl[bucket].lock); if (rv) - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); } void dlm_put_rsb(struct dlm_rsb *r) @@ -358,17 +388,17 @@ static int pre_rsb_struct(struct dlm_ls *ls) struct dlm_rsb *r1, *r2; int count = 0; - spin_lock(&ls->ls_new_rsb_spin); + spin_lock_bh(&ls->ls_new_rsb_spin); if (ls->ls_new_rsb_count > dlm_config.ci_new_rsb_count / 2) { - spin_unlock(&ls->ls_new_rsb_spin); + spin_unlock_bh(&ls->ls_new_rsb_spin); return 0; } - spin_unlock(&ls->ls_new_rsb_spin); + spin_unlock_bh(&ls->ls_new_rsb_spin); r1 = dlm_allocate_rsb(ls); r2 = dlm_allocate_rsb(ls); - spin_lock(&ls->ls_new_rsb_spin); + spin_lock_bh(&ls->ls_new_rsb_spin); if (r1) { list_add(&r1->res_hashchain, &ls->ls_new_rsb); ls->ls_new_rsb_count++; @@ -378,7 +408,7 @@ static int pre_rsb_struct(struct dlm_ls *ls) ls->ls_new_rsb_count++; } count = ls->ls_new_rsb_count; - spin_unlock(&ls->ls_new_rsb_spin); + spin_unlock_bh(&ls->ls_new_rsb_spin); if (!count) return -ENOMEM; @@ -395,10 +425,10 @@ static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len, struct dlm_rsb *r; int count; - spin_lock(&ls->ls_new_rsb_spin); + spin_lock_bh(&ls->ls_new_rsb_spin); if (list_empty(&ls->ls_new_rsb)) { count = ls->ls_new_rsb_count; - spin_unlock(&ls->ls_new_rsb_spin); + spin_unlock_bh(&ls->ls_new_rsb_spin); log_debug(ls, "find_rsb retry %d %d %s", count, dlm_config.ci_new_rsb_count, (const char *)name); @@ -410,7 +440,7 @@ static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len, /* Convert the empty list_head to a NULL rb_node for tree usage: */ memset(&r->res_hashnode, 0, sizeof(struct rb_node)); ls->ls_new_rsb_count--; - spin_unlock(&ls->ls_new_rsb_spin); + spin_unlock_bh(&ls->ls_new_rsb_spin); r->res_ls = ls; r->res_length = len; @@ -585,7 +615,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, goto out; } - spin_lock(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl[b].lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (error) @@ -655,7 +685,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, error = get_rsb_struct(ls, name, len, &r); if (error == -EAGAIN) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); goto retry; } if (error) @@ -704,7 +734,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, out_add: error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); out_unlock: - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); out: *r_ret = r; return error; @@ -729,7 +759,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, if (error < 0) goto out; - spin_lock(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl[b].lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (error) @@ -787,7 +817,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, error = get_rsb_struct(ls, name, len, &r); if (error == -EAGAIN) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); goto retry; } if (error) @@ -802,7 +832,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); out_unlock: - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); out: *r_ret = r; return error; @@ -1019,7 +1049,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, if (error < 0) return error; - spin_lock(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl[b].lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (!error) { /* because the rsb is active, we need to lock_rsb before @@ -1027,7 +1057,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, */ hold_rsb(r); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); lock_rsb(r); __dlm_master_lookup(ls, r, our_nodeid, from_nodeid, false, @@ -1053,14 +1083,14 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, r->res_toss_time = jiffies; /* the rsb was inactive (on toss list) */ - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); return 0; not_found: error = get_rsb_struct(ls, name, len, &r); if (error == -EAGAIN) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); goto retry; } if (error) @@ -1078,7 +1108,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, if (error) { /* should never happen */ dlm_free_rsb(r); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); goto retry; } @@ -1086,7 +1116,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, *result = DLM_LU_ADD; *r_nodeid = from_nodeid; out_unlock: - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); return error; } @@ -1097,13 +1127,13 @@ static void dlm_dump_rsb_hash(struct dlm_ls *ls, uint32_t hash) int i; for (i = 0; i < ls->ls_rsbtbl_size; i++) { - spin_lock(&ls->ls_rsbtbl[i].lock); + spin_lock_bh(&ls->ls_rsbtbl[i].lock); for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); if (r->res_hash == hash) dlm_dump_rsb(r); } - spin_unlock(&ls->ls_rsbtbl[i].lock); + spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } } @@ -1116,7 +1146,7 @@ void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) hash = jhash(name, len, 0); b = hash & (ls->ls_rsbtbl_size - 1); - spin_lock(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl[b].lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (!error) goto out_dump; @@ -1127,7 +1157,7 @@ void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) out_dump: dlm_dump_rsb(r); out: - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); } static void toss_rsb(struct kref *kref) @@ -1208,11 +1238,11 @@ static int _create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret, INIT_LIST_HEAD(&lkb->lkb_ownqueue); INIT_LIST_HEAD(&lkb->lkb_rsb_lookup); - spin_lock(&ls->ls_lkbidr_spin); + spin_lock_bh(&ls->ls_lkbidr_spin); rv = idr_alloc(&ls->ls_lkbidr, lkb, start, end, GFP_NOWAIT); if (rv >= 0) lkb->lkb_id = rv; - spin_unlock(&ls->ls_lkbidr_spin); + spin_unlock_bh(&ls->ls_lkbidr_spin); if (rv < 0) { log_error(ls, "create_lkb idr error %d", rv); @@ -1233,11 +1263,11 @@ static int find_lkb(struct dlm_ls *ls, uint32_t lkid, struct dlm_lkb **lkb_ret) { struct dlm_lkb *lkb; - spin_lock(&ls->ls_lkbidr_spin); + spin_lock_bh(&ls->ls_lkbidr_spin); lkb = idr_find(&ls->ls_lkbidr, lkid); if (lkb) kref_get(&lkb->lkb_ref); - spin_unlock(&ls->ls_lkbidr_spin); + spin_unlock_bh(&ls->ls_lkbidr_spin); *lkb_ret = lkb; return lkb ? 0 : -ENOENT; @@ -1261,11 +1291,11 @@ static int __put_lkb(struct dlm_ls *ls, struct dlm_lkb *lkb) uint32_t lkid = lkb->lkb_id; int rv; - rv = kref_put_lock(&lkb->lkb_ref, kill_lkb, - &ls->ls_lkbidr_spin); + rv = dlm_kref_put_lock_bh(&lkb->lkb_ref, kill_lkb, + &ls->ls_lkbidr_spin); if (rv) { idr_remove(&ls->ls_lkbidr, lkid); - spin_unlock(&ls->ls_lkbidr_spin); + spin_unlock_bh(&ls->ls_lkbidr_spin); detach_lkb(lkb); @@ -1406,7 +1436,7 @@ static int add_to_waiters(struct dlm_lkb *lkb, int mstype, int to_nodeid) struct dlm_ls *ls = lkb->lkb_resource->res_ls; int error = 0; - spin_lock(&ls->ls_waiters_lock); + spin_lock_bh(&ls->ls_waiters_lock); if (is_overlap_unlock(lkb) || (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL))) { @@ -1449,7 +1479,7 @@ static int add_to_waiters(struct dlm_lkb *lkb, int mstype, int to_nodeid) log_error(ls, "addwait error %x %d flags %x %d %d %s", lkb->lkb_id, error, dlm_iflags_val(lkb), mstype, lkb->lkb_wait_type, lkb->lkb_resource->res_name); - spin_unlock(&ls->ls_waiters_lock); + spin_unlock_bh(&ls->ls_waiters_lock); return error; } @@ -1549,9 +1579,9 @@ static int remove_from_waiters(struct dlm_lkb *lkb, int mstype) struct dlm_ls *ls = lkb->lkb_resource->res_ls; int error; - spin_lock(&ls->ls_waiters_lock); + spin_lock_bh(&ls->ls_waiters_lock); error = _remove_from_waiters(lkb, mstype, NULL); - spin_unlock(&ls->ls_waiters_lock); + spin_unlock_bh(&ls->ls_waiters_lock); return error; } @@ -1569,13 +1599,13 @@ static int remove_from_waiters_ms(struct dlm_lkb *lkb, int error; if (!local) - spin_lock(&ls->ls_waiters_lock); + spin_lock_bh(&ls->ls_waiters_lock); else WARN_ON_ONCE(!rwsem_is_locked(&ls->ls_in_recovery) || !dlm_locking_stopped(ls)); error = _remove_from_waiters(lkb, le32_to_cpu(ms->m_type), ms); if (!local) - spin_unlock(&ls->ls_waiters_lock); + spin_unlock_bh(&ls->ls_waiters_lock); return error; } @@ -1591,10 +1621,10 @@ static void shrink_bucket(struct dlm_ls *ls, int b) memset(&ls->ls_remove_lens, 0, sizeof(int) * DLM_REMOVE_NAMES_MAX); - spin_lock(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl[b].lock); if (!test_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags)) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); return; } @@ -1651,7 +1681,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags); else clear_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); /* * While searching for rsb's to free, we found some that require @@ -1666,16 +1696,16 @@ static void shrink_bucket(struct dlm_ls *ls, int b) name = ls->ls_remove_names[i]; len = ls->ls_remove_lens[i]; - spin_lock(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl[b].lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); if (rv) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); log_debug(ls, "remove_name not toss %s", name); continue; } if (r->res_master_nodeid != our_nodeid) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); log_debug(ls, "remove_name master %d dir %d our %d %s", r->res_master_nodeid, r->res_dir_nodeid, our_nodeid, name); @@ -1684,7 +1714,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) if (r->res_dir_nodeid == our_nodeid) { /* should never happen */ - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); log_error(ls, "remove_name dir %d master %d our %d %s", r->res_dir_nodeid, r->res_master_nodeid, our_nodeid, name); @@ -1693,21 +1723,21 @@ static void shrink_bucket(struct dlm_ls *ls, int b) if (!time_after_eq(jiffies, r->res_toss_time + dlm_config.ci_toss_secs * HZ)) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); log_debug(ls, "remove_name toss_time %lu now %lu %s", r->res_toss_time, jiffies, name); continue; } if (!kref_put(&r->res_ref, kill_rsb)) { - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); log_error(ls, "remove_name in use %s", name); continue; } rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); send_remove(r); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); dlm_free_rsb(r); } @@ -4171,7 +4201,7 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) hash = jhash(name, len, 0); b = hash & (ls->ls_rsbtbl_size - 1); - spin_lock(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl[b].lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); if (rv) { @@ -4181,7 +4211,7 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) /* should not happen */ log_error(ls, "receive_remove from %d not found %s", from_nodeid, name); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); return; } if (r->res_master_nodeid != from_nodeid) { @@ -4189,14 +4219,14 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) log_error(ls, "receive_remove keep from %d master %d", from_nodeid, r->res_master_nodeid); dlm_print_rsb(r); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); return; } log_debug(ls, "receive_remove from %d master %d first %x %s", from_nodeid, r->res_master_nodeid, r->res_first_lkid, name); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); return; } @@ -4204,19 +4234,19 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) log_error(ls, "receive_remove toss from %d master %d", from_nodeid, r->res_master_nodeid); dlm_print_rsb(r); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); return; } if (kref_put(&r->res_ref, kill_rsb)) { rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); dlm_free_rsb(r); } else { log_error(ls, "receive_remove from %d rsb ref error", from_nodeid); dlm_print_rsb(r); - spin_unlock(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl[b].lock); } } @@ -4752,20 +4782,20 @@ static void dlm_receive_message(struct dlm_ls *ls, const struct dlm_message *ms, int nodeid) { try_again: - read_lock(&ls->ls_requestqueue_lock); + read_lock_bh(&ls->ls_requestqueue_lock); if (test_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags)) { /* If we were a member of this lockspace, left, and rejoined, other nodes may still be sending us messages from the lockspace generation before we left. */ if (WARN_ON_ONCE(!ls->ls_generation)) { - read_unlock(&ls->ls_requestqueue_lock); + read_unlock_bh(&ls->ls_requestqueue_lock); log_limit(ls, "receive %d from %d ignore old gen", le32_to_cpu(ms->m_type), nodeid); return; } - read_unlock(&ls->ls_requestqueue_lock); - write_lock(&ls->ls_requestqueue_lock); + read_unlock_bh(&ls->ls_requestqueue_lock); + write_lock_bh(&ls->ls_requestqueue_lock); /* recheck because we hold writelock now */ if (!test_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags)) { write_unlock_bh(&ls->ls_requestqueue_lock); @@ -4773,10 +4803,10 @@ try_again: } dlm_add_requestqueue(ls, nodeid, ms); - write_unlock(&ls->ls_requestqueue_lock); + write_unlock_bh(&ls->ls_requestqueue_lock); } else { _receive_message(ls, ms, 0); - read_unlock(&ls->ls_requestqueue_lock); + read_unlock_bh(&ls->ls_requestqueue_lock); } } @@ -4836,7 +4866,7 @@ void dlm_receive_buffer(const union dlm_packet *p, int nodeid) /* this rwsem allows dlm_ls_stop() to wait for all dlm_recv threads to be inactive (in this ls) before transitioning to recovery mode */ - read_lock(&ls->ls_recv_active); + read_lock_bh(&ls->ls_recv_active); if (hd->h_cmd == DLM_MSG) dlm_receive_message(ls, &p->message, nodeid); else if (hd->h_cmd == DLM_RCOM) @@ -4844,7 +4874,7 @@ void dlm_receive_buffer(const union dlm_packet *p, int nodeid) else log_error(ls, "invalid h_cmd %d from %d lockspace %x", hd->h_cmd, nodeid, le32_to_cpu(hd->u.h_lockspace)); - read_unlock(&ls->ls_recv_active); + read_unlock_bh(&ls->ls_recv_active); dlm_put_lockspace(ls); } @@ -5004,7 +5034,7 @@ static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls) { struct dlm_lkb *lkb = NULL, *iter; - spin_lock(&ls->ls_waiters_lock); + spin_lock_bh(&ls->ls_waiters_lock); list_for_each_entry(iter, &ls->ls_waiters, lkb_wait_reply) { if (test_bit(DLM_IFL_RESEND_BIT, &iter->lkb_iflags)) { hold_lkb(iter); @@ -5012,7 +5042,7 @@ static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls) break; } } - spin_unlock(&ls->ls_waiters_lock); + spin_unlock_bh(&ls->ls_waiters_lock); return lkb; } @@ -5112,9 +5142,9 @@ int dlm_recover_waiters_post(struct dlm_ls *ls) } /* Forcibly remove from waiters list */ - spin_lock(&ls->ls_waiters_lock); + spin_lock_bh(&ls->ls_waiters_lock); list_del_init(&lkb->lkb_wait_reply); - spin_unlock(&ls->ls_waiters_lock); + spin_unlock_bh(&ls->ls_waiters_lock); /* * The lkb is now clear of all prior waiters state and can be @@ -5284,7 +5314,7 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket) struct rb_node *n; struct dlm_rsb *r; - spin_lock(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); for (n = rb_first(&ls->ls_rsbtbl[bucket].keep); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); @@ -5295,10 +5325,10 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket) continue; } hold_rsb(r); - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); return r; } - spin_unlock(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); return NULL; } @@ -5642,10 +5672,10 @@ int dlm_user_request(struct dlm_ls *ls, struct dlm_user_args *ua, } /* add this new lkb to the per-process list of locks */ - spin_lock(&ua->proc->locks_spin); + spin_lock_bh(&ua->proc->locks_spin); hold_lkb(lkb); list_add_tail(&lkb->lkb_ownqueue, &ua->proc->locks); - spin_unlock(&ua->proc->locks_spin); + spin_unlock_bh(&ua->proc->locks_spin); do_put = false; out_put: trace_dlm_lock_end(ls, lkb, name, namelen, mode, flags, error, false); @@ -5775,9 +5805,9 @@ int dlm_user_adopt_orphan(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, * for the proc locks list. */ - spin_lock(&ua->proc->locks_spin); + spin_lock_bh(&ua->proc->locks_spin); list_add_tail(&lkb->lkb_ownqueue, &ua->proc->locks); - spin_unlock(&ua->proc->locks_spin); + spin_unlock_bh(&ua->proc->locks_spin); out: kfree(ua_tmp); return rv; @@ -5821,11 +5851,11 @@ int dlm_user_unlock(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, if (error) goto out_put; - spin_lock(&ua->proc->locks_spin); + spin_lock_bh(&ua->proc->locks_spin); /* dlm_user_add_cb() may have already taken lkb off the proc list */ if (!list_empty(&lkb->lkb_ownqueue)) list_move(&lkb->lkb_ownqueue, &ua->proc->unlocking); - spin_unlock(&ua->proc->locks_spin); + spin_unlock_bh(&ua->proc->locks_spin); out_put: trace_dlm_unlock_end(ls, lkb, flags, error); dlm_put_lkb(lkb); @@ -5976,7 +6006,7 @@ static struct dlm_lkb *del_proc_lock(struct dlm_ls *ls, { struct dlm_lkb *lkb = NULL; - spin_lock(&ls->ls_clear_proc_locks); + spin_lock_bh(&ls->ls_clear_proc_locks); if (list_empty(&proc->locks)) goto out; @@ -5988,7 +6018,7 @@ static struct dlm_lkb *del_proc_lock(struct dlm_ls *ls, else set_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags); out: - spin_unlock(&ls->ls_clear_proc_locks); + spin_unlock_bh(&ls->ls_clear_proc_locks); return lkb; } @@ -6025,7 +6055,7 @@ void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) dlm_put_lkb(lkb); } - spin_lock(&ls->ls_clear_proc_locks); + spin_lock_bh(&ls->ls_clear_proc_locks); /* in-progress unlocks */ list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) { @@ -6039,7 +6069,7 @@ void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) dlm_free_cb(cb); } - spin_unlock(&ls->ls_clear_proc_locks); + spin_unlock_bh(&ls->ls_clear_proc_locks); dlm_unlock_recovery(ls); } @@ -6050,13 +6080,13 @@ static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) while (1) { lkb = NULL; - spin_lock(&proc->locks_spin); + spin_lock_bh(&proc->locks_spin); if (!list_empty(&proc->locks)) { lkb = list_entry(proc->locks.next, struct dlm_lkb, lkb_ownqueue); list_del_init(&lkb->lkb_ownqueue); } - spin_unlock(&proc->locks_spin); + spin_unlock_bh(&proc->locks_spin); if (!lkb) break; @@ -6066,20 +6096,20 @@ static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc) dlm_put_lkb(lkb); /* ref from proc->locks list */ } - spin_lock(&proc->locks_spin); + spin_lock_bh(&proc->locks_spin); list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) { list_del_init(&lkb->lkb_ownqueue); set_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags); dlm_put_lkb(lkb); } - spin_unlock(&proc->locks_spin); + spin_unlock_bh(&proc->locks_spin); - spin_lock(&proc->asts_spin); + spin_lock_bh(&proc->asts_spin); list_for_each_entry_safe(cb, cb_safe, &proc->asts, list) { list_del(&cb->list); dlm_free_cb(cb); } - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); } /* pid of 0 means purge all orphans */ diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index 0f6b2700c0da..45a74869810a 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -69,12 +69,12 @@ static inline int is_master(struct dlm_rsb *r) static inline void lock_rsb(struct dlm_rsb *r) { - spin_lock(&r->res_lock); + spin_lock_bh(&r->res_lock); } static inline void unlock_rsb(struct dlm_rsb *r) { - spin_unlock(&r->res_lock); + spin_unlock_bh(&r->res_lock); } #endif diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index c021bf684fbc..c3681a50decb 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -251,15 +251,15 @@ static struct dlm_ls *find_ls_to_scan(void) { struct dlm_ls *ls; - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); list_for_each_entry(ls, &lslist, ls_list) { if (time_after_eq(jiffies, ls->ls_scan_time + dlm_config.ci_scan_secs * HZ)) { - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); return ls; } } - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); return NULL; } @@ -306,7 +306,7 @@ struct dlm_ls *dlm_find_lockspace_global(uint32_t id) { struct dlm_ls *ls; - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); list_for_each_entry(ls, &lslist, ls_list) { if (ls->ls_global_id == id) { @@ -316,7 +316,7 @@ struct dlm_ls *dlm_find_lockspace_global(uint32_t id) } ls = NULL; out: - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); return ls; } @@ -324,7 +324,7 @@ struct dlm_ls *dlm_find_lockspace_local(dlm_lockspace_t *lockspace) { struct dlm_ls *ls; - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); list_for_each_entry(ls, &lslist, ls_list) { if (ls->ls_local_handle == lockspace) { atomic_inc(&ls->ls_count); @@ -333,7 +333,7 @@ struct dlm_ls *dlm_find_lockspace_local(dlm_lockspace_t *lockspace) } ls = NULL; out: - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); return ls; } @@ -341,7 +341,7 @@ struct dlm_ls *dlm_find_lockspace_device(int minor) { struct dlm_ls *ls; - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); list_for_each_entry(ls, &lslist, ls_list) { if (ls->ls_device.minor == minor) { atomic_inc(&ls->ls_count); @@ -350,7 +350,7 @@ struct dlm_ls *dlm_find_lockspace_device(int minor) } ls = NULL; out: - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); return ls; } @@ -365,15 +365,15 @@ static void remove_lockspace(struct dlm_ls *ls) retry: wait_event(ls->ls_count_wait, atomic_read(&ls->ls_count) == 0); - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); if (atomic_read(&ls->ls_count) != 0) { - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); goto retry; } WARN_ON(ls->ls_create_count != 0); list_del(&ls->ls_list); - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); } static int threads_start(void) @@ -448,7 +448,7 @@ static int new_lockspace(const char *name, const char *cluster, error = 0; - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); list_for_each_entry(ls, &lslist, ls_list) { WARN_ON(ls->ls_create_count <= 0); if (ls->ls_namelen != namelen) @@ -464,7 +464,7 @@ static int new_lockspace(const char *name, const char *cluster, error = 1; break; } - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); if (error) goto out; @@ -583,10 +583,10 @@ static int new_lockspace(const char *name, const char *cluster, INIT_LIST_HEAD(&ls->ls_dir_dump_list); rwlock_init(&ls->ls_dir_dump_lock); - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); ls->ls_create_count = 1; list_add(&ls->ls_list, &lslist); - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); if (flags & DLM_LSFL_FS) { error = dlm_callback_start(ls); @@ -655,9 +655,9 @@ static int new_lockspace(const char *name, const char *cluster, out_callback: dlm_callback_stop(ls); out_delist: - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); list_del(&ls->ls_list); - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); idr_destroy(&ls->ls_recover_idr); kfree(ls->ls_recover_buf); out_lkbidr: @@ -756,7 +756,7 @@ static int lockspace_busy(struct dlm_ls *ls, int force) { int rv; - spin_lock(&ls->ls_lkbidr_spin); + spin_lock_bh(&ls->ls_lkbidr_spin); if (force == 0) { rv = idr_for_each(&ls->ls_lkbidr, lkb_idr_is_any, ls); } else if (force == 1) { @@ -764,7 +764,7 @@ static int lockspace_busy(struct dlm_ls *ls, int force) } else { rv = 0; } - spin_unlock(&ls->ls_lkbidr_spin); + spin_unlock_bh(&ls->ls_lkbidr_spin); return rv; } @@ -776,7 +776,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) busy = lockspace_busy(ls, force); - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); if (ls->ls_create_count == 1) { if (busy) { rv = -EBUSY; @@ -790,7 +790,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) } else { rv = -EINVAL; } - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); if (rv) { log_debug(ls, "release_lockspace no remove %d", rv); @@ -918,20 +918,19 @@ void dlm_stop_lockspaces(void) restart: count = 0; - spin_lock(&lslist_lock); + spin_lock_bh(&lslist_lock); list_for_each_entry(ls, &lslist, ls_list) { if (!test_bit(LSFL_RUNNING, &ls->ls_flags)) { count++; continue; } - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); log_error(ls, "no userland control daemon, stopping lockspace"); dlm_ls_stop(ls); goto restart; } - spin_unlock(&lslist_lock); + spin_unlock_bh(&lslist_lock); if (count) log_print("dlm user daemon left %d lockspaces", count); } - diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index ab2cfbd2ea77..444dc858c4a4 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -867,36 +867,36 @@ static void process_dlm_messages(struct work_struct *work) { struct processqueue_entry *pentry; - spin_lock(&processqueue_lock); + spin_lock_bh(&processqueue_lock); pentry = list_first_entry_or_null(&processqueue, struct processqueue_entry, list); if (WARN_ON_ONCE(!pentry)) { process_dlm_messages_pending = false; - spin_unlock(&processqueue_lock); + spin_unlock_bh(&processqueue_lock); return; } list_del(&pentry->list); atomic_dec(&processqueue_count); - spin_unlock(&processqueue_lock); + spin_unlock_bh(&processqueue_lock); for (;;) { dlm_process_incoming_buffer(pentry->nodeid, pentry->buf, pentry->buflen); free_processqueue_entry(pentry); - spin_lock(&processqueue_lock); + spin_lock_bh(&processqueue_lock); pentry = list_first_entry_or_null(&processqueue, struct processqueue_entry, list); if (!pentry) { process_dlm_messages_pending = false; - spin_unlock(&processqueue_lock); + spin_unlock_bh(&processqueue_lock); break; } list_del(&pentry->list); atomic_dec(&processqueue_count); - spin_unlock(&processqueue_lock); + spin_unlock_bh(&processqueue_lock); } } @@ -966,14 +966,14 @@ again: memmove(con->rx_leftover_buf, pentry->buf + ret, con->rx_leftover); - spin_lock(&processqueue_lock); + spin_lock_bh(&processqueue_lock); ret = atomic_inc_return(&processqueue_count); list_add_tail(&pentry->list, &processqueue); if (!process_dlm_messages_pending) { process_dlm_messages_pending = true; queue_work(process_workqueue, &process_work); } - spin_unlock(&processqueue_lock); + spin_unlock_bh(&processqueue_lock); if (ret > DLM_MAX_PROCESS_BUFFERS) return DLM_IO_FLUSH; diff --git a/fs/dlm/member.c b/fs/dlm/member.c index ac1b555af9d6..6401916a97ef 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -630,7 +630,7 @@ int dlm_ls_stop(struct dlm_ls *ls) * message to the requestqueue without races. */ - write_lock(&ls->ls_recv_active); + write_lock_bh(&ls->ls_recv_active); /* * Abort any recovery that's in progress (see RECOVER_STOP, @@ -638,23 +638,23 @@ int dlm_ls_stop(struct dlm_ls *ls) * dlm to quit any processing (see RUNNING, dlm_locking_stopped()). */ - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); set_bit(LSFL_RECOVER_STOP, &ls->ls_flags); new = test_and_clear_bit(LSFL_RUNNING, &ls->ls_flags); ls->ls_recover_seq++; /* activate requestqueue and stop processing */ - write_lock(&ls->ls_requestqueue_lock); + write_lock_bh(&ls->ls_requestqueue_lock); set_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags); - write_unlock(&ls->ls_requestqueue_lock); - spin_unlock(&ls->ls_recover_lock); + write_unlock_bh(&ls->ls_requestqueue_lock); + spin_unlock_bh(&ls->ls_recover_lock); /* * Let dlm_recv run again, now any normal messages will be saved on the * requestqueue for later. */ - write_unlock(&ls->ls_recv_active); + write_unlock_bh(&ls->ls_recv_active); /* * This in_recovery lock does two things: @@ -679,13 +679,13 @@ int dlm_ls_stop(struct dlm_ls *ls) dlm_recoverd_suspend(ls); - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); kfree(ls->ls_slots); ls->ls_slots = NULL; ls->ls_num_slots = 0; ls->ls_slots_size = 0; ls->ls_recover_status = 0; - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); dlm_recoverd_resume(ls); @@ -719,12 +719,12 @@ int dlm_ls_start(struct dlm_ls *ls) if (error < 0) goto fail_rv; - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); /* the lockspace needs to be stopped before it can be started */ if (!dlm_locking_stopped(ls)) { - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); log_error(ls, "start ignored: lockspace running"); error = -EINVAL; goto fail; @@ -735,7 +735,7 @@ int dlm_ls_start(struct dlm_ls *ls) rv->seq = ++ls->ls_recover_seq; rv_old = ls->ls_recover_args; ls->ls_recover_args = rv; - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); if (rv_old) { log_error(ls, "unused recovery %llx %d", diff --git a/fs/dlm/midcomms.c b/fs/dlm/midcomms.c index ed6fb9b9a582..c34f38e9ee5c 100644 --- a/fs/dlm/midcomms.c +++ b/fs/dlm/midcomms.c @@ -364,9 +364,9 @@ int dlm_midcomms_addr(int nodeid, struct sockaddr_storage *addr, int len) node->users = 0; midcomms_node_reset(node); - spin_lock(&nodes_lock); + spin_lock_bh(&nodes_lock); hlist_add_head_rcu(&node->hlist, &node_hash[r]); - spin_unlock(&nodes_lock); + spin_unlock_bh(&nodes_lock); node->debugfs = dlm_create_debug_comms_file(nodeid, node); return 0; @@ -477,7 +477,7 @@ static void dlm_receive_ack(struct midcomms_node *node, uint32_t seq) static void dlm_pas_fin_ack_rcv(struct midcomms_node *node) { - spin_lock(&node->state_lock); + spin_lock_bh(&node->state_lock); pr_debug("receive passive fin ack from node %d with state %s\n", node->nodeid, dlm_state_str(node->state)); @@ -491,13 +491,13 @@ static void dlm_pas_fin_ack_rcv(struct midcomms_node *node) wake_up(&node->shutdown_wait); break; default: - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); log_print("%s: unexpected state: %d", __func__, node->state); WARN_ON_ONCE(1); return; } - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); } static void dlm_receive_buffer_3_2_trace(uint32_t seq, @@ -534,7 +534,7 @@ static void dlm_midcomms_receive_buffer(const union dlm_packet *p, if (is_expected_seq) { switch (p->header.h_cmd) { case DLM_FIN: - spin_lock(&node->state_lock); + spin_lock_bh(&node->state_lock); pr_debug("receive fin msg from node %d with state %s\n", node->nodeid, dlm_state_str(node->state)); @@ -575,13 +575,13 @@ static void dlm_midcomms_receive_buffer(const union dlm_packet *p, /* probably remove_member caught it, do nothing */ break; default: - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); log_print("%s: unexpected state: %d", __func__, node->state); WARN_ON_ONCE(1); return; } - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); break; default: WARN_ON_ONCE(test_bit(DLM_NODE_FLAG_STOP_RX, &node->flags)); @@ -1182,7 +1182,7 @@ void dlm_midcomms_exit(void) static void dlm_act_fin_ack_rcv(struct midcomms_node *node) { - spin_lock(&node->state_lock); + spin_lock_bh(&node->state_lock); pr_debug("receive active fin ack from node %d with state %s\n", node->nodeid, dlm_state_str(node->state)); @@ -1202,13 +1202,13 @@ static void dlm_act_fin_ack_rcv(struct midcomms_node *node) wake_up(&node->shutdown_wait); break; default: - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); log_print("%s: unexpected state: %d", __func__, node->state); WARN_ON_ONCE(1); return; } - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); } void dlm_midcomms_add_member(int nodeid) @@ -1223,7 +1223,7 @@ void dlm_midcomms_add_member(int nodeid) return; } - spin_lock(&node->state_lock); + spin_lock_bh(&node->state_lock); if (!node->users) { pr_debug("receive add member from node %d with state %s\n", node->nodeid, dlm_state_str(node->state)); @@ -1251,7 +1251,7 @@ void dlm_midcomms_add_member(int nodeid) node->users++; pr_debug("node %d users inc count %d\n", nodeid, node->users); - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); srcu_read_unlock(&nodes_srcu, idx); } @@ -1269,13 +1269,13 @@ void dlm_midcomms_remove_member(int nodeid) return; } - spin_lock(&node->state_lock); + spin_lock_bh(&node->state_lock); /* case of dlm_midcomms_addr() created node but * was not added before because dlm_midcomms_close() * removed the node */ if (!node->users) { - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); srcu_read_unlock(&nodes_srcu, idx); return; } @@ -1313,7 +1313,7 @@ void dlm_midcomms_remove_member(int nodeid) break; } } - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); srcu_read_unlock(&nodes_srcu, idx); } @@ -1351,7 +1351,7 @@ static void midcomms_shutdown(struct midcomms_node *node) return; } - spin_lock(&node->state_lock); + spin_lock_bh(&node->state_lock); pr_debug("receive active shutdown for node %d with state %s\n", node->nodeid, dlm_state_str(node->state)); switch (node->state) { @@ -1370,7 +1370,7 @@ static void midcomms_shutdown(struct midcomms_node *node) */ break; } - spin_unlock(&node->state_lock); + spin_unlock_bh(&node->state_lock); if (DLM_DEBUG_FENCE_TERMINATION) msleep(5000); @@ -1441,9 +1441,9 @@ int dlm_midcomms_close(int nodeid) ret = dlm_lowcomms_close(nodeid); dlm_delete_debug_comms_file(node->debugfs); - spin_lock(&nodes_lock); + spin_lock_bh(&nodes_lock); hlist_del_rcu(&node->hlist); - spin_unlock(&nodes_lock); + spin_unlock_bh(&nodes_lock); srcu_read_unlock(&nodes_srcu, idx); /* wait that all readers left until flush send queue */ diff --git a/fs/dlm/rcom.c b/fs/dlm/rcom.c index 2e3f529f3ff2..be1a71a6303a 100644 --- a/fs/dlm/rcom.c +++ b/fs/dlm/rcom.c @@ -143,18 +143,18 @@ static int check_rcom_config(struct dlm_ls *ls, struct dlm_rcom *rc, int nodeid) static void allow_sync_reply(struct dlm_ls *ls, __le64 *new_seq) { - spin_lock(&ls->ls_rcom_spin); + spin_lock_bh(&ls->ls_rcom_spin); *new_seq = cpu_to_le64(++ls->ls_rcom_seq); set_bit(LSFL_RCOM_WAIT, &ls->ls_flags); - spin_unlock(&ls->ls_rcom_spin); + spin_unlock_bh(&ls->ls_rcom_spin); } static void disallow_sync_reply(struct dlm_ls *ls) { - spin_lock(&ls->ls_rcom_spin); + spin_lock_bh(&ls->ls_rcom_spin); clear_bit(LSFL_RCOM_WAIT, &ls->ls_flags); clear_bit(LSFL_RCOM_READY, &ls->ls_flags); - spin_unlock(&ls->ls_rcom_spin); + spin_unlock_bh(&ls->ls_rcom_spin); } /* @@ -245,10 +245,10 @@ static void receive_rcom_status(struct dlm_ls *ls, goto do_create; } - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); status = ls->ls_recover_status; num_slots = ls->ls_num_slots; - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); len += num_slots * sizeof(struct rcom_slot); do_create: @@ -266,9 +266,9 @@ static void receive_rcom_status(struct dlm_ls *ls, if (!num_slots) goto do_send; - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); if (ls->ls_num_slots != num_slots) { - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); log_debug(ls, "receive_rcom_status num_slots %d to %d", num_slots, ls->ls_num_slots); rc->rc_result = 0; @@ -277,7 +277,7 @@ static void receive_rcom_status(struct dlm_ls *ls, } dlm_slots_copy_out(ls, rc); - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); do_send: send_rcom_stateless(msg, rc); @@ -285,7 +285,7 @@ static void receive_rcom_status(struct dlm_ls *ls, static void receive_sync_reply(struct dlm_ls *ls, const struct dlm_rcom *rc_in) { - spin_lock(&ls->ls_rcom_spin); + spin_lock_bh(&ls->ls_rcom_spin); if (!test_bit(LSFL_RCOM_WAIT, &ls->ls_flags) || le64_to_cpu(rc_in->rc_id) != ls->ls_rcom_seq) { log_debug(ls, "reject reply %d from %d seq %llx expect %llx", @@ -301,7 +301,7 @@ static void receive_sync_reply(struct dlm_ls *ls, const struct dlm_rcom *rc_in) clear_bit(LSFL_RCOM_WAIT, &ls->ls_flags); wake_up(&ls->ls_wait_general); out: - spin_unlock(&ls->ls_rcom_spin); + spin_unlock_bh(&ls->ls_rcom_spin); } int dlm_rcom_names(struct dlm_ls *ls, int nodeid, char *last_name, @@ -613,11 +613,11 @@ void dlm_receive_rcom(struct dlm_ls *ls, const struct dlm_rcom *rc, int nodeid) break; } - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); status = ls->ls_recover_status; stop = dlm_recovery_stopped(ls); seq = ls->ls_recover_seq; - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); if (stop && (rc->rc_type != cpu_to_le32(DLM_RCOM_STATUS))) goto ignore; diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 172c6b73f37a..13bc845fa305 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -74,9 +74,9 @@ int dlm_wait_function(struct dlm_ls *ls, int (*testfn) (struct dlm_ls *ls)) uint32_t dlm_recover_status(struct dlm_ls *ls) { uint32_t status; - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); status = ls->ls_recover_status; - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); return status; } @@ -87,9 +87,9 @@ static void _set_recover_status(struct dlm_ls *ls, uint32_t status) void dlm_set_recover_status(struct dlm_ls *ls, uint32_t status) { - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); _set_recover_status(ls, status); - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); } static int wait_status_all(struct dlm_ls *ls, uint32_t wait_status, @@ -188,13 +188,13 @@ int dlm_recover_members_wait(struct dlm_ls *ls, uint64_t seq) rv = dlm_slots_assign(ls, &num_slots, &slots_size, &slots, &gen); if (!rv) { - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); _set_recover_status(ls, DLM_RS_NODES_ALL); ls->ls_num_slots = num_slots; ls->ls_slots_size = slots_size; ls->ls_slots = slots; ls->ls_generation = gen; - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); } else { dlm_set_recover_status(ls, DLM_RS_NODES_ALL); } @@ -241,9 +241,9 @@ static int recover_list_empty(struct dlm_ls *ls) { int empty; - spin_lock(&ls->ls_recover_list_lock); + spin_lock_bh(&ls->ls_recover_list_lock); empty = list_empty(&ls->ls_recover_list); - spin_unlock(&ls->ls_recover_list_lock); + spin_unlock_bh(&ls->ls_recover_list_lock); return empty; } @@ -252,23 +252,23 @@ static void recover_list_add(struct dlm_rsb *r) { struct dlm_ls *ls = r->res_ls; - spin_lock(&ls->ls_recover_list_lock); + spin_lock_bh(&ls->ls_recover_list_lock); if (list_empty(&r->res_recover_list)) { list_add_tail(&r->res_recover_list, &ls->ls_recover_list); ls->ls_recover_list_count++; dlm_hold_rsb(r); } - spin_unlock(&ls->ls_recover_list_lock); + spin_unlock_bh(&ls->ls_recover_list_lock); } static void recover_list_del(struct dlm_rsb *r) { struct dlm_ls *ls = r->res_ls; - spin_lock(&ls->ls_recover_list_lock); + spin_lock_bh(&ls->ls_recover_list_lock); list_del_init(&r->res_recover_list); ls->ls_recover_list_count--; - spin_unlock(&ls->ls_recover_list_lock); + spin_unlock_bh(&ls->ls_recover_list_lock); dlm_put_rsb(r); } @@ -277,7 +277,7 @@ static void recover_list_clear(struct dlm_ls *ls) { struct dlm_rsb *r, *s; - spin_lock(&ls->ls_recover_list_lock); + spin_lock_bh(&ls->ls_recover_list_lock); list_for_each_entry_safe(r, s, &ls->ls_recover_list, res_recover_list) { list_del_init(&r->res_recover_list); r->res_recover_locks_count = 0; @@ -290,17 +290,17 @@ static void recover_list_clear(struct dlm_ls *ls) ls->ls_recover_list_count); ls->ls_recover_list_count = 0; } - spin_unlock(&ls->ls_recover_list_lock); + spin_unlock_bh(&ls->ls_recover_list_lock); } static int recover_idr_empty(struct dlm_ls *ls) { int empty = 1; - spin_lock(&ls->ls_recover_idr_lock); + spin_lock_bh(&ls->ls_recover_idr_lock); if (ls->ls_recover_list_count) empty = 0; - spin_unlock(&ls->ls_recover_idr_lock); + spin_unlock_bh(&ls->ls_recover_idr_lock); return empty; } @@ -310,7 +310,7 @@ static int recover_idr_add(struct dlm_rsb *r) struct dlm_ls *ls = r->res_ls; int rv; - spin_lock(&ls->ls_recover_idr_lock); + spin_lock_bh(&ls->ls_recover_idr_lock); if (r->res_id) { rv = -1; goto out_unlock; @@ -324,7 +324,7 @@ static int recover_idr_add(struct dlm_rsb *r) dlm_hold_rsb(r); rv = 0; out_unlock: - spin_unlock(&ls->ls_recover_idr_lock); + spin_unlock_bh(&ls->ls_recover_idr_lock); return rv; } @@ -332,11 +332,11 @@ static void recover_idr_del(struct dlm_rsb *r) { struct dlm_ls *ls = r->res_ls; - spin_lock(&ls->ls_recover_idr_lock); + spin_lock_bh(&ls->ls_recover_idr_lock); idr_remove(&ls->ls_recover_idr, r->res_id); r->res_id = 0; ls->ls_recover_list_count--; - spin_unlock(&ls->ls_recover_idr_lock); + spin_unlock_bh(&ls->ls_recover_idr_lock); dlm_put_rsb(r); } @@ -345,9 +345,9 @@ static struct dlm_rsb *recover_idr_find(struct dlm_ls *ls, uint64_t id) { struct dlm_rsb *r; - spin_lock(&ls->ls_recover_idr_lock); + spin_lock_bh(&ls->ls_recover_idr_lock); r = idr_find(&ls->ls_recover_idr, (int)id); - spin_unlock(&ls->ls_recover_idr_lock); + spin_unlock_bh(&ls->ls_recover_idr_lock); return r; } @@ -356,7 +356,7 @@ static void recover_idr_clear(struct dlm_ls *ls) struct dlm_rsb *r; int id; - spin_lock(&ls->ls_recover_idr_lock); + spin_lock_bh(&ls->ls_recover_idr_lock); idr_for_each_entry(&ls->ls_recover_idr, r, id) { idr_remove(&ls->ls_recover_idr, id); @@ -372,7 +372,7 @@ static void recover_idr_clear(struct dlm_ls *ls) ls->ls_recover_list_count); ls->ls_recover_list_count = 0; } - spin_unlock(&ls->ls_recover_idr_lock); + spin_unlock_bh(&ls->ls_recover_idr_lock); } @@ -887,7 +887,7 @@ void dlm_clear_toss(struct dlm_ls *ls) int i; for (i = 0; i < ls->ls_rsbtbl_size; i++) { - spin_lock(&ls->ls_rsbtbl[i].lock); + spin_lock_bh(&ls->ls_rsbtbl[i].lock); for (n = rb_first(&ls->ls_rsbtbl[i].toss); n; n = next) { next = rb_next(n); r = rb_entry(n, struct dlm_rsb, res_hashnode); @@ -895,7 +895,7 @@ void dlm_clear_toss(struct dlm_ls *ls) dlm_free_rsb(r); count++; } - spin_unlock(&ls->ls_rsbtbl[i].lock); + spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } if (count) diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index a11ae1da2f60..c82cc48988c6 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -26,7 +26,7 @@ static int dlm_create_masters_list(struct dlm_ls *ls) struct dlm_rsb *r; int i, error = 0; - write_lock(&ls->ls_masters_lock); + write_lock_bh(&ls->ls_masters_lock); if (!list_empty(&ls->ls_masters_list)) { log_error(ls, "root list not empty"); error = -EINVAL; @@ -46,7 +46,7 @@ static int dlm_create_masters_list(struct dlm_ls *ls) spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } out: - write_unlock(&ls->ls_masters_lock); + write_unlock_bh(&ls->ls_masters_lock); return error; } @@ -54,12 +54,12 @@ static void dlm_release_masters_list(struct dlm_ls *ls) { struct dlm_rsb *r, *safe; - write_lock(&ls->ls_masters_lock); + write_lock_bh(&ls->ls_masters_lock); list_for_each_entry_safe(r, safe, &ls->ls_masters_list, res_masters_list) { list_del_init(&r->res_masters_list); dlm_put_rsb(r); } - write_unlock(&ls->ls_masters_lock); + write_unlock_bh(&ls->ls_masters_lock); } static void dlm_create_root_list(struct dlm_ls *ls, struct list_head *root_list) @@ -103,9 +103,9 @@ static int enable_locking(struct dlm_ls *ls, uint64_t seq) { int error = -EINTR; - write_lock(&ls->ls_recv_active); + write_lock_bh(&ls->ls_recv_active); - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); if (ls->ls_recover_seq == seq) { set_bit(LSFL_RUNNING, &ls->ls_flags); /* unblocks processes waiting to enter the dlm */ @@ -113,9 +113,9 @@ static int enable_locking(struct dlm_ls *ls, uint64_t seq) clear_bit(LSFL_RECOVER_LOCK, &ls->ls_flags); error = 0; } - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); - write_unlock(&ls->ls_recv_active); + write_unlock_bh(&ls->ls_recv_active); return error; } @@ -349,12 +349,12 @@ static void do_ls_recovery(struct dlm_ls *ls) struct dlm_recover *rv = NULL; int error; - spin_lock(&ls->ls_recover_lock); + spin_lock_bh(&ls->ls_recover_lock); rv = ls->ls_recover_args; ls->ls_recover_args = NULL; if (rv && ls->ls_recover_seq == rv->seq) clear_bit(LSFL_RECOVER_STOP, &ls->ls_flags); - spin_unlock(&ls->ls_recover_lock); + spin_unlock_bh(&ls->ls_recover_lock); if (rv) { error = ls_recover(ls, rv); diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c index 9b646026df46..719a5243a069 100644 --- a/fs/dlm/requestqueue.c +++ b/fs/dlm/requestqueue.c @@ -68,7 +68,7 @@ int dlm_process_requestqueue(struct dlm_ls *ls) struct dlm_message *ms; int error = 0; - write_lock(&ls->ls_requestqueue_lock); + write_lock_bh(&ls->ls_requestqueue_lock); for (;;) { if (list_empty(&ls->ls_requestqueue)) { clear_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags); @@ -96,11 +96,11 @@ int dlm_process_requestqueue(struct dlm_ls *ls) error = -EINTR; break; } - write_unlock(&ls->ls_requestqueue_lock); + write_unlock_bh(&ls->ls_requestqueue_lock); schedule(); - write_lock(&ls->ls_requestqueue_lock); + write_lock_bh(&ls->ls_requestqueue_lock); } - write_unlock(&ls->ls_requestqueue_lock); + write_unlock_bh(&ls->ls_requestqueue_lock); return error; } @@ -135,7 +135,7 @@ void dlm_purge_requestqueue(struct dlm_ls *ls) struct dlm_message *ms; struct rq_entry *e, *safe; - write_lock(&ls->ls_requestqueue_lock); + write_lock_bh(&ls->ls_requestqueue_lock); list_for_each_entry_safe(e, safe, &ls->ls_requestqueue, list) { ms = &e->request; @@ -144,6 +144,6 @@ void dlm_purge_requestqueue(struct dlm_ls *ls) kfree(e); } } - write_unlock(&ls->ls_requestqueue_lock); + write_unlock_bh(&ls->ls_requestqueue_lock); } diff --git a/fs/dlm/user.c b/fs/dlm/user.c index b4971ba4bdd6..3173b974e8c8 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -189,7 +189,7 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode, return; ls = lkb->lkb_resource->res_ls; - spin_lock(&ls->ls_clear_proc_locks); + spin_lock_bh(&ls->ls_clear_proc_locks); /* If ORPHAN/DEAD flag is set, it means the process is dead so an ast can't be delivered. For ORPHAN's, dlm_clear_proc_locks() freed @@ -211,7 +211,7 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode, if ((flags & DLM_CB_CAST) && lkb_is_endoflife(mode, status)) set_bit(DLM_IFL_ENDOFLIFE_BIT, &lkb->lkb_iflags); - spin_lock(&proc->asts_spin); + spin_lock_bh(&proc->asts_spin); rv = dlm_queue_lkb_callback(lkb, flags, mode, status, sbflags, &cb); switch (rv) { @@ -232,23 +232,23 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode, case DLM_ENQUEUE_CALLBACK_FAILURE: fallthrough; default: - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); WARN_ON_ONCE(1); goto out; } - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); if (test_bit(DLM_IFL_ENDOFLIFE_BIT, &lkb->lkb_iflags)) { /* N.B. spin_lock locks_spin, not asts_spin */ - spin_lock(&proc->locks_spin); + spin_lock_bh(&proc->locks_spin); if (!list_empty(&lkb->lkb_ownqueue)) { list_del_init(&lkb->lkb_ownqueue); dlm_put_lkb(lkb); } - spin_unlock(&proc->locks_spin); + spin_unlock_bh(&proc->locks_spin); } out: - spin_unlock(&ls->ls_clear_proc_locks); + spin_unlock_bh(&ls->ls_clear_proc_locks); } static int device_user_lock(struct dlm_user_proc *proc, @@ -817,10 +817,10 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, if (test_bit(DLM_PROC_FLAGS_CLOSING, &proc->flags)) return -EINVAL; - spin_lock(&proc->asts_spin); + spin_lock_bh(&proc->asts_spin); if (list_empty(&proc->asts)) { if (file->f_flags & O_NONBLOCK) { - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); return -EAGAIN; } @@ -829,16 +829,16 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, repeat: set_current_state(TASK_INTERRUPTIBLE); if (list_empty(&proc->asts) && !signal_pending(current)) { - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); schedule(); - spin_lock(&proc->asts_spin); + spin_lock_bh(&proc->asts_spin); goto repeat; } set_current_state(TASK_RUNNING); remove_wait_queue(&proc->wait, &wait); if (signal_pending(current)) { - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); return -ERESTARTSYS; } } @@ -849,7 +849,7 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count, cb = list_first_entry(&proc->asts, struct dlm_callback, list); list_del(&cb->list); - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); if (cb->flags & DLM_CB_BAST) { trace_dlm_bast(cb->ls_id, cb->lkb_id, cb->mode, cb->res_name, @@ -874,12 +874,12 @@ static __poll_t device_poll(struct file *file, poll_table *wait) poll_wait(file, &proc->wait, wait); - spin_lock(&proc->asts_spin); + spin_lock_bh(&proc->asts_spin); if (!list_empty(&proc->asts)) { - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); return EPOLLIN | EPOLLRDNORM; } - spin_unlock(&proc->asts_spin); + spin_unlock_bh(&proc->asts_spin); return 0; } -- cgit v1.2.3-59-g8ed1b From 92d59adfaf710f34ae7788fa54f0731a7640833b Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 2 Apr 2024 15:18:10 -0400 Subject: dlm: do message processing in softirq context Move dlm message processing from an ordered workqueue context to an ordered softirq context. Handling dlm messages in softirq will allow requests to be cleared more quickly and efficiently, and should avoid longer queues of incomplete requests. Later patches are expected to run completion/blocking callbacks directly from this message processing context, further reducing context switches required to complete a request. In the longer term, concurrent message processing could be implemented. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lowcomms.c | 28 ++++++++++++++++++++-------- 1 file changed, 20 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 444dc858c4a4..6b8078085e56 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -204,6 +204,7 @@ static void process_dlm_messages(struct work_struct *work); static DECLARE_WORK(process_work, process_dlm_messages); static DEFINE_SPINLOCK(processqueue_lock); static bool process_dlm_messages_pending; +static DECLARE_WAIT_QUEUE_HEAD(processqueue_wq); static atomic_t processqueue_count; static LIST_HEAD(processqueue); @@ -877,7 +878,8 @@ static void process_dlm_messages(struct work_struct *work) } list_del(&pentry->list); - atomic_dec(&processqueue_count); + if (atomic_dec_and_test(&processqueue_count)) + wake_up(&processqueue_wq); spin_unlock_bh(&processqueue_lock); for (;;) { @@ -895,7 +897,8 @@ static void process_dlm_messages(struct work_struct *work) } list_del(&pentry->list); - atomic_dec(&processqueue_count); + if (atomic_dec_and_test(&processqueue_count)) + wake_up(&processqueue_wq); spin_unlock_bh(&processqueue_lock); } } @@ -1511,7 +1514,20 @@ static void process_recv_sockets(struct work_struct *work) /* CF_RECV_PENDING cleared */ break; case DLM_IO_FLUSH: - flush_workqueue(process_workqueue); + /* we can't flush the process_workqueue here because a + * WQ_MEM_RECLAIM workequeue can occurr a deadlock for a non + * WQ_MEM_RECLAIM workqueue such as process_workqueue. Instead + * we have a waitqueue to wait until all messages are + * processed. + * + * This handling is only necessary to backoff the sender and + * not queue all messages from the socket layer into DLM + * processqueue. When DLM is capable to parse multiple messages + * on an e.g. per socket basis this handling can might be + * removed. Especially in a message burst we are too slow to + * process messages and the queue will fill up memory. + */ + wait_event(processqueue_wq, !atomic_read(&processqueue_count)); fallthrough; case DLM_IO_RESCHED: cond_resched(); @@ -1701,11 +1717,7 @@ static int work_start(void) return -ENOMEM; } - /* ordered dlm message process queue, - * should be converted to a tasklet - */ - process_workqueue = alloc_ordered_workqueue("dlm_process", - WQ_HIGHPRI | WQ_MEM_RECLAIM); + process_workqueue = alloc_workqueue("dlm_process", WQ_HIGHPRI | WQ_BH, 0); if (!process_workqueue) { log_print("can't start dlm_process"); destroy_workqueue(io_workqueue); -- cgit v1.2.3-59-g8ed1b From 3253804773c0613a1bad5bfea2edf172b760d8b6 Mon Sep 17 00:00:00 2001 From: Stefan Berger Date: Fri, 23 Feb 2024 12:25:05 -0500 Subject: security: allow finer granularity in permitting copy-up of security xattrs Copying up xattrs is solely based on the security xattr name. For finer granularity add a dentry parameter to the security_inode_copy_up_xattr hook definition, allowing decisions to be based on the xattr content as well. Co-developed-by: Mimi Zohar Signed-off-by: Stefan Berger Acked-by: Amir Goldstein Acked-by: Paul Moore (LSM,SELinux) Signed-off-by: Mimi Zohar --- fs/overlayfs/copy_up.c | 2 +- include/linux/lsm_hook_defs.h | 3 ++- include/linux/security.h | 4 ++-- security/integrity/evm/evm_main.c | 2 +- security/security.c | 5 +++-- security/selinux/hooks.c | 2 +- security/smack/smack_lsm.c | 2 +- 7 files changed, 11 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c index 0762575a1e70..a5ef2005a2cc 100644 --- a/fs/overlayfs/copy_up.c +++ b/fs/overlayfs/copy_up.c @@ -114,7 +114,7 @@ int ovl_copy_xattr(struct super_block *sb, const struct path *oldpath, struct de if (ovl_is_private_xattr(sb, name)) continue; - error = security_inode_copy_up_xattr(name); + error = security_inode_copy_up_xattr(old, name); if (error < 0 && error != -EOPNOTSUPP) break; if (error == 1) { diff --git a/include/linux/lsm_hook_defs.h b/include/linux/lsm_hook_defs.h index 334e00efbde4..f804b76cde44 100644 --- a/include/linux/lsm_hook_defs.h +++ b/include/linux/lsm_hook_defs.h @@ -176,7 +176,8 @@ LSM_HOOK(int, 0, inode_listsecurity, struct inode *inode, char *buffer, size_t buffer_size) LSM_HOOK(void, LSM_RET_VOID, inode_getsecid, struct inode *inode, u32 *secid) LSM_HOOK(int, 0, inode_copy_up, struct dentry *src, struct cred **new) -LSM_HOOK(int, -EOPNOTSUPP, inode_copy_up_xattr, const char *name) +LSM_HOOK(int, -EOPNOTSUPP, inode_copy_up_xattr, struct dentry *src, + const char *name) LSM_HOOK(int, 0, kernfs_init_security, struct kernfs_node *kn_dir, struct kernfs_node *kn) LSM_HOOK(int, 0, file_permission, struct file *file, int mask) diff --git a/include/linux/security.h b/include/linux/security.h index 41a8f667bdfa..21cf70346b33 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -398,7 +398,7 @@ int security_inode_setsecurity(struct inode *inode, const char *name, const void int security_inode_listsecurity(struct inode *inode, char *buffer, size_t buffer_size); void security_inode_getsecid(struct inode *inode, u32 *secid); int security_inode_copy_up(struct dentry *src, struct cred **new); -int security_inode_copy_up_xattr(const char *name); +int security_inode_copy_up_xattr(struct dentry *src, const char *name); int security_kernfs_init_security(struct kernfs_node *kn_dir, struct kernfs_node *kn); int security_file_permission(struct file *file, int mask); @@ -1016,7 +1016,7 @@ static inline int security_kernfs_init_security(struct kernfs_node *kn_dir, return 0; } -static inline int security_inode_copy_up_xattr(const char *name) +static inline int security_inode_copy_up_xattr(struct dentry *src, const char *name) { return -EOPNOTSUPP; } diff --git a/security/integrity/evm/evm_main.c b/security/integrity/evm/evm_main.c index 81dbade5b9b3..b0a862bfd74a 100644 --- a/security/integrity/evm/evm_main.c +++ b/security/integrity/evm/evm_main.c @@ -946,7 +946,7 @@ static void evm_inode_post_setattr(struct mnt_idmap *idmap, evm_update_evmxattr(dentry, NULL, NULL, 0); } -static int evm_inode_copy_up_xattr(const char *name) +static int evm_inode_copy_up_xattr(struct dentry *src, const char *name) { if (strcmp(name, XATTR_NAME_EVM) == 0) return 1; /* Discard */ diff --git a/security/security.c b/security/security.c index 0a9a0ac3f266..e5da848c50b9 100644 --- a/security/security.c +++ b/security/security.c @@ -2628,6 +2628,7 @@ EXPORT_SYMBOL(security_inode_copy_up); /** * security_inode_copy_up_xattr() - Filter xattrs in an overlayfs copy-up op + * @src: union dentry of copy-up file * @name: xattr name * * Filter the xattrs being copied up when a unioned file is copied up from a @@ -2638,7 +2639,7 @@ EXPORT_SYMBOL(security_inode_copy_up); * if the security module does not know about attribute, or a negative * error code to abort the copy up. */ -int security_inode_copy_up_xattr(const char *name) +int security_inode_copy_up_xattr(struct dentry *src, const char *name) { int rc; @@ -2647,7 +2648,7 @@ int security_inode_copy_up_xattr(const char *name) * xattr), -EOPNOTSUPP if it does not know anything about the xattr or * any other error code in case of an error. */ - rc = call_int_hook(inode_copy_up_xattr, name); + rc = call_int_hook(inode_copy_up_xattr, src, name); if (rc != LSM_RET_DEFAULT(inode_copy_up_xattr)) return rc; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 3448454c82d0..9f018f51e47a 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3530,7 +3530,7 @@ static int selinux_inode_copy_up(struct dentry *src, struct cred **new) return 0; } -static int selinux_inode_copy_up_xattr(const char *name) +static int selinux_inode_copy_up_xattr(struct dentry *dentry, const char *name) { /* The copy_up hook above sets the initial context on an inode, but we * don't then want to overwrite it by blindly copying all the lower diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 146667937811..79d8dc633b9a 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -4885,7 +4885,7 @@ static int smack_inode_copy_up(struct dentry *dentry, struct cred **new) return 0; } -static int smack_inode_copy_up_xattr(const char *name) +static int smack_inode_copy_up_xattr(struct dentry *src, const char *name) { /* * Return 1 if this is the smack access Smack attribute. -- cgit v1.2.3-59-g8ed1b From 1f65e57dc5417b166843438bef31c70b9a5208fe Mon Sep 17 00:00:00 2001 From: Stefan Berger Date: Fri, 23 Feb 2024 12:25:12 -0500 Subject: fs: Rename SB_I_EVM_UNSUPPORTED to SB_I_EVM_HMAC_UNSUPPORTED Now that EVM supports RSA signatures for previously completely unsupported filesystems rename the flag SB_I_EVM_UNSUPPORTED to SB_I_EVM_HMAC_UNSUPPORTED to reflect that only HMAC is not supported. Suggested-by: Amir Goldstein Suggested-by: Mimi Zohar Signed-off-by: Stefan Berger Acked-by: Amir Goldstein Signed-off-by: Mimi Zohar --- fs/overlayfs/super.c | 2 +- include/linux/fs.h | 2 +- security/integrity/evm/evm_main.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index a40fc7e05525..06a231970cb5 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -1460,7 +1460,7 @@ int ovl_fill_super(struct super_block *sb, struct fs_context *fc) * lead to unexpected results. */ sb->s_iflags |= SB_I_NOUMASK; - sb->s_iflags |= SB_I_EVM_UNSUPPORTED; + sb->s_iflags |= SB_I_EVM_HMAC_UNSUPPORTED; err = -ENOMEM; root_dentry = ovl_get_root(sb, ctx->upper.dentry, oe); diff --git a/include/linux/fs.h b/include/linux/fs.h index 8dfd53b52744..7bbd5e2d2a20 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1177,7 +1177,7 @@ extern int send_sigurg(struct fown_struct *fown); #define SB_I_USERNS_VISIBLE 0x00000010 /* fstype already mounted */ #define SB_I_IMA_UNVERIFIABLE_SIGNATURE 0x00000020 #define SB_I_UNTRUSTED_MOUNTER 0x00000040 -#define SB_I_EVM_UNSUPPORTED 0x00000080 +#define SB_I_EVM_HMAC_UNSUPPORTED 0x00000080 #define SB_I_SKIP_SYNC 0x00000100 /* Skip superblock at global sync */ #define SB_I_PERSB_BDI 0x00000200 /* has a per-sb bdi */ diff --git a/security/integrity/evm/evm_main.c b/security/integrity/evm/evm_main.c index cfb4f9809369..c4a6081ce596 100644 --- a/security/integrity/evm/evm_main.c +++ b/security/integrity/evm/evm_main.c @@ -155,7 +155,7 @@ static int is_unsupported_fs(struct dentry *dentry) { struct inode *inode = d_backing_inode(dentry); - if (inode->i_sb->s_iflags & SB_I_EVM_UNSUPPORTED) { + if (inode->i_sb->s_iflags & SB_I_EVM_HMAC_UNSUPPORTED) { pr_info_once("%s not supported\n", inode->i_sb->s_type->name); return 1; } -- cgit v1.2.3-59-g8ed1b From 68879386180c0efd5a11e800b0525a01068c9457 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Tue, 26 Mar 2024 14:39:20 +0900 Subject: btrfs: zoned: do not flag ZEROOUT on non-dirty extent buffer Btrfs clears the content of an extent buffer marked as EXTENT_BUFFER_ZONED_ZEROOUT before the bio submission. This mechanism is introduced to prevent a write hole of an extent buffer, which is once allocated, marked dirty, but turns out unnecessary and cleaned up within one transaction operation. Currently, btrfs_clear_buffer_dirty() marks the extent buffer as EXTENT_BUFFER_ZONED_ZEROOUT, and skips the entry function. If this call happens while the buffer is under IO (with the WRITEBACK flag set, without the DIRTY flag), we can add the ZEROOUT flag and clear the buffer's content just before a bio submission. As a result: 1) it can lead to adding faulty delayed reference item which leads to a FS corrupted (EUCLEAN) error, and 2) it writes out cleared tree node on disk The former issue is previously discussed in [1]. The corruption happens when it runs a delayed reference update. So, on-disk data is safe. [1] https://lore.kernel.org/linux-btrfs/3f4f2a0ff1a6c818050434288925bdcf3cd719e5.1709124777.git.naohiro.aota@wdc.com/ The latter one can reach on-disk data. But, as that node is already processed by btrfs_clear_buffer_dirty(), that will be invalidated in the next transaction commit anyway. So, the chance of hitting the corruption is relatively small. Anyway, we should skip flagging ZEROOUT on a non-DIRTY extent buffer, to keep the content under IO intact. Fixes: aa6313e6ff2b ("btrfs: zoned: don't clear dirty flag of extent buffer") CC: stable@vger.kernel.org # 6.8 Link: https://lore.kernel.org/linux-btrfs/oadvdekkturysgfgi4qzuemd57zudeasynswurjxw3ocdfsef6@sjyufeugh63f/ Reviewed-by: Johannes Thumshirn Signed-off-by: Naohiro Aota Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 61594eaf1f89..df3fe36126f9 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4154,7 +4154,7 @@ void btrfs_clear_buffer_dirty(struct btrfs_trans_handle *trans, * The actual zeroout of the buffer will happen later in * btree_csum_one_bio. */ - if (btrfs_is_zoned(fs_info)) { + if (btrfs_is_zoned(fs_info) && test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) { set_bit(EXTENT_BUFFER_ZONED_ZEROOUT, &eb->bflags); return; } -- cgit v1.2.3-59-g8ed1b From 073bda7a541731f41ed08f32d286394236c74005 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Tue, 26 Mar 2024 14:39:21 +0900 Subject: btrfs: zoned: add ASSERT and WARN for EXTENT_BUFFER_ZONED_ZEROOUT handling Add an ASSERT to catch a faulty delayed reference item resulting from prematurely cleared extent buffer. Also, add a WARN to detect if we try to dirty a ZEROOUT buffer again, which is suspicious as its update will be lost. Reviewed-by: Johannes Thumshirn Signed-off-by: Naohiro Aota Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 8 ++++++++ fs/btrfs/extent_io.c | 1 + 2 files changed, 9 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index beedd6ed64d3..257d044bca91 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3464,6 +3464,14 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, if (root_id != BTRFS_TREE_LOG_OBJECTID) { struct btrfs_ref generic_ref = { 0 }; + /* + * Assert that the extent buffer is not cleared due to + * EXTENT_BUFFER_ZONED_ZEROOUT. Please refer + * btrfs_clear_buffer_dirty() and btree_csum_one_bio() for + * detail. + */ + ASSERT(btrfs_header_bytenr(buf) != 0); + btrfs_init_generic_ref(&generic_ref, BTRFS_DROP_DELAYED_REF, buf->start, buf->len, parent, btrfs_header_owner(buf)); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index df3fe36126f9..b18034f2ab80 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4193,6 +4193,7 @@ void set_extent_buffer_dirty(struct extent_buffer *eb) num_folios = num_extent_folios(eb); WARN_ON(atomic_read(&eb->refs) == 0); WARN_ON(!test_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags)); + WARN_ON(test_bit(EXTENT_BUFFER_ZONED_ZEROOUT, &eb->bflags)); if (!was_dirty) { bool subpage = eb->fs_info->nodesize < PAGE_SIZE; -- cgit v1.2.3-59-g8ed1b From 1db7959aacd905e6487d0478ac01d89f86eb1e51 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Tue, 26 Mar 2024 09:16:46 +1030 Subject: btrfs: do not wait for short bulk allocation [BUG] There is a recent report that when memory pressure is high (including cached pages), btrfs can spend most of its time on memory allocation in btrfs_alloc_page_array() for compressed read/write. [CAUSE] For btrfs_alloc_page_array() we always go alloc_pages_bulk_array(), and even if the bulk allocation failed (fell back to single page allocation) we still retry but with extra memalloc_retry_wait(). If the bulk alloc only returned one page a time, we would spend a lot of time on the retry wait. The behavior was introduced in commit 395cb57e8560 ("btrfs: wait between incomplete batch memory allocations"). [FIX] Although the commit mentioned that other filesystems do the wait, it's not the case at least nowadays. All the mainlined filesystems only call memalloc_retry_wait() if they failed to allocate any page (not only for bulk allocation). If there is any progress, they won't call memalloc_retry_wait() at all. For example, xfs_buf_alloc_pages() would only call memalloc_retry_wait() if there is no allocation progress at all, and the call is not for metadata readahead. So I don't believe we should call memalloc_retry_wait() unconditionally for short allocation. Call memalloc_retry_wait() if it fails to allocate any page for tree block allocation (which goes with __GFP_NOFAIL and may not need the special handling anyway), and reduce the latency for btrfs_alloc_page_array(). Reported-by: Julian Taylor Tested-by: Julian Taylor Link: https://lore.kernel.org/all/8966c095-cbe7-4d22-9784-a647d1bf27c3@1und1.de/ Fixes: 395cb57e8560 ("btrfs: wait between incomplete batch memory allocations") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Sweet Tea Dorminy Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 18 ++++-------------- 1 file changed, 4 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index b18034f2ab80..2776112dbdf8 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -681,31 +681,21 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) int btrfs_alloc_page_array(unsigned int nr_pages, struct page **page_array, gfp_t extra_gfp) { + const gfp_t gfp = GFP_NOFS | extra_gfp; unsigned int allocated; for (allocated = 0; allocated < nr_pages;) { unsigned int last = allocated; - allocated = alloc_pages_bulk_array(GFP_NOFS | extra_gfp, - nr_pages, page_array); - - if (allocated == nr_pages) - return 0; - - /* - * During this iteration, no page could be allocated, even - * though alloc_pages_bulk_array() falls back to alloc_page() - * if it could not bulk-allocate. So we must be out of memory. - */ - if (allocated == last) { + allocated = alloc_pages_bulk_array(gfp, nr_pages, page_array); + if (unlikely(allocated == last)) { + /* No progress, fail and do cleanup. */ for (int i = 0; i < allocated; i++) { __free_page(page_array[i]); page_array[i] = NULL; } return -ENOMEM; } - - memalloc_retry_wait(GFP_NOFS); } return 0; } -- cgit v1.2.3-59-g8ed1b From 60b703c71fa80de0c2e14af66e57e234019b7da2 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Tue, 2 Apr 2024 12:17:16 +0200 Subject: zonefs: Use str_plural() to fix Coccinelle warning Fixes the following Coccinelle/coccicheck warning reported by string_choices.cocci: opportunity for str_plural(zgroup->g_nr_zones) Signed-off-by: Thorsten Blum Signed-off-by: Damien Le Moal --- fs/zonefs/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c index c6a124e8d565..964fa7f24003 100644 --- a/fs/zonefs/super.c +++ b/fs/zonefs/super.c @@ -1048,7 +1048,7 @@ static int zonefs_init_zgroup(struct super_block *sb, zonefs_info(sb, "Zone group \"%s\" has %u file%s\n", zonefs_zgroup_name(ztype), zgroup->g_nr_zones, - zgroup->g_nr_zones > 1 ? "s" : ""); + str_plural(zgroup->g_nr_zones)); return 0; } -- cgit v1.2.3-59-g8ed1b From 9b31152fd74eeb10a20345909e542fef6f1d98e2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 18:50:27 -0400 Subject: bcachefs: btree_node_scan: Respect member.data_allowed If a device wasn't used for btree nodes, no need to scan for them. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_node_scan.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index a7d0593b3871..556f76f5c84e 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -216,6 +216,9 @@ static int read_btree_nodes(struct find_btree_nodes *f) closure_init_stack(&cl); for_each_online_member(c, ca) { + if (!(ca->mi.data_allowed & BIT(BCH_DATA_btree))) + continue; + struct find_btree_nodes_worker *w = kmalloc(sizeof(*w), GFP_KERNEL); struct task_struct *t; -- cgit v1.2.3-59-g8ed1b From 6e45a30fe5e7cf5d42ac07262a3d97644f23dc68 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Tue, 9 Apr 2024 23:53:00 +0000 Subject: fs/9p: remove erroneous nlink init from legacy stat2inode In 9p2000 legacy mode, stat2inode initializes nlink to 1, which is redundant with what alloc_inode should have already set. 9p2000.u overrides this with extensions if present in the stat structure, and 9p2000.L incorporates nlink into its stat structure. At the very least this probably messes with directory nlink accounting in legacy mode. Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_inode.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index c5b4d3631c47..47bd77199e20 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -1064,8 +1064,6 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, struct v9fs_session_info *v9ses = sb->s_fs_info; struct v9fs_inode *v9inode = V9FS_I(inode); - set_nlink(inode, 1); - inode_set_atime(inode, stat->atime, 0); inode_set_mtime(inode, stat->mtime, 0); inode_set_ctime(inode, stat->mtime, 0); -- cgit v1.2.3-59-g8ed1b From 3b84adf460381169c085e4bc09e7b57e9e16db0a Mon Sep 17 00:00:00 2001 From: Roman Smirnov Date: Wed, 27 Mar 2024 16:27:55 +0300 Subject: udf: udftime: prevent overflow in udf_disk_stamp_to_time() An overflow can occur in a situation where src.centiseconds takes the value of 255. This situation is unlikely, but there is no validation check anywere in the code. Found by Linux Verification Center (linuxtesting.org) with Svace. Suggested-by: Jan Kara Signed-off-by: Roman Smirnov Reviewed-by: Sergey Shtylyov Signed-off-by: Jan Kara Message-Id: <20240327132755.13945-1-r.smirnov@omp.ru> --- fs/udf/udftime.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/udf/udftime.c b/fs/udf/udftime.c index 758163af39c2..78ecc633606f 100644 --- a/fs/udf/udftime.c +++ b/fs/udf/udftime.c @@ -46,13 +46,18 @@ udf_disk_stamp_to_time(struct timespec64 *dest, struct timestamp src) dest->tv_sec = mktime64(year, src.month, src.day, src.hour, src.minute, src.second); dest->tv_sec -= offset * 60; - dest->tv_nsec = 1000 * (src.centiseconds * 10000 + - src.hundredsOfMicroseconds * 100 + src.microseconds); + /* * Sanitize nanosecond field since reportedly some filesystems are * recorded with bogus sub-second values. */ - dest->tv_nsec %= NSEC_PER_SEC; + if (src.centiseconds < 100 && src.hundredsOfMicroseconds < 100 && + src.microseconds < 100) { + dest->tv_nsec = 1000 * (src.centiseconds * 10000 + + src.hundredsOfMicroseconds * 100 + src.microseconds); + } else { + dest->tv_nsec = 0; + } } void -- cgit v1.2.3-59-g8ed1b From d94979904105a7ad8dca6fdcd8cb3fbecada22f1 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 1 Apr 2024 11:48:37 -0600 Subject: timerfd: convert to ->read_iter() Switch timerfd to using fops->read_iter(), so it can support not just O_NONBLOCK but IOCB_NOWAIT as well. With the latter, users like io_uring interact with timerfds a lot better, as they can be driven purely by the poll trigger. Manually get and install the required fd, so that FMODE_NOWAIT can be set before the file is installed into the file table. No functional changes intended in this patch, it's purely a straight conversion to using the read iterator method. Signed-off-by: Jens Axboe --- fs/timerfd.c | 36 ++++++++++++++++++++++++++---------- 1 file changed, 26 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/timerfd.c b/fs/timerfd.c index e9c96a0c79f1..4bf2f8bfec11 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -262,17 +262,18 @@ static __poll_t timerfd_poll(struct file *file, poll_table *wait) return events; } -static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count, - loff_t *ppos) +static ssize_t timerfd_read_iter(struct kiocb *iocb, struct iov_iter *to) { + struct file *file = iocb->ki_filp; struct timerfd_ctx *ctx = file->private_data; ssize_t res; u64 ticks = 0; - if (count < sizeof(ticks)) + if (iov_iter_count(to) < sizeof(ticks)) return -EINVAL; + spin_lock_irq(&ctx->wqh.lock); - if (file->f_flags & O_NONBLOCK) + if (file->f_flags & O_NONBLOCK || iocb->ki_flags & IOCB_NOWAIT) res = -EAGAIN; else res = wait_event_interruptible_locked_irq(ctx->wqh, ctx->ticks); @@ -312,8 +313,11 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count, ctx->ticks = 0; } spin_unlock_irq(&ctx->wqh.lock); - if (ticks) - res = put_user(ticks, (u64 __user *) buf) ? -EFAULT: sizeof(ticks); + if (ticks) { + res = copy_to_iter(&ticks, sizeof(ticks), to); + if (!res) + res = -EFAULT; + } return res; } @@ -384,7 +388,7 @@ static long timerfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg static const struct file_operations timerfd_fops = { .release = timerfd_release, .poll = timerfd_poll, - .read = timerfd_read, + .read_iter = timerfd_read_iter, .llseek = noop_llseek, .show_fdinfo = timerfd_show, .unlocked_ioctl = timerfd_ioctl, @@ -407,6 +411,7 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) { int ufd; struct timerfd_ctx *ctx; + struct file *file; /* Check the TFD_* constants for consistency. */ BUILD_BUG_ON(TFD_CLOEXEC != O_CLOEXEC); @@ -443,11 +448,22 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) ctx->moffs = ktime_mono_to_real(0); - ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, - O_RDWR | (flags & TFD_SHARED_FCNTL_FLAGS)); - if (ufd < 0) + ufd = get_unused_fd_flags(flags & TFD_SHARED_FCNTL_FLAGS); + if (ufd < 0) { + kfree(ctx); + return ufd; + } + + file = anon_inode_getfile("[timerfd]", &timerfd_fops, ctx, + O_RDWR | (flags & TFD_SHARED_FCNTL_FLAGS)); + if (IS_ERR(file)) { + put_unused_fd(ufd); kfree(ctx); + return PTR_ERR(file); + } + file->f_mode |= FMODE_NOWAIT; + fd_install(ufd, file); return ufd; } -- cgit v1.2.3-59-g8ed1b From 40f45fe8eb7efd70e772447dc98bb50c5e323ccb Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 2 Apr 2024 12:26:41 -0600 Subject: userfaultfd: convert to ->read_iter() Rather than use the older style ->read() hook, use ->read_iter() so that userfaultfd can support both O_NONBLOCK and IOCB_NOWAIT for non-blocking read attempts. Split the fd setup into two parts, so that userfaultfd can mark the file mode with FMODE_NOWAIT before installing it into the process table. With that, we can also defer grabbing the mm until we know the rest will succeed, as the fd isn't visible before then. Signed-off-by: Jens Axboe --- fs/userfaultfd.c | 44 ++++++++++++++++++++++++++++---------------- 1 file changed, 28 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 60dcfafdc11a..6d963402c835 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -31,6 +31,7 @@ #include #include #include +#include static int sysctl_unprivileged_userfaultfd __read_mostly; @@ -282,7 +283,7 @@ static inline bool userfaultfd_huge_must_wait(struct userfaultfd_ctx *ctx, /* * Verify the pagetables are still not ok after having reigstered into * the fault_pending_wqh to avoid userland having to UFFDIO_WAKE any - * userfault that has already been resolved, if userfaultfd_read and + * userfault that has already been resolved, if userfaultfd_read_iter and * UFFDIO_COPY|ZEROPAGE are being run simultaneously on two different * threads. */ @@ -1177,34 +1178,34 @@ static ssize_t userfaultfd_ctx_read(struct userfaultfd_ctx *ctx, int no_wait, return ret; } -static ssize_t userfaultfd_read(struct file *file, char __user *buf, - size_t count, loff_t *ppos) +static ssize_t userfaultfd_read_iter(struct kiocb *iocb, struct iov_iter *to) { + struct file *file = iocb->ki_filp; struct userfaultfd_ctx *ctx = file->private_data; ssize_t _ret, ret = 0; struct uffd_msg msg; - int no_wait = file->f_flags & O_NONBLOCK; struct inode *inode = file_inode(file); + bool no_wait; if (!userfaultfd_is_initialized(ctx)) return -EINVAL; + no_wait = file->f_flags & O_NONBLOCK || iocb->ki_flags & IOCB_NOWAIT; for (;;) { - if (count < sizeof(msg)) + if (iov_iter_count(to) < sizeof(msg)) return ret ? ret : -EINVAL; _ret = userfaultfd_ctx_read(ctx, no_wait, &msg, inode); if (_ret < 0) return ret ? ret : _ret; - if (copy_to_user((__u64 __user *) buf, &msg, sizeof(msg))) + _ret = !copy_to_iter_full(&msg, sizeof(msg), to); + if (_ret) return ret ? ret : -EFAULT; ret += sizeof(msg); - buf += sizeof(msg); - count -= sizeof(msg); /* * Allow to read more than one fault at time but only * block if waiting for the very first one. */ - no_wait = O_NONBLOCK; + no_wait = true; } } @@ -2172,7 +2173,7 @@ static const struct file_operations userfaultfd_fops = { #endif .release = userfaultfd_release, .poll = userfaultfd_poll, - .read = userfaultfd_read, + .read_iter = userfaultfd_read_iter, .unlocked_ioctl = userfaultfd_ioctl, .compat_ioctl = compat_ptr_ioctl, .llseek = noop_llseek, @@ -2192,6 +2193,7 @@ static void init_once_userfaultfd_ctx(void *mem) static int new_userfaultfd(int flags) { struct userfaultfd_ctx *ctx; + struct file *file; int fd; BUG_ON(!current->mm); @@ -2215,16 +2217,26 @@ static int new_userfaultfd(int flags) init_rwsem(&ctx->map_changing_lock); atomic_set(&ctx->mmap_changing, 0); ctx->mm = current->mm; - /* prevent the mm struct to be freed */ - mmgrab(ctx->mm); + + fd = get_unused_fd_flags(flags & UFFD_SHARED_FCNTL_FLAGS); + if (fd < 0) + goto err_out; /* Create a new inode so that the LSM can block the creation. */ - fd = anon_inode_create_getfd("[userfaultfd]", &userfaultfd_fops, ctx, + file = anon_inode_create_getfile("[userfaultfd]", &userfaultfd_fops, ctx, O_RDONLY | (flags & UFFD_SHARED_FCNTL_FLAGS), NULL); - if (fd < 0) { - mmdrop(ctx->mm); - kmem_cache_free(userfaultfd_ctx_cachep, ctx); + if (IS_ERR(file)) { + put_unused_fd(fd); + fd = PTR_ERR(file); + goto err_out; } + /* prevent the mm struct to be freed */ + mmgrab(ctx->mm); + file->f_mode |= FMODE_NOWAIT; + fd_install(fd, file); + return fd; +err_out: + kmem_cache_free(userfaultfd_ctx_cachep, ctx); return fd; } -- cgit v1.2.3-59-g8ed1b From fbe38120eb1dec94280d0381ce4aea52c44367b1 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 2 Apr 2024 12:12:18 -0600 Subject: signalfd: convert to ->read_iter() Rather than use the older style ->read() hook, use ->read_iter() so that signalfd can support both O_NONBLOCK and IOCB_NOWAIT for non-blocking read attempts. Split the fd setup into two parts, so that signalfd can mark the file mode with FMODE_NOWAIT before installing it into the process table. Signed-off-by: Jens Axboe --- fs/signalfd.c | 44 ++++++++++++++++++++++++++++---------------- 1 file changed, 28 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/signalfd.c b/fs/signalfd.c index e20d1484c663..4a5614442dbf 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -68,8 +68,7 @@ static __poll_t signalfd_poll(struct file *file, poll_table *wait) /* * Copied from copy_siginfo_to_user() in kernel/signal.c */ -static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, - kernel_siginfo_t const *kinfo) +static int signalfd_copyinfo(struct iov_iter *to, kernel_siginfo_t const *kinfo) { struct signalfd_siginfo new; @@ -146,10 +145,10 @@ static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, break; } - if (copy_to_user(uinfo, &new, sizeof(struct signalfd_siginfo))) + if (!copy_to_iter_full(&new, sizeof(struct signalfd_siginfo), to)) return -EFAULT; - return sizeof(*uinfo); + return sizeof(struct signalfd_siginfo); } static ssize_t signalfd_dequeue(struct signalfd_ctx *ctx, kernel_siginfo_t *info, @@ -199,28 +198,27 @@ static ssize_t signalfd_dequeue(struct signalfd_ctx *ctx, kernel_siginfo_t *info * error code. The "count" parameter must be at least the size of a * "struct signalfd_siginfo". */ -static ssize_t signalfd_read(struct file *file, char __user *buf, size_t count, - loff_t *ppos) +static ssize_t signalfd_read_iter(struct kiocb *iocb, struct iov_iter *to) { + struct file *file = iocb->ki_filp; struct signalfd_ctx *ctx = file->private_data; - struct signalfd_siginfo __user *siginfo; - int nonblock = file->f_flags & O_NONBLOCK; + size_t count = iov_iter_count(to); ssize_t ret, total = 0; kernel_siginfo_t info; + bool nonblock; count /= sizeof(struct signalfd_siginfo); if (!count) return -EINVAL; - siginfo = (struct signalfd_siginfo __user *) buf; + nonblock = file->f_flags & O_NONBLOCK || iocb->ki_flags & IOCB_NOWAIT; do { ret = signalfd_dequeue(ctx, &info, nonblock); if (unlikely(ret <= 0)) break; - ret = signalfd_copyinfo(siginfo, &info); + ret = signalfd_copyinfo(to, &info); if (ret < 0) break; - siginfo++; total += ret; nonblock = 1; } while (--count); @@ -246,7 +244,7 @@ static const struct file_operations signalfd_fops = { #endif .release = signalfd_release, .poll = signalfd_poll, - .read = signalfd_read, + .read_iter = signalfd_read_iter, .llseek = noop_llseek, }; @@ -265,20 +263,34 @@ static int do_signalfd4(int ufd, sigset_t *mask, int flags) signotset(mask); if (ufd == -1) { + struct file *file; + ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); if (!ctx) return -ENOMEM; ctx->sigmask = *mask; + ufd = get_unused_fd_flags(flags & O_CLOEXEC); + if (ufd < 0) { + kfree(ctx); + return ufd; + } + + file = anon_inode_getfile("[signalfd]", &signalfd_fops, ctx, + O_RDWR | (flags & O_NONBLOCK)); + if (IS_ERR(file)) { + put_unused_fd(ufd); + kfree(ctx); + return ufd; + } + file->f_mode |= FMODE_NOWAIT; + /* * When we call this, the initialization must be complete, since * anon_inode_getfd() will install the fd. */ - ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx, - O_RDWR | (flags & (O_CLOEXEC | O_NONBLOCK))); - if (ufd < 0) - kfree(ctx); + fd_install(ufd, file); } else { struct fd f = fdget(ufd); if (!f.file) -- cgit v1.2.3-59-g8ed1b From ec4535b2a1d709d3a1fbec26739c672f13c98a7b Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Mon, 8 Apr 2024 18:32:17 -0300 Subject: smb: client: fix NULL ptr deref in cifs_mark_open_handles_for_deleted_file() cifs_get_fattr() may be called with a NULL inode, so check for a non-NULL inode before calling cifs_mark_open_handles_for_deleted_file(). This fixes the following oops: mount.cifs //srv/share /mnt -o ...,vers=3.1.1 cd /mnt touch foo; tail -f foo & rm foo cat foo BUG: kernel NULL pointer dereference, address: 00000000000005c0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: 0000 [#1] PREEMPT SMP NOPTI CPU: 2 PID: 696 Comm: cat Not tainted 6.9.0-rc2 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-1.fc39 04/01/2014 RIP: 0010:__lock_acquire+0x5d/0x1c70 Code: 00 00 44 8b a4 24 a0 00 00 00 45 85 f6 0f 84 bb 06 00 00 8b 2d 48 e2 95 01 45 89 c3 41 89 d2 45 89 c8 85 ed 0 0 <48> 81 3f 40 7a 76 83 44 0f 44 d8 83 fe 01 0f 86 1b 03 00 00 31 d2 RSP: 0018:ffffc90000b37490 EFLAGS: 00010002 RAX: 0000000000000000 RBX: ffff888110021ec0 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 00000000000005c0 RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000001 R15: 0000000000000200 FS: 00007f2a1fa08740(0000) GS:ffff888157a00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000000005c0 CR3: 000000011ac7c000 CR4: 0000000000750ef0 PKRU: 55555554 Call Trace: ? __die+0x23/0x70 ? page_fault_oops+0x180/0x490 ? srso_alias_return_thunk+0x5/0xfbef5 ? exc_page_fault+0x70/0x230 ? asm_exc_page_fault+0x26/0x30 ? __lock_acquire+0x5d/0x1c70 ? srso_alias_return_thunk+0x5/0xfbef5 ? srso_alias_return_thunk+0x5/0xfbef5 lock_acquire+0xc0/0x2d0 ? cifs_mark_open_handles_for_deleted_file+0x3a/0x100 [cifs] ? srso_alias_return_thunk+0x5/0xfbef5 ? kmem_cache_alloc+0x2d9/0x370 _raw_spin_lock+0x34/0x80 ? cifs_mark_open_handles_for_deleted_file+0x3a/0x100 [cifs] cifs_mark_open_handles_for_deleted_file+0x3a/0x100 [cifs] cifs_get_fattr+0x24c/0x940 [cifs] ? srso_alias_return_thunk+0x5/0xfbef5 cifs_get_inode_info+0x96/0x120 [cifs] cifs_lookup+0x16e/0x800 [cifs] cifs_atomic_open+0xc7/0x5d0 [cifs] ? lookup_open.isra.0+0x3ce/0x5f0 ? __pfx_cifs_atomic_open+0x10/0x10 [cifs] lookup_open.isra.0+0x3ce/0x5f0 path_openat+0x42b/0xc30 ? srso_alias_return_thunk+0x5/0xfbef5 ? srso_alias_return_thunk+0x5/0xfbef5 ? srso_alias_return_thunk+0x5/0xfbef5 do_filp_open+0xc4/0x170 do_sys_openat2+0xab/0xe0 __x64_sys_openat+0x57/0xa0 do_syscall_64+0xc1/0x1e0 entry_SYSCALL_64_after_hwframe+0x72/0x7a Fixes: ffceb7640cbf ("smb: client: do not defer close open handles to deleted files") Reviewed-by: Meetakshi Setiya Reviewed-by: Bharath SM Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/inode.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index 91b07ef9e25c..60afab5c83d4 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -1105,7 +1105,8 @@ static int cifs_get_fattr(struct cifs_open_info_data *data, } else { cifs_open_info_to_fattr(fattr, data, sb); } - if (!rc && fattr->cf_flags & CIFS_FATTR_DELETE_PENDING) + if (!rc && *inode && + (fattr->cf_flags & CIFS_FATTR_DELETE_PENDING)) cifs_mark_open_handles_for_deleted_file(*inode, full_path); break; case -EREMOTE: -- cgit v1.2.3-59-g8ed1b From beccf29114886f1604e26f739cd108f048878ca8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 12:53:28 -0400 Subject: bcachefs: Fix a race in btree_update_nodes_written() One btree update might have terminated in a node update, and then while it is in flight another btree update might free that original node. This race has to be handled in btree_update_nodes_written() - we were missing a READ_ONCE(). Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index a4a63e363047..c4a5e83a56a4 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -704,9 +704,13 @@ static void btree_update_nodes_written(struct btree_update *as) bch2_fs_fatal_err_on(ret && !bch2_journal_error(&c->journal), c, "%s", bch2_err_str(ret)); err: - if (as->b) { - - b = as->b; + /* + * We have to be careful because another thread might be getting ready + * to free as->b and calling btree_update_reparent() on us - we'll + * recheck under btree_update_lock below: + */ + b = READ_ONCE(as->b); + if (b) { btree_path_idx_t path_idx = get_unlocked_mut_path(trans, as->btree_id, b->c.level, b->key.k.p); struct btree_path *path = trans->paths + path_idx; -- cgit v1.2.3-59-g8ed1b From 517236cb3e2f77bc785f06802dfbcca19dffd9ad Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 00:10:18 -0400 Subject: bcachefs: Kill read lock dropping in bch2_btree_node_lock_write_nofail() dropping read locks in bch2_btree_node_lock_write_nofail() dates from before we had the cycle detector; we can now tell the cycle detector directly when taking a lock may not fail because we can't handle transaction restarts. This is needed for adding should_be_locked asserts. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_locking.c | 28 +--------------------------- 1 file changed, 1 insertion(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index b9b151e693ed..f2caf491957e 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -440,33 +440,7 @@ void bch2_btree_node_lock_write_nofail(struct btree_trans *trans, struct btree_path *path, struct btree_bkey_cached_common *b) { - struct btree_path *linked; - unsigned i, iter; - int ret; - - /* - * XXX BIG FAT NOTICE - * - * Drop all read locks before taking a write lock: - * - * This is a hack, because bch2_btree_node_lock_write_nofail() is a - * hack - but by dropping read locks first, this should never fail, and - * we only use this in code paths where whatever read locks we've - * already taken are no longer needed: - */ - - trans_for_each_path(trans, linked, iter) { - if (!linked->nodes_locked) - continue; - - for (i = 0; i < BTREE_MAX_DEPTH; i++) - if (btree_node_read_locked(linked, i)) { - btree_node_unlock(trans, linked, i); - btree_path_set_dirty(linked, BTREE_ITER_NEED_RELOCK); - } - } - - ret = __btree_node_lock_write(trans, path, b, true); + int ret = __btree_node_lock_write(trans, path, b, true); BUG_ON(ret); } -- cgit v1.2.3-59-g8ed1b From 1189bdda6c991cbf9342d84410042dd5f3a792e0 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 01:30:22 -0400 Subject: bcachefs: Fix __bch2_btree_and_journal_iter_init_node_iter() We weren't respecting trans->journal_replay_not_finished - we shouldn't be searching the journal keys unless we have a ref on them. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_journal_iter.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index a6413a8747d3..1e8cf49a6935 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -469,10 +469,15 @@ void __bch2_btree_and_journal_iter_init_node_iter(struct btree_trans *trans, iter->trans = trans; iter->b = b; iter->node_iter = node_iter; - bch2_journal_iter_init(trans->c, &iter->journal, b->c.btree_id, b->c.level, pos); - INIT_LIST_HEAD(&iter->journal.list); iter->pos = b->data->min_key; iter->at_end = false; + INIT_LIST_HEAD(&iter->journal.list); + + if (trans->journal_replay_not_finished) { + bch2_journal_iter_init(trans->c, &iter->journal, b->c.btree_id, b->c.level, pos); + if (!test_bit(BCH_FS_may_go_rw, &trans->c->flags)) + list_add(&iter->journal.list, &trans->c->journal_iters); + } } /* @@ -487,9 +492,6 @@ void bch2_btree_and_journal_iter_init_node_iter(struct btree_trans *trans, bch2_btree_node_iter_init_from_start(&node_iter, b); __bch2_btree_and_journal_iter_init_node_iter(trans, iter, b, node_iter, b->data->min_key); - if (trans->journal_replay_not_finished && - !test_bit(BCH_FS_may_go_rw, &trans->c->flags)) - list_add(&iter->journal.list, &trans->c->journal_iters); } /* sort and dedup all keys in the journal: */ -- cgit v1.2.3-59-g8ed1b From fd0a133ef6edb8e49e74af505b38c93af4ba0b1e Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 10 Apr 2024 17:22:21 +0100 Subject: fs/direct-io: remove redundant assignment to variable retval The variable retval is being assigned a value that is not being read, it is being re-assigned later on in the function. The assignment is redundant and can be removed. Cleans up clang scan build warning: fs/direct-io.c:1220:2: warning: Value stored to 'retval' is never read [deadcode.DeadStores] Signed-off-by: Colin Ian King Link: https://lore.kernel.org/r/20240410162221.292485-1-colin.i.king@gmail.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/direct-io.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/direct-io.c b/fs/direct-io.c index 62c97ff9e852..b0aafe640fa4 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -1217,7 +1217,6 @@ ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, */ inode_dio_begin(inode); - retval = 0; sdio.blkbits = blkbits; sdio.blkfactor = i_blkbits - blkbits; sdio.block_in_file = offset >> blkbits; -- cgit v1.2.3-59-g8ed1b From a90bca2228c0646fc29a72689d308e5fe03e6d78 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 13 Mar 2024 17:43:41 -0400 Subject: fs: sysfs: Fix reference leak in sysfs_break_active_protection() The sysfs_break_active_protection() routine has an obvious reference leak in its error path. If the call to kernfs_find_and_get() fails then kn will be NULL, so the companion sysfs_unbreak_active_protection() routine won't get called (and would only cause an access violation by trying to dereference kn->parent if it was called). As a result, the reference to kobj acquired at the start of the function will never be released. Fix the leak by adding an explicit kobject_put() call when kn is NULL. Signed-off-by: Alan Stern Fixes: 2afc9166f79b ("scsi: sysfs: Introduce sysfs_{un,}break_active_protection()") Cc: Bart Van Assche Cc: stable@vger.kernel.org Reviewed-by: Bart Van Assche Acked-by: Tejun Heo Link: https://lore.kernel.org/r/8a4d3f0f-c5e3-4b70-a188-0ca433f9e6f9@rowland.harvard.edu Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/file.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 6b7652fb8050..7cd64021d453 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -463,6 +463,8 @@ struct kernfs_node *sysfs_break_active_protection(struct kobject *kobj, kn = kernfs_find_and_get(kobj->sd, attr->name); if (kn) kernfs_break_active_protection(kn); + else + kobject_put(kobj); return kn; } EXPORT_SYMBOL_GPL(sysfs_break_active_protection); -- cgit v1.2.3-59-g8ed1b From f488138b526715c6d2568d7329c4477911be4210 Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Thu, 11 Apr 2024 11:45:57 +0200 Subject: NFSD: fix endianness issue in nfsd4_encode_fattr4 The nfs4 mount fails with EIO on 64-bit big endian architectures since v6.7. The issue arises from employing a union in the nfsd4_encode_fattr4() function to overlay a 32-bit array with a 64-bit values based bitmap, which does not function as intended. Address the endianness issue by utilizing bitmap_from_arr32() to copy 32-bit attribute masks into a bitmap in an endianness-agnostic manner. Cc: stable@vger.kernel.org Fixes: fce7913b13d0 ("NFSD: Use a bitmask loop to encode FATTR4 results") Link: https://bugs.launchpad.net/ubuntu/+source/nfs-utils/+bug/2060217 Signed-off-by: Vasily Gorbik Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs4xdr.c | 47 +++++++++++++++++++++++------------------------ 1 file changed, 23 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index fac938f563ad..1955481832e0 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -3490,11 +3490,13 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct dentry *dentry, const u32 *bmval, int ignore_crossmnt) { + DECLARE_BITMAP(attr_bitmap, ARRAY_SIZE(nfsd4_enc_fattr4_encode_ops)); struct nfsd4_fattr_args args; struct svc_fh *tempfh = NULL; int starting_len = xdr->buf->len; __be32 *attrlen_p, status; int attrlen_offset; + u32 attrmask[3]; int err; struct nfsd4_compoundres *resp = rqstp->rq_resp; u32 minorversion = resp->cstate.minorversion; @@ -3502,10 +3504,6 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, .mnt = exp->ex_path.mnt, .dentry = dentry, }; - union { - u32 attrmask[3]; - unsigned long mask[2]; - } u; unsigned long bit; bool file_modified = false; u64 size = 0; @@ -3521,20 +3519,19 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, /* * Make a local copy of the attribute bitmap that can be modified. */ - memset(&u, 0, sizeof(u)); - u.attrmask[0] = bmval[0]; - u.attrmask[1] = bmval[1]; - u.attrmask[2] = bmval[2]; + attrmask[0] = bmval[0]; + attrmask[1] = bmval[1]; + attrmask[2] = bmval[2]; args.rdattr_err = 0; if (exp->ex_fslocs.migrated) { - status = fattr_handle_absent_fs(&u.attrmask[0], &u.attrmask[1], - &u.attrmask[2], &args.rdattr_err); + status = fattr_handle_absent_fs(&attrmask[0], &attrmask[1], + &attrmask[2], &args.rdattr_err); if (status) goto out; } args.size = 0; - if (u.attrmask[0] & (FATTR4_WORD0_CHANGE | FATTR4_WORD0_SIZE)) { + if (attrmask[0] & (FATTR4_WORD0_CHANGE | FATTR4_WORD0_SIZE)) { status = nfsd4_deleg_getattr_conflict(rqstp, d_inode(dentry), &file_modified, &size); if (status) @@ -3553,16 +3550,16 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, if (!(args.stat.result_mask & STATX_BTIME)) /* underlying FS does not offer btime so we can't share it */ - u.attrmask[1] &= ~FATTR4_WORD1_TIME_CREATE; - if ((u.attrmask[0] & (FATTR4_WORD0_FILES_AVAIL | FATTR4_WORD0_FILES_FREE | + attrmask[1] &= ~FATTR4_WORD1_TIME_CREATE; + if ((attrmask[0] & (FATTR4_WORD0_FILES_AVAIL | FATTR4_WORD0_FILES_FREE | FATTR4_WORD0_FILES_TOTAL | FATTR4_WORD0_MAXNAME)) || - (u.attrmask[1] & (FATTR4_WORD1_SPACE_AVAIL | FATTR4_WORD1_SPACE_FREE | + (attrmask[1] & (FATTR4_WORD1_SPACE_AVAIL | FATTR4_WORD1_SPACE_FREE | FATTR4_WORD1_SPACE_TOTAL))) { err = vfs_statfs(&path, &args.statfs); if (err) goto out_nfserr; } - if ((u.attrmask[0] & (FATTR4_WORD0_FILEHANDLE | FATTR4_WORD0_FSID)) && + if ((attrmask[0] & (FATTR4_WORD0_FILEHANDLE | FATTR4_WORD0_FSID)) && !fhp) { tempfh = kmalloc(sizeof(struct svc_fh), GFP_KERNEL); status = nfserr_jukebox; @@ -3577,10 +3574,10 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, args.fhp = fhp; args.acl = NULL; - if (u.attrmask[0] & FATTR4_WORD0_ACL) { + if (attrmask[0] & FATTR4_WORD0_ACL) { err = nfsd4_get_nfs4_acl(rqstp, dentry, &args.acl); if (err == -EOPNOTSUPP) - u.attrmask[0] &= ~FATTR4_WORD0_ACL; + attrmask[0] &= ~FATTR4_WORD0_ACL; else if (err == -EINVAL) { status = nfserr_attrnotsupp; goto out; @@ -3592,17 +3589,17 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, #ifdef CONFIG_NFSD_V4_SECURITY_LABEL args.context = NULL; - if ((u.attrmask[2] & FATTR4_WORD2_SECURITY_LABEL) || - u.attrmask[0] & FATTR4_WORD0_SUPPORTED_ATTRS) { + if ((attrmask[2] & FATTR4_WORD2_SECURITY_LABEL) || + attrmask[0] & FATTR4_WORD0_SUPPORTED_ATTRS) { if (exp->ex_flags & NFSEXP_SECURITY_LABEL) err = security_inode_getsecctx(d_inode(dentry), &args.context, &args.contextlen); else err = -EOPNOTSUPP; args.contextsupport = (err == 0); - if (u.attrmask[2] & FATTR4_WORD2_SECURITY_LABEL) { + if (attrmask[2] & FATTR4_WORD2_SECURITY_LABEL) { if (err == -EOPNOTSUPP) - u.attrmask[2] &= ~FATTR4_WORD2_SECURITY_LABEL; + attrmask[2] &= ~FATTR4_WORD2_SECURITY_LABEL; else if (err) goto out_nfserr; } @@ -3610,8 +3607,8 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, #endif /* CONFIG_NFSD_V4_SECURITY_LABEL */ /* attrmask */ - status = nfsd4_encode_bitmap4(xdr, u.attrmask[0], - u.attrmask[1], u.attrmask[2]); + status = nfsd4_encode_bitmap4(xdr, attrmask[0], attrmask[1], + attrmask[2]); if (status) goto out; @@ -3620,7 +3617,9 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, attrlen_p = xdr_reserve_space(xdr, XDR_UNIT); if (!attrlen_p) goto out_resource; - for_each_set_bit(bit, (const unsigned long *)&u.mask, + bitmap_from_arr32(attr_bitmap, attrmask, + ARRAY_SIZE(nfsd4_enc_fattr4_encode_ops)); + for_each_set_bit(bit, attr_bitmap, ARRAY_SIZE(nfsd4_enc_fattr4_encode_ops)) { status = nfsd4_enc_fattr4_encode_ops[bit](xdr, &args); if (status != nfs_ok) -- cgit v1.2.3-59-g8ed1b From d48c03198a92edf41e89477dab4f602df15165ee Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Sat, 6 Apr 2024 15:52:01 +0200 Subject: sysfs: Add sysfs_bin_attr_simple_read() helper When drivers expose a bin_attribute in sysfs which is backed by a buffer in memory, a common pattern is to set the @private and @size members in struct bin_attribute to the buffer's location and size. The ->read() callback then merely consists of a single memcpy() call. It's not even necessary to perform bounds checks as these are already handled by sysfs_kf_bin_read(). However each driver is so far providing its own ->read() implementation. The pattern is sufficiently frequent to merit a public helper, so add sysfs_bin_attr_simple_read() as well as BIN_ATTR_SIMPLE_RO() and BIN_ATTR_SIMPLE_ADMIN_RO() macros to ease declaration of such bin_attributes and reduce LoC and .text section size. Signed-off-by: Lukas Wunner Reviewed-by: Greg Kroah-Hartman Acked-by: Rafael J. Wysocki Acked-by: Ard Biesheuvel Link: https://lore.kernel.org/r/5ed62b197a442ec6db53d8746d9d806dd0576e2d.1712410202.git.lukas@wunner.de Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/file.c | 27 +++++++++++++++++++++++++++ include/linux/sysfs.h | 15 +++++++++++++++ 2 files changed, 42 insertions(+) (limited to 'fs') diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 6b7652fb8050..289b57d32c89 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -783,3 +783,30 @@ int sysfs_emit_at(char *buf, int at, const char *fmt, ...) return len; } EXPORT_SYMBOL_GPL(sysfs_emit_at); + +/** + * sysfs_bin_attr_simple_read - read callback to simply copy from memory. + * @file: attribute file which is being read. + * @kobj: object to which the attribute belongs. + * @attr: attribute descriptor. + * @buf: destination buffer. + * @off: offset in bytes from which to read. + * @count: maximum number of bytes to read. + * + * Simple ->read() callback for bin_attributes backed by a buffer in memory. + * The @private and @size members in struct bin_attribute must be set to the + * buffer's location and size before the bin_attribute is created in sysfs. + * + * Bounds check for @off and @count is done in sysfs_kf_bin_read(). + * Negative value check for @off is done in vfs_setpos() and default_llseek(). + * + * Returns number of bytes written to @buf. + */ +ssize_t sysfs_bin_attr_simple_read(struct file *file, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t off, size_t count) +{ + memcpy(buf, attr->private + off, count); + return count; +} +EXPORT_SYMBOL_GPL(sysfs_bin_attr_simple_read); diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 326341c62385..a7d725fbf739 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -371,6 +371,17 @@ struct bin_attribute bin_attr_##_name = __BIN_ATTR_ADMIN_RO(_name, _size) #define BIN_ATTR_ADMIN_RW(_name, _size) \ struct bin_attribute bin_attr_##_name = __BIN_ATTR_ADMIN_RW(_name, _size) +#define __BIN_ATTR_SIMPLE_RO(_name, _mode) { \ + .attr = { .name = __stringify(_name), .mode = _mode }, \ + .read = sysfs_bin_attr_simple_read, \ +} + +#define BIN_ATTR_SIMPLE_RO(_name) \ +struct bin_attribute bin_attr_##_name = __BIN_ATTR_SIMPLE_RO(_name, 0444) + +#define BIN_ATTR_SIMPLE_ADMIN_RO(_name) \ +struct bin_attribute bin_attr_##_name = __BIN_ATTR_SIMPLE_RO(_name, 0400) + struct sysfs_ops { ssize_t (*show)(struct kobject *, struct attribute *, char *); ssize_t (*store)(struct kobject *, struct attribute *, const char *, size_t); @@ -478,6 +489,10 @@ int sysfs_emit(char *buf, const char *fmt, ...); __printf(3, 4) int sysfs_emit_at(char *buf, int at, const char *fmt, ...); +ssize_t sysfs_bin_attr_simple_read(struct file *file, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t off, size_t count); + #else /* CONFIG_SYSFS */ static inline int sysfs_create_dir_ns(struct kobject *kobj, const void *ns) -- cgit v1.2.3-59-g8ed1b From c82389947d90c8b0dd206d9ef082191eb58df8fc Mon Sep 17 00:00:00 2001 From: Marco Elver Date: Thu, 11 Apr 2024 12:20:57 +0200 Subject: tracing: Add sched_prepare_exec tracepoint Add "sched_prepare_exec" tracepoint, which is run right after the point of no return but before the current task assumes its new exec identity. Unlike the tracepoint "sched_process_exec", the "sched_prepare_exec" tracepoint runs before flushing the old exec, i.e. while the task still has the original state (such as original MM), but when the new exec either succeeds or crashes (but never returns to the original exec). Being able to trace this event can be helpful in a number of use cases: * allowing tracing eBPF programs access to the original MM on exec, before current->mm is replaced; * counting exec in the original task (via perf event); * profiling flush time ("sched_prepare_exec" to "sched_process_exec"). Example of tracing output: $ cat /sys/kernel/debug/tracing/trace_pipe <...>-379 [003] ..... 179.626921: sched_prepare_exec: interp=/usr/bin/sshd filename=/usr/bin/sshd pid=379 comm=sshd <...>-381 [002] ..... 180.048580: sched_prepare_exec: interp=/bin/bash filename=/bin/bash pid=381 comm=sshd <...>-385 [001] ..... 180.068277: sched_prepare_exec: interp=/usr/bin/tty filename=/usr/bin/tty pid=385 comm=bash <...>-389 [006] ..... 192.020147: sched_prepare_exec: interp=/usr/bin/dmesg filename=/usr/bin/dmesg pid=389 comm=bash Signed-off-by: Marco Elver Acked-by: Steven Rostedt (Google) Reviewed-by: Masami Hiramatsu (Google) Link: https://lore.kernel.org/r/20240411102158.1272267-1-elver@google.com Signed-off-by: Kees Cook --- fs/exec.c | 8 ++++++++ include/trace/events/sched.h | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 43 insertions(+) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index cf1df7f16e55..b3c40fbb325f 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1267,6 +1267,14 @@ int begin_new_exec(struct linux_binprm * bprm) if (retval) return retval; + /* + * This tracepoint marks the point before flushing the old exec where + * the current task is still unchanged, but errors are fatal (point of + * no return). The later "sched_process_exec" tracepoint is called after + * the current task has successfully switched to the new exec. + */ + trace_sched_prepare_exec(current, bprm); + /* * Ensure all future errors are fatal. */ diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h index dbb01b4b7451..226f47c6939c 100644 --- a/include/trace/events/sched.h +++ b/include/trace/events/sched.h @@ -420,6 +420,41 @@ TRACE_EVENT(sched_process_exec, __entry->pid, __entry->old_pid) ); +/** + * sched_prepare_exec - called before setting up new exec + * @task: pointer to the current task + * @bprm: pointer to linux_binprm used for new exec + * + * Called before flushing the old exec, where @task is still unchanged, but at + * the point of no return during switching to the new exec. At the point it is + * called the exec will either succeed, or on failure terminate the task. Also + * see the "sched_process_exec" tracepoint, which is called right after @task + * has successfully switched to the new exec. + */ +TRACE_EVENT(sched_prepare_exec, + + TP_PROTO(struct task_struct *task, struct linux_binprm *bprm), + + TP_ARGS(task, bprm), + + TP_STRUCT__entry( + __string( interp, bprm->interp ) + __string( filename, bprm->filename ) + __field( pid_t, pid ) + __string( comm, task->comm ) + ), + + TP_fast_assign( + __assign_str(interp, bprm->interp); + __assign_str(filename, bprm->filename); + __entry->pid = task->pid; + __assign_str(comm, task->comm); + ), + + TP_printk("interp=%s filename=%s pid=%d comm=%s", + __get_str(interp), __get_str(filename), + __entry->pid, __get_str(comm)) +); #ifdef CONFIG_SCHEDSTATS #define DEFINE_EVENT_SCHEDSTAT DEFINE_EVENT -- cgit v1.2.3-59-g8ed1b From b372e96bd0a32729d55d27f613c8bc80708a82e1 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 25 Mar 2024 09:21:20 +1100 Subject: ceph: redirty page before returning AOP_WRITEPAGE_ACTIVATE The page has been marked clean before writepage is called. If we don't redirty it before postponing the write, it might never get written. Cc: stable@vger.kernel.org Fixes: 503d4fa6ee28 ("ceph: remove reliance on bdi congestion") Signed-off-by: NeilBrown Reviewed-by: Jeff Layton Reviewed-by: Xiubo Li Signed-off-by: Ilya Dryomov --- fs/ceph/addr.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 1340d77124ae..ee9caf7916fb 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -795,8 +795,10 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc) ihold(inode); if (wbc->sync_mode == WB_SYNC_NONE && - ceph_inode_to_fs_client(inode)->write_congested) + ceph_inode_to_fs_client(inode)->write_congested) { + redirty_page_for_writepage(wbc, page); return AOP_WRITEPAGE_ACTIVATE; + } wait_on_page_fscache(page); -- cgit v1.2.3-59-g8ed1b From 17f8dc2db52185460f212052f3a692c1fdc167ba Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Tue, 9 Apr 2024 08:56:03 +0800 Subject: ceph: switch to use cap_delay_lock for the unlink delay list The same list item will be used in both cap_delay_list and cap_unlink_delay_list, so it's buggy to use two different locks to protect them. Cc: stable@vger.kernel.org Fixes: dbc347ef7f0c ("ceph: add ceph_cap_unlink_work to fire check_caps() immediately") Link: https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/AODC76VXRAMXKLFDCTK4TKFDDPWUSCN5 Reported-by: Marc Ruhmann Signed-off-by: Xiubo Li Reviewed-by: Ilya Dryomov Tested-by: Marc Ruhmann Signed-off-by: Ilya Dryomov --- fs/ceph/caps.c | 4 ++-- fs/ceph/mds_client.c | 9 ++++----- fs/ceph/mds_client.h | 3 +-- 3 files changed, 7 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 55051ad09c19..c4941ba245ac 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -4783,13 +4783,13 @@ int ceph_drop_caps_for_unlink(struct inode *inode) doutc(mdsc->fsc->client, "%p %llx.%llx\n", inode, ceph_vinop(inode)); - spin_lock(&mdsc->cap_unlink_delay_lock); + spin_lock(&mdsc->cap_delay_lock); ci->i_ceph_flags |= CEPH_I_FLUSH; if (!list_empty(&ci->i_cap_delay_list)) list_del_init(&ci->i_cap_delay_list); list_add_tail(&ci->i_cap_delay_list, &mdsc->cap_unlink_delay_list); - spin_unlock(&mdsc->cap_unlink_delay_lock); + spin_unlock(&mdsc->cap_delay_lock); /* * Fire the work immediately, because the MDS maybe diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 3ab9c268a8bb..360b686c3c67 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -2504,7 +2504,7 @@ static void ceph_cap_unlink_work(struct work_struct *work) struct ceph_client *cl = mdsc->fsc->client; doutc(cl, "begin\n"); - spin_lock(&mdsc->cap_unlink_delay_lock); + spin_lock(&mdsc->cap_delay_lock); while (!list_empty(&mdsc->cap_unlink_delay_list)) { struct ceph_inode_info *ci; struct inode *inode; @@ -2516,15 +2516,15 @@ static void ceph_cap_unlink_work(struct work_struct *work) inode = igrab(&ci->netfs.inode); if (inode) { - spin_unlock(&mdsc->cap_unlink_delay_lock); + spin_unlock(&mdsc->cap_delay_lock); doutc(cl, "on %p %llx.%llx\n", inode, ceph_vinop(inode)); ceph_check_caps(ci, CHECK_CAPS_FLUSH); iput(inode); - spin_lock(&mdsc->cap_unlink_delay_lock); + spin_lock(&mdsc->cap_delay_lock); } } - spin_unlock(&mdsc->cap_unlink_delay_lock); + spin_unlock(&mdsc->cap_delay_lock); doutc(cl, "done\n"); } @@ -5404,7 +5404,6 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) INIT_LIST_HEAD(&mdsc->cap_wait_list); spin_lock_init(&mdsc->cap_delay_lock); INIT_LIST_HEAD(&mdsc->cap_unlink_delay_list); - spin_lock_init(&mdsc->cap_unlink_delay_lock); INIT_LIST_HEAD(&mdsc->snap_flush_list); spin_lock_init(&mdsc->snap_flush_lock); mdsc->last_cap_flush_tid = 1; diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 03f8ff00874f..b88e80415224 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -461,9 +461,8 @@ struct ceph_mds_client { struct delayed_work delayed_work; /* delayed work */ unsigned long last_renew_caps; /* last time we renewed our caps */ struct list_head cap_delay_list; /* caps with delayed release */ - spinlock_t cap_delay_lock; /* protects cap_delay_list */ struct list_head cap_unlink_delay_list; /* caps with delayed release for unlink */ - spinlock_t cap_unlink_delay_lock; /* protects cap_unlink_delay_list */ + spinlock_t cap_delay_lock; /* protects cap_delay_list and cap_unlink_delay_list */ struct list_head snap_flush_list; /* cap_snaps ready to flush */ spinlock_t snap_flush_lock; -- cgit v1.2.3-59-g8ed1b From 28e0947651ce6a2200b9a7eceb93282e97d7e51a Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 6 Apr 2024 23:16:08 -0500 Subject: smb3: fix Open files on server counter going negative We were decrementing the count of open files on server twice for the case where we were closing cached directories. Fixes: 8e843bf38f7b ("cifs: return a single-use cfid if we did not get a lease") Cc: stable@vger.kernel.org Acked-by: Bharath SM Signed-off-by: Steve French --- fs/smb/client/cached_dir.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cached_dir.c b/fs/smb/client/cached_dir.c index 13a9d7acf8f8..0ff2491c311d 100644 --- a/fs/smb/client/cached_dir.c +++ b/fs/smb/client/cached_dir.c @@ -433,8 +433,8 @@ smb2_close_cached_fid(struct kref *ref) if (cfid->is_open) { rc = SMB2_close(0, cfid->tcon, cfid->fid.persistent_fid, cfid->fid.volatile_fid); - if (rc != -EBUSY && rc != -EAGAIN) - atomic_dec(&cfid->tcon->num_remote_opens); + if (rc) /* should we retry on -EBUSY or -EAGAIN? */ + cifs_dbg(VFS, "close cached dir rc %d\n", rc); } free_cached_dir(cfid); -- cgit v1.2.3-59-g8ed1b From c6ff459037b2e35450af2351037eac4c8aca1d6b Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Tue, 9 Apr 2024 11:28:59 -0300 Subject: smb: client: instantiate when creating SFU files In cifs_sfu_make_node(), on success, instantiate rather than leave it with dentry unhashed negative to support callers that expect mknod(2) to always instantiate. This fixes the following test case: mount.cifs //srv/share /mnt -o ...,sfu mkfifo /mnt/fifo ./xfstests/ltp/growfiles -b -W test -e 1 -u -i 0 -L 30 /mnt/fifo ... BUG: unable to handle page fault for address: 000000034cec4e58 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: 0000 1 PREEMPT SMP PTI CPU: 0 PID: 138098 Comm: growfiles Kdump: loaded Not tainted 5.14.0-436.3987_1240945149.el9.x86_64 #1 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 RIP: 0010:_raw_callee_save__kvm_vcpu_is_preempted+0x0/0x20 Code: e8 15 d9 61 00 e9 63 ff ff ff 41 bd ea ff ff ff e9 58 ff ff ff e8 d0 71 c0 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 <48> 8b 04 fd 60 2b c1 99 80 b8 90 50 03 00 00 0f 95 c0 c3 cc cc cc RSP: 0018:ffffb6a143cf7cf8 EFLAGS: 00010206 RAX: ffff8a9bc30fb038 RBX: ffff8a9bc666a200 RCX: ffff8a9cc0260000 RDX: 00000000736f622e RSI: ffff8a9bc30fb038 RDI: 000000007665645f RBP: ffffb6a143cf7d70 R08: 0000000000001000 R09: 0000000000000001 R10: 0000000000000001 R11: 0000000000000000 R12: ffff8a9bc666a200 R13: 0000559a302a12b0 R14: 0000000000001000 R15: 0000000000000000 FS: 00007fbed1dbb740(0000) GS:ffff8a9cf0000000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000034cec4e58 CR3: 0000000128ec6006 CR4: 0000000000770ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: ? show_trace_log_lvl+0x1c4/0x2df ? show_trace_log_lvl+0x1c4/0x2df ? __mutex_lock.constprop.0+0x5f7/0x6a0 ? __die_body.cold+0x8/0xd ? page_fault_oops+0x134/0x170 ? exc_page_fault+0x62/0x150 ? asm_exc_page_fault+0x22/0x30 ? _pfx_raw_callee_save__kvm_vcpu_is_preempted+0x10/0x10 __mutex_lock.constprop.0+0x5f7/0x6a0 ? __mod_memcg_lruvec_state+0x84/0xd0 pipe_write+0x47/0x650 ? do_anonymous_page+0x258/0x410 ? inode_security+0x22/0x60 ? selinux_file_permission+0x108/0x150 vfs_write+0x2cb/0x410 ksys_write+0x5f/0xe0 do_syscall_64+0x5c/0xf0 ? syscall_exit_to_user_mode+0x22/0x40 ? do_syscall_64+0x6b/0xf0 ? sched_clock_cpu+0x9/0xc0 ? exc_page_fault+0x62/0x150 entry_SYSCALL_64_after_hwframe+0x6e/0x76 Cc: stable@vger.kernel.org Fixes: 72bc63f5e23a ("smb3: fix creating FIFOs when mounting with "sfu" mount option") Suggested-by: Al Viro Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/smb2ops.c | 94 +++++++++++++++++++++++++++++-------------------- 1 file changed, 55 insertions(+), 39 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index b156eefa75d7..78c94d0350fe 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -4964,68 +4964,84 @@ static int smb2_next_header(struct TCP_Server_Info *server, char *buf, return 0; } -int cifs_sfu_make_node(unsigned int xid, struct inode *inode, - struct dentry *dentry, struct cifs_tcon *tcon, - const char *full_path, umode_t mode, dev_t dev) +static int __cifs_sfu_make_node(unsigned int xid, struct inode *inode, + struct dentry *dentry, struct cifs_tcon *tcon, + const char *full_path, umode_t mode, dev_t dev) { - struct cifs_open_info_data buf = {}; struct TCP_Server_Info *server = tcon->ses->server; struct cifs_open_parms oparms; struct cifs_io_parms io_parms = {}; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifs_fid fid; unsigned int bytes_written; - struct win_dev *pdev; + struct win_dev pdev = {}; struct kvec iov[2]; __u32 oplock = server->oplocks ? REQ_OPLOCK : 0; int rc; - if (!S_ISCHR(mode) && !S_ISBLK(mode) && !S_ISFIFO(mode)) + switch (mode & S_IFMT) { + case S_IFCHR: + strscpy(pdev.type, "IntxCHR"); + pdev.major = cpu_to_le64(MAJOR(dev)); + pdev.minor = cpu_to_le64(MINOR(dev)); + break; + case S_IFBLK: + strscpy(pdev.type, "IntxBLK"); + pdev.major = cpu_to_le64(MAJOR(dev)); + pdev.minor = cpu_to_le64(MINOR(dev)); + break; + case S_IFIFO: + strscpy(pdev.type, "LnxFIFO"); + break; + default: return -EPERM; + } - oparms = (struct cifs_open_parms) { - .tcon = tcon, - .cifs_sb = cifs_sb, - .desired_access = GENERIC_WRITE, - .create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR | - CREATE_OPTION_SPECIAL), - .disposition = FILE_CREATE, - .path = full_path, - .fid = &fid, - }; + oparms = CIFS_OPARMS(cifs_sb, tcon, full_path, GENERIC_WRITE, + FILE_CREATE, CREATE_NOT_DIR | + CREATE_OPTION_SPECIAL, ACL_NO_MODE); + oparms.fid = &fid; - rc = server->ops->open(xid, &oparms, &oplock, &buf); + rc = server->ops->open(xid, &oparms, &oplock, NULL); if (rc) return rc; - /* - * BB Do not bother to decode buf since no local inode yet to put - * timestamps in, but we can reuse it safely. - */ - pdev = (struct win_dev *)&buf.fi; io_parms.pid = current->tgid; io_parms.tcon = tcon; - io_parms.length = sizeof(*pdev); - iov[1].iov_base = pdev; - iov[1].iov_len = sizeof(*pdev); - if (S_ISCHR(mode)) { - memcpy(pdev->type, "IntxCHR", 8); - pdev->major = cpu_to_le64(MAJOR(dev)); - pdev->minor = cpu_to_le64(MINOR(dev)); - } else if (S_ISBLK(mode)) { - memcpy(pdev->type, "IntxBLK", 8); - pdev->major = cpu_to_le64(MAJOR(dev)); - pdev->minor = cpu_to_le64(MINOR(dev)); - } else if (S_ISFIFO(mode)) { - memcpy(pdev->type, "LnxFIFO", 8); - } + io_parms.length = sizeof(pdev); + iov[1].iov_base = &pdev; + iov[1].iov_len = sizeof(pdev); rc = server->ops->sync_write(xid, &fid, &io_parms, &bytes_written, iov, 1); server->ops->close(xid, tcon, &fid); - d_drop(dentry); - /* FIXME: add code here to set EAs */ - cifs_free_open_info(&buf); + return rc; +} + +int cifs_sfu_make_node(unsigned int xid, struct inode *inode, + struct dentry *dentry, struct cifs_tcon *tcon, + const char *full_path, umode_t mode, dev_t dev) +{ + struct inode *new = NULL; + int rc; + + rc = __cifs_sfu_make_node(xid, inode, dentry, tcon, + full_path, mode, dev); + if (rc) + return rc; + + if (tcon->posix_extensions) { + rc = smb311_posix_get_inode_info(&new, full_path, NULL, + inode->i_sb, xid); + } else if (tcon->unix_ext) { + rc = cifs_get_inode_info_unix(&new, full_path, + inode->i_sb, xid); + } else { + rc = cifs_get_inode_info(&new, full_path, NULL, + inode->i_sb, xid, NULL); + } + if (!rc) + d_instantiate(dentry, new); return rc; } -- cgit v1.2.3-59-g8ed1b From 35f834265e0dc78b003aa0d1af65cafb89666b76 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 4 Apr 2024 18:06:56 -0500 Subject: smb3: fix broken reconnect when password changing on the server by allowing password rotation There are various use cases that are becoming more common in which password changes are scheduled on a server(s) periodically but the clients connected to this server need to stay connected (even in the face of brief network reconnects) due to mounts which can not be easily unmounted and mounted at will, and servers that do password rotation do not always have the ability to tell the clients exactly when to the new password will be effective, so add support for an alt password ("password2=") on mount (and also remount) so that we can anticipate the upcoming change to the server without risking breaking existing mounts. An alternative would have been to use the kernel keyring for this but the processes doing the reconnect do not have access to the keyring but do have access to the ses structure. Reviewed-by: Bharath SM Signed-off-by: Steve French --- fs/smb/client/cifsglob.h | 1 + fs/smb/client/connect.c | 8 ++++++++ fs/smb/client/fs_context.c | 21 +++++++++++++++++++++ fs/smb/client/fs_context.h | 2 ++ fs/smb/client/misc.c | 1 + fs/smb/client/smb2pdu.c | 11 +++++++++++ 6 files changed, 44 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index f6a302205f89..d6669ce4ae87 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1077,6 +1077,7 @@ struct cifs_ses { and after mount option parsing we fill it */ char *domainName; char *password; + char *password2; /* When key rotation used, new password may be set before it expires */ char workstation_name[CIFS_MAX_WORKSTATION_LEN]; struct session_key auth_key; struct ntlmssp_auth *ntlmssp; /* ciphertext, flags, server challenge */ diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 85679ae106fd..4e35970681bf 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -2183,6 +2183,7 @@ cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) } ++delim; + /* BB consider adding support for password2 (Key Rotation) for multiuser in future */ ctx->password = kstrndup(delim, len, GFP_KERNEL); if (!ctx->password) { cifs_dbg(FYI, "Unable to allocate %zd bytes for password\n", @@ -2206,6 +2207,7 @@ cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) kfree(ctx->username); ctx->username = NULL; kfree_sensitive(ctx->password); + /* no need to free ctx->password2 since not allocated in this path */ ctx->password = NULL; goto out_key_put; } @@ -2317,6 +2319,12 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) if (!ses->password) goto get_ses_fail; } + /* ctx->password freed at unmount */ + if (ctx->password2) { + ses->password2 = kstrdup(ctx->password2, GFP_KERNEL); + if (!ses->password2) + goto get_ses_fail; + } if (ctx->domainname) { ses->domainName = kstrdup(ctx->domainname, GFP_KERNEL); if (!ses->domainName) diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c index b7bfe705b2c4..6c727d8c31e8 100644 --- a/fs/smb/client/fs_context.c +++ b/fs/smb/client/fs_context.c @@ -162,6 +162,7 @@ const struct fs_parameter_spec smb3_fs_parameters[] = { fsparam_string("username", Opt_user), fsparam_string("pass", Opt_pass), fsparam_string("password", Opt_pass), + fsparam_string("password2", Opt_pass2), fsparam_string("ip", Opt_ip), fsparam_string("addr", Opt_ip), fsparam_string("domain", Opt_domain), @@ -345,6 +346,7 @@ smb3_fs_context_dup(struct smb3_fs_context *new_ctx, struct smb3_fs_context *ctx new_ctx->nodename = NULL; new_ctx->username = NULL; new_ctx->password = NULL; + new_ctx->password2 = NULL; new_ctx->server_hostname = NULL; new_ctx->domainname = NULL; new_ctx->UNC = NULL; @@ -357,6 +359,7 @@ smb3_fs_context_dup(struct smb3_fs_context *new_ctx, struct smb3_fs_context *ctx DUP_CTX_STR(prepath); DUP_CTX_STR(username); DUP_CTX_STR(password); + DUP_CTX_STR(password2); DUP_CTX_STR(server_hostname); DUP_CTX_STR(UNC); DUP_CTX_STR(source); @@ -905,6 +908,8 @@ static int smb3_reconfigure(struct fs_context *fc) else { kfree_sensitive(ses->password); ses->password = kstrdup(ctx->password, GFP_KERNEL); + kfree_sensitive(ses->password2); + ses->password2 = kstrdup(ctx->password2, GFP_KERNEL); } STEAL_STRING(cifs_sb, ctx, domainname); STEAL_STRING(cifs_sb, ctx, nodename); @@ -1305,6 +1310,18 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, goto cifs_parse_mount_err; } break; + case Opt_pass2: + kfree_sensitive(ctx->password2); + ctx->password2 = NULL; + if (strlen(param->string) == 0) + break; + + ctx->password2 = kstrdup(param->string, GFP_KERNEL); + if (ctx->password2 == NULL) { + cifs_errorf(fc, "OOM when copying password2 string\n"); + goto cifs_parse_mount_err; + } + break; case Opt_ip: if (strlen(param->string) == 0) { ctx->got_ip = false; @@ -1608,6 +1625,8 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, cifs_parse_mount_err: kfree_sensitive(ctx->password); ctx->password = NULL; + kfree_sensitive(ctx->password2); + ctx->password2 = NULL; return -EINVAL; } @@ -1713,6 +1732,8 @@ smb3_cleanup_fs_context_contents(struct smb3_fs_context *ctx) ctx->username = NULL; kfree_sensitive(ctx->password); ctx->password = NULL; + kfree_sensitive(ctx->password2); + ctx->password2 = NULL; kfree(ctx->server_hostname); ctx->server_hostname = NULL; kfree(ctx->UNC); diff --git a/fs/smb/client/fs_context.h b/fs/smb/client/fs_context.h index 8a35645e0b65..a947bddeba27 100644 --- a/fs/smb/client/fs_context.h +++ b/fs/smb/client/fs_context.h @@ -145,6 +145,7 @@ enum cifs_param { Opt_source, Opt_user, Opt_pass, + Opt_pass2, Opt_ip, Opt_domain, Opt_srcaddr, @@ -177,6 +178,7 @@ struct smb3_fs_context { char *username; char *password; + char *password2; char *domainname; char *source; char *server_hostname; diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c index 33ac4f8f5050..7d15a1969b81 100644 --- a/fs/smb/client/misc.c +++ b/fs/smb/client/misc.c @@ -98,6 +98,7 @@ sesInfoFree(struct cifs_ses *buf_to_free) kfree(buf_to_free->serverDomain); kfree(buf_to_free->serverNOS); kfree_sensitive(buf_to_free->password); + kfree_sensitive(buf_to_free->password2); kfree(buf_to_free->user_name); kfree(buf_to_free->domainName); kfree_sensitive(buf_to_free->auth_key.response); diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index c0c4933af5fc..86c647a947cc 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -367,6 +367,17 @@ again: } rc = cifs_setup_session(0, ses, server, nls_codepage); + if ((rc == -EACCES) || (rc == -EKEYEXPIRED) || (rc == -EKEYREVOKED)) { + /* + * Try alternate password for next reconnect (key rotation + * could be enabled on the server e.g.) if an alternate + * password is available and the current password is expired, + * but do not swap on non pwd related errors like host down + */ + if (ses->password2) + swap(ses->password2, ses->password); + } + if ((rc == -EACCES) && !tcon->retry) { mutex_unlock(&ses->session_mutex); rc = -EHOSTDOWN; -- cgit v1.2.3-59-g8ed1b From a8fa658eebe8b17fc852482da52f8841be8931d6 Mon Sep 17 00:00:00 2001 From: Yang Li Date: Fri, 22 Mar 2024 14:26:04 +0800 Subject: eventfs: Fix kernel-doc comments to functions This commit fix kernel-doc style comments with complete parameter descriptions for the lookup_file(),lookup_dir_entry() and lookup_file_dentry(). Link: https://lore.kernel.org/linux-trace-kernel/20240322062604.28862-1-yang.lee@linux.alibaba.com Signed-off-by: Yang Li Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index dc067eeb6387..894c6ca1e500 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -336,6 +336,7 @@ static void update_inode_attr(struct dentry *dentry, struct inode *inode, /** * lookup_file - look up a file in the tracefs filesystem + * @parent_ei: Pointer to the eventfs_inode that represents parent of the file * @dentry: the dentry to look up * @mode: the permission that the file should have. * @attr: saved attributes changed by user @@ -389,6 +390,7 @@ static struct dentry *lookup_file(struct eventfs_inode *parent_ei, /** * lookup_dir_entry - look up a dir in the tracefs filesystem * @dentry: the directory to look up + * @pei: Pointer to the parent eventfs_inode if available * @ei: the eventfs_inode that represents the directory to create * * This function will look up a dentry for a directory represented by @@ -478,16 +480,20 @@ void eventfs_d_release(struct dentry *dentry) /** * lookup_file_dentry - create a dentry for a file of an eventfs_inode + * @dentry: The parent dentry under which the new file's dentry will be created * @ei: the eventfs_inode that the file will be created under * @idx: the index into the entry_attrs[] of the @ei - * @parent: The parent dentry of the created file. - * @name: The name of the file to create * @mode: The mode of the file. * @data: The data to use to set the inode of the file with on open() * @fops: The fops of the file to be created. * - * Create a dentry for a file of an eventfs_inode @ei and place it into the - * address located at @e_dentry. + * This function creates a dentry for a file associated with an + * eventfs_inode @ei. It uses the entry attributes specified by @idx, + * if available. The file will have the specified @mode and its inode will be + * set up with @data upon open. The file operations will be set to @fops. + * + * Return: Returns a pointer to the newly created file's dentry or an error + * pointer. */ static struct dentry * lookup_file_dentry(struct dentry *dentry, -- cgit v1.2.3-59-g8ed1b From 824f06ff81464f823cd6259ff2ec8fbeceb2afa5 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Thu, 11 Apr 2024 23:36:33 +0000 Subject: fs/9p: Revert "fs/9p: fix dups even in uncached mode" This reverts commit be57855f505003c5cafff40338d5d0f23b00ba4d. It caused a regression involving duplicate inode numbers in some tester trees. The bad behavior seems to be dependent on inode reuse policy in underlying file system, so it did not trigger in my test setup. Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_super.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'fs') diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index 4236058c7bbd..eeac06c2a84c 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -244,6 +244,21 @@ done: return res; } +static int v9fs_drop_inode(struct inode *inode) +{ + struct v9fs_session_info *v9ses; + + v9ses = v9fs_inode2v9ses(inode); + if (v9ses->cache & (CACHE_META|CACHE_LOOSE)) + return generic_drop_inode(inode); + /* + * in case of non cached mode always drop the + * inode because we want the inode attribute + * to always match that on the server. + */ + return 1; +} + static int v9fs_write_inode(struct inode *inode, struct writeback_control *wbc) { @@ -278,6 +293,7 @@ static const struct super_operations v9fs_super_ops_dotl = { .alloc_inode = v9fs_alloc_inode, .free_inode = v9fs_free_inode, .statfs = v9fs_statfs, + .drop_inode = v9fs_drop_inode, .evict_inode = v9fs_evict_inode, .show_options = v9fs_show_options, .umount_begin = v9fs_umount_begin, -- cgit v1.2.3-59-g8ed1b From 7fd524b9bd1be210fe79035800f4bd78a41b349f Mon Sep 17 00:00:00 2001 From: Joakim Sindholt Date: Mon, 18 Mar 2024 12:22:32 +0100 Subject: fs/9p: drop inodes immediately on non-.L too Signed-off-by: Joakim Sindholt Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_super.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index eeac06c2a84c..55e67e36ae68 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -283,6 +283,7 @@ static const struct super_operations v9fs_super_ops = { .alloc_inode = v9fs_alloc_inode, .free_inode = v9fs_free_inode, .statfs = simple_statfs, + .drop_inode = v9fs_drop_inode, .evict_inode = v9fs_evict_inode, .show_options = v9fs_show_options, .umount_begin = v9fs_umount_begin, -- cgit v1.2.3-59-g8ed1b From 2b3e79fea66e166622a454715ce981432ac8c6e3 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 01:01:11 -0400 Subject: bcachefs: Don't use bch2_btree_node_lock_write_nofail() in btree split path It turns out - btree splits happen with the rest of the transaction still locked, to avoid unnecessary restarts, which means using nofail doesn't work here - we can deadlock. Fortunately, we now have the ability to return errors here. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 41 +++++++++++++++++++++++-------------- 1 file changed, 26 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index c4a5e83a56a4..29a5bc7d8789 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1280,23 +1280,29 @@ static void bch2_btree_set_root_inmem(struct bch_fs *c, struct btree *b) bch2_recalc_btree_reserve(c); } -static void bch2_btree_set_root(struct btree_update *as, - struct btree_trans *trans, - struct btree_path *path, - struct btree *b) +static int bch2_btree_set_root(struct btree_update *as, + struct btree_trans *trans, + struct btree_path *path, + struct btree *b, + bool nofail) { struct bch_fs *c = as->c; - struct btree *old; trace_and_count(c, btree_node_set_root, trans, b); - old = btree_node_root(c, b); + struct btree *old = btree_node_root(c, b); /* * Ensure no one is using the old root while we switch to the * new root: */ - bch2_btree_node_lock_write_nofail(trans, path, &old->c); + if (nofail) { + bch2_btree_node_lock_write_nofail(trans, path, &old->c); + } else { + int ret = bch2_btree_node_lock_write(trans, path, &old->c); + if (ret) + return ret; + } bch2_btree_set_root_inmem(c, b); @@ -1310,6 +1316,7 @@ static void bch2_btree_set_root(struct btree_update *as, * depend on the new root would have to update the new root. */ bch2_btree_node_unlock_write(trans, path, old); + return 0; } /* Interior node updates: */ @@ -1652,15 +1659,16 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans, if (parent) { /* Split a non root node */ ret = bch2_btree_insert_node(as, trans, path, parent, &as->parent_keys); - if (ret) - goto err; } else if (n3) { - bch2_btree_set_root(as, trans, trans->paths + path, n3); + ret = bch2_btree_set_root(as, trans, trans->paths + path, n3, false); } else { /* Root filled up but didn't need to be split */ - bch2_btree_set_root(as, trans, trans->paths + path, n1); + ret = bch2_btree_set_root(as, trans, trans->paths + path, n1, false); } + if (ret) + goto err; + if (n3) { bch2_btree_update_get_open_buckets(as, n3); bch2_btree_node_write(c, n3, SIX_LOCK_intent, 0); @@ -1863,7 +1871,9 @@ static void __btree_increase_depth(struct btree_update *as, struct btree_trans * bch2_keylist_add(&as->parent_keys, &b->key); btree_split_insert_keys(as, trans, path_idx, n, &as->parent_keys); - bch2_btree_set_root(as, trans, path, n); + int ret = bch2_btree_set_root(as, trans, path, n, true); + BUG_ON(ret); + bch2_btree_update_get_open_buckets(as, n); bch2_btree_node_write(c, n, SIX_LOCK_intent, 0); bch2_trans_node_add(trans, path, n); @@ -2106,12 +2116,13 @@ int bch2_btree_node_rewrite(struct btree_trans *trans, if (parent) { bch2_keylist_add(&as->parent_keys, &n->key); ret = bch2_btree_insert_node(as, trans, iter->path, parent, &as->parent_keys); - if (ret) - goto err; } else { - bch2_btree_set_root(as, trans, btree_iter_path(trans, iter), n); + ret = bch2_btree_set_root(as, trans, btree_iter_path(trans, iter), n, false); } + if (ret) + goto err; + bch2_btree_update_get_open_buckets(as, n); bch2_btree_node_write(c, n, SIX_LOCK_intent, 0); -- cgit v1.2.3-59-g8ed1b From 4f8cf60ac18bee62e8c58654a300eb44b96caf09 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 5 Mar 2024 18:52:05 +0000 Subject: reiserfs: Convert to writepages Use buffer_migrate_folio to handle folio migration instead of writing out dirty pages and reading them back in again. Use writepages to write out folios more efficiently. We now only do that wait_on_write_block check once per call to writepages instead of once per page. It would be possible to do one transaction per writeback run, but that's a bit of a big change to do to this old filesystem, so leave it as one transaction per folio (and leave reiserfs supporting only one page per folio). Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Jan Kara Signed-off-by: Jan Kara Message-Id: <20240305185208.1200166-1-willy@infradead.org> --- fs/reiserfs/inode.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 1d825459ee6e..c1daedc50f4c 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2503,8 +2503,8 @@ out: * start/recovery path as __block_write_full_folio, along with special * code to handle reiserfs tails. */ -static int reiserfs_write_full_folio(struct folio *folio, - struct writeback_control *wbc) +static int reiserfs_write_folio(struct folio *folio, + struct writeback_control *wbc, void *data) { struct inode *inode = folio->mapping->host; unsigned long end_index = inode->i_size >> PAGE_SHIFT; @@ -2721,12 +2721,11 @@ static int reiserfs_read_folio(struct file *f, struct folio *folio) return block_read_full_folio(folio, reiserfs_get_block); } -static int reiserfs_writepage(struct page *page, struct writeback_control *wbc) +static int reiserfs_writepages(struct address_space *mapping, + struct writeback_control *wbc) { - struct folio *folio = page_folio(page); - struct inode *inode = folio->mapping->host; - reiserfs_wait_on_write_block(inode->i_sb); - return reiserfs_write_full_folio(folio, wbc); + reiserfs_wait_on_write_block(mapping->host->i_sb); + return write_cache_pages(mapping, wbc, reiserfs_write_folio, NULL); } static void reiserfs_truncate_failed_write(struct inode *inode) @@ -3405,7 +3404,7 @@ out: } const struct address_space_operations reiserfs_address_space_operations = { - .writepage = reiserfs_writepage, + .writepages = reiserfs_writepages, .read_folio = reiserfs_read_folio, .readahead = reiserfs_readahead, .release_folio = reiserfs_release_folio, @@ -3415,4 +3414,5 @@ const struct address_space_operations reiserfs_address_space_operations = { .bmap = reiserfs_aop_bmap, .direct_IO = reiserfs_direct_IO, .dirty_folio = reiserfs_dirty_folio, + .migrate_folio = buffer_migrate_folio, }; -- cgit v1.2.3-59-g8ed1b From 8e2e0a79a3349b6853f572e4be4885b482f7d837 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 12 Apr 2024 17:49:42 +0800 Subject: quota: fix to propagate error of mark_dquot_dirty() to caller in order to let caller be aware of failure of mark_dquot_dirty(). Signed-off-by: Chao Yu Signed-off-by: Jan Kara Message-Id: <20240412094942.2131243-1-chao@kernel.org> --- fs/quota/dquot.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 808544f74e5e..627eb2f72ef3 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -410,7 +410,7 @@ static inline int mark_all_dquot_dirty(struct dquot __rcu * const *dquots) if (dquot) /* Even in case of error we have to continue */ ret = mark_dquot_dirty(dquot); - if (!err) + if (!err && ret < 0) err = ret; } return err; @@ -1737,7 +1737,7 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number, int flags) if (reserve) goto out_flush_warn; - mark_all_dquot_dirty(dquots); + ret = mark_all_dquot_dirty(dquots); out_flush_warn: srcu_read_unlock(&dquot_srcu, index); flush_warnings(warn); @@ -1786,7 +1786,7 @@ int dquot_alloc_inode(struct inode *inode) warn_put_all: spin_unlock(&inode->i_lock); if (ret == 0) - mark_all_dquot_dirty(dquots); + ret = mark_all_dquot_dirty(dquots); srcu_read_unlock(&dquot_srcu, index); flush_warnings(warn); return ret; @@ -1990,7 +1990,7 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to) qsize_t inode_usage = 1; struct dquot __rcu **dquots; struct dquot *transfer_from[MAXQUOTAS] = {}; - int cnt, index, ret = 0; + int cnt, index, ret = 0, err; char is_valid[MAXQUOTAS] = {}; struct dquot_warn warn_to[MAXQUOTAS]; struct dquot_warn warn_from_inodes[MAXQUOTAS]; @@ -2087,8 +2087,12 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to) * mark_all_dquot_dirty(). */ index = srcu_read_lock(&dquot_srcu); - mark_all_dquot_dirty((struct dquot __rcu **)transfer_from); - mark_all_dquot_dirty((struct dquot __rcu **)transfer_to); + err = mark_all_dquot_dirty((struct dquot __rcu **)transfer_from); + if (err < 0) + ret = err; + err = mark_all_dquot_dirty((struct dquot __rcu **)transfer_to); + if (err < 0) + ret = err; srcu_read_unlock(&dquot_srcu, index); flush_warnings(warn_to); @@ -2098,7 +2102,7 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to) for (cnt = 0; cnt < MAXQUOTAS; cnt++) if (is_valid[cnt]) transfer_to[cnt] = transfer_from[cnt]; - return 0; + return ret; over_quota: /* Back out changes we already did */ for (cnt--; cnt >= 0; cnt--) { @@ -2726,6 +2730,7 @@ static int do_set_dqblk(struct dquot *dquot, struct qc_dqblk *di) struct mem_dqblk *dm = &dquot->dq_dqb; int check_blim = 0, check_ilim = 0; struct mem_dqinfo *dqi = &sb_dqopt(dquot->dq_sb)->info[dquot->dq_id.type]; + int ret; if (di->d_fieldmask & ~VFS_QC_MASK) return -EINVAL; @@ -2807,8 +2812,9 @@ static int do_set_dqblk(struct dquot *dquot, struct qc_dqblk *di) else set_bit(DQ_FAKE_B, &dquot->dq_flags); spin_unlock(&dquot->dq_dqb_lock); - mark_dquot_dirty(dquot); - + ret = mark_dquot_dirty(dquot); + if (ret < 0) + return ret; return 0; } -- cgit v1.2.3-59-g8ed1b From 9ec2b350160104175241229ac711ffdde8c10078 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Fri, 12 Apr 2024 14:32:12 +0200 Subject: reiserfs: Trim some README bits Remove some bits of README based on authors request. References: https://lore.kernel.org/lkml/b98b29cf-27d9-49e0-b10b-1848399badfd@kittens.ph/ Signed-off-by: Jan Kara --- fs/reiserfs/README | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/README b/fs/reiserfs/README index e2f7a264e3ff..11e9ecf24b63 100644 --- a/fs/reiserfs/README +++ b/fs/reiserfs/README @@ -102,19 +102,9 @@ that start on a node aligned boundary (there are reasons to want to node align files), and he invented and implemented indirect items and unformatted nodes as the solution. -Konstantin Shvachko, with the help of the Russian version of a VC, -tried to put me in a position where I was forced into giving control -of the project to him. (Fortunately, as the person paying the money -for all salaries from my dayjob I owned all copyrights, and you can't -really force takeovers of sole proprietorships.) This was something -curious, because he never really understood the value of our project, -why we should do what we do, or why innovation was possible in -general, but he was sure that he ought to be controlling it. Every -innovation had to be forced past him while he was with us. He added -two years to the time required to complete reiserfs, and was a net -loss for me. Mikhail Gilula was a brilliant innovator who also left -in a destructive way that erased the value of his contributions, and -that he was shown much generosity just makes it more painful. +Konstantin Shvachko was taking part in the early days. + +Mikhail Gilula was a brilliant innovator that has shown much generosity. Grigory Zaigralin was an extremely effective system administrator for our group. -- cgit v1.2.3-59-g8ed1b From 7c5dffb3d90c5921b91981cc663e02757d90526e Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Sun, 7 Apr 2024 15:26:03 +0800 Subject: f2fs: compress: fix to relocate check condition in f2fs_{release,reserve}_compress_blocks() Compress flag should be checked after inode lock held to avoid racing w/ f2fs_setflags_common(), fix it. Fixes: 4c8ff7095bef ("f2fs: support data compression") Reported-by: Zhiguo Niu Closes: https://lore.kernel.org/linux-f2fs-devel/CAHJ8P3LdZXLc2rqeYjvymgYHr2+YLuJ0sLG9DdsJZmwO7deuhw@mail.gmail.com Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index bd55bbb4fa67..e4435461aeeb 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -3543,9 +3543,6 @@ static int f2fs_release_compress_blocks(struct file *filp, unsigned long arg) if (!f2fs_sb_has_compression(sbi)) return -EOPNOTSUPP; - if (!f2fs_compressed_file(inode)) - return -EINVAL; - if (f2fs_readonly(sbi->sb)) return -EROFS; @@ -3564,7 +3561,8 @@ static int f2fs_release_compress_blocks(struct file *filp, unsigned long arg) goto out; } - if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { + if (!f2fs_compressed_file(inode) || + is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { ret = -EINVAL; goto out; } @@ -3725,9 +3723,6 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg) if (!f2fs_sb_has_compression(sbi)) return -EOPNOTSUPP; - if (!f2fs_compressed_file(inode)) - return -EINVAL; - if (f2fs_readonly(sbi->sb)) return -EROFS; @@ -3739,7 +3734,8 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg) inode_lock(inode); - if (!is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { + if (!f2fs_compressed_file(inode) || + !is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { ret = -EINVAL; goto unlock_inode; } -- cgit v1.2.3-59-g8ed1b From bd9ae4ae9e585061acfd4a169f2321706f900246 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Sun, 7 Apr 2024 15:26:04 +0800 Subject: f2fs: compress: fix to relocate check condition in f2fs_ioc_{,de}compress_file() Compress flag should be checked after inode lock held to avoid racing w/ f2fs_setflags_common() , fix it. Fixes: 5fdb322ff2c2 ("f2fs: add F2FS_IOC_DECOMPRESS_FILE and F2FS_IOC_COMPRESS_FILE") Reported-by: Zhiguo Niu Closes: https://lore.kernel.org/linux-f2fs-devel/CAHJ8P3LdZXLc2rqeYjvymgYHr2+YLuJ0sLG9DdsJZmwO7deuhw@mail.gmail.com Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index e4435461aeeb..f88dff065ff1 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4136,9 +4136,6 @@ static int f2fs_ioc_decompress_file(struct file *filp) if (!(filp->f_mode & FMODE_WRITE)) return -EBADF; - if (!f2fs_compressed_file(inode)) - return -EINVAL; - f2fs_balance_fs(sbi, true); file_start_write(filp); @@ -4149,7 +4146,8 @@ static int f2fs_ioc_decompress_file(struct file *filp) goto out; } - if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { + if (!f2fs_compressed_file(inode) || + is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { ret = -EINVAL; goto out; } @@ -4214,9 +4212,6 @@ static int f2fs_ioc_compress_file(struct file *filp) if (!(filp->f_mode & FMODE_WRITE)) return -EBADF; - if (!f2fs_compressed_file(inode)) - return -EINVAL; - f2fs_balance_fs(sbi, true); file_start_write(filp); @@ -4227,7 +4222,8 @@ static int f2fs_ioc_compress_file(struct file *filp) goto out; } - if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { + if (!f2fs_compressed_file(inode) || + is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { ret = -EINVAL; goto out; } -- cgit v1.2.3-59-g8ed1b From 278a6253a673611dbc8ab72a3b34b151a8e75822 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Wed, 3 Apr 2024 22:24:19 +0800 Subject: f2fs: fix to relocate check condition in f2fs_fallocate() compress and pinfile flag should be checked after inode lock held to avoid race condition, fix it. Fixes: 4c8ff7095bef ("f2fs: support data compression") Fixes: 5fed0be8583f ("f2fs: do not allow partial truncation on pinned file") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index f88dff065ff1..fc3d4d69e72c 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1820,15 +1820,6 @@ static long f2fs_fallocate(struct file *file, int mode, (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE))) return -EOPNOTSUPP; - /* - * Pinned file should not support partial truncation since the block - * can be used by applications. - */ - if ((f2fs_compressed_file(inode) || f2fs_is_pinned_file(inode)) && - (mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE | - FALLOC_FL_ZERO_RANGE | FALLOC_FL_INSERT_RANGE))) - return -EOPNOTSUPP; - if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE | FALLOC_FL_INSERT_RANGE)) @@ -1836,6 +1827,17 @@ static long f2fs_fallocate(struct file *file, int mode, inode_lock(inode); + /* + * Pinned file should not support partial truncation since the block + * can be used by applications. + */ + if ((f2fs_compressed_file(inode) || f2fs_is_pinned_file(inode)) && + (mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE | + FALLOC_FL_ZERO_RANGE | FALLOC_FL_INSERT_RANGE))) { + ret = -EOPNOTSUPP; + goto out; + } + ret = file_modified(file); if (ret) goto out; -- cgit v1.2.3-59-g8ed1b From e07230da0500e0919a765037c5e81583b519be2c Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Wed, 3 Apr 2024 22:24:20 +0800 Subject: f2fs: fix to check pinfile flag in f2fs_move_file_range() ioctl(F2FS_IOC_MOVE_RANGE) can truncate or punch hole on pinned file, fix to disallow it. Fixes: 5fed0be8583f ("f2fs: do not allow partial truncation on pinned file") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index fc3d4d69e72c..681dfa8775ab 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2860,7 +2860,8 @@ static int f2fs_move_file_range(struct file *file_in, loff_t pos_in, goto out; } - if (f2fs_compressed_file(src) || f2fs_compressed_file(dst)) { + if (f2fs_compressed_file(src) || f2fs_compressed_file(dst) || + f2fs_is_pinned_file(src) || f2fs_is_pinned_file(dst)) { ret = -EOPNOTSUPP; goto out_unlock; } -- cgit v1.2.3-59-g8ed1b From 3bdb7f161697e2d5123b89fe1778ef17a44858e7 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Wed, 3 Apr 2024 23:07:53 +0000 Subject: f2fs: don't set RO when shutting down f2fs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Shutdown does not check the error of thaw_super due to readonly, which causes a deadlock like below. f2fs_ioc_shutdown(F2FS_GOING_DOWN_FULLSYNC) issue_discard_thread - bdev_freeze - freeze_super - f2fs_stop_checkpoint() - f2fs_handle_critical_error - sb_start_write - set RO - waiting - bdev_thaw - thaw_super_locked - return -EINVAL, if sb_rdonly() - f2fs_stop_discard_thread -> wait for kthread_stop(discard_thread); Reported-by: "Light Hsieh (謝明燈)" Reviewed-by: Daeho Jeong Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/super.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 8ac4734c2df6..df32573d1f62 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -4159,9 +4159,15 @@ void f2fs_handle_critical_error(struct f2fs_sb_info *sbi, unsigned char reason, if (shutdown) set_sbi_flag(sbi, SBI_IS_SHUTDOWN); - /* continue filesystem operators if errors=continue */ - if (continue_fs || f2fs_readonly(sb)) + /* + * Continue filesystem operators if errors=continue. Should not set + * RO by shutdown, since RO bypasses thaw_super which can hang the + * system. + */ + if (continue_fs || f2fs_readonly(sb) || shutdown) { + f2fs_warn(sbi, "Stopped filesystem due to reason: %d", reason); return; + } f2fs_warn(sbi, "Remounting filesystem read-only"); /* -- cgit v1.2.3-59-g8ed1b From b084403cfc3295b59a1b6bcc94efaf870fc3c2c9 Mon Sep 17 00:00:00 2001 From: Daeho Jeong Date: Tue, 9 Apr 2024 16:34:11 -0700 Subject: f2fs: write missing last sum blk of file pinning section While do not allocating a new section in advance for file pinning area, I missed that we should write the sum block for the last segment of a file pinning section. Fixes: 9703d69d9d15 ("f2fs: support file pinning for zoned devices") Signed-off-by: Daeho Jeong Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 4fd76e867e0a..6474b7338e81 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3559,6 +3559,8 @@ int f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, if (segment_full) { if (type == CURSEG_COLD_DATA_PINNED && !((curseg->segno + 1) % sbi->segs_per_sec)) { + write_sum_page(sbi, curseg->sum_blk, + GET_SUM_BLOCK(sbi, curseg->segno)); reset_curseg_fields(curseg); goto skip_new_segment; } -- cgit v1.2.3-59-g8ed1b From fa18d87cb20fbe597047704babae3850a7a65219 Mon Sep 17 00:00:00 2001 From: Zhiguo Niu Date: Wed, 20 Mar 2024 14:22:16 +0800 Subject: f2fs: add REQ_TIME time update for some user behaviors some user behaviors requested filesystem operations, which will cause filesystem not idle. Meanwhile adjust some f2fs_update_time(REQ_TIME) positions. Signed-off-by: Zhiguo Niu Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 681dfa8775ab..0d2626a6a21f 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2375,13 +2375,14 @@ static bool uuid_is_nonzero(__u8 u[16]) static int f2fs_ioc_set_encryption_policy(struct file *filp, unsigned long arg) { struct inode *inode = file_inode(filp); + int ret; if (!f2fs_sb_has_encrypt(F2FS_I_SB(inode))) return -EOPNOTSUPP; + ret = fscrypt_ioctl_set_policy(filp, (const void __user *)arg); f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); - - return fscrypt_ioctl_set_policy(filp, (const void __user *)arg); + return ret; } static int f2fs_ioc_get_encryption_policy(struct file *filp, unsigned long arg) @@ -2809,7 +2810,8 @@ static int f2fs_ioc_defragment(struct file *filp, unsigned long arg) err = f2fs_defragment_range(sbi, filp, &range); mnt_drop_write_file(filp); - f2fs_update_time(sbi, REQ_TIME); + if (range.len) + f2fs_update_time(sbi, REQ_TIME); if (err < 0) return err; @@ -3622,6 +3624,8 @@ static int f2fs_release_compress_blocks(struct file *filp, unsigned long arg) filemap_invalidate_unlock(inode->i_mapping); f2fs_up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); out: + if (released_blocks) + f2fs_update_time(sbi, REQ_TIME); inode_unlock(inode); mnt_drop_write_file(filp); @@ -3790,6 +3794,8 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg) f2fs_mark_inode_dirty_sync(inode, true); } unlock_inode: + if (reserved_blocks) + f2fs_update_time(sbi, REQ_TIME); inode_unlock(inode); mnt_drop_write_file(filp); @@ -3986,6 +3992,7 @@ static int f2fs_sec_trim_file(struct file *filp, unsigned long arg) if (len) ret = f2fs_secure_erase(prev_bdev, inode, prev_index, prev_block, len, range.flags); + f2fs_update_time(sbi, REQ_TIME); out: filemap_invalidate_unlock(mapping); f2fs_up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); @@ -4193,6 +4200,7 @@ static int f2fs_ioc_decompress_file(struct file *filp) if (ret) f2fs_warn(sbi, "%s: The file might be partially decompressed (errno=%d). Please delete the file.", __func__, ret); + f2fs_update_time(sbi, REQ_TIME); out: inode_unlock(inode); file_end_write(filp); @@ -4270,6 +4278,7 @@ static int f2fs_ioc_compress_file(struct file *filp) if (ret) f2fs_warn(sbi, "%s: The file might be partially compressed (errno=%d). Please delete the file.", __func__, ret); + f2fs_update_time(sbi, REQ_TIME); out: inode_unlock(inode); file_end_write(filp); -- cgit v1.2.3-59-g8ed1b From 16778aea91869756cad3e07c9ce8ef32a660358c Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Wed, 27 Mar 2024 01:27:02 +0000 Subject: f2fs: use folio_test_writeback Let's convert PageWriteback to folio_test_writeback. Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 2 +- fs/f2fs/data.c | 3 +-- fs/f2fs/f2fs.h | 2 +- fs/f2fs/gc.c | 2 +- fs/f2fs/inline.c | 2 +- fs/f2fs/inode.c | 3 ++- fs/f2fs/node.c | 2 +- fs/f2fs/segment.c | 10 +++++----- 8 files changed, 13 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index 8892c8262141..d67c471ab5df 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -1484,7 +1484,7 @@ continue_unlock: if (!PageDirty(cc->rpages[i])) goto continue_unlock; - if (PageWriteback(cc->rpages[i])) { + if (folio_test_writeback(page_folio(cc->rpages[i]))) { if (wbc->sync_mode == WB_SYNC_NONE) goto continue_unlock; f2fs_wait_on_page_writeback(cc->rpages[i], DATA, true, true); diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 0d88649c60a5..5d641fac02ba 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2704,8 +2704,7 @@ got_it: if (err) { if (fscrypt_inode_uses_fs_layer_crypto(inode)) fscrypt_finalize_bounce_page(&fio->encrypted_page); - if (PageWriteback(page)) - end_page_writeback(page); + end_page_writeback(page); } else { set_inode_flag(inode, FI_UPDATE_WRITE); } diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index e9ef971f4dba..dd530dc70005 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -4660,7 +4660,7 @@ static inline void f2fs_truncate_meta_inode_pages(struct f2fs_sb_info *sbi, page = find_get_page(META_MAPPING(sbi), blkaddr + i); if (page) { - if (PageWriteback(page)) + if (folio_test_writeback(page_folio(page))) need_submit = true; f2fs_put_page(page, 0); } diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 8852814dab7f..ac4cbbe50c2f 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -1434,7 +1434,7 @@ static int move_data_page(struct inode *inode, block_t bidx, int gc_type, goto out; if (gc_type == BG_GC) { - if (PageWriteback(page)) { + if (folio_test_writeback(page_folio(page))) { err = -EAGAIN; goto out; } diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index ac00423f117b..3d3218a4b29d 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -164,7 +164,7 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page) return -EFSCORRUPTED; } - f2fs_bug_on(F2FS_P_SB(page), PageWriteback(page)); + f2fs_bug_on(F2FS_P_SB(page), folio_test_writeback(page_folio(page))); f2fs_do_read_inline_data(page, dn->inode_page); set_page_dirty(page); diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 12b1fef31f43..d7a5a88a1a5e 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -161,7 +161,8 @@ bool f2fs_inode_chksum_verify(struct f2fs_sb_info *sbi, struct page *page) if (!f2fs_enable_inode_chksum(sbi, page)) #else if (!f2fs_enable_inode_chksum(sbi, page) || - PageDirty(page) || PageWriteback(page)) + PageDirty(page) || + folio_test_writeback(page_folio(page))) #endif return true; diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index bb57bbaff7b4..3b9eb5693683 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -1743,7 +1743,7 @@ int f2fs_move_node_page(struct page *node_page, int gc_type) goto release_page; } else { /* set page dirty and write it */ - if (!PageWriteback(node_page)) + if (!folio_test_writeback(page_folio(node_page))) set_page_dirty(node_page); } out_page: diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 6474b7338e81..f0da516ba8dc 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3614,13 +3614,13 @@ skip_new_segment: mutex_unlock(&curseg->curseg_mutex); f2fs_up_read(&SM_I(sbi)->curseg_lock); return 0; + out_err: *new_blkaddr = NULL_ADDR; up_write(&sit_i->sentry_lock); mutex_unlock(&curseg->curseg_mutex); f2fs_up_read(&SM_I(sbi)->curseg_lock); return ret; - } void f2fs_update_device_state(struct f2fs_sb_info *sbi, nid_t ino, @@ -3662,8 +3662,7 @@ static void do_write_page(struct f2fs_summary *sum, struct f2fs_io_info *fio) &fio->new_blkaddr, sum, type, fio)) { if (fscrypt_inode_uses_fs_layer_crypto(fio->page->mapping->host)) fscrypt_finalize_bounce_page(&fio->encrypted_page); - if (PageWriteback(fio->page)) - end_page_writeback(fio->page); + end_page_writeback(fio->page); if (f2fs_in_warm_node_list(fio->sbi, fio->page)) f2fs_del_fsync_node_entry(fio->sbi, fio->page); goto out; @@ -3906,7 +3905,7 @@ void f2fs_replace_block(struct f2fs_sb_info *sbi, struct dnode_of_data *dn, void f2fs_wait_on_page_writeback(struct page *page, enum page_type type, bool ordered, bool locked) { - if (PageWriteback(page)) { + if (folio_test_writeback(page_folio(page))) { struct f2fs_sb_info *sbi = F2FS_P_SB(page); /* submit cached LFS IO */ @@ -3915,7 +3914,8 @@ void f2fs_wait_on_page_writeback(struct page *page, f2fs_submit_merged_ipu_write(sbi, NULL, page); if (ordered) { wait_on_page_writeback(page); - f2fs_bug_on(sbi, locked && PageWriteback(page)); + f2fs_bug_on(sbi, locked && + folio_test_writeback(page_folio(page))); } else { wait_for_stable_page(page); } -- cgit v1.2.3-59-g8ed1b From 89ea21d70d9c9968dfd10c7e30520d0f03d465c2 Mon Sep 17 00:00:00 2001 From: Tim Schumacher Date: Thu, 28 Mar 2024 21:50:32 +0100 Subject: efivarfs: Remove unused internal struct members The structure was moved to the efivarfs internals in commit 2d82e6227ea1 ("efi: vars: Move efivar caching layer into efivarfs") after previously being used as the data ABI for efivars until its removal in commit 0f5b2c69a4cb ("efi: vars: Remove deprecated 'efivars' sysfs interface"). As efivarfs only uses the structure for the variable name caching layer, the data-related members were never in use. Remove them to avoid implying that efivarfs is bound by the same restrictions that efivars once had. While at it, remove the packed attribute, since we no longer have to guarantee a stable layout. Signed-off-by: Tim Schumacher Signed-off-by: Ard Biesheuvel --- fs/efivarfs/internal.h | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/efivarfs/internal.h b/fs/efivarfs/internal.h index f7206158ee81..d71d2e08422f 100644 --- a/fs/efivarfs/internal.h +++ b/fs/efivarfs/internal.h @@ -24,11 +24,8 @@ struct efivarfs_fs_info { struct efi_variable { efi_char16_t VariableName[EFI_VAR_NAME_LEN/sizeof(efi_char16_t)]; efi_guid_t VendorGuid; - unsigned long DataSize; - __u8 Data[1024]; - efi_status_t Status; __u32 Attributes; -} __attribute__((packed)); +}; struct efivar_entry { struct efi_variable var; -- cgit v1.2.3-59-g8ed1b From cda30c6542c8bb445bc84f6616cac8d012547f0a Mon Sep 17 00:00:00 2001 From: Tim Schumacher Date: Thu, 28 Mar 2024 21:50:33 +0100 Subject: efi: Clear up misconceptions about a maximum variable name size The UEFI specification does not make any mention of a maximum variable name size, so the headers and implementation shouldn't claim that one exists either. Comments referring to this limit have been removed or rewritten, as this is an implementation detail local to the Linux kernel. Where appropriate, the magic value of 1024 has been replaced with EFI_VAR_NAME_LEN, as this is used for the efi_variable struct definition. This in itself does not change any behavior, but should serve as points of interest when making future changes in the same area. A related build-time check has been added to ensure that the special 512 byte sized buffer will not overflow with a potentially decreased EFI_VAR_NAME_LEN. Signed-off-by: Tim Schumacher Signed-off-by: Ard Biesheuvel --- drivers/firmware/efi/vars.c | 2 +- fs/efivarfs/vars.c | 5 +++-- include/linux/efi.h | 9 ++++----- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/drivers/firmware/efi/vars.c b/drivers/firmware/efi/vars.c index f654e6f6af87..4056ba7f3440 100644 --- a/drivers/firmware/efi/vars.c +++ b/drivers/firmware/efi/vars.c @@ -215,7 +215,7 @@ efi_status_t efivar_set_variable_locked(efi_char16_t *name, efi_guid_t *vendor, if (data_size > 0) { status = check_var_size(nonblocking, attr, - data_size + ucs2_strsize(name, 1024)); + data_size + ucs2_strsize(name, EFI_VAR_NAME_LEN)); if (status != EFI_SUCCESS) return status; } diff --git a/fs/efivarfs/vars.c b/fs/efivarfs/vars.c index 4d722af1014f..3cc89bb624f0 100644 --- a/fs/efivarfs/vars.c +++ b/fs/efivarfs/vars.c @@ -295,9 +295,9 @@ static bool variable_is_present(efi_char16_t *variable_name, efi_guid_t *vendor, unsigned long strsize1, strsize2; bool found = false; - strsize1 = ucs2_strsize(variable_name, 1024); + strsize1 = ucs2_strsize(variable_name, EFI_VAR_NAME_LEN); list_for_each_entry_safe(entry, n, head, list) { - strsize2 = ucs2_strsize(entry->var.VariableName, 1024); + strsize2 = ucs2_strsize(entry->var.VariableName, EFI_VAR_NAME_LEN); if (strsize1 == strsize2 && !memcmp(variable_name, &(entry->var.VariableName), strsize2) && @@ -396,6 +396,7 @@ int efivar_init(int (*func)(efi_char16_t *, efi_guid_t, unsigned long, void *, do { variable_name_size = 512; + BUILD_BUG_ON(EFI_VAR_NAME_LEN < 512); status = efivar_get_next_variable(&variable_name_size, variable_name, diff --git a/include/linux/efi.h b/include/linux/efi.h index d59b0947fba0..418e555459da 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -1072,12 +1072,11 @@ static inline u64 efivar_reserved_space(void) { return 0; } #endif /* - * The maximum size of VariableName + Data = 1024 - * Therefore, it's reasonable to save that much - * space in each part of the structure, - * and we use a page for reading/writing. + * There is no actual upper limit specified for the variable name size. + * + * This limit exists only for practical purposes, since name conversions + * are bounds-checked and name data is occasionally stored in-line. */ - #define EFI_VAR_NAME_LEN 1024 int efivars_register(struct efivars *efivars, -- cgit v1.2.3-59-g8ed1b From 42bd2af5950456d46fdaa91c3a8fb02e680f19f5 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 10 Apr 2024 17:10:12 -0700 Subject: vfs: relax linkat() AT_EMPTY_PATH - aka flink() - requirements MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit "The definition of insanity is doing the same thing over and over again and expecting different results” We've tried to do this before, most recently with commit bb2314b47996 ("fs: Allow unprivileged linkat(..., AT_EMPTY_PATH) aka flink") about a decade ago. But the effort goes back even further than that, eg this thread back from 1998 that is so old that we don't even have it archived in lore: https://lkml.org/lkml/1998/3/10/108 which also points out some of the reasons why it's dangerous. Or, how about then in 2003: https://lkml.org/lkml/2003/4/6/112 where we went through some of the same arguments, just wirh different people involved. In particular, having access to a file descriptor does not necessarily mean that you have access to the path that was used for lookup, and there may be very good reasons why you absolutely must not have access to a path to said file. For example, if we were passed a file descriptor from the outside into some limited environment (think chroot, but also user namespaces etc) a 'flink()' system call could now make that file visible inside a context where it's not supposed to be visible. In the process the user may also be able to re-open it with permissions that the original file descriptor did not have (eg a read-only file descriptor may be associated with an underlying file that is writable). Another variation on this is if somebody else (typically root) opens a file in a directory that is not accessible to others, and passes the file descriptor on as a read-only file. Again, the access to the file descriptor does not imply that you should have access to a path to the file in the filesystem. So while we have tried this several times in the past, it never works. The last time we did this, that commit bb2314b47996 quickly got reverted again in commit f0cc6ffb8ce8 (Revert "fs: Allow unprivileged linkat(..., AT_EMPTY_PATH) aka flink"), with a note saying "We may re-do this once the whole discussion about the interface is done". Well, the discussion is long done, and didn't come to any resolution. There's no question that 'flink()' would be a useful operation, but it's a dangerous one. However, it does turn out that since 2008 (commit d76b0d9b2d87: "CRED: Use creds in file structs") we have had a fairly straightforward way to check whether the file descriptor was opened by the same credentials as the credentials of the flink(). That allows the most common patterns that people want to use, which tend to be to either open the source carefully (ie using the openat2() RESOLVE_xyz flags, and/or checking ownership with fstat() before linking), or to use O_TMPFILE and fill in the file contents before it's exposed to the world with linkat(). But it also means that if the file descriptor was opened by somebody else, or we've gone through a credentials change since, the operation no longer works (unless we have CAP_DAC_READ_SEARCH capabilities in the opener's user namespace, as before). Note that the credential equality check is done by using pointer equality, which means that it's not enough that you have effectively the same user - they have to be literally identical, since our credentials are using copy-on-write semantics. So you can't change your credentials to something else and try to change it back to the same ones between the open() and the linkat(). This is not meant to be some kind of generic permission check, this is literally meant as a "the open and link calls are 'atomic' wrt user credentials" check. It also means that you can't just move things between namespaces, because the credentials aren't just a list of uid's and gid's: they includes the pointer to the user_ns that the capabilities are relative to. So let's try this one more time and see if maybe this approach ends up being workable after all. Cc: Andrew Lutomirski Cc: Al Viro Cc: Christian Brauner Cc: Peter Anvin Cc: Jan Kara Signed-off-by: Linus Torvalds Link: https://lore.kernel.org/r/20240411001012.12513-1-torvalds@linux-foundation.org [brauner: relax capability check to opener of the file] Link: https://lore.kernel.org/all/20231113-undenkbar-gediegen-efde5f1c34bc@brauner Signed-off-by: Christian Brauner --- fs/namei.c | 19 +++++++++++++------ include/linux/namei.h | 1 + 2 files changed, 14 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/namei.c b/fs/namei.c index ceb9ddf8dfdd..8c1ff0b388e2 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2422,6 +2422,14 @@ static const char *path_init(struct nameidata *nd, unsigned flags) if (!f.file) return ERR_PTR(-EBADF); + if (flags & LOOKUP_LINKAT_EMPTY) { + if (f.file->f_cred != current_cred() && + !ns_capable(f.file->f_cred->user_ns, CAP_DAC_READ_SEARCH)) { + fdput(f); + return ERR_PTR(-ENOENT); + } + } + dentry = f.file->f_path.dentry; if (*s && unlikely(!d_can_lookup(dentry))) { @@ -4644,14 +4652,13 @@ int do_linkat(int olddfd, struct filename *old, int newdfd, goto out_putnames; } /* - * To use null names we require CAP_DAC_READ_SEARCH + * To use null names we require CAP_DAC_READ_SEARCH or + * that the open-time creds of the dfd matches current. * This ensures that not everyone will be able to create - * handlink using the passed filedescriptor. + * a hardlink using the passed file descriptor. */ - if (flags & AT_EMPTY_PATH && !capable(CAP_DAC_READ_SEARCH)) { - error = -ENOENT; - goto out_putnames; - } + if (flags & AT_EMPTY_PATH) + how |= LOOKUP_LINKAT_EMPTY; if (flags & AT_SYMLINK_FOLLOW) how |= LOOKUP_FOLLOW; diff --git a/include/linux/namei.h b/include/linux/namei.h index 74e0cc14ebf8..967aa9ea9f96 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -44,6 +44,7 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT}; #define LOOKUP_BENEATH 0x080000 /* No escaping from starting point. */ #define LOOKUP_IN_ROOT 0x100000 /* Treat dirfd as fs root. */ #define LOOKUP_CACHED 0x200000 /* Only do cached lookup */ +#define LOOKUP_LINKAT_EMPTY 0x400000 /* Linkat request with empty path. */ /* LOOKUP_* flags which do scope-related checks based on the dirfd. */ #define LOOKUP_IS_SCOPED (LOOKUP_BENEATH | LOOKUP_IN_ROOT) -- cgit v1.2.3-59-g8ed1b From 58caa786f1c02fd84919fb6db9eaecb22e8f7983 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 17:47:42 -0400 Subject: bcachefs: Fix UAFs of btree_insert_entry array The btree paths array is now dynamically resizable - and as well the btree_insert_entries array, as it needs to be the same size. The merge path (and interior update path) allocates new btree paths, thus can trigger a resize; thus we need to not retain direct pointers after invoking merge; similarly when running btree node triggers. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_trans_commit.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index aa9da4970740..9609ad370d71 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -499,9 +499,8 @@ static int run_one_trans_trigger(struct btree_trans *trans, struct btree_insert_ } static int run_btree_triggers(struct btree_trans *trans, enum btree_id btree_id, - struct btree_insert_entry *btree_id_start) + unsigned btree_id_start) { - struct btree_insert_entry *i; bool trans_trigger_run; int ret, overwrite; @@ -514,13 +513,13 @@ static int run_btree_triggers(struct btree_trans *trans, enum btree_id btree_id, do { trans_trigger_run = false; - for (i = btree_id_start; - i < trans->updates + trans->nr_updates && i->btree_id <= btree_id; + for (unsigned i = btree_id_start; + i < trans->nr_updates && trans->updates[i].btree_id <= btree_id; i++) { - if (i->btree_id != btree_id) + if (trans->updates[i].btree_id != btree_id) continue; - ret = run_one_trans_trigger(trans, i, overwrite); + ret = run_one_trans_trigger(trans, trans->updates + i, overwrite); if (ret < 0) return ret; if (ret) @@ -534,8 +533,7 @@ static int run_btree_triggers(struct btree_trans *trans, enum btree_id btree_id, static int bch2_trans_commit_run_triggers(struct btree_trans *trans) { - struct btree_insert_entry *btree_id_start = trans->updates; - unsigned btree_id = 0; + unsigned btree_id = 0, btree_id_start = 0; int ret = 0; /* @@ -549,8 +547,8 @@ static int bch2_trans_commit_run_triggers(struct btree_trans *trans) if (btree_id == BTREE_ID_alloc) continue; - while (btree_id_start < trans->updates + trans->nr_updates && - btree_id_start->btree_id < btree_id) + while (btree_id_start < trans->nr_updates && + trans->updates[btree_id_start].btree_id < btree_id) btree_id_start++; ret = run_btree_triggers(trans, btree_id, btree_id_start); @@ -558,11 +556,13 @@ static int bch2_trans_commit_run_triggers(struct btree_trans *trans) return ret; } - trans_for_each_update(trans, i) { + for (unsigned idx = 0; idx < trans->nr_updates; idx++) { + struct btree_insert_entry *i = trans->updates + idx; + if (i->btree_id > BTREE_ID_alloc) break; if (i->btree_id == BTREE_ID_alloc) { - ret = run_btree_triggers(trans, BTREE_ID_alloc, i); + ret = run_btree_triggers(trans, BTREE_ID_alloc, idx); if (ret) return ret; break; @@ -826,7 +826,8 @@ static inline int do_bch2_trans_commit(struct btree_trans *trans, unsigned flags struct bch_fs *c = trans->c; int ret = 0, u64s_delta = 0; - trans_for_each_update(trans, i) { + for (unsigned idx = 0; idx < trans->nr_updates; idx++) { + struct btree_insert_entry *i = trans->updates + idx; if (i->cached) continue; -- cgit v1.2.3-59-g8ed1b From 031ad9e7dbd18c63e671fc4d98be3082189b8a63 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 21:30:43 -0400 Subject: bcachefs: Check for packed bkeys that are too big add missing validation; fixes assertion pop in bkey unpack Signed-off-by: Kent Overstreet --- fs/bcachefs/bkey.h | 6 ++++++ fs/bcachefs/btree_io.c | 15 ++++++++------- 2 files changed, 14 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bkey.h b/fs/bcachefs/bkey.h index cf23ff47bed8..3a45d128f608 100644 --- a/fs/bcachefs/bkey.h +++ b/fs/bcachefs/bkey.h @@ -314,6 +314,12 @@ static inline unsigned bkeyp_key_u64s(const struct bkey_format *format, return bkey_packed(k) ? format->key_u64s : BKEY_U64s; } +static inline bool bkeyp_u64s_valid(const struct bkey_format *f, + const struct bkey_packed *k) +{ + return ((unsigned) k->u64s - bkeyp_key_u64s(f, k) <= U8_MAX - BKEY_U64s); +} + static inline unsigned bkeyp_key_bytes(const struct bkey_format *format, const struct bkey_packed *k) { diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index d7de82ac3893..b1aaf0550644 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -831,7 +831,7 @@ static int bset_key_invalid(struct bch_fs *c, struct btree *b, (rw == WRITE ? bch2_bkey_val_invalid(c, k, READ, err) : 0); } -static bool __bkey_valid(struct bch_fs *c, struct btree *b, +static bool bkey_packed_valid(struct bch_fs *c, struct btree *b, struct bset *i, struct bkey_packed *k) { if (bkey_p_next(k) > vstruct_last(i)) @@ -840,7 +840,7 @@ static bool __bkey_valid(struct bch_fs *c, struct btree *b, if (k->format > KEY_FORMAT_CURRENT) return false; - if (k->u64s < bkeyp_key_u64s(&b->format, k)) + if (!bkeyp_u64s_valid(&b->format, k)) return false; struct printbuf buf = PRINTBUF; @@ -884,11 +884,13 @@ static int validate_bset_keys(struct bch_fs *c, struct btree *b, "invalid bkey format %u", k->format)) goto drop_this_key; - if (btree_err_on(k->u64s < bkeyp_key_u64s(&b->format, k), + if (btree_err_on(!bkeyp_u64s_valid(&b->format, k), -BCH_ERR_btree_node_read_err_fixable, c, NULL, b, i, btree_node_bkey_bad_u64s, - "k->u64s too small (%u < %u)", k->u64s, bkeyp_key_u64s(&b->format, k))) + "bad k->u64s %u (min %u max %lu)", k->u64s, + bkeyp_key_u64s(&b->format, k), + U8_MAX - BKEY_U64s + bkeyp_key_u64s(&b->format, k))) goto drop_this_key; if (!write) @@ -947,13 +949,12 @@ drop_this_key: * do */ - if (!__bkey_valid(c, b, i, (void *) ((u64 *) k + next_good_key))) { + if (!bkey_packed_valid(c, b, i, (void *) ((u64 *) k + next_good_key))) { for (next_good_key = 1; next_good_key < (u64 *) vstruct_last(i) - (u64 *) k; next_good_key++) - if (__bkey_valid(c, b, i, (void *) ((u64 *) k + next_good_key))) + if (bkey_packed_valid(c, b, i, (void *) ((u64 *) k + next_good_key))) goto got_good_key; - } /* -- cgit v1.2.3-59-g8ed1b From 87cb0239c87f608fd48fb50f9f53f129dcfd73f4 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 23:38:07 -0400 Subject: bcachefs: btree node scan: handle encrypted nodes Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_node_scan.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index 556f76f5c84e..20f2b37c4474 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -133,9 +133,19 @@ static void try_read_btree_node(struct find_btree_nodes *f, struct bch_dev *ca, if (le64_to_cpu(bn->magic) != bset_magic(c)) return; + if (bch2_csum_type_is_encryption(BSET_CSUM_TYPE(&bn->keys))) { + struct nonce nonce = btree_nonce(&bn->keys, 0); + unsigned bytes = (void *) &bn->keys - (void *) &bn->flags; + + bch2_encrypt(c, BSET_CSUM_TYPE(&bn->keys), nonce, &bn->flags, bytes); + } + if (btree_id_is_alloc(BTREE_NODE_ID(bn))) return; + if (BTREE_NODE_LEVEL(bn) >= BTREE_MAX_DEPTH) + return; + rcu_read_lock(); struct found_btree_node n = { .btree_id = BTREE_NODE_ID(bn), -- cgit v1.2.3-59-g8ed1b From dc32c118ec6b1032693c489a0aa9e011f0acdb1a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 21:20:27 -0400 Subject: bcachefs: fix unsafety in bch2_extent_ptr_to_text() Need to check if we have a valid bucket before checking if ptr is stale Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 0e3ca99fbd2d..36d12d2adb81 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -998,7 +998,9 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struc prt_str(out, " cached"); if (ptr->unwritten) prt_str(out, " unwritten"); - if (ca && ptr_stale(ca, ptr)) + if (b >= ca->mi.first_bucket && + b < ca->mi.nbuckets && + ptr_stale(ca, ptr)) prt_printf(out, " stale"); } } -- cgit v1.2.3-59-g8ed1b From 2aeed876d7c2c2fb4c6b92f59bdf7e73cfe5e098 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 23:37:24 -0400 Subject: bcachefs: fix unsafety in bch2_stripe_to_text() .to_text() functions need to work on key values that didn't pass .valid Signed-off-by: Kent Overstreet --- fs/bcachefs/ec.c | 46 +++++++++++++++++++++++++--------------------- fs/bcachefs/ec.h | 2 ++ 2 files changed, 27 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 082075244e16..c0b2b2f180e0 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -131,29 +131,33 @@ fsck_err: void bch2_stripe_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) { - const struct bch_stripe *s = bkey_s_c_to_stripe(k).v; - unsigned i, nr_data = s->nr_blocks - s->nr_redundant; + const struct bch_stripe *sp = bkey_s_c_to_stripe(k).v; + struct bch_stripe s = {}; + + memcpy(&s, sp, min(sizeof(s), bkey_val_bytes(k.k))); + + unsigned nr_data = s.nr_blocks - s.nr_redundant; prt_printf(out, "algo %u sectors %u blocks %u:%u csum %u gran %u", - s->algorithm, - le16_to_cpu(s->sectors), - nr_data, - s->nr_redundant, - s->csum_type, - 1U << s->csum_granularity_bits); - - for (i = 0; i < s->nr_blocks; i++) { - const struct bch_extent_ptr *ptr = s->ptrs + i; - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); - u32 offset; - u64 b = sector_to_bucket_and_offset(ca, ptr->offset, &offset); - - prt_printf(out, " %u:%llu:%u", ptr->dev, b, offset); - if (i < nr_data) - prt_printf(out, "#%u", stripe_blockcount_get(s, i)); - prt_printf(out, " gen %u", ptr->gen); - if (ptr_stale(ca, ptr)) - prt_printf(out, " stale"); + s.algorithm, + le16_to_cpu(s.sectors), + nr_data, + s.nr_redundant, + s.csum_type, + 1U << s.csum_granularity_bits); + + for (unsigned i = 0; i < s.nr_blocks; i++) { + const struct bch_extent_ptr *ptr = sp->ptrs + i; + + if ((void *) ptr >= bkey_val_end(k)) + break; + + bch2_extent_ptr_to_text(out, c, ptr); + + if (s.csum_type < BCH_CSUM_NR && + i < nr_data && + stripe_blockcount_offset(&s, i) < bkey_val_bytes(k.k)) + prt_printf(out, "#%u", stripe_blockcount_get(sp, i)); } } diff --git a/fs/bcachefs/ec.h b/fs/bcachefs/ec.h index f4369b02e805..f042616888b0 100644 --- a/fs/bcachefs/ec.h +++ b/fs/bcachefs/ec.h @@ -32,6 +32,8 @@ static inline unsigned stripe_csums_per_device(const struct bch_stripe *s) static inline unsigned stripe_csum_offset(const struct bch_stripe *s, unsigned dev, unsigned csum_idx) { + EBUG_ON(s->csum_type >= BCH_CSUM_NR); + unsigned csum_bytes = bch_crc_bytes[s->csum_type]; return sizeof(struct bch_stripe) + -- cgit v1.2.3-59-g8ed1b From 7b4c4ccf848b359762c1525bab5a12f008f14a65 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 23:58:36 -0400 Subject: bcachefs: fix race in bch2_btree_node_evict() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 84474324dba9..c7f156320a35 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -1148,6 +1148,8 @@ wait_on_io: btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_intent); btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_write); + if (unlikely(b->hash_val != btree_ptr_hash_val(k))) + goto out; if (btree_node_dirty(b)) { __bch2_btree_node_write(c, b, BTREE_WRITE_cache_reclaim); @@ -1162,7 +1164,7 @@ wait_on_io: btree_node_data_free(c, b); bch2_btree_node_hash_remove(bc, b); mutex_unlock(&bc->lock); - +out: six_unlock_write(&b->c.lock); six_unlock_intent(&b->c.lock); } -- cgit v1.2.3-59-g8ed1b From ba8ed36e72033dd6b89d44145f323e6c4508bfc9 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 00:09:08 -0400 Subject: bcachefs: don't queue btree nodes for rewrites during scan many nodes found during scan will be old nodes, overwritten by newer nodes Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_io.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index b1aaf0550644..9678b2375bed 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1340,7 +1340,9 @@ start: rb->start_time); bio_put(&rb->bio); - if (saw_error && !btree_node_read_error(b)) { + if (saw_error && + !btree_node_read_error(b) && + c->curr_recovery_pass != BCH_RECOVERY_PASS_scan_for_btree_nodes) { printbuf_reset(&buf); bch2_bpos_to_text(&buf, b->key.k.p); bch_err_ratelimited(c, "%s: rewriting btree node at btree=%s level=%u %s due to error", -- cgit v1.2.3-59-g8ed1b From 9abb6dd7ce5a261f7aebf6f396b50a63db71133f Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 15:17:00 -0400 Subject: bcachefs: Standardize helpers for printing enum strs with bounds checks Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 4 ++-- fs/bcachefs/buckets.h | 8 -------- fs/bcachefs/checksum.c | 23 ++++++++++++++--------- fs/bcachefs/checksum.h | 5 +++-- fs/bcachefs/compress.h | 8 -------- fs/bcachefs/ec.c | 14 ++++++-------- fs/bcachefs/extents.c | 7 ++++--- fs/bcachefs/journal_io.c | 17 +++++++++-------- fs/bcachefs/opts.c | 29 +++++++++++++++++++++++++---- fs/bcachefs/opts.h | 10 ++++++---- 10 files changed, 69 insertions(+), 56 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 364ae42022af..9480f1b44f07 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -1314,7 +1314,7 @@ static inline __u64 __bset_magic(struct bch_sb *sb) x(write_buffer_keys, 11) \ x(datetime, 12) -enum { +enum bch_jset_entry_type { #define x(f, nr) BCH_JSET_ENTRY_##f = nr, BCH_JSET_ENTRY_TYPES() #undef x @@ -1360,7 +1360,7 @@ struct jset_entry_blacklist_v2 { x(inodes, 1) \ x(key_version, 2) -enum { +enum bch_fs_usage_type { #define x(f, nr) BCH_FS_USAGE_##f = nr, BCH_FS_USAGE_TYPES() #undef x diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index 00aaf4bb5139..f9af5adabe83 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -395,14 +395,6 @@ static inline const char *bch2_data_type_str(enum bch_data_type type) : "(invalid data type)"; } -static inline void bch2_prt_data_type(struct printbuf *out, enum bch_data_type type) -{ - if (type < BCH_DATA_NR) - prt_str(out, __bch2_data_types[type]); - else - prt_printf(out, "(invalid data type %u)", type); -} - /* disk reservations: */ static inline void bch2_disk_reservation_put(struct bch_fs *c, diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index 4701457f6381..7ed779b411f6 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -429,15 +429,20 @@ int bch2_rechecksum_bio(struct bch_fs *c, struct bio *bio, extent_nonce(version, crc_old), bio); if (bch2_crc_cmp(merged, crc_old.csum) && !c->opts.no_data_io) { - bch_err(c, "checksum error in %s() (memory corruption or bug?)\n" - "expected %0llx:%0llx got %0llx:%0llx (old type %s new type %s)", - __func__, - crc_old.csum.hi, - crc_old.csum.lo, - merged.hi, - merged.lo, - bch2_csum_types[crc_old.csum_type], - bch2_csum_types[new_csum_type]); + struct printbuf buf = PRINTBUF; + prt_printf(&buf, "checksum error in %s() (memory corruption or bug?)\n" + "expected %0llx:%0llx got %0llx:%0llx (old type ", + __func__, + crc_old.csum.hi, + crc_old.csum.lo, + merged.hi, + merged.lo); + bch2_prt_csum_type(&buf, crc_old.csum_type); + prt_str(&buf, " new type "); + bch2_prt_csum_type(&buf, new_csum_type); + prt_str(&buf, ")"); + bch_err(c, "%s", buf.buf); + printbuf_exit(&buf); return -EIO; } diff --git a/fs/bcachefs/checksum.h b/fs/bcachefs/checksum.h index 1b8c2c1016dc..e40499fde9a4 100644 --- a/fs/bcachefs/checksum.h +++ b/fs/bcachefs/checksum.h @@ -61,11 +61,12 @@ static inline void bch2_csum_err_msg(struct printbuf *out, struct bch_csum expected, struct bch_csum got) { - prt_printf(out, "checksum error: got "); + prt_str(out, "checksum error, type "); + bch2_prt_csum_type(out, type); + prt_str(out, ": got "); bch2_csum_to_text(out, type, got); prt_str(out, " should be "); bch2_csum_to_text(out, type, expected); - prt_printf(out, " type %s", bch2_csum_types[type]); } int bch2_chacha_encrypt_key(struct bch_key *, struct nonce, void *, size_t); diff --git a/fs/bcachefs/compress.h b/fs/bcachefs/compress.h index 58c2eb45570f..607fd5e232c9 100644 --- a/fs/bcachefs/compress.h +++ b/fs/bcachefs/compress.h @@ -47,14 +47,6 @@ static inline enum bch_compression_type bch2_compression_opt_to_type(unsigned v) return __bch2_compression_opt_to_type[bch2_compression_decode(v).type]; } -static inline void bch2_prt_compression_type(struct printbuf *out, enum bch_compression_type type) -{ - if (type < BCH_COMPRESSION_TYPE_NR) - prt_str(out, __bch2_compression_types[type]); - else - prt_printf(out, "(invalid compression type %u)", type); -} - int bch2_bio_uncompress_inplace(struct bch_fs *, struct bio *, struct bch_extent_crc_unpacked *); int bch2_bio_uncompress(struct bch_fs *, struct bio *, struct bio *, diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index c0b2b2f180e0..556a217108d3 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -138,13 +138,13 @@ void bch2_stripe_to_text(struct printbuf *out, struct bch_fs *c, unsigned nr_data = s.nr_blocks - s.nr_redundant; - prt_printf(out, "algo %u sectors %u blocks %u:%u csum %u gran %u", + prt_printf(out, "algo %u sectors %u blocks %u:%u csum ", s.algorithm, le16_to_cpu(s.sectors), nr_data, - s.nr_redundant, - s.csum_type, - 1U << s.csum_granularity_bits); + s.nr_redundant); + bch2_prt_csum_type(out, s.csum_type); + prt_printf(out, " gran %u", 1U << s.csum_granularity_bits); for (unsigned i = 0; i < s.nr_blocks; i++) { const struct bch_extent_ptr *ptr = sp->ptrs + i; @@ -611,10 +611,8 @@ static void ec_validate_checksums(struct bch_fs *c, struct ec_stripe_buf *buf) struct printbuf err = PRINTBUF; struct bch_dev *ca = bch_dev_bkey_exists(c, v->ptrs[i].dev); - prt_printf(&err, "stripe checksum error: expected %0llx:%0llx got %0llx:%0llx (type %s)\n", - want.hi, want.lo, - got.hi, got.lo, - bch2_csum_types[v->csum_type]); + prt_str(&err, "stripe "); + bch2_csum_err_msg(&err, v->csum_type, want, got); prt_printf(&err, " for %ps at %u of\n ", (void *) _RET_IP_, i); bch2_bkey_val_to_text(&err, c, bkey_i_to_s_c(&buf->key)); bch_err_ratelimited(ca, "%s", err.buf); diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 36d12d2adb81..1a331e539204 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -1030,11 +1030,12 @@ void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, struct bch_extent_crc_unpacked crc = bch2_extent_crc_unpack(k.k, entry_to_crc(entry)); - prt_printf(out, "crc: c_size %u size %u offset %u nonce %u csum %s compress ", + prt_printf(out, "crc: c_size %u size %u offset %u nonce %u csum ", crc.compressed_size, crc.uncompressed_size, - crc.offset, crc.nonce, - bch2_csum_types[crc.csum_type]); + crc.offset, crc.nonce); + bch2_prt_csum_type(out, crc.csum_type); + prt_str(out, " compress "); bch2_prt_compression_type(out, crc.compression_type); break; } diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 725fcf46f631..9aa28b52ab92 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -247,7 +247,7 @@ static void journal_entry_err_msg(struct printbuf *out, if (entry) { prt_str(out, " type="); - prt_str(out, bch2_jset_entry_types[entry->type]); + bch2_prt_jset_entry_type(out, entry->type); } if (!jset) { @@ -403,7 +403,8 @@ static void journal_entry_btree_keys_to_text(struct printbuf *out, struct bch_fs jset_entry_for_each_key(entry, k) { if (!first) { prt_newline(out); - prt_printf(out, "%s: ", bch2_jset_entry_types[entry->type]); + bch2_prt_jset_entry_type(out, entry->type); + prt_str(out, ": "); } prt_printf(out, "btree=%s l=%u ", bch2_btree_id_str(entry->btree_id), entry->level); bch2_bkey_val_to_text(out, c, bkey_i_to_s_c(k)); @@ -563,9 +564,9 @@ static void journal_entry_usage_to_text(struct printbuf *out, struct bch_fs *c, struct jset_entry_usage *u = container_of(entry, struct jset_entry_usage, entry); - prt_printf(out, "type=%s v=%llu", - bch2_fs_usage_types[u->entry.btree_id], - le64_to_cpu(u->v)); + prt_str(out, "type="); + bch2_prt_fs_usage_type(out, u->entry.btree_id); + prt_printf(out, " v=%llu", le64_to_cpu(u->v)); } static int journal_entry_data_usage_validate(struct bch_fs *c, @@ -827,11 +828,11 @@ int bch2_journal_entry_validate(struct bch_fs *c, void bch2_journal_entry_to_text(struct printbuf *out, struct bch_fs *c, struct jset_entry *entry) { + bch2_prt_jset_entry_type(out, entry->type); + if (entry->type < BCH_JSET_ENTRY_NR) { - prt_printf(out, "%s: ", bch2_jset_entry_types[entry->type]); + prt_str(out, ": "); bch2_jset_entry_ops[entry->type].to_text(out, c, entry); - } else { - prt_printf(out, "(unknown type %u)", entry->type); } } diff --git a/fs/bcachefs/opts.c b/fs/bcachefs/opts.c index e1800c4119b5..bb068fd72465 100644 --- a/fs/bcachefs/opts.c +++ b/fs/bcachefs/opts.c @@ -43,7 +43,7 @@ const char * const __bch2_btree_ids[] = { NULL }; -const char * const bch2_csum_types[] = { +static const char * const __bch2_csum_types[] = { BCH_CSUM_TYPES() NULL }; @@ -53,7 +53,7 @@ const char * const bch2_csum_opts[] = { NULL }; -const char * const __bch2_compression_types[] = { +static const char * const __bch2_compression_types[] = { BCH_COMPRESSION_TYPES() NULL }; @@ -83,18 +83,39 @@ const char * const bch2_member_states[] = { NULL }; -const char * const bch2_jset_entry_types[] = { +static const char * const __bch2_jset_entry_types[] = { BCH_JSET_ENTRY_TYPES() NULL }; -const char * const bch2_fs_usage_types[] = { +static const char * const __bch2_fs_usage_types[] = { BCH_FS_USAGE_TYPES() NULL }; #undef x +static void prt_str_opt_boundscheck(struct printbuf *out, const char * const opts[], + unsigned nr, const char *type, unsigned idx) +{ + if (idx < nr) + prt_str(out, opts[idx]); + else + prt_printf(out, "(unknown %s %u)", type, idx); +} + +#define PRT_STR_OPT_BOUNDSCHECKED(name, type) \ +void bch2_prt_##name(struct printbuf *out, type t) \ +{ \ + prt_str_opt_boundscheck(out, __bch2_##name##s, ARRAY_SIZE(__bch2_##name##s) - 1, #name, t);\ +} + +PRT_STR_OPT_BOUNDSCHECKED(jset_entry_type, enum bch_jset_entry_type); +PRT_STR_OPT_BOUNDSCHECKED(fs_usage_type, enum bch_fs_usage_type); +PRT_STR_OPT_BOUNDSCHECKED(data_type, enum bch_data_type); +PRT_STR_OPT_BOUNDSCHECKED(csum_type, enum bch_csum_type); +PRT_STR_OPT_BOUNDSCHECKED(compression_type, enum bch_compression_type); + static int bch2_opt_fix_errors_parse(struct bch_fs *c, const char *val, u64 *res, struct printbuf *err) { diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index 1ac4135cca1c..84e452835a17 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -16,18 +16,20 @@ extern const char * const bch2_version_upgrade_opts[]; extern const char * const bch2_sb_features[]; extern const char * const bch2_sb_compat[]; extern const char * const __bch2_btree_ids[]; -extern const char * const bch2_csum_types[]; extern const char * const bch2_csum_opts[]; -extern const char * const __bch2_compression_types[]; extern const char * const bch2_compression_opts[]; extern const char * const bch2_str_hash_types[]; extern const char * const bch2_str_hash_opts[]; extern const char * const __bch2_data_types[]; extern const char * const bch2_member_states[]; -extern const char * const bch2_jset_entry_types[]; -extern const char * const bch2_fs_usage_types[]; extern const char * const bch2_d_types[]; +void bch2_prt_jset_entry_type(struct printbuf *, enum bch_jset_entry_type); +void bch2_prt_fs_usage_type(struct printbuf *, enum bch_fs_usage_type); +void bch2_prt_data_type(struct printbuf *, enum bch_data_type); +void bch2_prt_csum_type(struct printbuf *, enum bch_csum_type); +void bch2_prt_compression_type(struct printbuf *, enum bch_compression_type); + static inline const char *bch2_d_type_str(unsigned d_type) { return (d_type < BCH_DT_MAX ? bch2_d_types[d_type] : NULL) ?: "(bad d_type)"; -- cgit v1.2.3-59-g8ed1b From 4518e80adfdbfdec1cc79c98bc73677ff44d96bd Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 14:05:36 -0400 Subject: bcachefs: Go rw if running any explicit recovery passes This fixes a bug where we fail to start when upgrading/downgrading because we forgot we needed to go rw. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery_passes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index cb501460d615..0cec0f7d9703 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -44,7 +44,7 @@ static int bch2_set_may_go_rw(struct bch_fs *c) set_bit(BCH_FS_may_go_rw, &c->flags); - if (keys->nr || c->opts.fsck || !c->sb.clean) + if (keys->nr || c->opts.fsck || !c->sb.clean || c->recovery_passes_explicit) return bch2_fs_read_write_early(c); return 0; } -- cgit v1.2.3-59-g8ed1b From 82cf18f23e1ae17053983e325e550194f947e1f1 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 21:07:05 -0400 Subject: bcachefs: Fix deadlock in journal replay btree_key_can_insert_cached() should be checking the watermark - BCH_TRANS_COMMIT_journal_replay really means nonblocking mode when watermark < reclaim, it was being used incorrectly. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_trans_commit.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 9609ad370d71..bbec91e8e650 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -397,12 +397,13 @@ static int btree_key_can_insert_cached(struct btree_trans *trans, unsigned flags struct bkey_cached *ck = (void *) path->l[0].b; unsigned new_u64s; struct bkey_i *new_k; + unsigned watermark = flags & BCH_WATERMARK_MASK; EBUG_ON(path->level); - if (!test_bit(BKEY_CACHED_DIRTY, &ck->flags) && - bch2_btree_key_cache_must_wait(c) && - !(flags & BCH_TRANS_COMMIT_journal_reclaim)) + if (watermark < BCH_WATERMARK_reclaim && + !test_bit(BKEY_CACHED_DIRTY, &ck->flags) && + bch2_btree_key_cache_must_wait(c)) return -BCH_ERR_btree_insert_need_journal_reclaim; /* -- cgit v1.2.3-59-g8ed1b From 9e203c43dc1cbaefb3888ee0ba885b2d20d47526 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 00:26:01 -0400 Subject: bcachefs: Fix missing write refs in fs fio paths bch2_journal_flush_seq requires us to have a write ref Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 2 ++ fs/bcachefs/fs-io-direct.c | 19 +++++++++++++------ fs/bcachefs/fs-io.c | 16 ++++++++-------- 3 files changed, 23 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index a31a5f706929..91c3c1fef233 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -709,6 +709,8 @@ struct btree_trans_buf { x(stripe_delete) \ x(reflink) \ x(fallocate) \ + x(fsync) \ + x(dio_write) \ x(discard) \ x(discard_fast) \ x(invalidate) \ diff --git a/fs/bcachefs/fs-io-direct.c b/fs/bcachefs/fs-io-direct.c index f49e6c0f0f68..b889370a5088 100644 --- a/fs/bcachefs/fs-io-direct.c +++ b/fs/bcachefs/fs-io-direct.c @@ -387,6 +387,8 @@ static __always_inline long bch2_dio_write_done(struct dio_write *dio) ret = dio->op.error ?: ((long) dio->written << 9); bio_put(&dio->op.wbio.bio); + bch2_write_ref_put(dio->op.c, BCH_WRITE_REF_dio_write); + /* inode->i_dio_count is our ref on inode and thus bch_fs */ inode_dio_end(&inode->v); @@ -590,22 +592,25 @@ ssize_t bch2_direct_write(struct kiocb *req, struct iov_iter *iter) prefetch(&inode->ei_inode); prefetch((void *) &inode->ei_inode + 64); + if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_dio_write)) + return -EROFS; + inode_lock(&inode->v); ret = generic_write_checks(req, iter); if (unlikely(ret <= 0)) - goto err; + goto err_put_write_ref; ret = file_remove_privs(file); if (unlikely(ret)) - goto err; + goto err_put_write_ref; ret = file_update_time(file); if (unlikely(ret)) - goto err; + goto err_put_write_ref; if (unlikely((req->ki_pos|iter->count) & (block_bytes(c) - 1))) - goto err; + goto err_put_write_ref; inode_dio_begin(&inode->v); bch2_pagecache_block_get(inode); @@ -645,7 +650,7 @@ ssize_t bch2_direct_write(struct kiocb *req, struct iov_iter *iter) } ret = bch2_dio_write_loop(dio); -err: +out: if (locked) inode_unlock(&inode->v); return ret; @@ -653,7 +658,9 @@ err_put_bio: bch2_pagecache_block_put(inode); bio_put(bio); inode_dio_end(&inode->v); - goto err; +err_put_write_ref: + bch2_write_ref_put(c, BCH_WRITE_REF_dio_write); + goto out; } void bch2_fs_fs_io_direct_exit(struct bch_fs *c) diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c index 8c70123b6a0c..20b40477425f 100644 --- a/fs/bcachefs/fs-io.c +++ b/fs/bcachefs/fs-io.c @@ -174,18 +174,18 @@ void __bch2_i_sectors_acct(struct bch_fs *c, struct bch_inode_info *inode, static int bch2_flush_inode(struct bch_fs *c, struct bch_inode_info *inode) { - struct bch_inode_unpacked u; - int ret; - if (c->opts.journal_flush_disabled) return 0; - ret = bch2_inode_find_by_inum(c, inode_inum(inode), &u); - if (ret) - return ret; + if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_fsync)) + return -EROFS; - return bch2_journal_flush_seq(&c->journal, u.bi_journal_seq) ?: - bch2_inode_flush_nocow_writes(c, inode); + struct bch_inode_unpacked u; + int ret = bch2_inode_find_by_inum(c, inode_inum(inode), &u) ?: + bch2_journal_flush_seq(&c->journal, u.bi_journal_seq) ?: + bch2_inode_flush_nocow_writes(c, inode); + bch2_write_ref_put(c, BCH_WRITE_REF_fsync); + return ret; } int bch2_fsync(struct file *file, loff_t start, loff_t end, int datasync) -- cgit v1.2.3-59-g8ed1b From 9054ef2ea944528b7935d0ce3f540d4dc0bc37ba Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 16:13:13 -0400 Subject: bcachefs: Run merges at BCH_WATERMARK_btree This fixes a deadlock where the interior update path during journal replay ends up doing a ton of merges on the backpointers btree, and deadlocking. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 29a5bc7d8789..1a7537ee4329 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1926,6 +1926,8 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, BUG_ON(!trans->paths[path].should_be_locked); BUG_ON(!btree_node_locked(&trans->paths[path], level)); + flags &= ~BCH_WATERMARK_MASK; + b = trans->paths[path].l[level].b; if ((sib == btree_prev_sib && bpos_eq(b->data->min_key, POS_MIN)) || @@ -2071,6 +2073,10 @@ err: bch2_path_put(trans, new_path, true); bch2_path_put(trans, sib_path, true); bch2_trans_verify_locks(trans); + if (ret == -BCH_ERR_journal_reclaim_would_deadlock) + ret = 0; + if (!ret) + ret = bch2_trans_relock(trans); return ret; err_free_update: bch2_btree_node_free_never_used(as, trans, n); -- cgit v1.2.3-59-g8ed1b From 3f10048973c8366e27147d72bf3fd8d0e1d929f2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 18:39:03 -0400 Subject: bcachefs: Disable merges from interior update path There's been a bug in the btree write buffer where it wasn't triggering btree node merges - and leaving behind a bunch of nearly empty btree nodes. Then during journal replay, when updates to the backpointers btree aren't using the btree write buffer (because we require synchronization with journal replay), we end up doing those merges all at once. Then if it's the interior update path running them, we deadlock because those run with the highest watermark. There's no real need for the interior update path to be doing btree node merges; other code paths can handle that at lower watermarks. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 1a7537ee4329..82ac00aa0f7b 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1926,6 +1926,16 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, BUG_ON(!trans->paths[path].should_be_locked); BUG_ON(!btree_node_locked(&trans->paths[path], level)); + /* + * Work around a deadlock caused by the btree write buffer not doing + * merges and leaving tons of merges for us to do - we really don't need + * to be doing merges at all from the interior update path, and if the + * interior update path is generating too many new interior updates we + * deadlock: + */ + if ((flags & BCH_WATERMARK_MASK) == BCH_WATERMARK_interior_updates) + return 0; + flags &= ~BCH_WATERMARK_MASK; b = trans->paths[path].l[level].b; -- cgit v1.2.3-59-g8ed1b From 86dbf8c566417afd62087ecd3bedcf84f91ee4e4 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 26 Dec 2023 22:42:34 -0500 Subject: bcachefs: Fix btree node merging on write buffer btrees The btree write buffer flush fastpath that avoids the main transaction commit path had the unfortunate side effect of not doing btree node merging. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_write_buffer.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_write_buffer.c b/fs/bcachefs/btree_write_buffer.c index baf63e2fddb6..36a6f42aba5e 100644 --- a/fs/bcachefs/btree_write_buffer.c +++ b/fs/bcachefs/btree_write_buffer.c @@ -316,6 +316,16 @@ static int bch2_btree_write_buffer_flush_locked(struct btree_trans *trans) bpos_gt(k->k.k.p, path->l[0].b->key.k.p)) { bch2_btree_node_unlock_write(trans, path, path->l[0].b); write_locked = false; + + ret = lockrestart_do(trans, + bch2_btree_iter_traverse(&iter) ?: + bch2_foreground_maybe_merge(trans, iter.path, 0, + BCH_WATERMARK_reclaim| + BCH_TRANS_COMMIT_journal_reclaim| + BCH_TRANS_COMMIT_no_check_rw| + BCH_TRANS_COMMIT_no_enospc)); + if (ret) + goto err; } } @@ -382,10 +392,10 @@ static int bch2_btree_write_buffer_flush_locked(struct btree_trans *trans) ret = commit_do(trans, NULL, NULL, BCH_WATERMARK_reclaim| + BCH_TRANS_COMMIT_journal_reclaim| BCH_TRANS_COMMIT_no_check_rw| BCH_TRANS_COMMIT_no_enospc| - BCH_TRANS_COMMIT_no_journal_res| - BCH_TRANS_COMMIT_journal_reclaim, + BCH_TRANS_COMMIT_no_journal_res , btree_write_buffered_insert(trans, i)); if (ret) goto err; -- cgit v1.2.3-59-g8ed1b From 16b52bbee4823b01ab7fe3919373c981a38f3797 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Fri, 5 Apr 2024 17:56:35 +0300 Subject: kernfs: annotate different lockdep class for of->mutex of writable files The writable file /sys/power/resume may call vfs lookup helpers for arbitrary paths and readonly files can be read by overlayfs from vfs helpers when sysfs is a lower layer of overalyfs. To avoid a lockdep warning of circular dependency between overlayfs inode lock and kernfs of->mutex, use a different lockdep class for writable and readonly kernfs files. Reported-by: syzbot+9a5b0ced8b1bfb238b56@syzkaller.appspotmail.com Fixes: 0fedefd4c4e3 ("kernfs: sysfs: support custom llseek method for sysfs entries") Suggested-by: Al Viro Signed-off-by: Amir Goldstein Signed-off-by: Al Viro --- fs/kernfs/file.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/kernfs/file.c b/fs/kernfs/file.c index e9df2f87072c..8502ef68459b 100644 --- a/fs/kernfs/file.c +++ b/fs/kernfs/file.c @@ -636,11 +636,18 @@ static int kernfs_fop_open(struct inode *inode, struct file *file) * each file a separate locking class. Let's differentiate on * whether the file has mmap or not for now. * - * Both paths of the branch look the same. They're supposed to + * For similar reasons, writable and readonly files are given different + * lockdep key, because the writable file /sys/power/resume may call vfs + * lookup helpers for arbitrary paths and readonly files can be read by + * overlayfs from vfs helpers when sysfs is a lower layer of overalyfs. + * + * All three cases look the same. They're supposed to * look that way and give @of->mutex different static lockdep keys. */ if (has_mmap) mutex_init(&of->mutex); + else if (file->f_mode & FMODE_WRITE) + mutex_init(&of->mutex); else mutex_init(&of->mutex); -- cgit v1.2.3-59-g8ed1b From 3fdd89b452c2ea5e2195d6e315bef122769584c9 Mon Sep 17 00:00:00 2001 From: Daeho Jeong Date: Thu, 11 Apr 2024 10:54:10 -0700 Subject: f2fs: prevent writing without fallocate() for pinned files In a case writing without fallocate(), we can't guarantee it's allocated in the conventional area for zoned stroage. To make it consistent across storage devices, we disallow it regardless of storage device types. Signed-off-by: Daeho Jeong Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 0d2626a6a21f..7fbdb7863778 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -58,7 +58,7 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf) struct inode *inode = file_inode(vmf->vma->vm_file); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); struct dnode_of_data dn; - bool need_alloc = true; + bool need_alloc = !f2fs_is_pinned_file(inode); int err = 0; vm_fault_t ret; @@ -115,19 +115,18 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf) goto out_sem; } + set_new_dnode(&dn, inode, NULL, NULL, 0); if (need_alloc) { /* block allocation */ - set_new_dnode(&dn, inode, NULL, NULL, 0); err = f2fs_get_block_locked(&dn, page->index); - } - -#ifdef CONFIG_F2FS_FS_COMPRESSION - if (!need_alloc) { - set_new_dnode(&dn, inode, NULL, NULL, 0); + } else { err = f2fs_get_dnode_of_data(&dn, page->index, LOOKUP_NODE); f2fs_put_dnode(&dn); + if (f2fs_is_pinned_file(inode) && + !__is_valid_data_blkaddr(dn.data_blkaddr)) + err = -EIO; } -#endif + if (err) { unlock_page(page); goto out_sem; @@ -3260,7 +3259,7 @@ static int f2fs_ioc_set_pin_file(struct file *filp, unsigned long arg) goto done; } - if (f2fs_sb_has_blkzoned(sbi) && F2FS_HAS_BLOCKS(inode)) { + if (F2FS_HAS_BLOCKS(inode)) { ret = -EFBIG; goto out; } @@ -4823,6 +4822,8 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) bool dio; bool may_need_sync = true; int preallocated; + const loff_t pos = iocb->ki_pos; + const ssize_t count = iov_iter_count(from); ssize_t ret; if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) { @@ -4844,6 +4845,12 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) inode_lock(inode); } + if (f2fs_is_pinned_file(inode) && + !f2fs_overwrite_io(inode, pos, count)) { + ret = -EIO; + goto out_unlock; + } + ret = f2fs_write_checks(iocb, from); if (ret <= 0) goto out_unlock; -- cgit v1.2.3-59-g8ed1b From b2cf5a1ff236fcd0eb9dcbb188df30b50bb1af0f Mon Sep 17 00:00:00 2001 From: Daeho Jeong Date: Thu, 11 Apr 2024 11:37:53 -0700 Subject: f2fs: allow direct io of pinned files for zoned storage Since the allocation happens in conventional LU for zoned storage, we can allow direct io for that. Signed-off-by: Daeho Jeong Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 7fbdb7863778..ac1ae85f3cc3 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -833,7 +833,8 @@ static bool f2fs_force_buffered_io(struct inode *inode, int rw) * for blkzoned device, fallback direct IO to buffered IO, so * all IOs can be serialized by log-structured write. */ - if (f2fs_sb_has_blkzoned(sbi) && (rw == WRITE)) + if (f2fs_sb_has_blkzoned(sbi) && (rw == WRITE) && + !f2fs_is_pinned_file(inode)) return true; if (is_sbi_flag_set(sbi, SBI_CP_DISABLED)) return true; -- cgit v1.2.3-59-g8ed1b From bceb86be9e970bc6d77bfd1f8e5272d9a2007c16 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 23:59:06 -0400 Subject: bcachefs: add missing bounds check in __bch2_bkey_val_invalid() Signed-off-by: Kent Overstreet --- fs/bcachefs/bkey_methods.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c index 5e52684764eb..e4481e4d5f38 100644 --- a/fs/bcachefs/bkey_methods.c +++ b/fs/bcachefs/bkey_methods.c @@ -175,7 +175,10 @@ int __bch2_bkey_invalid(struct bch_fs *c, struct bkey_s_c k, !(bch2_key_types_allowed[type] & BIT_ULL(k.k->type)), c, err, bkey_invalid_type_for_btree, "invalid key type for btree %s (%s)", - bch2_btree_node_type_str(type), bch2_bkey_types[k.k->type]); + bch2_btree_node_type_str(type), + k.k->type < KEY_TYPE_MAX + ? bch2_bkey_types[k.k->type] + : "(unknown)"); if (btree_node_type_is_extents(type) && !bkey_whiteout(k.k)) { bkey_fsck_err_on(k.k->size == 0, c, err, -- cgit v1.2.3-59-g8ed1b From d789e9a7d5e2799f4d5425b0b620210d2fcad529 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 23:59:28 -0400 Subject: bcachefs: Interior known are required to have known key types For forwards compatibilyt, we allow bkeys of unknown type in leaf nodes; we can simply ignore metadata we don't understand. Pointers to btree nodes must always be of known types, howwever. Signed-off-by: Kent Overstreet --- fs/bcachefs/bkey_methods.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c index e4481e4d5f38..db336a43fc08 100644 --- a/fs/bcachefs/bkey_methods.c +++ b/fs/bcachefs/bkey_methods.c @@ -171,7 +171,8 @@ int __bch2_bkey_invalid(struct bch_fs *c, struct bkey_s_c k, if (type >= BKEY_TYPE_NR) return 0; - bkey_fsck_err_on((flags & BKEY_INVALID_COMMIT) && + bkey_fsck_err_on((type == BKEY_TYPE_btree || + (flags & BKEY_INVALID_COMMIT)) && !(bch2_key_types_allowed[type] & BIT_ULL(k.k->type)), c, err, bkey_invalid_type_for_btree, "invalid key type for btree %s (%s)", -- cgit v1.2.3-59-g8ed1b From 8cf2036e7b557282667a437d409e6307d55366ab Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 15:34:14 -0400 Subject: bcachefs: add safety checks in bch2_btree_node_fill() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index c7f156320a35..c347d08d80bc 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -711,7 +711,30 @@ static noinline struct btree *bch2_btree_node_fill(struct btree_trans *trans, struct btree *b; u32 seq; - BUG_ON(level + 1 >= BTREE_MAX_DEPTH); + if (unlikely(level >= BTREE_MAX_DEPTH)) { + int ret = bch2_fs_topology_error(c, "attempting to get btree node at level %u, >= max depth %u", + level, BTREE_MAX_DEPTH); + return ERR_PTR(ret); + } + + if (unlikely(!bkey_is_btree_ptr(&k->k))) { + struct printbuf buf = PRINTBUF; + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(k)); + + int ret = bch2_fs_topology_error(c, "attempting to get btree node with non-btree key %s", buf.buf); + printbuf_exit(&buf); + return ERR_PTR(ret); + } + + if (unlikely(k->k.u64s > BKEY_BTREE_PTR_U64s_MAX)) { + struct printbuf buf = PRINTBUF; + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(k)); + + int ret = bch2_fs_topology_error(c, "attempting to get btree node with too big key %s", buf.buf); + printbuf_exit(&buf); + return ERR_PTR(ret); + } + /* * Parent node must be locked, else we could read in a btree node that's * been freed: -- cgit v1.2.3-59-g8ed1b From e879389f5777de5d94f27f8d88d7e92341afa3ef Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 15:54:33 -0400 Subject: bcachefs: Fix bch2_btree_node_fill() for !path We shouldn't be doing the unlock/relock dance when we're not using a path - this fixes an assertion pop when called from btree node scan. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 44 ++++++++++++++++++-------------------------- 1 file changed, 18 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index c347d08d80bc..02c70e813fac 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -709,7 +709,6 @@ static noinline struct btree *bch2_btree_node_fill(struct btree_trans *trans, struct bch_fs *c = trans->c; struct btree_cache *bc = &c->btree_cache; struct btree *b; - u32 seq; if (unlikely(level >= BTREE_MAX_DEPTH)) { int ret = bch2_fs_topology_error(c, "attempting to get btree node at level %u, >= max depth %u", @@ -775,34 +774,26 @@ static noinline struct btree *bch2_btree_node_fill(struct btree_trans *trans, } set_btree_node_read_in_flight(b); - six_unlock_write(&b->c.lock); - seq = six_lock_seq(&b->c.lock); - six_unlock_intent(&b->c.lock); - /* Unlock before doing IO: */ - if (path && sync) - bch2_trans_unlock_noassert(trans); - - bch2_btree_node_read(trans, b, sync); + if (path) { + u32 seq = six_lock_seq(&b->c.lock); - if (!sync) - return NULL; + /* Unlock before doing IO: */ + six_unlock_intent(&b->c.lock); + bch2_trans_unlock_noassert(trans); - if (path) { - int ret = bch2_trans_relock(trans) ?: - bch2_btree_path_relock_intent(trans, path); - if (ret) { - BUG_ON(!trans->restarted); - return ERR_PTR(ret); - } - } + bch2_btree_node_read(trans, b, sync); - if (!six_relock_type(&b->c.lock, lock_type, seq)) { - BUG_ON(!path); + if (!sync) + return NULL; - trace_and_count(c, trans_restart_relock_after_fill, trans, _THIS_IP_, path); - return ERR_PTR(btree_trans_restart(trans, BCH_ERR_transaction_restart_relock_after_fill)); + if (!six_relock_type(&b->c.lock, lock_type, seq)) + b = NULL; + } else { + bch2_btree_node_read(trans, b, sync); + if (lock_type == SIX_LOCK_read) + six_lock_downgrade(&b->c.lock); } return b; @@ -1135,18 +1126,19 @@ int bch2_btree_node_prefetch(struct btree_trans *trans, { struct bch_fs *c = trans->c; struct btree_cache *bc = &c->btree_cache; - struct btree *b; BUG_ON(path && !btree_node_locked(path, level + 1)); BUG_ON(level >= BTREE_MAX_DEPTH); - b = btree_cache_find(bc, k); + struct btree *b = btree_cache_find(bc, k); if (b) return 0; b = bch2_btree_node_fill(trans, path, k, btree_id, level, SIX_LOCK_read, false); - return PTR_ERR_OR_ZERO(b); + if (!IS_ERR_OR_NULL(b)) + six_unlock_read(&b->c.lock); + return bch2_trans_relock(trans) ?: PTR_ERR_OR_ZERO(b); } void bch2_btree_node_evict(struct btree_trans *trans, const struct bkey_i *k) -- cgit v1.2.3-59-g8ed1b From bdae2a7e6020e1cf864a30dab2af323575569dc7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 22:43:11 -0400 Subject: bcachefs: sysfs internal/trigger_journal_flush Add a sysfs knob for immediately flushing the entire journal. Signed-off-by: Kent Overstreet --- fs/bcachefs/sysfs.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index b18b0cc81b59..5be92fe3f4ea 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -25,6 +25,7 @@ #include "ec.h" #include "inode.h" #include "journal.h" +#include "journal_reclaim.h" #include "keylist.h" #include "move.h" #include "movinggc.h" @@ -138,6 +139,7 @@ do { \ write_attribute(trigger_gc); write_attribute(trigger_discards); write_attribute(trigger_invalidates); +write_attribute(trigger_journal_flush); write_attribute(prune_cache); write_attribute(btree_wakeup); rw_attribute(btree_gc_periodic); @@ -500,7 +502,7 @@ STORE(bch2_fs) /* Debugging: */ - if (!test_bit(BCH_FS_rw, &c->flags)) + if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_sysfs)) return -EROFS; if (attr == &sysfs_prune_cache) { @@ -533,6 +535,11 @@ STORE(bch2_fs) if (attr == &sysfs_trigger_invalidates) bch2_do_invalidates(c); + if (attr == &sysfs_trigger_journal_flush) { + bch2_journal_flush_all_pins(&c->journal); + bch2_journal_meta(&c->journal); + } + #ifdef CONFIG_BCACHEFS_TESTS if (attr == &sysfs_perf_test) { char *tmp = kstrdup(buf, GFP_KERNEL), *p = tmp; @@ -553,6 +560,7 @@ STORE(bch2_fs) size = ret; } #endif + bch2_write_ref_put(c, BCH_WRITE_REF_sysfs); return size; } SYSFS_OPS(bch2_fs); @@ -651,6 +659,7 @@ struct attribute *bch2_fs_internal_files[] = { &sysfs_trigger_gc, &sysfs_trigger_discards, &sysfs_trigger_invalidates, + &sysfs_trigger_journal_flush, &sysfs_prune_cache, &sysfs_btree_wakeup, -- cgit v1.2.3-59-g8ed1b From 27c15ed297cb71c2e7a839439b5a097081a32605 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 18:45:47 -0400 Subject: bcachefs: bch_member.btree_allocated_bitmap This adds a small (64 bit) per-device bitmap that tracks ranges that have btree nodes, for accelerating btree node scan if it is ever needed. - New helpers, bch2_dev_btree_bitmap_marked() and bch2_dev_bitmap_mark(), for checking and updating the bitmap - Interior btree update path updates the bitmaps when required - The check_allocations pass has a new fsck_err check, btree_bitmap_not_marked - New on disk format version, mi_btree_mitmap, which indicates the new bitmap is present - Upgrade table lists the required recovery pass and expected fsck error - Btree node scan uses the bitmap to skip ranges if we're on the new version Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 7 +++-- fs/bcachefs/btree_gc.c | 13 +++++++++ fs/bcachefs/btree_node_scan.c | 9 +++++-- fs/bcachefs/btree_update_interior.c | 24 +++++++++++++++++ fs/bcachefs/sb-downgrade.c | 5 +++- fs/bcachefs/sb-errors_types.h | 3 ++- fs/bcachefs/sb-members.c | 53 +++++++++++++++++++++++++++++++++++++ fs/bcachefs/sb-members.h | 21 +++++++++++++++ fs/bcachefs/super_types.h | 2 ++ 9 files changed, 131 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 9480f1b44f07..085987435a5e 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -578,7 +578,8 @@ struct bch_member { __le64 nbuckets; /* device size */ __le16 first_bucket; /* index of first bucket used */ __le16 bucket_size; /* sectors */ - __le32 pad; + __u8 btree_bitmap_shift; + __u8 pad[3]; __le64 last_mount; /* time_t */ __le64 flags; @@ -587,6 +588,7 @@ struct bch_member { __le64 errors_at_reset[BCH_MEMBER_ERROR_NR]; __le64 errors_reset_time; __le64 seq; + __le64 btree_allocated_bitmap; }; #define BCH_MEMBER_V1_BYTES 56 @@ -876,7 +878,8 @@ struct bch_sb_field_downgrade { x(rebalance_work, BCH_VERSION(1, 3)) \ x(member_seq, BCH_VERSION(1, 4)) \ x(subvolume_fs_parent, BCH_VERSION(1, 5)) \ - x(btree_subvolume_children, BCH_VERSION(1, 6)) + x(btree_subvolume_children, BCH_VERSION(1, 6)) \ + x(mi_btree_bitmap, BCH_VERSION(1, 7)) enum bcachefs_metadata_version { bcachefs_metadata_version_min = 9, diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index d2555da55c6d..ecbd9598f69f 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -828,6 +828,7 @@ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, struct bch_fs *c = trans->c; struct bkey deleted = KEY(0, 0, 0); struct bkey_s_c old = (struct bkey_s_c) { &deleted, NULL }; + struct printbuf buf = PRINTBUF; int ret = 0; deleted.p = k->k->p; @@ -848,11 +849,23 @@ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, if (ret) goto err; + if (mustfix_fsck_err_on(level && !bch2_dev_btree_bitmap_marked(c, *k), + c, btree_bitmap_not_marked, + "btree ptr not marked in member info btree allocated bitmap\n %s", + (bch2_bkey_val_to_text(&buf, c, *k), + buf.buf))) { + mutex_lock(&c->sb_lock); + bch2_dev_btree_bitmap_mark(c, *k); + bch2_write_super(c); + mutex_unlock(&c->sb_lock); + } + ret = commit_do(trans, NULL, NULL, 0, bch2_key_trigger(trans, btree_id, level, old, unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_GC)); fsck_err: err: + printbuf_exit(&buf); bch_err_fn(c, ret); return ret; } diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index 20f2b37c4474..866bd278439f 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -205,8 +205,13 @@ static int read_btree_nodes_worker(void *p) last_print = jiffies; } - try_read_btree_node(w->f, ca, bio, buf, - bucket * ca->mi.bucket_size + bucket_offset); + u64 sector = bucket * ca->mi.bucket_size + bucket_offset; + + if (c->sb.version_upgrade_complete >= bcachefs_metadata_version_mi_btree_bitmap && + !bch2_dev_btree_bitmap_marked_sectors(ca, sector, btree_sectors(c))) + continue; + + try_read_btree_node(w->f, ca, bio, buf, sector); } err: bio_put(bio); diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 82ac00aa0f7b..6030c396754f 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -21,6 +21,7 @@ #include "keylist.h" #include "recovery_passes.h" #include "replicas.h" +#include "sb-members.h" #include "super-io.h" #include "trace.h" @@ -605,6 +606,26 @@ static void btree_update_add_key(struct btree_update *as, bch2_keylist_push(keys); } +static bool btree_update_new_nodes_marked_sb(struct btree_update *as) +{ + for_each_keylist_key(&as->new_keys, k) + if (!bch2_dev_btree_bitmap_marked(as->c, bkey_i_to_s_c(k))) + return false; + return true; +} + +static void btree_update_new_nodes_mark_sb(struct btree_update *as) +{ + struct bch_fs *c = as->c; + + mutex_lock(&c->sb_lock); + for_each_keylist_key(&as->new_keys, k) + bch2_dev_btree_bitmap_mark(c, bkey_i_to_s_c(k)); + + bch2_write_super(c); + mutex_unlock(&c->sb_lock); +} + /* * The transactional part of an interior btree node update, where we journal the * update we did to the interior node and update alloc info: @@ -662,6 +683,9 @@ static void btree_update_nodes_written(struct btree_update *as) if (ret) goto err; + if (!btree_update_new_nodes_marked_sb(as)) + btree_update_new_nodes_mark_sb(as); + /* * Wait for any in flight writes to finish before we free the old nodes * on disk: diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c index d6f81179c3a2..a98ef940b7a3 100644 --- a/fs/bcachefs/sb-downgrade.c +++ b/fs/bcachefs/sb-downgrade.c @@ -51,7 +51,10 @@ BCH_FSCK_ERR_subvol_fs_path_parent_wrong) \ x(btree_subvolume_children, \ BIT_ULL(BCH_RECOVERY_PASS_check_subvols), \ - BCH_FSCK_ERR_subvol_children_not_set) + BCH_FSCK_ERR_subvol_children_not_set) \ + x(mi_btree_bitmap, \ + BIT_ULL(BCH_RECOVERY_PASS_check_allocations), \ + BCH_FSCK_ERR_btree_bitmap_not_marked) #define DOWNGRADE_TABLE() diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index d7d609131030..5b600c6d7aca 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -270,7 +270,8 @@ x(btree_ptr_v2_min_key_bad, 262) \ x(btree_root_unreadable_and_scan_found_nothing, 263) \ x(snapshot_node_missing, 264) \ - x(dup_backpointer_to_bad_csum_extent, 265) + x(dup_backpointer_to_bad_csum_extent, 265) \ + x(btree_bitmap_not_marked, 266) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index eff5ce18c69c..522a969345e5 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" +#include "btree_cache.h" #include "disk_groups.h" #include "opts.h" #include "replicas.h" @@ -426,3 +427,55 @@ void bch2_dev_errors_reset(struct bch_dev *ca) bch2_write_super(c); mutex_unlock(&c->sb_lock); } + +/* + * Per member "range has btree nodes" bitmap: + * + * This is so that if we ever have to run the btree node scan to repair we don't + * have to scan full devices: + */ + +bool bch2_dev_btree_bitmap_marked(struct bch_fs *c, struct bkey_s_c k) +{ + bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) + if (!bch2_dev_btree_bitmap_marked_sectors(bch_dev_bkey_exists(c, ptr->dev), + ptr->offset, btree_sectors(c))) + return false; + return true; +} + +static void __bch2_dev_btree_bitmap_mark(struct bch_sb_field_members_v2 *mi, unsigned dev, + u64 start, unsigned sectors) +{ + struct bch_member *m = __bch2_members_v2_get_mut(mi, dev); + u64 bitmap = le64_to_cpu(m->btree_allocated_bitmap); + + u64 end = start + sectors; + + int resize = ilog2(roundup_pow_of_two(end)) - (m->btree_bitmap_shift + 6); + if (resize > 0) { + u64 new_bitmap = 0; + + for (unsigned i = 0; i < 64; i++) + if (bitmap & BIT_ULL(i)) + new_bitmap |= BIT_ULL(i >> resize); + bitmap = new_bitmap; + m->btree_bitmap_shift += resize; + } + + for (unsigned bit = sectors >> m->btree_bitmap_shift; + bit << m->btree_bitmap_shift < end; + bit++) + bitmap |= BIT_ULL(bit); + + m->btree_allocated_bitmap = cpu_to_le64(bitmap); +} + +void bch2_dev_btree_bitmap_mark(struct bch_fs *c, struct bkey_s_c k) +{ + lockdep_assert_held(&c->sb_lock); + + struct bch_sb_field_members_v2 *mi = bch2_sb_field_get(c->disk_sb.sb, members_v2); + bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) + __bch2_dev_btree_bitmap_mark(mi, ptr->dev, ptr->offset, btree_sectors(c)); +} diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index be0a94183271..b27c3e4467cf 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -3,6 +3,7 @@ #define _BCACHEFS_SB_MEMBERS_H #include "darray.h" +#include "bkey_types.h" extern char * const bch2_member_error_strs[]; @@ -220,6 +221,8 @@ static inline struct bch_member_cpu bch2_mi_to_cpu(struct bch_member *mi) : 1, .freespace_initialized = BCH_MEMBER_FREESPACE_INITIALIZED(mi), .valid = bch2_member_exists(mi), + .btree_bitmap_shift = mi->btree_bitmap_shift, + .btree_allocated_bitmap = le64_to_cpu(mi->btree_allocated_bitmap), }; } @@ -228,4 +231,22 @@ void bch2_sb_members_from_cpu(struct bch_fs *); void bch2_dev_io_errors_to_text(struct printbuf *, struct bch_dev *); void bch2_dev_errors_reset(struct bch_dev *); +static inline bool bch2_dev_btree_bitmap_marked_sectors(struct bch_dev *ca, u64 start, unsigned sectors) +{ + u64 end = start + sectors; + + if (end > 64 << ca->mi.btree_bitmap_shift) + return false; + + for (unsigned bit = sectors >> ca->mi.btree_bitmap_shift; + bit << ca->mi.btree_bitmap_shift < end; + bit++) + if (!(ca->mi.btree_allocated_bitmap & BIT_ULL(bit))) + return false; + return true; +} + +bool bch2_dev_btree_bitmap_marked(struct bch_fs *, struct bkey_s_c); +void bch2_dev_btree_bitmap_mark(struct bch_fs *, struct bkey_s_c); + #endif /* _BCACHEFS_SB_MEMBERS_H */ diff --git a/fs/bcachefs/super_types.h b/fs/bcachefs/super_types.h index ec784d975f66..11bcef170c2c 100644 --- a/fs/bcachefs/super_types.h +++ b/fs/bcachefs/super_types.h @@ -37,6 +37,8 @@ struct bch_member_cpu { u8 durability; u8 freespace_initialized; u8 valid; + u8 btree_bitmap_shift; + u64 btree_allocated_bitmap; }; #endif /* _BCACHEFS_SUPER_TYPES_H */ -- cgit v1.2.3-59-g8ed1b From f0a73d4fde5b285d94a702026216d9fd1fd2733d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 14 Apr 2024 00:51:48 -0400 Subject: bcachefs: Check for backpointer bucket_offset >= bucket size Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 8 +++++--- fs/bcachefs/backpointers.h | 9 +++------ fs/bcachefs/sb-errors_types.h | 2 +- 3 files changed, 9 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 114328acde72..fadb1078903d 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -49,13 +49,15 @@ int bch2_backpointer_invalid(struct bch_fs *c, struct bkey_s_c k, if (!bch2_dev_exists2(c, bp.k->p.inode)) return 0; + struct bch_dev *ca = bch_dev_bkey_exists(c, bp.k->p.inode); struct bpos bucket = bp_pos_to_bucket(c, bp.k->p); int ret = 0; - bkey_fsck_err_on(!bpos_eq(bp.k->p, bucket_pos_to_bp(c, bucket, bp.v->bucket_offset)), + bkey_fsck_err_on((bp.v->bucket_offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT) >= ca->mi.bucket_size || + !bpos_eq(bp.k->p, bucket_pos_to_bp(c, bucket, bp.v->bucket_offset)), c, err, - backpointer_pos_wrong, - "backpointer at wrong pos"); + backpointer_bucket_offset_wrong, + "backpointer bucket_offset wrong"); fsck_err: return ret; } diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index da012ca7daee..85949b9fd880 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -53,14 +53,11 @@ static inline struct bpos bucket_pos_to_bp(const struct bch_fs *c, u64 bucket_offset) { struct bch_dev *ca = bch_dev_bkey_exists(c, bucket.inode); - struct bpos ret; - - ret = POS(bucket.inode, - (bucket_to_sector(ca, bucket.offset) << - MAX_EXTENT_COMPRESS_RATIO_SHIFT) + bucket_offset); + struct bpos ret = POS(bucket.inode, + (bucket_to_sector(ca, bucket.offset) << + MAX_EXTENT_COMPRESS_RATIO_SHIFT) + bucket_offset); EBUG_ON(!bkey_eq(bucket, bp_pos_to_bucket(c, ret))); - return ret; } diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 5b600c6d7aca..4ca6e7b0d8aa 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -130,7 +130,7 @@ x(bucket_gens_nonzero_for_invalid_buckets, 122) \ x(need_discard_freespace_key_to_invalid_dev_bucket, 123) \ x(need_discard_freespace_key_bad, 124) \ - x(backpointer_pos_wrong, 125) \ + x(backpointer_bucket_offset_wrong, 125) \ x(backpointer_to_missing_device, 126) \ x(backpointer_to_missing_alloc, 127) \ x(backpointer_to_missing_ptr, 128) \ -- cgit v1.2.3-59-g8ed1b From 4864a6dd8320ad856698f93009c89f66ccb1653f Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 7 Apr 2024 18:57:56 +0300 Subject: fuse: fix wrong ff->iomode state changes from parallel dio write There is a confusion with fuse_file_uncached_io_{start,end} interface. These helpers do two things when called from passthrough open()/release(): 1. Take/drop negative refcount of fi->iocachectr (inode uncached io mode) 2. State change ff->iomode IOM_NONE <-> IOM_UNCACHED (file uncached open) The calls from parallel dio write path need to take a reference on fi->iocachectr, but they should not be changing ff->iomode state, because in this case, the fi->iocachectr reference does not stick around until file release(). Factor out helpers fuse_inode_uncached_io_{start,end}, to be used from parallel dio write path and rename fuse_file_*cached_io_{start,end} helpers to fuse_file_*cached_io_{open,release} to clarify the difference. Fixes: 205c1d802683 ("fuse: allow parallel dio writes with FUSE_DIRECT_IO_ALLOW_MMAP") Signed-off-by: Amir Goldstein Signed-off-by: Miklos Szeredi --- fs/fuse/file.c | 12 +++++++----- fs/fuse/fuse_i.h | 7 ++++--- fs/fuse/inode.c | 1 + fs/fuse/iomode.c | 56 +++++++++++++++++++++++++++++++++++++++----------------- 4 files changed, 51 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index a56e7bffd000..b57ce4157640 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1362,7 +1362,7 @@ static void fuse_dio_lock(struct kiocb *iocb, struct iov_iter *from, bool *exclusive) { struct inode *inode = file_inode(iocb->ki_filp); - struct fuse_file *ff = iocb->ki_filp->private_data; + struct fuse_inode *fi = get_fuse_inode(inode); *exclusive = fuse_dio_wr_exclusive_lock(iocb, from); if (*exclusive) { @@ -1377,7 +1377,7 @@ static void fuse_dio_lock(struct kiocb *iocb, struct iov_iter *from, * have raced, so check it again. */ if (fuse_io_past_eof(iocb, from) || - fuse_file_uncached_io_start(inode, ff, NULL) != 0) { + fuse_inode_uncached_io_start(fi, NULL) != 0) { inode_unlock_shared(inode); inode_lock(inode); *exclusive = true; @@ -1388,13 +1388,13 @@ static void fuse_dio_lock(struct kiocb *iocb, struct iov_iter *from, static void fuse_dio_unlock(struct kiocb *iocb, bool exclusive) { struct inode *inode = file_inode(iocb->ki_filp); - struct fuse_file *ff = iocb->ki_filp->private_data; + struct fuse_inode *fi = get_fuse_inode(inode); if (exclusive) { inode_unlock(inode); } else { /* Allow opens in caching mode after last parallel dio end */ - fuse_file_uncached_io_end(inode, ff); + fuse_inode_uncached_io_end(fi); inode_unlock_shared(inode); } } @@ -2574,8 +2574,10 @@ static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma) * First mmap of direct_io file enters caching inode io mode. * Also waits for parallel dio writers to go into serial mode * (exclusive instead of shared lock). + * After first mmap, the inode stays in caching io mode until + * the direct_io file release. */ - rc = fuse_file_cached_io_start(inode, ff); + rc = fuse_file_cached_io_open(inode, ff); if (rc) return rc; } diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index b24084b60864..f23919610313 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -1394,9 +1394,10 @@ int fuse_fileattr_set(struct mnt_idmap *idmap, struct dentry *dentry, struct fileattr *fa); /* iomode.c */ -int fuse_file_cached_io_start(struct inode *inode, struct fuse_file *ff); -int fuse_file_uncached_io_start(struct inode *inode, struct fuse_file *ff, struct fuse_backing *fb); -void fuse_file_uncached_io_end(struct inode *inode, struct fuse_file *ff); +int fuse_file_cached_io_open(struct inode *inode, struct fuse_file *ff); +int fuse_inode_uncached_io_start(struct fuse_inode *fi, + struct fuse_backing *fb); +void fuse_inode_uncached_io_end(struct fuse_inode *fi); int fuse_file_io_open(struct file *file, struct inode *inode); void fuse_file_io_release(struct fuse_file *ff, struct inode *inode); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 3a5d88878335..99e44ea7d875 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -175,6 +175,7 @@ static void fuse_evict_inode(struct inode *inode) } } if (S_ISREG(inode->i_mode) && !fuse_is_bad(inode)) { + WARN_ON(fi->iocachectr != 0); WARN_ON(!list_empty(&fi->write_files)); WARN_ON(!list_empty(&fi->queued_writes)); } diff --git a/fs/fuse/iomode.c b/fs/fuse/iomode.c index c653ddcf0578..98f1fd523dae 100644 --- a/fs/fuse/iomode.c +++ b/fs/fuse/iomode.c @@ -21,12 +21,13 @@ static inline bool fuse_is_io_cache_wait(struct fuse_inode *fi) } /* - * Start cached io mode. + * Called on cached file open() and on first mmap() of direct_io file. + * Takes cached_io inode mode reference to be dropped on file release. * * Blocks new parallel dio writes and waits for the in-progress parallel dio * writes to complete. */ -int fuse_file_cached_io_start(struct inode *inode, struct fuse_file *ff) +int fuse_file_cached_io_open(struct inode *inode, struct fuse_file *ff) { struct fuse_inode *fi = get_fuse_inode(inode); @@ -67,10 +68,9 @@ int fuse_file_cached_io_start(struct inode *inode, struct fuse_file *ff) return 0; } -static void fuse_file_cached_io_end(struct inode *inode, struct fuse_file *ff) +static void fuse_file_cached_io_release(struct fuse_file *ff, + struct fuse_inode *fi) { - struct fuse_inode *fi = get_fuse_inode(inode); - spin_lock(&fi->lock); WARN_ON(fi->iocachectr <= 0); WARN_ON(ff->iomode != IOM_CACHED); @@ -82,9 +82,8 @@ static void fuse_file_cached_io_end(struct inode *inode, struct fuse_file *ff) } /* Start strictly uncached io mode where cache access is not allowed */ -int fuse_file_uncached_io_start(struct inode *inode, struct fuse_file *ff, struct fuse_backing *fb) +int fuse_inode_uncached_io_start(struct fuse_inode *fi, struct fuse_backing *fb) { - struct fuse_inode *fi = get_fuse_inode(inode); struct fuse_backing *oldfb; int err = 0; @@ -99,9 +98,7 @@ int fuse_file_uncached_io_start(struct inode *inode, struct fuse_file *ff, struc err = -ETXTBSY; goto unlock; } - WARN_ON(ff->iomode != IOM_NONE); fi->iocachectr--; - ff->iomode = IOM_UNCACHED; /* fuse inode holds a single refcount of backing file */ if (!oldfb) { @@ -115,15 +112,29 @@ unlock: return err; } -void fuse_file_uncached_io_end(struct inode *inode, struct fuse_file *ff) +/* Takes uncached_io inode mode reference to be dropped on file release */ +static int fuse_file_uncached_io_open(struct inode *inode, + struct fuse_file *ff, + struct fuse_backing *fb) { struct fuse_inode *fi = get_fuse_inode(inode); + int err; + + err = fuse_inode_uncached_io_start(fi, fb); + if (err) + return err; + + WARN_ON(ff->iomode != IOM_NONE); + ff->iomode = IOM_UNCACHED; + return 0; +} + +void fuse_inode_uncached_io_end(struct fuse_inode *fi) +{ struct fuse_backing *oldfb = NULL; spin_lock(&fi->lock); WARN_ON(fi->iocachectr >= 0); - WARN_ON(ff->iomode != IOM_UNCACHED); - ff->iomode = IOM_NONE; fi->iocachectr++; if (!fi->iocachectr) { wake_up(&fi->direct_io_waitq); @@ -134,6 +145,15 @@ void fuse_file_uncached_io_end(struct inode *inode, struct fuse_file *ff) fuse_backing_put(oldfb); } +/* Drop uncached_io reference from passthrough open */ +static void fuse_file_uncached_io_release(struct fuse_file *ff, + struct fuse_inode *fi) +{ + WARN_ON(ff->iomode != IOM_UNCACHED); + ff->iomode = IOM_NONE; + fuse_inode_uncached_io_end(fi); +} + /* * Open flags that are allowed in combination with FOPEN_PASSTHROUGH. * A combination of FOPEN_PASSTHROUGH and FOPEN_DIRECT_IO means that read/write @@ -163,7 +183,7 @@ static int fuse_file_passthrough_open(struct inode *inode, struct file *file) return PTR_ERR(fb); /* First passthrough file open denies caching inode io mode */ - err = fuse_file_uncached_io_start(inode, ff, fb); + err = fuse_file_uncached_io_open(inode, ff, fb); if (!err) return 0; @@ -216,7 +236,7 @@ int fuse_file_io_open(struct file *file, struct inode *inode) if (ff->open_flags & FOPEN_PASSTHROUGH) err = fuse_file_passthrough_open(inode, file); else - err = fuse_file_cached_io_start(inode, ff); + err = fuse_file_cached_io_open(inode, ff); if (err) goto fail; @@ -236,8 +256,10 @@ fail: /* No more pending io and no new io possible to inode via open/mmapped file */ void fuse_file_io_release(struct fuse_file *ff, struct inode *inode) { + struct fuse_inode *fi = get_fuse_inode(inode); + /* - * Last parallel dio close allows caching inode io mode. + * Last passthrough file close allows caching inode io mode. * Last caching file close exits caching inode io mode. */ switch (ff->iomode) { @@ -245,10 +267,10 @@ void fuse_file_io_release(struct fuse_file *ff, struct inode *inode) /* Nothing to do */ break; case IOM_UNCACHED: - fuse_file_uncached_io_end(inode, ff); + fuse_file_uncached_io_release(ff, fi); break; case IOM_CACHED: - fuse_file_cached_io_end(inode, ff); + fuse_file_cached_io_release(ff, fi); break; } } -- cgit v1.2.3-59-g8ed1b From 7cc911262835419fe469ebfae89891c0e97c62ef Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Sun, 7 Apr 2024 18:57:57 +0300 Subject: fuse: fix parallel dio write on file open in passthrough mode Parallel dio write takes a negative refcount of fi->iocachectr and so does open of file in passthrough mode. The refcount of passthrough mode is associated with attach/detach of a fuse_backing object to fuse inode. For parallel dio write, the backing file is irrelevant, so the call to fuse_inode_uncached_io_start() passes a NULL fuse_backing object. Passing a NULL fuse_backing will result in false -EBUSY error if the file is already open in passthrough mode. Allow taking negative fi->iocachectr refcount with NULL fuse_backing, because it does not conflict with an already attached fuse_backing object. Fixes: 4a90451bbc7f ("fuse: implement open in passthrough mode") Signed-off-by: Amir Goldstein Signed-off-by: Miklos Szeredi --- fs/fuse/iomode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/fuse/iomode.c b/fs/fuse/iomode.c index 98f1fd523dae..c99e285f3183 100644 --- a/fs/fuse/iomode.c +++ b/fs/fuse/iomode.c @@ -90,7 +90,7 @@ int fuse_inode_uncached_io_start(struct fuse_inode *fi, struct fuse_backing *fb) spin_lock(&fi->lock); /* deny conflicting backing files on same fuse inode */ oldfb = fuse_inode_backing(fi); - if (oldfb && oldfb != fb) { + if (fb && oldfb && oldfb != fb) { err = -EBUSY; goto unlock; } @@ -101,7 +101,7 @@ int fuse_inode_uncached_io_start(struct fuse_inode *fi, struct fuse_backing *fb) fi->iocachectr--; /* fuse inode holds a single refcount of backing file */ - if (!oldfb) { + if (fb && !oldfb) { oldfb = fuse_inode_backing_set(fi, fb); WARN_ON_ONCE(oldfb != NULL); } else { -- cgit v1.2.3-59-g8ed1b From eb4b691b9115fae4c844f5941418335575cf667f Mon Sep 17 00:00:00 2001 From: Danny Lin Date: Sat, 13 Apr 2024 17:34:31 -0700 Subject: fuse: fix leaked ENOSYS error on first statx call FUSE attempts to detect server support for statx by trying it once and setting no_statx=1 if it fails with ENOSYS, but consider the following scenario: - Userspace (e.g. sh) calls stat() on a file * succeeds - Userspace (e.g. lsd) calls statx(BTIME) on the same file - request_mask = STATX_BASIC_STATS | STATX_BTIME - first pass: sync=true due to differing cache_mask - statx fails and returns ENOSYS - set no_statx and retry - retry sets mask = STATX_BASIC_STATS - now mask == cache_mask; sync=false (time_before: still valid) - so we take the "else if (stat)" path - "err" is still ENOSYS from the failed statx call Fix this by zeroing "err" before retrying the failed call. Fixes: d3045530bdd2 ("fuse: implement statx") Cc: stable@vger.kernel.org # v6.6 Signed-off-by: Danny Lin Signed-off-by: Miklos Szeredi --- fs/fuse/dir.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 4a6df591add6..2b0d4781f394 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1321,6 +1321,7 @@ retry: err = fuse_do_statx(inode, file, stat); if (err == -ENOSYS) { fc->no_statx = 1; + err = 0; goto retry; } } else { -- cgit v1.2.3-59-g8ed1b From 09492cb45100cab909cabe164deb7cdc14e38634 Mon Sep 17 00:00:00 2001 From: Yang Li Date: Fri, 15 Mar 2024 16:02:53 +0800 Subject: cuse: add kernel-doc comments to cuse_process_init_reply() This commit adds kernel-doc style comments with complete parameter descriptions for the function cuse_process_init_reply. Signed-off-by: Yang Li Signed-off-by: Miklos Szeredi --- fs/fuse/cuse.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c index b6cad106c37e..0b2da7b7e2ad 100644 --- a/fs/fuse/cuse.c +++ b/fs/fuse/cuse.c @@ -310,6 +310,10 @@ struct cuse_init_args { /** * cuse_process_init_reply - finish initializing CUSE channel * + * @fm: The fuse mount information containing the CUSE connection. + * @args: The arguments passed to the init reply. + * @error: The error code signifying if any error occurred during the process. + * * This function creates the character device and sets up all the * required data structures for it. Please read the comment at the * top of this file for high level overview. -- cgit v1.2.3-59-g8ed1b From 0fac04e4e0ea9ca51ac16d1a3d0e5dfe2919a6dd Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 12 Apr 2024 08:16:14 +0200 Subject: iomap: convert iomap_writepages to writeack_iter This removes one indirect function call per folio, and adds type safety by not casting through a void pointer. Based on a patch by Matthew Wilcox. Signed-off-by: Christoph Hellwig Link: https://lore.kernel.org/r/20240412061614.1511629-1-hch@lst.de Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 4e8e41c8b3c0..e09441f4fceb 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -1958,18 +1958,13 @@ static int iomap_writepage_map(struct iomap_writepage_ctx *wpc, return error; } -static int iomap_do_writepage(struct folio *folio, - struct writeback_control *wbc, void *data) -{ - return iomap_writepage_map(data, wbc, folio); -} - int iomap_writepages(struct address_space *mapping, struct writeback_control *wbc, struct iomap_writepage_ctx *wpc, const struct iomap_writeback_ops *ops) { - int ret; + struct folio *folio = NULL; + int error; /* * Writeback from reclaim context should never happen except in the case @@ -1980,8 +1975,9 @@ iomap_writepages(struct address_space *mapping, struct writeback_control *wbc, return -EIO; wpc->ops = ops; - ret = write_cache_pages(mapping, wbc, iomap_do_writepage, wpc); - return iomap_submit_ioend(wpc, ret); + while ((folio = writeback_iter(mapping, wbc, folio, &error))) + error = iomap_writepage_map(wpc, wbc, folio); + return iomap_submit_ioend(wpc, error); } EXPORT_SYMBOL_GPL(iomap_writepages); -- cgit v1.2.3-59-g8ed1b From 3078e059a5e984663c5c2b04485375c84c2700f9 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 12 Apr 2024 14:36:38 +0800 Subject: bcachefs: fix error path of __bch2_read_super() In __bch2_read_super(), if kstrdup() fails, it needs to release memory in sb->holder, fix to call bch2_free_super() in the error path. Signed-off-by: Chao Yu Reviewed-by: Hongbo Li Signed-off-by: Kent Overstreet --- fs/bcachefs/super-io.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index e0aa3655b63b..648986a7dbde 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -700,8 +700,11 @@ retry: return -ENOMEM; sb->sb_name = kstrdup(path, GFP_KERNEL); - if (!sb->sb_name) - return -ENOMEM; + if (!sb->sb_name) { + ret = -ENOMEM; + prt_printf(&err, "error allocating memory for sb_name"); + goto err; + } #ifndef __KERNEL__ if (opt_get(*opts, direct_io) == false) -- cgit v1.2.3-59-g8ed1b From ad29cf999a9161e7849aa229d2028854f90728c2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 18:02:15 -0400 Subject: bcachefs: set_btree_iter_dontneed also clears should_be_locked This is part of a larger series cleaning up the semantics of should_be_locked and adding assertions around it; if we don't need an iterator/path anymore, it clearly doesn't need to be locked. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 1d58d447b386..1c70836dd7cc 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -498,8 +498,13 @@ static inline void set_btree_iter_dontneed(struct btree_iter *iter) { struct btree_trans *trans = iter->trans; - if (!trans->restarted) - btree_iter_path(trans, iter)->preserve = false; + if (!iter->path || trans->restarted) + return; + + struct btree_path *path = btree_iter_path(trans, iter); + path->preserve = false; + if (path->ref == 1) + path->should_be_locked = false; } void *__bch2_trans_kmalloc(struct btree_trans *, size_t); -- cgit v1.2.3-59-g8ed1b From f60d374d2cc88034385265d193a38e3f4a4b430c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 4 Jan 2024 21:35:38 -0500 Subject: close_on_exec(): pass files_struct instead of fdtable both callers are happier that way... Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/file.c | 5 +---- fs/proc/fd.c | 4 +--- include/linux/fdtable.h | 10 +++++----- 3 files changed, 7 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/file.c b/fs/file.c index 3b683b9101d8..eff5ce79f66a 100644 --- a/fs/file.c +++ b/fs/file.c @@ -1219,12 +1219,9 @@ void set_close_on_exec(unsigned int fd, int flag) bool get_close_on_exec(unsigned int fd) { - struct files_struct *files = current->files; - struct fdtable *fdt; bool res; rcu_read_lock(); - fdt = files_fdtable(files); - res = close_on_exec(fd, fdt); + res = close_on_exec(fd, current->files); rcu_read_unlock(); return res; } diff --git a/fs/proc/fd.c b/fs/proc/fd.c index 6e72e5ad42bc..0139aae19651 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -39,10 +39,8 @@ static int seq_show(struct seq_file *m, void *v) spin_lock(&files->file_lock); file = files_lookup_fd_locked(files, fd); if (file) { - struct fdtable *fdt = files_fdtable(files); - f_flags = file->f_flags; - if (close_on_exec(fd, fdt)) + if (close_on_exec(fd, files)) f_flags |= O_CLOEXEC; get_file(file); diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h index 78c8326d74ae..cc060b20502b 100644 --- a/include/linux/fdtable.h +++ b/include/linux/fdtable.h @@ -33,11 +33,6 @@ struct fdtable { struct rcu_head rcu; }; -static inline bool close_on_exec(unsigned int fd, const struct fdtable *fdt) -{ - return test_bit(fd, fdt->close_on_exec); -} - static inline bool fd_is_open(unsigned int fd, const struct fdtable *fdt) { return test_bit(fd, fdt->open_fds); @@ -107,6 +102,11 @@ struct file *lookup_fdget_rcu(unsigned int fd); struct file *task_lookup_fdget_rcu(struct task_struct *task, unsigned int fd); struct file *task_lookup_next_fdget_rcu(struct task_struct *task, unsigned int *fd); +static inline bool close_on_exec(unsigned int fd, const struct files_struct *files) +{ + return test_bit(fd, files_fdtable(files)->close_on_exec); +} + struct task_struct; void put_files_struct(struct files_struct *fs); -- cgit v1.2.3-59-g8ed1b From c4aab26253cd1f302279b8d6b5b66ccf1b120520 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 4 Jan 2024 21:45:47 -0500 Subject: fd_is_open(): move to fs/file.c no users outside that... Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/file.c | 5 +++++ include/linux/fdtable.h | 5 ----- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/file.c b/fs/file.c index eff5ce79f66a..ab38b005633c 100644 --- a/fs/file.c +++ b/fs/file.c @@ -271,6 +271,11 @@ static inline void __clear_open_fd(unsigned int fd, struct fdtable *fdt) __clear_bit(fd / BITS_PER_LONG, fdt->full_fds_bits); } +static inline bool fd_is_open(unsigned int fd, const struct fdtable *fdt) +{ + return test_bit(fd, fdt->open_fds); +} + static unsigned int count_open_files(struct fdtable *fdt) { unsigned int size = fdt->max_fds; diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h index cc060b20502b..2944d4aa413b 100644 --- a/include/linux/fdtable.h +++ b/include/linux/fdtable.h @@ -33,11 +33,6 @@ struct fdtable { struct rcu_head rcu; }; -static inline bool fd_is_open(unsigned int fd, const struct fdtable *fdt) -{ - return test_bit(fd, fdt->open_fds); -} - /* * Open file table structure */ -- cgit v1.2.3-59-g8ed1b From 613aee94dddf07de9fde8dd4fcb6e4579cde8a65 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Jan 2024 06:20:36 -0500 Subject: get_file_rcu(): no need to check for NULL separately IS_ERR(NULL) is false and IS_ERR() already comes with unlikely()... Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/file.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/file.c b/fs/file.c index ab38b005633c..8076aef9c210 100644 --- a/fs/file.c +++ b/fs/file.c @@ -920,13 +920,8 @@ struct file *get_file_rcu(struct file __rcu **f) struct file __rcu *file; file = __get_file_rcu(f); - if (unlikely(!file)) - return NULL; - - if (unlikely(IS_ERR(file))) - continue; - - return file; + if (!IS_ERR(file)) + return file; } } EXPORT_SYMBOL_GPL(get_file_rcu); -- cgit v1.2.3-59-g8ed1b From af58dc1f50c1946018773beca23ebaad587b9cc9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Jan 2024 06:24:55 -0500 Subject: kernel_file_open(): get rid of inode argument always equal to ->dentry->d_inode of the path argument these days. Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/cachefiles/namei.c | 3 +-- fs/open.c | 5 ++--- fs/overlayfs/util.c | 2 +- include/linux/fs.h | 2 +- 4 files changed, 5 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index 7ade836beb58..f53977169db4 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -563,8 +563,7 @@ static bool cachefiles_open_file(struct cachefiles_object *object, */ path.mnt = cache->mnt; path.dentry = dentry; - file = kernel_file_open(&path, O_RDWR | O_LARGEFILE | O_DIRECT, - d_backing_inode(dentry), cache->cache_cred); + file = kernel_file_open(&path, O_RDWR | O_LARGEFILE | O_DIRECT, cache->cache_cred); if (IS_ERR(file)) { trace_cachefiles_vfs_error(object, d_backing_inode(dentry), PTR_ERR(file), diff --git a/fs/open.c b/fs/open.c index ee8460c83c77..ec287ac67e7f 100644 --- a/fs/open.c +++ b/fs/open.c @@ -1155,7 +1155,6 @@ EXPORT_SYMBOL(dentry_create); * kernel_file_open - open a file for kernel internal use * @path: path of the file to open * @flags: open flags - * @inode: the inode * @cred: credentials for open * * Open a file for use by in-kernel consumers. The file is not accounted @@ -1165,7 +1164,7 @@ EXPORT_SYMBOL(dentry_create); * Return: Opened file on success, an error pointer on failure. */ struct file *kernel_file_open(const struct path *path, int flags, - struct inode *inode, const struct cred *cred) + const struct cred *cred) { struct file *f; int error; @@ -1175,7 +1174,7 @@ struct file *kernel_file_open(const struct path *path, int flags, return f; f->f_path = *path; - error = do_dentry_open(f, inode, NULL); + error = do_dentry_open(f, d_inode(path->dentry), NULL); if (error) { fput(f); f = ERR_PTR(error); diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c index d285d1d7baad..edc9216f6e27 100644 --- a/fs/overlayfs/util.c +++ b/fs/overlayfs/util.c @@ -1376,7 +1376,7 @@ int ovl_ensure_verity_loaded(struct path *datapath) * If this inode was not yet opened, the verity info hasn't been * loaded yet, so we need to do that here to force it into memory. */ - filp = kernel_file_open(datapath, O_RDONLY, inode, current_cred()); + filp = kernel_file_open(datapath, O_RDONLY, current_cred()); if (IS_ERR(filp)) return PTR_ERR(filp); fput(filp); diff --git a/include/linux/fs.h b/include/linux/fs.h index 00fc429b0af0..143e967a3af2 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1906,7 +1906,7 @@ struct file *kernel_tmpfile_open(struct mnt_idmap *idmap, umode_t mode, int open_flag, const struct cred *cred); struct file *kernel_file_open(const struct path *path, int flags, - struct inode *inode, const struct cred *cred); + const struct cred *cred); int vfs_mkobj(struct dentry *, umode_t, int (*f)(struct dentry *, umode_t, void *), -- cgit v1.2.3-59-g8ed1b From 0f4a2cebe256dab4e9b8836b87ce4c909cd585eb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 20 Jan 2024 06:28:52 -0500 Subject: do_dentry_open(): kill inode argument should've been done as soon as overlayfs stopped messing with fake paths... Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/open.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/open.c b/fs/open.c index ec287ac67e7f..89cafb572061 100644 --- a/fs/open.c +++ b/fs/open.c @@ -902,10 +902,10 @@ cleanup_inode: } static int do_dentry_open(struct file *f, - struct inode *inode, int (*open)(struct inode *, struct file *)) { static const struct file_operations empty_fops = {}; + struct inode *inode = f->f_path.dentry->d_inode; int error; path_get(&f->f_path); @@ -1047,7 +1047,7 @@ int finish_open(struct file *file, struct dentry *dentry, BUG_ON(file->f_mode & FMODE_OPENED); /* once it's opened, it's opened */ file->f_path.dentry = dentry; - return do_dentry_open(file, d_backing_inode(dentry), open); + return do_dentry_open(file, open); } EXPORT_SYMBOL(finish_open); @@ -1086,7 +1086,7 @@ EXPORT_SYMBOL(file_path); int vfs_open(const struct path *path, struct file *file) { file->f_path = *path; - return do_dentry_open(file, d_backing_inode(path->dentry), NULL); + return do_dentry_open(file, NULL); } struct file *dentry_open(const struct path *path, int flags, @@ -1174,7 +1174,7 @@ struct file *kernel_file_open(const struct path *path, int flags, return f; f->f_path = *path; - error = do_dentry_open(f, d_inode(path->dentry), NULL); + error = do_dentry_open(f, NULL); if (error) { fput(f); f = ERR_PTR(error); -- cgit v1.2.3-59-g8ed1b From 7c98f7cb8fda964fbc60b9307ad35e94735fa35f Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Mon, 28 Aug 2023 17:13:18 +0200 Subject: remove call_{read,write}_iter() functions These have no clear purpose. This is effectively a revert of commit bb7462b6fd64 ("vfs: use helpers for calling f_op->{read,write}_iter()"). The patch was created with the help of a coccinelle script. Fixes: bb7462b6fd64 ("vfs: use helpers for calling f_op->{read,write}_iter()") Reviewed-by: Christian Brauner Signed-off-by: Miklos Szeredi Signed-off-by: Al Viro --- drivers/block/loop.c | 4 ++-- drivers/target/target_core_file.c | 4 ++-- fs/aio.c | 4 ++-- fs/read_write.c | 12 ++++++------ fs/splice.c | 4 ++-- include/linux/fs.h | 12 ------------ io_uring/rw.c | 4 ++-- 7 files changed, 16 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 28a95fd366fe..93780f41646b 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -445,9 +445,9 @@ static int lo_rw_aio(struct loop_device *lo, struct loop_cmd *cmd, cmd->iocb.ki_ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_NONE, 0); if (rw == ITER_SOURCE) - ret = call_write_iter(file, &cmd->iocb, &iter); + ret = file->f_op->write_iter(&cmd->iocb, &iter); else - ret = call_read_iter(file, &cmd->iocb, &iter); + ret = file->f_op->read_iter(&cmd->iocb, &iter); lo_rw_aio_do_completion(cmd); diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c index 4d447520bab8..94e6cd4e7e43 100644 --- a/drivers/target/target_core_file.c +++ b/drivers/target/target_core_file.c @@ -299,9 +299,9 @@ fd_execute_rw_aio(struct se_cmd *cmd, struct scatterlist *sgl, u32 sgl_nents, aio_cmd->iocb.ki_flags |= IOCB_DSYNC; if (is_write) - ret = call_write_iter(file, &aio_cmd->iocb, &iter); + ret = file->f_op->write_iter(&aio_cmd->iocb, &iter); else - ret = call_read_iter(file, &aio_cmd->iocb, &iter); + ret = file->f_op->read_iter(&aio_cmd->iocb, &iter); if (ret != -EIOCBQUEUED) cmd_rw_aio_complete(&aio_cmd->iocb, ret); diff --git a/fs/aio.c b/fs/aio.c index 9cdaa2faa536..13ca81c7c744 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1605,7 +1605,7 @@ static int aio_read(struct kiocb *req, const struct iocb *iocb, return ret; ret = rw_verify_area(READ, file, &req->ki_pos, iov_iter_count(&iter)); if (!ret) - aio_rw_done(req, call_read_iter(file, req, &iter)); + aio_rw_done(req, file->f_op->read_iter(req, &iter)); kfree(iovec); return ret; } @@ -1636,7 +1636,7 @@ static int aio_write(struct kiocb *req, const struct iocb *iocb, if (S_ISREG(file_inode(file)->i_mode)) kiocb_start_write(req); req->ki_flags |= IOCB_WRITE; - aio_rw_done(req, call_write_iter(file, req, &iter)); + aio_rw_done(req, file->f_op->write_iter(req, &iter)); } kfree(iovec); return ret; diff --git a/fs/read_write.c b/fs/read_write.c index d4c036e82b6c..2de7f6adb33d 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -392,7 +392,7 @@ static ssize_t new_sync_read(struct file *filp, char __user *buf, size_t len, lo kiocb.ki_pos = (ppos ? *ppos : 0); iov_iter_ubuf(&iter, ITER_DEST, buf, len); - ret = call_read_iter(filp, &kiocb, &iter); + ret = filp->f_op->read_iter(&kiocb, &iter); BUG_ON(ret == -EIOCBQUEUED); if (ppos) *ppos = kiocb.ki_pos; @@ -494,7 +494,7 @@ static ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t kiocb.ki_pos = (ppos ? *ppos : 0); iov_iter_ubuf(&iter, ITER_SOURCE, (void __user *)buf, len); - ret = call_write_iter(filp, &kiocb, &iter); + ret = filp->f_op->write_iter(&kiocb, &iter); BUG_ON(ret == -EIOCBQUEUED); if (ret > 0 && ppos) *ppos = kiocb.ki_pos; @@ -736,9 +736,9 @@ static ssize_t do_iter_readv_writev(struct file *filp, struct iov_iter *iter, kiocb.ki_pos = (ppos ? *ppos : 0); if (type == READ) - ret = call_read_iter(filp, &kiocb, iter); + ret = filp->f_op->read_iter(&kiocb, iter); else - ret = call_write_iter(filp, &kiocb, iter); + ret = filp->f_op->write_iter(&kiocb, iter); BUG_ON(ret == -EIOCBQUEUED); if (ppos) *ppos = kiocb.ki_pos; @@ -799,7 +799,7 @@ ssize_t vfs_iocb_iter_read(struct file *file, struct kiocb *iocb, if (ret < 0) return ret; - ret = call_read_iter(file, iocb, iter); + ret = file->f_op->read_iter(iocb, iter); out: if (ret >= 0) fsnotify_access(file); @@ -860,7 +860,7 @@ ssize_t vfs_iocb_iter_write(struct file *file, struct kiocb *iocb, return ret; kiocb_start_write(iocb); - ret = call_write_iter(file, iocb, iter); + ret = file->f_op->write_iter(iocb, iter); if (ret != -EIOCBQUEUED) kiocb_end_write(iocb); if (ret > 0) diff --git a/fs/splice.c b/fs/splice.c index 218e24b1ac40..60aed8de21f8 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -362,7 +362,7 @@ ssize_t copy_splice_read(struct file *in, loff_t *ppos, iov_iter_bvec(&to, ITER_DEST, bv, npages, len); init_sync_kiocb(&kiocb, in); kiocb.ki_pos = *ppos; - ret = call_read_iter(in, &kiocb, &to); + ret = in->f_op->read_iter(&kiocb, &to); if (ret > 0) { keep = DIV_ROUND_UP(ret, PAGE_SIZE); @@ -740,7 +740,7 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out, iov_iter_bvec(&from, ITER_SOURCE, array, n, sd.total_len - left); init_sync_kiocb(&kiocb, out); kiocb.ki_pos = sd.pos; - ret = call_write_iter(out, &kiocb, &from); + ret = out->f_op->write_iter(&kiocb, &from); sd.pos = kiocb.ki_pos; if (ret <= 0) break; diff --git a/include/linux/fs.h b/include/linux/fs.h index 143e967a3af2..a94343f567cb 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2096,18 +2096,6 @@ struct inode_operations { struct offset_ctx *(*get_offset_ctx)(struct inode *inode); } ____cacheline_aligned; -static inline ssize_t call_read_iter(struct file *file, struct kiocb *kio, - struct iov_iter *iter) -{ - return file->f_op->read_iter(kio, iter); -} - -static inline ssize_t call_write_iter(struct file *file, struct kiocb *kio, - struct iov_iter *iter) -{ - return file->f_op->write_iter(kio, iter); -} - static inline int call_mmap(struct file *file, struct vm_area_struct *vma) { return file->f_op->mmap(file, vma); diff --git a/io_uring/rw.c b/io_uring/rw.c index 0585ebcc9773..7d335b7e00ed 100644 --- a/io_uring/rw.c +++ b/io_uring/rw.c @@ -701,7 +701,7 @@ static inline int io_iter_do_read(struct io_rw *rw, struct iov_iter *iter) struct file *file = rw->kiocb.ki_filp; if (likely(file->f_op->read_iter)) - return call_read_iter(file, &rw->kiocb, iter); + return file->f_op->read_iter(&rw->kiocb, iter); else if (file->f_op->read) return loop_rw_iter(READ, rw, iter); else @@ -1047,7 +1047,7 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags) kiocb->ki_flags |= IOCB_WRITE; if (likely(req->file->f_op->write_iter)) - ret2 = call_write_iter(req->file, kiocb, &s->iter); + ret2 = req->file->f_op->write_iter(kiocb, &s->iter); else if (req->file->f_op->write) ret2 = loop_rw_iter(WRITE, rw, &s->iter); else -- cgit v1.2.3-59-g8ed1b From 549d3c9a29921f388ef3bcfd1d4f669b7dd4eed2 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:03 -0700 Subject: xfs: pass xfs_buf lookup flags to xfs_*read_agi Allow callers to pass buffer lookup flags to xfs_read_agi and xfs_ialloc_read_agi. This will be used in the next patch to fix a deadlock in the online fsck inode scanner. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_ag.c | 8 ++++---- fs/xfs/libxfs/xfs_ialloc.c | 16 ++++++++++------ fs/xfs/libxfs/xfs_ialloc.h | 5 +++-- fs/xfs/libxfs/xfs_ialloc_btree.c | 4 ++-- fs/xfs/scrub/common.c | 4 ++-- fs/xfs/scrub/fscounters.c | 2 +- fs/xfs/scrub/iscan.c | 2 +- fs/xfs/scrub/repair.c | 6 +++--- fs/xfs/xfs_inode.c | 8 ++++---- fs/xfs/xfs_iwalk.c | 4 ++-- fs/xfs/xfs_log_recover.c | 4 ++-- 11 files changed, 34 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index dc1873f76bff..09fe9412eab4 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -194,7 +194,7 @@ xfs_initialize_perag_data( pag = xfs_perag_get(mp, index); error = xfs_alloc_read_agf(pag, NULL, 0, NULL); if (!error) - error = xfs_ialloc_read_agi(pag, NULL, NULL); + error = xfs_ialloc_read_agi(pag, NULL, 0, NULL); if (error) { xfs_perag_put(pag); return error; @@ -931,7 +931,7 @@ xfs_ag_shrink_space( int error, err2; ASSERT(pag->pag_agno == mp->m_sb.sb_agcount - 1); - error = xfs_ialloc_read_agi(pag, *tpp, &agibp); + error = xfs_ialloc_read_agi(pag, *tpp, 0, &agibp); if (error) return error; @@ -1062,7 +1062,7 @@ xfs_ag_extend_space( ASSERT(pag->pag_agno == pag->pag_mount->m_sb.sb_agcount - 1); - error = xfs_ialloc_read_agi(pag, tp, &bp); + error = xfs_ialloc_read_agi(pag, tp, 0, &bp); if (error) return error; @@ -1119,7 +1119,7 @@ xfs_ag_get_geometry( int error; /* Lock the AG headers. */ - error = xfs_ialloc_read_agi(pag, NULL, &agi_bp); + error = xfs_ialloc_read_agi(pag, NULL, 0, &agi_bp); if (error) return error; error = xfs_alloc_read_agf(pag, NULL, 0, &agf_bp); diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index e5ac3e5430c4..cb37f0007731 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -1699,7 +1699,7 @@ xfs_dialloc_good_ag( return false; if (!xfs_perag_initialised_agi(pag)) { - error = xfs_ialloc_read_agi(pag, tp, NULL); + error = xfs_ialloc_read_agi(pag, tp, 0, NULL); if (error) return false; } @@ -1768,7 +1768,7 @@ xfs_dialloc_try_ag( * Then read in the AGI buffer and recheck with the AGI buffer * lock held. */ - error = xfs_ialloc_read_agi(pag, *tpp, &agbp); + error = xfs_ialloc_read_agi(pag, *tpp, 0, &agbp); if (error) return error; @@ -2286,7 +2286,7 @@ xfs_difree( /* * Get the allocation group header. */ - error = xfs_ialloc_read_agi(pag, tp, &agbp); + error = xfs_ialloc_read_agi(pag, tp, 0, &agbp); if (error) { xfs_warn(mp, "%s: xfs_ialloc_read_agi() returned error %d.", __func__, error); @@ -2332,7 +2332,7 @@ xfs_imap_lookup( int error; int i; - error = xfs_ialloc_read_agi(pag, tp, &agbp); + error = xfs_ialloc_read_agi(pag, tp, 0, &agbp); if (error) { xfs_alert(mp, "%s: xfs_ialloc_read_agi() returned error %d, agno %d", @@ -2675,6 +2675,7 @@ int xfs_read_agi( struct xfs_perag *pag, struct xfs_trans *tp, + xfs_buf_flags_t flags, struct xfs_buf **agibpp) { struct xfs_mount *mp = pag->pag_mount; @@ -2684,7 +2685,7 @@ xfs_read_agi( error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, XFS_AG_DADDR(mp, pag->pag_agno, XFS_AGI_DADDR(mp)), - XFS_FSS_TO_BB(mp, 1), 0, agibpp, &xfs_agi_buf_ops); + XFS_FSS_TO_BB(mp, 1), flags, agibpp, &xfs_agi_buf_ops); if (xfs_metadata_is_sick(error)) xfs_ag_mark_sick(pag, XFS_SICK_AG_AGI); if (error) @@ -2704,6 +2705,7 @@ int xfs_ialloc_read_agi( struct xfs_perag *pag, struct xfs_trans *tp, + int flags, struct xfs_buf **agibpp) { struct xfs_buf *agibp; @@ -2712,7 +2714,9 @@ xfs_ialloc_read_agi( trace_xfs_ialloc_read_agi(pag->pag_mount, pag->pag_agno); - error = xfs_read_agi(pag, tp, &agibp); + error = xfs_read_agi(pag, tp, + (flags & XFS_IALLOC_FLAG_TRYLOCK) ? XBF_TRYLOCK : 0, + &agibp); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h index f1412183bb44..b549627e3a61 100644 --- a/fs/xfs/libxfs/xfs_ialloc.h +++ b/fs/xfs/libxfs/xfs_ialloc.h @@ -63,10 +63,11 @@ xfs_ialloc_log_agi( struct xfs_buf *bp, /* allocation group header buffer */ uint32_t fields); /* bitmask of fields to log */ -int xfs_read_agi(struct xfs_perag *pag, struct xfs_trans *tp, +int xfs_read_agi(struct xfs_perag *pag, struct xfs_trans *tp, xfs_buf_flags_t flags, struct xfs_buf **agibpp); int xfs_ialloc_read_agi(struct xfs_perag *pag, struct xfs_trans *tp, - struct xfs_buf **agibpp); + int flags, struct xfs_buf **agibpp); +#define XFS_IALLOC_FLAG_TRYLOCK (1U << 0) /* use trylock for buffer locking */ /* * Lookup a record by ino in the btree given by cur. diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index cc661fca6ff5..42e9fd47f6c7 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -745,7 +745,7 @@ xfs_finobt_count_blocks( struct xfs_btree_cur *cur; int error; - error = xfs_ialloc_read_agi(pag, tp, &agbp); + error = xfs_ialloc_read_agi(pag, tp, 0, &agbp); if (error) return error; @@ -768,7 +768,7 @@ xfs_finobt_read_blocks( struct xfs_agi *agi; int error; - error = xfs_ialloc_read_agi(pag, tp, &agbp); + error = xfs_ialloc_read_agi(pag, tp, 0, &agbp); if (error) return error; diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index 47a20cf5205f..a27d33b6f464 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -445,7 +445,7 @@ xchk_perag_read_headers( { int error; - error = xfs_ialloc_read_agi(sa->pag, sc->tp, &sa->agi_bp); + error = xfs_ialloc_read_agi(sa->pag, sc->tp, 0, &sa->agi_bp); if (error && want_ag_read_header_failure(sc, XFS_SCRUB_TYPE_AGI)) return error; @@ -827,7 +827,7 @@ again: * in the iget cache miss path. */ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, inum)); - error = xfs_ialloc_read_agi(pag, tp, agi_bpp); + error = xfs_ialloc_read_agi(pag, tp, 0, agi_bpp); xfs_perag_put(pag); if (error) return error; diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c index d310737c8823..da2f6729699d 100644 --- a/fs/xfs/scrub/fscounters.c +++ b/fs/xfs/scrub/fscounters.c @@ -85,7 +85,7 @@ xchk_fscount_warmup( continue; /* Lock both AG headers. */ - error = xfs_ialloc_read_agi(pag, sc->tp, &agi_bp); + error = xfs_ialloc_read_agi(pag, sc->tp, 0, &agi_bp); if (error) break; error = xfs_alloc_read_agf(pag, sc->tp, 0, &agf_bp); diff --git a/fs/xfs/scrub/iscan.c b/fs/xfs/scrub/iscan.c index ec3478bc505e..66ba0fbd059e 100644 --- a/fs/xfs/scrub/iscan.c +++ b/fs/xfs/scrub/iscan.c @@ -281,7 +281,7 @@ xchk_iscan_advance( if (!pag) return -ECANCELED; - ret = xfs_ialloc_read_agi(pag, sc->tp, &agi_bp); + ret = xfs_ialloc_read_agi(pag, sc->tp, 0, &agi_bp); if (ret) goto out_pag; diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index f43dce771cdd..443e62f72481 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -290,7 +290,7 @@ xrep_calc_ag_resblks( icount = pag->pagi_count; } else { /* Try to get the actual counters from disk. */ - error = xfs_ialloc_read_agi(pag, NULL, &bp); + error = xfs_ialloc_read_agi(pag, NULL, 0, &bp); if (!error) { icount = pag->pagi_count; xfs_buf_relse(bp); @@ -908,7 +908,7 @@ xrep_reinit_pagi( ASSERT(xfs_perag_initialised_agi(pag)); clear_bit(XFS_AGSTATE_AGI_INIT, &pag->pag_opstate); - error = xfs_ialloc_read_agi(pag, sc->tp, &bp); + error = xfs_ialloc_read_agi(pag, sc->tp, 0, &bp); if (error) return error; @@ -934,7 +934,7 @@ xrep_ag_init( ASSERT(!sa->pag); - error = xfs_ialloc_read_agi(pag, sc->tp, &sa->agi_bp); + error = xfs_ialloc_read_agi(pag, sc->tp, 0, &sa->agi_bp); if (error) return error; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index d55b42b2480d..3e667a19b80b 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2167,7 +2167,7 @@ xfs_iunlink( pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino)); /* Get the agi buffer first. It ensures lock ordering on the list. */ - error = xfs_read_agi(pag, tp, &agibp); + error = xfs_read_agi(pag, tp, 0, &agibp); if (error) goto out; @@ -2264,7 +2264,7 @@ xfs_iunlink_remove( trace_xfs_iunlink_remove(ip); /* Get the agi buffer first. It ensures lock ordering on the list. */ - error = xfs_read_agi(pag, tp, &agibp); + error = xfs_read_agi(pag, tp, 0, &agibp); if (error) return error; @@ -3142,7 +3142,7 @@ retry: pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, inodes[i]->i_ino)); - error = xfs_read_agi(pag, tp, &bp); + error = xfs_read_agi(pag, tp, 0, &bp); xfs_perag_put(pag); if (error) goto out_trans_cancel; @@ -3814,7 +3814,7 @@ xfs_inode_reload_unlinked_bucket( /* Grab the first inode in the list */ pag = xfs_perag_get(mp, agno); - error = xfs_ialloc_read_agi(pag, tp, &agibp); + error = xfs_ialloc_read_agi(pag, tp, 0, &agibp); xfs_perag_put(pag); if (error) return error; diff --git a/fs/xfs/xfs_iwalk.c b/fs/xfs/xfs_iwalk.c index 01b55f03a102..730c8d48da28 100644 --- a/fs/xfs/xfs_iwalk.c +++ b/fs/xfs/xfs_iwalk.c @@ -268,7 +268,7 @@ xfs_iwalk_ag_start( /* Set up a fresh cursor and empty the inobt cache. */ iwag->nr_recs = 0; - error = xfs_ialloc_read_agi(pag, tp, agi_bpp); + error = xfs_ialloc_read_agi(pag, tp, 0, agi_bpp); if (error) return error; *curpp = xfs_inobt_init_cursor(pag, tp, *agi_bpp); @@ -386,7 +386,7 @@ xfs_iwalk_run_callbacks( } /* ...and recreate the cursor just past where we left off. */ - error = xfs_ialloc_read_agi(iwag->pag, iwag->tp, agi_bpp); + error = xfs_ialloc_read_agi(iwag->pag, iwag->tp, 0, agi_bpp); if (error) return error; *curpp = xfs_inobt_init_cursor(iwag->pag, iwag->tp, *agi_bpp); diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 13f1d2e91540..1b1f0a4cd494 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2656,7 +2656,7 @@ xlog_recover_clear_agi_bucket( if (error) goto out_error; - error = xfs_read_agi(pag, tp, &agibp); + error = xfs_read_agi(pag, tp, 0, &agibp); if (error) goto out_abort; @@ -2772,7 +2772,7 @@ xlog_recover_iunlink_ag( int bucket; int error; - error = xfs_read_agi(pag, NULL, &agibp); + error = xfs_read_agi(pag, NULL, 0, &agibp); if (error) { /* * AGI is b0rked. Don't process it. -- cgit v1.2.3-59-g8ed1b From 2afd5276d314d775ae0bdaa6ec22069515bd9c70 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:04 -0700 Subject: xfs: fix an AGI lock acquisition ordering problem in xrep_dinode_findmode While reviewing the next patch which fixes an ABBA deadlock between the AGI and a directory ILOCK, someone asked a question about why we're holding the AGI in the first place. The reason for that is to quiesce the inode structures for that AG while we do a repair. I then realized that the xrep_dinode_findmode invokes xchk_iscan_iter, which walks the inobts (and hence the AGIs) to find all the inodes. This itself is also an ABBA vector, since the damaged inode could be in AG 5, which we hold while we scan AG 0 for directories. 5 -> 0 is not allowed. To address this, modify the iscan to allow trylock of the AGI buffer using the flags argument to xfs_ialloc_read_agi that the previous patch added. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/inode_repair.c | 1 + fs/xfs/scrub/iscan.c | 36 +++++++++++++++++++++++++++++++++++- fs/xfs/scrub/iscan.h | 15 +++++++++++++++ fs/xfs/scrub/trace.h | 10 ++++++++-- 4 files changed, 59 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c index eab380e95ef4..35da0193c919 100644 --- a/fs/xfs/scrub/inode_repair.c +++ b/fs/xfs/scrub/inode_repair.c @@ -356,6 +356,7 @@ xrep_dinode_find_mode( * so there's a real possibility that _iscan_iter can return EBUSY. */ xchk_iscan_start(sc, 5000, 100, &ri->ftype_iscan); + xchk_iscan_set_agi_trylock(&ri->ftype_iscan); ri->ftype_iscan.skip_ino = sc->sm->sm_ino; ri->alleged_ftype = XFS_DIR3_FT_UNKNOWN; while ((error = xchk_iscan_iter(&ri->ftype_iscan, &dp)) == 1) { diff --git a/fs/xfs/scrub/iscan.c b/fs/xfs/scrub/iscan.c index 66ba0fbd059e..c643b7d79b60 100644 --- a/fs/xfs/scrub/iscan.c +++ b/fs/xfs/scrub/iscan.c @@ -243,6 +243,40 @@ xchk_iscan_finish( mutex_unlock(&iscan->lock); } +/* + * Grab the AGI to advance the inode scan. Returns 0 if *agi_bpp is now set, + * -ECANCELED if the live scan aborted, -EBUSY if the AGI could not be grabbed, + * or the usual negative errno. + */ +STATIC int +xchk_iscan_read_agi( + struct xchk_iscan *iscan, + struct xfs_perag *pag, + struct xfs_buf **agi_bpp) +{ + struct xfs_scrub *sc = iscan->sc; + unsigned long relax; + int ret; + + if (!xchk_iscan_agi_needs_trylock(iscan)) + return xfs_ialloc_read_agi(pag, sc->tp, 0, agi_bpp); + + relax = msecs_to_jiffies(iscan->iget_retry_delay); + do { + ret = xfs_ialloc_read_agi(pag, sc->tp, XFS_IALLOC_FLAG_TRYLOCK, + agi_bpp); + if (ret != -EAGAIN) + return ret; + if (!iscan->iget_timeout || + time_is_before_jiffies(iscan->__iget_deadline)) + return -EBUSY; + + trace_xchk_iscan_agi_retry_wait(iscan); + } while (!schedule_timeout_killable(relax) && + !xchk_iscan_aborted(iscan)); + return -ECANCELED; +} + /* * Advance ino to the next inode that the inobt thinks is allocated, being * careful to jump to the next AG if we've reached the right end of this AG's @@ -281,7 +315,7 @@ xchk_iscan_advance( if (!pag) return -ECANCELED; - ret = xfs_ialloc_read_agi(pag, sc->tp, 0, &agi_bp); + ret = xchk_iscan_read_agi(iscan, pag, &agi_bp); if (ret) goto out_pag; diff --git a/fs/xfs/scrub/iscan.h b/fs/xfs/scrub/iscan.h index 71f657552dfa..5e0e4ed9dea6 100644 --- a/fs/xfs/scrub/iscan.h +++ b/fs/xfs/scrub/iscan.h @@ -59,6 +59,9 @@ struct xchk_iscan { /* Set if the scan has been aborted due to some event in the fs. */ #define XCHK_ISCAN_OPSTATE_ABORTED (1) +/* Use trylock to acquire the AGI */ +#define XCHK_ISCAN_OPSTATE_TRYLOCK_AGI (2) + static inline bool xchk_iscan_aborted(const struct xchk_iscan *iscan) { @@ -71,6 +74,18 @@ xchk_iscan_abort(struct xchk_iscan *iscan) set_bit(XCHK_ISCAN_OPSTATE_ABORTED, &iscan->__opstate); } +static inline bool +xchk_iscan_agi_needs_trylock(const struct xchk_iscan *iscan) +{ + return test_bit(XCHK_ISCAN_OPSTATE_TRYLOCK_AGI, &iscan->__opstate); +} + +static inline void +xchk_iscan_set_agi_trylock(struct xchk_iscan *iscan) +{ + set_bit(XCHK_ISCAN_OPSTATE_TRYLOCK_AGI, &iscan->__opstate); +} + void xchk_iscan_start(struct xfs_scrub *sc, unsigned int iget_timeout, unsigned int iget_retry_delay, struct xchk_iscan *iscan); void xchk_iscan_teardown(struct xchk_iscan *iscan); diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 5b294be52c55..b1c7c79760d4 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -1300,7 +1300,7 @@ TRACE_EVENT(xchk_iscan_iget_batch, __entry->unavail) ); -TRACE_EVENT(xchk_iscan_iget_retry_wait, +DECLARE_EVENT_CLASS(xchk_iscan_retry_wait_class, TP_PROTO(struct xchk_iscan *iscan), TP_ARGS(iscan), TP_STRUCT__entry( @@ -1326,7 +1326,13 @@ TRACE_EVENT(xchk_iscan_iget_retry_wait, __entry->remaining, __entry->iget_timeout, __entry->retry_delay) -); +) +#define DEFINE_ISCAN_RETRY_WAIT_EVENT(name) \ +DEFINE_EVENT(xchk_iscan_retry_wait_class, name, \ + TP_PROTO(struct xchk_iscan *iscan), \ + TP_ARGS(iscan)) +DEFINE_ISCAN_RETRY_WAIT_EVENT(xchk_iscan_iget_retry_wait); +DEFINE_ISCAN_RETRY_WAIT_EVENT(xchk_iscan_agi_retry_wait); TRACE_EVENT(xchk_nlinks_collect_dirent, TP_PROTO(struct xfs_mount *mp, struct xfs_inode *dp, -- cgit v1.2.3-59-g8ed1b From 21ad2d03641ae70a7acdcf1212cd135dafb2a798 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:05 -0700 Subject: xfs: fix potential AGI <-> ILOCK ABBA deadlock in xrep_dinode_findmode_walk_directory xfs/399 found the following deadlock when fuzzing core.mode = ones: /proc/20506/task/20558/stack : [<0>] xfs_ilock+0xa0/0x240 [xfs] [<0>] xfs_ilock_data_map_shared+0x1b/0x20 [xfs] [<0>] xrep_dinode_findmode_walk_directory+0x69/0xe0 [xfs] [<0>] xrep_dinode_find_mode+0x103/0x2a0 [xfs] [<0>] xrep_dinode_mode+0x7c/0x120 [xfs] [<0>] xrep_dinode_core+0xed/0x2b0 [xfs] [<0>] xrep_dinode_problems+0x10/0x80 [xfs] [<0>] xrep_inode+0x6c/0xc0 [xfs] [<0>] xrep_attempt+0x64/0x1d0 [xfs] [<0>] xfs_scrub_metadata+0x365/0x840 [xfs] [<0>] xfs_scrubv_metadata+0x282/0x430 [xfs] [<0>] xfs_ioc_scrubv_metadata+0x149/0x1a0 [xfs] [<0>] xfs_file_ioctl+0xc68/0x1780 [xfs] /proc/20506/task/20559/stack : [<0>] xfs_buf_lock+0x3b/0x110 [xfs] [<0>] xfs_buf_find_lock+0x66/0x1c0 [xfs] [<0>] xfs_buf_get_map+0x208/0xc00 [xfs] [<0>] xfs_buf_read_map+0x5d/0x2c0 [xfs] [<0>] xfs_trans_read_buf_map+0x1b0/0x4c0 [xfs] [<0>] xfs_read_agi+0xbd/0x190 [xfs] [<0>] xfs_ialloc_read_agi+0x47/0x160 [xfs] [<0>] xfs_imap_lookup+0x69/0x1f0 [xfs] [<0>] xfs_imap+0x1fc/0x3d0 [xfs] [<0>] xfs_iget+0x357/0xd50 [xfs] [<0>] xchk_dir_actor+0x16e/0x330 [xfs] [<0>] xchk_dir_walk_block+0x164/0x1e0 [xfs] [<0>] xchk_dir_walk+0x13a/0x190 [xfs] [<0>] xchk_directory+0x1a2/0x2b0 [xfs] [<0>] xfs_scrub_metadata+0x2f4/0x840 [xfs] [<0>] xfs_scrubv_metadata+0x282/0x430 [xfs] [<0>] xfs_ioc_scrubv_metadata+0x149/0x1a0 [xfs] [<0>] xfs_file_ioctl+0xc68/0x1780 [xfs] Thread 20558 holds an AGI buffer and is trying to grab the ILOCK of the root directory. Thread 20559 holds the root directory ILOCK and is trying to grab the AGI of an inode that is one of the root directory's children. The AGI held by 20558 is the same buffer that 20559 is trying to acquire. In other words, this is an ABBA deadlock. In general, the lock order is ILOCK and then AGI -- rename does this while preparing for an operation involving whiteouts or renaming files out of existence; and unlink does this when moving an inode to the unlinked list. The only place where we do it in the opposite order is on the child during an icreate, but at that point the child is marked INEW and is not visible to other threads. Work around this deadlock by replacing the blocking ilock attempt with a nonblocking loop that aborts after 30 seconds. Relax for a jiffy after a failed lock attempt. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/inode_repair.c | 49 ++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 48 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c index 35da0193c919..097afba3043f 100644 --- a/fs/xfs/scrub/inode_repair.c +++ b/fs/xfs/scrub/inode_repair.c @@ -282,6 +282,51 @@ xrep_dinode_findmode_dirent( return 0; } +/* Try to lock a directory, or wait a jiffy. */ +static inline int +xrep_dinode_ilock_nowait( + struct xfs_inode *dp, + unsigned int lock_mode) +{ + if (xfs_ilock_nowait(dp, lock_mode)) + return true; + + schedule_timeout_killable(1); + return false; +} + +/* + * Try to lock a directory to look for ftype hints. Since we already hold the + * AGI buffer, we cannot block waiting for the ILOCK because rename can take + * the ILOCK and then try to lock AGIs. + */ +STATIC int +xrep_dinode_trylock_directory( + struct xrep_inode *ri, + struct xfs_inode *dp, + unsigned int *lock_modep) +{ + unsigned long deadline = jiffies + msecs_to_jiffies(30000); + unsigned int lock_mode; + int error = 0; + + do { + if (xchk_should_terminate(ri->sc, &error)) + return error; + + if (xfs_need_iread_extents(&dp->i_df)) + lock_mode = XFS_ILOCK_EXCL; + else + lock_mode = XFS_ILOCK_SHARED; + + if (xrep_dinode_ilock_nowait(dp, lock_mode)) { + *lock_modep = lock_mode; + return 0; + } + } while (!time_is_before_jiffies(deadline)); + return -EBUSY; +} + /* * If this is a directory, walk the dirents looking for any that point to the * scrub target inode. @@ -299,7 +344,9 @@ xrep_dinode_findmode_walk_directory( * Scan the directory to see if there it contains an entry pointing to * the directory that we are repairing. */ - lock_mode = xfs_ilock_data_map_shared(dp); + error = xrep_dinode_trylock_directory(ri, dp, &lock_mode); + if (error) + return error; /* * If this directory is known to be sick, we cannot scan it reliably -- cgit v1.2.3-59-g8ed1b From 98a778b42514dcd00c0356b2fbfd458636cbff87 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:06 -0700 Subject: xfs: fix error bailout in xrep_abt_build_new_trees Dan Carpenter reports: "Commit 4bdfd7d15747 ("xfs: repair free space btrees") from Dec 15, 2023 (linux-next), leads to the following Smatch static checker warning: fs/xfs/scrub/alloc_repair.c:781 xrep_abt_build_new_trees() warn: missing unwind goto?" That's a bug, so let's fix it. Reported-by: Dan Carpenter Fixes: 4bdfd7d15747 ("xfs: repair free space btrees") Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/alloc_repair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/alloc_repair.c b/fs/xfs/scrub/alloc_repair.c index d421b253923e..30295898cc8a 100644 --- a/fs/xfs/scrub/alloc_repair.c +++ b/fs/xfs/scrub/alloc_repair.c @@ -778,7 +778,7 @@ xrep_abt_build_new_trees( error = xrep_bnobt_sort_records(ra); if (error) - return error; + goto err_levels; /* Load the free space by block number tree. */ ra->array_cur = XFARRAY_CURSOR_INIT; -- cgit v1.2.3-59-g8ed1b From 5302a5c8beb21d01b7b8d92cc73b6871bc27d7bf Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:06 -0700 Subject: xfs: only clear log incompat flags at clean unmount While reviewing the online fsck patchset, someone spied the xfs_swapext_can_use_without_log_assistance function and wondered why we go through this inverted-bitmask dance to avoid setting the XFS_SB_FEAT_INCOMPAT_LOG_SWAPEXT feature. (The same principles apply to the logged extended attribute update feature bit in the since-merged LARP series.) The reason for this dance is that xfs_add_incompat_log_feature is an expensive operation -- it forces the log, pushes the AIL, and then if nobody's beaten us to it, sets the feature bit and issues a synchronous write of the primary superblock. That could be a one-time cost amortized over the life of the filesystem, but the log quiesce and cover operations call xfs_clear_incompat_log_features to remove feature bits opportunistically. On a moderately loaded filesystem this leads to us cycling those bits on and off over and over, which hurts performance. Why do we clear the log incompat bits? Back in ~2020 I think Dave and I had a conversation on IRC[2] about what the log incompat bits represent. IIRC in that conversation we decided that the log incompat bits protect unrecovered log items so that old kernels won't try to recover them and barf. Since a clean log has no protected log items, we could clear the bits at cover/quiesce time. As Dave Chinner pointed out in the thread, clearing log incompat bits at unmount time has positive effects for golden root disk image generator setups, since the generator could be running a newer kernel than what gets written to the golden image -- if there are log incompat fields set in the golden image that was generated by a newer kernel/OS image builder then the provisioning host cannot mount the filesystem even though the log is clean and recovery is unnecessary to mount the filesystem. Given that it's expensive to set log incompat bits, we really only want to do that once per bit per mount. Therefore, I propose that we only clear log incompat bits as part of writing a clean unmount record. Do this by adding an operational state flag to the xfs mount that guards whether or not the feature bit clearing can actually take place. This eliminates the l_incompat_users rwsem that we use to protect a log cleaning operation from clearing a feature bit that a frontend thread is trying to set -- this lock adds another way to fail w.r.t. locking. For the swapext series, I shard that into multiple locks just to work around the lockdep complaints, and that's fugly. Link: https://lore.kernel.org/linux-xfs/20240131230043.GA6180@frogsfrogsfrogs/ Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig Reviewed-by: Dave Chinner --- .../filesystems/xfs/xfs-online-fsck-design.rst | 3 -- fs/xfs/xfs_log.c | 28 +-------------- fs/xfs/xfs_log.h | 2 -- fs/xfs/xfs_log_priv.h | 3 -- fs/xfs/xfs_log_recover.c | 15 -------- fs/xfs/xfs_mount.c | 8 ++++- fs/xfs/xfs_mount.h | 6 +++- fs/xfs/xfs_xattr.c | 42 ++++------------------ 8 files changed, 19 insertions(+), 88 deletions(-) (limited to 'fs') diff --git a/Documentation/filesystems/xfs/xfs-online-fsck-design.rst b/Documentation/filesystems/xfs/xfs-online-fsck-design.rst index 6333697ba3e8..1d161752f09e 100644 --- a/Documentation/filesystems/xfs/xfs-online-fsck-design.rst +++ b/Documentation/filesystems/xfs/xfs-online-fsck-design.rst @@ -4047,9 +4047,6 @@ series. | one ``struct rw_semaphore`` for each feature. | | The log cleaning code tries to take this rwsem in exclusive mode to | | clear the bit; if the lock attempt fails, the feature bit remains set. | -| Filesystem code signals its intention to use a log incompat feature in a | -| transaction by calling ``xlog_use_incompat_feat``, which takes the rwsem | -| in shared mode. | | The code supporting a log incompat feature should create wrapper | | functions to obtain the log feature and call | | ``xfs_add_incompat_log_feature`` to set the feature bits in the primary | diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 5004f23d344e..416c15494983 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1448,7 +1448,7 @@ xfs_log_work_queue( * Clear the log incompat flags if we have the opportunity. * * This only happens if we're about to log the second dummy transaction as part - * of covering the log and we can get the log incompat feature usage lock. + * of covering the log. */ static inline void xlog_clear_incompat( @@ -1463,11 +1463,7 @@ xlog_clear_incompat( if (log->l_covered_state != XLOG_STATE_COVER_DONE2) return; - if (!down_write_trylock(&log->l_incompat_users)) - return; - xfs_clear_incompat_log_features(mp); - up_write(&log->l_incompat_users); } /* @@ -1585,8 +1581,6 @@ xlog_alloc_log( } log->l_sectBBsize = 1 << log2_size; - init_rwsem(&log->l_incompat_users); - xlog_get_iclog_buffer_size(mp, log); spin_lock_init(&log->l_icloglock); @@ -3871,23 +3865,3 @@ xfs_log_check_lsn( return valid; } - -/* - * Notify the log that we're about to start using a feature that is protected - * by a log incompat feature flag. This will prevent log covering from - * clearing those flags. - */ -void -xlog_use_incompat_feat( - struct xlog *log) -{ - down_read(&log->l_incompat_users); -} - -/* Notify the log that we've finished using log incompat features. */ -void -xlog_drop_incompat_feat( - struct xlog *log) -{ - up_read(&log->l_incompat_users); -} diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 2728886c2963..d69acf881153 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -159,8 +159,6 @@ bool xfs_log_check_lsn(struct xfs_mount *, xfs_lsn_t); xfs_lsn_t xlog_grant_push_threshold(struct xlog *log, int need_bytes); bool xlog_force_shutdown(struct xlog *log, uint32_t shutdown_flags); -void xlog_use_incompat_feat(struct xlog *log); -void xlog_drop_incompat_feat(struct xlog *log); int xfs_attr_use_log_assist(struct xfs_mount *mp); #endif /* __XFS_LOG_H__ */ diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index e30c06ec20e3..43881575cd49 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -450,9 +450,6 @@ struct xlog { xfs_lsn_t l_recovery_lsn; uint32_t l_iclog_roundoff;/* padding roundoff */ - - /* Users of log incompat features should take a read lock. */ - struct rw_semaphore l_incompat_users; }; /* diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 1b1f0a4cd494..41aec991433c 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -3496,21 +3496,6 @@ xlog_recover_finish( */ xfs_log_force(log->l_mp, XFS_LOG_SYNC); - /* - * Now that we've recovered the log and all the intents, we can clear - * the log incompat feature bits in the superblock because there's no - * longer anything to protect. We rely on the AIL push to write out the - * updated superblock after everything else. - */ - if (xfs_clear_incompat_log_features(log->l_mp)) { - error = xfs_sync_sb(log->l_mp, false); - if (error < 0) { - xfs_alert(log->l_mp, - "Failed to clear log incompat features on recovery"); - goto out_error; - } - } - xlog_recover_process_iunlinks(log); /* diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index df370eb5dc15..d37ba10f5fa3 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1095,6 +1095,11 @@ xfs_unmountfs( "Freespace may not be correct on next mount."); xfs_unmount_check(mp); + /* + * Indicate that it's ok to clear log incompat bits before cleaning + * the log and writing the unmount record. + */ + xfs_set_done_with_log_incompat(mp); xfs_log_unmount(mp); xfs_da_unmount(mp); xfs_uuid_unmount(mp); @@ -1364,7 +1369,8 @@ xfs_clear_incompat_log_features( if (!xfs_has_crc(mp) || !xfs_sb_has_incompat_log_feature(&mp->m_sb, XFS_SB_FEAT_INCOMPAT_LOG_ALL) || - xfs_is_shutdown(mp)) + xfs_is_shutdown(mp) || + !xfs_is_done_with_log_incompat(mp)) return false; /* diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index e880aa48de68..6ec038b88454 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -412,6 +412,8 @@ __XFS_HAS_FEAT(nouuid, NOUUID) #define XFS_OPSTATE_WARNED_LARP 9 /* Mount time quotacheck is running */ #define XFS_OPSTATE_QUOTACHECK_RUNNING 10 +/* Do we want to clear log incompat flags? */ +#define XFS_OPSTATE_UNSET_LOG_INCOMPAT 11 #define __XFS_IS_OPSTATE(name, NAME) \ static inline bool xfs_is_ ## name (struct xfs_mount *mp) \ @@ -439,6 +441,7 @@ __XFS_IS_OPSTATE(quotacheck_running, QUOTACHECK_RUNNING) #else # define xfs_is_quotacheck_running(mp) (false) #endif +__XFS_IS_OPSTATE(done_with_log_incompat, UNSET_LOG_INCOMPAT) static inline bool xfs_should_warn(struct xfs_mount *mp, long nr) @@ -457,7 +460,8 @@ xfs_should_warn(struct xfs_mount *mp, long nr) { (1UL << XFS_OPSTATE_WARNED_SCRUB), "wscrub" }, \ { (1UL << XFS_OPSTATE_WARNED_SHRINK), "wshrink" }, \ { (1UL << XFS_OPSTATE_WARNED_LARP), "wlarp" }, \ - { (1UL << XFS_OPSTATE_QUOTACHECK_RUNNING), "quotacheck" } + { (1UL << XFS_OPSTATE_QUOTACHECK_RUNNING), "quotacheck" }, \ + { (1UL << XFS_OPSTATE_UNSET_LOG_INCOMPAT), "unset_log_incompat" } /* * Max and min values for mount-option defined I/O diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 364104e1b38a..4ebf7052eb67 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -22,10 +22,7 @@ /* * Get permission to use log-assisted atomic exchange of file extents. - * - * Callers must not be running any transactions or hold any inode locks, and - * they must release the permission by calling xlog_drop_incompat_feat - * when they're done. + * Callers must not be running any transactions or hold any ILOCKs. */ static inline int xfs_attr_grab_log_assist( @@ -33,16 +30,7 @@ xfs_attr_grab_log_assist( { int error = 0; - /* - * Protect ourselves from an idle log clearing the logged xattrs log - * incompat feature bit. - */ - xlog_use_incompat_feat(mp->m_log); - - /* - * If log-assisted xattrs are already enabled, the caller can use the - * log assisted swap functions with the log-incompat reference we got. - */ + /* xattr update log intent items are already enabled */ if (xfs_sb_version_haslogxattrs(&mp->m_sb)) return 0; @@ -52,31 +40,19 @@ xfs_attr_grab_log_assist( * a V5 filesystem for the superblock field, but we'll require rmap * or reflink to avoid having to deal with really old kernels. */ - if (!xfs_has_reflink(mp) && !xfs_has_rmapbt(mp)) { - error = -EOPNOTSUPP; - goto drop_incompat; - } + if (!xfs_has_reflink(mp) && !xfs_has_rmapbt(mp)) + return -EOPNOTSUPP; /* Enable log-assisted xattrs. */ error = xfs_add_incompat_log_feature(mp, XFS_SB_FEAT_INCOMPAT_LOG_XATTRS); if (error) - goto drop_incompat; + return error; xfs_warn_mount(mp, XFS_OPSTATE_WARNED_LARP, "EXPERIMENTAL logged extended attributes feature in use. Use at your own risk!"); return 0; -drop_incompat: - xlog_drop_incompat_feat(mp->m_log); - return error; -} - -static inline void -xfs_attr_rele_log_assist( - struct xfs_mount *mp) -{ - xlog_drop_incompat_feat(mp->m_log); } static inline bool @@ -100,7 +76,6 @@ xfs_attr_change( struct xfs_da_args *args) { struct xfs_mount *mp = args->dp->i_mount; - bool use_logging = false; int error; ASSERT(!(args->op_flags & XFS_DA_OP_LOGGED)); @@ -111,14 +86,9 @@ xfs_attr_change( return error; args->op_flags |= XFS_DA_OP_LOGGED; - use_logging = true; } - error = xfs_attr_set(args); - - if (use_logging) - xfs_attr_rele_log_assist(mp); - return error; + return xfs_attr_set(args); } -- cgit v1.2.3-59-g8ed1b From a4db266a705c5518f3049074fd233b6c57daab00 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:07 -0700 Subject: xfs: move inode lease breaking functions to xfs_inode.c The lease breaking functions operate at the scope of the entire VFS inode, not subranges of a file. Move them to xfs_inode.c since they're already declared in xfs_inode.h. This cleanup moves us closer to having xfs_FOO.h declare only the symbols in xfs_FOO.c. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_file.c | 61 ----------------------------------------------------- fs/xfs/xfs_inode.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_inode.h | 1 - 3 files changed, 62 insertions(+), 62 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 632653e00906..40b778415f5f 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -861,67 +861,6 @@ xfs_file_write_iter( return xfs_file_buffered_write(iocb, from); } -static void -xfs_wait_dax_page( - struct inode *inode) -{ - struct xfs_inode *ip = XFS_I(inode); - - xfs_iunlock(ip, XFS_MMAPLOCK_EXCL); - schedule(); - xfs_ilock(ip, XFS_MMAPLOCK_EXCL); -} - -int -xfs_break_dax_layouts( - struct inode *inode, - bool *retry) -{ - struct page *page; - - xfs_assert_ilocked(XFS_I(inode), XFS_MMAPLOCK_EXCL); - - page = dax_layout_busy_page(inode->i_mapping); - if (!page) - return 0; - - *retry = true; - return ___wait_var_event(&page->_refcount, - atomic_read(&page->_refcount) == 1, TASK_INTERRUPTIBLE, - 0, 0, xfs_wait_dax_page(inode)); -} - -int -xfs_break_layouts( - struct inode *inode, - uint *iolock, - enum layout_break_reason reason) -{ - bool retry; - int error; - - xfs_assert_ilocked(XFS_I(inode), XFS_IOLOCK_SHARED | XFS_IOLOCK_EXCL); - - do { - retry = false; - switch (reason) { - case BREAK_UNMAP: - error = xfs_break_dax_layouts(inode, &retry); - if (error || retry) - break; - fallthrough; - case BREAK_WRITE: - error = xfs_break_leased_layouts(inode, iolock, &retry); - break; - default: - WARN_ON_ONCE(1); - error = -EINVAL; - } - } while (error == 0 && retry); - - return error; -} - /* Does this file, inode, or mount want synchronous writes? */ static inline bool xfs_file_sync_writes(struct file *filp) { diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 3e667a19b80b..39e6f88e9691 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -38,6 +38,7 @@ #include "xfs_ag.h" #include "xfs_log_priv.h" #include "xfs_health.h" +#include "xfs_pnfs.h" struct kmem_cache *xfs_inode_cache; @@ -3946,3 +3947,64 @@ xfs_inode_count_blocks( xfs_bmap_count_leaves(ifp, rblocks); *dblocks = ip->i_nblocks - *rblocks; } + +static void +xfs_wait_dax_page( + struct inode *inode) +{ + struct xfs_inode *ip = XFS_I(inode); + + xfs_iunlock(ip, XFS_MMAPLOCK_EXCL); + schedule(); + xfs_ilock(ip, XFS_MMAPLOCK_EXCL); +} + +int +xfs_break_dax_layouts( + struct inode *inode, + bool *retry) +{ + struct page *page; + + xfs_assert_ilocked(XFS_I(inode), XFS_MMAPLOCK_EXCL); + + page = dax_layout_busy_page(inode->i_mapping); + if (!page) + return 0; + + *retry = true; + return ___wait_var_event(&page->_refcount, + atomic_read(&page->_refcount) == 1, TASK_INTERRUPTIBLE, + 0, 0, xfs_wait_dax_page(inode)); +} + +int +xfs_break_layouts( + struct inode *inode, + uint *iolock, + enum layout_break_reason reason) +{ + bool retry; + int error; + + xfs_assert_ilocked(XFS_I(inode), XFS_IOLOCK_SHARED | XFS_IOLOCK_EXCL); + + do { + retry = false; + switch (reason) { + case BREAK_UNMAP: + error = xfs_break_dax_layouts(inode, &retry); + if (error || retry) + break; + fallthrough; + case BREAK_WRITE: + error = xfs_break_leased_layouts(inode, iolock, &retry); + break; + default: + WARN_ON_ONCE(1); + error = -EINVAL; + } + } while (error == 0 && retry); + + return error; +} diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index ab46ffb3ac19..5164c5d3e549 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -565,7 +565,6 @@ xfs_itruncate_extents( return xfs_itruncate_extents_flags(tpp, ip, whichfork, new_size, 0); } -/* from xfs_file.c */ int xfs_break_dax_layouts(struct inode *inode, bool *retry); int xfs_break_layouts(struct inode *inode, uint *iolock, enum layout_break_reason reason); -- cgit v1.2.3-59-g8ed1b From 3fc4844585c761e69654a237df41ffc5e51262e9 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:08 -0700 Subject: xfs: move xfs_iops.c declarations out of xfs_inode.h Similarly, move declarations of public symbols of xfs_iops.c from xfs_inode.h to xfs_iops.h. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_inode.h | 5 ----- fs/xfs/xfs_iops.h | 4 ++++ 2 files changed, 4 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 5164c5d3e549..b2dde0e0f265 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -569,11 +569,6 @@ int xfs_break_dax_layouts(struct inode *inode, bool *retry); int xfs_break_layouts(struct inode *inode, uint *iolock, enum layout_break_reason reason); -/* from xfs_iops.c */ -extern void xfs_setup_inode(struct xfs_inode *ip); -extern void xfs_setup_iops(struct xfs_inode *ip); -extern void xfs_diflags_to_iflags(struct xfs_inode *ip, bool init); - static inline void xfs_update_stable_writes(struct xfs_inode *ip) { if (bdev_stable_writes(xfs_inode_buftarg(ip)->bt_bdev)) diff --git a/fs/xfs/xfs_iops.h b/fs/xfs/xfs_iops.h index 7f84a0843b24..8a38c3e2ed0e 100644 --- a/fs/xfs/xfs_iops.h +++ b/fs/xfs/xfs_iops.h @@ -19,4 +19,8 @@ int xfs_vn_setattr_size(struct mnt_idmap *idmap, int xfs_inode_init_security(struct inode *inode, struct inode *dir, const struct qstr *qstr); +extern void xfs_setup_inode(struct xfs_inode *ip); +extern void xfs_setup_iops(struct xfs_inode *ip); +extern void xfs_diflags_to_iflags(struct xfs_inode *ip, bool init); + #endif /* __XFS_IOPS_H__ */ -- cgit v1.2.3-59-g8ed1b From 00acb28d96746f78389f23a7b5309a917b45c12f Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:09 -0700 Subject: xfs: declare xfs_file.c symbols in xfs_file.h Move the two public symbols in xfs_file.c to xfs_file.h. We're about to add more public symbols in that source file, so let's finally create the header file. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_file.c | 1 + fs/xfs/xfs_file.h | 12 ++++++++++++ fs/xfs/xfs_ioctl.c | 1 + fs/xfs/xfs_iops.c | 1 + fs/xfs/xfs_iops.h | 3 --- 5 files changed, 15 insertions(+), 3 deletions(-) create mode 100644 fs/xfs/xfs_file.h (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 40b778415f5f..9961d4b5efbe 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -24,6 +24,7 @@ #include "xfs_pnfs.h" #include "xfs_iomap.h" #include "xfs_reflink.h" +#include "xfs_file.h" #include #include diff --git a/fs/xfs/xfs_file.h b/fs/xfs/xfs_file.h new file mode 100644 index 000000000000..7d39e3eca56d --- /dev/null +++ b/fs/xfs/xfs_file.h @@ -0,0 +1,12 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + */ +#ifndef __XFS_FILE_H__ +#define __XFS_FILE_H__ + +extern const struct file_operations xfs_file_operations; +extern const struct file_operations xfs_dir_file_operations; + +#endif /* __XFS_FILE_H__ */ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index d0e2cec6210d..1397edea20f1 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -39,6 +39,7 @@ #include "xfs_ioctl.h" #include "xfs_xattr.h" #include "xfs_rtbitmap.h" +#include "xfs_file.h" #include #include diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 66f8c47642e8..55ed2d1023d6 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -25,6 +25,7 @@ #include "xfs_error.h" #include "xfs_ioctl.h" #include "xfs_xattr.h" +#include "xfs_file.h" #include #include diff --git a/fs/xfs/xfs_iops.h b/fs/xfs/xfs_iops.h index 8a38c3e2ed0e..3c1a2605ffd2 100644 --- a/fs/xfs/xfs_iops.h +++ b/fs/xfs/xfs_iops.h @@ -8,9 +8,6 @@ struct xfs_inode; -extern const struct file_operations xfs_file_operations; -extern const struct file_operations xfs_dir_file_operations; - extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size); int xfs_vn_setattr_size(struct mnt_idmap *idmap, -- cgit v1.2.3-59-g8ed1b From ee20808d848c87a51e176706d81b95a21747d6cf Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:10 -0700 Subject: xfs: create a new helper to return a file's allocation unit Create a new helper function to calculate the fundamental allocation unit (i.e. the smallest unit of space we can allocate) of a file. Things are going to get hairy with range-exchange on the realtime device, so prepare for this now. Remove the static attribute from xfs_is_falloc_aligned since the next patch will need it. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_file.c | 32 ++++++++++++-------------------- fs/xfs/xfs_file.h | 3 +++ fs/xfs/xfs_inode.c | 13 +++++++++++++ fs/xfs/xfs_inode.h | 1 + 4 files changed, 29 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 9961d4b5efbe..64278f8acaee 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -39,33 +39,25 @@ static const struct vm_operations_struct xfs_file_vm_ops; * Decide if the given file range is aligned to the size of the fundamental * allocation unit for the file. */ -static bool +bool xfs_is_falloc_aligned( struct xfs_inode *ip, loff_t pos, long long int len) { - struct xfs_mount *mp = ip->i_mount; - uint64_t mask; - - if (XFS_IS_REALTIME_INODE(ip)) { - if (!is_power_of_2(mp->m_sb.sb_rextsize)) { - u64 rextbytes; - u32 mod; - - rextbytes = XFS_FSB_TO_B(mp, mp->m_sb.sb_rextsize); - div_u64_rem(pos, rextbytes, &mod); - if (mod) - return false; - div_u64_rem(len, rextbytes, &mod); - return mod == 0; - } - mask = XFS_FSB_TO_B(mp, mp->m_sb.sb_rextsize) - 1; - } else { - mask = mp->m_sb.sb_blocksize - 1; + unsigned int alloc_unit = xfs_inode_alloc_unitsize(ip); + + if (!is_power_of_2(alloc_unit)) { + u32 mod; + + div_u64_rem(pos, alloc_unit, &mod); + if (mod) + return false; + div_u64_rem(len, alloc_unit, &mod); + return mod == 0; } - return !((pos | len) & mask); + return !((pos | len) & (alloc_unit - 1)); } /* diff --git a/fs/xfs/xfs_file.h b/fs/xfs/xfs_file.h index 7d39e3eca56d..2ad91f755caf 100644 --- a/fs/xfs/xfs_file.h +++ b/fs/xfs/xfs_file.h @@ -9,4 +9,7 @@ extern const struct file_operations xfs_file_operations; extern const struct file_operations xfs_dir_file_operations; +bool xfs_is_falloc_aligned(struct xfs_inode *ip, loff_t pos, + long long int len); + #endif /* __XFS_FILE_H__ */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 39e6f88e9691..492dae0efad2 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -4008,3 +4008,16 @@ xfs_break_layouts( return error; } + +/* Returns the size of fundamental allocation unit for a file, in bytes. */ +unsigned int +xfs_inode_alloc_unitsize( + struct xfs_inode *ip) +{ + unsigned int blocks = 1; + + if (XFS_IS_REALTIME_INODE(ip)) + blocks = ip->i_mount->m_sb.sb_rextsize; + + return XFS_FSB_TO_B(ip->i_mount, blocks); +} diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index b2dde0e0f265..fa3e605901e2 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -625,6 +625,7 @@ int xfs_inode_reload_unlinked(struct xfs_inode *ip); bool xfs_ifork_zapped(const struct xfs_inode *ip, int whichfork); void xfs_inode_count_blocks(struct xfs_trans *tp, struct xfs_inode *ip, xfs_filblks_t *dblocks, xfs_filblks_t *rblocks); +unsigned int xfs_inode_alloc_unitsize(struct xfs_inode *ip); struct xfs_dir_update_params { const struct xfs_inode *dp; -- cgit v1.2.3-59-g8ed1b From 6b700a5be9b3b69419474622336c63fdc1cc3ca4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:11 -0700 Subject: xfs: hoist multi-fsb allocation unit detection to a helper Replace the open-coded logic to decide if a file has a multi-fsb allocation unit to a helper to make the code easier to read. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_bmap_util.c | 4 ++-- fs/xfs/xfs_inode.h | 9 +++++++++ 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 19e11d1da660..53aa90a0ee3a 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -542,7 +542,7 @@ xfs_can_free_eofblocks( * forever. */ end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_ISIZE(ip)); - if (XFS_IS_REALTIME_INODE(ip) && mp->m_sb.sb_rextsize > 1) + if (xfs_inode_has_bigrtalloc(ip)) end_fsb = xfs_rtb_roundup_rtx(mp, end_fsb); last_fsb = XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes); if (last_fsb <= end_fsb) @@ -843,7 +843,7 @@ xfs_free_file_space( endoffset_fsb = XFS_B_TO_FSBT(mp, offset + len); /* We can only free complete realtime extents. */ - if (XFS_IS_REALTIME_INODE(ip) && mp->m_sb.sb_rextsize > 1) { + if (xfs_inode_has_bigrtalloc(ip)) { startoffset_fsb = xfs_rtb_roundup_rtx(mp, startoffset_fsb); endoffset_fsb = xfs_rtb_rounddown_rtx(mp, endoffset_fsb); } diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index fa3e605901e2..f559e68ee707 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -311,6 +311,15 @@ static inline bool xfs_inode_has_large_extent_counts(struct xfs_inode *ip) return ip->i_diflags2 & XFS_DIFLAG2_NREXT64; } +/* + * Decide if this file is a realtime file whose data allocation unit is larger + * than a single filesystem block. + */ +static inline bool xfs_inode_has_bigrtalloc(struct xfs_inode *ip) +{ + return XFS_IS_REALTIME_INODE(ip) && ip->i_mount->m_sb.sb_rextsize > 1; +} + /* * Return the buftarg used for data allocations on a given inode. */ -- cgit v1.2.3-59-g8ed1b From ac5cebeed61351c0a60c65bd20c70120469c46ff Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:12 -0700 Subject: xfs: refactor non-power-of-two alignment checks Create a helper function that can compute if a 64-bit number is an integer multiple of a 32-bit number, where the 32-bit number is not required to be an even power of two. This is needed for some new code for the realtime device, where we can set 37k allocation units and then have to remap them. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_file.c | 12 +++--------- fs/xfs/xfs_linux.h | 5 +++++ 2 files changed, 8 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 64278f8acaee..d1d4158441bd 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -47,15 +47,9 @@ xfs_is_falloc_aligned( { unsigned int alloc_unit = xfs_inode_alloc_unitsize(ip); - if (!is_power_of_2(alloc_unit)) { - u32 mod; - - div_u64_rem(pos, alloc_unit, &mod); - if (mod) - return false; - div_u64_rem(len, alloc_unit, &mod); - return mod == 0; - } + if (!is_power_of_2(alloc_unit)) + return isaligned_64(pos, alloc_unit) && + isaligned_64(len, alloc_unit); return !((pos | len) & (alloc_unit - 1)); } diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h index 8f07c9f6157f..ac355328121a 100644 --- a/fs/xfs/xfs_linux.h +++ b/fs/xfs/xfs_linux.h @@ -198,6 +198,11 @@ static inline uint64_t howmany_64(uint64_t x, uint32_t y) return x; } +static inline bool isaligned_64(uint64_t x, uint32_t y) +{ + return do_div(x, y) == 0; +} + /* If @b is a power of 2, return log2(b). Else return -1. */ static inline int8_t log2_if_power2(unsigned long b) { -- cgit v1.2.3-59-g8ed1b From 15f78aa3eb07645e7bef15a53b4ae1c757907d2c Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:12 -0700 Subject: xfs: constify xfs_bmap_is_written_extent This predicate doesn't modify the structure that's being passed in, so we can mark it const. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_bmap.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index f7662595309d..b8bdbf1560e6 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -158,7 +158,7 @@ static inline bool xfs_bmap_is_real_extent(const struct xfs_bmbt_irec *irec) * Return true if the extent is a real, allocated extent, or false if it is a * delayed allocation, and unwritten extent or a hole. */ -static inline bool xfs_bmap_is_written_extent(struct xfs_bmbt_irec *irec) +static inline bool xfs_bmap_is_written_extent(const struct xfs_bmbt_irec *irec) { return xfs_bmap_is_real_extent(irec) && irec->br_state != XFS_EXT_UNWRITTEN; -- cgit v1.2.3-59-g8ed1b From 5b9932f6001c70b984e8c9c2fe09e443beb4baba Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:13 -0700 Subject: vfs: export remap and write check helpers Export these functions so that the next patch can use them to check the file ranges being passed to the XFS_IOC_EXCHANGE_RANGE operation. Cc: linux-fsdevel@vger.kernel.org Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/read_write.c | 1 + fs/remap_range.c | 4 ++-- include/linux/fs.h | 1 + 3 files changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/read_write.c b/fs/read_write.c index d4c036e82b6c..85c096f2c0d0 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -1667,6 +1667,7 @@ int generic_write_check_limits(struct file *file, loff_t pos, loff_t *count) return 0; } +EXPORT_SYMBOL_GPL(generic_write_check_limits); /* Like generic_write_checks(), but takes size of write instead of iter. */ int generic_write_checks_count(struct kiocb *iocb, loff_t *count) diff --git a/fs/remap_range.c b/fs/remap_range.c index de07f978ce3e..28246dfc8485 100644 --- a/fs/remap_range.c +++ b/fs/remap_range.c @@ -99,8 +99,7 @@ static int generic_remap_checks(struct file *file_in, loff_t pos_in, return 0; } -static int remap_verify_area(struct file *file, loff_t pos, loff_t len, - bool write) +int remap_verify_area(struct file *file, loff_t pos, loff_t len, bool write) { int mask = write ? MAY_WRITE : MAY_READ; loff_t tmp; @@ -118,6 +117,7 @@ static int remap_verify_area(struct file *file, loff_t pos, loff_t len, return fsnotify_file_area_perm(file, mask, &pos, len); } +EXPORT_SYMBOL_GPL(remap_verify_area); /* * Ensure that we don't remap a partial EOF block in the middle of something diff --git a/include/linux/fs.h b/include/linux/fs.h index 8dfd53b52744..0835faeebe7b 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2119,6 +2119,7 @@ extern ssize_t vfs_read(struct file *, char __user *, size_t, loff_t *); extern ssize_t vfs_write(struct file *, const char __user *, size_t, loff_t *); extern ssize_t vfs_copy_file_range(struct file *, loff_t , struct file *, loff_t, size_t, unsigned int); +int remap_verify_area(struct file *file, loff_t pos, loff_t len, bool write); int __generic_remap_file_range_prep(struct file *file_in, loff_t pos_in, struct file *file_out, loff_t pos_out, loff_t *len, unsigned int remap_flags, -- cgit v1.2.3-59-g8ed1b From 9a64d9b3109d01cca0b83c1d36538b7a37c5284e Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:14 -0700 Subject: xfs: introduce new file range exchange ioctl Introduce a new ioctl to handle exchanging ranges of bytes between files. The goal here is to perform the exchange atomically with respect to applications -- either they see the file contents before the exchange or they see that A-B is now B-A, even if the kernel crashes. My original goal with all this code was to make it so that online repair can build a replacement directory or xattr structure in a temporary file and commit the repair by atomically exchanging all the data blocks between the two files. However, I needed a way to test this mechanism thoroughly, so I've been evolving an ioctl interface since then. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/libxfs/xfs_fs.h | 41 ++++++ fs/xfs/xfs_exchrange.c | 339 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_exchrange.h | 30 +++++ fs/xfs/xfs_ioctl.c | 4 + 5 files changed, 415 insertions(+) create mode 100644 fs/xfs/xfs_exchrange.c create mode 100644 fs/xfs/xfs_exchrange.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 76674ad5833e..2474242f5a05 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -67,6 +67,7 @@ xfs-y += xfs_aops.o \ xfs_dir2_readdir.o \ xfs_discard.o \ xfs_error.o \ + xfs_exchrange.o \ xfs_export.o \ xfs_extent_busy.o \ xfs_file.o \ diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index ca1b17d01437..8a1e30cf4dc8 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -772,6 +772,46 @@ struct xfs_scrub_metadata { # define XFS_XATTR_LIST_MAX 65536 #endif +/* + * Exchange part of file1 with part of the file that this ioctl that is being + * called against (which we'll call file2). Filesystems must be able to + * restart and complete the operation even after the system goes down. + */ +struct xfs_exchange_range { + __s32 file1_fd; + __u32 pad; /* must be zeroes */ + __u64 file1_offset; /* file1 offset, bytes */ + __u64 file2_offset; /* file2 offset, bytes */ + __u64 length; /* bytes to exchange */ + + __u64 flags; /* see XFS_EXCHANGE_RANGE_* below */ +}; + +/* + * Exchange file data all the way to the ends of both files, and then exchange + * the file sizes. This flag can be used to replace a file's contents with a + * different amount of data. length will be ignored. + */ +#define XFS_EXCHANGE_RANGE_TO_EOF (1ULL << 0) + +/* Flush all changes in file data and file metadata to disk before returning. */ +#define XFS_EXCHANGE_RANGE_DSYNC (1ULL << 1) + +/* Dry run; do all the parameter verification but do not change anything. */ +#define XFS_EXCHANGE_RANGE_DRY_RUN (1ULL << 2) + +/* + * Exchange only the parts of the two files where the file allocation units + * mapped to file1's range have been written to. This can accelerate + * scatter-gather atomic writes with a temp file if all writes are aligned to + * the file allocation unit. + */ +#define XFS_EXCHANGE_RANGE_FILE1_WRITTEN (1ULL << 3) + +#define XFS_EXCHANGE_RANGE_ALL_FLAGS (XFS_EXCHANGE_RANGE_TO_EOF | \ + XFS_EXCHANGE_RANGE_DSYNC | \ + XFS_EXCHANGE_RANGE_DRY_RUN | \ + XFS_EXCHANGE_RANGE_FILE1_WRITTEN) /* * ioctl commands that are used by Linux filesystems @@ -843,6 +883,7 @@ struct xfs_scrub_metadata { #define XFS_IOC_FSGEOMETRY _IOR ('X', 126, struct xfs_fsop_geom) #define XFS_IOC_BULKSTAT _IOR ('X', 127, struct xfs_bulkstat_req) #define XFS_IOC_INUMBERS _IOR ('X', 128, struct xfs_inumbers_req) +#define XFS_IOC_EXCHANGE_RANGE _IOWR('X', 129, struct xfs_exchange_range) /* XFS_IOC_GETFSUUID ---------- deprecated 140 */ diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c new file mode 100644 index 000000000000..4cd824e47f75 --- /dev/null +++ b/fs/xfs/xfs_exchrange.c @@ -0,0 +1,339 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_inode.h" +#include "xfs_trans.h" +#include "xfs_exchrange.h" +#include + +/* + * Generic code for exchanging ranges of two files via XFS_IOC_EXCHANGE_RANGE. + * This part deals with struct file objects and byte ranges and does not deal + * with XFS-specific data structures such as xfs_inodes and block ranges. This + * separation may some day facilitate porting to another filesystem. + * + * The goal is to exchange fxr.length bytes starting at fxr.file1_offset in + * file1 with the same number of bytes starting at fxr.file2_offset in file2. + * Implementations must call xfs_exchange_range_prep to prepare the two + * files prior to taking locks; and they must update the inode change and mod + * times of both files as part of the metadata update. The timestamp update + * and freshness checks must be done atomically as part of the data exchange + * operation to ensure correctness of the freshness check. + * xfs_exchange_range_finish must be called after the operation completes + * successfully but before locks are dropped. + */ + +/* Verify that we have security clearance to perform this operation. */ +static int +xfs_exchange_range_verify_area( + struct xfs_exchrange *fxr) +{ + int ret; + + ret = remap_verify_area(fxr->file1, fxr->file1_offset, fxr->length, + true); + if (ret) + return ret; + + return remap_verify_area(fxr->file2, fxr->file2_offset, fxr->length, + true); +} + +/* + * Performs necessary checks before doing a range exchange, having stabilized + * mutable inode attributes via i_rwsem. + */ +static inline int +xfs_exchange_range_checks( + struct xfs_exchrange *fxr, + unsigned int alloc_unit) +{ + struct inode *inode1 = file_inode(fxr->file1); + struct inode *inode2 = file_inode(fxr->file2); + uint64_t allocmask = alloc_unit - 1; + int64_t test_len; + uint64_t blen; + loff_t size1, size2, tmp; + int error; + + /* Don't touch certain kinds of inodes */ + if (IS_IMMUTABLE(inode1) || IS_IMMUTABLE(inode2)) + return -EPERM; + if (IS_SWAPFILE(inode1) || IS_SWAPFILE(inode2)) + return -ETXTBSY; + + size1 = i_size_read(inode1); + size2 = i_size_read(inode2); + + /* Ranges cannot start after EOF. */ + if (fxr->file1_offset > size1 || fxr->file2_offset > size2) + return -EINVAL; + + /* + * If the caller said to exchange to EOF, we set the length of the + * request large enough to cover everything to the end of both files. + */ + if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) { + fxr->length = max_t(int64_t, size1 - fxr->file1_offset, + size2 - fxr->file2_offset); + + error = xfs_exchange_range_verify_area(fxr); + if (error) + return error; + } + + /* + * The start of both ranges must be aligned to the file allocation + * unit. + */ + if (!IS_ALIGNED(fxr->file1_offset, alloc_unit) || + !IS_ALIGNED(fxr->file2_offset, alloc_unit)) + return -EINVAL; + + /* Ensure offsets don't wrap. */ + if (check_add_overflow(fxr->file1_offset, fxr->length, &tmp) || + check_add_overflow(fxr->file2_offset, fxr->length, &tmp)) + return -EINVAL; + + /* + * We require both ranges to end within EOF, unless we're exchanging + * to EOF. + */ + if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) && + (fxr->file1_offset + fxr->length > size1 || + fxr->file2_offset + fxr->length > size2)) + return -EINVAL; + + /* + * Make sure we don't hit any file size limits. If we hit any size + * limits such that test_length was adjusted, we abort the whole + * operation. + */ + test_len = fxr->length; + error = generic_write_check_limits(fxr->file2, fxr->file2_offset, + &test_len); + if (error) + return error; + error = generic_write_check_limits(fxr->file1, fxr->file1_offset, + &test_len); + if (error) + return error; + if (test_len != fxr->length) + return -EINVAL; + + /* + * If the user wanted us to exchange up to the infile's EOF, round up + * to the next allocation unit boundary for this check. Do the same + * for the outfile. + * + * Otherwise, reject the range length if it's not aligned to an + * allocation unit. + */ + if (fxr->file1_offset + fxr->length == size1) + blen = ALIGN(size1, alloc_unit) - fxr->file1_offset; + else if (fxr->file2_offset + fxr->length == size2) + blen = ALIGN(size2, alloc_unit) - fxr->file2_offset; + else if (!IS_ALIGNED(fxr->length, alloc_unit)) + return -EINVAL; + else + blen = fxr->length; + + /* Don't allow overlapped exchanges within the same file. */ + if (inode1 == inode2 && + fxr->file2_offset + blen > fxr->file1_offset && + fxr->file1_offset + blen > fxr->file2_offset) + return -EINVAL; + + /* + * Ensure that we don't exchange a partial EOF block into the middle of + * another file. + */ + if ((fxr->length & allocmask) == 0) + return 0; + + blen = fxr->length; + if (fxr->file2_offset + blen < size2) + blen &= ~allocmask; + + if (fxr->file1_offset + blen < size1) + blen &= ~allocmask; + + return blen == fxr->length ? 0 : -EINVAL; +} + +/* + * Check that the two inodes are eligible for range exchanges, the ranges make + * sense, and then flush all dirty data. Caller must ensure that the inodes + * have been locked against any other modifications. + */ +static inline int +xfs_exchange_range_prep( + struct xfs_exchrange *fxr, + unsigned int alloc_unit) +{ + struct inode *inode1 = file_inode(fxr->file1); + struct inode *inode2 = file_inode(fxr->file2); + bool same_inode = (inode1 == inode2); + int error; + + /* Check that we don't violate system file offset limits. */ + error = xfs_exchange_range_checks(fxr, alloc_unit); + if (error || fxr->length == 0) + return error; + + /* Wait for the completion of any pending IOs on both files */ + inode_dio_wait(inode1); + if (!same_inode) + inode_dio_wait(inode2); + + error = filemap_write_and_wait_range(inode1->i_mapping, + fxr->file1_offset, + fxr->file1_offset + fxr->length - 1); + if (error) + return error; + + error = filemap_write_and_wait_range(inode2->i_mapping, + fxr->file2_offset, + fxr->file2_offset + fxr->length - 1); + if (error) + return error; + + /* + * If the files or inodes involved require synchronous writes, amend + * the request to force the filesystem to flush all data and metadata + * to disk after the operation completes. + */ + if (((fxr->file1->f_flags | fxr->file2->f_flags) & O_SYNC) || + IS_SYNC(inode1) || IS_SYNC(inode2)) + fxr->flags |= XFS_EXCHANGE_RANGE_DSYNC; + + return 0; +} + +/* + * Finish a range exchange operation, if it was successful. Caller must ensure + * that the inodes are still locked against any other modifications. + */ +static inline int +xfs_exchange_range_finish( + struct xfs_exchrange *fxr) +{ + int error; + + error = file_remove_privs(fxr->file1); + if (error) + return error; + if (file_inode(fxr->file1) == file_inode(fxr->file2)) + return 0; + + return file_remove_privs(fxr->file2); +} + +/* Exchange parts of two files. */ +static int +xfs_exchange_range( + struct xfs_exchrange *fxr) +{ + struct inode *inode1 = file_inode(fxr->file1); + struct inode *inode2 = file_inode(fxr->file2); + int ret; + + BUILD_BUG_ON(XFS_EXCHANGE_RANGE_ALL_FLAGS & + XFS_EXCHANGE_RANGE_PRIV_FLAGS); + + /* Both files must be on the same mount/filesystem. */ + if (fxr->file1->f_path.mnt != fxr->file2->f_path.mnt) + return -EXDEV; + + if (fxr->flags & ~XFS_EXCHANGE_RANGE_ALL_FLAGS) + return -EINVAL; + + /* Userspace requests only honored for regular files. */ + if (S_ISDIR(inode1->i_mode) || S_ISDIR(inode2->i_mode)) + return -EISDIR; + if (!S_ISREG(inode1->i_mode) || !S_ISREG(inode2->i_mode)) + return -EINVAL; + + /* Both files must be opened for read and write. */ + if (!(fxr->file1->f_mode & FMODE_READ) || + !(fxr->file1->f_mode & FMODE_WRITE) || + !(fxr->file2->f_mode & FMODE_READ) || + !(fxr->file2->f_mode & FMODE_WRITE)) + return -EBADF; + + /* Neither file can be opened append-only. */ + if ((fxr->file1->f_flags & O_APPEND) || + (fxr->file2->f_flags & O_APPEND)) + return -EBADF; + + /* + * If we're not exchanging to EOF, we can check the areas before + * stabilizing both files' i_size. + */ + if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)) { + ret = xfs_exchange_range_verify_area(fxr); + if (ret) + return ret; + } + + /* Update cmtime if the fd/inode don't forbid it. */ + if (!(fxr->file1->f_mode & FMODE_NOCMTIME) && !IS_NOCMTIME(inode1)) + fxr->flags |= __XFS_EXCHANGE_RANGE_UPD_CMTIME1; + if (!(fxr->file2->f_mode & FMODE_NOCMTIME) && !IS_NOCMTIME(inode2)) + fxr->flags |= __XFS_EXCHANGE_RANGE_UPD_CMTIME2; + + file_start_write(fxr->file2); + ret = -EOPNOTSUPP; /* XXX call out to lower level code */ + file_end_write(fxr->file2); + if (ret) + return ret; + + fsnotify_modify(fxr->file1); + if (fxr->file2 != fxr->file1) + fsnotify_modify(fxr->file2); + return 0; +} + +/* Collect exchange-range arguments from userspace. */ +long +xfs_ioc_exchange_range( + struct file *file, + struct xfs_exchange_range __user *argp) +{ + struct xfs_exchrange fxr = { + .file2 = file, + }; + struct xfs_exchange_range args; + struct fd file1; + int error; + + if (copy_from_user(&args, argp, sizeof(args))) + return -EFAULT; + if (memchr_inv(&args.pad, 0, sizeof(args.pad))) + return -EINVAL; + if (args.flags & ~XFS_EXCHANGE_RANGE_ALL_FLAGS) + return -EINVAL; + + fxr.file1_offset = args.file1_offset; + fxr.file2_offset = args.file2_offset; + fxr.length = args.length; + fxr.flags = args.flags; + + file1 = fdget(args.file1_fd); + if (!file1.file) + return -EBADF; + fxr.file1 = file1.file; + + error = xfs_exchange_range(&fxr); + fdput(file1); + return error; +} diff --git a/fs/xfs/xfs_exchrange.h b/fs/xfs/xfs_exchrange.h new file mode 100644 index 000000000000..f80369c7df5d --- /dev/null +++ b/fs/xfs/xfs_exchrange.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_EXCHRANGE_H__ +#define __XFS_EXCHRANGE_H__ + +/* Update the mtime/cmtime of file1 and file2 */ +#define __XFS_EXCHANGE_RANGE_UPD_CMTIME1 (1ULL << 63) +#define __XFS_EXCHANGE_RANGE_UPD_CMTIME2 (1ULL << 62) + +#define XFS_EXCHANGE_RANGE_PRIV_FLAGS (__XFS_EXCHANGE_RANGE_UPD_CMTIME1 | \ + __XFS_EXCHANGE_RANGE_UPD_CMTIME2) + +struct xfs_exchrange { + struct file *file1; + struct file *file2; + + loff_t file1_offset; + loff_t file2_offset; + u64 length; + + u64 flags; /* XFS_EXCHANGE_RANGE flags */ +}; + +long xfs_ioc_exchange_range(struct file *file, + struct xfs_exchange_range __user *argp); + +#endif /* __XFS_EXCHRANGE_H__ */ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 1397edea20f1..efa95892655d 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -40,6 +40,7 @@ #include "xfs_xattr.h" #include "xfs_rtbitmap.h" #include "xfs_file.h" +#include "xfs_exchrange.h" #include #include @@ -2170,6 +2171,9 @@ xfs_file_ioctl( return error; } + case XFS_IOC_EXCHANGE_RANGE: + return xfs_ioc_exchange_range(filp, arg); + default: return -ENOTTY; } -- cgit v1.2.3-59-g8ed1b From 1518646eef26c220e9256906260ecaaa64503522 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:15 -0700 Subject: xfs: create a incompat flag for atomic file mapping exchanges Create a incompat flag so that we only attempt to process file mapping exchange log items if the filesystem supports it, and a geometry flag to advertise support if it's present. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_format.h | 23 ++++++++++++----------- fs/xfs/libxfs/xfs_fs.h | 1 + fs/xfs/libxfs/xfs_sb.c | 5 +++++ fs/xfs/xfs_mount.h | 2 ++ fs/xfs/xfs_super.c | 4 ++++ 5 files changed, 24 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index 2b2f9050fbfb..ff1e28316e1b 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -367,18 +367,19 @@ xfs_sb_has_ro_compat_feature( return (sbp->sb_features_ro_compat & feature) != 0; } -#define XFS_SB_FEAT_INCOMPAT_FTYPE (1 << 0) /* filetype in dirent */ -#define XFS_SB_FEAT_INCOMPAT_SPINODES (1 << 1) /* sparse inode chunks */ -#define XFS_SB_FEAT_INCOMPAT_META_UUID (1 << 2) /* metadata UUID */ -#define XFS_SB_FEAT_INCOMPAT_BIGTIME (1 << 3) /* large timestamps */ -#define XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR (1 << 4) /* needs xfs_repair */ -#define XFS_SB_FEAT_INCOMPAT_NREXT64 (1 << 5) /* large extent counters */ +#define XFS_SB_FEAT_INCOMPAT_FTYPE (1 << 0) /* filetype in dirent */ +#define XFS_SB_FEAT_INCOMPAT_SPINODES (1 << 1) /* sparse inode chunks */ +#define XFS_SB_FEAT_INCOMPAT_META_UUID (1 << 2) /* metadata UUID */ +#define XFS_SB_FEAT_INCOMPAT_BIGTIME (1 << 3) /* large timestamps */ +#define XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR (1 << 4) /* needs xfs_repair */ +#define XFS_SB_FEAT_INCOMPAT_NREXT64 (1 << 5) /* large extent counters */ +#define XFS_SB_FEAT_INCOMPAT_EXCHRANGE (1 << 6) /* exchangerange supported */ #define XFS_SB_FEAT_INCOMPAT_ALL \ - (XFS_SB_FEAT_INCOMPAT_FTYPE| \ - XFS_SB_FEAT_INCOMPAT_SPINODES| \ - XFS_SB_FEAT_INCOMPAT_META_UUID| \ - XFS_SB_FEAT_INCOMPAT_BIGTIME| \ - XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR| \ + (XFS_SB_FEAT_INCOMPAT_FTYPE | \ + XFS_SB_FEAT_INCOMPAT_SPINODES | \ + XFS_SB_FEAT_INCOMPAT_META_UUID | \ + XFS_SB_FEAT_INCOMPAT_BIGTIME | \ + XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR | \ XFS_SB_FEAT_INCOMPAT_NREXT64) #define XFS_SB_FEAT_INCOMPAT_UNKNOWN ~XFS_SB_FEAT_INCOMPAT_ALL diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index 8a1e30cf4dc8..53526fca7386 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -239,6 +239,7 @@ typedef struct xfs_fsop_resblks { #define XFS_FSOP_GEOM_FLAGS_BIGTIME (1 << 21) /* 64-bit nsec timestamps */ #define XFS_FSOP_GEOM_FLAGS_INOBTCNT (1 << 22) /* inobt btree counter */ #define XFS_FSOP_GEOM_FLAGS_NREXT64 (1 << 23) /* large extent counters */ +#define XFS_FSOP_GEOM_FLAGS_EXCHANGE_RANGE (1 << 24) /* exchange range */ /* * Minimum and maximum sizes need for growth checks. diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index 73a4b895de67..c350e259b685 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -26,6 +26,7 @@ #include "xfs_health.h" #include "xfs_ag.h" #include "xfs_rtbitmap.h" +#include "xfs_exchrange.h" /* * Physical superblock buffer manipulations. Shared with libxfs in userspace. @@ -175,6 +176,8 @@ xfs_sb_version_to_features( features |= XFS_FEAT_NEEDSREPAIR; if (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_NREXT64) features |= XFS_FEAT_NREXT64; + if (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_EXCHRANGE) + features |= XFS_FEAT_EXCHANGE_RANGE; return features; } @@ -1259,6 +1262,8 @@ xfs_fs_geometry( } if (xfs_has_large_extent_counts(mp)) geo->flags |= XFS_FSOP_GEOM_FLAGS_NREXT64; + if (xfs_has_exchange_range(mp)) + geo->flags |= XFS_FSOP_GEOM_FLAGS_EXCHANGE_RANGE; geo->rtsectsize = sbp->sb_blocksize; geo->dirblocksize = xfs_dir2_dirblock_bytes(sbp); diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 6ec038b88454..b022e5120dc4 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -292,6 +292,7 @@ typedef struct xfs_mount { #define XFS_FEAT_BIGTIME (1ULL << 24) /* large timestamps */ #define XFS_FEAT_NEEDSREPAIR (1ULL << 25) /* needs xfs_repair */ #define XFS_FEAT_NREXT64 (1ULL << 26) /* large extent counters */ +#define XFS_FEAT_EXCHANGE_RANGE (1ULL << 27) /* exchange range */ /* Mount features */ #define XFS_FEAT_NOATTR2 (1ULL << 48) /* disable attr2 creation */ @@ -355,6 +356,7 @@ __XFS_HAS_FEAT(inobtcounts, INOBTCNT) __XFS_HAS_FEAT(bigtime, BIGTIME) __XFS_HAS_FEAT(needsrepair, NEEDSREPAIR) __XFS_HAS_FEAT(large_extent_counts, NREXT64) +__XFS_HAS_FEAT(exchange_range, EXCHANGE_RANGE) /* * Mount features diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index bce020374c5e..dbda72df3419 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1727,6 +1727,10 @@ xfs_fs_fill_super( goto out_filestream_unmount; } + if (xfs_has_exchange_range(mp)) + xfs_warn(mp, + "EXPERIMENTAL exchange-range feature enabled. Use at your own risk!"); + error = xfs_mountfs(mp); if (error) goto out_filestream_unmount; -- cgit v1.2.3-59-g8ed1b From 6c08f434bd33f88cf169e9e43c7a5e42fb3f2118 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:16 -0700 Subject: xfs: introduce a file mapping exchange log intent item Introduce a new intent log item to handle exchanging mappings between the forks of two files. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/libxfs/xfs_log_format.h | 42 ++++++- fs/xfs/libxfs/xfs_log_recover.h | 2 + fs/xfs/xfs_exchmaps_item.c | 235 ++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_exchmaps_item.h | 59 ++++++++++ fs/xfs/xfs_log_recover.c | 2 + fs/xfs/xfs_super.c | 19 ++++ 7 files changed, 357 insertions(+), 3 deletions(-) create mode 100644 fs/xfs/xfs_exchmaps_item.c create mode 100644 fs/xfs/xfs_exchmaps_item.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 2474242f5a05..68ca9726e7b7 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -102,6 +102,7 @@ xfs-y += xfs_log.o \ xfs_buf_item.o \ xfs_buf_item_recover.o \ xfs_dquot_item_recover.o \ + xfs_exchmaps_item.o \ xfs_extfree_item.o \ xfs_attr_item.o \ xfs_icreate_item.o \ diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index 16872972e1e9..09024431cae9 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -117,8 +117,9 @@ struct xfs_unmount_log_format { #define XLOG_REG_TYPE_ATTRD_FORMAT 28 #define XLOG_REG_TYPE_ATTR_NAME 29 #define XLOG_REG_TYPE_ATTR_VALUE 30 -#define XLOG_REG_TYPE_MAX 30 - +#define XLOG_REG_TYPE_XMI_FORMAT 31 +#define XLOG_REG_TYPE_XMD_FORMAT 32 +#define XLOG_REG_TYPE_MAX 32 /* * Flags to log operation header @@ -243,6 +244,8 @@ typedef struct xfs_trans_header { #define XFS_LI_BUD 0x1245 #define XFS_LI_ATTRI 0x1246 /* attr set/remove intent*/ #define XFS_LI_ATTRD 0x1247 /* attr set/remove done */ +#define XFS_LI_XMI 0x1248 /* mapping exchange intent */ +#define XFS_LI_XMD 0x1249 /* mapping exchange done */ #define XFS_LI_TYPE_DESC \ { XFS_LI_EFI, "XFS_LI_EFI" }, \ @@ -260,7 +263,9 @@ typedef struct xfs_trans_header { { XFS_LI_BUI, "XFS_LI_BUI" }, \ { XFS_LI_BUD, "XFS_LI_BUD" }, \ { XFS_LI_ATTRI, "XFS_LI_ATTRI" }, \ - { XFS_LI_ATTRD, "XFS_LI_ATTRD" } + { XFS_LI_ATTRD, "XFS_LI_ATTRD" }, \ + { XFS_LI_XMI, "XFS_LI_XMI" }, \ + { XFS_LI_XMD, "XFS_LI_XMD" } /* * Inode Log Item Format definitions. @@ -878,6 +883,37 @@ struct xfs_bud_log_format { uint64_t bud_bui_id; /* id of corresponding bui */ }; +/* + * XMI/XMD (file mapping exchange) log format definitions + */ + +/* This is the structure used to lay out an mapping exchange log item. */ +struct xfs_xmi_log_format { + uint16_t xmi_type; /* xmi log item type */ + uint16_t xmi_size; /* size of this item */ + uint32_t __pad; /* must be zero */ + uint64_t xmi_id; /* xmi identifier */ + + uint64_t xmi_inode1; /* inumber of first file */ + uint64_t xmi_inode2; /* inumber of second file */ + uint64_t xmi_startoff1; /* block offset into file1 */ + uint64_t xmi_startoff2; /* block offset into file2 */ + uint64_t xmi_blockcount; /* number of blocks */ + uint64_t xmi_flags; /* XFS_EXCHMAPS_* */ + uint64_t xmi_isize1; /* intended file1 size */ + uint64_t xmi_isize2; /* intended file2 size */ +}; + +#define XFS_EXCHMAPS_LOGGED_FLAGS (0) + +/* This is the structure used to lay out an mapping exchange done log item. */ +struct xfs_xmd_log_format { + uint16_t xmd_type; /* xmd log item type */ + uint16_t xmd_size; /* size of this item */ + uint32_t __pad; + uint64_t xmd_xmi_id; /* id of corresponding xmi */ +}; + /* * Dquot Log format definitions. * diff --git a/fs/xfs/libxfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h index 9fe7a9564bca..47b758b49cb3 100644 --- a/fs/xfs/libxfs/xfs_log_recover.h +++ b/fs/xfs/libxfs/xfs_log_recover.h @@ -75,6 +75,8 @@ extern const struct xlog_recover_item_ops xlog_cui_item_ops; extern const struct xlog_recover_item_ops xlog_cud_item_ops; extern const struct xlog_recover_item_ops xlog_attri_item_ops; extern const struct xlog_recover_item_ops xlog_attrd_item_ops; +extern const struct xlog_recover_item_ops xlog_xmi_item_ops; +extern const struct xlog_recover_item_ops xlog_xmd_item_ops; /* * Macros, structures, prototypes for internal log manager use. diff --git a/fs/xfs/xfs_exchmaps_item.c b/fs/xfs/xfs_exchmaps_item.c new file mode 100644 index 000000000000..65b0ade41b3d --- /dev/null +++ b/fs/xfs/xfs_exchmaps_item.c @@ -0,0 +1,235 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" +#include "xfs_bit.h" +#include "xfs_shared.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_inode.h" +#include "xfs_trans.h" +#include "xfs_trans_priv.h" +#include "xfs_exchmaps_item.h" +#include "xfs_log.h" +#include "xfs_bmap.h" +#include "xfs_icache.h" +#include "xfs_trans_space.h" +#include "xfs_error.h" +#include "xfs_log_priv.h" +#include "xfs_log_recover.h" + +struct kmem_cache *xfs_xmi_cache; +struct kmem_cache *xfs_xmd_cache; + +static const struct xfs_item_ops xfs_xmi_item_ops; + +static inline struct xfs_xmi_log_item *XMI_ITEM(struct xfs_log_item *lip) +{ + return container_of(lip, struct xfs_xmi_log_item, xmi_item); +} + +STATIC void +xfs_xmi_item_free( + struct xfs_xmi_log_item *xmi_lip) +{ + kvfree(xmi_lip->xmi_item.li_lv_shadow); + kmem_cache_free(xfs_xmi_cache, xmi_lip); +} + +/* + * Freeing the XMI requires that we remove it from the AIL if it has already + * been placed there. However, the XMI may not yet have been placed in the AIL + * when called by xfs_xmi_release() from XMD processing due to the ordering of + * committed vs unpin operations in bulk insert operations. Hence the reference + * count to ensure only the last caller frees the XMI. + */ +STATIC void +xfs_xmi_release( + struct xfs_xmi_log_item *xmi_lip) +{ + ASSERT(atomic_read(&xmi_lip->xmi_refcount) > 0); + if (atomic_dec_and_test(&xmi_lip->xmi_refcount)) { + xfs_trans_ail_delete(&xmi_lip->xmi_item, 0); + xfs_xmi_item_free(xmi_lip); + } +} + + +STATIC void +xfs_xmi_item_size( + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) +{ + *nvecs += 1; + *nbytes += sizeof(struct xfs_xmi_log_format); +} + +/* + * This is called to fill in the vector of log iovecs for the given xmi log + * item. We use only 1 iovec, and we point that at the xmi_log_format structure + * embedded in the xmi item. + */ +STATIC void +xfs_xmi_item_format( + struct xfs_log_item *lip, + struct xfs_log_vec *lv) +{ + struct xfs_xmi_log_item *xmi_lip = XMI_ITEM(lip); + struct xfs_log_iovec *vecp = NULL; + + xmi_lip->xmi_format.xmi_type = XFS_LI_XMI; + xmi_lip->xmi_format.xmi_size = 1; + + xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_XMI_FORMAT, + &xmi_lip->xmi_format, + sizeof(struct xfs_xmi_log_format)); +} + +/* + * The unpin operation is the last place an XMI is manipulated in the log. It + * is either inserted in the AIL or aborted in the event of a log I/O error. In + * either case, the XMI transaction has been successfully committed to make it + * this far. Therefore, we expect whoever committed the XMI to either construct + * and commit the XMD or drop the XMD's reference in the event of error. Simply + * drop the log's XMI reference now that the log is done with it. + */ +STATIC void +xfs_xmi_item_unpin( + struct xfs_log_item *lip, + int remove) +{ + struct xfs_xmi_log_item *xmi_lip = XMI_ITEM(lip); + + xfs_xmi_release(xmi_lip); +} + +/* + * The XMI has been either committed or aborted if the transaction has been + * cancelled. If the transaction was cancelled, an XMD isn't going to be + * constructed and thus we free the XMI here directly. + */ +STATIC void +xfs_xmi_item_release( + struct xfs_log_item *lip) +{ + xfs_xmi_release(XMI_ITEM(lip)); +} + +/* Allocate and initialize an xmi item. */ +STATIC struct xfs_xmi_log_item * +xfs_xmi_init( + struct xfs_mount *mp) + +{ + struct xfs_xmi_log_item *xmi_lip; + + xmi_lip = kmem_cache_zalloc(xfs_xmi_cache, GFP_KERNEL | __GFP_NOFAIL); + + xfs_log_item_init(mp, &xmi_lip->xmi_item, XFS_LI_XMI, &xfs_xmi_item_ops); + xmi_lip->xmi_format.xmi_id = (uintptr_t)(void *)xmi_lip; + atomic_set(&xmi_lip->xmi_refcount, 2); + + return xmi_lip; +} + +static inline struct xfs_xmd_log_item *XMD_ITEM(struct xfs_log_item *lip) +{ + return container_of(lip, struct xfs_xmd_log_item, xmd_item); +} + +STATIC bool +xfs_xmi_item_match( + struct xfs_log_item *lip, + uint64_t intent_id) +{ + return XMI_ITEM(lip)->xmi_format.xmi_id == intent_id; +} + +static const struct xfs_item_ops xfs_xmi_item_ops = { + .flags = XFS_ITEM_INTENT, + .iop_size = xfs_xmi_item_size, + .iop_format = xfs_xmi_item_format, + .iop_unpin = xfs_xmi_item_unpin, + .iop_release = xfs_xmi_item_release, + .iop_match = xfs_xmi_item_match, +}; + +/* + * This routine is called to create an in-core file mapping exchange item from + * the xmi format structure which was logged on disk. It allocates an in-core + * xmi, copies the exchange information from the format structure into it, and + * adds the xmi to the AIL with the given LSN. + */ +STATIC int +xlog_recover_xmi_commit_pass2( + struct xlog *log, + struct list_head *buffer_list, + struct xlog_recover_item *item, + xfs_lsn_t lsn) +{ + struct xfs_mount *mp = log->l_mp; + struct xfs_xmi_log_item *xmi_lip; + struct xfs_xmi_log_format *xmi_formatp; + size_t len; + + len = sizeof(struct xfs_xmi_log_format); + if (item->ri_buf[0].i_len != len) { + XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp); + return -EFSCORRUPTED; + } + + xmi_formatp = item->ri_buf[0].i_addr; + if (xmi_formatp->__pad != 0) { + XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp); + return -EFSCORRUPTED; + } + + xmi_lip = xfs_xmi_init(mp); + memcpy(&xmi_lip->xmi_format, xmi_formatp, len); + + /* not implemented yet */ + return -EIO; +} + +const struct xlog_recover_item_ops xlog_xmi_item_ops = { + .item_type = XFS_LI_XMI, + .commit_pass2 = xlog_recover_xmi_commit_pass2, +}; + +/* + * This routine is called when an XMD format structure is found in a committed + * transaction in the log. Its purpose is to cancel the corresponding XMI if it + * was still in the log. To do this it searches the AIL for the XMI with an id + * equal to that in the XMD format structure. If we find it we drop the XMD + * reference, which removes the XMI from the AIL and frees it. + */ +STATIC int +xlog_recover_xmd_commit_pass2( + struct xlog *log, + struct list_head *buffer_list, + struct xlog_recover_item *item, + xfs_lsn_t lsn) +{ + struct xfs_xmd_log_format *xmd_formatp; + + xmd_formatp = item->ri_buf[0].i_addr; + if (item->ri_buf[0].i_len != sizeof(struct xfs_xmd_log_format)) { + XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp); + return -EFSCORRUPTED; + } + + xlog_recover_release_intent(log, XFS_LI_XMI, xmd_formatp->xmd_xmi_id); + return 0; +} + +const struct xlog_recover_item_ops xlog_xmd_item_ops = { + .item_type = XFS_LI_XMD, + .commit_pass2 = xlog_recover_xmd_commit_pass2, +}; diff --git a/fs/xfs/xfs_exchmaps_item.h b/fs/xfs/xfs_exchmaps_item.h new file mode 100644 index 000000000000..ada1eb314e65 --- /dev/null +++ b/fs/xfs/xfs_exchmaps_item.h @@ -0,0 +1,59 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_EXCHMAPS_ITEM_H__ +#define __XFS_EXCHMAPS_ITEM_H__ + +/* + * The file mapping exchange intent item helps us exchange multiple file + * mappings between two inode forks. It does this by tracking the range of + * file block offsets that still need to be exchanged, and relogs as progress + * happens. + * + * *I items should be recorded in the *first* of a series of rolled + * transactions, and the *D items should be recorded in the same transaction + * that records the associated bmbt updates. + * + * Should the system crash after the commit of the first transaction but + * before the commit of the final transaction in a series, log recovery will + * use the redo information recorded by the intent items to replay the + * rest of the mapping exchanges. + */ + +/* kernel only XMI/XMD definitions */ + +struct xfs_mount; +struct kmem_cache; + +/* + * This is the incore file mapping exchange intent log item. It is used to log + * the fact that we are exchanging mappings between two files. It is used in + * conjunction with the incore file mapping exchange done log item described + * below. + * + * These log items follow the same rules as struct xfs_efi_log_item; see the + * comments about that structure (in xfs_extfree_item.h) for more details. + */ +struct xfs_xmi_log_item { + struct xfs_log_item xmi_item; + atomic_t xmi_refcount; + struct xfs_xmi_log_format xmi_format; +}; + +/* + * This is the incore file mapping exchange done log item. It is used to log + * the fact that an exchange mentioned in an earlier xmi item have been + * performed. + */ +struct xfs_xmd_log_item { + struct xfs_log_item xmd_item; + struct xfs_xmi_log_item *xmd_intent_log_item; + struct xfs_xmd_log_format xmd_format; +}; + +extern struct kmem_cache *xfs_xmi_cache; +extern struct kmem_cache *xfs_xmd_cache; + +#endif /* __XFS_EXCHMAPS_ITEM_H__ */ diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 41aec991433c..1e5ba95adf2c 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1789,6 +1789,8 @@ static const struct xlog_recover_item_ops *xlog_recover_item_ops[] = { &xlog_bud_item_ops, &xlog_attri_item_ops, &xlog_attrd_item_ops, + &xlog_xmi_item_ops, + &xlog_xmd_item_ops, }; static const struct xlog_recover_item_ops * diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index dbda72df3419..5c9ba974252d 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -43,6 +43,7 @@ #include "xfs_iunlink_item.h" #include "xfs_dahash_test.h" #include "xfs_rtbitmap.h" +#include "xfs_exchmaps_item.h" #include "scrub/stats.h" #include "scrub/rcbag_btree.h" @@ -2189,8 +2190,24 @@ xfs_init_caches(void) if (!xfs_iunlink_cache) goto out_destroy_attri_cache; + xfs_xmd_cache = kmem_cache_create("xfs_xmd_item", + sizeof(struct xfs_xmd_log_item), + 0, 0, NULL); + if (!xfs_xmd_cache) + goto out_destroy_iul_cache; + + xfs_xmi_cache = kmem_cache_create("xfs_xmi_item", + sizeof(struct xfs_xmi_log_item), + 0, 0, NULL); + if (!xfs_xmi_cache) + goto out_destroy_xmd_cache; + return 0; + out_destroy_xmd_cache: + kmem_cache_destroy(xfs_xmd_cache); + out_destroy_iul_cache: + kmem_cache_destroy(xfs_iunlink_cache); out_destroy_attri_cache: kmem_cache_destroy(xfs_attri_cache); out_destroy_attrd_cache: @@ -2247,6 +2264,8 @@ xfs_destroy_caches(void) * destroy caches. */ rcu_barrier(); + kmem_cache_destroy(xfs_xmd_cache); + kmem_cache_destroy(xfs_xmi_cache); kmem_cache_destroy(xfs_iunlink_cache); kmem_cache_destroy(xfs_attri_cache); kmem_cache_destroy(xfs_attrd_cache); -- cgit v1.2.3-59-g8ed1b From 966ceafc7a437105ecfe1cadb3747b2965a260ca Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:17 -0700 Subject: xfs: create deferred log items for file mapping exchanges Now that we've created the skeleton of a log intent item to track and restart file mapping exchange operations, add the upper level logic to commit intent items and turn them into concrete work recorded in the log. This builds on the existing bmap update intent items that have been around for a while now. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/libxfs/xfs_defer.c | 6 + fs/xfs/libxfs/xfs_defer.h | 2 +- fs/xfs/libxfs/xfs_exchmaps.c | 1045 +++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_exchmaps.h | 118 +++++ fs/xfs/libxfs/xfs_log_format.h | 24 +- fs/xfs/libxfs/xfs_trans_space.h | 4 + fs/xfs/xfs_exchmaps_item.c | 368 +++++++++++++- fs/xfs/xfs_exchmaps_item.h | 5 + fs/xfs/xfs_exchrange.c | 49 ++ fs/xfs/xfs_exchrange.h | 8 + fs/xfs/xfs_trace.c | 1 + fs/xfs/xfs_trace.h | 217 ++++++++ 13 files changed, 1844 insertions(+), 4 deletions(-) create mode 100644 fs/xfs/libxfs/xfs_exchmaps.c create mode 100644 fs/xfs/libxfs/xfs_exchmaps.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 68ca9726e7b7..b547a3dc03f8 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -34,6 +34,7 @@ xfs-y += $(addprefix libxfs/, \ xfs_dir2_node.o \ xfs_dir2_sf.o \ xfs_dquot_buf.o \ + xfs_exchmaps.o \ xfs_ialloc.o \ xfs_ialloc_btree.o \ xfs_iext_tree.o \ diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c index c13276095cc0..061cc01245a9 100644 --- a/fs/xfs/libxfs/xfs_defer.c +++ b/fs/xfs/libxfs/xfs_defer.c @@ -27,6 +27,7 @@ #include "xfs_da_btree.h" #include "xfs_attr.h" #include "xfs_trans_priv.h" +#include "xfs_exchmaps.h" static struct kmem_cache *xfs_defer_pending_cache; @@ -1176,6 +1177,10 @@ xfs_defer_init_item_caches(void) error = xfs_attr_intent_init_cache(); if (error) goto err; + error = xfs_exchmaps_intent_init_cache(); + if (error) + goto err; + return 0; err: xfs_defer_destroy_item_caches(); @@ -1186,6 +1191,7 @@ err: void xfs_defer_destroy_item_caches(void) { + xfs_exchmaps_intent_destroy_cache(); xfs_attr_intent_destroy_cache(); xfs_extfree_intent_destroy_cache(); xfs_bmap_intent_destroy_cache(); diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h index 18a9fb92dde8..81cca60d70a3 100644 --- a/fs/xfs/libxfs/xfs_defer.h +++ b/fs/xfs/libxfs/xfs_defer.h @@ -72,7 +72,7 @@ extern const struct xfs_defer_op_type xfs_rmap_update_defer_type; extern const struct xfs_defer_op_type xfs_extent_free_defer_type; extern const struct xfs_defer_op_type xfs_agfl_free_defer_type; extern const struct xfs_defer_op_type xfs_attr_defer_type; - +extern const struct xfs_defer_op_type xfs_exchmaps_defer_type; /* * Deferred operation item relogging limits. diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c new file mode 100644 index 000000000000..b8e9450cc175 --- /dev/null +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -0,0 +1,1045 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_inode.h" +#include "xfs_trans.h" +#include "xfs_bmap.h" +#include "xfs_icache.h" +#include "xfs_quota.h" +#include "xfs_exchmaps.h" +#include "xfs_trace.h" +#include "xfs_bmap_btree.h" +#include "xfs_trans_space.h" +#include "xfs_error.h" +#include "xfs_errortag.h" +#include "xfs_health.h" +#include "xfs_exchmaps_item.h" + +struct kmem_cache *xfs_exchmaps_intent_cache; + +/* bmbt mappings adjacent to a pair of records. */ +struct xfs_exchmaps_adjacent { + struct xfs_bmbt_irec left1; + struct xfs_bmbt_irec right1; + struct xfs_bmbt_irec left2; + struct xfs_bmbt_irec right2; +}; + +#define ADJACENT_INIT { \ + .left1 = { .br_startblock = HOLESTARTBLOCK }, \ + .right1 = { .br_startblock = HOLESTARTBLOCK }, \ + .left2 = { .br_startblock = HOLESTARTBLOCK }, \ + .right2 = { .br_startblock = HOLESTARTBLOCK }, \ +} + +/* Information to reset reflink flag / CoW fork state after an exchange. */ + +/* + * If the reflink flag is set on either inode, make sure it has an incore CoW + * fork, since all reflink inodes must have them. If there's a CoW fork and it + * has mappings in it, make sure the inodes are tagged appropriately so that + * speculative preallocations can be GC'd if we run low of space. + */ +static inline void +xfs_exchmaps_ensure_cowfork( + struct xfs_inode *ip) +{ + struct xfs_ifork *cfork; + + if (xfs_is_reflink_inode(ip)) + xfs_ifork_init_cow(ip); + + cfork = xfs_ifork_ptr(ip, XFS_COW_FORK); + if (!cfork) + return; + if (cfork->if_bytes > 0) + xfs_inode_set_cowblocks_tag(ip); + else + xfs_inode_clear_cowblocks_tag(ip); +} + +/* + * Adjust the on-disk inode size upwards if needed so that we never add + * mappings into the file past EOF. This is crucial so that log recovery won't + * get confused by the sudden appearance of post-eof mappings. + */ +STATIC void +xfs_exchmaps_update_size( + struct xfs_trans *tp, + struct xfs_inode *ip, + struct xfs_bmbt_irec *imap, + xfs_fsize_t new_isize) +{ + struct xfs_mount *mp = tp->t_mountp; + xfs_fsize_t len; + + if (new_isize < 0) + return; + + len = min(XFS_FSB_TO_B(mp, imap->br_startoff + imap->br_blockcount), + new_isize); + + if (len <= ip->i_disk_size) + return; + + trace_xfs_exchmaps_update_inode_size(ip, len); + + ip->i_disk_size = len; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); +} + +/* Advance the incore state tracking after exchanging a mapping. */ +static inline void +xmi_advance( + struct xfs_exchmaps_intent *xmi, + const struct xfs_bmbt_irec *irec) +{ + xmi->xmi_startoff1 += irec->br_blockcount; + xmi->xmi_startoff2 += irec->br_blockcount; + xmi->xmi_blockcount -= irec->br_blockcount; +} + +/* Do we still have more mappings to exchange? */ +static inline bool +xmi_has_more_exchange_work(const struct xfs_exchmaps_intent *xmi) +{ + return xmi->xmi_blockcount > 0; +} + +/* Do we have post-operation cleanups to perform? */ +static inline bool +xmi_has_postop_work(const struct xfs_exchmaps_intent *xmi) +{ + return xmi->xmi_flags & (XFS_EXCHMAPS_CLEAR_INO1_REFLINK | + XFS_EXCHMAPS_CLEAR_INO2_REFLINK); +} + +/* Check all mappings to make sure we can actually exchange them. */ +int +xfs_exchmaps_check_forks( + struct xfs_mount *mp, + const struct xfs_exchmaps_req *req) +{ + struct xfs_ifork *ifp1, *ifp2; + int whichfork = xfs_exchmaps_reqfork(req); + + /* No fork? */ + ifp1 = xfs_ifork_ptr(req->ip1, whichfork); + ifp2 = xfs_ifork_ptr(req->ip2, whichfork); + if (!ifp1 || !ifp2) + return -EINVAL; + + /* We don't know how to exchange local format forks. */ + if (ifp1->if_format == XFS_DINODE_FMT_LOCAL || + ifp2->if_format == XFS_DINODE_FMT_LOCAL) + return -EINVAL; + + /* We don't support realtime data forks yet. */ + if (!XFS_IS_REALTIME_INODE(req->ip1)) + return 0; + if (whichfork == XFS_ATTR_FORK) + return 0; + return -EINVAL; +} + +#ifdef CONFIG_XFS_QUOTA +/* Log the actual updates to the quota accounting. */ +static inline void +xfs_exchmaps_update_quota( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi, + struct xfs_bmbt_irec *irec1, + struct xfs_bmbt_irec *irec2) +{ + int64_t ip1_delta = 0, ip2_delta = 0; + unsigned int qflag; + + qflag = XFS_IS_REALTIME_INODE(xmi->xmi_ip1) ? XFS_TRANS_DQ_RTBCOUNT : + XFS_TRANS_DQ_BCOUNT; + + if (xfs_bmap_is_real_extent(irec1)) { + ip1_delta -= irec1->br_blockcount; + ip2_delta += irec1->br_blockcount; + } + + if (xfs_bmap_is_real_extent(irec2)) { + ip1_delta += irec2->br_blockcount; + ip2_delta -= irec2->br_blockcount; + } + + xfs_trans_mod_dquot_byino(tp, xmi->xmi_ip1, qflag, ip1_delta); + xfs_trans_mod_dquot_byino(tp, xmi->xmi_ip2, qflag, ip2_delta); +} +#else +# define xfs_exchmaps_update_quota(tp, xmi, irec1, irec2) ((void)0) +#endif + +/* Decide if we want to skip this mapping from file1. */ +static inline bool +xfs_exchmaps_can_skip_mapping( + struct xfs_exchmaps_intent *xmi, + struct xfs_bmbt_irec *irec) +{ + /* Do not skip this mapping if the caller did not tell us to. */ + if (!(xmi->xmi_flags & XFS_EXCHMAPS_INO1_WRITTEN)) + return false; + + /* Do not skip mapped, written mappings. */ + if (xfs_bmap_is_written_extent(irec)) + return false; + + /* + * The mapping is unwritten or a hole. It cannot be a delalloc + * reservation because we already excluded those. It cannot be an + * unwritten mapping with dirty page cache because we flushed the page + * cache. We don't support realtime files yet, so we needn't (yet) + * deal with them. + */ + return true; +} + +/* + * Walk forward through the file ranges in @xmi until we find two different + * mappings to exchange. If there is work to do, return the mappings; + * otherwise we've reached the end of the range and xmi_blockcount will be + * zero. + * + * If the walk skips over a pair of mappings to the same storage, save them as + * the left records in @adj (if provided) so that the simulation phase can + * avoid an extra lookup. + */ +static int +xfs_exchmaps_find_mappings( + struct xfs_exchmaps_intent *xmi, + struct xfs_bmbt_irec *irec1, + struct xfs_bmbt_irec *irec2, + struct xfs_exchmaps_adjacent *adj) +{ + int nimaps; + int bmap_flags; + int error; + + bmap_flags = xfs_bmapi_aflag(xfs_exchmaps_whichfork(xmi)); + + for (; xmi_has_more_exchange_work(xmi); xmi_advance(xmi, irec1)) { + /* Read mapping from the first file */ + nimaps = 1; + error = xfs_bmapi_read(xmi->xmi_ip1, xmi->xmi_startoff1, + xmi->xmi_blockcount, irec1, &nimaps, + bmap_flags); + if (error) + return error; + if (nimaps != 1 || + irec1->br_startblock == DELAYSTARTBLOCK || + irec1->br_startoff != xmi->xmi_startoff1) { + /* + * We should never get no mapping or a delalloc mapping + * or something that doesn't match what we asked for, + * since the caller flushed both inodes and we hold the + * ILOCKs for both inodes. + */ + ASSERT(0); + return -EINVAL; + } + + if (xfs_exchmaps_can_skip_mapping(xmi, irec1)) { + trace_xfs_exchmaps_mapping1_skip(xmi->xmi_ip1, irec1); + continue; + } + + /* Read mapping from the second file */ + nimaps = 1; + error = xfs_bmapi_read(xmi->xmi_ip2, xmi->xmi_startoff2, + irec1->br_blockcount, irec2, &nimaps, + bmap_flags); + if (error) + return error; + if (nimaps != 1 || + irec2->br_startblock == DELAYSTARTBLOCK || + irec2->br_startoff != xmi->xmi_startoff2) { + /* + * We should never get no mapping or a delalloc mapping + * or something that doesn't match what we asked for, + * since the caller flushed both inodes and we hold the + * ILOCKs for both inodes. + */ + ASSERT(0); + return -EINVAL; + } + + /* + * We can only exchange as many blocks as the smaller of the + * two mapping maps. + */ + irec1->br_blockcount = min(irec1->br_blockcount, + irec2->br_blockcount); + + trace_xfs_exchmaps_mapping1(xmi->xmi_ip1, irec1); + trace_xfs_exchmaps_mapping2(xmi->xmi_ip2, irec2); + + /* We found something to exchange, so return it. */ + if (irec1->br_startblock != irec2->br_startblock) + return 0; + + /* + * Two mappings pointing to the same physical block must not + * have different states; that's filesystem corruption. Move + * on to the next mapping if they're both holes or both point + * to the same physical space extent. + */ + if (irec1->br_state != irec2->br_state) { + xfs_bmap_mark_sick(xmi->xmi_ip1, + xfs_exchmaps_whichfork(xmi)); + xfs_bmap_mark_sick(xmi->xmi_ip2, + xfs_exchmaps_whichfork(xmi)); + return -EFSCORRUPTED; + } + + /* + * Save the mappings if we're estimating work and skipping + * these identical mappings. + */ + if (adj) { + memcpy(&adj->left1, irec1, sizeof(*irec1)); + memcpy(&adj->left2, irec2, sizeof(*irec2)); + } + } + + return 0; +} + +/* Exchange these two mappings. */ +static void +xfs_exchmaps_one_step( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi, + struct xfs_bmbt_irec *irec1, + struct xfs_bmbt_irec *irec2) +{ + int whichfork = xfs_exchmaps_whichfork(xmi); + + xfs_exchmaps_update_quota(tp, xmi, irec1, irec2); + + /* Remove both mappings. */ + xfs_bmap_unmap_extent(tp, xmi->xmi_ip1, whichfork, irec1); + xfs_bmap_unmap_extent(tp, xmi->xmi_ip2, whichfork, irec2); + + /* + * Re-add both mappings. We exchange the file offsets between the two + * maps and add the opposite map, which has the effect of filling the + * logical offsets we just unmapped, but with with the physical mapping + * information exchanged. + */ + swap(irec1->br_startoff, irec2->br_startoff); + xfs_bmap_map_extent(tp, xmi->xmi_ip1, whichfork, irec2); + xfs_bmap_map_extent(tp, xmi->xmi_ip2, whichfork, irec1); + + /* Make sure we're not adding mappings past EOF. */ + if (whichfork == XFS_DATA_FORK) { + xfs_exchmaps_update_size(tp, xmi->xmi_ip1, irec2, + xmi->xmi_isize1); + xfs_exchmaps_update_size(tp, xmi->xmi_ip2, irec1, + xmi->xmi_isize2); + } + + /* + * Advance our cursor and exit. The caller (either defer ops or log + * recovery) will log the XMD item, and if *blockcount is nonzero, it + * will log a new XMI item for the remainder and call us back. + */ + xmi_advance(xmi, irec1); +} + +/* Clear the reflink flag after an exchange. */ +static inline void +xfs_exchmaps_clear_reflink( + struct xfs_trans *tp, + struct xfs_inode *ip) +{ + trace_xfs_reflink_unset_inode_flag(ip); + + ip->i_diflags2 &= ~XFS_DIFLAG2_REFLINK; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); +} + +/* Finish whatever work might come after an exchange operation. */ +static int +xfs_exchmaps_do_postop_work( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi) +{ + if (xmi->xmi_flags & XFS_EXCHMAPS_CLEAR_INO1_REFLINK) { + xfs_exchmaps_clear_reflink(tp, xmi->xmi_ip1); + xmi->xmi_flags &= ~XFS_EXCHMAPS_CLEAR_INO1_REFLINK; + } + + if (xmi->xmi_flags & XFS_EXCHMAPS_CLEAR_INO2_REFLINK) { + xfs_exchmaps_clear_reflink(tp, xmi->xmi_ip2); + xmi->xmi_flags &= ~XFS_EXCHMAPS_CLEAR_INO2_REFLINK; + } + + return 0; +} + +/* Finish one step in a mapping exchange operation, possibly relogging. */ +int +xfs_exchmaps_finish_one( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi) +{ + struct xfs_bmbt_irec irec1, irec2; + int error; + + if (xmi_has_more_exchange_work(xmi)) { + /* + * If the operation state says that some range of the files + * have not yet been exchanged, look for mappings in that range + * to exchange. If we find some mappings, exchange them. + */ + error = xfs_exchmaps_find_mappings(xmi, &irec1, &irec2, NULL); + if (error) + return error; + + if (xmi_has_more_exchange_work(xmi)) + xfs_exchmaps_one_step(tp, xmi, &irec1, &irec2); + + /* + * If the caller asked us to exchange the file sizes after the + * exchange and either we just exchanged the last mappings in + * the range or we didn't find anything to exchange, update the + * ondisk file sizes. + */ + if ((xmi->xmi_flags & XFS_EXCHMAPS_SET_SIZES) && + !xmi_has_more_exchange_work(xmi)) { + xmi->xmi_ip1->i_disk_size = xmi->xmi_isize1; + xmi->xmi_ip2->i_disk_size = xmi->xmi_isize2; + + xfs_trans_log_inode(tp, xmi->xmi_ip1, XFS_ILOG_CORE); + xfs_trans_log_inode(tp, xmi->xmi_ip2, XFS_ILOG_CORE); + } + } else if (xmi_has_postop_work(xmi)) { + /* + * Now that we're finished with the exchange operation, + * complete the post-op cleanup work. + */ + error = xfs_exchmaps_do_postop_work(tp, xmi); + if (error) + return error; + } + + /* If we still have work to do, ask for a new transaction. */ + if (xmi_has_more_exchange_work(xmi) || xmi_has_postop_work(xmi)) { + trace_xfs_exchmaps_defer(tp->t_mountp, xmi); + return -EAGAIN; + } + + /* + * If we reach here, we've finished all the exchange work and the post + * operation work. The last thing we need to do before returning to + * the caller is to make sure that COW forks are set up correctly. + */ + if (!(xmi->xmi_flags & XFS_EXCHMAPS_ATTR_FORK)) { + xfs_exchmaps_ensure_cowfork(xmi->xmi_ip1); + xfs_exchmaps_ensure_cowfork(xmi->xmi_ip2); + } + + return 0; +} + +/* + * Compute the amount of bmbt blocks we should reserve for each file. In the + * worst case, each exchange will fill a hole with a new mapping, which could + * result in a btree split every time we add a new leaf block. + */ +static inline uint64_t +xfs_exchmaps_bmbt_blocks( + struct xfs_mount *mp, + const struct xfs_exchmaps_req *req) +{ + return howmany_64(req->nr_exchanges, + XFS_MAX_CONTIG_BMAPS_PER_BLOCK(mp)) * + XFS_EXTENTADD_SPACE_RES(mp, xfs_exchmaps_reqfork(req)); +} + +/* Compute the space we should reserve for the rmap btree expansions. */ +static inline uint64_t +xfs_exchmaps_rmapbt_blocks( + struct xfs_mount *mp, + const struct xfs_exchmaps_req *req) +{ + if (!xfs_has_rmapbt(mp)) + return 0; + if (XFS_IS_REALTIME_INODE(req->ip1)) + return 0; + + return howmany_64(req->nr_exchanges, + XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp)) * + XFS_RMAPADD_SPACE_RES(mp); +} + +/* Estimate the bmbt and rmapbt overhead required to exchange mappings. */ +static int +xfs_exchmaps_estimate_overhead( + struct xfs_exchmaps_req *req) +{ + struct xfs_mount *mp = req->ip1->i_mount; + xfs_filblks_t bmbt_blocks; + xfs_filblks_t rmapbt_blocks; + xfs_filblks_t resblks = req->resblks; + + /* + * Compute the number of bmbt and rmapbt blocks we might need to handle + * the estimated number of exchanges. + */ + bmbt_blocks = xfs_exchmaps_bmbt_blocks(mp, req); + rmapbt_blocks = xfs_exchmaps_rmapbt_blocks(mp, req); + + trace_xfs_exchmaps_overhead(mp, bmbt_blocks, rmapbt_blocks); + + /* Make sure the change in file block count doesn't overflow. */ + if (check_add_overflow(req->ip1_bcount, bmbt_blocks, &req->ip1_bcount)) + return -EFBIG; + if (check_add_overflow(req->ip2_bcount, bmbt_blocks, &req->ip2_bcount)) + return -EFBIG; + + /* + * Add together the number of blocks we need to handle btree growth, + * then add it to the number of blocks we need to reserve to this + * transaction. + */ + if (check_add_overflow(resblks, bmbt_blocks, &resblks)) + return -ENOSPC; + if (check_add_overflow(resblks, bmbt_blocks, &resblks)) + return -ENOSPC; + if (check_add_overflow(resblks, rmapbt_blocks, &resblks)) + return -ENOSPC; + if (check_add_overflow(resblks, rmapbt_blocks, &resblks)) + return -ENOSPC; + + /* Can't actually reserve more than UINT_MAX blocks. */ + if (req->resblks > UINT_MAX) + return -ENOSPC; + + req->resblks = resblks; + trace_xfs_exchmaps_final_estimate(req); + return 0; +} + +/* Decide if we can merge two real mappings. */ +static inline bool +xmi_can_merge( + const struct xfs_bmbt_irec *b1, + const struct xfs_bmbt_irec *b2) +{ + /* Don't merge holes. */ + if (b1->br_startblock == HOLESTARTBLOCK || + b2->br_startblock == HOLESTARTBLOCK) + return false; + + /* We don't merge holes. */ + if (!xfs_bmap_is_real_extent(b1) || !xfs_bmap_is_real_extent(b2)) + return false; + + if (b1->br_startoff + b1->br_blockcount == b2->br_startoff && + b1->br_startblock + b1->br_blockcount == b2->br_startblock && + b1->br_state == b2->br_state && + b1->br_blockcount + b2->br_blockcount <= XFS_MAX_BMBT_EXTLEN) + return true; + + return false; +} + +/* + * Decide if we can merge three mappings. Caller must ensure all three + * mappings must not be holes or delalloc reservations. + */ +static inline bool +xmi_can_merge_all( + const struct xfs_bmbt_irec *l, + const struct xfs_bmbt_irec *m, + const struct xfs_bmbt_irec *r) +{ + xfs_filblks_t new_len; + + new_len = l->br_blockcount + m->br_blockcount + r->br_blockcount; + return new_len <= XFS_MAX_BMBT_EXTLEN; +} + +#define CLEFT_CONTIG 0x01 +#define CRIGHT_CONTIG 0x02 +#define CHOLE 0x04 +#define CBOTH_CONTIG (CLEFT_CONTIG | CRIGHT_CONTIG) + +#define NLEFT_CONTIG 0x10 +#define NRIGHT_CONTIG 0x20 +#define NHOLE 0x40 +#define NBOTH_CONTIG (NLEFT_CONTIG | NRIGHT_CONTIG) + +/* Estimate the effect of a single exchange on mapping count. */ +static inline int +xmi_delta_nextents_step( + struct xfs_mount *mp, + const struct xfs_bmbt_irec *left, + const struct xfs_bmbt_irec *curr, + const struct xfs_bmbt_irec *new, + const struct xfs_bmbt_irec *right) +{ + bool lhole, rhole, chole, nhole; + unsigned int state = 0; + int ret = 0; + + lhole = left->br_startblock == HOLESTARTBLOCK; + rhole = right->br_startblock == HOLESTARTBLOCK; + chole = curr->br_startblock == HOLESTARTBLOCK; + nhole = new->br_startblock == HOLESTARTBLOCK; + + if (chole) + state |= CHOLE; + if (!lhole && !chole && xmi_can_merge(left, curr)) + state |= CLEFT_CONTIG; + if (!rhole && !chole && xmi_can_merge(curr, right)) + state |= CRIGHT_CONTIG; + if ((state & CBOTH_CONTIG) == CBOTH_CONTIG && + !xmi_can_merge_all(left, curr, right)) + state &= ~CRIGHT_CONTIG; + + if (nhole) + state |= NHOLE; + if (!lhole && !nhole && xmi_can_merge(left, new)) + state |= NLEFT_CONTIG; + if (!rhole && !nhole && xmi_can_merge(new, right)) + state |= NRIGHT_CONTIG; + if ((state & NBOTH_CONTIG) == NBOTH_CONTIG && + !xmi_can_merge_all(left, new, right)) + state &= ~NRIGHT_CONTIG; + + switch (state & (CLEFT_CONTIG | CRIGHT_CONTIG | CHOLE)) { + case CLEFT_CONTIG | CRIGHT_CONTIG: + /* + * left/curr/right are the same mapping, so deleting curr + * causes 2 new mappings to be created. + */ + ret += 2; + break; + case 0: + /* + * curr is not contiguous with any mapping, so we remove curr + * completely + */ + ret--; + break; + case CHOLE: + /* hole, do nothing */ + break; + case CLEFT_CONTIG: + case CRIGHT_CONTIG: + /* trim either left or right, no change */ + break; + } + + switch (state & (NLEFT_CONTIG | NRIGHT_CONTIG | NHOLE)) { + case NLEFT_CONTIG | NRIGHT_CONTIG: + /* + * left/curr/right will become the same mapping, so adding + * curr causes the deletion of right. + */ + ret--; + break; + case 0: + /* new is not contiguous with any mapping */ + ret++; + break; + case NHOLE: + /* hole, do nothing. */ + break; + case NLEFT_CONTIG: + case NRIGHT_CONTIG: + /* new is absorbed into left or right, no change */ + break; + } + + trace_xfs_exchmaps_delta_nextents_step(mp, left, curr, new, right, ret, + state); + return ret; +} + +/* Make sure we don't overflow the extent (mapping) counters. */ +static inline int +xmi_ensure_delta_nextents( + struct xfs_exchmaps_req *req, + struct xfs_inode *ip, + int64_t delta) +{ + struct xfs_mount *mp = ip->i_mount; + int whichfork = xfs_exchmaps_reqfork(req); + struct xfs_ifork *ifp = xfs_ifork_ptr(ip, whichfork); + uint64_t new_nextents; + xfs_extnum_t max_nextents; + + if (delta < 0) + return 0; + + /* + * It's always an error if the delta causes integer overflow. delta + * needs an explicit cast here to avoid warnings about implicit casts + * coded into the overflow check. + */ + if (check_add_overflow(ifp->if_nextents, (uint64_t)delta, + &new_nextents)) + return -EFBIG; + + if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_REDUCE_MAX_IEXTENTS) && + new_nextents > 10) + return -EFBIG; + + /* + * We always promote both inodes to have large extent counts if the + * superblock feature is enabled, so we only need to check against the + * theoretical maximum. + */ + max_nextents = xfs_iext_max_nextents(xfs_has_large_extent_counts(mp), + whichfork); + if (new_nextents > max_nextents) + return -EFBIG; + + return 0; +} + +/* Find the next mapping after irec. */ +static inline int +xmi_next( + struct xfs_inode *ip, + int bmap_flags, + const struct xfs_bmbt_irec *irec, + struct xfs_bmbt_irec *nrec) +{ + xfs_fileoff_t off; + xfs_filblks_t blockcount; + int nimaps = 1; + int error; + + off = irec->br_startoff + irec->br_blockcount; + blockcount = XFS_MAX_FILEOFF - off; + error = xfs_bmapi_read(ip, off, blockcount, nrec, &nimaps, bmap_flags); + if (error) + return error; + if (nrec->br_startblock == DELAYSTARTBLOCK || + nrec->br_startoff != off) { + /* + * If we don't get the mapping we want, return a zero-length + * mapping, which our estimator function will pretend is a hole. + * We shouldn't get delalloc reservations. + */ + nrec->br_startblock = HOLESTARTBLOCK; + } + + return 0; +} + +int __init +xfs_exchmaps_intent_init_cache(void) +{ + xfs_exchmaps_intent_cache = kmem_cache_create("xfs_exchmaps_intent", + sizeof(struct xfs_exchmaps_intent), + 0, 0, NULL); + + return xfs_exchmaps_intent_cache != NULL ? 0 : -ENOMEM; +} + +void +xfs_exchmaps_intent_destroy_cache(void) +{ + kmem_cache_destroy(xfs_exchmaps_intent_cache); + xfs_exchmaps_intent_cache = NULL; +} + +/* + * Decide if we will exchange the reflink flags between the two files after the + * exchange. The only time we want to do this is if we're exchanging all + * mappings under EOF and the inode reflink flags have different states. + */ +static inline bool +xmi_can_exchange_reflink_flags( + const struct xfs_exchmaps_req *req, + unsigned int reflink_state) +{ + struct xfs_mount *mp = req->ip1->i_mount; + + if (hweight32(reflink_state) != 1) + return false; + if (req->startoff1 != 0 || req->startoff2 != 0) + return false; + if (req->blockcount != XFS_B_TO_FSB(mp, req->ip1->i_disk_size)) + return false; + if (req->blockcount != XFS_B_TO_FSB(mp, req->ip2->i_disk_size)) + return false; + return true; +} + + +/* Allocate and initialize a new incore intent item from a request. */ +struct xfs_exchmaps_intent * +xfs_exchmaps_init_intent( + const struct xfs_exchmaps_req *req) +{ + struct xfs_exchmaps_intent *xmi; + unsigned int rs = 0; + + xmi = kmem_cache_zalloc(xfs_exchmaps_intent_cache, + GFP_NOFS | __GFP_NOFAIL); + INIT_LIST_HEAD(&xmi->xmi_list); + xmi->xmi_ip1 = req->ip1; + xmi->xmi_ip2 = req->ip2; + xmi->xmi_startoff1 = req->startoff1; + xmi->xmi_startoff2 = req->startoff2; + xmi->xmi_blockcount = req->blockcount; + xmi->xmi_isize1 = xmi->xmi_isize2 = -1; + xmi->xmi_flags = req->flags & XFS_EXCHMAPS_PARAMS; + + if (xfs_exchmaps_whichfork(xmi) == XFS_ATTR_FORK) + return xmi; + + if (req->flags & XFS_EXCHMAPS_SET_SIZES) { + xmi->xmi_flags |= XFS_EXCHMAPS_SET_SIZES; + xmi->xmi_isize1 = req->ip2->i_disk_size; + xmi->xmi_isize2 = req->ip1->i_disk_size; + } + + /* Record the state of each inode's reflink flag before the op. */ + if (xfs_is_reflink_inode(req->ip1)) + rs |= 1; + if (xfs_is_reflink_inode(req->ip2)) + rs |= 2; + + /* + * Figure out if we're clearing the reflink flags (which effectively + * exchanges them) after the operation. + */ + if (xmi_can_exchange_reflink_flags(req, rs)) { + if (rs & 1) + xmi->xmi_flags |= XFS_EXCHMAPS_CLEAR_INO1_REFLINK; + if (rs & 2) + xmi->xmi_flags |= XFS_EXCHMAPS_CLEAR_INO2_REFLINK; + } + + return xmi; +} + +/* + * Estimate the number of exchange operations and the number of file blocks + * in each file that will be affected by the exchange operation. + */ +int +xfs_exchmaps_estimate( + struct xfs_exchmaps_req *req) +{ + struct xfs_exchmaps_intent *xmi; + struct xfs_bmbt_irec irec1, irec2; + struct xfs_exchmaps_adjacent adj = ADJACENT_INIT; + xfs_filblks_t ip1_blocks = 0, ip2_blocks = 0; + int64_t d_nexts1, d_nexts2; + int bmap_flags; + int error; + + ASSERT(!(req->flags & ~XFS_EXCHMAPS_PARAMS)); + + bmap_flags = xfs_bmapi_aflag(xfs_exchmaps_reqfork(req)); + xmi = xfs_exchmaps_init_intent(req); + + /* + * To guard against the possibility of overflowing the extent counters, + * we have to estimate an upper bound on the potential increase in that + * counter. We can split the mapping at each end of the range, and for + * each step of the exchange we can split the mapping that we're + * working on if the mappings do not align. + */ + d_nexts1 = d_nexts2 = 3; + + while (xmi_has_more_exchange_work(xmi)) { + /* + * Walk through the file ranges until we find something to + * exchange. Because we're simulating the exchange, pass in + * adj to capture skipped mappings for correct estimation of + * bmbt record merges. + */ + error = xfs_exchmaps_find_mappings(xmi, &irec1, &irec2, &adj); + if (error) + goto out_free; + if (!xmi_has_more_exchange_work(xmi)) + break; + + /* Update accounting. */ + if (xfs_bmap_is_real_extent(&irec1)) + ip1_blocks += irec1.br_blockcount; + if (xfs_bmap_is_real_extent(&irec2)) + ip2_blocks += irec2.br_blockcount; + req->nr_exchanges++; + + /* Read the next mappings from both files. */ + error = xmi_next(req->ip1, bmap_flags, &irec1, &adj.right1); + if (error) + goto out_free; + + error = xmi_next(req->ip2, bmap_flags, &irec2, &adj.right2); + if (error) + goto out_free; + + /* Update extent count deltas. */ + d_nexts1 += xmi_delta_nextents_step(req->ip1->i_mount, + &adj.left1, &irec1, &irec2, &adj.right1); + + d_nexts2 += xmi_delta_nextents_step(req->ip1->i_mount, + &adj.left2, &irec2, &irec1, &adj.right2); + + /* Now pretend we exchanged the mappings. */ + if (xmi_can_merge(&adj.left2, &irec1)) + adj.left2.br_blockcount += irec1.br_blockcount; + else + memcpy(&adj.left2, &irec1, sizeof(irec1)); + + if (xmi_can_merge(&adj.left1, &irec2)) + adj.left1.br_blockcount += irec2.br_blockcount; + else + memcpy(&adj.left1, &irec2, sizeof(irec2)); + + xmi_advance(xmi, &irec1); + } + + /* Account for the blocks that are being exchanged. */ + if (XFS_IS_REALTIME_INODE(req->ip1) && + xfs_exchmaps_reqfork(req) == XFS_DATA_FORK) { + req->ip1_rtbcount = ip1_blocks; + req->ip2_rtbcount = ip2_blocks; + } else { + req->ip1_bcount = ip1_blocks; + req->ip2_bcount = ip2_blocks; + } + + /* + * Make sure that both forks have enough slack left in their extent + * counters that the exchange operation will not overflow. + */ + trace_xfs_exchmaps_delta_nextents(req, d_nexts1, d_nexts2); + if (req->ip1 == req->ip2) { + error = xmi_ensure_delta_nextents(req, req->ip1, + d_nexts1 + d_nexts2); + } else { + error = xmi_ensure_delta_nextents(req, req->ip1, d_nexts1); + if (error) + goto out_free; + error = xmi_ensure_delta_nextents(req, req->ip2, d_nexts2); + } + if (error) + goto out_free; + + trace_xfs_exchmaps_initial_estimate(req); + error = xfs_exchmaps_estimate_overhead(req); +out_free: + kmem_cache_free(xfs_exchmaps_intent_cache, xmi); + return error; +} + +/* Set the reflink flag before an operation. */ +static inline void +xfs_exchmaps_set_reflink( + struct xfs_trans *tp, + struct xfs_inode *ip) +{ + trace_xfs_reflink_set_inode_flag(ip); + + ip->i_diflags2 |= XFS_DIFLAG2_REFLINK; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); +} + +/* + * If either file has shared blocks and we're exchanging data forks, we must + * flag the other file as having shared blocks so that we get the shared-block + * rmap functions if we need to fix up the rmaps. + */ +void +xfs_exchmaps_ensure_reflink( + struct xfs_trans *tp, + const struct xfs_exchmaps_intent *xmi) +{ + unsigned int rs = 0; + + if (xfs_is_reflink_inode(xmi->xmi_ip1)) + rs |= 1; + if (xfs_is_reflink_inode(xmi->xmi_ip2)) + rs |= 2; + + if ((rs & 1) && !xfs_is_reflink_inode(xmi->xmi_ip2)) + xfs_exchmaps_set_reflink(tp, xmi->xmi_ip2); + + if ((rs & 2) && !xfs_is_reflink_inode(xmi->xmi_ip1)) + xfs_exchmaps_set_reflink(tp, xmi->xmi_ip1); +} + +/* Set the large extent count flag before an operation if needed. */ +static inline void +xfs_exchmaps_ensure_large_extent_counts( + struct xfs_trans *tp, + struct xfs_inode *ip) +{ + if (xfs_inode_has_large_extent_counts(ip)) + return; + + ip->i_diflags2 |= XFS_DIFLAG2_NREXT64; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); +} + +/* Widen the extent counter fields of both inodes if necessary. */ +void +xfs_exchmaps_upgrade_extent_counts( + struct xfs_trans *tp, + const struct xfs_exchmaps_intent *xmi) +{ + if (!xfs_has_large_extent_counts(tp->t_mountp)) + return; + + xfs_exchmaps_ensure_large_extent_counts(tp, xmi->xmi_ip1); + xfs_exchmaps_ensure_large_extent_counts(tp, xmi->xmi_ip2); +} + +/* + * Schedule an exchange a range of mappings from one inode to another. + * + * The use of file mapping exchange log intent items ensures the operation can + * be resumed even if the system goes down. The caller must commit the + * transaction to start the work. + * + * The caller must ensure the inodes must be joined to the transaction and + * ILOCKd; they will still be joined to the transaction at exit. + */ +void +xfs_exchange_mappings( + struct xfs_trans *tp, + const struct xfs_exchmaps_req *req) +{ + struct xfs_exchmaps_intent *xmi; + + xfs_assert_ilocked(req->ip1, XFS_ILOCK_EXCL); + xfs_assert_ilocked(req->ip2, XFS_ILOCK_EXCL); + ASSERT(!(req->flags & ~XFS_EXCHMAPS_LOGGED_FLAGS)); + if (req->flags & XFS_EXCHMAPS_SET_SIZES) + ASSERT(!(req->flags & XFS_EXCHMAPS_ATTR_FORK)); + ASSERT(xfs_has_exchange_range(tp->t_mountp)); + + if (req->blockcount == 0) + return; + + xmi = xfs_exchmaps_init_intent(req); + xfs_exchmaps_defer_add(tp, xmi); + xfs_exchmaps_ensure_reflink(tp, xmi); + xfs_exchmaps_upgrade_extent_counts(tp, xmi); +} diff --git a/fs/xfs/libxfs/xfs_exchmaps.h b/fs/xfs/libxfs/xfs_exchmaps.h new file mode 100644 index 000000000000..e8fc3f80c68c --- /dev/null +++ b/fs/xfs/libxfs/xfs_exchmaps.h @@ -0,0 +1,118 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_EXCHMAPS_H__ +#define __XFS_EXCHMAPS_H__ + +/* In-core deferred operation info about a file mapping exchange request. */ +struct xfs_exchmaps_intent { + /* List of other incore deferred work. */ + struct list_head xmi_list; + + /* Inodes participating in the operation. */ + struct xfs_inode *xmi_ip1; + struct xfs_inode *xmi_ip2; + + /* File offset range information. */ + xfs_fileoff_t xmi_startoff1; + xfs_fileoff_t xmi_startoff2; + xfs_filblks_t xmi_blockcount; + + /* Set these file sizes after the operation, unless negative. */ + xfs_fsize_t xmi_isize1; + xfs_fsize_t xmi_isize2; + + uint64_t xmi_flags; /* XFS_EXCHMAPS_* flags */ +}; + +/* flags that can be passed to xfs_exchmaps_{estimate,mappings} */ +#define XFS_EXCHMAPS_PARAMS (XFS_EXCHMAPS_ATTR_FORK | \ + XFS_EXCHMAPS_SET_SIZES | \ + XFS_EXCHMAPS_INO1_WRITTEN) + +static inline int +xfs_exchmaps_whichfork(const struct xfs_exchmaps_intent *xmi) +{ + if (xmi->xmi_flags & XFS_EXCHMAPS_ATTR_FORK) + return XFS_ATTR_FORK; + return XFS_DATA_FORK; +} + +/* Parameters for a mapping exchange request. */ +struct xfs_exchmaps_req { + /* Inodes participating in the operation. */ + struct xfs_inode *ip1; + struct xfs_inode *ip2; + + /* File offset range information. */ + xfs_fileoff_t startoff1; + xfs_fileoff_t startoff2; + xfs_filblks_t blockcount; + + /* XFS_EXCHMAPS_* operation flags */ + uint64_t flags; + + /* + * Fields below this line are filled out by xfs_exchmaps_estimate; + * callers should initialize this part of the struct to zero. + */ + + /* + * Data device blocks to be moved out of ip1, and free space needed to + * handle the bmbt changes. + */ + xfs_filblks_t ip1_bcount; + + /* + * Data device blocks to be moved out of ip2, and free space needed to + * handle the bmbt changes. + */ + xfs_filblks_t ip2_bcount; + + /* rt blocks to be moved out of ip1. */ + xfs_filblks_t ip1_rtbcount; + + /* rt blocks to be moved out of ip2. */ + xfs_filblks_t ip2_rtbcount; + + /* Free space needed to handle the bmbt changes */ + unsigned long long resblks; + + /* Number of exchanges needed to complete the operation */ + unsigned long long nr_exchanges; +}; + +static inline int +xfs_exchmaps_reqfork(const struct xfs_exchmaps_req *req) +{ + if (req->flags & XFS_EXCHMAPS_ATTR_FORK) + return XFS_ATTR_FORK; + return XFS_DATA_FORK; +} + +int xfs_exchmaps_estimate(struct xfs_exchmaps_req *req); + +extern struct kmem_cache *xfs_exchmaps_intent_cache; + +int __init xfs_exchmaps_intent_init_cache(void); +void xfs_exchmaps_intent_destroy_cache(void); + +struct xfs_exchmaps_intent *xfs_exchmaps_init_intent( + const struct xfs_exchmaps_req *req); +void xfs_exchmaps_ensure_reflink(struct xfs_trans *tp, + const struct xfs_exchmaps_intent *xmi); +void xfs_exchmaps_upgrade_extent_counts(struct xfs_trans *tp, + const struct xfs_exchmaps_intent *xmi); + +int xfs_exchmaps_finish_one(struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi); + +int xfs_exchmaps_check_forks(struct xfs_mount *mp, + const struct xfs_exchmaps_req *req); + +void xfs_exchange_mappings(struct xfs_trans *tp, + const struct xfs_exchmaps_req *req); + +#endif /* __XFS_EXCHMAPS_H__ */ diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index 09024431cae9..8dbe1f997dfd 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -904,7 +904,29 @@ struct xfs_xmi_log_format { uint64_t xmi_isize2; /* intended file2 size */ }; -#define XFS_EXCHMAPS_LOGGED_FLAGS (0) +/* Exchange mappings between extended attribute forks instead of data forks. */ +#define XFS_EXCHMAPS_ATTR_FORK (1ULL << 0) + +/* Set the file sizes when finished. */ +#define XFS_EXCHMAPS_SET_SIZES (1ULL << 1) + +/* + * Exchange the mappings of the two files only if the file allocation units + * mapped to file1's range have been written. + */ +#define XFS_EXCHMAPS_INO1_WRITTEN (1ULL << 2) + +/* Clear the reflink flag from inode1 after the operation. */ +#define XFS_EXCHMAPS_CLEAR_INO1_REFLINK (1ULL << 3) + +/* Clear the reflink flag from inode2 after the operation. */ +#define XFS_EXCHMAPS_CLEAR_INO2_REFLINK (1ULL << 4) + +#define XFS_EXCHMAPS_LOGGED_FLAGS (XFS_EXCHMAPS_ATTR_FORK | \ + XFS_EXCHMAPS_SET_SIZES | \ + XFS_EXCHMAPS_INO1_WRITTEN | \ + XFS_EXCHMAPS_CLEAR_INO1_REFLINK | \ + XFS_EXCHMAPS_CLEAR_INO2_REFLINK) /* This is the structure used to lay out an mapping exchange done log item. */ struct xfs_xmd_log_format { diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h index 87b31c69a773..9640fc232c14 100644 --- a/fs/xfs/libxfs/xfs_trans_space.h +++ b/fs/xfs/libxfs/xfs_trans_space.h @@ -10,6 +10,10 @@ * Components of space reservations. */ +/* Worst case number of bmaps that can be held in a block. */ +#define XFS_MAX_CONTIG_BMAPS_PER_BLOCK(mp) \ + (((mp)->m_bmap_dmxr[0]) - ((mp)->m_bmap_dmnr[0])) + /* Worst case number of rmaps that can be held in a block. */ #define XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp) \ (((mp)->m_rmap_mxr[0]) - ((mp)->m_rmap_mnr[0])) diff --git a/fs/xfs/xfs_exchmaps_item.c b/fs/xfs/xfs_exchmaps_item.c index 65b0ade41b3d..a40216f33214 100644 --- a/fs/xfs/xfs_exchmaps_item.c +++ b/fs/xfs/xfs_exchmaps_item.c @@ -16,13 +16,17 @@ #include "xfs_trans.h" #include "xfs_trans_priv.h" #include "xfs_exchmaps_item.h" +#include "xfs_exchmaps.h" #include "xfs_log.h" #include "xfs_bmap.h" #include "xfs_icache.h" +#include "xfs_bmap_btree.h" #include "xfs_trans_space.h" #include "xfs_error.h" #include "xfs_log_priv.h" #include "xfs_log_recover.h" +#include "xfs_exchrange.h" +#include "xfs_trace.h" struct kmem_cache *xfs_xmi_cache; struct kmem_cache *xfs_xmd_cache; @@ -144,6 +148,365 @@ static inline struct xfs_xmd_log_item *XMD_ITEM(struct xfs_log_item *lip) return container_of(lip, struct xfs_xmd_log_item, xmd_item); } +STATIC void +xfs_xmd_item_size( + struct xfs_log_item *lip, + int *nvecs, + int *nbytes) +{ + *nvecs += 1; + *nbytes += sizeof(struct xfs_xmd_log_format); +} + +/* + * This is called to fill in the vector of log iovecs for the given xmd log + * item. We use only 1 iovec, and we point that at the xmd_log_format structure + * embedded in the xmd item. + */ +STATIC void +xfs_xmd_item_format( + struct xfs_log_item *lip, + struct xfs_log_vec *lv) +{ + struct xfs_xmd_log_item *xmd_lip = XMD_ITEM(lip); + struct xfs_log_iovec *vecp = NULL; + + xmd_lip->xmd_format.xmd_type = XFS_LI_XMD; + xmd_lip->xmd_format.xmd_size = 1; + + xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_XMD_FORMAT, &xmd_lip->xmd_format, + sizeof(struct xfs_xmd_log_format)); +} + +/* + * The XMD is either committed or aborted if the transaction is cancelled. If + * the transaction is cancelled, drop our reference to the XMI and free the + * XMD. + */ +STATIC void +xfs_xmd_item_release( + struct xfs_log_item *lip) +{ + struct xfs_xmd_log_item *xmd_lip = XMD_ITEM(lip); + + xfs_xmi_release(xmd_lip->xmd_intent_log_item); + kvfree(xmd_lip->xmd_item.li_lv_shadow); + kmem_cache_free(xfs_xmd_cache, xmd_lip); +} + +static struct xfs_log_item * +xfs_xmd_item_intent( + struct xfs_log_item *lip) +{ + return &XMD_ITEM(lip)->xmd_intent_log_item->xmi_item; +} + +static const struct xfs_item_ops xfs_xmd_item_ops = { + .flags = XFS_ITEM_RELEASE_WHEN_COMMITTED | + XFS_ITEM_INTENT_DONE, + .iop_size = xfs_xmd_item_size, + .iop_format = xfs_xmd_item_format, + .iop_release = xfs_xmd_item_release, + .iop_intent = xfs_xmd_item_intent, +}; + +/* Log file mapping exchange information in the intent item. */ +STATIC struct xfs_log_item * +xfs_exchmaps_create_intent( + struct xfs_trans *tp, + struct list_head *items, + unsigned int count, + bool sort) +{ + struct xfs_xmi_log_item *xmi_lip; + struct xfs_exchmaps_intent *xmi; + struct xfs_xmi_log_format *xlf; + + ASSERT(count == 1); + + xmi = list_first_entry_or_null(items, struct xfs_exchmaps_intent, + xmi_list); + + xmi_lip = xfs_xmi_init(tp->t_mountp); + xlf = &xmi_lip->xmi_format; + + xlf->xmi_inode1 = xmi->xmi_ip1->i_ino; + xlf->xmi_inode2 = xmi->xmi_ip2->i_ino; + xlf->xmi_startoff1 = xmi->xmi_startoff1; + xlf->xmi_startoff2 = xmi->xmi_startoff2; + xlf->xmi_blockcount = xmi->xmi_blockcount; + xlf->xmi_isize1 = xmi->xmi_isize1; + xlf->xmi_isize2 = xmi->xmi_isize2; + xlf->xmi_flags = xmi->xmi_flags & XFS_EXCHMAPS_LOGGED_FLAGS; + + return &xmi_lip->xmi_item; +} + +STATIC struct xfs_log_item * +xfs_exchmaps_create_done( + struct xfs_trans *tp, + struct xfs_log_item *intent, + unsigned int count) +{ + struct xfs_xmi_log_item *xmi_lip = XMI_ITEM(intent); + struct xfs_xmd_log_item *xmd_lip; + + xmd_lip = kmem_cache_zalloc(xfs_xmd_cache, GFP_KERNEL | __GFP_NOFAIL); + xfs_log_item_init(tp->t_mountp, &xmd_lip->xmd_item, XFS_LI_XMD, + &xfs_xmd_item_ops); + xmd_lip->xmd_intent_log_item = xmi_lip; + xmd_lip->xmd_format.xmd_xmi_id = xmi_lip->xmi_format.xmi_id; + + return &xmd_lip->xmd_item; +} + +/* Add this deferred XMI to the transaction. */ +void +xfs_exchmaps_defer_add( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi) +{ + trace_xfs_exchmaps_defer(tp->t_mountp, xmi); + + xfs_defer_add(tp, &xmi->xmi_list, &xfs_exchmaps_defer_type); +} + +static inline struct xfs_exchmaps_intent *xmi_entry(const struct list_head *e) +{ + return list_entry(e, struct xfs_exchmaps_intent, xmi_list); +} + +/* Cancel a deferred file mapping exchange. */ +STATIC void +xfs_exchmaps_cancel_item( + struct list_head *item) +{ + struct xfs_exchmaps_intent *xmi = xmi_entry(item); + + kmem_cache_free(xfs_exchmaps_intent_cache, xmi); +} + +/* Process a deferred file mapping exchange. */ +STATIC int +xfs_exchmaps_finish_item( + struct xfs_trans *tp, + struct xfs_log_item *done, + struct list_head *item, + struct xfs_btree_cur **state) +{ + struct xfs_exchmaps_intent *xmi = xmi_entry(item); + int error; + + /* + * Exchange one more mappings between two files. If there's still more + * work to do, we want to requeue ourselves after all other pending + * deferred operations have finished. This includes all of the dfops + * that we queued directly as well as any new ones created in the + * process of finishing the others. Doing so prevents us from queuing + * a large number of XMI log items in kernel memory, which in turn + * prevents us from pinning the tail of the log (while logging those + * new XMI items) until the first XMI items can be processed. + */ + error = xfs_exchmaps_finish_one(tp, xmi); + if (error != -EAGAIN) + xfs_exchmaps_cancel_item(item); + return error; +} + +/* Abort all pending XMIs. */ +STATIC void +xfs_exchmaps_abort_intent( + struct xfs_log_item *intent) +{ + xfs_xmi_release(XMI_ITEM(intent)); +} + +/* Is this recovered XMI ok? */ +static inline bool +xfs_xmi_validate( + struct xfs_mount *mp, + struct xfs_xmi_log_item *xmi_lip) +{ + struct xfs_xmi_log_format *xlf = &xmi_lip->xmi_format; + + if (!xfs_has_exchange_range(mp)) + return false; + + if (xmi_lip->xmi_format.__pad != 0) + return false; + + if (xlf->xmi_flags & ~XFS_EXCHMAPS_LOGGED_FLAGS) + return false; + + if (!xfs_verify_ino(mp, xlf->xmi_inode1) || + !xfs_verify_ino(mp, xlf->xmi_inode2)) + return false; + + if (!xfs_verify_fileext(mp, xlf->xmi_startoff1, xlf->xmi_blockcount)) + return false; + + return xfs_verify_fileext(mp, xlf->xmi_startoff2, xlf->xmi_blockcount); +} + +/* + * Use the recovered log state to create a new request, estimate resource + * requirements, and create a new incore intent state. + */ +STATIC struct xfs_exchmaps_intent * +xfs_xmi_item_recover_intent( + struct xfs_mount *mp, + struct xfs_defer_pending *dfp, + const struct xfs_xmi_log_format *xlf, + struct xfs_exchmaps_req *req, + struct xfs_inode **ipp1, + struct xfs_inode **ipp2) +{ + struct xfs_inode *ip1, *ip2; + struct xfs_exchmaps_intent *xmi; + int error; + + /* + * Grab both inodes and set IRECOVERY to prevent trimming of post-eof + * mappings and freeing of unlinked inodes until we're totally done + * processing files. + */ + error = xlog_recover_iget(mp, xlf->xmi_inode1, &ip1); + if (error) + return ERR_PTR(error); + error = xlog_recover_iget(mp, xlf->xmi_inode2, &ip2); + if (error) + goto err_rele1; + + req->ip1 = ip1; + req->ip2 = ip2; + req->startoff1 = xlf->xmi_startoff1; + req->startoff2 = xlf->xmi_startoff2; + req->blockcount = xlf->xmi_blockcount; + req->flags = xlf->xmi_flags & XFS_EXCHMAPS_PARAMS; + + xfs_exchrange_ilock(NULL, ip1, ip2); + error = xfs_exchmaps_estimate(req); + xfs_exchrange_iunlock(ip1, ip2); + if (error) + goto err_rele2; + + *ipp1 = ip1; + *ipp2 = ip2; + xmi = xfs_exchmaps_init_intent(req); + xfs_defer_add_item(dfp, &xmi->xmi_list); + return xmi; + +err_rele2: + xfs_irele(ip2); +err_rele1: + xfs_irele(ip1); + req->ip2 = req->ip1 = NULL; + return ERR_PTR(error); +} + +/* Process a file mapping exchange item that was recovered from the log. */ +STATIC int +xfs_exchmaps_recover_work( + struct xfs_defer_pending *dfp, + struct list_head *capture_list) +{ + struct xfs_exchmaps_req req = { .flags = 0 }; + struct xfs_trans_res resv; + struct xfs_exchmaps_intent *xmi; + struct xfs_log_item *lip = dfp->dfp_intent; + struct xfs_xmi_log_item *xmi_lip = XMI_ITEM(lip); + struct xfs_mount *mp = lip->li_log->l_mp; + struct xfs_trans *tp; + struct xfs_inode *ip1, *ip2; + int error = 0; + + if (!xfs_xmi_validate(mp, xmi_lip)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + &xmi_lip->xmi_format, + sizeof(xmi_lip->xmi_format)); + return -EFSCORRUPTED; + } + + xmi = xfs_xmi_item_recover_intent(mp, dfp, &xmi_lip->xmi_format, &req, + &ip1, &ip2); + if (IS_ERR(xmi)) + return PTR_ERR(xmi); + + trace_xfs_exchmaps_recover(mp, xmi); + + resv = xlog_recover_resv(&M_RES(mp)->tr_write); + error = xfs_trans_alloc(mp, &resv, req.resblks, 0, 0, &tp); + if (error) + goto err_rele; + + xfs_exchrange_ilock(tp, ip1, ip2); + + xfs_exchmaps_ensure_reflink(tp, xmi); + xfs_exchmaps_upgrade_extent_counts(tp, xmi); + error = xlog_recover_finish_intent(tp, dfp); + if (error == -EFSCORRUPTED) + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + &xmi_lip->xmi_format, + sizeof(xmi_lip->xmi_format)); + if (error) + goto err_cancel; + + /* + * Commit transaction, which frees the transaction and saves the inodes + * for later replay activities. + */ + error = xfs_defer_ops_capture_and_commit(tp, capture_list); + goto err_unlock; + +err_cancel: + xfs_trans_cancel(tp); +err_unlock: + xfs_exchrange_iunlock(ip1, ip2); +err_rele: + xfs_irele(ip2); + xfs_irele(ip1); + return error; +} + +/* Relog an intent item to push the log tail forward. */ +static struct xfs_log_item * +xfs_exchmaps_relog_intent( + struct xfs_trans *tp, + struct xfs_log_item *intent, + struct xfs_log_item *done_item) +{ + struct xfs_xmi_log_item *xmi_lip; + struct xfs_xmi_log_format *old_xlf, *new_xlf; + + old_xlf = &XMI_ITEM(intent)->xmi_format; + + xmi_lip = xfs_xmi_init(tp->t_mountp); + new_xlf = &xmi_lip->xmi_format; + + new_xlf->xmi_inode1 = old_xlf->xmi_inode1; + new_xlf->xmi_inode2 = old_xlf->xmi_inode2; + new_xlf->xmi_startoff1 = old_xlf->xmi_startoff1; + new_xlf->xmi_startoff2 = old_xlf->xmi_startoff2; + new_xlf->xmi_blockcount = old_xlf->xmi_blockcount; + new_xlf->xmi_flags = old_xlf->xmi_flags; + new_xlf->xmi_isize1 = old_xlf->xmi_isize1; + new_xlf->xmi_isize2 = old_xlf->xmi_isize2; + + return &xmi_lip->xmi_item; +} + +const struct xfs_defer_op_type xfs_exchmaps_defer_type = { + .name = "exchmaps", + .max_items = 1, + .create_intent = xfs_exchmaps_create_intent, + .abort_intent = xfs_exchmaps_abort_intent, + .create_done = xfs_exchmaps_create_done, + .finish_item = xfs_exchmaps_finish_item, + .cancel_item = xfs_exchmaps_cancel_item, + .recover_work = xfs_exchmaps_recover_work, + .relog_intent = xfs_exchmaps_relog_intent, +}; + STATIC bool xfs_xmi_item_match( struct xfs_log_item *lip, @@ -194,8 +557,9 @@ xlog_recover_xmi_commit_pass2( xmi_lip = xfs_xmi_init(mp); memcpy(&xmi_lip->xmi_format, xmi_formatp, len); - /* not implemented yet */ - return -EIO; + xlog_recover_intent_item(log, &xmi_lip->xmi_item, lsn, + &xfs_exchmaps_defer_type); + return 0; } const struct xlog_recover_item_ops xlog_xmi_item_ops = { diff --git a/fs/xfs/xfs_exchmaps_item.h b/fs/xfs/xfs_exchmaps_item.h index ada1eb314e65..efa368d25d09 100644 --- a/fs/xfs/xfs_exchmaps_item.h +++ b/fs/xfs/xfs_exchmaps_item.h @@ -56,4 +56,9 @@ struct xfs_xmd_log_item { extern struct kmem_cache *xfs_xmi_cache; extern struct kmem_cache *xfs_xmd_cache; +struct xfs_exchmaps_intent; + +void xfs_exchmaps_defer_add(struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi); + #endif /* __XFS_EXCHMAPS_ITEM_H__ */ diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c index 4cd824e47f75..35351b973521 100644 --- a/fs/xfs/xfs_exchrange.c +++ b/fs/xfs/xfs_exchrange.c @@ -13,8 +13,57 @@ #include "xfs_inode.h" #include "xfs_trans.h" #include "xfs_exchrange.h" +#include "xfs_exchmaps.h" #include +/* Lock (and optionally join) two inodes for a file range exchange. */ +void +xfs_exchrange_ilock( + struct xfs_trans *tp, + struct xfs_inode *ip1, + struct xfs_inode *ip2) +{ + if (ip1 != ip2) + xfs_lock_two_inodes(ip1, XFS_ILOCK_EXCL, + ip2, XFS_ILOCK_EXCL); + else + xfs_ilock(ip1, XFS_ILOCK_EXCL); + if (tp) { + xfs_trans_ijoin(tp, ip1, 0); + if (ip2 != ip1) + xfs_trans_ijoin(tp, ip2, 0); + } + +} + +/* Unlock two inodes after a file range exchange operation. */ +void +xfs_exchrange_iunlock( + struct xfs_inode *ip1, + struct xfs_inode *ip2) +{ + if (ip2 != ip1) + xfs_iunlock(ip2, XFS_ILOCK_EXCL); + xfs_iunlock(ip1, XFS_ILOCK_EXCL); +} + +/* + * Estimate the resource requirements to exchange file contents between the two + * files. The caller is required to hold the IOLOCK and the MMAPLOCK and to + * have flushed both inodes' pagecache and active direct-ios. + */ +int +xfs_exchrange_estimate( + struct xfs_exchmaps_req *req) +{ + int error; + + xfs_exchrange_ilock(NULL, req->ip1, req->ip2); + error = xfs_exchmaps_estimate(req); + xfs_exchrange_iunlock(req->ip1, req->ip2); + return error; +} + /* * Generic code for exchanging ranges of two files via XFS_IOC_EXCHANGE_RANGE. * This part deals with struct file objects and byte ranges and does not deal diff --git a/fs/xfs/xfs_exchrange.h b/fs/xfs/xfs_exchrange.h index f80369c7df5d..039abcca546e 100644 --- a/fs/xfs/xfs_exchrange.h +++ b/fs/xfs/xfs_exchrange.h @@ -27,4 +27,12 @@ struct xfs_exchrange { long xfs_ioc_exchange_range(struct file *file, struct xfs_exchange_range __user *argp); +struct xfs_exchmaps_req; + +void xfs_exchrange_ilock(struct xfs_trans *tp, struct xfs_inode *ip1, + struct xfs_inode *ip2); +void xfs_exchrange_iunlock(struct xfs_inode *ip1, struct xfs_inode *ip2); + +int xfs_exchrange_estimate(struct xfs_exchmaps_req *req); + #endif /* __XFS_EXCHRANGE_H__ */ diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c index 1a963382e5e9..9f38e69f1ce4 100644 --- a/fs/xfs/xfs_trace.c +++ b/fs/xfs/xfs_trace.c @@ -39,6 +39,7 @@ #include "xfs_buf_mem.h" #include "xfs_btree_mem.h" #include "xfs_bmap.h" +#include "xfs_exchmaps.h" /* * We include this last to have the helpers above available for the trace diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index aea97fc074f8..7c17d1f80fec 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -82,6 +82,8 @@ struct xfs_perag; struct xfbtree; struct xfs_btree_ops; struct xfs_bmap_intent; +struct xfs_exchmaps_intent; +struct xfs_exchmaps_req; #define XFS_ATTR_FILTER_FLAGS \ { XFS_ATTR_ROOT, "ROOT" }, \ @@ -4770,6 +4772,221 @@ DEFINE_XFBTREE_FREESP_EVENT(xfbtree_alloc_block); DEFINE_XFBTREE_FREESP_EVENT(xfbtree_free_block); #endif /* CONFIG_XFS_BTREE_IN_MEM */ +/* exchmaps tracepoints */ +#define XFS_EXCHMAPS_STRINGS \ + { XFS_EXCHMAPS_ATTR_FORK, "ATTRFORK" }, \ + { XFS_EXCHMAPS_SET_SIZES, "SETSIZES" }, \ + { XFS_EXCHMAPS_INO1_WRITTEN, "INO1_WRITTEN" }, \ + { XFS_EXCHMAPS_CLEAR_INO1_REFLINK, "CLEAR_INO1_REFLINK" }, \ + { XFS_EXCHMAPS_CLEAR_INO2_REFLINK, "CLEAR_INO2_REFLINK" } + +DEFINE_INODE_IREC_EVENT(xfs_exchmaps_mapping1_skip); +DEFINE_INODE_IREC_EVENT(xfs_exchmaps_mapping1); +DEFINE_INODE_IREC_EVENT(xfs_exchmaps_mapping2); +DEFINE_ITRUNC_EVENT(xfs_exchmaps_update_inode_size); + +TRACE_EVENT(xfs_exchmaps_overhead, + TP_PROTO(struct xfs_mount *mp, unsigned long long bmbt_blocks, + unsigned long long rmapbt_blocks), + TP_ARGS(mp, bmbt_blocks, rmapbt_blocks), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned long long, bmbt_blocks) + __field(unsigned long long, rmapbt_blocks) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->bmbt_blocks = bmbt_blocks; + __entry->rmapbt_blocks = rmapbt_blocks; + ), + TP_printk("dev %d:%d bmbt_blocks 0x%llx rmapbt_blocks 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->bmbt_blocks, + __entry->rmapbt_blocks) +); + +DECLARE_EVENT_CLASS(xfs_exchmaps_estimate_class, + TP_PROTO(const struct xfs_exchmaps_req *req), + TP_ARGS(req), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino1) + __field(xfs_ino_t, ino2) + __field(xfs_fileoff_t, startoff1) + __field(xfs_fileoff_t, startoff2) + __field(xfs_filblks_t, blockcount) + __field(uint64_t, flags) + __field(xfs_filblks_t, ip1_bcount) + __field(xfs_filblks_t, ip2_bcount) + __field(xfs_filblks_t, ip1_rtbcount) + __field(xfs_filblks_t, ip2_rtbcount) + __field(unsigned long long, resblks) + __field(unsigned long long, nr_exchanges) + ), + TP_fast_assign( + __entry->dev = req->ip1->i_mount->m_super->s_dev; + __entry->ino1 = req->ip1->i_ino; + __entry->ino2 = req->ip2->i_ino; + __entry->startoff1 = req->startoff1; + __entry->startoff2 = req->startoff2; + __entry->blockcount = req->blockcount; + __entry->flags = req->flags; + __entry->ip1_bcount = req->ip1_bcount; + __entry->ip2_bcount = req->ip2_bcount; + __entry->ip1_rtbcount = req->ip1_rtbcount; + __entry->ip2_rtbcount = req->ip2_rtbcount; + __entry->resblks = req->resblks; + __entry->nr_exchanges = req->nr_exchanges; + ), + TP_printk("dev %d:%d ino1 0x%llx fileoff1 0x%llx ino2 0x%llx fileoff2 0x%llx fsbcount 0x%llx flags (%s) bcount1 0x%llx rtbcount1 0x%llx bcount2 0x%llx rtbcount2 0x%llx resblks 0x%llx nr_exchanges %llu", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino1, __entry->startoff1, + __entry->ino2, __entry->startoff2, + __entry->blockcount, + __print_flags_u64(__entry->flags, "|", XFS_EXCHMAPS_STRINGS), + __entry->ip1_bcount, + __entry->ip1_rtbcount, + __entry->ip2_bcount, + __entry->ip2_rtbcount, + __entry->resblks, + __entry->nr_exchanges) +); + +#define DEFINE_EXCHMAPS_ESTIMATE_EVENT(name) \ +DEFINE_EVENT(xfs_exchmaps_estimate_class, name, \ + TP_PROTO(const struct xfs_exchmaps_req *req), \ + TP_ARGS(req)) +DEFINE_EXCHMAPS_ESTIMATE_EVENT(xfs_exchmaps_initial_estimate); +DEFINE_EXCHMAPS_ESTIMATE_EVENT(xfs_exchmaps_final_estimate); + +DECLARE_EVENT_CLASS(xfs_exchmaps_intent_class, + TP_PROTO(struct xfs_mount *mp, const struct xfs_exchmaps_intent *xmi), + TP_ARGS(mp, xmi), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino1) + __field(xfs_ino_t, ino2) + __field(uint64_t, flags) + __field(xfs_fileoff_t, startoff1) + __field(xfs_fileoff_t, startoff2) + __field(xfs_filblks_t, blockcount) + __field(xfs_fsize_t, isize1) + __field(xfs_fsize_t, isize2) + __field(xfs_fsize_t, new_isize1) + __field(xfs_fsize_t, new_isize2) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->ino1 = xmi->xmi_ip1->i_ino; + __entry->ino2 = xmi->xmi_ip2->i_ino; + __entry->flags = xmi->xmi_flags; + __entry->startoff1 = xmi->xmi_startoff1; + __entry->startoff2 = xmi->xmi_startoff2; + __entry->blockcount = xmi->xmi_blockcount; + __entry->isize1 = xmi->xmi_ip1->i_disk_size; + __entry->isize2 = xmi->xmi_ip2->i_disk_size; + __entry->new_isize1 = xmi->xmi_isize1; + __entry->new_isize2 = xmi->xmi_isize2; + ), + TP_printk("dev %d:%d ino1 0x%llx fileoff1 0x%llx ino2 0x%llx fileoff2 0x%llx fsbcount 0x%llx flags (%s) isize1 0x%llx newisize1 0x%llx isize2 0x%llx newisize2 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino1, __entry->startoff1, + __entry->ino2, __entry->startoff2, + __entry->blockcount, + __print_flags_u64(__entry->flags, "|", XFS_EXCHMAPS_STRINGS), + __entry->isize1, __entry->new_isize1, + __entry->isize2, __entry->new_isize2) +); + +#define DEFINE_EXCHMAPS_INTENT_EVENT(name) \ +DEFINE_EVENT(xfs_exchmaps_intent_class, name, \ + TP_PROTO(struct xfs_mount *mp, const struct xfs_exchmaps_intent *xmi), \ + TP_ARGS(mp, xmi)) +DEFINE_EXCHMAPS_INTENT_EVENT(xfs_exchmaps_defer); +DEFINE_EXCHMAPS_INTENT_EVENT(xfs_exchmaps_recover); + +TRACE_EVENT(xfs_exchmaps_delta_nextents_step, + TP_PROTO(struct xfs_mount *mp, + const struct xfs_bmbt_irec *left, + const struct xfs_bmbt_irec *curr, + const struct xfs_bmbt_irec *new, + const struct xfs_bmbt_irec *right, + int delta, unsigned int state), + TP_ARGS(mp, left, curr, new, right, delta, state), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_fileoff_t, loff) + __field(xfs_fsblock_t, lstart) + __field(xfs_filblks_t, lcount) + __field(xfs_fileoff_t, coff) + __field(xfs_fsblock_t, cstart) + __field(xfs_filblks_t, ccount) + __field(xfs_fileoff_t, noff) + __field(xfs_fsblock_t, nstart) + __field(xfs_filblks_t, ncount) + __field(xfs_fileoff_t, roff) + __field(xfs_fsblock_t, rstart) + __field(xfs_filblks_t, rcount) + __field(int, delta) + __field(unsigned int, state) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->loff = left->br_startoff; + __entry->lstart = left->br_startblock; + __entry->lcount = left->br_blockcount; + __entry->coff = curr->br_startoff; + __entry->cstart = curr->br_startblock; + __entry->ccount = curr->br_blockcount; + __entry->noff = new->br_startoff; + __entry->nstart = new->br_startblock; + __entry->ncount = new->br_blockcount; + __entry->roff = right->br_startoff; + __entry->rstart = right->br_startblock; + __entry->rcount = right->br_blockcount; + __entry->delta = delta; + __entry->state = state; + ), + TP_printk("dev %d:%d left 0x%llx:0x%llx:0x%llx; curr 0x%llx:0x%llx:0x%llx <- new 0x%llx:0x%llx:0x%llx; right 0x%llx:0x%llx:0x%llx delta %d state 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->loff, __entry->lstart, __entry->lcount, + __entry->coff, __entry->cstart, __entry->ccount, + __entry->noff, __entry->nstart, __entry->ncount, + __entry->roff, __entry->rstart, __entry->rcount, + __entry->delta, __entry->state) +); + +TRACE_EVENT(xfs_exchmaps_delta_nextents, + TP_PROTO(const struct xfs_exchmaps_req *req, int64_t d_nexts1, + int64_t d_nexts2), + TP_ARGS(req, d_nexts1, d_nexts2), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino1) + __field(xfs_ino_t, ino2) + __field(xfs_extnum_t, nexts1) + __field(xfs_extnum_t, nexts2) + __field(int64_t, d_nexts1) + __field(int64_t, d_nexts2) + ), + TP_fast_assign( + int whichfork = xfs_exchmaps_reqfork(req); + + __entry->dev = req->ip1->i_mount->m_super->s_dev; + __entry->ino1 = req->ip1->i_ino; + __entry->ino2 = req->ip2->i_ino; + __entry->nexts1 = xfs_ifork_ptr(req->ip1, whichfork)->if_nextents; + __entry->nexts2 = xfs_ifork_ptr(req->ip2, whichfork)->if_nextents; + __entry->d_nexts1 = d_nexts1; + __entry->d_nexts2 = d_nexts2; + ), + TP_printk("dev %d:%d ino1 0x%llx nexts %llu ino2 0x%llx nexts %llu delta1 %lld delta2 %lld", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino1, __entry->nexts1, + __entry->ino2, __entry->nexts2, + __entry->d_nexts1, __entry->d_nexts2) +); + #endif /* _TRACE_XFS_H */ #undef TRACE_INCLUDE_PATH -- cgit v1.2.3-59-g8ed1b From 42672471f938cdab2573f32ce23915b78f0578f4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:18 -0700 Subject: xfs: bind together the front and back ends of the file range exchange code So far, we've constructed the front end of the file range exchange code that does all the checking; and the back end of the file mapping exchange code that actually does the work. Glue these two pieces together so that we can turn on the functionality. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_exchrange.c | 334 ++++++++++++++++++++++++++++++++++++++++++++++++- fs/xfs/xfs_trace.c | 1 + fs/xfs/xfs_trace.h | 109 ++++++++++++++++ 3 files changed, 443 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c index 35351b973521..0fc95e6471cb 100644 --- a/fs/xfs/xfs_exchrange.c +++ b/fs/xfs/xfs_exchrange.c @@ -12,8 +12,15 @@ #include "xfs_defer.h" #include "xfs_inode.h" #include "xfs_trans.h" +#include "xfs_quota.h" +#include "xfs_bmap_util.h" +#include "xfs_reflink.h" +#include "xfs_trace.h" #include "xfs_exchrange.h" #include "xfs_exchmaps.h" +#include "xfs_sb.h" +#include "xfs_icache.h" +#include "xfs_log.h" #include /* Lock (and optionally join) two inodes for a file range exchange. */ @@ -64,6 +71,207 @@ xfs_exchrange_estimate( return error; } +#define QRETRY_IP1 (0x1) +#define QRETRY_IP2 (0x2) + +/* + * Obtain a quota reservation to make sure we don't hit EDQUOT. We can skip + * this if quota enforcement is disabled or if both inodes' dquots are the + * same. The qretry structure must be initialized to zeroes before the first + * call to this function. + */ +STATIC int +xfs_exchrange_reserve_quota( + struct xfs_trans *tp, + const struct xfs_exchmaps_req *req, + unsigned int *qretry) +{ + int64_t ddelta, rdelta; + int ip1_error = 0; + int error; + + /* + * Don't bother with a quota reservation if we're not enforcing them + * or the two inodes have the same dquots. + */ + if (!XFS_IS_QUOTA_ON(tp->t_mountp) || req->ip1 == req->ip2 || + (req->ip1->i_udquot == req->ip2->i_udquot && + req->ip1->i_gdquot == req->ip2->i_gdquot && + req->ip1->i_pdquot == req->ip2->i_pdquot)) + return 0; + + *qretry = 0; + + /* + * For each file, compute the net gain in the number of regular blocks + * that will be mapped into that file and reserve that much quota. The + * quota counts must be able to absorb at least that much space. + */ + ddelta = req->ip2_bcount - req->ip1_bcount; + rdelta = req->ip2_rtbcount - req->ip1_rtbcount; + if (ddelta > 0 || rdelta > 0) { + error = xfs_trans_reserve_quota_nblks(tp, req->ip1, + ddelta > 0 ? ddelta : 0, + rdelta > 0 ? rdelta : 0, + false); + if (error == -EDQUOT || error == -ENOSPC) { + /* + * Save this error and see what happens if we try to + * reserve quota for ip2. Then report both. + */ + *qretry |= QRETRY_IP1; + ip1_error = error; + error = 0; + } + if (error) + return error; + } + if (ddelta < 0 || rdelta < 0) { + error = xfs_trans_reserve_quota_nblks(tp, req->ip2, + ddelta < 0 ? -ddelta : 0, + rdelta < 0 ? -rdelta : 0, + false); + if (error == -EDQUOT || error == -ENOSPC) + *qretry |= QRETRY_IP2; + if (error) + return error; + } + if (ip1_error) + return ip1_error; + + /* + * For each file, forcibly reserve the gross gain in mapped blocks so + * that we don't trip over any quota block reservation assertions. + * We must reserve the gross gain because the quota code subtracts from + * bcount the number of blocks that we unmap; it does not add that + * quantity back to the quota block reservation. + */ + error = xfs_trans_reserve_quota_nblks(tp, req->ip1, req->ip1_bcount, + req->ip1_rtbcount, true); + if (error) + return error; + + return xfs_trans_reserve_quota_nblks(tp, req->ip2, req->ip2_bcount, + req->ip2_rtbcount, true); +} + +/* Exchange the mappings (and hence the contents) of two files' forks. */ +STATIC int +xfs_exchrange_mappings( + const struct xfs_exchrange *fxr, + struct xfs_inode *ip1, + struct xfs_inode *ip2) +{ + struct xfs_mount *mp = ip1->i_mount; + struct xfs_exchmaps_req req = { + .ip1 = ip1, + .ip2 = ip2, + .startoff1 = XFS_B_TO_FSBT(mp, fxr->file1_offset), + .startoff2 = XFS_B_TO_FSBT(mp, fxr->file2_offset), + .blockcount = XFS_B_TO_FSB(mp, fxr->length), + }; + struct xfs_trans *tp; + unsigned int qretry; + bool retried = false; + int error; + + trace_xfs_exchrange_mappings(fxr, ip1, ip2); + + if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) + req.flags |= XFS_EXCHMAPS_SET_SIZES; + if (fxr->flags & XFS_EXCHANGE_RANGE_FILE1_WRITTEN) + req.flags |= XFS_EXCHMAPS_INO1_WRITTEN; + + error = xfs_exchrange_estimate(&req); + if (error) + return error; + +retry: + /* Allocate the transaction, lock the inodes, and join them. */ + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, req.resblks, 0, + XFS_TRANS_RES_FDBLKS, &tp); + if (error) + return error; + + xfs_exchrange_ilock(tp, ip1, ip2); + + trace_xfs_exchrange_before(ip2, 2); + trace_xfs_exchrange_before(ip1, 1); + + error = xfs_exchmaps_check_forks(mp, &req); + if (error) + goto out_trans_cancel; + + /* + * Reserve ourselves some quota if any of them are in enforcing mode. + * In theory we only need enough to satisfy the change in the number + * of blocks between the two ranges being remapped. + */ + error = xfs_exchrange_reserve_quota(tp, &req, &qretry); + if ((error == -EDQUOT || error == -ENOSPC) && !retried) { + xfs_trans_cancel(tp); + xfs_exchrange_iunlock(ip1, ip2); + if (qretry & QRETRY_IP1) + xfs_blockgc_free_quota(ip1, 0); + if (qretry & QRETRY_IP2) + xfs_blockgc_free_quota(ip2, 0); + retried = true; + goto retry; + } + if (error) + goto out_trans_cancel; + + /* If we got this far on a dry run, all parameters are ok. */ + if (fxr->flags & XFS_EXCHANGE_RANGE_DRY_RUN) + goto out_trans_cancel; + + /* Update the mtime and ctime of both files. */ + if (fxr->flags & __XFS_EXCHANGE_RANGE_UPD_CMTIME1) + xfs_trans_ichgtime(tp, ip1, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + if (fxr->flags & __XFS_EXCHANGE_RANGE_UPD_CMTIME2) + xfs_trans_ichgtime(tp, ip2, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + + xfs_exchange_mappings(tp, &req); + + /* + * Force the log to persist metadata updates if the caller or the + * administrator requires this. The generic prep function already + * flushed the relevant parts of the page cache. + */ + if (xfs_has_wsync(mp) || (fxr->flags & XFS_EXCHANGE_RANGE_DSYNC)) + xfs_trans_set_sync(tp); + + error = xfs_trans_commit(tp); + + trace_xfs_exchrange_after(ip2, 2); + trace_xfs_exchrange_after(ip1, 1); + + if (error) + goto out_unlock; + + /* + * If the caller wanted us to exchange the contents of two complete + * files of unequal length, exchange the incore sizes now. This should + * be safe because we flushed both files' page caches, exchanged all + * the mappings, and updated the ondisk sizes. + */ + if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) { + loff_t temp; + + temp = i_size_read(VFS_I(ip2)); + i_size_write(VFS_I(ip2), i_size_read(VFS_I(ip1))); + i_size_write(VFS_I(ip1), temp); + } + +out_unlock: + xfs_exchrange_iunlock(ip1, ip2); + return error; + +out_trans_cancel: + xfs_trans_cancel(tp); + goto out_unlock; +} + /* * Generic code for exchanging ranges of two files via XFS_IOC_EXCHANGE_RANGE. * This part deals with struct file objects and byte ranges and does not deal @@ -287,6 +495,130 @@ xfs_exchange_range_finish( return file_remove_privs(fxr->file2); } +/* Prepare two files to have their data exchanged. */ +STATIC int +xfs_exchrange_prep( + struct xfs_exchrange *fxr, + struct xfs_inode *ip1, + struct xfs_inode *ip2) +{ + unsigned int alloc_unit = xfs_inode_alloc_unitsize(ip2); + int error; + + trace_xfs_exchrange_prep(fxr, ip1, ip2); + + /* Verify both files are either real-time or non-realtime */ + if (XFS_IS_REALTIME_INODE(ip1) != XFS_IS_REALTIME_INODE(ip2)) + return -EINVAL; + + /* + * The alignment checks in the generic helpers cannot deal with + * allocation units that are not powers of 2. This can happen with the + * realtime volume if the extent size is set. + */ + if (!is_power_of_2(alloc_unit)) + return -EOPNOTSUPP; + + error = xfs_exchange_range_prep(fxr, alloc_unit); + if (error || fxr->length == 0) + return error; + + /* Attach dquots to both inodes before changing block maps. */ + error = xfs_qm_dqattach(ip2); + if (error) + return error; + error = xfs_qm_dqattach(ip1); + if (error) + return error; + + trace_xfs_exchrange_flush(fxr, ip1, ip2); + + /* Flush the relevant ranges of both files. */ + error = xfs_flush_unmap_range(ip2, fxr->file2_offset, fxr->length); + if (error) + return error; + error = xfs_flush_unmap_range(ip1, fxr->file1_offset, fxr->length); + if (error) + return error; + + /* + * Cancel CoW fork preallocations for the ranges of both files. The + * prep function should have flushed all the dirty data, so the only + * CoW mappings remaining should be speculative. + */ + if (xfs_inode_has_cow_data(ip1)) { + error = xfs_reflink_cancel_cow_range(ip1, fxr->file1_offset, + fxr->length, true); + if (error) + return error; + } + + if (xfs_inode_has_cow_data(ip2)) { + error = xfs_reflink_cancel_cow_range(ip2, fxr->file2_offset, + fxr->length, true); + if (error) + return error; + } + + return 0; +} + +/* + * Exchange contents of files. This is the binding between the generic + * file-level concepts and the XFS inode-specific implementation. + */ +STATIC int +xfs_exchrange_contents( + struct xfs_exchrange *fxr) +{ + struct inode *inode1 = file_inode(fxr->file1); + struct inode *inode2 = file_inode(fxr->file2); + struct xfs_inode *ip1 = XFS_I(inode1); + struct xfs_inode *ip2 = XFS_I(inode2); + struct xfs_mount *mp = ip1->i_mount; + int error; + + if (!xfs_has_exchange_range(mp)) + return -EOPNOTSUPP; + + if (fxr->flags & ~(XFS_EXCHANGE_RANGE_ALL_FLAGS | + XFS_EXCHANGE_RANGE_PRIV_FLAGS)) + return -EINVAL; + + if (xfs_is_shutdown(mp)) + return -EIO; + + /* Lock both files against IO */ + error = xfs_ilock2_io_mmap(ip1, ip2); + if (error) + goto out_err; + + /* Prepare and then exchange file contents. */ + error = xfs_exchrange_prep(fxr, ip1, ip2); + if (error) + goto out_unlock; + + error = xfs_exchrange_mappings(fxr, ip1, ip2); + if (error) + goto out_unlock; + + /* + * Finish the exchange by removing special file privileges like any + * other file write would do. This may involve turning on support for + * logged xattrs if either file has security capabilities. + */ + error = xfs_exchange_range_finish(fxr); + if (error) + goto out_unlock; + +out_unlock: + xfs_iunlock2_io_mmap(ip1, ip2); +out_err: + if (error) + trace_xfs_exchrange_error(ip2, error, _RET_IP_); + return error; +} + /* Exchange parts of two files. */ static int xfs_exchange_range( @@ -341,7 +673,7 @@ xfs_exchange_range( fxr->flags |= __XFS_EXCHANGE_RANGE_UPD_CMTIME2; file_start_write(fxr->file2); - ret = -EOPNOTSUPP; /* XXX call out to lower level code */ + ret = xfs_exchrange_contents(fxr); file_end_write(fxr->file2); if (ret) return ret; diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c index 9f38e69f1ce4..cf92a3bd56c7 100644 --- a/fs/xfs/xfs_trace.c +++ b/fs/xfs/xfs_trace.c @@ -40,6 +40,7 @@ #include "xfs_btree_mem.h" #include "xfs_bmap.h" #include "xfs_exchmaps.h" +#include "xfs_exchrange.h" /* * We include this last to have the helpers above available for the trace diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 7c17d1f80fec..729e728c2076 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -84,6 +84,7 @@ struct xfs_btree_ops; struct xfs_bmap_intent; struct xfs_exchmaps_intent; struct xfs_exchmaps_req; +struct xfs_exchrange; #define XFS_ATTR_FILTER_FLAGS \ { XFS_ATTR_ROOT, "ROOT" }, \ @@ -4785,6 +4786,114 @@ DEFINE_INODE_IREC_EVENT(xfs_exchmaps_mapping1); DEFINE_INODE_IREC_EVENT(xfs_exchmaps_mapping2); DEFINE_ITRUNC_EVENT(xfs_exchmaps_update_inode_size); +#define XFS_EXCHRANGE_INODES \ + { 1, "file1" }, \ + { 2, "file2" } + +DECLARE_EVENT_CLASS(xfs_exchrange_inode_class, + TP_PROTO(struct xfs_inode *ip, int whichfile), + TP_ARGS(ip, whichfile), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(int, whichfile) + __field(xfs_ino_t, ino) + __field(int, format) + __field(xfs_extnum_t, nex) + __field(int, broot_size) + __field(int, fork_off) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->whichfile = whichfile; + __entry->ino = ip->i_ino; + __entry->format = ip->i_df.if_format; + __entry->nex = ip->i_df.if_nextents; + __entry->fork_off = xfs_inode_fork_boff(ip); + ), + TP_printk("dev %d:%d ino 0x%llx whichfile %s format %s num_extents %llu forkoff 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_symbolic(__entry->whichfile, XFS_EXCHRANGE_INODES), + __print_symbolic(__entry->format, XFS_INODE_FORMAT_STR), + __entry->nex, + __entry->fork_off) +) + +#define DEFINE_EXCHRANGE_INODE_EVENT(name) \ +DEFINE_EVENT(xfs_exchrange_inode_class, name, \ + TP_PROTO(struct xfs_inode *ip, int whichfile), \ + TP_ARGS(ip, whichfile)) + +DEFINE_EXCHRANGE_INODE_EVENT(xfs_exchrange_before); +DEFINE_EXCHRANGE_INODE_EVENT(xfs_exchrange_after); +DEFINE_INODE_ERROR_EVENT(xfs_exchrange_error); + +#define XFS_EXCHANGE_RANGE_FLAGS_STRS \ + { XFS_EXCHANGE_RANGE_TO_EOF, "TO_EOF" }, \ + { XFS_EXCHANGE_RANGE_DSYNC , "DSYNC" }, \ + { XFS_EXCHANGE_RANGE_DRY_RUN, "DRY_RUN" }, \ + { XFS_EXCHANGE_RANGE_FILE1_WRITTEN, "F1_WRITTEN" }, \ + { __XFS_EXCHANGE_RANGE_UPD_CMTIME1, "CMTIME1" }, \ + { __XFS_EXCHANGE_RANGE_UPD_CMTIME2, "CMTIME2" } + +/* file exchange-range tracepoint class */ +DECLARE_EVENT_CLASS(xfs_exchrange_class, + TP_PROTO(const struct xfs_exchrange *fxr, struct xfs_inode *ip1, + struct xfs_inode *ip2), + TP_ARGS(fxr, ip1, ip2), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ip1_ino) + __field(loff_t, ip1_isize) + __field(loff_t, ip1_disize) + __field(xfs_ino_t, ip2_ino) + __field(loff_t, ip2_isize) + __field(loff_t, ip2_disize) + + __field(loff_t, file1_offset) + __field(loff_t, file2_offset) + __field(unsigned long long, length) + __field(unsigned long long, flags) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip1)->i_sb->s_dev; + __entry->ip1_ino = ip1->i_ino; + __entry->ip1_isize = VFS_I(ip1)->i_size; + __entry->ip1_disize = ip1->i_disk_size; + __entry->ip2_ino = ip2->i_ino; + __entry->ip2_isize = VFS_I(ip2)->i_size; + __entry->ip2_disize = ip2->i_disk_size; + + __entry->file1_offset = fxr->file1_offset; + __entry->file2_offset = fxr->file2_offset; + __entry->length = fxr->length; + __entry->flags = fxr->flags; + ), + TP_printk("dev %d:%d flags %s bytecount 0x%llx " + "ino1 0x%llx isize 0x%llx disize 0x%llx pos 0x%llx -> " + "ino2 0x%llx isize 0x%llx disize 0x%llx pos 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __print_flags_u64(__entry->flags, "|", XFS_EXCHANGE_RANGE_FLAGS_STRS), + __entry->length, + __entry->ip1_ino, + __entry->ip1_isize, + __entry->ip1_disize, + __entry->file1_offset, + __entry->ip2_ino, + __entry->ip2_isize, + __entry->ip2_disize, + __entry->file2_offset) +) + +#define DEFINE_EXCHRANGE_EVENT(name) \ +DEFINE_EVENT(xfs_exchrange_class, name, \ + TP_PROTO(const struct xfs_exchrange *fxr, struct xfs_inode *ip1, \ + struct xfs_inode *ip2), \ + TP_ARGS(fxr, ip1, ip2)) +DEFINE_EXCHRANGE_EVENT(xfs_exchrange_prep); +DEFINE_EXCHRANGE_EVENT(xfs_exchrange_flush); +DEFINE_EXCHRANGE_EVENT(xfs_exchrange_mappings); + TRACE_EVENT(xfs_exchmaps_overhead, TP_PROTO(struct xfs_mount *mp, unsigned long long bmbt_blocks, unsigned long long rmapbt_blocks), -- cgit v1.2.3-59-g8ed1b From 5fd022ec7d420dfca1eaaf997923a5d4dd0dcf62 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:19 -0700 Subject: xfs: add error injection to test file mapping exchange recovery Add an errortag so that we can test recovery of exchmaps log items. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_errortag.h | 4 +++- fs/xfs/libxfs/xfs_exchmaps.c | 3 +++ fs/xfs/xfs_error.c | 3 +++ 3 files changed, 9 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_errortag.h b/fs/xfs/libxfs/xfs_errortag.h index 01a9e86b3037..7002d7676a78 100644 --- a/fs/xfs/libxfs/xfs_errortag.h +++ b/fs/xfs/libxfs/xfs_errortag.h @@ -63,7 +63,8 @@ #define XFS_ERRTAG_ATTR_LEAF_TO_NODE 41 #define XFS_ERRTAG_WB_DELAY_MS 42 #define XFS_ERRTAG_WRITE_DELAY_MS 43 -#define XFS_ERRTAG_MAX 44 +#define XFS_ERRTAG_EXCHMAPS_FINISH_ONE 44 +#define XFS_ERRTAG_MAX 45 /* * Random factors for above tags, 1 means always, 2 means 1/2 time, etc. @@ -111,5 +112,6 @@ #define XFS_RANDOM_ATTR_LEAF_TO_NODE 1 #define XFS_RANDOM_WB_DELAY_MS 3000 #define XFS_RANDOM_WRITE_DELAY_MS 3000 +#define XFS_RANDOM_EXCHMAPS_FINISH_ONE 1 #endif /* __XFS_ERRORTAG_H_ */ diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index b8e9450cc175..3b1f29e95fea 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -437,6 +437,9 @@ xfs_exchmaps_finish_one( return error; } + if (XFS_TEST_ERROR(false, tp->t_mountp, XFS_ERRTAG_EXCHMAPS_FINISH_ONE)) + return -EIO; + /* If we still have work to do, ask for a new transaction. */ if (xmi_has_more_exchange_work(xmi) || xmi_has_postop_work(xmi)) { trace_xfs_exchmaps_defer(tp->t_mountp, xmi); diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index 7ad0e92c6b5b..78cdc5064a8c 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -62,6 +62,7 @@ static unsigned int xfs_errortag_random_default[] = { XFS_RANDOM_ATTR_LEAF_TO_NODE, XFS_RANDOM_WB_DELAY_MS, XFS_RANDOM_WRITE_DELAY_MS, + XFS_RANDOM_EXCHMAPS_FINISH_ONE, }; struct xfs_errortag_attr { @@ -179,6 +180,7 @@ XFS_ERRORTAG_ATTR_RW(da_leaf_split, XFS_ERRTAG_DA_LEAF_SPLIT); XFS_ERRORTAG_ATTR_RW(attr_leaf_to_node, XFS_ERRTAG_ATTR_LEAF_TO_NODE); XFS_ERRORTAG_ATTR_RW(wb_delay_ms, XFS_ERRTAG_WB_DELAY_MS); XFS_ERRORTAG_ATTR_RW(write_delay_ms, XFS_ERRTAG_WRITE_DELAY_MS); +XFS_ERRORTAG_ATTR_RW(exchmaps_finish_one, XFS_ERRTAG_EXCHMAPS_FINISH_ONE); static struct attribute *xfs_errortag_attrs[] = { XFS_ERRORTAG_ATTR_LIST(noerror), @@ -224,6 +226,7 @@ static struct attribute *xfs_errortag_attrs[] = { XFS_ERRORTAG_ATTR_LIST(attr_leaf_to_node), XFS_ERRORTAG_ATTR_LIST(wb_delay_ms), XFS_ERRORTAG_ATTR_LIST(write_delay_ms), + XFS_ERRORTAG_ATTR_LIST(exchmaps_finish_one), NULL, }; ATTRIBUTE_GROUPS(xfs_errortag); -- cgit v1.2.3-59-g8ed1b From 497d7a2608f8b7329e92bdaaf745ca127a582ad9 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:20 -0700 Subject: xfs: condense extended attributes after a mapping exchange operation Add a new file mapping exchange flag that enables us to perform post-exchange processing on file2 once we're done exchanging the extent mappings. If we were swapping mappings between extended attribute forks, we want to be able to convert file2's attr fork from block to inline format. (This implies that all fork contents are exchanged.) This isn't used anywhere right now, but we need to have the basic ondisk flags in place so that a future online xattr repair feature can create salvaged attrs in a temporary file and exchange the attr fork mappings when ready. If one file is in extents format and the other is inline, we will have to promote both to extents format to perform the exchange. After the exchange, we can try to condense the fixed file's attr fork back down to inline format if possible. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_exchmaps.c | 53 ++++++++++++++++++++++++++++++++++++++++++-- fs/xfs/libxfs/xfs_exchmaps.h | 5 +++++ fs/xfs/xfs_trace.h | 3 ++- 3 files changed, 58 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index 3b1f29e95fea..e46b314fa0cf 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -24,6 +24,10 @@ #include "xfs_errortag.h" #include "xfs_health.h" #include "xfs_exchmaps_item.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_attr_leaf.h" +#include "xfs_attr.h" struct kmem_cache *xfs_exchmaps_intent_cache; @@ -121,7 +125,8 @@ static inline bool xmi_has_postop_work(const struct xfs_exchmaps_intent *xmi) { return xmi->xmi_flags & (XFS_EXCHMAPS_CLEAR_INO1_REFLINK | - XFS_EXCHMAPS_CLEAR_INO2_REFLINK); + XFS_EXCHMAPS_CLEAR_INO2_REFLINK | + __XFS_EXCHMAPS_INO2_SHORTFORM); } /* Check all mappings to make sure we can actually exchange them. */ @@ -360,6 +365,36 @@ xfs_exchmaps_one_step( xmi_advance(xmi, irec1); } +/* Convert inode2's leaf attr fork back to shortform, if possible.. */ +STATIC int +xfs_exchmaps_attr_to_sf( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi) +{ + struct xfs_da_args args = { + .dp = xmi->xmi_ip2, + .geo = tp->t_mountp->m_attr_geo, + .whichfork = XFS_ATTR_FORK, + .trans = tp, + }; + struct xfs_buf *bp; + int forkoff; + int error; + + if (!xfs_attr_is_leaf(xmi->xmi_ip2)) + return 0; + + error = xfs_attr3_leaf_read(tp, xmi->xmi_ip2, 0, &bp); + if (error) + return error; + + forkoff = xfs_attr_shortform_allfit(bp, xmi->xmi_ip2); + if (forkoff == 0) + return 0; + + return xfs_attr3_leaf_to_shortform(bp, &args, forkoff); +} + /* Clear the reflink flag after an exchange. */ static inline void xfs_exchmaps_clear_reflink( @@ -378,6 +413,16 @@ xfs_exchmaps_do_postop_work( struct xfs_trans *tp, struct xfs_exchmaps_intent *xmi) { + if (xmi->xmi_flags & __XFS_EXCHMAPS_INO2_SHORTFORM) { + int error = 0; + + if (xmi->xmi_flags & XFS_EXCHMAPS_ATTR_FORK) + error = xfs_exchmaps_attr_to_sf(tp, xmi); + xmi->xmi_flags &= ~__XFS_EXCHMAPS_INO2_SHORTFORM; + if (error) + return error; + } + if (xmi->xmi_flags & XFS_EXCHMAPS_CLEAR_INO1_REFLINK) { xfs_exchmaps_clear_reflink(tp, xmi->xmi_ip1); xmi->xmi_flags &= ~XFS_EXCHMAPS_CLEAR_INO1_REFLINK; @@ -809,8 +854,10 @@ xfs_exchmaps_init_intent( xmi->xmi_isize1 = xmi->xmi_isize2 = -1; xmi->xmi_flags = req->flags & XFS_EXCHMAPS_PARAMS; - if (xfs_exchmaps_whichfork(xmi) == XFS_ATTR_FORK) + if (xfs_exchmaps_whichfork(xmi) == XFS_ATTR_FORK) { + xmi->xmi_flags |= __XFS_EXCHMAPS_INO2_SHORTFORM; return xmi; + } if (req->flags & XFS_EXCHMAPS_SET_SIZES) { xmi->xmi_flags |= XFS_EXCHMAPS_SET_SIZES; @@ -1031,6 +1078,8 @@ xfs_exchange_mappings( { struct xfs_exchmaps_intent *xmi; + BUILD_BUG_ON(XFS_EXCHMAPS_INTERNAL_FLAGS & XFS_EXCHMAPS_LOGGED_FLAGS); + xfs_assert_ilocked(req->ip1, XFS_ILOCK_EXCL); xfs_assert_ilocked(req->ip2, XFS_ILOCK_EXCL); ASSERT(!(req->flags & ~XFS_EXCHMAPS_LOGGED_FLAGS)); diff --git a/fs/xfs/libxfs/xfs_exchmaps.h b/fs/xfs/libxfs/xfs_exchmaps.h index e8fc3f80c68c..d8718fca606e 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.h +++ b/fs/xfs/libxfs/xfs_exchmaps.h @@ -27,6 +27,11 @@ struct xfs_exchmaps_intent { uint64_t xmi_flags; /* XFS_EXCHMAPS_* flags */ }; +/* Try to convert inode2 from block to short format at the end, if possible. */ +#define __XFS_EXCHMAPS_INO2_SHORTFORM (1ULL << 63) + +#define XFS_EXCHMAPS_INTERNAL_FLAGS (__XFS_EXCHMAPS_INO2_SHORTFORM) + /* flags that can be passed to xfs_exchmaps_{estimate,mappings} */ #define XFS_EXCHMAPS_PARAMS (XFS_EXCHMAPS_ATTR_FORK | \ XFS_EXCHMAPS_SET_SIZES | \ diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 729e728c2076..caef95f2c87c 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -4779,7 +4779,8 @@ DEFINE_XFBTREE_FREESP_EVENT(xfbtree_free_block); { XFS_EXCHMAPS_SET_SIZES, "SETSIZES" }, \ { XFS_EXCHMAPS_INO1_WRITTEN, "INO1_WRITTEN" }, \ { XFS_EXCHMAPS_CLEAR_INO1_REFLINK, "CLEAR_INO1_REFLINK" }, \ - { XFS_EXCHMAPS_CLEAR_INO2_REFLINK, "CLEAR_INO2_REFLINK" } + { XFS_EXCHMAPS_CLEAR_INO2_REFLINK, "CLEAR_INO2_REFLINK" }, \ + { __XFS_EXCHMAPS_INO2_SHORTFORM, "INO2_SF" } DEFINE_INODE_IREC_EVENT(xfs_exchmaps_mapping1_skip); DEFINE_INODE_IREC_EVENT(xfs_exchmaps_mapping1); -- cgit v1.2.3-59-g8ed1b From da165fbde23b84591b6ccdf6749277d2d767b770 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:20 -0700 Subject: xfs: condense directories after a mapping exchange operation The previous commit added a new file mapping exchange flag that enables us to perform post-swap processing on file2 once we're done exchanging extent mappings. Now add this ability for directories. This isn't used anywhere right now, but we need to have the basic ondisk flags in place so that a future online directory repair feature can create salvaged dirents in a temporary directory and exchange the data fork mappings when ready. If one file is in extents format and the other is inline, we will have to promote both to extents format to perform the exchange. After the exchange, we can try to condense the fixed directory down to inline format if possible. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_exchmaps.c | 43 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index e46b314fa0cf..f199629adbf0 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -28,6 +28,8 @@ #include "xfs_da_btree.h" #include "xfs_attr_leaf.h" #include "xfs_attr.h" +#include "xfs_dir2_priv.h" +#include "xfs_dir2.h" struct kmem_cache *xfs_exchmaps_intent_cache; @@ -395,6 +397,42 @@ xfs_exchmaps_attr_to_sf( return xfs_attr3_leaf_to_shortform(bp, &args, forkoff); } +/* Convert inode2's block dir fork back to shortform, if possible.. */ +STATIC int +xfs_exchmaps_dir_to_sf( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi) +{ + struct xfs_da_args args = { + .dp = xmi->xmi_ip2, + .geo = tp->t_mountp->m_dir_geo, + .whichfork = XFS_DATA_FORK, + .trans = tp, + }; + struct xfs_dir2_sf_hdr sfh; + struct xfs_buf *bp; + bool isblock; + int size; + int error; + + error = xfs_dir2_isblock(&args, &isblock); + if (error) + return error; + + if (!isblock) + return 0; + + error = xfs_dir3_block_read(tp, xmi->xmi_ip2, &bp); + if (error) + return error; + + size = xfs_dir2_block_sfsize(xmi->xmi_ip2, bp->b_addr, &sfh); + if (size > xfs_inode_data_fork_size(xmi->xmi_ip2)) + return 0; + + return xfs_dir2_block_to_sf(&args, bp, size, &sfh); +} + /* Clear the reflink flag after an exchange. */ static inline void xfs_exchmaps_clear_reflink( @@ -418,6 +456,8 @@ xfs_exchmaps_do_postop_work( if (xmi->xmi_flags & XFS_EXCHMAPS_ATTR_FORK) error = xfs_exchmaps_attr_to_sf(tp, xmi); + else if (S_ISDIR(VFS_I(xmi->xmi_ip2)->i_mode)) + error = xfs_exchmaps_dir_to_sf(tp, xmi); xmi->xmi_flags &= ~__XFS_EXCHMAPS_INO2_SHORTFORM; if (error) return error; @@ -882,6 +922,9 @@ xfs_exchmaps_init_intent( xmi->xmi_flags |= XFS_EXCHMAPS_CLEAR_INO2_REFLINK; } + if (S_ISDIR(VFS_I(xmi->xmi_ip2)->i_mode)) + xmi->xmi_flags |= __XFS_EXCHMAPS_INO2_SHORTFORM; + return xmi; } -- cgit v1.2.3-59-g8ed1b From 33a9be2b7016e79f47c4c1b523a0aa59d41893c0 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:21 -0700 Subject: xfs: condense symbolic links after a mapping exchange operation The previous commit added a new file mapping exchange flag that enables us to perform post-exchange processing on file2 once we're done exchanging the extent mappings. Now add this ability for symlinks. This isn't used anywhere right now, but we need to have the basic ondisk flags in place so that a future online symlink repair feature can salvage the remote target in a temporary link and exchange the data fork mappings when ready. If one file is in extents format and the other is inline, we will have to promote both to extents format to perform the exchange. After the exchange, we can try to condense the fixed symlink down to inline format if possible. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_exchmaps.c | 49 +++++++++++++++++++++++++++++++++++- fs/xfs/libxfs/xfs_symlink_remote.c | 47 +++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_symlink_remote.h | 1 + fs/xfs/xfs_symlink.c | 51 ++++++-------------------------------- 4 files changed, 103 insertions(+), 45 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index f199629adbf0..f58240466b1c 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -30,6 +30,7 @@ #include "xfs_attr.h" #include "xfs_dir2_priv.h" #include "xfs_dir2.h" +#include "xfs_symlink_remote.h" struct kmem_cache *xfs_exchmaps_intent_cache; @@ -433,6 +434,49 @@ xfs_exchmaps_dir_to_sf( return xfs_dir2_block_to_sf(&args, bp, size, &sfh); } +/* Convert inode2's remote symlink target back to shortform, if possible. */ +STATIC int +xfs_exchmaps_link_to_sf( + struct xfs_trans *tp, + struct xfs_exchmaps_intent *xmi) +{ + struct xfs_inode *ip = xmi->xmi_ip2; + struct xfs_ifork *ifp = xfs_ifork_ptr(ip, XFS_DATA_FORK); + char *buf; + int error; + + if (ifp->if_format == XFS_DINODE_FMT_LOCAL || + ip->i_disk_size > xfs_inode_data_fork_size(ip)) + return 0; + + /* Read the current symlink target into a buffer. */ + buf = kmalloc(ip->i_disk_size + 1, + GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_NOFAIL); + if (!buf) { + ASSERT(0); + return -ENOMEM; + } + + error = xfs_symlink_remote_read(ip, buf); + if (error) + goto free; + + /* Remove the blocks. */ + error = xfs_symlink_remote_truncate(tp, ip); + if (error) + goto free; + + /* Convert fork to local format and log our changes. */ + xfs_idestroy_fork(ifp); + ifp->if_bytes = 0; + ifp->if_format = XFS_DINODE_FMT_LOCAL; + xfs_init_local_fork(ip, XFS_DATA_FORK, buf, ip->i_disk_size); + xfs_trans_log_inode(tp, ip, XFS_ILOG_DDATA | XFS_ILOG_CORE); +free: + kfree(buf); + return error; +} + /* Clear the reflink flag after an exchange. */ static inline void xfs_exchmaps_clear_reflink( @@ -458,6 +502,8 @@ xfs_exchmaps_do_postop_work( error = xfs_exchmaps_attr_to_sf(tp, xmi); else if (S_ISDIR(VFS_I(xmi->xmi_ip2)->i_mode)) error = xfs_exchmaps_dir_to_sf(tp, xmi); + else if (S_ISLNK(VFS_I(xmi->xmi_ip2)->i_mode)) + error = xfs_exchmaps_link_to_sf(tp, xmi); xmi->xmi_flags &= ~__XFS_EXCHMAPS_INO2_SHORTFORM; if (error) return error; @@ -922,7 +968,8 @@ xfs_exchmaps_init_intent( xmi->xmi_flags |= XFS_EXCHMAPS_CLEAR_INO2_REFLINK; } - if (S_ISDIR(VFS_I(xmi->xmi_ip2)->i_mode)) + if (S_ISDIR(VFS_I(xmi->xmi_ip2)->i_mode) || + S_ISLNK(VFS_I(xmi->xmi_ip2)->i_mode)) xmi->xmi_flags |= __XFS_EXCHMAPS_INO2_SHORTFORM; return xmi; diff --git a/fs/xfs/libxfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c index ffb1317a9212..8f0d5c584f46 100644 --- a/fs/xfs/libxfs/xfs_symlink_remote.c +++ b/fs/xfs/libxfs/xfs_symlink_remote.c @@ -380,3 +380,50 @@ xfs_symlink_write_target( ASSERT(pathlen == 0); return 0; } + +/* Remove all the blocks from a symlink and invalidate buffers. */ +int +xfs_symlink_remote_truncate( + struct xfs_trans *tp, + struct xfs_inode *ip) +{ + struct xfs_bmbt_irec mval[XFS_SYMLINK_MAPS]; + struct xfs_mount *mp = tp->t_mountp; + struct xfs_buf *bp; + int nmaps = XFS_SYMLINK_MAPS; + int done = 0; + int i; + int error; + + /* Read mappings and invalidate buffers. */ + error = xfs_bmapi_read(ip, 0, XFS_MAX_FILEOFF, mval, &nmaps, 0); + if (error) + return error; + + for (i = 0; i < nmaps; i++) { + if (!xfs_bmap_is_real_extent(&mval[i])) + break; + + error = xfs_trans_get_buf(tp, mp->m_ddev_targp, + XFS_FSB_TO_DADDR(mp, mval[i].br_startblock), + XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0, + &bp); + if (error) + return error; + + xfs_trans_binval(tp, bp); + } + + /* Unmap the remote blocks. */ + error = xfs_bunmapi(tp, ip, 0, XFS_MAX_FILEOFF, 0, nmaps, &done); + if (error) + return error; + if (!done) { + ASSERT(done); + xfs_inode_mark_sick(ip, XFS_SICK_INO_SYMLINK); + return -EFSCORRUPTED; + } + + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + return 0; +} diff --git a/fs/xfs/libxfs/xfs_symlink_remote.h b/fs/xfs/libxfs/xfs_symlink_remote.h index a63bd38ae4fa..ac3dac8f617e 100644 --- a/fs/xfs/libxfs/xfs_symlink_remote.h +++ b/fs/xfs/libxfs/xfs_symlink_remote.h @@ -22,5 +22,6 @@ int xfs_symlink_remote_read(struct xfs_inode *ip, char *link); int xfs_symlink_write_target(struct xfs_trans *tp, struct xfs_inode *ip, const char *target_path, int pathlen, xfs_fsblock_t fs_blocks, uint resblks); +int xfs_symlink_remote_truncate(struct xfs_trans *tp, struct xfs_inode *ip); #endif /* __XFS_SYMLINK_REMOTE_H */ diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 3e376d24c7c1..3daeebff4bb4 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -250,19 +250,12 @@ out_release_dquots: */ STATIC int xfs_inactive_symlink_rmt( - struct xfs_inode *ip) + struct xfs_inode *ip) { - struct xfs_buf *bp; - int done; - int error; - int i; - xfs_mount_t *mp; - xfs_bmbt_irec_t mval[XFS_SYMLINK_MAPS]; - int nmaps; - int size; - xfs_trans_t *tp; - - mp = ip->i_mount; + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + int error; + ASSERT(!xfs_need_iread_extents(&ip->i_df)); /* * We're freeing a symlink that has some @@ -286,44 +279,14 @@ xfs_inactive_symlink_rmt( * locked for the second transaction. In the error paths we need it * held so the cancel won't rele it, see below. */ - size = (int)ip->i_disk_size; ip->i_disk_size = 0; VFS_I(ip)->i_mode = (VFS_I(ip)->i_mode & ~S_IFMT) | S_IFREG; xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - /* - * Find the block(s) so we can inval and unmap them. - */ - done = 0; - nmaps = ARRAY_SIZE(mval); - error = xfs_bmapi_read(ip, 0, xfs_symlink_blocks(mp, size), - mval, &nmaps, 0); - if (error) - goto error_trans_cancel; - /* - * Invalidate the block(s). No validation is done. - */ - for (i = 0; i < nmaps; i++) { - error = xfs_trans_get_buf(tp, mp->m_ddev_targp, - XFS_FSB_TO_DADDR(mp, mval[i].br_startblock), - XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0, - &bp); - if (error) - goto error_trans_cancel; - xfs_trans_binval(tp, bp); - } - /* - * Unmap the dead block(s) to the dfops. - */ - error = xfs_bunmapi(tp, ip, 0, size, 0, nmaps, &done); + + error = xfs_symlink_remote_truncate(tp, ip); if (error) goto error_trans_cancel; - ASSERT(done); - /* - * Commit the transaction. This first logs the EFI and the inode, then - * rolls and commits the transaction that frees the extents. - */ - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); error = xfs_trans_commit(tp); if (error) { ASSERT(xfs_is_shutdown(mp)); -- cgit v1.2.3-59-g8ed1b From e62941103faa2eedba6a155316e059a490c743a6 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:22 -0700 Subject: xfs: make file range exchange support realtime files Now that bmap items support the realtime device, we can add the necessary pieces to the file range exchange code to support exchanging mappings. All we really need to do here is adjust the blockcount upwards to the end of the rt extent and remove the inode checks. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_exchmaps.c | 70 +++++++++++++++++++++++++++++++++++++------- fs/xfs/xfs_exchrange.c | 9 ++++++ 2 files changed, 69 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index f58240466b1c..7fa244228750 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -152,12 +152,7 @@ xfs_exchmaps_check_forks( ifp2->if_format == XFS_DINODE_FMT_LOCAL) return -EINVAL; - /* We don't support realtime data forks yet. */ - if (!XFS_IS_REALTIME_INODE(req->ip1)) - return 0; - if (whichfork == XFS_ATTR_FORK) - return 0; - return -EINVAL; + return 0; } #ifdef CONFIG_XFS_QUOTA @@ -198,6 +193,8 @@ xfs_exchmaps_can_skip_mapping( struct xfs_exchmaps_intent *xmi, struct xfs_bmbt_irec *irec) { + struct xfs_mount *mp = xmi->xmi_ip1->i_mount; + /* Do not skip this mapping if the caller did not tell us to. */ if (!(xmi->xmi_flags & XFS_EXCHMAPS_INO1_WRITTEN)) return false; @@ -209,11 +206,64 @@ xfs_exchmaps_can_skip_mapping( /* * The mapping is unwritten or a hole. It cannot be a delalloc * reservation because we already excluded those. It cannot be an - * unwritten mapping with dirty page cache because we flushed the page - * cache. We don't support realtime files yet, so we needn't (yet) - * deal with them. + * unwritten extent with dirty page cache because we flushed the page + * cache. For files where the allocation unit is 1FSB (files on the + * data dev, rt files if the extent size is 1FSB), we can safely + * skip this mapping. */ - return true; + if (!xfs_inode_has_bigrtalloc(xmi->xmi_ip1)) + return true; + + /* + * For a realtime file with a multi-fsb allocation unit, the decision + * is trickier because we can only swap full allocation units. + * Unwritten mappings can appear in the middle of an rtx if the rtx is + * partially written, but they can also appear for preallocations. + * + * If the mapping is a hole, skip it entirely. Holes should align with + * rtx boundaries. + */ + if (!xfs_bmap_is_real_extent(irec)) + return true; + + /* + * All mappings below this point are unwritten. + * + * - If the beginning is not aligned to an rtx, trim the end of the + * mapping so that it does not cross an rtx boundary, and swap it. + * + * - If both ends are aligned to an rtx, skip the entire mapping. + */ + if (!isaligned_64(irec->br_startoff, mp->m_sb.sb_rextsize)) { + xfs_fileoff_t new_end; + + new_end = roundup_64(irec->br_startoff, mp->m_sb.sb_rextsize); + irec->br_blockcount = min(irec->br_blockcount, + new_end - irec->br_startoff); + return false; + } + if (isaligned_64(irec->br_blockcount, mp->m_sb.sb_rextsize)) + return true; + + /* + * All mappings below this point are unwritten, start on an rtx + * boundary, and do not end on an rtx boundary. + * + * - If the mapping is longer than one rtx, trim the end of the mapping + * down to an rtx boundary and skip it. + * + * - The mapping is shorter than one rtx. Swap it. + */ + if (irec->br_blockcount > mp->m_sb.sb_rextsize) { + xfs_fileoff_t new_end; + + new_end = rounddown_64(irec->br_startoff + irec->br_blockcount, + mp->m_sb.sb_rextsize); + irec->br_blockcount = new_end - irec->br_startoff; + return true; + } + + return false; } /* diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c index 0fc95e6471cb..90baf12bd97f 100644 --- a/fs/xfs/xfs_exchrange.c +++ b/fs/xfs/xfs_exchrange.c @@ -21,6 +21,7 @@ #include "xfs_sb.h" #include "xfs_icache.h" #include "xfs_log.h" +#include "xfs_rtbitmap.h" #include /* Lock (and optionally join) two inodes for a file range exchange. */ @@ -182,6 +183,14 @@ xfs_exchrange_mappings( if (fxr->flags & XFS_EXCHANGE_RANGE_FILE1_WRITTEN) req.flags |= XFS_EXCHMAPS_INO1_WRITTEN; + /* + * Round the request length up to the nearest file allocation unit. + * The prep function already checked that the request offsets and + * length in @fxr are safe to round up. + */ + if (xfs_inode_has_bigrtalloc(ip2)) + req.blockcount = xfs_rtb_roundup_rtx(mp, req.blockcount); + error = xfs_exchrange_estimate(&req); if (error) return error; -- cgit v1.2.3-59-g8ed1b From b3e60f84838d5abc3a73d7ef0fc595dd1041c565 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:23 -0700 Subject: xfs: support non-power-of-two rtextsize with exchange-range The generic exchange-range alignment checks use (fast) bitmasking operations to perform block alignment checks on the exchange parameters. Unfortunately, bitmasks require that the alignment size be a power of two. This isn't true for realtime devices with a non-power-of-two extent size, so we have to copy-pasta the generic checks using long division for this to work properly. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_exchrange.c | 89 ++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 82 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c index 90baf12bd97f..c8a655c92c92 100644 --- a/fs/xfs/xfs_exchrange.c +++ b/fs/xfs/xfs_exchrange.c @@ -504,6 +504,75 @@ xfs_exchange_range_finish( return file_remove_privs(fxr->file2); } +/* + * Check the alignment of an exchange request when the allocation unit size + * isn't a power of two. The generic file-level helpers use (fast) + * bitmask-based alignment checks, but here we have to use slow long division. + */ +static int +xfs_exchrange_check_rtalign( + const struct xfs_exchrange *fxr, + struct xfs_inode *ip1, + struct xfs_inode *ip2, + unsigned int alloc_unit) +{ + uint64_t length = fxr->length; + uint64_t blen; + loff_t size1, size2; + + size1 = i_size_read(VFS_I(ip1)); + size2 = i_size_read(VFS_I(ip2)); + + /* The start of both ranges must be aligned to a rt extent. */ + if (!isaligned_64(fxr->file1_offset, alloc_unit) || + !isaligned_64(fxr->file2_offset, alloc_unit)) + return -EINVAL; + + if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) + length = max_t(int64_t, size1 - fxr->file1_offset, + size2 - fxr->file2_offset); + + /* + * If the user wanted us to exchange up to the infile's EOF, round up + * to the next rt extent boundary for this check. Do the same for the + * outfile. + * + * Otherwise, reject the range length if it's not rt extent aligned. + * We already confirmed the starting offsets' rt extent block + * alignment. + */ + if (fxr->file1_offset + length == size1) + blen = roundup_64(size1, alloc_unit) - fxr->file1_offset; + else if (fxr->file2_offset + length == size2) + blen = roundup_64(size2, alloc_unit) - fxr->file2_offset; + else if (!isaligned_64(length, alloc_unit)) + return -EINVAL; + else + blen = length; + + /* Don't allow overlapped exchanges within the same file. */ + if (ip1 == ip2 && + fxr->file2_offset + blen > fxr->file1_offset && + fxr->file1_offset + blen > fxr->file2_offset) + return -EINVAL; + + /* + * Ensure that we don't exchange a partial EOF rt extent into the + * middle of another file. + */ + if (isaligned_64(length, alloc_unit)) + return 0; + + blen = length; + if (fxr->file2_offset + length < size2) + blen = rounddown_64(blen, alloc_unit); + + if (fxr->file1_offset + blen < size1) + blen = rounddown_64(blen, alloc_unit); + + return blen == length ? 0 : -EINVAL; +} + /* Prepare two files to have their data exchanged. */ STATIC int xfs_exchrange_prep( @@ -511,6 +580,7 @@ xfs_exchrange_prep( struct xfs_inode *ip1, struct xfs_inode *ip2) { + struct xfs_mount *mp = ip2->i_mount; unsigned int alloc_unit = xfs_inode_alloc_unitsize(ip2); int error; @@ -520,13 +590,18 @@ xfs_exchrange_prep( if (XFS_IS_REALTIME_INODE(ip1) != XFS_IS_REALTIME_INODE(ip2)) return -EINVAL; - /* - * The alignment checks in the generic helpers cannot deal with - * allocation units that are not powers of 2. This can happen with the - * realtime volume if the extent size is set. - */ - if (!is_power_of_2(alloc_unit)) - return -EOPNOTSUPP; + /* Check non-power of two alignment issues, if necessary. */ + if (!is_power_of_2(alloc_unit)) { + error = xfs_exchrange_check_rtalign(fxr, ip1, ip2, alloc_unit); + if (error) + return error; + + /* + * Do the generic file-level checks with the regular block + * alignment. + */ + alloc_unit = mp->m_sb.sb_blocksize; + } error = xfs_exchange_range_prep(fxr, alloc_unit); if (error || fxr->length == 0) -- cgit v1.2.3-59-g8ed1b From 14f19991020b3c712d626727c17599f134cc6efa Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:24 -0700 Subject: xfs: capture inode generation numbers in the ondisk exchmaps log item Per some very late review comments, capture the generation numbers of both inodes involved in a file content exchange operation so that we don't accidentally target files with have been reallocated. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_log_format.h | 2 ++ fs/xfs/libxfs/xfs_log_recover.h | 2 ++ fs/xfs/xfs_exchmaps_item.c | 25 ++++++++++++++++++++----- fs/xfs/xfs_log_recover.c | 31 +++++++++++++++++++++++++++++++ 4 files changed, 55 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index 8dbe1f997dfd..accba2acd623 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -896,6 +896,8 @@ struct xfs_xmi_log_format { uint64_t xmi_inode1; /* inumber of first file */ uint64_t xmi_inode2; /* inumber of second file */ + uint32_t xmi_igen1; /* generation of first file */ + uint32_t xmi_igen2; /* generation of second file */ uint64_t xmi_startoff1; /* block offset into file1 */ uint64_t xmi_startoff2; /* block offset into file2 */ uint64_t xmi_blockcount; /* number of blocks */ diff --git a/fs/xfs/libxfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h index 47b758b49cb3..521d327e4c89 100644 --- a/fs/xfs/libxfs/xfs_log_recover.h +++ b/fs/xfs/libxfs/xfs_log_recover.h @@ -123,6 +123,8 @@ bool xlog_is_buffer_cancelled(struct xlog *log, xfs_daddr_t blkno, uint len); int xlog_recover_iget(struct xfs_mount *mp, xfs_ino_t ino, struct xfs_inode **ipp); +int xlog_recover_iget_handle(struct xfs_mount *mp, xfs_ino_t ino, uint32_t gen, + struct xfs_inode **ipp); void xlog_recover_release_intent(struct xlog *log, unsigned short intent_type, uint64_t intent_id); int xlog_alloc_buf_cancel_table(struct xlog *log); diff --git a/fs/xfs/xfs_exchmaps_item.c b/fs/xfs/xfs_exchmaps_item.c index a40216f33214..264a121c5e16 100644 --- a/fs/xfs/xfs_exchmaps_item.c +++ b/fs/xfs/xfs_exchmaps_item.c @@ -231,7 +231,9 @@ xfs_exchmaps_create_intent( xlf = &xmi_lip->xmi_format; xlf->xmi_inode1 = xmi->xmi_ip1->i_ino; + xlf->xmi_igen1 = VFS_I(xmi->xmi_ip1)->i_generation; xlf->xmi_inode2 = xmi->xmi_ip2->i_ino; + xlf->xmi_igen2 = VFS_I(xmi->xmi_ip2)->i_generation; xlf->xmi_startoff1 = xmi->xmi_startoff1; xlf->xmi_startoff2 = xmi->xmi_startoff2; xlf->xmi_blockcount = xmi->xmi_blockcount; @@ -368,14 +370,25 @@ xfs_xmi_item_recover_intent( /* * Grab both inodes and set IRECOVERY to prevent trimming of post-eof * mappings and freeing of unlinked inodes until we're totally done - * processing files. + * processing files. The ondisk format of this new log item contains + * file handle information, which is why recovery for other items do + * not check the inode generation number. */ - error = xlog_recover_iget(mp, xlf->xmi_inode1, &ip1); - if (error) + error = xlog_recover_iget_handle(mp, xlf->xmi_inode1, xlf->xmi_igen1, + &ip1); + if (error) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, xlf, + sizeof(*xlf)); return ERR_PTR(error); - error = xlog_recover_iget(mp, xlf->xmi_inode2, &ip2); - if (error) + } + + error = xlog_recover_iget_handle(mp, xlf->xmi_inode2, xlf->xmi_igen2, + &ip2); + if (error) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, xlf, + sizeof(*xlf)); goto err_rele1; + } req->ip1 = ip1; req->ip2 = ip2; @@ -485,6 +498,8 @@ xfs_exchmaps_relog_intent( new_xlf->xmi_inode1 = old_xlf->xmi_inode1; new_xlf->xmi_inode2 = old_xlf->xmi_inode2; + new_xlf->xmi_igen1 = old_xlf->xmi_igen1; + new_xlf->xmi_igen2 = old_xlf->xmi_igen2; new_xlf->xmi_startoff1 = old_xlf->xmi_startoff1; new_xlf->xmi_startoff2 = old_xlf->xmi_startoff2; new_xlf->xmi_blockcount = old_xlf->xmi_blockcount; diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 1e5ba95adf2c..b445e8ce4a7d 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1767,6 +1767,37 @@ xlog_recover_iget( return 0; } +/* + * Get an inode so that we can recover a log operation. + * + * Log intent items that target inodes effectively contain a file handle. + * Check that the generation number matches the intent item like we do for + * other file handles. Log intent items defined after this validation weakness + * was identified must use this function. + */ +int +xlog_recover_iget_handle( + struct xfs_mount *mp, + xfs_ino_t ino, + uint32_t gen, + struct xfs_inode **ipp) +{ + struct xfs_inode *ip; + int error; + + error = xlog_recover_iget(mp, ino, &ip); + if (error) + return error; + + if (VFS_I(ip)->i_generation != gen) { + xfs_irele(ip); + return -EFSCORRUPTED; + } + + *ipp = ip; + return 0; +} + /****************************************************************************** * * Log recover routines -- cgit v1.2.3-59-g8ed1b From 0730e8d8ba1d1507f1d7fd719e1f835ce69961fe Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:26 -0700 Subject: xfs: enable logged file mapping exchange feature Add the XFS_SB_FEAT_INCOMPAT_EXCHRANGE feature to the set of features that we will permit when mounting a filesystem. This turns on support for the file range exchange feature. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_format.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index ff1e28316e1b..10153ce116d4 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -380,7 +380,8 @@ xfs_sb_has_ro_compat_feature( XFS_SB_FEAT_INCOMPAT_META_UUID | \ XFS_SB_FEAT_INCOMPAT_BIGTIME | \ XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR | \ - XFS_SB_FEAT_INCOMPAT_NREXT64) + XFS_SB_FEAT_INCOMPAT_NREXT64 | \ + XFS_SB_FEAT_INCOMPAT_EXCHRANGE) #define XFS_SB_FEAT_INCOMPAT_UNKNOWN ~XFS_SB_FEAT_INCOMPAT_ALL static inline bool -- cgit v1.2.3-59-g8ed1b From cab23a4233c601668da51dd53776f1f83d0dccee Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:27 -0700 Subject: xfs: hide private inodes from bulkstat and handle functions We're about to start adding functionality that uses internal inodes that are private to XFS. What this means is that userspace should never be able to access any information about these files, and should not be able to open these files by handle. To prevent users from ever finding the file or mis-interactions with the security apparatus, set S_PRIVATE on the inode. Don't allow bulkstat, open-by-handle, or linking of S_PRIVATE files into the directory tree. This should keep private inodes actually private. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_export.c | 2 +- fs/xfs/xfs_iops.c | 3 +++ fs/xfs/xfs_itable.c | 8 ++++++++ 3 files changed, 12 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c index 7cd09c3a82cb..4b03221351c0 100644 --- a/fs/xfs/xfs_export.c +++ b/fs/xfs/xfs_export.c @@ -160,7 +160,7 @@ xfs_nfs_get_inode( } } - if (VFS_I(ip)->i_generation != generation) { + if (VFS_I(ip)->i_generation != generation || IS_PRIVATE(VFS_I(ip))) { xfs_irele(ip); return ERR_PTR(-ESTALE); } diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 55ed2d1023d6..7f0c840f0fd2 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -365,6 +365,9 @@ xfs_vn_link( if (unlikely(error)) return error; + if (IS_PRIVATE(inode)) + return -EPERM; + error = xfs_link(XFS_I(dir), XFS_I(inode), &name); if (unlikely(error)) return error; diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 95fc31b9f87d..c0757ab99495 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -97,6 +97,14 @@ xfs_bulkstat_one_int( vfsuid = i_uid_into_vfsuid(idmap, inode); vfsgid = i_gid_into_vfsgid(idmap, inode); + /* If this is a private inode, don't leak its details to userspace. */ + if (IS_PRIVATE(inode)) { + xfs_iunlock(ip, XFS_ILOCK_SHARED); + xfs_irele(ip); + error = -EINVAL; + goto out_advance; + } + /* xfs_iget returns the following without needing * further change. */ -- cgit v1.2.3-59-g8ed1b From 84c14ee39dd388d73054181c70b3f42af7ddc238 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:28 -0700 Subject: xfs: create temporary files and directories for online repair Teach the online repair code how to create temporary files or directories. These temporary files can be used to stage reconstructed information until we're ready to perform an atomic extent swap to commit the new metadata. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/parent.c | 2 +- fs/xfs/scrub/scrub.c | 3 + fs/xfs/scrub/scrub.h | 4 + fs/xfs/scrub/tempfile.c | 251 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/tempfile.h | 28 ++++++ fs/xfs/scrub/trace.h | 33 +++++++ fs/xfs/xfs_inode.c | 3 +- fs/xfs/xfs_inode.h | 2 + 9 files changed, 324 insertions(+), 3 deletions(-) create mode 100644 fs/xfs/scrub/tempfile.c create mode 100644 fs/xfs/scrub/tempfile.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index b547a3dc03f8..ae8488ab4d6b 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -207,6 +207,7 @@ xfs-y += $(addprefix scrub/, \ refcount_repair.o \ repair.o \ rmap_repair.o \ + tempfile.o \ ) xfs-$(CONFIG_XFS_RT) += $(addprefix scrub/, \ diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index 7db873672146..5da10ed1fe8c 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -143,7 +143,7 @@ xchk_parent_validate( } if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error)) return error; - if (dp == sc->ip || !S_ISDIR(VFS_I(dp)->i_mode)) { + if (dp == sc->ip || dp == sc->tempip || !S_ISDIR(VFS_I(dp)->i_mode)) { xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); goto out_rele; } diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 20fac9723c08..d9012e9a6afd 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -17,6 +17,7 @@ #include "xfs_scrub.h" #include "xfs_buf_mem.h" #include "xfs_rmap.h" +#include "xfs_exchrange.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" @@ -24,6 +25,7 @@ #include "scrub/health.h" #include "scrub/stats.h" #include "scrub/xfile.h" +#include "scrub/tempfile.h" /* * Online Scrub and Repair @@ -211,6 +213,7 @@ xchk_teardown( sc->buf = NULL; } + xrep_tempfile_rele(sc); xchk_fsgates_disable(sc); return error; } diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index 9ad65b604fe1..e37d8599718e 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -105,6 +105,10 @@ struct xfs_scrub { /* Lock flags for @ip. */ uint ilock_flags; + /* A temporary file on this filesystem, for staging new metadata. */ + struct xfs_inode *tempip; + uint temp_ilock_flags; + /* See the XCHK/XREP state flags below. */ unsigned int flags; diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c new file mode 100644 index 000000000000..68d245749bc1 --- /dev/null +++ b/fs/xfs/scrub/tempfile.c @@ -0,0 +1,251 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2021-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_quota.h" +#include "xfs_bmap_btree.h" +#include "xfs_trans_space.h" +#include "xfs_dir2.h" +#include "xfs_exchrange.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/tempfile.h" + +/* + * Create a temporary file for reconstructing metadata, with the intention of + * atomically exchanging the temporary file's contents with the file that's + * being repaired. + */ +int +xrep_tempfile_create( + struct xfs_scrub *sc, + uint16_t mode) +{ + struct xfs_mount *mp = sc->mp; + struct xfs_trans *tp = NULL; + struct xfs_dquot *udqp = NULL; + struct xfs_dquot *gdqp = NULL; + struct xfs_dquot *pdqp = NULL; + struct xfs_trans_res *tres; + struct xfs_inode *dp = mp->m_rootip; + xfs_ino_t ino; + unsigned int resblks; + bool is_dir = S_ISDIR(mode); + int error; + + if (xfs_is_shutdown(mp)) + return -EIO; + if (xfs_is_readonly(mp)) + return -EROFS; + + ASSERT(sc->tp == NULL); + ASSERT(sc->tempip == NULL); + + /* + * Make sure that we have allocated dquot(s) on disk. The temporary + * inode should be completely root owned so that we don't fail due to + * quota limits. + */ + error = xfs_qm_vop_dqalloc(dp, GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, 0, + XFS_QMOPT_QUOTALL, &udqp, &gdqp, &pdqp); + if (error) + return error; + + if (is_dir) { + resblks = XFS_MKDIR_SPACE_RES(mp, 0); + tres = &M_RES(mp)->tr_mkdir; + } else { + resblks = XFS_IALLOC_SPACE_RES(mp); + tres = &M_RES(mp)->tr_create_tmpfile; + } + + error = xfs_trans_alloc_icreate(mp, tres, udqp, gdqp, pdqp, resblks, + &tp); + if (error) + goto out_release_dquots; + + /* Allocate inode, set up directory. */ + error = xfs_dialloc(&tp, dp->i_ino, mode, &ino); + if (error) + goto out_trans_cancel; + error = xfs_init_new_inode(&nop_mnt_idmap, tp, dp, ino, mode, 0, 0, + 0, false, &sc->tempip); + if (error) + goto out_trans_cancel; + + /* Change the ownership of the inode to root. */ + VFS_I(sc->tempip)->i_uid = GLOBAL_ROOT_UID; + VFS_I(sc->tempip)->i_gid = GLOBAL_ROOT_GID; + sc->tempip->i_diflags &= ~(XFS_DIFLAG_REALTIME | XFS_DIFLAG_RTINHERIT); + xfs_trans_log_inode(tp, sc->tempip, XFS_ILOG_CORE); + + /* + * Mark our temporary file as private so that LSMs and the ACL code + * don't try to add their own metadata or reason about these files. + * The file should never be exposed to userspace. + */ + VFS_I(sc->tempip)->i_flags |= S_PRIVATE; + VFS_I(sc->tempip)->i_opflags &= ~IOP_XATTR; + + if (is_dir) { + error = xfs_dir_init(tp, sc->tempip, dp); + if (error) + goto out_trans_cancel; + } + + /* + * Attach the dquot(s) to the inodes and modify them incore. + * These ids of the inode couldn't have changed since the new + * inode has been locked ever since it was created. + */ + xfs_qm_vop_create_dqattach(tp, sc->tempip, udqp, gdqp, pdqp); + + /* + * Put our temp file on the unlinked list so it's purged automatically. + * All file-based metadata being reconstructed using this file must be + * atomically exchanged with the original file because the contents + * here will be purged when the inode is dropped or log recovery cleans + * out the unlinked list. + */ + error = xfs_iunlink(tp, sc->tempip); + if (error) + goto out_trans_cancel; + + error = xfs_trans_commit(tp); + if (error) + goto out_release_inode; + + trace_xrep_tempfile_create(sc); + + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); + + /* Finish setting up the incore / vfs context. */ + xfs_setup_iops(sc->tempip); + xfs_finish_inode_setup(sc->tempip); + + sc->temp_ilock_flags = 0; + return error; + +out_trans_cancel: + xfs_trans_cancel(tp); +out_release_inode: + /* + * Wait until after the current transaction is aborted to finish the + * setup of the inode and release the inode. This prevents recursive + * transactions and deadlocks from xfs_inactive. + */ + if (sc->tempip) { + xfs_finish_inode_setup(sc->tempip); + xchk_irele(sc, sc->tempip); + } +out_release_dquots: + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); + + return error; +} + +/* Take IOLOCK_EXCL on the temporary file, maybe. */ +bool +xrep_tempfile_iolock_nowait( + struct xfs_scrub *sc) +{ + if (xfs_ilock_nowait(sc->tempip, XFS_IOLOCK_EXCL)) { + sc->temp_ilock_flags |= XFS_IOLOCK_EXCL; + return true; + } + + return false; +} + +/* + * Take the temporary file's IOLOCK while holding a different inode's IOLOCK. + * In theory nobody else should hold the tempfile's IOLOCK, but we use trylock + * to avoid deadlocks and lockdep complaints. + */ +int +xrep_tempfile_iolock_polled( + struct xfs_scrub *sc) +{ + int error = 0; + + while (!xrep_tempfile_iolock_nowait(sc)) { + if (xchk_should_terminate(sc, &error)) + return error; + delay(1); + } + + return 0; +} + +/* Release IOLOCK_EXCL on the temporary file. */ +void +xrep_tempfile_iounlock( + struct xfs_scrub *sc) +{ + xfs_iunlock(sc->tempip, XFS_IOLOCK_EXCL); + sc->temp_ilock_flags &= ~XFS_IOLOCK_EXCL; +} + +/* Prepare the temporary file for metadata updates by grabbing ILOCK_EXCL. */ +void +xrep_tempfile_ilock( + struct xfs_scrub *sc) +{ + sc->temp_ilock_flags |= XFS_ILOCK_EXCL; + xfs_ilock(sc->tempip, XFS_ILOCK_EXCL); +} + +/* Try to grab ILOCK_EXCL on the temporary file. */ +bool +xrep_tempfile_ilock_nowait( + struct xfs_scrub *sc) +{ + if (xfs_ilock_nowait(sc->tempip, XFS_ILOCK_EXCL)) { + sc->temp_ilock_flags |= XFS_ILOCK_EXCL; + return true; + } + + return false; +} + +/* Unlock ILOCK_EXCL on the temporary file after an update. */ +void +xrep_tempfile_iunlock( + struct xfs_scrub *sc) +{ + xfs_iunlock(sc->tempip, XFS_ILOCK_EXCL); + sc->temp_ilock_flags &= ~XFS_ILOCK_EXCL; +} + +/* Release the temporary file. */ +void +xrep_tempfile_rele( + struct xfs_scrub *sc) +{ + if (!sc->tempip) + return; + + if (sc->temp_ilock_flags) { + xfs_iunlock(sc->tempip, sc->temp_ilock_flags); + sc->temp_ilock_flags = 0; + } + + xchk_irele(sc, sc->tempip); + sc->tempip = NULL; +} diff --git a/fs/xfs/scrub/tempfile.h b/fs/xfs/scrub/tempfile.h new file mode 100644 index 000000000000..e165e0a3faf6 --- /dev/null +++ b/fs/xfs/scrub/tempfile.h @@ -0,0 +1,28 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2021-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_TEMPFILE_H__ +#define __XFS_SCRUB_TEMPFILE_H__ + +#ifdef CONFIG_XFS_ONLINE_REPAIR +int xrep_tempfile_create(struct xfs_scrub *sc, uint16_t mode); +void xrep_tempfile_rele(struct xfs_scrub *sc); + +bool xrep_tempfile_iolock_nowait(struct xfs_scrub *sc); +int xrep_tempfile_iolock_polled(struct xfs_scrub *sc); +void xrep_tempfile_iounlock(struct xfs_scrub *sc); + +void xrep_tempfile_ilock(struct xfs_scrub *sc); +bool xrep_tempfile_ilock_nowait(struct xfs_scrub *sc); +void xrep_tempfile_iunlock(struct xfs_scrub *sc); +#else +static inline void xrep_tempfile_iolock_both(struct xfs_scrub *sc) +{ + xchk_ilock(sc, XFS_IOLOCK_EXCL); +} +# define xrep_tempfile_rele(sc) +#endif /* CONFIG_XFS_ONLINE_REPAIR */ + +#endif /* __XFS_SCRUB_TEMPFILE_H__ */ diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index b1c7c79760d4..020b029b7988 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2279,6 +2279,39 @@ TRACE_EVENT(xrep_rmap_live_update, __entry->flags) ); +TRACE_EVENT(xrep_tempfile_create, + TP_PROTO(struct xfs_scrub *sc), + TP_ARGS(sc), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(unsigned int, type) + __field(xfs_agnumber_t, agno) + __field(xfs_ino_t, inum) + __field(unsigned int, gen) + __field(unsigned int, flags) + __field(xfs_ino_t, temp_inum) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->ino = sc->file ? XFS_I(file_inode(sc->file))->i_ino : 0; + __entry->type = sc->sm->sm_type; + __entry->agno = sc->sm->sm_agno; + __entry->inum = sc->sm->sm_ino; + __entry->gen = sc->sm->sm_gen; + __entry->flags = sc->sm->sm_flags; + __entry->temp_inum = sc->tempip->i_ino; + ), + TP_printk("dev %d:%d ino 0x%llx type %s inum 0x%llx gen 0x%x flags 0x%x temp_inum 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_symbolic(__entry->type, XFS_SCRUB_TYPE_STRINGS), + __entry->inum, + __entry->gen, + __entry->flags, + __entry->temp_inum) +); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 492dae0efad2..ac92c0525d9b 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -42,7 +42,6 @@ struct kmem_cache *xfs_inode_cache; -STATIC int xfs_iunlink(struct xfs_trans *, struct xfs_inode *); STATIC int xfs_iunlink_remove(struct xfs_trans *tp, struct xfs_perag *pag, struct xfs_inode *); @@ -2151,7 +2150,7 @@ xfs_iunlink_insert_inode( * We place the on-disk inode on a list in the AGI. It will be pulled from this * list when the inode is freed. */ -STATIC int +int xfs_iunlink( struct xfs_trans *tp, struct xfs_inode *ip) diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index f559e68ee707..596eec715675 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -616,6 +616,8 @@ extern struct kmem_cache *xfs_inode_cache; bool xfs_inode_needs_inactive(struct xfs_inode *ip); +int xfs_iunlink(struct xfs_trans *tp, struct xfs_inode *ip); + void xfs_end_io(struct work_struct *work); int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2); -- cgit v1.2.3-59-g8ed1b From 20a3c1ecc35d2aa9ee1276e1cfd485689f29b662 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:29 -0700 Subject: xfs: refactor live buffer invalidation for repairs In an upcoming patch, we will need to be able to look for xfs_buf objects caching file-based metadata blocks without needing to walk the (possibly corrupt) structures to find all the buffers. Repair already has most of the code needed to scan the buffer cache, so hoist these utility functions. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/reap.c | 73 +++++++++++++++++++++++++++++++++++++---------------- fs/xfs/scrub/reap.h | 20 +++++++++++++++ 2 files changed, 71 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/reap.c b/fs/xfs/scrub/reap.c index 0252a3b5b65a..7ae6253395e7 100644 --- a/fs/xfs/scrub/reap.c +++ b/fs/xfs/scrub/reap.c @@ -211,6 +211,48 @@ static inline void xreap_defer_finish_reset(struct xreap_state *rs) rs->force_roll = false; } +/* + * Compute the maximum length of a buffer cache scan (in units of sectors), + * given a quantity of fs blocks. + */ +xfs_daddr_t +xrep_bufscan_max_sectors( + struct xfs_mount *mp, + xfs_extlen_t fsblocks) +{ + int max_fsbs; + + /* Remote xattr values are the largest buffers that we support. */ + max_fsbs = xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); + + return XFS_FSB_TO_BB(mp, min_t(xfs_extlen_t, fsblocks, max_fsbs)); +} + +/* + * Return an incore buffer from a sector scan, or NULL if there are no buffers + * left to return. + */ +struct xfs_buf * +xrep_bufscan_advance( + struct xfs_mount *mp, + struct xrep_bufscan *scan) +{ + scan->__sector_count += scan->daddr_step; + while (scan->__sector_count <= scan->max_sectors) { + struct xfs_buf *bp = NULL; + int error; + + error = xfs_buf_incore(mp->m_ddev_targp, scan->daddr, + scan->__sector_count, XBF_LIVESCAN, &bp); + if (!error) + return bp; + + scan->__sector_count += scan->daddr_step; + } + + return NULL; +} + /* Try to invalidate the incore buffers for an extent that we're freeing. */ STATIC void xreap_agextent_binval( @@ -241,28 +283,15 @@ xreap_agextent_binval( * of any plausible size. */ while (bno < agbno_next) { - xfs_agblock_t fsbcount; - xfs_agblock_t max_fsbs; - - /* - * Max buffer size is the max remote xattr buffer size, which - * is one fs block larger than 64k. - */ - max_fsbs = min_t(xfs_agblock_t, agbno_next - bno, - xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX)); - - for (fsbcount = 1; fsbcount <= max_fsbs; fsbcount++) { - struct xfs_buf *bp = NULL; - xfs_daddr_t daddr; - int error; - - daddr = XFS_AGB_TO_DADDR(mp, agno, bno); - error = xfs_buf_incore(mp->m_ddev_targp, daddr, - XFS_FSB_TO_BB(mp, fsbcount), - XBF_LIVESCAN, &bp); - if (error) - continue; - + struct xrep_bufscan scan = { + .daddr = XFS_AGB_TO_DADDR(mp, agno, bno), + .max_sectors = xrep_bufscan_max_sectors(mp, + agbno_next - bno), + .daddr_step = XFS_FSB_TO_BB(mp, 1), + }; + struct xfs_buf *bp; + + while ((bp = xrep_bufscan_advance(mp, &scan)) != NULL) { xfs_trans_bjoin(sc->tp, bp); xfs_trans_binval(sc->tp, bp); rs->invalidated++; diff --git a/fs/xfs/scrub/reap.h b/fs/xfs/scrub/reap.h index 0b69f16dd98f..bb09e21fcb17 100644 --- a/fs/xfs/scrub/reap.h +++ b/fs/xfs/scrub/reap.h @@ -14,4 +14,24 @@ int xrep_reap_agblocks(struct xfs_scrub *sc, struct xagb_bitmap *bitmap, int xrep_reap_fsblocks(struct xfs_scrub *sc, struct xfsb_bitmap *bitmap, const struct xfs_owner_info *oinfo); +/* Buffer cache scan context. */ +struct xrep_bufscan { + /* Disk address for the buffers we want to scan. */ + xfs_daddr_t daddr; + + /* Maximum number of sectors to scan. */ + xfs_daddr_t max_sectors; + + /* Each round, increment the search length by this number of sectors. */ + xfs_daddr_t daddr_step; + + /* Internal scan state; initialize to zero. */ + xfs_daddr_t __sector_count; +}; + +xfs_daddr_t xrep_bufscan_max_sectors(struct xfs_mount *mp, + xfs_extlen_t fsblocks); +struct xfs_buf *xrep_bufscan_advance(struct xfs_mount *mp, + struct xrep_bufscan *scan); + #endif /* __XFS_SCRUB_REAP_H__ */ -- cgit v1.2.3-59-g8ed1b From 5befb047b9f4de1747bf48c63cab997a97e0088b Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:30 -0700 Subject: xfs: add the ability to reap entire inode forks In preparation for supporting repair of indexed file-based metadata (such as realtime bitmaps, directories, and extended attribute data), add a function to reap the old blocks after a metadata repair finishes. IOWs, this is an elaborate bunmapi call that deals with crosslinked blocks by unmapping them without freeing them, and also scans for incore buffers to invalidate. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/reap.c | 372 +++++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/reap.h | 1 + fs/xfs/scrub/trace.h | 63 +++++++++ 3 files changed, 436 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/reap.c b/fs/xfs/scrub/reap.c index 7ae6253395e7..01ceaa4efa16 100644 --- a/fs/xfs/scrub/reap.c +++ b/fs/xfs/scrub/reap.c @@ -675,3 +675,375 @@ xrep_reap_fsblocks( return 0; } + +/* + * Metadata files are not supposed to share blocks with anything else. + * If blocks are shared, we remove the reverse mapping (thus reducing the + * crosslink factor); if blocks are not shared, we also need to free them. + * + * This first step determines the longest subset of the passed-in imap + * (starting at its beginning) that is either crosslinked or not crosslinked. + * The blockcount will be adjust down as needed. + */ +STATIC int +xreap_bmapi_select( + struct xfs_scrub *sc, + struct xfs_inode *ip, + int whichfork, + struct xfs_bmbt_irec *imap, + bool *crosslinked) +{ + struct xfs_owner_info oinfo; + struct xfs_btree_cur *cur; + xfs_filblks_t len = 1; + xfs_agblock_t bno; + xfs_agblock_t agbno; + xfs_agblock_t agbno_next; + int error; + + agbno = XFS_FSB_TO_AGBNO(sc->mp, imap->br_startblock); + agbno_next = agbno + imap->br_blockcount; + + cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, sc->sa.agf_bp, + sc->sa.pag); + + xfs_rmap_ino_owner(&oinfo, ip->i_ino, whichfork, imap->br_startoff); + error = xfs_rmap_has_other_keys(cur, agbno, 1, &oinfo, crosslinked); + if (error) + goto out_cur; + + bno = agbno + 1; + while (bno < agbno_next) { + bool also_crosslinked; + + oinfo.oi_offset++; + error = xfs_rmap_has_other_keys(cur, bno, 1, &oinfo, + &also_crosslinked); + if (error) + goto out_cur; + + if (also_crosslinked != *crosslinked) + break; + + len++; + bno++; + } + + imap->br_blockcount = len; + trace_xreap_bmapi_select(sc->sa.pag, agbno, len, *crosslinked); +out_cur: + xfs_btree_del_cursor(cur, error); + return error; +} + +/* + * Decide if this buffer can be joined to a transaction. This is true for most + * buffers, but there are two cases that we want to catch: large remote xattr + * value buffers are not logged and can overflow the buffer log item dirty + * bitmap size; and oversized cached buffers if things have really gone + * haywire. + */ +static inline bool +xreap_buf_loggable( + const struct xfs_buf *bp) +{ + int i; + + for (i = 0; i < bp->b_map_count; i++) { + int chunks; + int map_size; + + chunks = DIV_ROUND_UP(BBTOB(bp->b_maps[i].bm_len), + XFS_BLF_CHUNK); + map_size = DIV_ROUND_UP(chunks, NBWORD); + if (map_size > XFS_BLF_DATAMAP_SIZE) + return false; + } + + return true; +} + +/* + * Invalidate any buffers for this file mapping. The @imap blockcount may be + * adjusted downward if we need to roll the transaction. + */ +STATIC int +xreap_bmapi_binval( + struct xfs_scrub *sc, + struct xfs_inode *ip, + int whichfork, + struct xfs_bmbt_irec *imap) +{ + struct xfs_mount *mp = sc->mp; + struct xfs_perag *pag = sc->sa.pag; + int bmap_flags = xfs_bmapi_aflag(whichfork); + xfs_fileoff_t off; + xfs_fileoff_t max_off; + xfs_extlen_t scan_blocks; + xfs_agnumber_t agno = sc->sa.pag->pag_agno; + xfs_agblock_t bno; + xfs_agblock_t agbno; + xfs_agblock_t agbno_next; + unsigned int invalidated = 0; + int error; + + /* + * Avoid invalidating AG headers and post-EOFS blocks because we never + * own those. + */ + agbno = bno = XFS_FSB_TO_AGBNO(sc->mp, imap->br_startblock); + agbno_next = agbno + imap->br_blockcount; + if (!xfs_verify_agbno(pag, agbno) || + !xfs_verify_agbno(pag, agbno_next - 1)) + return 0; + + /* + * Buffers for file blocks can span multiple contiguous mappings. This + * means that for each block in the mapping, there could exist an + * xfs_buf indexed by that block with any length up to the maximum + * buffer size (remote xattr values) or to the next hole in the fork. + * To set up our binval scan, first we need to figure out the location + * of the next hole. + */ + off = imap->br_startoff + imap->br_blockcount; + max_off = off + xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); + while (off < max_off) { + struct xfs_bmbt_irec hmap; + int nhmaps = 1; + + error = xfs_bmapi_read(ip, off, max_off - off, &hmap, + &nhmaps, bmap_flags); + if (error) + return error; + if (nhmaps != 1 || hmap.br_startblock == DELAYSTARTBLOCK) { + ASSERT(0); + return -EFSCORRUPTED; + } + + if (!xfs_bmap_is_real_extent(&hmap)) + break; + + off = hmap.br_startoff + hmap.br_blockcount; + } + scan_blocks = off - imap->br_startoff; + + trace_xreap_bmapi_binval_scan(sc, imap, scan_blocks); + + /* + * If there are incore buffers for these blocks, invalidate them. If + * we can't (try)lock the buffer we assume it's owned by someone else + * and leave it alone. The buffer cache cannot detect aliasing, so + * employ nested loops to detect incore buffers of any plausible size. + */ + while (bno < agbno_next) { + struct xrep_bufscan scan = { + .daddr = XFS_AGB_TO_DADDR(mp, agno, bno), + .max_sectors = xrep_bufscan_max_sectors(mp, + scan_blocks), + .daddr_step = XFS_FSB_TO_BB(mp, 1), + }; + struct xfs_buf *bp; + + while ((bp = xrep_bufscan_advance(mp, &scan)) != NULL) { + if (xreap_buf_loggable(bp)) { + xfs_trans_bjoin(sc->tp, bp); + xfs_trans_binval(sc->tp, bp); + } else { + xfs_buf_stale(bp); + xfs_buf_relse(bp); + } + invalidated++; + + /* + * Stop invalidating if we've hit the limit; we should + * still have enough reservation left to free however + * much of the mapping we've seen so far. + */ + if (invalidated > XREAP_MAX_BINVAL) { + imap->br_blockcount = agbno_next - bno; + goto out; + } + } + + bno++; + scan_blocks--; + } + +out: + trace_xreap_bmapi_binval(sc->sa.pag, agbno, imap->br_blockcount); + return 0; +} + +/* + * Dispose of as much of the beginning of this file fork mapping as possible. + * The number of blocks disposed of is returned in @imap->br_blockcount. + */ +STATIC int +xrep_reap_bmapi_iter( + struct xfs_scrub *sc, + struct xfs_inode *ip, + int whichfork, + struct xfs_bmbt_irec *imap, + bool crosslinked) +{ + int error; + + if (crosslinked) { + /* + * If there are other rmappings, this block is cross linked and + * must not be freed. Remove the reverse mapping, leave the + * buffer cache in its possibly confused state, and move on. + * We don't want to risk discarding valid data buffers from + * anybody else who thinks they own the block, even though that + * runs the risk of stale buffer warnings in the future. + */ + trace_xreap_dispose_unmap_extent(sc->sa.pag, + XFS_FSB_TO_AGBNO(sc->mp, imap->br_startblock), + imap->br_blockcount); + + /* + * Schedule removal of the mapping from the fork. We use + * deferred log intents in this function to control the exact + * sequence of metadata updates. + */ + xfs_bmap_unmap_extent(sc->tp, ip, whichfork, imap); + xfs_trans_mod_dquot_byino(sc->tp, ip, XFS_TRANS_DQ_BCOUNT, + -(int64_t)imap->br_blockcount); + xfs_rmap_unmap_extent(sc->tp, ip, whichfork, imap); + return 0; + } + + /* + * If the block is not crosslinked, we can invalidate all the incore + * buffers for the extent, and then free the extent. This is a bit of + * a mess since we don't detect discontiguous buffers that are indexed + * by a block starting before the first block of the extent but overlap + * anyway. + */ + trace_xreap_dispose_free_extent(sc->sa.pag, + XFS_FSB_TO_AGBNO(sc->mp, imap->br_startblock), + imap->br_blockcount); + + /* + * Invalidate as many buffers as we can, starting at the beginning of + * this mapping. If this function sets blockcount to zero, the + * transaction is full of logged buffer invalidations, so we need to + * return early so that we can roll and retry. + */ + error = xreap_bmapi_binval(sc, ip, whichfork, imap); + if (error || imap->br_blockcount == 0) + return error; + + /* + * Schedule removal of the mapping from the fork. We use deferred log + * intents in this function to control the exact sequence of metadata + * updates. + */ + xfs_bmap_unmap_extent(sc->tp, ip, whichfork, imap); + xfs_trans_mod_dquot_byino(sc->tp, ip, XFS_TRANS_DQ_BCOUNT, + -(int64_t)imap->br_blockcount); + return xfs_free_extent_later(sc->tp, imap->br_startblock, + imap->br_blockcount, NULL, XFS_AG_RESV_NONE, true); +} + +/* + * Dispose of as much of this file extent as we can. Upon successful return, + * the imap will reflect the mapping that was removed from the fork. + */ +STATIC int +xreap_ifork_extent( + struct xfs_scrub *sc, + struct xfs_inode *ip, + int whichfork, + struct xfs_bmbt_irec *imap) +{ + xfs_agnumber_t agno; + bool crosslinked; + int error; + + ASSERT(sc->sa.pag == NULL); + + trace_xreap_ifork_extent(sc, ip, whichfork, imap); + + agno = XFS_FSB_TO_AGNO(sc->mp, imap->br_startblock); + sc->sa.pag = xfs_perag_get(sc->mp, agno); + if (!sc->sa.pag) + return -EFSCORRUPTED; + + error = xfs_alloc_read_agf(sc->sa.pag, sc->tp, 0, &sc->sa.agf_bp); + if (error) + goto out_pag; + + /* + * Decide the fate of the blocks at the beginning of the mapping, then + * update the mapping to use it with the unmap calls. + */ + error = xreap_bmapi_select(sc, ip, whichfork, imap, &crosslinked); + if (error) + goto out_agf; + + error = xrep_reap_bmapi_iter(sc, ip, whichfork, imap, crosslinked); + if (error) + goto out_agf; + +out_agf: + xfs_trans_brelse(sc->tp, sc->sa.agf_bp); + sc->sa.agf_bp = NULL; +out_pag: + xfs_perag_put(sc->sa.pag); + sc->sa.pag = NULL; + return error; +} + +/* + * Dispose of each block mapped to the given fork of the given file. Callers + * must hold ILOCK_EXCL, and ip can only be sc->ip or sc->tempip. The fork + * must not have any delalloc reservations. + */ +int +xrep_reap_ifork( + struct xfs_scrub *sc, + struct xfs_inode *ip, + int whichfork) +{ + xfs_fileoff_t off = 0; + int bmap_flags = xfs_bmapi_aflag(whichfork); + int error; + + ASSERT(xfs_has_rmapbt(sc->mp)); + ASSERT(ip == sc->ip || ip == sc->tempip); + ASSERT(whichfork == XFS_ATTR_FORK || !XFS_IS_REALTIME_INODE(ip)); + + while (off < XFS_MAX_FILEOFF) { + struct xfs_bmbt_irec imap; + int nimaps = 1; + + /* Read the next extent, skip past holes and delalloc. */ + error = xfs_bmapi_read(ip, off, XFS_MAX_FILEOFF - off, &imap, + &nimaps, bmap_flags); + if (error) + return error; + if (nimaps != 1 || imap.br_startblock == DELAYSTARTBLOCK) { + ASSERT(0); + return -EFSCORRUPTED; + } + + /* + * If this is a real space mapping, reap as much of it as we + * can in a single transaction. + */ + if (xfs_bmap_is_real_extent(&imap)) { + error = xreap_ifork_extent(sc, ip, whichfork, &imap); + if (error) + return error; + + error = xfs_defer_finish(&sc->tp); + if (error) + return error; + } + + off = imap.br_startoff + imap.br_blockcount; + } + + return 0; +} diff --git a/fs/xfs/scrub/reap.h b/fs/xfs/scrub/reap.h index bb09e21fcb17..3f2f1775e29d 100644 --- a/fs/xfs/scrub/reap.h +++ b/fs/xfs/scrub/reap.h @@ -13,6 +13,7 @@ int xrep_reap_agblocks(struct xfs_scrub *sc, struct xagb_bitmap *bitmap, const struct xfs_owner_info *oinfo, enum xfs_ag_resv_type type); int xrep_reap_fsblocks(struct xfs_scrub *sc, struct xfsb_bitmap *bitmap, const struct xfs_owner_info *oinfo); +int xrep_reap_ifork(struct xfs_scrub *sc, struct xfs_inode *ip, int whichfork); /* Buffer cache scan context. */ struct xrep_bufscan { diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 020b029b7988..cbd70ecd3011 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -1539,6 +1539,7 @@ DEFINE_EVENT(xrep_extent_class, name, \ DEFINE_REPAIR_EXTENT_EVENT(xreap_dispose_unmap_extent); DEFINE_REPAIR_EXTENT_EVENT(xreap_dispose_free_extent); DEFINE_REPAIR_EXTENT_EVENT(xreap_agextent_binval); +DEFINE_REPAIR_EXTENT_EVENT(xreap_bmapi_binval); DEFINE_REPAIR_EXTENT_EVENT(xrep_agfl_insert); DECLARE_EVENT_CLASS(xrep_reap_find_class, @@ -1572,6 +1573,7 @@ DEFINE_EVENT(xrep_reap_find_class, name, \ bool crosslinked), \ TP_ARGS(pag, agbno, len, crosslinked)) DEFINE_REPAIR_REAP_FIND_EVENT(xreap_agextent_select); +DEFINE_REPAIR_REAP_FIND_EVENT(xreap_bmapi_select); DECLARE_EVENT_CLASS(xrep_rmap_class, TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, @@ -2312,6 +2314,67 @@ TRACE_EVENT(xrep_tempfile_create, __entry->temp_inum) ); +TRACE_EVENT(xreap_ifork_extent, + TP_PROTO(struct xfs_scrub *sc, struct xfs_inode *ip, int whichfork, + const struct xfs_bmbt_irec *irec), + TP_ARGS(sc, ip, whichfork, irec), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(int, whichfork) + __field(xfs_fileoff_t, fileoff) + __field(xfs_filblks_t, len) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(int, state) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->whichfork = whichfork; + __entry->fileoff = irec->br_startoff; + __entry->len = irec->br_blockcount; + __entry->agno = XFS_FSB_TO_AGNO(sc->mp, irec->br_startblock); + __entry->agbno = XFS_FSB_TO_AGBNO(sc->mp, irec->br_startblock); + __entry->state = irec->br_state; + ), + TP_printk("dev %d:%d ip 0x%llx whichfork %s agno 0x%x agbno 0x%x fileoff 0x%llx fsbcount 0x%llx state 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_symbolic(__entry->whichfork, XFS_WHICHFORK_STRINGS), + __entry->agno, + __entry->agbno, + __entry->fileoff, + __entry->len, + __entry->state) +); + +TRACE_EVENT(xreap_bmapi_binval_scan, + TP_PROTO(struct xfs_scrub *sc, const struct xfs_bmbt_irec *irec, + xfs_extlen_t scan_blocks), + TP_ARGS(sc, irec, scan_blocks), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_filblks_t, len) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, scan_blocks) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->len = irec->br_blockcount; + __entry->agno = XFS_FSB_TO_AGNO(sc->mp, irec->br_startblock); + __entry->agbno = XFS_FSB_TO_AGBNO(sc->mp, irec->br_startblock); + __entry->scan_blocks = scan_blocks; + ), + TP_printk("dev %d:%d agno 0x%x agbno 0x%x fsbcount 0x%llx scan_blocks 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->len, + __entry->scan_blocks) +); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ -- cgit v1.2.3-59-g8ed1b From e81ce4241318d5ec943b4fe205953498a66ca071 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:31 -0700 Subject: xfs: support preallocating and copying content into temporary files Create the routines we need to preallocate space in a temporary ondisk file and then copy the contents of an xfile into the tempfile. The upcoming rtsummary repair feature will construct the contents of a realtime summary file in memory, after which it will want to copy all that into the ondisk temporary file before atomically committing the new rtsummary contents. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/tempfile.c | 197 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/tempfile.h | 15 ++++ fs/xfs/scrub/trace.h | 39 ++++++++++ 3 files changed, 251 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index 68d245749bc1..83e683e16561 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -14,14 +14,18 @@ #include "xfs_inode.h" #include "xfs_ialloc.h" #include "xfs_quota.h" +#include "xfs_bmap.h" #include "xfs_bmap_btree.h" #include "xfs_trans_space.h" #include "xfs_dir2.h" #include "xfs_exchrange.h" +#include "xfs_defer.h" #include "scrub/scrub.h" #include "scrub/common.h" +#include "scrub/repair.h" #include "scrub/trace.h" #include "scrub/tempfile.h" +#include "scrub/xfile.h" /* * Create a temporary file for reconstructing metadata, with the intention of @@ -249,3 +253,196 @@ xrep_tempfile_rele( xchk_irele(sc, sc->tempip); sc->tempip = NULL; } + +/* + * Make sure that the given range of the data fork of the temporary file is + * mapped to written blocks. The caller must ensure that both inodes are + * joined to the transaction. + */ +int +xrep_tempfile_prealloc( + struct xfs_scrub *sc, + xfs_fileoff_t off, + xfs_filblks_t len) +{ + struct xfs_bmbt_irec map; + xfs_fileoff_t end = off + len; + int error; + + ASSERT(sc->tempip != NULL); + ASSERT(!XFS_NOT_DQATTACHED(sc->mp, sc->tempip)); + + for (; off < end; off = map.br_startoff + map.br_blockcount) { + int nmaps = 1; + + /* + * If we have a real extent mapping this block then we're + * in ok shape. + */ + error = xfs_bmapi_read(sc->tempip, off, end - off, &map, &nmaps, + XFS_DATA_FORK); + if (error) + return error; + if (nmaps == 0) { + ASSERT(nmaps != 0); + return -EFSCORRUPTED; + } + + if (xfs_bmap_is_written_extent(&map)) + continue; + + /* + * If we find a delalloc reservation then something is very + * very wrong. Bail out. + */ + if (map.br_startblock == DELAYSTARTBLOCK) + return -EFSCORRUPTED; + + /* + * Make sure this block has a real zeroed extent allocated to + * it. + */ + nmaps = 1; + error = xfs_bmapi_write(sc->tp, sc->tempip, off, end - off, + XFS_BMAPI_CONVERT | XFS_BMAPI_ZERO, 0, &map, + &nmaps); + if (error) + return error; + if (nmaps != 1) + return -EFSCORRUPTED; + + trace_xrep_tempfile_prealloc(sc, XFS_DATA_FORK, &map); + + /* Commit new extent and all deferred work. */ + error = xfs_defer_finish(&sc->tp); + if (error) + return error; + } + + return 0; +} + +/* + * Write data to each block of a file. The given range of the tempfile's data + * fork must already be populated with written extents. + */ +int +xrep_tempfile_copyin( + struct xfs_scrub *sc, + xfs_fileoff_t off, + xfs_filblks_t len, + xrep_tempfile_copyin_fn prep_fn, + void *data) +{ + LIST_HEAD(buffers_list); + struct xfs_mount *mp = sc->mp; + struct xfs_buf *bp; + xfs_fileoff_t flush_mask; + xfs_fileoff_t end = off + len; + loff_t pos = XFS_FSB_TO_B(mp, off); + int error = 0; + + ASSERT(S_ISREG(VFS_I(sc->tempip)->i_mode)); + + /* Flush buffers to disk every 512K */ + flush_mask = XFS_B_TO_FSBT(mp, (1U << 19)) - 1; + + for (; off < end; off++, pos += mp->m_sb.sb_blocksize) { + struct xfs_bmbt_irec map; + int nmaps = 1; + + /* Read block mapping for this file block. */ + error = xfs_bmapi_read(sc->tempip, off, 1, &map, &nmaps, 0); + if (error) + goto out_err; + if (nmaps == 0 || !xfs_bmap_is_written_extent(&map)) { + error = -EFSCORRUPTED; + goto out_err; + } + + /* Get the metadata buffer for this offset in the file. */ + error = xfs_trans_get_buf(sc->tp, mp->m_ddev_targp, + XFS_FSB_TO_DADDR(mp, map.br_startblock), + mp->m_bsize, 0, &bp); + if (error) + goto out_err; + + trace_xrep_tempfile_copyin(sc, XFS_DATA_FORK, &map); + + /* Read in a block's worth of data from the xfile. */ + error = prep_fn(sc, bp, data); + if (error) { + xfs_trans_brelse(sc->tp, bp); + goto out_err; + } + + /* Queue buffer, and flush if we have too much dirty data. */ + xfs_buf_delwri_queue_here(bp, &buffers_list); + xfs_trans_brelse(sc->tp, bp); + + if (!(off & flush_mask)) { + error = xfs_buf_delwri_submit(&buffers_list); + if (error) + goto out_err; + } + } + + /* + * Write the new blocks to disk. If the ordered list isn't empty after + * that, then something went wrong and we have to fail. This should + * never happen, but we'll check anyway. + */ + error = xfs_buf_delwri_submit(&buffers_list); + if (error) + goto out_err; + + if (!list_empty(&buffers_list)) { + ASSERT(list_empty(&buffers_list)); + error = -EIO; + goto out_err; + } + + return 0; + +out_err: + xfs_buf_delwri_cancel(&buffers_list); + return error; +} + +/* + * Set the temporary file's size. Caller must join the tempfile to the scrub + * transaction and is responsible for adjusting block mappings as needed. + */ +int +xrep_tempfile_set_isize( + struct xfs_scrub *sc, + unsigned long long isize) +{ + if (sc->tempip->i_disk_size == isize) + return 0; + + sc->tempip->i_disk_size = isize; + i_size_write(VFS_I(sc->tempip), isize); + return xrep_tempfile_roll_trans(sc); +} + +/* + * Roll a repair transaction involving the temporary file. Caller must join + * both the temporary file and the file being scrubbed to the transaction. + * This function return with both inodes joined to a new scrub transaction, + * or the usual negative errno. + */ +int +xrep_tempfile_roll_trans( + struct xfs_scrub *sc) +{ + int error; + + xfs_trans_log_inode(sc->tp, sc->tempip, XFS_ILOG_CORE); + error = xrep_roll_trans(sc); + if (error) + return error; + + xfs_trans_ijoin(sc->tp, sc->tempip, 0); + return 0; +} diff --git a/fs/xfs/scrub/tempfile.h b/fs/xfs/scrub/tempfile.h index e165e0a3faf6..7980f9c4de55 100644 --- a/fs/xfs/scrub/tempfile.h +++ b/fs/xfs/scrub/tempfile.h @@ -17,6 +17,21 @@ void xrep_tempfile_iounlock(struct xfs_scrub *sc); void xrep_tempfile_ilock(struct xfs_scrub *sc); bool xrep_tempfile_ilock_nowait(struct xfs_scrub *sc); void xrep_tempfile_iunlock(struct xfs_scrub *sc); + +int xrep_tempfile_prealloc(struct xfs_scrub *sc, xfs_fileoff_t off, + xfs_filblks_t len); + +enum xfs_blft; + +typedef int (*xrep_tempfile_copyin_fn)(struct xfs_scrub *sc, + struct xfs_buf *bp, void *data); + +int xrep_tempfile_copyin(struct xfs_scrub *sc, xfs_fileoff_t off, + xfs_filblks_t len, xrep_tempfile_copyin_fn fn, void *data); + +int xrep_tempfile_set_isize(struct xfs_scrub *sc, unsigned long long isize); + +int xrep_tempfile_roll_trans(struct xfs_scrub *sc); #else static inline void xrep_tempfile_iolock_both(struct xfs_scrub *sc) { diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index cbd70ecd3011..ae90731bf6ad 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2314,6 +2314,45 @@ TRACE_EVENT(xrep_tempfile_create, __entry->temp_inum) ); +DECLARE_EVENT_CLASS(xrep_tempfile_class, + TP_PROTO(struct xfs_scrub *sc, int whichfork, + struct xfs_bmbt_irec *irec), + TP_ARGS(sc, whichfork, irec), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(int, whichfork) + __field(xfs_fileoff_t, lblk) + __field(xfs_filblks_t, len) + __field(xfs_fsblock_t, pblk) + __field(int, state) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->ino = sc->tempip->i_ino; + __entry->whichfork = whichfork; + __entry->lblk = irec->br_startoff; + __entry->len = irec->br_blockcount; + __entry->pblk = irec->br_startblock; + __entry->state = irec->br_state; + ), + TP_printk("dev %d:%d ino 0x%llx whichfork %s fileoff 0x%llx fsbcount 0x%llx startblock 0x%llx state %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_symbolic(__entry->whichfork, XFS_WHICHFORK_STRINGS), + __entry->lblk, + __entry->len, + __entry->pblk, + __entry->state) +); +#define DEFINE_XREP_TEMPFILE_EVENT(name) \ +DEFINE_EVENT(xrep_tempfile_class, name, \ + TP_PROTO(struct xfs_scrub *sc, int whichfork, \ + struct xfs_bmbt_irec *irec), \ + TP_ARGS(sc, whichfork, irec)) +DEFINE_XREP_TEMPFILE_EVENT(xrep_tempfile_prealloc); +DEFINE_XREP_TEMPFILE_EVENT(xrep_tempfile_copyin); + TRACE_EVENT(xreap_ifork_extent, TP_PROTO(struct xfs_scrub *sc, struct xfs_inode *ip, int whichfork, const struct xfs_bmbt_irec *irec), -- cgit v1.2.3-59-g8ed1b From 56596d8bffd2f3c80d9844835e6c118d89b67ca9 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:32 -0700 Subject: xfs: teach the tempfile to set up atomic file content exchanges Create some new routines to exchange the contents of a temporary file created to stage a repair with another ondisk file. This will be used by the realtime summary repair function to commit atomically the new rtsummary data, which will be staged in the tempfile. The rest of XFS coordinates access to the realtime metadata inodes solely through the ILOCK. For repair to hold its exclusive access to the realtime summary file, it has to allocate a single large transaction and roll it repeatedly throughout the repair while holding the ILOCK. In turn, this means that for now there's only a partial file mapping exchange implementation for the temporary file because we can only work within an existing transaction. For now, the only tempswap functions needed here are to estimate the resource requirements of the exchange, reserve more space/quota to an existing transaction, and kick off the actual exchange. The rest will be added in a later patch in preparation for repairing xattrs and directories. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/scrub.c | 8 +- fs/xfs/scrub/scrub.h | 7 ++ fs/xfs/scrub/tempexch.h | 21 ++++++ fs/xfs/scrub/tempfile.c | 191 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/trace.h | 1 + 5 files changed, 225 insertions(+), 3 deletions(-) create mode 100644 fs/xfs/scrub/tempexch.h (limited to 'fs') diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index d9012e9a6afd..ff156edf49a0 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -149,14 +149,15 @@ xchk_probe( /* Scrub setup and teardown */ +#define FSGATES_MASK (XCHK_FSGATES_ALL | XREP_FSGATES_ALL) static inline void xchk_fsgates_disable( struct xfs_scrub *sc) { - if (!(sc->flags & XCHK_FSGATES_ALL)) + if (!(sc->flags & FSGATES_MASK)) return; - trace_xchk_fsgates_disable(sc, sc->flags & XCHK_FSGATES_ALL); + trace_xchk_fsgates_disable(sc, sc->flags & FSGATES_MASK); if (sc->flags & XCHK_FSGATES_DRAIN) xfs_drain_wait_disable(); @@ -170,8 +171,9 @@ xchk_fsgates_disable( if (sc->flags & XCHK_FSGATES_RMAP) xfs_rmap_hook_disable(); - sc->flags &= ~XCHK_FSGATES_ALL; + sc->flags &= ~FSGATES_MASK; } +#undef FSGATES_MASK /* Free all the resources and finish the transactions. */ STATIC int diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index e37d8599718e..d38f0b30416c 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -131,6 +131,7 @@ struct xfs_scrub { #define XCHK_FSGATES_QUOTA (1U << 4) /* quota live update enabled */ #define XCHK_FSGATES_DIRENTS (1U << 5) /* directory live update enabled */ #define XCHK_FSGATES_RMAP (1U << 6) /* rmapbt live update enabled */ +#define XREP_FSGATES_EXCHANGE_RANGE (1U << 29) /* uses file content exchange */ #define XREP_RESET_PERAG_RESV (1U << 30) /* must reset AG space reservation */ #define XREP_ALREADY_FIXED (1U << 31) /* checking our repair work */ @@ -145,6 +146,12 @@ struct xfs_scrub { XCHK_FSGATES_DIRENTS | \ XCHK_FSGATES_RMAP) +/* + * The sole XREP_FSGATES* flag reflects a log intent item that is protected + * by a log-incompat feature flag. No code patching in use here. + */ +#define XREP_FSGATES_ALL (XREP_FSGATES_EXCHANGE_RANGE) + /* Metadata scrubbers */ int xchk_tester(struct xfs_scrub *sc); int xchk_superblock(struct xfs_scrub *sc); diff --git a/fs/xfs/scrub/tempexch.h b/fs/xfs/scrub/tempexch.h new file mode 100644 index 000000000000..98222b684b6a --- /dev/null +++ b/fs/xfs/scrub/tempexch.h @@ -0,0 +1,21 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2022-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_TEMPEXCH_H__ +#define __XFS_SCRUB_TEMPEXCH_H__ + +#ifdef CONFIG_XFS_ONLINE_REPAIR +struct xrep_tempexch { + struct xfs_exchmaps_req req; +}; + +int xrep_tempexch_enable(struct xfs_scrub *sc); +int xrep_tempexch_trans_reserve(struct xfs_scrub *sc, int whichfork, + struct xrep_tempexch *ti); + +int xrep_tempexch_contents(struct xfs_scrub *sc, struct xrep_tempexch *ti); +#endif /* CONFIG_XFS_ONLINE_REPAIR */ + +#endif /* __XFS_SCRUB_TEMPEXCH_H__ */ diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index 83e683e16561..7791336ca820 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -19,12 +19,14 @@ #include "xfs_trans_space.h" #include "xfs_dir2.h" #include "xfs_exchrange.h" +#include "xfs_exchmaps.h" #include "xfs_defer.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/repair.h" #include "scrub/trace.h" #include "scrub/tempfile.h" +#include "scrub/tempexch.h" #include "scrub/xfile.h" /* @@ -446,3 +448,192 @@ xrep_tempfile_roll_trans( xfs_trans_ijoin(sc->tp, sc->tempip, 0); return 0; } + +/* Enable file content exchanges. */ +int +xrep_tempexch_enable( + struct xfs_scrub *sc) +{ + if (sc->flags & XREP_FSGATES_EXCHANGE_RANGE) + return 0; + + if (!xfs_has_exchange_range(sc->mp)) + return -EOPNOTSUPP; + + trace_xchk_fsgates_enable(sc, XREP_FSGATES_EXCHANGE_RANGE); + + sc->flags |= XREP_FSGATES_EXCHANGE_RANGE; + return 0; +} + +/* + * Fill out the mapping exchange request in preparation for atomically + * committing the contents of a metadata file that we've rebuilt in the temp + * file. + */ +STATIC int +xrep_tempexch_prep_request( + struct xfs_scrub *sc, + int whichfork, + struct xrep_tempexch *tx) +{ + struct xfs_exchmaps_req *req = &tx->req; + + memset(tx, 0, sizeof(struct xrep_tempexch)); + + /* COW forks don't exist on disk. */ + if (whichfork == XFS_COW_FORK) { + ASSERT(0); + return -EINVAL; + } + + /* Both files should have the relevant forks. */ + if (!xfs_ifork_ptr(sc->ip, whichfork) || + !xfs_ifork_ptr(sc->tempip, whichfork)) { + ASSERT(xfs_ifork_ptr(sc->ip, whichfork) != NULL); + ASSERT(xfs_ifork_ptr(sc->tempip, whichfork) != NULL); + return -EINVAL; + } + + /* Exchange all mappings in both forks. */ + req->ip1 = sc->tempip; + req->ip2 = sc->ip; + req->startoff1 = 0; + req->startoff2 = 0; + switch (whichfork) { + case XFS_ATTR_FORK: + req->flags |= XFS_EXCHMAPS_ATTR_FORK; + break; + case XFS_DATA_FORK: + /* Always exchange sizes when exchanging data fork mappings. */ + req->flags |= XFS_EXCHMAPS_SET_SIZES; + break; + } + req->blockcount = XFS_MAX_FILEOFF; + + return 0; +} + +/* + * Obtain a quota reservation to make sure we don't hit EDQUOT. We can skip + * this if quota enforcement is disabled or if both inodes' dquots are the + * same. The qretry structure must be initialized to zeroes before the first + * call to this function. + */ +STATIC int +xrep_tempexch_reserve_quota( + struct xfs_scrub *sc, + const struct xrep_tempexch *tx) +{ + struct xfs_trans *tp = sc->tp; + const struct xfs_exchmaps_req *req = &tx->req; + int64_t ddelta, rdelta; + int error; + + /* + * Don't bother with a quota reservation if we're not enforcing them + * or the two inodes have the same dquots. + */ + if (!XFS_IS_QUOTA_ON(tp->t_mountp) || req->ip1 == req->ip2 || + (req->ip1->i_udquot == req->ip2->i_udquot && + req->ip1->i_gdquot == req->ip2->i_gdquot && + req->ip1->i_pdquot == req->ip2->i_pdquot)) + return 0; + + /* + * Quota reservation for each file comes from two sources. First, we + * need to account for any net gain in mapped blocks during the + * exchange. Second, we need reservation for the gross gain in mapped + * blocks so that we don't trip over any quota block reservation + * assertions. We must reserve the gross gain because the quota code + * subtracts from bcount the number of blocks that we unmap; it does + * not add that quantity back to the quota block reservation. + */ + ddelta = max_t(int64_t, 0, req->ip2_bcount - req->ip1_bcount); + rdelta = max_t(int64_t, 0, req->ip2_rtbcount - req->ip1_rtbcount); + error = xfs_trans_reserve_quota_nblks(tp, req->ip1, + ddelta + req->ip1_bcount, rdelta + req->ip1_rtbcount, + true); + if (error) + return error; + + ddelta = max_t(int64_t, 0, req->ip1_bcount - req->ip2_bcount); + rdelta = max_t(int64_t, 0, req->ip1_rtbcount - req->ip2_rtbcount); + return xfs_trans_reserve_quota_nblks(tp, req->ip2, + ddelta + req->ip2_bcount, rdelta + req->ip2_rtbcount, + true); +} + +/* + * Prepare an existing transaction for an atomic file contents exchange. + * + * This function fills out the mapping exchange request and resource estimation + * structures in preparation for exchanging the contents of a metadata file + * that has been rebuilt in the temp file. Next, it reserves space and quota + * for the transaction. + * + * The caller must hold ILOCK_EXCL of the scrub target file and the temporary + * file. The caller must join both inodes to the transaction with no unlock + * flags, and is responsible for dropping both ILOCKs when appropriate. Only + * use this when those ILOCKs cannot be dropped. + */ +int +xrep_tempexch_trans_reserve( + struct xfs_scrub *sc, + int whichfork, + struct xrep_tempexch *tx) +{ + int error; + + ASSERT(sc->tp != NULL); + xfs_assert_ilocked(sc->ip, XFS_ILOCK_EXCL); + xfs_assert_ilocked(sc->tempip, XFS_ILOCK_EXCL); + + error = xrep_tempexch_prep_request(sc, whichfork, tx); + if (error) + return error; + + error = xfs_exchmaps_estimate(&tx->req); + if (error) + return error; + + error = xfs_trans_reserve_more(sc->tp, tx->req.resblks, 0); + if (error) + return error; + + return xrep_tempexch_reserve_quota(sc, tx); +} + +/* + * Exchange file mappings (and hence file contents) between the file being + * repaired and the temporary file. Returns with both inodes locked and joined + * to a clean scrub transaction. + */ +int +xrep_tempexch_contents( + struct xfs_scrub *sc, + struct xrep_tempexch *tx) +{ + int error; + + ASSERT(sc->flags & XREP_FSGATES_EXCHANGE_RANGE); + + xfs_exchange_mappings(sc->tp, &tx->req); + error = xfs_defer_finish(&sc->tp); + if (error) + return error; + + /* + * If we exchanged the ondisk sizes of two metadata files, we must + * exchanged the incore sizes as well. + */ + if (tx->req.flags & XFS_EXCHMAPS_SET_SIZES) { + loff_t temp; + + temp = i_size_read(VFS_I(sc->ip)); + i_size_write(VFS_I(sc->ip), i_size_read(VFS_I(sc->tempip))); + i_size_write(VFS_I(sc->tempip), temp); + } + + return 0; +} diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index ae90731bf6ad..8d05f2adae3d 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -114,6 +114,7 @@ TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_HEALTHY); { XCHK_FSGATES_QUOTA, "fsgates_quota" }, \ { XCHK_FSGATES_DIRENTS, "fsgates_dirents" }, \ { XCHK_FSGATES_RMAP, "fsgates_rmap" }, \ + { XREP_FSGATES_EXCHANGE_RANGE, "fsgates_exchrange" }, \ { XREP_RESET_PERAG_RESV, "reset_perag_resv" }, \ { XREP_ALREADY_FIXED, "already_fixed" } -- cgit v1.2.3-59-g8ed1b From abf039e2e4afde98e448253f9a7ecc784a87924d Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:33 -0700 Subject: xfs: online repair of realtime summaries Repair the realtime summary data by constructing a new rtsummary file in the scrub temporary file, then atomically swapping the contents. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/common.c | 1 + fs/xfs/scrub/repair.h | 3 + fs/xfs/scrub/rtsummary.c | 33 ++++---- fs/xfs/scrub/rtsummary.h | 37 +++++++++ fs/xfs/scrub/rtsummary_repair.c | 177 ++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/scrub.c | 3 +- 7 files changed, 239 insertions(+), 16 deletions(-) create mode 100644 fs/xfs/scrub/rtsummary.h create mode 100644 fs/xfs/scrub/rtsummary_repair.c (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index ae8488ab4d6b..5e3ac7ec8fa5 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -212,6 +212,7 @@ xfs-y += $(addprefix scrub/, \ xfs-$(CONFIG_XFS_RT) += $(addprefix scrub/, \ rtbitmap_repair.o \ + rtsummary_repair.o \ ) xfs-$(CONFIG_XFS_QUOTA) += $(addprefix scrub/, \ diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index a27d33b6f464..a2da2bef509a 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -31,6 +31,7 @@ #include "xfs_ag.h" #include "xfs_error.h" #include "xfs_quota.h" +#include "xfs_exchmaps.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index ce082d941459..0e2b695ab8f6 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -126,8 +126,10 @@ int xrep_fscounters(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xrep_rtbitmap(struct xfs_scrub *sc); +int xrep_rtsummary(struct xfs_scrub *sc); #else # define xrep_rtbitmap xrep_notsupported +# define xrep_rtsummary xrep_notsupported #endif /* CONFIG_XFS_RT */ #ifdef CONFIG_XFS_QUOTA @@ -212,6 +214,7 @@ xrep_setup_nothing( #define xrep_quotacheck xrep_notsupported #define xrep_nlinks xrep_notsupported #define xrep_fscounters xrep_notsupported +#define xrep_rtsummary xrep_notsupported #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/rtsummary.c b/fs/xfs/scrub/rtsummary.c index 5055092bd9e8..3fee603f5244 100644 --- a/fs/xfs/scrub/rtsummary.c +++ b/fs/xfs/scrub/rtsummary.c @@ -17,10 +17,14 @@ #include "xfs_bit.h" #include "xfs_bmap.h" #include "xfs_sb.h" +#include "xfs_exchmaps.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" #include "scrub/xfile.h" +#include "scrub/repair.h" +#include "scrub/tempexch.h" +#include "scrub/rtsummary.h" /* * Realtime Summary @@ -32,18 +36,6 @@ * (potentially large) amount of data in pageable memory. */ -struct xchk_rtsummary { - struct xfs_rtalloc_args args; - - uint64_t rextents; - uint64_t rbmblocks; - uint64_t rsumsize; - unsigned int rsumlevels; - - /* Memory buffer for the summary comparison. */ - union xfs_suminfo_raw words[]; -}; - /* Set us up to check the rtsummary file. */ int xchk_setup_rtsummary( @@ -60,6 +52,12 @@ xchk_setup_rtsummary( return -ENOMEM; sc->buf = rts; + if (xchk_could_repair(sc)) { + error = xrep_setup_rtsummary(sc, rts); + if (error) + return error; + } + /* * Create an xfile to construct a new rtsummary file. The xfile allows * us to avoid pinning kernel memory for this purpose. @@ -70,7 +68,7 @@ xchk_setup_rtsummary( if (error) return error; - error = xchk_trans_alloc(sc, 0); + error = xchk_trans_alloc(sc, rts->resblks); if (error) return error; @@ -135,7 +133,7 @@ xfsum_store( sumoff << XFS_WORDLOG); } -static inline int +inline int xfsum_copyout( struct xfs_scrub *sc, xfs_rtsumoff_t sumoff, @@ -362,7 +360,12 @@ xchk_rtsummary( error = xchk_rtsum_compare(sc); out_rbm: - /* Unlock the rtbitmap since we're done with it. */ + /* + * Unlock the rtbitmap since we're done with it. All other writers of + * the rt free space metadata grab the bitmap and summary ILOCKs in + * that order, so we're still protected against allocation activities + * even if we continue on to the repair function. + */ xfs_iunlock(mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); return error; } diff --git a/fs/xfs/scrub/rtsummary.h b/fs/xfs/scrub/rtsummary.h new file mode 100644 index 000000000000..e1d50304d8d4 --- /dev/null +++ b/fs/xfs/scrub/rtsummary.h @@ -0,0 +1,37 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_RTSUMMARY_H__ +#define __XFS_SCRUB_RTSUMMARY_H__ + +struct xchk_rtsummary { +#ifdef CONFIG_XFS_ONLINE_REPAIR + struct xrep_tempexch tempexch; +#endif + struct xfs_rtalloc_args args; + + uint64_t rextents; + uint64_t rbmblocks; + uint64_t rsumsize; + unsigned int rsumlevels; + unsigned int resblks; + + /* suminfo position of xfile as we write buffers to disk. */ + xfs_rtsumoff_t prep_wordoff; + + /* Memory buffer for the summary comparison. */ + union xfs_suminfo_raw words[]; +}; + +int xfsum_copyout(struct xfs_scrub *sc, xfs_rtsumoff_t sumoff, + union xfs_suminfo_raw *rawinfo, unsigned int nr_words); + +#ifdef CONFIG_XFS_ONLINE_REPAIR +int xrep_setup_rtsummary(struct xfs_scrub *sc, struct xchk_rtsummary *rts); +#else +# define xrep_setup_rtsummary(sc, rts) (0) +#endif /* CONFIG_XFS_ONLINE_REPAIR */ + +#endif /* __XFS_SCRUB_RTSUMMARY_H__ */ diff --git a/fs/xfs/scrub/rtsummary_repair.c b/fs/xfs/scrub/rtsummary_repair.c new file mode 100644 index 000000000000..c8bb6c4f15d0 --- /dev/null +++ b/fs/xfs/scrub/rtsummary_repair.c @@ -0,0 +1,177 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_btree.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_rtalloc.h" +#include "xfs_inode.h" +#include "xfs_bit.h" +#include "xfs_bmap.h" +#include "xfs_bmap_btree.h" +#include "xfs_exchmaps.h" +#include "xfs_rtbitmap.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/tempfile.h" +#include "scrub/tempexch.h" +#include "scrub/reap.h" +#include "scrub/xfile.h" +#include "scrub/rtsummary.h" + +/* Set us up to repair the rtsummary file. */ +int +xrep_setup_rtsummary( + struct xfs_scrub *sc, + struct xchk_rtsummary *rts) +{ + struct xfs_mount *mp = sc->mp; + unsigned long long blocks; + int error; + + error = xrep_tempfile_create(sc, S_IFREG); + if (error) + return error; + + /* + * If we're doing a repair, we reserve enough blocks to write out a + * completely new summary file, plus twice as many blocks as we would + * need if we can only allocate one block per data fork mapping. This + * should cover the preallocation of the temporary file and exchanging + * the extent mappings. + * + * We cannot use xfs_exchmaps_estimate because we have not yet + * constructed the replacement rtsummary and therefore do not know how + * many extents it will use. By the time we do, we will have a dirty + * transaction (which we cannot drop because we cannot drop the + * rtsummary ILOCK) and cannot ask for more reservation. + */ + blocks = XFS_B_TO_FSB(mp, mp->m_rsumsize); + blocks += xfs_bmbt_calc_size(mp, blocks) * 2; + if (blocks > UINT_MAX) + return -EOPNOTSUPP; + + rts->resblks += blocks; + + /* + * Grab support for atomic file content exchanges before we allocate + * any transactions or grab ILOCKs. + */ + return xrep_tempexch_enable(sc); +} + +static int +xrep_rtsummary_prep_buf( + struct xfs_scrub *sc, + struct xfs_buf *bp, + void *data) +{ + struct xchk_rtsummary *rts = data; + struct xfs_mount *mp = sc->mp; + union xfs_suminfo_raw *ondisk; + int error; + + rts->args.mp = sc->mp; + rts->args.tp = sc->tp; + rts->args.sumbp = bp; + ondisk = xfs_rsumblock_infoptr(&rts->args, 0); + rts->args.sumbp = NULL; + + bp->b_ops = &xfs_rtbuf_ops; + + error = xfsum_copyout(sc, rts->prep_wordoff, ondisk, mp->m_blockwsize); + if (error) + return error; + + rts->prep_wordoff += mp->m_blockwsize; + xfs_trans_buf_set_type(sc->tp, bp, XFS_BLFT_RTSUMMARY_BUF); + return 0; +} + +/* Repair the realtime summary. */ +int +xrep_rtsummary( + struct xfs_scrub *sc) +{ + struct xchk_rtsummary *rts = sc->buf; + struct xfs_mount *mp = sc->mp; + xfs_filblks_t rsumblocks; + int error; + + /* We require the rmapbt to rebuild anything. */ + if (!xfs_has_rmapbt(mp)) + return -EOPNOTSUPP; + + /* Walk away if we disagree on the size of the rt bitmap. */ + if (rts->rbmblocks != mp->m_sb.sb_rbmblocks) + return 0; + + /* Make sure any problems with the fork are fixed. */ + error = xrep_metadata_inode_forks(sc); + if (error) + return error; + + /* + * Try to take ILOCK_EXCL of the temporary file. We had better be the + * only ones holding onto this inode, but we can't block while holding + * the rtsummary file's ILOCK_EXCL. + */ + while (!xrep_tempfile_ilock_nowait(sc)) { + if (xchk_should_terminate(sc, &error)) + return error; + delay(1); + } + + /* Make sure we have space allocated for the entire summary file. */ + rsumblocks = XFS_B_TO_FSB(mp, rts->rsumsize); + xfs_trans_ijoin(sc->tp, sc->ip, 0); + xfs_trans_ijoin(sc->tp, sc->tempip, 0); + error = xrep_tempfile_prealloc(sc, 0, rsumblocks); + if (error) + return error; + + /* Last chance to abort before we start committing fixes. */ + if (xchk_should_terminate(sc, &error)) + return error; + + /* Copy the rtsummary file that we generated. */ + error = xrep_tempfile_copyin(sc, 0, rsumblocks, + xrep_rtsummary_prep_buf, rts); + if (error) + return error; + error = xrep_tempfile_set_isize(sc, rts->rsumsize); + if (error) + return error; + + /* + * Now exchange the contents. Nothing in repair uses the temporary + * buffer, so we can reuse it for the tempfile exchrange information. + */ + error = xrep_tempexch_trans_reserve(sc, XFS_DATA_FORK, &rts->tempexch); + if (error) + return error; + + error = xrep_tempexch_contents(sc, &rts->tempexch); + if (error) + return error; + + /* Reset incore state and blow out the summary cache. */ + if (mp->m_rsum_cache) + memset(mp->m_rsum_cache, 0xFF, mp->m_sb.sb_rbmblocks); + + mp->m_rsumlevels = rts->rsumlevels; + mp->m_rsumsize = rts->rsumsize; + + /* Free the old rtsummary blocks if they're not in use. */ + return xrep_reap_ifork(sc, sc->tempip, XFS_DATA_FORK); +} diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index ff156edf49a0..62a064c1a5d3 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -18,6 +18,7 @@ #include "xfs_buf_mem.h" #include "xfs_rmap.h" #include "xfs_exchrange.h" +#include "xfs_exchmaps.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" @@ -354,7 +355,7 @@ static const struct xchk_meta_ops meta_scrub_ops[] = { .type = ST_FS, .setup = xchk_setup_rtsummary, .scrub = xchk_rtsummary, - .repair = xrep_notsupported, + .repair = xrep_rtsummary, }, [XFS_SCRUB_TYPE_UQUOTA] = { /* user quota */ .type = ST_FS, -- cgit v1.2.3-59-g8ed1b From 9eef772f3a194f6841850e45dacdf4207ec7da84 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:34 -0700 Subject: xfs: add an explicit owner field to xfs_da_args Add an explicit owner field to xfs_da_args, which will make it easier for online fsck to set the owner field of the temporary directory and xattr structures that it builds to repair damaged metadata. Note: I hopefully found all the xfs_da_args definitions by looking for automatic stack variable declarations and xfs_da_args.dp assignments: git grep -E '(args.*dp =|struct xfs_da_args[[:space:]]*[a-z0-9][a-z0-9]*)' Note that callers of xfs_attr_{get,set,change} can set the owner to zero (or leave it unset) to have the default set to args->dp. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 4 ++++ fs/xfs/libxfs/xfs_attr_leaf.c | 2 ++ fs/xfs/libxfs/xfs_bmap.c | 1 + fs/xfs/libxfs/xfs_da_btree.h | 1 + fs/xfs/libxfs/xfs_dir2.c | 5 +++++ fs/xfs/libxfs/xfs_exchmaps.c | 2 ++ fs/xfs/scrub/attr.c | 1 + fs/xfs/scrub/dabtree.c | 1 + fs/xfs/scrub/dir.c | 3 ++- fs/xfs/scrub/readdir.c | 2 ++ fs/xfs/xfs_attr_item.c | 1 + fs/xfs/xfs_dir2_readdir.c | 1 + fs/xfs/xfs_trace.h | 7 +++++-- 13 files changed, 28 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 673a4b6d2e8d..74d769461443 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -264,6 +264,8 @@ xfs_attr_get( if (xfs_is_shutdown(args->dp->i_mount)) return -EIO; + if (!args->owner) + args->owner = args->dp->i_ino; args->geo = args->dp->i_mount->m_attr_geo; args->whichfork = XFS_ATTR_FORK; args->hashval = xfs_da_hashname(args->name, args->namelen); @@ -937,6 +939,8 @@ xfs_attr_set( if (error) return error; + if (!args->owner) + args->owner = args->dp->i_ino; args->geo = mp->m_attr_geo; args->whichfork = XFS_ATTR_FORK; args->hashval = xfs_da_hashname(args->name, args->namelen); diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index ac904cc1a97b..e606eae8d377 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -904,6 +904,7 @@ xfs_attr_shortform_to_leaf( nargs.whichfork = XFS_ATTR_FORK; nargs.trans = args->trans; nargs.op_flags = XFS_DA_OP_OKNOENT; + nargs.owner = args->owner; sfe = xfs_attr_sf_firstentry(sf); for (i = 0; i < sf->count; i++) { @@ -1106,6 +1107,7 @@ xfs_attr3_leaf_to_shortform( nargs.whichfork = XFS_ATTR_FORK; nargs.trans = args->trans; nargs.op_flags = XFS_DA_OP_OKNOENT; + nargs.owner = args->owner; for (i = 0; i < ichdr.count; entry++, i++) { if (entry->flags & XFS_ATTR_INCOMPLETE) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 656c95a22f2e..46bbc9f0a117 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -976,6 +976,7 @@ xfs_bmap_add_attrfork_local( dargs.total = dargs.geo->fsbcount; dargs.whichfork = XFS_DATA_FORK; dargs.trans = tp; + dargs.owner = ip->i_ino; return xfs_dir2_sf_to_block(&dargs); } diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 706baf36e175..7fb13f26edaa 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -79,6 +79,7 @@ typedef struct xfs_da_args { int rmtvaluelen2; /* remote attr value length in bytes */ uint32_t op_flags; /* operation flags */ enum xfs_dacmp cmpresult; /* name compare result for lookups */ + xfs_ino_t owner; /* inode that owns the dir/attr data */ } xfs_da_args_t; /* diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index 4821519efad4..9da99fa20c75 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -250,6 +250,7 @@ xfs_dir_init( args->geo = dp->i_mount->m_dir_geo; args->dp = dp; args->trans = tp; + args->owner = dp->i_ino; error = xfs_dir2_sf_create(args, pdp->i_ino); kfree(args); return error; @@ -295,6 +296,7 @@ xfs_dir_createname( args->whichfork = XFS_DATA_FORK; args->trans = tp; args->op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; + args->owner = dp->i_ino; if (!inum) args->op_flags |= XFS_DA_OP_JUSTCHECK; @@ -383,6 +385,7 @@ xfs_dir_lookup( args->whichfork = XFS_DATA_FORK; args->trans = tp; args->op_flags = XFS_DA_OP_OKNOENT; + args->owner = dp->i_ino; if (ci_name) args->op_flags |= XFS_DA_OP_CILOOKUP; @@ -456,6 +459,7 @@ xfs_dir_removename( args->total = total; args->whichfork = XFS_DATA_FORK; args->trans = tp; + args->owner = dp->i_ino; if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { rval = xfs_dir2_sf_removename(args); @@ -517,6 +521,7 @@ xfs_dir_replace( args->total = total; args->whichfork = XFS_DATA_FORK; args->trans = tp; + args->owner = dp->i_ino; if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { rval = xfs_dir2_sf_replace(args); diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index 7fa244228750..8d28e8cce5e9 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -429,6 +429,7 @@ xfs_exchmaps_attr_to_sf( .geo = tp->t_mountp->m_attr_geo, .whichfork = XFS_ATTR_FORK, .trans = tp, + .owner = xmi->xmi_ip2->i_ino, }; struct xfs_buf *bp; int forkoff; @@ -459,6 +460,7 @@ xfs_exchmaps_dir_to_sf( .geo = tp->t_mountp->m_dir_geo, .whichfork = XFS_DATA_FORK, .trans = tp, + .owner = xmi->xmi_ip2->i_ino, }; struct xfs_dir2_sf_hdr sfh; struct xfs_buf *bp; diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 83c7feb38714..0c467f4f8e77 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -169,6 +169,7 @@ xchk_xattr_listent( .hashval = xfs_da_hashname(name, namelen), .trans = context->tp, .valuelen = valuelen, + .owner = context->dp->i_ino, }; struct xchk_xattr_buf *ab; struct xchk_xattr *sx; diff --git a/fs/xfs/scrub/dabtree.c b/fs/xfs/scrub/dabtree.c index 82b150d3b8b7..fa6385a99ac4 100644 --- a/fs/xfs/scrub/dabtree.c +++ b/fs/xfs/scrub/dabtree.c @@ -494,6 +494,7 @@ xchk_da_btree( ds->dargs.whichfork = whichfork; ds->dargs.trans = sc->tp; ds->dargs.op_flags = XFS_DA_OP_OKNOENT; + ds->dargs.owner = sc->ip->i_ino; ds->state = xfs_da_state_alloc(&ds->dargs); ds->sc = sc; ds->private = private; diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 076a310b8eb0..042e28547e04 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -621,10 +621,11 @@ xchk_directory_blocks( { struct xfs_bmbt_irec got; struct xfs_da_args args = { - .dp = sc ->ip, + .dp = sc->ip, .whichfork = XFS_DATA_FORK, .geo = sc->mp->m_dir_geo, .trans = sc->tp, + .owner = sc->ip->i_ino, }; struct xfs_ifork *ifp = xfs_ifork_ptr(sc->ip, XFS_DATA_FORK); struct xfs_mount *mp = sc->mp; diff --git a/fs/xfs/scrub/readdir.c b/fs/xfs/scrub/readdir.c index dfdcb96b6c16..fb98b7624994 100644 --- a/fs/xfs/scrub/readdir.c +++ b/fs/xfs/scrub/readdir.c @@ -273,6 +273,7 @@ xchk_dir_walk( .dp = dp, .geo = dp->i_mount->m_dir_geo, .trans = sc->tp, + .owner = dp->i_ino, }; bool isblock; int error; @@ -324,6 +325,7 @@ xchk_dir_lookup( .hashval = xfs_dir2_hashname(dp->i_mount, name), .whichfork = XFS_DATA_FORK, .op_flags = XFS_DA_OP_OKNOENT, + .owner = dp->i_ino, }; bool isblock, isleaf; int error; diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 9b4c61e1c22e..d46034705694 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -540,6 +540,7 @@ xfs_attri_recover_work( args->attr_filter = attrp->alfi_attr_filter & XFS_ATTRI_FILTER_MASK; args->op_flags = XFS_DA_OP_RECOVERY | XFS_DA_OP_OKNOENT | XFS_DA_OP_LOGGED; + args->owner = args->dp->i_ino; ASSERT(xfs_sb_version_haslogxattrs(&mp->m_sb)); diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c index cf9296b7e06f..4e811fa393ad 100644 --- a/fs/xfs/xfs_dir2_readdir.c +++ b/fs/xfs/xfs_dir2_readdir.c @@ -532,6 +532,7 @@ xfs_readdir( args.dp = dp; args.geo = dp->i_mount->m_dir_geo; args.trans = tp; + args.owner = dp->i_ino; if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) return xfs_dir2_sf_getdents(&args, ctx); diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index caef95f2c87c..939baf08331b 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -1931,6 +1931,7 @@ DECLARE_EVENT_CLASS(xfs_da_class, __field(xfs_dahash_t, hashval) __field(xfs_ino_t, inumber) __field(uint32_t, op_flags) + __field(xfs_ino_t, owner) ), TP_fast_assign( __entry->dev = VFS_I(args->dp)->i_sb->s_dev; @@ -1941,9 +1942,10 @@ DECLARE_EVENT_CLASS(xfs_da_class, __entry->hashval = args->hashval; __entry->inumber = args->inumber; __entry->op_flags = args->op_flags; + __entry->owner = args->owner; ), TP_printk("dev %d:%d ino 0x%llx name %.*s namelen %d hashval 0x%x " - "inumber 0x%llx op_flags %s", + "inumber 0x%llx op_flags %s owner 0x%llx", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->namelen, @@ -1951,7 +1953,8 @@ DECLARE_EVENT_CLASS(xfs_da_class, __entry->namelen, __entry->hashval, __entry->inumber, - __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS)) + __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS), + __entry->owner) ) #define DEFINE_DIR2_EVENT(name) \ -- cgit v1.2.3-59-g8ed1b From 17a85dc64ae0804d33a2293686fc987a830a462d Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:34 -0700 Subject: xfs: use the xfs_da_args owner field to set new dir/attr block owner When we're creating leaf, data, freespace, or dabtree blocks for directories and xattrs, use the explicit owner field (instead of the xfs_inode) to set the owner field. This will enable online repair to construct replacement data structures in a temporary file without having to change the owner fields prior to swapping the new and old structures. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_leaf.c | 2 +- fs/xfs/libxfs/xfs_attr_remote.c | 4 ++-- fs/xfs/libxfs/xfs_da_btree.c | 2 +- fs/xfs/libxfs/xfs_dir2_block.c | 19 ++++++++++--------- fs/xfs/libxfs/xfs_dir2_data.c | 2 +- fs/xfs/libxfs/xfs_dir2_leaf.c | 11 +++++------ fs/xfs/libxfs/xfs_dir2_node.c | 2 +- 7 files changed, 21 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index e606eae8d377..8937c034b330 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -1239,7 +1239,7 @@ xfs_attr3_leaf_create( ichdr.magic = XFS_ATTR3_LEAF_MAGIC; hdr3->blkno = cpu_to_be64(xfs_buf_daddr(bp)); - hdr3->owner = cpu_to_be64(dp->i_ino); + hdr3->owner = cpu_to_be64(args->owner); uuid_copy(&hdr3->uuid, &mp->m_sb.sb_meta_uuid); ichdr.freemap[0].base = sizeof(struct xfs_attr3_leaf_hdr); diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index ff0412828772..024895cc7029 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -522,8 +522,8 @@ xfs_attr_rmtval_set_value( return error; bp->b_ops = &xfs_attr3_rmt_buf_ops; - xfs_attr_rmtval_copyin(mp, bp, args->dp->i_ino, &offset, - &valuelen, &src); + xfs_attr_rmtval_copyin(mp, bp, args->owner, &offset, &valuelen, + &src); error = xfs_bwrite(bp); /* GROT: NOTE: synchronous write */ xfs_buf_relse(bp); diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index 718d071bb21a..743f6421cc04 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -486,7 +486,7 @@ xfs_da3_node_create( memset(hdr3, 0, sizeof(struct xfs_da3_node_hdr)); ichdr.magic = XFS_DA3_NODE_MAGIC; hdr3->info.blkno = cpu_to_be64(xfs_buf_daddr(bp)); - hdr3->info.owner = cpu_to_be64(args->dp->i_ino); + hdr3->info.owner = cpu_to_be64(args->owner); uuid_copy(&hdr3->info.uuid, &mp->m_sb.sb_meta_uuid); } else { ichdr.magic = XFS_DA_NODE_MAGIC; diff --git a/fs/xfs/libxfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c index a2da007adb46..61cbc668f228 100644 --- a/fs/xfs/libxfs/xfs_dir2_block.c +++ b/fs/xfs/libxfs/xfs_dir2_block.c @@ -163,12 +163,13 @@ xfs_dir3_block_read( static void xfs_dir3_block_init( - struct xfs_mount *mp, - struct xfs_trans *tp, - struct xfs_buf *bp, - struct xfs_inode *dp) + struct xfs_da_args *args, + struct xfs_buf *bp) { - struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr; + struct xfs_trans *tp = args->trans; + struct xfs_inode *dp = args->dp; + struct xfs_mount *mp = dp->i_mount; + struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr; bp->b_ops = &xfs_dir3_block_buf_ops; xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_BLOCK_BUF); @@ -177,7 +178,7 @@ xfs_dir3_block_init( memset(hdr3, 0, sizeof(*hdr3)); hdr3->magic = cpu_to_be32(XFS_DIR3_BLOCK_MAGIC); hdr3->blkno = cpu_to_be64(xfs_buf_daddr(bp)); - hdr3->owner = cpu_to_be64(dp->i_ino); + hdr3->owner = cpu_to_be64(args->owner); uuid_copy(&hdr3->uuid, &mp->m_sb.sb_meta_uuid); return; @@ -1009,7 +1010,7 @@ xfs_dir2_leaf_to_block( /* * Start converting it to block form. */ - xfs_dir3_block_init(mp, tp, dbp, dp); + xfs_dir3_block_init(args, dbp); needlog = 1; needscan = 0; @@ -1129,7 +1130,7 @@ xfs_dir2_sf_to_block( error = xfs_dir3_data_init(args, blkno, &bp); if (error) goto out_free; - xfs_dir3_block_init(mp, tp, bp, dp); + xfs_dir3_block_init(args, bp); hdr = bp->b_addr; /* @@ -1169,7 +1170,7 @@ xfs_dir2_sf_to_block( * Create entry for . */ dep = bp->b_addr + offset; - dep->inumber = cpu_to_be64(dp->i_ino); + dep->inumber = cpu_to_be64(args->owner); dep->namelen = 1; dep->name[0] = '.'; xfs_dir2_data_put_ftype(mp, dep, XFS_DIR3_FT_DIR); diff --git a/fs/xfs/libxfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c index 7a6d965bea71..c3ef720b5ff6 100644 --- a/fs/xfs/libxfs/xfs_dir2_data.c +++ b/fs/xfs/libxfs/xfs_dir2_data.c @@ -725,7 +725,7 @@ xfs_dir3_data_init( memset(hdr3, 0, sizeof(*hdr3)); hdr3->magic = cpu_to_be32(XFS_DIR3_DATA_MAGIC); hdr3->blkno = cpu_to_be64(xfs_buf_daddr(bp)); - hdr3->owner = cpu_to_be64(dp->i_ino); + hdr3->owner = cpu_to_be64(args->owner); uuid_copy(&hdr3->uuid, &mp->m_sb.sb_meta_uuid); } else diff --git a/fs/xfs/libxfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c index 08dda5ce9d91..20ce057d12e8 100644 --- a/fs/xfs/libxfs/xfs_dir2_leaf.c +++ b/fs/xfs/libxfs/xfs_dir2_leaf.c @@ -304,12 +304,12 @@ xfs_dir3_leafn_read( */ static void xfs_dir3_leaf_init( - struct xfs_mount *mp, - struct xfs_trans *tp, + struct xfs_da_args *args, struct xfs_buf *bp, - xfs_ino_t owner, uint16_t type) { + struct xfs_mount *mp = args->dp->i_mount; + struct xfs_trans *tp = args->trans; struct xfs_dir2_leaf *leaf = bp->b_addr; ASSERT(type == XFS_DIR2_LEAF1_MAGIC || type == XFS_DIR2_LEAFN_MAGIC); @@ -323,7 +323,7 @@ xfs_dir3_leaf_init( ? cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) : cpu_to_be16(XFS_DIR3_LEAFN_MAGIC); leaf3->info.blkno = cpu_to_be64(xfs_buf_daddr(bp)); - leaf3->info.owner = cpu_to_be64(owner); + leaf3->info.owner = cpu_to_be64(args->owner); uuid_copy(&leaf3->info.uuid, &mp->m_sb.sb_meta_uuid); } else { memset(leaf, 0, sizeof(*leaf)); @@ -356,7 +356,6 @@ xfs_dir3_leaf_get_buf( { struct xfs_inode *dp = args->dp; struct xfs_trans *tp = args->trans; - struct xfs_mount *mp = dp->i_mount; struct xfs_buf *bp; int error; @@ -369,7 +368,7 @@ xfs_dir3_leaf_get_buf( if (error) return error; - xfs_dir3_leaf_init(mp, tp, bp, dp->i_ino, magic); + xfs_dir3_leaf_init(args, bp, magic); xfs_dir3_leaf_log_header(args, bp); if (magic == XFS_DIR2_LEAF1_MAGIC) xfs_dir3_leaf_log_tail(args, bp); diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c index be0b8834028c..1ad7405f9c38 100644 --- a/fs/xfs/libxfs/xfs_dir2_node.c +++ b/fs/xfs/libxfs/xfs_dir2_node.c @@ -349,7 +349,7 @@ xfs_dir3_free_get_buf( hdr.magic = XFS_DIR3_FREE_MAGIC; hdr3->hdr.blkno = cpu_to_be64(xfs_buf_daddr(bp)); - hdr3->hdr.owner = cpu_to_be64(dp->i_ino); + hdr3->hdr.owner = cpu_to_be64(args->owner); uuid_copy(&hdr3->hdr.uuid, &mp->m_sb.sb_meta_uuid); } else hdr.magic = XFS_DIR2_FREE_MAGIC; -- cgit v1.2.3-59-g8ed1b From 33c028ffe36ad7a91930acf0bd3d6ee7340022bf Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:35 -0700 Subject: xfs: reduce indenting in xfs_attr_node_list Reduce the indentation here so that we can add some things in the next patch without going over the column limits. Suggested-by: Christoph Hellwig Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_list.c | 56 ++++++++++++++++++++++++++------------------------ 1 file changed, 29 insertions(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index a6819a642cc0..42a575db7267 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -310,46 +310,47 @@ xfs_attr_node_list( */ bp = NULL; if (cursor->blkno > 0) { + struct xfs_attr_leaf_entry *entries; + error = xfs_da3_node_read(context->tp, dp, cursor->blkno, &bp, XFS_ATTR_FORK); if (xfs_metadata_is_sick(error)) xfs_dirattr_mark_sick(dp, XFS_ATTR_FORK); - if ((error != 0) && (error != -EFSCORRUPTED)) + if (error != 0 && error != -EFSCORRUPTED) return error; - if (bp) { - struct xfs_attr_leaf_entry *entries; + if (!bp) + goto need_lookup; - node = bp->b_addr; - switch (be16_to_cpu(node->hdr.info.magic)) { - case XFS_DA_NODE_MAGIC: - case XFS_DA3_NODE_MAGIC: + node = bp->b_addr; + switch (be16_to_cpu(node->hdr.info.magic)) { + case XFS_DA_NODE_MAGIC: + case XFS_DA3_NODE_MAGIC: + trace_xfs_attr_list_wrong_blk(context); + xfs_trans_brelse(context->tp, bp); + bp = NULL; + break; + case XFS_ATTR_LEAF_MAGIC: + case XFS_ATTR3_LEAF_MAGIC: + leaf = bp->b_addr; + xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, + &leafhdr, leaf); + entries = xfs_attr3_leaf_entryp(leaf); + if (cursor->hashval > be32_to_cpu( + entries[leafhdr.count - 1].hashval)) { trace_xfs_attr_list_wrong_blk(context); xfs_trans_brelse(context->tp, bp); bp = NULL; - break; - case XFS_ATTR_LEAF_MAGIC: - case XFS_ATTR3_LEAF_MAGIC: - leaf = bp->b_addr; - xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, - &leafhdr, leaf); - entries = xfs_attr3_leaf_entryp(leaf); - if (cursor->hashval > be32_to_cpu( - entries[leafhdr.count - 1].hashval)) { - trace_xfs_attr_list_wrong_blk(context); - xfs_trans_brelse(context->tp, bp); - bp = NULL; - } else if (cursor->hashval <= be32_to_cpu( - entries[0].hashval)) { - trace_xfs_attr_list_wrong_blk(context); - xfs_trans_brelse(context->tp, bp); - bp = NULL; - } - break; - default: + } else if (cursor->hashval <= be32_to_cpu( + entries[0].hashval)) { trace_xfs_attr_list_wrong_blk(context); xfs_trans_brelse(context->tp, bp); bp = NULL; } + break; + default: + trace_xfs_attr_list_wrong_blk(context); + xfs_trans_brelse(context->tp, bp); + bp = NULL; } } @@ -359,6 +360,7 @@ xfs_attr_node_list( * Note that start of node block is same as start of leaf block. */ if (bp == NULL) { +need_lookup: error = xfs_attr_node_list_lookup(context, cursor, &bp); if (error || !bp) return error; -- cgit v1.2.3-59-g8ed1b From f4887fbc41dcb1560ec5da982ac7c6ad04b71de5 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:36 -0700 Subject: xfs: validate attr leaf buffer owners Create a leaf block header checking function to validate the owner field of xattr leaf blocks. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 10 ++++---- fs/xfs/libxfs/xfs_attr_leaf.c | 56 ++++++++++++++++++++++++++++++++++++------- fs/xfs/libxfs/xfs_attr_leaf.h | 4 +++- fs/xfs/libxfs/xfs_da_btree.c | 42 ++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_da_btree.h | 1 + fs/xfs/libxfs/xfs_exchmaps.c | 3 ++- fs/xfs/scrub/dabtree.c | 7 ++++++ fs/xfs/xfs_attr_list.c | 24 ++++++++++++++++--- 8 files changed, 128 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 74d769461443..b3c9666cd011 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -649,8 +649,8 @@ xfs_attr_leaf_remove_attr( int forkoff; int error; - error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, - &bp); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, + args->blkno, &bp); if (error) return error; @@ -681,7 +681,7 @@ xfs_attr_leaf_shrink( if (!xfs_attr_is_leaf(dp)) return 0; - error = xfs_attr3_leaf_read(args->trans, args->dp, 0, &bp); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, 0, &bp); if (error) return error; @@ -1158,7 +1158,7 @@ xfs_attr_leaf_try_add( struct xfs_buf *bp; int error; - error = xfs_attr3_leaf_read(args->trans, args->dp, 0, &bp); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, 0, &bp); if (error) return error; @@ -1206,7 +1206,7 @@ xfs_attr_leaf_hasname( { int error = 0; - error = xfs_attr3_leaf_read(args->trans, args->dp, 0, bp); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, 0, bp); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index 8937c034b330..17ec5ff5a4e3 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -388,6 +388,27 @@ xfs_attr3_leaf_verify( return NULL; } +xfs_failaddr_t +xfs_attr3_leaf_header_check( + struct xfs_buf *bp, + xfs_ino_t owner) +{ + struct xfs_mount *mp = bp->b_mount; + + if (xfs_has_crc(mp)) { + struct xfs_attr3_leafblock *hdr3 = bp->b_addr; + + if (hdr3->hdr.info.hdr.magic != + cpu_to_be16(XFS_ATTR3_LEAF_MAGIC)) + return __this_address; + + if (be64_to_cpu(hdr3->hdr.info.owner) != owner) + return __this_address; + } + + return NULL; +} + static void xfs_attr3_leaf_write_verify( struct xfs_buf *bp) @@ -448,16 +469,30 @@ int xfs_attr3_leaf_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t bno, struct xfs_buf **bpp) { + xfs_failaddr_t fa; int err; err = xfs_da_read_buf(tp, dp, bno, 0, bpp, XFS_ATTR_FORK, &xfs_attr3_leaf_buf_ops); - if (!err && tp && *bpp) + if (err || !(*bpp)) + return err; + + fa = xfs_attr3_leaf_header_check(*bpp, owner); + if (fa) { + __xfs_buf_mark_corrupt(*bpp, fa); + xfs_trans_brelse(tp, *bpp); + *bpp = NULL; + xfs_dirattr_mark_sick(dp, XFS_ATTR_FORK); + return -EFSCORRUPTED; + } + + if (tp) xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_ATTR_LEAF_BUF); - return err; + return 0; } /*======================================================================== @@ -1160,7 +1195,7 @@ xfs_attr3_leaf_to_node( error = xfs_da_grow_inode(args, &blkno); if (error) goto out; - error = xfs_attr3_leaf_read(args->trans, dp, 0, &bp1); + error = xfs_attr3_leaf_read(args->trans, dp, args->owner, 0, &bp1); if (error) goto out; @@ -1995,7 +2030,7 @@ xfs_attr3_leaf_toosmall( if (blkno == 0) continue; error = xfs_attr3_leaf_read(state->args->trans, state->args->dp, - blkno, &bp); + state->args->owner, blkno, &bp); if (error) return error; @@ -2717,7 +2752,8 @@ xfs_attr3_leaf_clearflag( /* * Set up the operation. */ - error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, + args->blkno, &bp); if (error) return error; @@ -2781,7 +2817,8 @@ xfs_attr3_leaf_setflag( /* * Set up the operation. */ - error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, + args->blkno, &bp); if (error) return error; @@ -2840,7 +2877,8 @@ xfs_attr3_leaf_flipflags( /* * Read the block containing the "old" attr */ - error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp1); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, + args->blkno, &bp1); if (error) return error; @@ -2848,8 +2886,8 @@ xfs_attr3_leaf_flipflags( * Read the block containing the "new" attr, if it is different */ if (args->blkno2 != args->blkno) { - error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno2, - &bp2); + error = xfs_attr3_leaf_read(args->trans, args->dp, args->owner, + args->blkno2, &bp2); if (error) return error; } else { diff --git a/fs/xfs/libxfs/xfs_attr_leaf.h b/fs/xfs/libxfs/xfs_attr_leaf.h index 9b9948639c0f..bac219589896 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.h +++ b/fs/xfs/libxfs/xfs_attr_leaf.h @@ -98,12 +98,14 @@ int xfs_attr_leaf_order(struct xfs_buf *leaf1_bp, struct xfs_buf *leaf2_bp); int xfs_attr_leaf_newentsize(struct xfs_da_args *args, int *local); int xfs_attr3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp, - xfs_dablk_t bno, struct xfs_buf **bpp); + xfs_ino_t owner, xfs_dablk_t bno, struct xfs_buf **bpp); void xfs_attr3_leaf_hdr_from_disk(struct xfs_da_geometry *geo, struct xfs_attr3_icleaf_hdr *to, struct xfs_attr_leafblock *from); void xfs_attr3_leaf_hdr_to_disk(struct xfs_da_geometry *geo, struct xfs_attr_leafblock *to, struct xfs_attr3_icleaf_hdr *from); +xfs_failaddr_t xfs_attr3_leaf_header_check(struct xfs_buf *bp, + xfs_ino_t owner); #endif /* __XFS_ATTR_LEAF_H__ */ diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index 743f6421cc04..65eef8775187 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -252,6 +252,25 @@ xfs_da3_node_verify( return NULL; } +xfs_failaddr_t +xfs_da3_header_check( + struct xfs_buf *bp, + xfs_ino_t owner) +{ + struct xfs_mount *mp = bp->b_mount; + struct xfs_da_blkinfo *hdr = bp->b_addr; + + if (!xfs_has_crc(mp)) + return NULL; + + switch (hdr->magic) { + case cpu_to_be16(XFS_ATTR3_LEAF_MAGIC): + return xfs_attr3_leaf_header_check(bp, owner); + } + + return NULL; +} + static void xfs_da3_node_write_verify( struct xfs_buf *bp) @@ -1591,6 +1610,7 @@ xfs_da3_node_lookup_int( struct xfs_da_node_entry *btree; struct xfs_da3_icnode_hdr nodehdr; struct xfs_da_args *args; + xfs_failaddr_t fa; xfs_dablk_t blkno; xfs_dahash_t hashval; xfs_dahash_t btreehashval; @@ -1629,6 +1649,12 @@ xfs_da3_node_lookup_int( if (magic == XFS_ATTR_LEAF_MAGIC || magic == XFS_ATTR3_LEAF_MAGIC) { + fa = xfs_attr3_leaf_header_check(blk->bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(blk->bp, fa); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } blk->magic = XFS_ATTR_LEAF_MAGIC; blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL); break; @@ -1996,6 +2022,7 @@ xfs_da3_path_shift( struct xfs_da_node_entry *btree; struct xfs_da3_icnode_hdr nodehdr; struct xfs_buf *bp; + xfs_failaddr_t fa; xfs_dablk_t blkno = 0; int level; int error; @@ -2087,6 +2114,12 @@ xfs_da3_path_shift( break; case XFS_ATTR_LEAF_MAGIC: case XFS_ATTR3_LEAF_MAGIC: + fa = xfs_attr3_leaf_header_check(blk->bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(blk->bp, fa); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } blk->magic = XFS_ATTR_LEAF_MAGIC; ASSERT(level == path->active-1); blk->index = 0; @@ -2290,6 +2323,7 @@ xfs_da3_swap_lastblock( struct xfs_buf *last_buf; struct xfs_buf *sib_buf; struct xfs_buf *par_buf; + xfs_failaddr_t fa; xfs_dahash_t dead_hash; xfs_fileoff_t lastoff; xfs_dablk_t dead_blkno; @@ -2326,6 +2360,14 @@ xfs_da3_swap_lastblock( error = xfs_da3_node_read(tp, dp, last_blkno, &last_buf, w); if (error) return error; + fa = xfs_da3_header_check(last_buf, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(last_buf, fa); + xfs_trans_brelse(tp, last_buf); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } + /* * Copy the last block into the dead buffer and log it. */ diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 7fb13f26edaa..99618e0c8a72 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -236,6 +236,7 @@ void xfs_da3_node_hdr_from_disk(struct xfs_mount *mp, struct xfs_da3_icnode_hdr *to, struct xfs_da_intnode *from); void xfs_da3_node_hdr_to_disk(struct xfs_mount *mp, struct xfs_da_intnode *to, struct xfs_da3_icnode_hdr *from); +xfs_failaddr_t xfs_da3_header_check(struct xfs_buf *bp, xfs_ino_t owner); extern struct kmem_cache *xfs_da_state_cache; diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index 8d28e8cce5e9..9c9cf2e998b2 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -438,7 +438,8 @@ xfs_exchmaps_attr_to_sf( if (!xfs_attr_is_leaf(xmi->xmi_ip2)) return 0; - error = xfs_attr3_leaf_read(tp, xmi->xmi_ip2, 0, &bp); + error = xfs_attr3_leaf_read(tp, xmi->xmi_ip2, xmi->xmi_ip2->i_ino, 0, + &bp); if (error) return error; diff --git a/fs/xfs/scrub/dabtree.c b/fs/xfs/scrub/dabtree.c index fa6385a99ac4..c71254088dff 100644 --- a/fs/xfs/scrub/dabtree.c +++ b/fs/xfs/scrub/dabtree.c @@ -320,6 +320,7 @@ xchk_da_btree_block( struct xfs_da3_blkinfo *hdr3; struct xfs_da_args *dargs = &ds->dargs; struct xfs_inode *ip = ds->dargs.dp; + xfs_failaddr_t fa; xfs_ino_t owner; int *pmaxrecs; struct xfs_da3_icnode_hdr nodehdr; @@ -442,6 +443,12 @@ xchk_da_btree_block( goto out_freebp; } + fa = xfs_da3_header_check(blk->bp, dargs->owner); + if (fa) { + xchk_da_set_corrupt(ds, level); + goto out_freebp; + } + /* * If we've been handed a block that is below the dabtree root, does * its hashval match what the parent block expected to see? diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index 42a575db7267..f6496e33ff91 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -214,6 +214,7 @@ xfs_attr_node_list_lookup( struct xfs_mount *mp = dp->i_mount; struct xfs_trans *tp = context->tp; struct xfs_buf *bp; + xfs_failaddr_t fa; int i; int error = 0; unsigned int expected_level = 0; @@ -273,6 +274,12 @@ xfs_attr_node_list_lookup( } } + fa = xfs_attr3_leaf_header_check(bp, dp->i_ino); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + goto out_releasebuf; + } + if (expected_level != 0) goto out_corruptbuf; @@ -281,6 +288,7 @@ xfs_attr_node_list_lookup( out_corruptbuf: xfs_buf_mark_corrupt(bp); +out_releasebuf: xfs_trans_brelse(tp, bp); xfs_dirattr_mark_sick(dp, XFS_ATTR_FORK); return -EFSCORRUPTED; @@ -297,6 +305,7 @@ xfs_attr_node_list( struct xfs_buf *bp; struct xfs_inode *dp = context->dp; struct xfs_mount *mp = dp->i_mount; + xfs_failaddr_t fa; int error = 0; trace_xfs_attr_node_list(context); @@ -332,6 +341,14 @@ xfs_attr_node_list( case XFS_ATTR_LEAF_MAGIC: case XFS_ATTR3_LEAF_MAGIC: leaf = bp->b_addr; + fa = xfs_attr3_leaf_header_check(bp, dp->i_ino); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_trans_brelse(context->tp, bp); + xfs_dirattr_mark_sick(dp, XFS_ATTR_FORK); + bp = NULL; + break; + } xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &leafhdr, leaf); entries = xfs_attr3_leaf_entryp(leaf); @@ -382,8 +399,8 @@ need_lookup: break; cursor->blkno = leafhdr.forw; xfs_trans_brelse(context->tp, bp); - error = xfs_attr3_leaf_read(context->tp, dp, cursor->blkno, - &bp); + error = xfs_attr3_leaf_read(context->tp, dp, dp->i_ino, + cursor->blkno, &bp); if (error) return error; } @@ -503,7 +520,8 @@ xfs_attr_leaf_list( trace_xfs_attr_leaf_list(context); context->cursor.blkno = 0; - error = xfs_attr3_leaf_read(context->tp, context->dp, 0, &bp); + error = xfs_attr3_leaf_read(context->tp, context->dp, + context->dp->i_ino, 0, &bp); if (error) return error; -- cgit v1.2.3-59-g8ed1b From 8c25dc728bd1ca9344001aa6ef4556885572baa4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:37 -0700 Subject: xfs: validate attr remote value buffer owners Check the owner field of xattr remote value blocks. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_remote.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 024895cc7029..a8de9dc1e998 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -280,12 +280,12 @@ xfs_attr_rmtval_copyout( struct xfs_mount *mp, struct xfs_buf *bp, struct xfs_inode *dp, + xfs_ino_t owner, int *offset, int *valuelen, uint8_t **dst) { char *src = bp->b_addr; - xfs_ino_t ino = dp->i_ino; xfs_daddr_t bno = xfs_buf_daddr(bp); int len = BBTOB(bp->b_length); int blksize = mp->m_attr_geo->blksize; @@ -299,11 +299,11 @@ xfs_attr_rmtval_copyout( byte_cnt = min(*valuelen, byte_cnt); if (xfs_has_crc(mp)) { - if (xfs_attr3_rmt_hdr_ok(src, ino, *offset, + if (xfs_attr3_rmt_hdr_ok(src, owner, *offset, byte_cnt, bno)) { xfs_alert(mp, "remote attribute header mismatch bno/off/len/owner (0x%llx/0x%x/Ox%x/0x%llx)", - bno, *offset, byte_cnt, ino); + bno, *offset, byte_cnt, owner); xfs_dirattr_mark_sick(dp, XFS_ATTR_FORK); return -EFSCORRUPTED; } @@ -427,8 +427,7 @@ xfs_attr_rmtval_get( return error; error = xfs_attr_rmtval_copyout(mp, bp, args->dp, - &offset, &valuelen, - &dst); + args->owner, &offset, &valuelen, &dst); xfs_buf_relse(bp); if (error) return error; -- cgit v1.2.3-59-g8ed1b From d44bea9b41ca25f91fd9f25ed2cc3bb2f6dab4bc Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:38 -0700 Subject: xfs: validate dabtree node buffer owners Check the owner field of dabtree node blocks. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_da_btree.c | 109 +++++++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_da_btree.h | 1 + fs/xfs/xfs_attr_list.c | 9 ++++ 3 files changed, 119 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index 65eef8775187..e6c28bccdbc0 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -252,6 +252,26 @@ xfs_da3_node_verify( return NULL; } +xfs_failaddr_t +xfs_da3_node_header_check( + struct xfs_buf *bp, + xfs_ino_t owner) +{ + struct xfs_mount *mp = bp->b_mount; + + if (xfs_has_crc(mp)) { + struct xfs_da3_blkinfo *hdr3 = bp->b_addr; + + if (hdr3->hdr.magic != cpu_to_be16(XFS_DA3_NODE_MAGIC)) + return __this_address; + + if (be64_to_cpu(hdr3->owner) != owner) + return __this_address; + } + + return NULL; +} + xfs_failaddr_t xfs_da3_header_check( struct xfs_buf *bp, @@ -266,6 +286,8 @@ xfs_da3_header_check( switch (hdr->magic) { case cpu_to_be16(XFS_ATTR3_LEAF_MAGIC): return xfs_attr3_leaf_header_check(bp, owner); + case cpu_to_be16(XFS_DA3_NODE_MAGIC): + return xfs_da3_node_header_check(bp, owner); } return NULL; @@ -1218,6 +1240,7 @@ xfs_da3_root_join( struct xfs_da3_icnode_hdr oldroothdr; int error; struct xfs_inode *dp = state->args->dp; + xfs_failaddr_t fa; trace_xfs_da_root_join(state->args); @@ -1244,6 +1267,13 @@ xfs_da3_root_join( error = xfs_da3_node_read(args->trans, dp, child, &bp, args->whichfork); if (error) return error; + fa = xfs_da3_header_check(bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_trans_brelse(args->trans, bp); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } xfs_da_blkinfo_onlychild_validate(bp->b_addr, oldroothdr.level); /* @@ -1278,6 +1308,7 @@ xfs_da3_node_toosmall( struct xfs_da_blkinfo *info; xfs_dablk_t blkno; struct xfs_buf *bp; + xfs_failaddr_t fa; struct xfs_da3_icnode_hdr nodehdr; int count; int forward; @@ -1352,6 +1383,13 @@ xfs_da3_node_toosmall( state->args->whichfork); if (error) return error; + fa = xfs_da3_node_header_check(bp, state->args->owner); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_trans_brelse(state->args->trans, bp); + xfs_da_mark_sick(state->args); + return -EFSCORRUPTED; + } node = bp->b_addr; xfs_da3_node_hdr_from_disk(dp->i_mount, &thdr, node); @@ -1674,6 +1712,13 @@ xfs_da3_node_lookup_int( return -EFSCORRUPTED; } + fa = xfs_da3_node_header_check(blk->bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(blk->bp, fa); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } + blk->magic = XFS_DA_NODE_MAGIC; /* @@ -1846,6 +1891,7 @@ xfs_da3_blk_link( struct xfs_da_blkinfo *tmp_info; struct xfs_da_args *args; struct xfs_buf *bp; + xfs_failaddr_t fa; int before = 0; int error; struct xfs_inode *dp = state->args->dp; @@ -1889,6 +1935,13 @@ xfs_da3_blk_link( &bp, args->whichfork); if (error) return error; + fa = xfs_da3_header_check(bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_trans_brelse(args->trans, bp); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } ASSERT(bp != NULL); tmp_info = bp->b_addr; ASSERT(tmp_info->magic == old_info->magic); @@ -1910,6 +1963,13 @@ xfs_da3_blk_link( &bp, args->whichfork); if (error) return error; + fa = xfs_da3_header_check(bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_trans_brelse(args->trans, bp); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } ASSERT(bp != NULL); tmp_info = bp->b_addr; ASSERT(tmp_info->magic == old_info->magic); @@ -1939,6 +1999,7 @@ xfs_da3_blk_unlink( struct xfs_da_blkinfo *tmp_info; struct xfs_da_args *args; struct xfs_buf *bp; + xfs_failaddr_t fa; int error; /* @@ -1969,6 +2030,13 @@ xfs_da3_blk_unlink( &bp, args->whichfork); if (error) return error; + fa = xfs_da3_header_check(bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_trans_brelse(args->trans, bp); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } ASSERT(bp != NULL); tmp_info = bp->b_addr; ASSERT(tmp_info->magic == save_info->magic); @@ -1986,6 +2054,13 @@ xfs_da3_blk_unlink( &bp, args->whichfork); if (error) return error; + fa = xfs_da3_header_check(bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_trans_brelse(args->trans, bp); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } ASSERT(bp != NULL); tmp_info = bp->b_addr; ASSERT(tmp_info->magic == save_info->magic); @@ -2101,6 +2176,12 @@ xfs_da3_path_shift( switch (be16_to_cpu(info->magic)) { case XFS_DA_NODE_MAGIC: case XFS_DA3_NODE_MAGIC: + fa = xfs_da3_node_header_check(blk->bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(blk->bp, fa); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } blk->magic = XFS_DA_NODE_MAGIC; xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, bp->b_addr); @@ -2406,6 +2487,13 @@ xfs_da3_swap_lastblock( error = xfs_da3_node_read(tp, dp, sib_blkno, &sib_buf, w); if (error) goto done; + fa = xfs_da3_header_check(sib_buf, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(sib_buf, fa); + xfs_da_mark_sick(args); + error = -EFSCORRUPTED; + goto done; + } sib_info = sib_buf->b_addr; if (XFS_IS_CORRUPT(mp, be32_to_cpu(sib_info->forw) != last_blkno || @@ -2427,6 +2515,13 @@ xfs_da3_swap_lastblock( error = xfs_da3_node_read(tp, dp, sib_blkno, &sib_buf, w); if (error) goto done; + fa = xfs_da3_header_check(sib_buf, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(sib_buf, fa); + xfs_da_mark_sick(args); + error = -EFSCORRUPTED; + goto done; + } sib_info = sib_buf->b_addr; if (XFS_IS_CORRUPT(mp, be32_to_cpu(sib_info->back) != last_blkno || @@ -2450,6 +2545,13 @@ xfs_da3_swap_lastblock( error = xfs_da3_node_read(tp, dp, par_blkno, &par_buf, w); if (error) goto done; + fa = xfs_da3_node_header_check(par_buf, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(par_buf, fa); + xfs_da_mark_sick(args); + error = -EFSCORRUPTED; + goto done; + } par_node = par_buf->b_addr; xfs_da3_node_hdr_from_disk(dp->i_mount, &par_hdr, par_node); if (XFS_IS_CORRUPT(mp, @@ -2499,6 +2601,13 @@ xfs_da3_swap_lastblock( error = xfs_da3_node_read(tp, dp, par_blkno, &par_buf, w); if (error) goto done; + fa = xfs_da3_node_header_check(par_buf, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(par_buf, fa); + xfs_da_mark_sick(args); + error = -EFSCORRUPTED; + goto done; + } par_node = par_buf->b_addr; xfs_da3_node_hdr_from_disk(dp->i_mount, &par_hdr, par_node); if (XFS_IS_CORRUPT(mp, par_hdr.level != level)) { diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 99618e0c8a72..7a004786ee0a 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -237,6 +237,7 @@ void xfs_da3_node_hdr_from_disk(struct xfs_mount *mp, void xfs_da3_node_hdr_to_disk(struct xfs_mount *mp, struct xfs_da_intnode *to, struct xfs_da3_icnode_hdr *from); xfs_failaddr_t xfs_da3_header_check(struct xfs_buf *bp, xfs_ino_t owner); +xfs_failaddr_t xfs_da3_node_header_check(struct xfs_buf *bp, xfs_ino_t owner); extern struct kmem_cache *xfs_da_state_cache; diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index f6496e33ff91..6a621f016f04 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -239,6 +239,10 @@ xfs_attr_node_list_lookup( goto out_corruptbuf; } + fa = xfs_da3_node_header_check(bp, dp->i_ino); + if (fa) + goto out_corruptbuf; + xfs_da3_node_hdr_from_disk(mp, &nodehdr, node); /* Tree taller than we can handle; bail out! */ @@ -335,6 +339,11 @@ xfs_attr_node_list( case XFS_DA_NODE_MAGIC: case XFS_DA3_NODE_MAGIC: trace_xfs_attr_list_wrong_blk(context); + fa = xfs_da3_node_header_check(bp, dp->i_ino); + if (fa) { + __xfs_buf_mark_corrupt(bp, fa); + xfs_dirattr_mark_sick(dp, XFS_ATTR_FORK); + } xfs_trans_brelse(context->tp, bp); bp = NULL; break; -- cgit v1.2.3-59-g8ed1b From 402eef10a1bab0b428c418cfbaaa0a62efc9c951 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:39 -0700 Subject: xfs: validate directory leaf buffer owners Check the owner field of directory leaf blocks. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_da_btree.c | 16 +++++++++++ fs/xfs/libxfs/xfs_dir2.h | 2 ++ fs/xfs/libxfs/xfs_dir2_leaf.c | 65 +++++++++++++++++++++++++++++++++++++++---- fs/xfs/libxfs/xfs_dir2_node.c | 3 +- fs/xfs/libxfs/xfs_dir2_priv.h | 4 +-- fs/xfs/scrub/dir.c | 2 +- 6 files changed, 82 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index e6c28bccdbc0..b13796629e22 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -288,8 +288,12 @@ xfs_da3_header_check( return xfs_attr3_leaf_header_check(bp, owner); case cpu_to_be16(XFS_DA3_NODE_MAGIC): return xfs_da3_node_header_check(bp, owner); + case cpu_to_be16(XFS_DIR3_LEAF1_MAGIC): + case cpu_to_be16(XFS_DIR3_LEAFN_MAGIC): + return xfs_dir3_leaf_header_check(bp, owner); } + ASSERT(0); return NULL; } @@ -1700,6 +1704,12 @@ xfs_da3_node_lookup_int( if (magic == XFS_DIR2_LEAFN_MAGIC || magic == XFS_DIR3_LEAFN_MAGIC) { + fa = xfs_dir3_leaf_header_check(blk->bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(blk->bp, fa); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } blk->magic = XFS_DIR2_LEAFN_MAGIC; blk->hashval = xfs_dir2_leaf_lasthash(args->dp, blk->bp, NULL); @@ -2208,6 +2218,12 @@ xfs_da3_path_shift( break; case XFS_DIR2_LEAFN_MAGIC: case XFS_DIR3_LEAFN_MAGIC: + fa = xfs_dir3_leaf_header_check(blk->bp, args->owner); + if (fa) { + __xfs_buf_mark_corrupt(blk->bp, fa); + xfs_da_mark_sick(args); + return -EFSCORRUPTED; + } blk->magic = XFS_DIR2_LEAFN_MAGIC; ASSERT(level == path->active-1); blk->index = 0; diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index 8497d041f316..2f728c26a416 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -101,6 +101,8 @@ extern struct xfs_dir2_data_free *xfs_dir2_data_freefind( extern int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino); +xfs_failaddr_t xfs_dir3_leaf_header_check(struct xfs_buf *bp, xfs_ino_t owner); + extern const struct xfs_buf_ops xfs_dir3_block_buf_ops; extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops; extern const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops; diff --git a/fs/xfs/libxfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c index 20ce057d12e8..53b808e2a5f0 100644 --- a/fs/xfs/libxfs/xfs_dir2_leaf.c +++ b/fs/xfs/libxfs/xfs_dir2_leaf.c @@ -208,6 +208,29 @@ xfs_dir3_leaf_verify( return xfs_dir3_leaf_check_int(mp, &leafhdr, bp->b_addr, true); } +xfs_failaddr_t +xfs_dir3_leaf_header_check( + struct xfs_buf *bp, + xfs_ino_t owner) +{ + struct xfs_mount *mp = bp->b_mount; + + if (xfs_has_crc(mp)) { + struct xfs_dir3_leaf *hdr3 = bp->b_addr; + + if (hdr3->hdr.info.hdr.magic != + cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) && + hdr3->hdr.info.hdr.magic != + cpu_to_be16(XFS_DIR3_LEAFN_MAGIC)) + return __this_address; + + if (be64_to_cpu(hdr3->hdr.info.owner) != owner) + return __this_address; + } + + return NULL; +} + static void xfs_dir3_leaf_read_verify( struct xfs_buf *bp) @@ -271,32 +294,60 @@ int xfs_dir3_leaf_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t fbno, struct xfs_buf **bpp) { + xfs_failaddr_t fa; int err; err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK, &xfs_dir3_leaf1_buf_ops); - if (!err && tp && *bpp) + if (err || !(*bpp)) + return err; + + fa = xfs_dir3_leaf_header_check(*bpp, owner); + if (fa) { + __xfs_buf_mark_corrupt(*bpp, fa); + xfs_trans_brelse(tp, *bpp); + *bpp = NULL; + xfs_dirattr_mark_sick(dp, XFS_DATA_FORK); + return -EFSCORRUPTED; + } + + if (tp) xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAF1_BUF); - return err; + return 0; } int xfs_dir3_leafn_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t fbno, struct xfs_buf **bpp) { + xfs_failaddr_t fa; int err; err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops); - if (!err && tp && *bpp) + if (err || !(*bpp)) + return err; + + fa = xfs_dir3_leaf_header_check(*bpp, owner); + if (fa) { + __xfs_buf_mark_corrupt(*bpp, fa); + xfs_trans_brelse(tp, *bpp); + *bpp = NULL; + xfs_dirattr_mark_sick(dp, XFS_DATA_FORK); + return -EFSCORRUPTED; + } + + if (tp) xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAFN_BUF); - return err; + return 0; } /* @@ -646,7 +697,8 @@ xfs_dir2_leaf_addname( trace_xfs_dir2_leaf_addname(args); - error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, &lbp); + error = xfs_dir3_leaf_read(tp, dp, args->owner, args->geo->leafblk, + &lbp); if (error) return error; @@ -1237,7 +1289,8 @@ xfs_dir2_leaf_lookup_int( tp = args->trans; mp = dp->i_mount; - error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, &lbp); + error = xfs_dir3_leaf_read(tp, dp, args->owner, args->geo->leafblk, + &lbp); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c index 1ad7405f9c38..e21965788188 100644 --- a/fs/xfs/libxfs/xfs_dir2_node.c +++ b/fs/xfs/libxfs/xfs_dir2_node.c @@ -1562,7 +1562,8 @@ xfs_dir2_leafn_toosmall( /* * Read the sibling leaf block. */ - error = xfs_dir3_leafn_read(state->args->trans, dp, blkno, &bp); + error = xfs_dir3_leafn_read(state->args->trans, dp, + state->args->owner, blkno, &bp); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h index 1db2e60ba827..2f0e3ad47b37 100644 --- a/fs/xfs/libxfs/xfs_dir2_priv.h +++ b/fs/xfs/libxfs/xfs_dir2_priv.h @@ -95,9 +95,9 @@ void xfs_dir2_leaf_hdr_from_disk(struct xfs_mount *mp, void xfs_dir2_leaf_hdr_to_disk(struct xfs_mount *mp, struct xfs_dir2_leaf *to, struct xfs_dir3_icleaf_hdr *from); int xfs_dir3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp, - xfs_dablk_t fbno, struct xfs_buf **bpp); + xfs_ino_t owner, xfs_dablk_t fbno, struct xfs_buf **bpp); int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp, - xfs_dablk_t fbno, struct xfs_buf **bpp); + xfs_ino_t owner, xfs_dablk_t fbno, struct xfs_buf **bpp); extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args, struct xfs_buf *dbp); extern int xfs_dir2_leaf_addname(struct xfs_da_args *args); diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 042e28547e04..d94e265a8e1f 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -470,7 +470,7 @@ xchk_directory_leaf1_bestfree( int error; /* Read the free space block. */ - error = xfs_dir3_leaf_read(sc->tp, sc->ip, lblk, &bp); + error = xfs_dir3_leaf_read(sc->tp, sc->ip, sc->ip->i_ino, lblk, &bp); if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error)) return error; xchk_buffer_recheck(sc, bp); -- cgit v1.2.3-59-g8ed1b From cc6740ddb423db2066f7669eaaa377fdbf84ab1e Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:40 -0700 Subject: xfs: validate explicit directory data buffer owners Port the existing directory data header checking function to accept an owner number instead of an xfs_inode, then update the callsites to use xfs_da_args.owner when possible. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_dir2.h | 1 + fs/xfs/libxfs/xfs_dir2_block.c | 3 ++- fs/xfs/libxfs/xfs_dir2_data.c | 16 ++++++++++------ fs/xfs/libxfs/xfs_dir2_leaf.c | 21 +++++++++++---------- fs/xfs/libxfs/xfs_dir2_node.c | 7 +++---- fs/xfs/libxfs/xfs_dir2_priv.h | 3 ++- fs/xfs/scrub/dir.c | 14 +++++++------- fs/xfs/scrub/readdir.c | 2 +- fs/xfs/xfs_dir2_readdir.c | 3 ++- 9 files changed, 39 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index 2f728c26a416..d623bfdcd421 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -102,6 +102,7 @@ extern struct xfs_dir2_data_free *xfs_dir2_data_freefind( extern int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino); xfs_failaddr_t xfs_dir3_leaf_header_check(struct xfs_buf *bp, xfs_ino_t owner); +xfs_failaddr_t xfs_dir3_data_header_check(struct xfs_buf *bp, xfs_ino_t owner); extern const struct xfs_buf_ops xfs_dir3_block_buf_ops; extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops; diff --git a/fs/xfs/libxfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c index 61cbc668f228..b20b08394aa0 100644 --- a/fs/xfs/libxfs/xfs_dir2_block.c +++ b/fs/xfs/libxfs/xfs_dir2_block.c @@ -982,7 +982,8 @@ xfs_dir2_leaf_to_block( * Read the data block if we don't already have it, give up if it fails. */ if (!dbp) { - error = xfs_dir3_data_read(tp, dp, args->geo->datablk, 0, &dbp); + error = xfs_dir3_data_read(tp, dp, args->owner, + args->geo->datablk, 0, &dbp); if (error) return error; } diff --git a/fs/xfs/libxfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c index c3ef720b5ff6..ea0b9628df18 100644 --- a/fs/xfs/libxfs/xfs_dir2_data.c +++ b/fs/xfs/libxfs/xfs_dir2_data.c @@ -395,17 +395,20 @@ static const struct xfs_buf_ops xfs_dir3_data_reada_buf_ops = { .verify_write = xfs_dir3_data_write_verify, }; -static xfs_failaddr_t +xfs_failaddr_t xfs_dir3_data_header_check( - struct xfs_inode *dp, - struct xfs_buf *bp) + struct xfs_buf *bp, + xfs_ino_t owner) { - struct xfs_mount *mp = dp->i_mount; + struct xfs_mount *mp = bp->b_mount; if (xfs_has_crc(mp)) { struct xfs_dir3_data_hdr *hdr3 = bp->b_addr; - if (be64_to_cpu(hdr3->hdr.owner) != dp->i_ino) + if (hdr3->hdr.magic != cpu_to_be32(XFS_DIR3_DATA_MAGIC)) + return __this_address; + + if (be64_to_cpu(hdr3->hdr.owner) != owner) return __this_address; } @@ -416,6 +419,7 @@ int xfs_dir3_data_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp) @@ -429,7 +433,7 @@ xfs_dir3_data_read( return err; /* Check things that we can't do in the verifier. */ - fa = xfs_dir3_data_header_check(dp, *bpp); + fa = xfs_dir3_data_header_check(*bpp, owner); if (fa) { __xfs_buf_mark_corrupt(*bpp, fa); xfs_trans_brelse(tp, *bpp); diff --git a/fs/xfs/libxfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c index 53b808e2a5f0..0b1b852f6178 100644 --- a/fs/xfs/libxfs/xfs_dir2_leaf.c +++ b/fs/xfs/libxfs/xfs_dir2_leaf.c @@ -885,9 +885,9 @@ xfs_dir2_leaf_addname( * Already had space in some data block. * Just read that one in. */ - error = xfs_dir3_data_read(tp, dp, - xfs_dir2_db_to_da(args->geo, use_block), - 0, &dbp); + error = xfs_dir3_data_read(tp, dp, args->owner, + xfs_dir2_db_to_da(args->geo, use_block), 0, + &dbp); if (error) { xfs_trans_brelse(tp, lbp); return error; @@ -1328,9 +1328,9 @@ xfs_dir2_leaf_lookup_int( if (newdb != curdb) { if (dbp) xfs_trans_brelse(tp, dbp); - error = xfs_dir3_data_read(tp, dp, - xfs_dir2_db_to_da(args->geo, newdb), - 0, &dbp); + error = xfs_dir3_data_read(tp, dp, args->owner, + xfs_dir2_db_to_da(args->geo, newdb), 0, + &dbp); if (error) { xfs_trans_brelse(tp, lbp); return error; @@ -1370,9 +1370,9 @@ xfs_dir2_leaf_lookup_int( ASSERT(cidb != -1); if (cidb != curdb) { xfs_trans_brelse(tp, dbp); - error = xfs_dir3_data_read(tp, dp, - xfs_dir2_db_to_da(args->geo, cidb), - 0, &dbp); + error = xfs_dir3_data_read(tp, dp, args->owner, + xfs_dir2_db_to_da(args->geo, cidb), 0, + &dbp); if (error) { xfs_trans_brelse(tp, lbp); return error; @@ -1666,7 +1666,8 @@ xfs_dir2_leaf_trim_data( /* * Read the offending data block. We need its buffer. */ - error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(geo, db), 0, &dbp); + error = xfs_dir3_data_read(tp, dp, args->owner, + xfs_dir2_db_to_da(geo, db), 0, &dbp); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c index e21965788188..dc85197b8448 100644 --- a/fs/xfs/libxfs/xfs_dir2_node.c +++ b/fs/xfs/libxfs/xfs_dir2_node.c @@ -863,7 +863,7 @@ xfs_dir2_leafn_lookup_for_entry( ASSERT(state->extravalid); curbp = state->extrablk.bp; } else { - error = xfs_dir3_data_read(tp, dp, + error = xfs_dir3_data_read(tp, dp, args->owner, xfs_dir2_db_to_da(args->geo, newdb), 0, &curbp); @@ -1949,9 +1949,8 @@ xfs_dir2_node_addname_int( &freehdr, &findex); } else { /* Read the data block in. */ - error = xfs_dir3_data_read(tp, dp, - xfs_dir2_db_to_da(args->geo, dbno), - 0, &dbp); + error = xfs_dir3_data_read(tp, dp, args->owner, + xfs_dir2_db_to_da(args->geo, dbno), 0, &dbp); } if (error) return error; diff --git a/fs/xfs/libxfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h index 2f0e3ad47b37..879aa2e9fd73 100644 --- a/fs/xfs/libxfs/xfs_dir2_priv.h +++ b/fs/xfs/libxfs/xfs_dir2_priv.h @@ -78,7 +78,8 @@ extern void xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp); extern xfs_failaddr_t __xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp); int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp, - xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp); + xfs_ino_t owner, xfs_dablk_t bno, unsigned int flags, + struct xfs_buf **bpp); int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno, unsigned int flags); diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index d94e265a8e1f..6b572196bb43 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -196,8 +196,8 @@ xchk_dir_rec( xchk_da_set_corrupt(ds, level); goto out; } - error = xfs_dir3_data_read(ds->dargs.trans, dp, rec_bno, - XFS_DABUF_MAP_HOLE_OK, &bp); + error = xfs_dir3_data_read(ds->dargs.trans, dp, ds->dargs.owner, + rec_bno, XFS_DABUF_MAP_HOLE_OK, &bp); if (!xchk_fblock_process_error(ds->sc, XFS_DATA_FORK, rec_bno, &error)) goto out; @@ -318,7 +318,8 @@ xchk_directory_data_bestfree( error = xfs_dir3_block_read(sc->tp, sc->ip, &bp); } else { /* dir data format */ - error = xfs_dir3_data_read(sc->tp, sc->ip, lblk, 0, &bp); + error = xfs_dir3_data_read(sc->tp, sc->ip, sc->ip->i_ino, lblk, + 0, &bp); } if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error)) goto out; @@ -531,10 +532,9 @@ xchk_directory_leaf1_bestfree( /* Check all the bestfree entries. */ for (i = 0; i < bestcount; i++, bestp++) { best = be16_to_cpu(*bestp); - error = xfs_dir3_data_read(sc->tp, sc->ip, + error = xfs_dir3_data_read(sc->tp, sc->ip, args->owner, xfs_dir2_db_to_da(args->geo, i), - XFS_DABUF_MAP_HOLE_OK, - &dbp); + XFS_DABUF_MAP_HOLE_OK, &dbp); if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error)) break; @@ -597,7 +597,7 @@ xchk_directory_free_bestfree( stale++; continue; } - error = xfs_dir3_data_read(sc->tp, sc->ip, + error = xfs_dir3_data_read(sc->tp, sc->ip, args->owner, (freehdr.firstdb + i) * args->geo->fsbcount, 0, &dbp); if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, diff --git a/fs/xfs/scrub/readdir.c b/fs/xfs/scrub/readdir.c index fb98b7624994..bed15a9524a2 100644 --- a/fs/xfs/scrub/readdir.c +++ b/fs/xfs/scrub/readdir.c @@ -175,7 +175,7 @@ xchk_read_leaf_dir_buf( if (new_off > *curoff) *curoff = new_off; - return xfs_dir3_data_read(tp, dp, map.br_startoff, 0, bpp); + return xfs_dir3_data_read(tp, dp, dp->i_ino, map.br_startoff, 0, bpp); } /* Call a function for every entry in a leaf directory. */ diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c index 4e811fa393ad..2c03371b542a 100644 --- a/fs/xfs/xfs_dir2_readdir.c +++ b/fs/xfs/xfs_dir2_readdir.c @@ -282,7 +282,8 @@ xfs_dir2_leaf_readbuf( new_off = xfs_dir2_da_to_byte(geo, map.br_startoff); if (new_off > *cur_off) *cur_off = new_off; - error = xfs_dir3_data_read(args->trans, dp, map.br_startoff, 0, &bp); + error = xfs_dir3_data_read(args->trans, dp, args->owner, + map.br_startoff, 0, &bp); if (error) goto out; -- cgit v1.2.3-59-g8ed1b From 29b41ce919b7f0b0c2220e088e450d9b132bec36 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:41 -0700 Subject: xfs: validate explicit directory block buffer owners Port the existing directory block header checking function to accept an owner number instead of an xfs_inode, then update the callsites to use xfs_da_args.owner when possible. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_dir2.h | 1 + fs/xfs/libxfs/xfs_dir2_block.c | 20 ++++++++++++-------- fs/xfs/libxfs/xfs_dir2_priv.h | 4 ++-- fs/xfs/libxfs/xfs_exchmaps.c | 2 +- fs/xfs/scrub/dir.c | 2 +- fs/xfs/scrub/readdir.c | 2 +- fs/xfs/xfs_dir2_readdir.c | 2 +- 7 files changed, 19 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index d623bfdcd421..eb3a5c35025b 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -103,6 +103,7 @@ extern int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino); xfs_failaddr_t xfs_dir3_leaf_header_check(struct xfs_buf *bp, xfs_ino_t owner); xfs_failaddr_t xfs_dir3_data_header_check(struct xfs_buf *bp, xfs_ino_t owner); +xfs_failaddr_t xfs_dir3_block_header_check(struct xfs_buf *bp, xfs_ino_t owner); extern const struct xfs_buf_ops xfs_dir3_block_buf_ops; extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops; diff --git a/fs/xfs/libxfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c index b20b08394aa0..0f93ed1a4a74 100644 --- a/fs/xfs/libxfs/xfs_dir2_block.c +++ b/fs/xfs/libxfs/xfs_dir2_block.c @@ -115,17 +115,20 @@ const struct xfs_buf_ops xfs_dir3_block_buf_ops = { .verify_struct = xfs_dir3_block_verify, }; -static xfs_failaddr_t +xfs_failaddr_t xfs_dir3_block_header_check( - struct xfs_inode *dp, - struct xfs_buf *bp) + struct xfs_buf *bp, + xfs_ino_t owner) { - struct xfs_mount *mp = dp->i_mount; + struct xfs_mount *mp = bp->b_mount; if (xfs_has_crc(mp)) { struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr; - if (be64_to_cpu(hdr3->owner) != dp->i_ino) + if (hdr3->magic != cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) + return __this_address; + + if (be64_to_cpu(hdr3->owner) != owner) return __this_address; } @@ -136,6 +139,7 @@ int xfs_dir3_block_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, struct xfs_buf **bpp) { struct xfs_mount *mp = dp->i_mount; @@ -148,7 +152,7 @@ xfs_dir3_block_read( return err; /* Check things that we can't do in the verifier. */ - fa = xfs_dir3_block_header_check(dp, *bpp); + fa = xfs_dir3_block_header_check(*bpp, owner); if (fa) { __xfs_buf_mark_corrupt(*bpp, fa); xfs_trans_brelse(tp, *bpp); @@ -383,7 +387,7 @@ xfs_dir2_block_addname( tp = args->trans; /* Read the (one and only) directory block into bp. */ - error = xfs_dir3_block_read(tp, dp, &bp); + error = xfs_dir3_block_read(tp, dp, args->owner, &bp); if (error) return error; @@ -698,7 +702,7 @@ xfs_dir2_block_lookup_int( dp = args->dp; tp = args->trans; - error = xfs_dir3_block_read(tp, dp, &bp); + error = xfs_dir3_block_read(tp, dp, args->owner, &bp); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h index 879aa2e9fd73..adbc544c9bef 100644 --- a/fs/xfs/libxfs/xfs_dir2_priv.h +++ b/fs/xfs/libxfs/xfs_dir2_priv.h @@ -50,8 +50,8 @@ extern int xfs_dir_cilookup_result(struct xfs_da_args *args, /* xfs_dir2_block.c */ -extern int xfs_dir3_block_read(struct xfs_trans *tp, struct xfs_inode *dp, - struct xfs_buf **bpp); +int xfs_dir3_block_read(struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, struct xfs_buf **bpp); extern int xfs_dir2_block_addname(struct xfs_da_args *args); extern int xfs_dir2_block_lookup(struct xfs_da_args *args); extern int xfs_dir2_block_removename(struct xfs_da_args *args); diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index 9c9cf2e998b2..3880ae32eecf 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -476,7 +476,7 @@ xfs_exchmaps_dir_to_sf( if (!isblock) return 0; - error = xfs_dir3_block_read(tp, xmi->xmi_ip2, &bp); + error = xfs_dir3_block_read(tp, xmi->xmi_ip2, xmi->xmi_ip2->i_ino, &bp); if (error) return error; diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 6b572196bb43..43f5bc8ce0d4 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -315,7 +315,7 @@ xchk_directory_data_bestfree( /* dir block format */ if (lblk != XFS_B_TO_FSBT(mp, XFS_DIR2_DATA_OFFSET)) xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk); - error = xfs_dir3_block_read(sc->tp, sc->ip, &bp); + error = xfs_dir3_block_read(sc->tp, sc->ip, sc->ip->i_ino, &bp); } else { /* dir data format */ error = xfs_dir3_data_read(sc->tp, sc->ip, sc->ip->i_ino, lblk, diff --git a/fs/xfs/scrub/readdir.c b/fs/xfs/scrub/readdir.c index bed15a9524a2..e94080469315 100644 --- a/fs/xfs/scrub/readdir.c +++ b/fs/xfs/scrub/readdir.c @@ -99,7 +99,7 @@ xchk_dir_walk_block( unsigned int off, next_off, end; int error; - error = xfs_dir3_block_read(sc->tp, dp, &bp); + error = xfs_dir3_block_read(sc->tp, dp, dp->i_ino, &bp); if (error) return error; diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c index 2c03371b542a..b3abad5a6cd8 100644 --- a/fs/xfs/xfs_dir2_readdir.c +++ b/fs/xfs/xfs_dir2_readdir.c @@ -157,7 +157,7 @@ xfs_dir2_block_getdents( if (xfs_dir2_dataptr_to_db(geo, ctx->pos) > geo->datablk) return 0; - error = xfs_dir3_block_read(args->trans, dp, &bp); + error = xfs_dir3_block_read(args->trans, dp, args->owner, &bp); if (error) return error; -- cgit v1.2.3-59-g8ed1b From fe6c9f8e48e0dcbfc3dba17edd88490c8579b34b Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:41 -0700 Subject: xfs: validate explicit directory free block owners Port the existing directory freespace block header checking function to accept an owner number instead of an xfs_inode, then update the callsites to use xfs_da_args.owner when possible. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_dir2_leaf.c | 3 ++- fs/xfs/libxfs/xfs_dir2_node.c | 32 ++++++++++++++++++-------------- fs/xfs/libxfs/xfs_dir2_priv.h | 4 ++-- fs/xfs/scrub/dir.c | 2 +- 4 files changed, 23 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c index 0b1b852f6178..71c2f22a3f6e 100644 --- a/fs/xfs/libxfs/xfs_dir2_leaf.c +++ b/fs/xfs/libxfs/xfs_dir2_leaf.c @@ -1806,7 +1806,8 @@ xfs_dir2_node_to_leaf( /* * Read the freespace block. */ - error = xfs_dir2_free_read(tp, dp, args->geo->freeblk, &fbp); + error = xfs_dir2_free_read(tp, dp, args->owner, args->geo->freeblk, + &fbp); if (error) return error; xfs_dir2_free_hdr_from_disk(mp, &freehdr, fbp->b_addr); diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c index dc85197b8448..fe8d4fa13128 100644 --- a/fs/xfs/libxfs/xfs_dir2_node.c +++ b/fs/xfs/libxfs/xfs_dir2_node.c @@ -175,11 +175,11 @@ const struct xfs_buf_ops xfs_dir3_free_buf_ops = { /* Everything ok in the free block header? */ static xfs_failaddr_t xfs_dir3_free_header_check( - struct xfs_inode *dp, - xfs_dablk_t fbno, - struct xfs_buf *bp) + struct xfs_buf *bp, + xfs_ino_t owner, + xfs_dablk_t fbno) { - struct xfs_mount *mp = dp->i_mount; + struct xfs_mount *mp = bp->b_mount; int maxbests = mp->m_dir_geo->free_max_bests; unsigned int firstdb; @@ -195,7 +195,7 @@ xfs_dir3_free_header_check( return __this_address; if (be32_to_cpu(hdr3->nvalid) < be32_to_cpu(hdr3->nused)) return __this_address; - if (be64_to_cpu(hdr3->hdr.owner) != dp->i_ino) + if (be64_to_cpu(hdr3->hdr.owner) != owner) return __this_address; } else { struct xfs_dir2_free_hdr *hdr = bp->b_addr; @@ -214,6 +214,7 @@ static int __xfs_dir3_free_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t fbno, unsigned int flags, struct xfs_buf **bpp) @@ -227,7 +228,7 @@ __xfs_dir3_free_read( return err; /* Check things that we can't do in the verifier. */ - fa = xfs_dir3_free_header_check(dp, fbno, *bpp); + fa = xfs_dir3_free_header_check(*bpp, owner, fbno); if (fa) { __xfs_buf_mark_corrupt(*bpp, fa); xfs_trans_brelse(tp, *bpp); @@ -299,20 +300,23 @@ int xfs_dir2_free_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t fbno, struct xfs_buf **bpp) { - return __xfs_dir3_free_read(tp, dp, fbno, 0, bpp); + return __xfs_dir3_free_read(tp, dp, owner, fbno, 0, bpp); } static int xfs_dir2_free_try_read( struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t fbno, struct xfs_buf **bpp) { - return __xfs_dir3_free_read(tp, dp, fbno, XFS_DABUF_MAP_HOLE_OK, bpp); + return __xfs_dir3_free_read(tp, dp, owner, fbno, XFS_DABUF_MAP_HOLE_OK, + bpp); } static int @@ -717,7 +721,7 @@ xfs_dir2_leafn_lookup_for_addname( if (curbp) xfs_trans_brelse(tp, curbp); - error = xfs_dir2_free_read(tp, dp, + error = xfs_dir2_free_read(tp, dp, args->owner, xfs_dir2_db_to_da(args->geo, newfdb), &curbp); @@ -1356,8 +1360,8 @@ xfs_dir2_leafn_remove( * read in the free block. */ fdb = xfs_dir2_db_to_fdb(geo, db); - error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(geo, fdb), - &fbp); + error = xfs_dir2_free_read(tp, dp, args->owner, + xfs_dir2_db_to_da(geo, fdb), &fbp); if (error) return error; free = fbp->b_addr; @@ -1716,7 +1720,7 @@ xfs_dir2_node_add_datablk( * that was just allocated. */ fbno = xfs_dir2_db_to_fdb(args->geo, *dbno); - error = xfs_dir2_free_try_read(tp, dp, + error = xfs_dir2_free_try_read(tp, dp, args->owner, xfs_dir2_db_to_da(args->geo, fbno), &fbp); if (error) return error; @@ -1863,7 +1867,7 @@ xfs_dir2_node_find_freeblk( * so this might not succeed. This should be really rare, so * there's no reason to avoid it. */ - error = xfs_dir2_free_try_read(tp, dp, + error = xfs_dir2_free_try_read(tp, dp, args->owner, xfs_dir2_db_to_da(args->geo, fbno), &fbp); if (error) @@ -2302,7 +2306,7 @@ xfs_dir2_node_trim_free( /* * Read the freespace block. */ - error = xfs_dir2_free_try_read(tp, dp, fo, &bp); + error = xfs_dir2_free_try_read(tp, dp, args->owner, fo, &bp); if (error) return error; /* diff --git a/fs/xfs/libxfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h index adbc544c9bef..3befb32509fa 100644 --- a/fs/xfs/libxfs/xfs_dir2_priv.h +++ b/fs/xfs/libxfs/xfs_dir2_priv.h @@ -155,8 +155,8 @@ extern int xfs_dir2_node_removename(struct xfs_da_args *args); extern int xfs_dir2_node_replace(struct xfs_da_args *args); extern int xfs_dir2_node_trim_free(struct xfs_da_args *args, xfs_fileoff_t fo, int *rvalp); -extern int xfs_dir2_free_read(struct xfs_trans *tp, struct xfs_inode *dp, - xfs_dablk_t fbno, struct xfs_buf **bpp); +int xfs_dir2_free_read(struct xfs_trans *tp, struct xfs_inode *dp, + xfs_ino_t owner, xfs_dablk_t fbno, struct xfs_buf **bpp); /* xfs_dir2_sf.c */ xfs_ino_t xfs_dir2_sf_get_ino(struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr, diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 43f5bc8ce0d4..7bac74621af7 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -577,7 +577,7 @@ xchk_directory_free_bestfree( int error; /* Read the free space block */ - error = xfs_dir2_free_read(sc->tp, sc->ip, lblk, &bp); + error = xfs_dir2_free_read(sc->tp, sc->ip, sc->ip->i_ino, lblk, &bp); if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error)) return error; xchk_buffer_recheck(sc, bp); -- cgit v1.2.3-59-g8ed1b From 98339edf0750e75e216521961c98c8105a830bf8 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:42 -0700 Subject: xfs: enable discarding of folios backing an xfile Create a new xfile function to discard the page cache that's backing part of an xfile. The next patch wil use this to drop parts of an xfile that aren't needed anymore. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/trace.h | 1 + fs/xfs/scrub/xfile.c | 12 ++++++++++++ fs/xfs/scrub/xfile.h | 1 + 3 files changed, 14 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 8d05f2adae3d..7d07912d8f75 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -948,6 +948,7 @@ DEFINE_XFILE_EVENT(xfile_store); DEFINE_XFILE_EVENT(xfile_seek_data); DEFINE_XFILE_EVENT(xfile_get_folio); DEFINE_XFILE_EVENT(xfile_put_folio); +DEFINE_XFILE_EVENT(xfile_discard); TRACE_EVENT(xfarray_create, TP_PROTO(struct xfarray *xfa, unsigned long long required_capacity), diff --git a/fs/xfs/scrub/xfile.c b/fs/xfs/scrub/xfile.c index 8cdd863db585..4e254a0ba003 100644 --- a/fs/xfs/scrub/xfile.c +++ b/fs/xfs/scrub/xfile.c @@ -310,3 +310,15 @@ xfile_put_folio( folio_unlock(folio); folio_put(folio); } + +/* Discard the page cache that's backing a range of the xfile. */ +void +xfile_discard( + struct xfile *xf, + loff_t pos, + u64 count) +{ + trace_xfile_discard(xf, pos, count); + + shmem_truncate_range(file_inode(xf->file), pos, pos + count - 1); +} diff --git a/fs/xfs/scrub/xfile.h b/fs/xfs/scrub/xfile.h index 76d78dba7e34..8dfbae1fe33a 100644 --- a/fs/xfs/scrub/xfile.h +++ b/fs/xfs/scrub/xfile.h @@ -17,6 +17,7 @@ int xfile_load(struct xfile *xf, void *buf, size_t count, loff_t pos); int xfile_store(struct xfile *xf, const void *buf, size_t count, loff_t pos); +void xfile_discard(struct xfile *xf, loff_t pos, u64 count); loff_t xfile_seek_data(struct xfile *xf, loff_t pos); #define XFILE_MAX_FOLIO_SIZE (PAGE_SIZE << MAX_PAGECACHE_ORDER) -- cgit v1.2.3-59-g8ed1b From d2bd7eef4f217fc3e1ddf18b4f4eac848886bfa0 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:43 -0700 Subject: xfs: create a blob array data structure Create a simple 'blob array' data structure for storage of arbitrarily sized metadata objects that will be used to reconstruct metadata. For the intended usage (temporarily storing extended attribute names and values) we only have to support storing objects and retrieving them. Use the xfile abstraction to store the attribute information in memory that can be swapped out. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/xfblob.c | 151 ++++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/xfblob.h | 24 ++++++++ 3 files changed, 176 insertions(+) create mode 100644 fs/xfs/scrub/xfblob.c create mode 100644 fs/xfs/scrub/xfblob.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 5e3ac7ec8fa5..bc27757702fe 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -208,6 +208,7 @@ xfs-y += $(addprefix scrub/, \ repair.o \ rmap_repair.o \ tempfile.o \ + xfblob.o \ ) xfs-$(CONFIG_XFS_RT) += $(addprefix scrub/, \ diff --git a/fs/xfs/scrub/xfblob.c b/fs/xfs/scrub/xfblob.c new file mode 100644 index 000000000000..cec668debce5 --- /dev/null +++ b/fs/xfs/scrub/xfblob.c @@ -0,0 +1,151 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2021-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "scrub/scrub.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" + +/* + * XFS Blob Storage + * ================ + * Stores and retrieves blobs using an xfile. Objects are appended to the file + * and the offset is returned as a magic cookie for retrieval. + */ + +#define XB_KEY_MAGIC 0xABAADDAD +struct xb_key { + uint32_t xb_magic; /* XB_KEY_MAGIC */ + uint32_t xb_size; /* size of the blob, in bytes */ + loff_t xb_offset; /* byte offset of this key */ + /* blob comes after here */ +} __packed; + +/* Initialize a blob storage object. */ +int +xfblob_create( + const char *description, + struct xfblob **blobp) +{ + struct xfblob *blob; + struct xfile *xfile; + int error; + + error = xfile_create(description, 0, &xfile); + if (error) + return error; + + blob = kmalloc(sizeof(struct xfblob), XCHK_GFP_FLAGS); + if (!blob) { + error = -ENOMEM; + goto out_xfile; + } + + blob->xfile = xfile; + blob->last_offset = PAGE_SIZE; + + *blobp = blob; + return 0; + +out_xfile: + xfile_destroy(xfile); + return error; +} + +/* Destroy a blob storage object. */ +void +xfblob_destroy( + struct xfblob *blob) +{ + xfile_destroy(blob->xfile); + kfree(blob); +} + +/* Retrieve a blob. */ +int +xfblob_load( + struct xfblob *blob, + xfblob_cookie cookie, + void *ptr, + uint32_t size) +{ + struct xb_key key; + int error; + + error = xfile_load(blob->xfile, &key, sizeof(key), cookie); + if (error) + return error; + + if (key.xb_magic != XB_KEY_MAGIC || key.xb_offset != cookie) { + ASSERT(0); + return -ENODATA; + } + if (size < key.xb_size) { + ASSERT(0); + return -EFBIG; + } + + return xfile_load(blob->xfile, ptr, key.xb_size, + cookie + sizeof(key)); +} + +/* Store a blob. */ +int +xfblob_store( + struct xfblob *blob, + xfblob_cookie *cookie, + const void *ptr, + uint32_t size) +{ + struct xb_key key = { + .xb_offset = blob->last_offset, + .xb_magic = XB_KEY_MAGIC, + .xb_size = size, + }; + loff_t pos = blob->last_offset; + int error; + + error = xfile_store(blob->xfile, &key, sizeof(key), pos); + if (error) + return error; + + pos += sizeof(key); + error = xfile_store(blob->xfile, ptr, size, pos); + if (error) + goto out_err; + + *cookie = blob->last_offset; + blob->last_offset += sizeof(key) + size; + return 0; +out_err: + xfile_discard(blob->xfile, blob->last_offset, sizeof(key)); + return error; +} + +/* Free a blob. */ +int +xfblob_free( + struct xfblob *blob, + xfblob_cookie cookie) +{ + struct xb_key key; + int error; + + error = xfile_load(blob->xfile, &key, sizeof(key), cookie); + if (error) + return error; + + if (key.xb_magic != XB_KEY_MAGIC || key.xb_offset != cookie) { + ASSERT(0); + return -ENODATA; + } + + xfile_discard(blob->xfile, cookie, sizeof(key) + key.xb_size); + return 0; +} diff --git a/fs/xfs/scrub/xfblob.h b/fs/xfs/scrub/xfblob.h new file mode 100644 index 000000000000..bd98647407f1 --- /dev/null +++ b/fs/xfs/scrub/xfblob.h @@ -0,0 +1,24 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2021-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_XFBLOB_H__ +#define __XFS_SCRUB_XFBLOB_H__ + +struct xfblob { + struct xfile *xfile; + loff_t last_offset; +}; + +typedef loff_t xfblob_cookie; + +int xfblob_create(const char *descr, struct xfblob **blobp); +void xfblob_destroy(struct xfblob *blob); +int xfblob_load(struct xfblob *blob, xfblob_cookie cookie, void *ptr, + uint32_t size); +int xfblob_store(struct xfblob *blob, xfblob_cookie *cookie, const void *ptr, + uint32_t size); +int xfblob_free(struct xfblob *blob, xfblob_cookie cookie); + +#endif /* __XFS_SCRUB_XFBLOB_H__ */ -- cgit v1.2.3-59-g8ed1b From 629fdaf5f5b1b7f7107ed4de04e0991a99501ced Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:44 -0700 Subject: xfs: use atomic extent swapping to fix user file fork data Build on the code that was recently added to the temporary repair file code so that we can atomically switch the contents of any file fork, even if the fork is in local format. The upcoming functions to repair xattrs, directories, and symlinks will need that capability. Repair can lock out access to these user files by holding IOLOCK_EXCL on these user files. Therefore, it is safe to drop the ILOCK of both the file being repaired and the tempfile being used for staging, and cancel the scrub transaction. We do this so that we can reuse the resource estimation and transaction allocation functions used by a regular file exchange operation. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_exchmaps.c | 2 +- fs/xfs/libxfs/xfs_exchmaps.h | 1 + fs/xfs/scrub/tempexch.h | 2 + fs/xfs/scrub/tempfile.c | 204 +++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/tempfile.h | 3 + 5 files changed, 211 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index 3880ae32eecf..44ab6a9235c0 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -675,7 +675,7 @@ xfs_exchmaps_rmapbt_blocks( } /* Estimate the bmbt and rmapbt overhead required to exchange mappings. */ -static int +int xfs_exchmaps_estimate_overhead( struct xfs_exchmaps_req *req) { diff --git a/fs/xfs/libxfs/xfs_exchmaps.h b/fs/xfs/libxfs/xfs_exchmaps.h index d8718fca606e..fa822dff202a 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.h +++ b/fs/xfs/libxfs/xfs_exchmaps.h @@ -97,6 +97,7 @@ xfs_exchmaps_reqfork(const struct xfs_exchmaps_req *req) return XFS_DATA_FORK; } +int xfs_exchmaps_estimate_overhead(struct xfs_exchmaps_req *req); int xfs_exchmaps_estimate(struct xfs_exchmaps_req *req); extern struct kmem_cache *xfs_exchmaps_intent_cache; diff --git a/fs/xfs/scrub/tempexch.h b/fs/xfs/scrub/tempexch.h index 98222b684b6a..c1dd4adec4f1 100644 --- a/fs/xfs/scrub/tempexch.h +++ b/fs/xfs/scrub/tempexch.h @@ -14,6 +14,8 @@ struct xrep_tempexch { int xrep_tempexch_enable(struct xfs_scrub *sc); int xrep_tempexch_trans_reserve(struct xfs_scrub *sc, int whichfork, struct xrep_tempexch *ti); +int xrep_tempexch_trans_alloc(struct xfs_scrub *sc, int whichfork, + struct xrep_tempexch *ti); int xrep_tempexch_contents(struct xfs_scrub *sc, struct xrep_tempexch *ti); #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index 7791336ca820..0b3060be938f 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -239,6 +239,28 @@ xrep_tempfile_iunlock( sc->temp_ilock_flags &= ~XFS_ILOCK_EXCL; } +/* + * Begin the process of making changes to both the file being scrubbed and + * the temporary file by taking ILOCK_EXCL on both. + */ +void +xrep_tempfile_ilock_both( + struct xfs_scrub *sc) +{ + xfs_lock_two_inodes(sc->ip, XFS_ILOCK_EXCL, sc->tempip, XFS_ILOCK_EXCL); + sc->ilock_flags |= XFS_ILOCK_EXCL; + sc->temp_ilock_flags |= XFS_ILOCK_EXCL; +} + +/* Unlock ILOCK_EXCL on both files. */ +void +xrep_tempfile_iunlock_both( + struct xfs_scrub *sc) +{ + xrep_tempfile_iunlock(sc); + xchk_iunlock(sc, XFS_ILOCK_EXCL); +} + /* Release the temporary file. */ void xrep_tempfile_rele( @@ -514,6 +536,89 @@ xrep_tempexch_prep_request( return 0; } +/* + * Fill out the mapping exchange resource estimation structures in preparation + * for exchanging the contents of a metadata file that we've rebuilt in the + * temp file. Caller must hold IOLOCK_EXCL but not ILOCK_EXCL on both files. + */ +STATIC int +xrep_tempexch_estimate( + struct xfs_scrub *sc, + struct xrep_tempexch *tx) +{ + struct xfs_exchmaps_req *req = &tx->req; + struct xfs_ifork *ifp; + struct xfs_ifork *tifp; + int whichfork = xfs_exchmaps_reqfork(req); + int state = 0; + + /* + * The exchmaps code only knows how to exchange file fork space + * mappings. Any fork data in local format must be promoted to a + * single block before the exchange can take place. + */ + ifp = xfs_ifork_ptr(sc->ip, whichfork); + if (ifp->if_format == XFS_DINODE_FMT_LOCAL) + state |= 1; + + tifp = xfs_ifork_ptr(sc->tempip, whichfork); + if (tifp->if_format == XFS_DINODE_FMT_LOCAL) + state |= 2; + + switch (state) { + case 0: + /* Both files have mapped extents; use the regular estimate. */ + return xfs_exchrange_estimate(req); + case 1: + /* + * The file being repaired is in local format, but the temp + * file has mapped extents. To perform the exchange, the file + * being repaired must have its shorform data converted to an + * ondisk block so that the forks will be in extents format. + * We need one resblk for the conversion; the number of + * exchanges is (worst case) the temporary file's extent count + * plus the block we converted. + */ + req->ip1_bcount = sc->tempip->i_nblocks; + req->ip2_bcount = 1; + req->nr_exchanges = 1 + tifp->if_nextents; + req->resblks = 1; + break; + case 2: + /* + * The temporary file is in local format, but the file being + * repaired has mapped extents. To perform the exchange, the + * temp file must have its shortform data converted to an + * ondisk block, and the fork changed to extents format. We + * need one resblk for the conversion; the number of exchanges + * is (worst case) the extent count of the file being repaired + * plus the block we converted. + */ + req->ip1_bcount = 1; + req->ip2_bcount = sc->ip->i_nblocks; + req->nr_exchanges = 1 + ifp->if_nextents; + req->resblks = 1; + break; + case 3: + /* + * Both forks are in local format. To perform the exchange, + * both files must have their shortform data converted to + * fsblocks, and both forks must be converted to extents + * format. We need two resblks for the two conversions, and + * the number of exchanges is 1 since there's only one block at + * fileoff 0. Presumably, the caller could not exchange the + * two inode fork areas directly. + */ + req->ip1_bcount = 1; + req->ip2_bcount = 1; + req->nr_exchanges = 1; + req->resblks = 2; + break; + } + + return xfs_exchmaps_estimate_overhead(req); +} + /* * Obtain a quota reservation to make sure we don't hit EDQUOT. We can skip * this if quota enforcement is disabled or if both inodes' dquots are the @@ -604,6 +709,55 @@ xrep_tempexch_trans_reserve( return xrep_tempexch_reserve_quota(sc, tx); } +/* + * Create a new transaction for a file contents exchange. + * + * This function fills out the mapping excahange request and resource + * estimation structures in preparation for exchanging the contents of a + * metadata file that has been rebuilt in the temp file. Next, it reserves + * space, takes ILOCK_EXCL of both inodes, joins them to the transaction and + * reserves quota for the transaction. + * + * The caller is responsible for dropping both ILOCKs when appropriate. + */ +int +xrep_tempexch_trans_alloc( + struct xfs_scrub *sc, + int whichfork, + struct xrep_tempexch *tx) +{ + unsigned int flags = 0; + int error; + + ASSERT(sc->tp == NULL); + + error = xrep_tempexch_prep_request(sc, whichfork, tx); + if (error) + return error; + + error = xrep_tempexch_estimate(sc, tx); + if (error) + return error; + + if (xfs_has_lazysbcount(sc->mp)) + flags |= XFS_TRANS_RES_FDBLKS; + + error = xrep_tempexch_enable(sc); + if (error) + return error; + + error = xfs_trans_alloc(sc->mp, &M_RES(sc->mp)->tr_itruncate, + tx->req.resblks, 0, flags, &sc->tp); + if (error) + return error; + + sc->temp_ilock_flags |= XFS_ILOCK_EXCL; + sc->ilock_flags |= XFS_ILOCK_EXCL; + xfs_exchrange_ilock(sc->tp, sc->ip, sc->tempip); + + return xrep_tempexch_reserve_quota(sc, tx); +} + /* * Exchange file mappings (and hence file contents) between the file being * repaired and the temporary file. Returns with both inodes locked and joined @@ -637,3 +791,53 @@ xrep_tempexch_contents( return 0; } + +/* + * Write local format data from one of the temporary file's forks into the same + * fork of file being repaired, and exchange the file sizes, if appropriate. + * Caller must ensure that the file being repaired has enough fork space to + * hold all the bytes. + */ +void +xrep_tempfile_copyout_local( + struct xfs_scrub *sc, + int whichfork) +{ + struct xfs_ifork *temp_ifp; + struct xfs_ifork *ifp; + unsigned int ilog_flags = XFS_ILOG_CORE; + + temp_ifp = xfs_ifork_ptr(sc->tempip, whichfork); + ifp = xfs_ifork_ptr(sc->ip, whichfork); + + ASSERT(temp_ifp != NULL); + ASSERT(ifp != NULL); + ASSERT(temp_ifp->if_format == XFS_DINODE_FMT_LOCAL); + ASSERT(ifp->if_format == XFS_DINODE_FMT_LOCAL); + + switch (whichfork) { + case XFS_DATA_FORK: + ASSERT(sc->tempip->i_disk_size <= + xfs_inode_data_fork_size(sc->ip)); + break; + case XFS_ATTR_FORK: + ASSERT(sc->tempip->i_forkoff >= sc->ip->i_forkoff); + break; + default: + ASSERT(0); + return; + } + + /* Recreate @sc->ip's incore fork (ifp) with data from temp_ifp. */ + xfs_idestroy_fork(ifp); + xfs_init_local_fork(sc->ip, whichfork, temp_ifp->if_data, + temp_ifp->if_bytes); + + if (whichfork == XFS_DATA_FORK) { + i_size_write(VFS_I(sc->ip), i_size_read(VFS_I(sc->tempip))); + sc->ip->i_disk_size = sc->tempip->i_disk_size; + } + + ilog_flags |= xfs_ilog_fdata(whichfork); + xfs_trans_log_inode(sc->tp, sc->ip, ilog_flags); +} diff --git a/fs/xfs/scrub/tempfile.h b/fs/xfs/scrub/tempfile.h index 7980f9c4de55..d57e4f145a7c 100644 --- a/fs/xfs/scrub/tempfile.h +++ b/fs/xfs/scrub/tempfile.h @@ -17,6 +17,8 @@ void xrep_tempfile_iounlock(struct xfs_scrub *sc); void xrep_tempfile_ilock(struct xfs_scrub *sc); bool xrep_tempfile_ilock_nowait(struct xfs_scrub *sc); void xrep_tempfile_iunlock(struct xfs_scrub *sc); +void xrep_tempfile_iunlock_both(struct xfs_scrub *sc); +void xrep_tempfile_ilock_both(struct xfs_scrub *sc); int xrep_tempfile_prealloc(struct xfs_scrub *sc, xfs_fileoff_t off, xfs_filblks_t len); @@ -32,6 +34,7 @@ int xrep_tempfile_copyin(struct xfs_scrub *sc, xfs_fileoff_t off, int xrep_tempfile_set_isize(struct xfs_scrub *sc, unsigned long long isize); int xrep_tempfile_roll_trans(struct xfs_scrub *sc); +void xrep_tempfile_copyout_local(struct xfs_scrub *sc, int whichfork); #else static inline void xrep_tempfile_iolock_both(struct xfs_scrub *sc) { -- cgit v1.2.3-59-g8ed1b From e47dcf113ae348678143cc935a1183059c02c9ad Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:45 -0700 Subject: xfs: repair extended attributes If the extended attributes look bad, try to sift through the rubble to find whatever keys/values we can, stage a new attribute structure in a temporary file and use the atomic extent swapping mechanism to commit the results in bulk. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/libxfs/xfs_attr.c | 2 +- fs/xfs/libxfs/xfs_attr.h | 2 + fs/xfs/libxfs/xfs_da_format.h | 5 + fs/xfs/scrub/attr.c | 20 +- fs/xfs/scrub/attr.h | 7 + fs/xfs/scrub/attr_repair.c | 1207 +++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/attr_repair.h | 11 + fs/xfs/scrub/repair.c | 46 ++ fs/xfs/scrub/repair.h | 6 + fs/xfs/scrub/scrub.c | 2 +- fs/xfs/scrub/trace.h | 83 +++ fs/xfs/scrub/xfarray.c | 17 + fs/xfs/scrub/xfarray.h | 2 + fs/xfs/scrub/xfblob.c | 17 + fs/xfs/scrub/xfblob.h | 2 + fs/xfs/scrub/xfile.h | 5 + fs/xfs/xfs_buf.c | 3 + fs/xfs/xfs_trace.h | 2 + 19 files changed, 1436 insertions(+), 4 deletions(-) create mode 100644 fs/xfs/scrub/attr_repair.c create mode 100644 fs/xfs/scrub/attr_repair.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index bc27757702fe..8647629ac7bf 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -194,6 +194,7 @@ ifeq ($(CONFIG_XFS_ONLINE_REPAIR),y) xfs-y += $(addprefix scrub/, \ agheader_repair.o \ alloc_repair.o \ + attr_repair.o \ bmap_repair.o \ cow_repair.o \ fscounters_repair.o \ diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index b3c9666cd011..05d22c5e3885 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -1055,7 +1055,7 @@ out_trans_cancel: * External routines when attribute list is inside the inode *========================================================================*/ -static inline int xfs_attr_sf_totsize(struct xfs_inode *dp) +int xfs_attr_sf_totsize(struct xfs_inode *dp) { struct xfs_attr_sf_hdr *sf = dp->i_af.if_data; diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index 81be9b3e4004..e4f55008552b 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -618,4 +618,6 @@ extern struct kmem_cache *xfs_attr_intent_cache; int __init xfs_attr_intent_init_cache(void); void xfs_attr_intent_destroy_cache(void); +int xfs_attr_sf_totsize(struct xfs_inode *dp); + #endif /* __XFS_ATTR_H__ */ diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h index 060e5c96b70f..aac3fe039614 100644 --- a/fs/xfs/libxfs/xfs_da_format.h +++ b/fs/xfs/libxfs/xfs_da_format.h @@ -721,6 +721,11 @@ struct xfs_attr3_leafblock { #define XFS_ATTR_INCOMPLETE (1u << XFS_ATTR_INCOMPLETE_BIT) #define XFS_ATTR_NSP_ONDISK_MASK (XFS_ATTR_ROOT | XFS_ATTR_SECURE) +#define XFS_ATTR_NAMESPACE_STR \ + { XFS_ATTR_LOCAL, "local" }, \ + { XFS_ATTR_ROOT, "root" }, \ + { XFS_ATTR_SECURE, "secure" } + /* * Alignment for namelist and valuelist entries (since they are mixed * there can be only one alignment value) diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 0c467f4f8e77..7621e548d730 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -10,6 +10,7 @@ #include "xfs_trans_resv.h" #include "xfs_mount.h" #include "xfs_log_format.h" +#include "xfs_trans.h" #include "xfs_inode.h" #include "xfs_da_format.h" #include "xfs_da_btree.h" @@ -20,6 +21,7 @@ #include "scrub/common.h" #include "scrub/dabtree.h" #include "scrub/attr.h" +#include "scrub/repair.h" /* Free the buffers linked from the xattr buffer. */ static void @@ -35,6 +37,8 @@ xchk_xattr_buf_cleanup( kvfree(ab->value); ab->value = NULL; ab->value_sz = 0; + kvfree(ab->name); + ab->name = NULL; } /* @@ -65,7 +69,7 @@ xchk_xattr_want_freemap( * reallocating the buffer if necessary. Buffer contents are not preserved * across a reallocation. */ -static int +int xchk_setup_xattr_buf( struct xfs_scrub *sc, size_t value_size) @@ -95,6 +99,12 @@ xchk_setup_xattr_buf( return -ENOMEM; } + if (xchk_could_repair(sc)) { + ab->name = kvmalloc(XATTR_NAME_MAX + 1, XCHK_GFP_FLAGS); + if (!ab->name) + return -ENOMEM; + } + resize_value: if (ab->value_sz >= value_size) return 0; @@ -121,6 +131,12 @@ xchk_setup_xattr( { int error; + if (xchk_could_repair(sc)) { + error = xrep_setup_xattr(sc); + if (error) + return error; + } + /* * We failed to get memory while checking attrs, so this time try to * get all the memory we're ever going to need. Allocate the buffer @@ -247,7 +263,7 @@ fail_xref: * Within a char, the lowest bit of the char represents the byte with * the smallest address */ -STATIC bool +bool xchk_xattr_set_map( struct xfs_scrub *sc, unsigned long *map, diff --git a/fs/xfs/scrub/attr.h b/fs/xfs/scrub/attr.h index 48fd9402c432..7db58af56646 100644 --- a/fs/xfs/scrub/attr.h +++ b/fs/xfs/scrub/attr.h @@ -16,9 +16,16 @@ struct xchk_xattr_buf { /* Bitmap of free space in xattr leaf blocks. */ unsigned long *freemap; + /* Memory buffer used to hold salvaged xattr names. */ + unsigned char *name; + /* Memory buffer used to extract xattr values. */ void *value; size_t value_sz; }; +bool xchk_xattr_set_map(struct xfs_scrub *sc, unsigned long *map, + unsigned int start, unsigned int len); +int xchk_setup_xattr_buf(struct xfs_scrub *sc, size_t value_size); + #endif /* __XFS_SCRUB_ATTR_H__ */ diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c new file mode 100644 index 000000000000..7b4318764d03 --- /dev/null +++ b/fs/xfs/scrub/attr_repair.c @@ -0,0 +1,1207 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2018-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_btree.h" +#include "xfs_bit.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_inode.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_dir2.h" +#include "xfs_attr.h" +#include "xfs_attr_leaf.h" +#include "xfs_attr_sf.h" +#include "xfs_attr_remote.h" +#include "xfs_bmap.h" +#include "xfs_bmap_util.h" +#include "xfs_exchmaps.h" +#include "xfs_exchrange.h" +#include "xfs_acl.h" +#include "scrub/xfs_scrub.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/tempfile.h" +#include "scrub/tempexch.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" +#include "scrub/attr.h" +#include "scrub/reap.h" +#include "scrub/attr_repair.h" + +/* + * Extended Attribute Repair + * ========================= + * + * We repair extended attributes by reading the attr leaf blocks looking for + * attributes entries that look salvageable (name passes verifiers, value can + * be retrieved, etc). Each extended attribute worth salvaging is stashed in + * memory, and the stashed entries are periodically replayed into a temporary + * file to constrain memory use. Batching the construction of the temporary + * extended attribute structure in this fashion reduces lock cycling of the + * file being repaired and the temporary file. + * + * When salvaging completes, the remaining stashed attributes are replayed to + * the temporary file. An atomic file contents exchange is used to commit the + * new xattr blocks to the file being repaired. This will disrupt attrmulti + * cursors. + */ + +struct xrep_xattr_key { + /* Cookie for retrieval of the xattr name. */ + xfblob_cookie name_cookie; + + /* Cookie for retrieval of the xattr value. */ + xfblob_cookie value_cookie; + + /* XFS_ATTR_* flags */ + int flags; + + /* Length of the value and name. */ + uint32_t valuelen; + uint16_t namelen; +}; + +/* + * Stash up to 8 pages of attrs in xattr_records/xattr_blobs before we write + * them to the temp file. + */ +#define XREP_XATTR_MAX_STASH_BYTES (PAGE_SIZE * 8) + +struct xrep_xattr { + struct xfs_scrub *sc; + + /* Information for exchanging attr fork mappings at the end. */ + struct xrep_tempexch tx; + + /* xattr keys */ + struct xfarray *xattr_records; + + /* xattr values */ + struct xfblob *xattr_blobs; + + /* Number of attributes that we are salvaging. */ + unsigned long long attrs_found; +}; + +/* Set up to recreate the extended attributes. */ +int +xrep_setup_xattr( + struct xfs_scrub *sc) +{ + return xrep_tempfile_create(sc, S_IFREG); +} + +/* + * Decide if we want to salvage this attribute. We don't bother with + * incomplete or oversized keys or values. The @value parameter can be null + * for remote attrs. + */ +STATIC int +xrep_xattr_want_salvage( + struct xrep_xattr *rx, + unsigned int attr_flags, + const void *name, + int namelen, + const void *value, + int valuelen) +{ + if (attr_flags & XFS_ATTR_INCOMPLETE) + return false; + if (namelen > XATTR_NAME_MAX || namelen <= 0) + return false; + if (!xfs_attr_namecheck(name, namelen)) + return false; + if (valuelen > XATTR_SIZE_MAX || valuelen < 0) + return false; + if (hweight32(attr_flags & XFS_ATTR_NSP_ONDISK_MASK) > 1) + return false; + return true; +} + +/* Allocate an in-core record to hold xattrs while we rebuild the xattr data. */ +STATIC int +xrep_xattr_salvage_key( + struct xrep_xattr *rx, + int flags, + unsigned char *name, + int namelen, + unsigned char *value, + int valuelen) +{ + struct xrep_xattr_key key = { + .valuelen = valuelen, + .flags = flags & XFS_ATTR_NSP_ONDISK_MASK, + }; + unsigned int i = 0; + int error = 0; + + if (xchk_should_terminate(rx->sc, &error)) + return error; + + /* + * Truncate the name to the first character that would trip namecheck. + * If we no longer have a name after that, ignore this attribute. + */ + while (i < namelen && name[i] != 0) + i++; + if (i == 0) + return 0; + key.namelen = i; + + trace_xrep_xattr_salvage_rec(rx->sc->ip, flags, name, key.namelen, + valuelen); + + error = xfblob_store(rx->xattr_blobs, &key.name_cookie, name, + key.namelen); + if (error) + return error; + + error = xfblob_store(rx->xattr_blobs, &key.value_cookie, value, + key.valuelen); + if (error) + return error; + + error = xfarray_append(rx->xattr_records, &key); + if (error) + return error; + + rx->attrs_found++; + return 0; +} + +/* + * Record a shortform extended attribute key & value for later reinsertion + * into the inode. + */ +STATIC int +xrep_xattr_salvage_sf_attr( + struct xrep_xattr *rx, + struct xfs_attr_sf_hdr *hdr, + struct xfs_attr_sf_entry *sfe) +{ + struct xfs_scrub *sc = rx->sc; + struct xchk_xattr_buf *ab = sc->buf; + unsigned char *name = sfe->nameval; + unsigned char *value = &sfe->nameval[sfe->namelen]; + + if (!xchk_xattr_set_map(sc, ab->usedmap, (char *)name - (char *)hdr, + sfe->namelen)) + return 0; + + if (!xchk_xattr_set_map(sc, ab->usedmap, (char *)value - (char *)hdr, + sfe->valuelen)) + return 0; + + if (!xrep_xattr_want_salvage(rx, sfe->flags, sfe->nameval, + sfe->namelen, value, sfe->valuelen)) + return 0; + + return xrep_xattr_salvage_key(rx, sfe->flags, sfe->nameval, + sfe->namelen, value, sfe->valuelen); +} + +/* + * Record a local format extended attribute key & value for later reinsertion + * into the inode. + */ +STATIC int +xrep_xattr_salvage_local_attr( + struct xrep_xattr *rx, + struct xfs_attr_leaf_entry *ent, + unsigned int nameidx, + const char *buf_end, + struct xfs_attr_leaf_name_local *lentry) +{ + struct xchk_xattr_buf *ab = rx->sc->buf; + unsigned char *value; + unsigned int valuelen; + unsigned int namesize; + + /* + * Decode the leaf local entry format. If something seems wrong, we + * junk the attribute. + */ + value = &lentry->nameval[lentry->namelen]; + valuelen = be16_to_cpu(lentry->valuelen); + namesize = xfs_attr_leaf_entsize_local(lentry->namelen, valuelen); + if ((char *)lentry + namesize > buf_end) + return 0; + if (!xrep_xattr_want_salvage(rx, ent->flags, lentry->nameval, + lentry->namelen, value, valuelen)) + return 0; + if (!xchk_xattr_set_map(rx->sc, ab->usedmap, nameidx, namesize)) + return 0; + + /* Try to save this attribute. */ + return xrep_xattr_salvage_key(rx, ent->flags, lentry->nameval, + lentry->namelen, value, valuelen); +} + +/* + * Record a remote format extended attribute key & value for later reinsertion + * into the inode. + */ +STATIC int +xrep_xattr_salvage_remote_attr( + struct xrep_xattr *rx, + struct xfs_attr_leaf_entry *ent, + unsigned int nameidx, + const char *buf_end, + struct xfs_attr_leaf_name_remote *rentry, + unsigned int ent_idx, + struct xfs_buf *leaf_bp) +{ + struct xchk_xattr_buf *ab = rx->sc->buf; + struct xfs_da_args args = { + .trans = rx->sc->tp, + .dp = rx->sc->ip, + .index = ent_idx, + .geo = rx->sc->mp->m_attr_geo, + .owner = rx->sc->ip->i_ino, + .attr_filter = ent->flags & XFS_ATTR_NSP_ONDISK_MASK, + .namelen = rentry->namelen, + .name = rentry->name, + .value = ab->value, + .valuelen = be32_to_cpu(rentry->valuelen), + }; + unsigned int namesize; + int error; + + /* + * Decode the leaf remote entry format. If something seems wrong, we + * junk the attribute. Note that we should never find a zero-length + * remote attribute value. + */ + namesize = xfs_attr_leaf_entsize_remote(rentry->namelen); + if ((char *)rentry + namesize > buf_end) + return 0; + if (args.valuelen == 0 || + !xrep_xattr_want_salvage(rx, ent->flags, rentry->name, + rentry->namelen, NULL, args.valuelen)) + return 0; + if (!xchk_xattr_set_map(rx->sc, ab->usedmap, nameidx, namesize)) + return 0; + + /* + * Enlarge the buffer (if needed) to hold the value that we're trying + * to salvage from the old extended attribute data. + */ + error = xchk_setup_xattr_buf(rx->sc, args.valuelen); + if (error == -ENOMEM) + error = -EDEADLOCK; + if (error) + return error; + + /* Look up the remote value and stash it for reconstruction. */ + error = xfs_attr3_leaf_getvalue(leaf_bp, &args); + if (error || args.rmtblkno == 0) + goto err_free; + + error = xfs_attr_rmtval_get(&args); + if (error) + goto err_free; + + /* Try to save this attribute. */ + error = xrep_xattr_salvage_key(rx, ent->flags, rentry->name, + rentry->namelen, ab->value, args.valuelen); +err_free: + /* remote value was garbage, junk it */ + if (error == -EFSBADCRC || error == -EFSCORRUPTED) + error = 0; + return error; +} + +/* Extract every xattr key that we can from this attr fork block. */ +STATIC int +xrep_xattr_recover_leaf( + struct xrep_xattr *rx, + struct xfs_buf *bp) +{ + struct xfs_attr3_icleaf_hdr leafhdr; + struct xfs_scrub *sc = rx->sc; + struct xfs_mount *mp = sc->mp; + struct xfs_attr_leafblock *leaf; + struct xfs_attr_leaf_name_local *lentry; + struct xfs_attr_leaf_name_remote *rentry; + struct xfs_attr_leaf_entry *ent; + struct xfs_attr_leaf_entry *entries; + struct xchk_xattr_buf *ab = rx->sc->buf; + char *buf_end; + size_t off; + unsigned int nameidx; + unsigned int hdrsize; + int i; + int error = 0; + + bitmap_zero(ab->usedmap, mp->m_attr_geo->blksize); + + /* Check the leaf header */ + leaf = bp->b_addr; + xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &leafhdr, leaf); + hdrsize = xfs_attr3_leaf_hdr_size(leaf); + xchk_xattr_set_map(sc, ab->usedmap, 0, hdrsize); + entries = xfs_attr3_leaf_entryp(leaf); + + buf_end = (char *)bp->b_addr + mp->m_attr_geo->blksize; + for (i = 0, ent = entries; i < leafhdr.count; ent++, i++) { + if (xchk_should_terminate(sc, &error)) + return error; + + /* Skip key if it conflicts with something else? */ + off = (char *)ent - (char *)leaf; + if (!xchk_xattr_set_map(sc, ab->usedmap, off, + sizeof(xfs_attr_leaf_entry_t))) + continue; + + /* Check the name information. */ + nameidx = be16_to_cpu(ent->nameidx); + if (nameidx < leafhdr.firstused || + nameidx >= mp->m_attr_geo->blksize) + continue; + + if (ent->flags & XFS_ATTR_LOCAL) { + lentry = xfs_attr3_leaf_name_local(leaf, i); + error = xrep_xattr_salvage_local_attr(rx, ent, nameidx, + buf_end, lentry); + } else { + rentry = xfs_attr3_leaf_name_remote(leaf, i); + error = xrep_xattr_salvage_remote_attr(rx, ent, nameidx, + buf_end, rentry, i, bp); + } + if (error) + return error; + } + + return 0; +} + +/* Try to recover shortform attrs. */ +STATIC int +xrep_xattr_recover_sf( + struct xrep_xattr *rx) +{ + struct xfs_scrub *sc = rx->sc; + struct xchk_xattr_buf *ab = sc->buf; + struct xfs_attr_sf_hdr *hdr; + struct xfs_attr_sf_entry *sfe; + struct xfs_attr_sf_entry *next; + struct xfs_ifork *ifp; + unsigned char *end; + int i; + int error = 0; + + ifp = xfs_ifork_ptr(rx->sc->ip, XFS_ATTR_FORK); + hdr = ifp->if_data; + + bitmap_zero(ab->usedmap, ifp->if_bytes); + end = (unsigned char *)ifp->if_data + ifp->if_bytes; + xchk_xattr_set_map(sc, ab->usedmap, 0, sizeof(*hdr)); + + sfe = xfs_attr_sf_firstentry(hdr); + if ((unsigned char *)sfe > end) + return 0; + + for (i = 0; i < hdr->count; i++) { + if (xchk_should_terminate(sc, &error)) + return error; + + next = xfs_attr_sf_nextentry(sfe); + if ((unsigned char *)next > end) + break; + + if (xchk_xattr_set_map(sc, ab->usedmap, + (char *)sfe - (char *)hdr, + sizeof(struct xfs_attr_sf_entry))) { + /* + * No conflicts with the sf entry; let's save this + * attribute. + */ + error = xrep_xattr_salvage_sf_attr(rx, hdr, sfe); + if (error) + return error; + } + + sfe = next; + } + + return 0; +} + +/* + * Try to return a buffer of xattr data for a given physical extent. + * + * Because the buffer cache get function complains if it finds a buffer + * matching the block number but not matching the length, we must be careful to + * look for incore buffers (up to the maximum length of a remote value) that + * could be hiding anywhere in the physical range. If we find an incore + * buffer, we can pass that to the caller. Optionally, read a single block and + * pass that back. + * + * Note the subtlety that remote attr value blocks for which there is no incore + * buffer will be passed to the callback one block at a time. These buffers + * will not have any ops attached and must be staled to prevent aliasing with + * multiblock buffers once we drop the ILOCK. + */ +STATIC int +xrep_xattr_find_buf( + struct xfs_mount *mp, + xfs_fsblock_t fsbno, + xfs_extlen_t max_len, + bool can_read, + struct xfs_buf **bpp) +{ + struct xrep_bufscan scan = { + .daddr = XFS_FSB_TO_DADDR(mp, fsbno), + .max_sectors = xrep_bufscan_max_sectors(mp, max_len), + .daddr_step = XFS_FSB_TO_BB(mp, 1), + }; + struct xfs_buf *bp; + + while ((bp = xrep_bufscan_advance(mp, &scan)) != NULL) { + *bpp = bp; + return 0; + } + + if (!can_read) { + *bpp = NULL; + return 0; + } + + return xfs_buf_read(mp->m_ddev_targp, scan.daddr, XFS_FSB_TO_BB(mp, 1), + XBF_TRYLOCK, bpp, NULL); +} + +/* + * Deal with a buffer that we found during our walk of the attr fork. + * + * Attribute leaf and node blocks are simple -- they're a single block, so we + * can walk them one at a time and we never have to worry about discontiguous + * multiblock buffers like we do for directories. + * + * Unfortunately, remote attr blocks add a lot of complexity here. Each disk + * block is totally self contained, in the sense that the v5 header provides no + * indication that there could be more data in the next block. The incore + * buffers can span multiple blocks, though they never cross extent records. + * However, they don't necessarily start or end on an extent record boundary. + * Therefore, we need a special buffer find function to walk the buffer cache + * for us. + * + * The caller must hold the ILOCK on the file being repaired. We use + * XBF_TRYLOCK here to skip any locked buffer on the assumption that we don't + * own the block and don't want to hang the system on a potentially garbage + * buffer. + */ +STATIC int +xrep_xattr_recover_block( + struct xrep_xattr *rx, + xfs_dablk_t dabno, + xfs_fsblock_t fsbno, + xfs_extlen_t max_len, + xfs_extlen_t *actual_len) +{ + struct xfs_da_blkinfo *info; + struct xfs_buf *bp; + int error; + + error = xrep_xattr_find_buf(rx->sc->mp, fsbno, max_len, true, &bp); + if (error) + return error; + info = bp->b_addr; + *actual_len = XFS_BB_TO_FSB(rx->sc->mp, bp->b_length); + + trace_xrep_xattr_recover_leafblock(rx->sc->ip, dabno, + be16_to_cpu(info->magic)); + + /* + * If the buffer has the right magic number for an attr leaf block and + * passes a structure check (we don't care about checksums), salvage + * as much as we can from the block. */ + if (info->magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC) && + xrep_buf_verify_struct(bp, &xfs_attr3_leaf_buf_ops) && + xfs_attr3_leaf_header_check(bp, rx->sc->ip->i_ino) == NULL) + error = xrep_xattr_recover_leaf(rx, bp); + + /* + * If the buffer didn't already have buffer ops set, it was read in by + * the _find_buf function and could very well be /part/ of a multiblock + * remote block. Mark it stale so that it doesn't hang around in + * memory to cause problems. + */ + if (bp->b_ops == NULL) + xfs_buf_stale(bp); + + xfs_buf_relse(bp); + return error; +} + +/* Insert one xattr key/value. */ +STATIC int +xrep_xattr_insert_rec( + struct xrep_xattr *rx, + const struct xrep_xattr_key *key) +{ + struct xfs_da_args args = { + .dp = rx->sc->tempip, + .attr_filter = key->flags, + .attr_flags = XATTR_CREATE, + .namelen = key->namelen, + .valuelen = key->valuelen, + .owner = rx->sc->ip->i_ino, + }; + struct xchk_xattr_buf *ab = rx->sc->buf; + int error; + + /* + * Grab pointers to the scrub buffer so that we can use them to insert + * attrs into the temp file. + */ + args.name = ab->name; + args.value = ab->value; + + /* + * The attribute name is stored near the end of the in-core buffer, + * though we reserve one more byte to ensure null termination. + */ + ab->name[XATTR_NAME_MAX] = 0; + + error = xfblob_load(rx->xattr_blobs, key->name_cookie, ab->name, + key->namelen); + if (error) + return error; + + error = xfblob_free(rx->xattr_blobs, key->name_cookie); + if (error) + return error; + + error = xfblob_load(rx->xattr_blobs, key->value_cookie, args.value, + key->valuelen); + if (error) + return error; + + error = xfblob_free(rx->xattr_blobs, key->value_cookie); + if (error) + return error; + + ab->name[key->namelen] = 0; + + trace_xrep_xattr_insert_rec(rx->sc->tempip, key->flags, ab->name, + key->namelen, key->valuelen); + + /* + * xfs_attr_set creates and commits its own transaction. If the attr + * already exists, we'll just drop it during the rebuild. + */ + error = xfs_attr_set(&args); + if (error == -EEXIST) + error = 0; + + return error; +} + +/* + * Periodically flush salvaged attributes to the temporary file. This is done + * to reduce the memory requirements of the xattr rebuild because files can + * contain millions of attributes. + */ +STATIC int +xrep_xattr_flush_stashed( + struct xrep_xattr *rx) +{ + xfarray_idx_t array_cur; + int error; + + /* + * Entering this function, the scrub context has a reference to the + * inode being repaired, the temporary file, and a scrub transaction + * that we use during xattr salvaging to avoid livelocking if there + * are cycles in the xattr structures. We hold ILOCK_EXCL on both + * the inode being repaired, though it is not ijoined to the scrub + * transaction. + * + * To constrain kernel memory use, we occasionally flush salvaged + * xattrs from the xfarray and xfblob structures into the temporary + * file in preparation for exchanging the xattr structures at the end. + * Updating the temporary file requires a transaction, so we commit the + * scrub transaction and drop the two ILOCKs so that xfs_attr_set can + * allocate whatever transaction it wants. + * + * We still hold IOLOCK_EXCL on the inode being repaired, which + * prevents anyone from modifying the damaged xattr data while we + * repair it. + */ + error = xrep_trans_commit(rx->sc); + if (error) + return error; + xchk_iunlock(rx->sc, XFS_ILOCK_EXCL); + + /* + * Take the IOLOCK of the temporary file while we modify xattrs. This + * isn't strictly required because the temporary file is never revealed + * to userspace, but we follow the same locking rules. We still hold + * sc->ip's IOLOCK. + */ + error = xrep_tempfile_iolock_polled(rx->sc); + if (error) + return error; + + /* Add all the salvaged attrs to the temporary file. */ + foreach_xfarray_idx(rx->xattr_records, array_cur) { + struct xrep_xattr_key key; + + error = xfarray_load(rx->xattr_records, array_cur, &key); + if (error) + return error; + + error = xrep_xattr_insert_rec(rx, &key); + if (error) + return error; + } + + /* Empty out both arrays now that we've added the entries. */ + xfarray_truncate(rx->xattr_records); + xfblob_truncate(rx->xattr_blobs); + + xrep_tempfile_iounlock(rx->sc); + + /* Recreate the salvage transaction and relock the inode. */ + error = xchk_trans_alloc(rx->sc, 0); + if (error) + return error; + xchk_ilock(rx->sc, XFS_ILOCK_EXCL); + return 0; +} + +/* Decide if we've stashed too much xattr data in memory. */ +static inline bool +xrep_xattr_want_flush_stashed( + struct xrep_xattr *rx) +{ + unsigned long long bytes; + + bytes = xfarray_bytes(rx->xattr_records) + + xfblob_bytes(rx->xattr_blobs); + return bytes > XREP_XATTR_MAX_STASH_BYTES; +} + +/* Extract as many attribute keys and values as we can. */ +STATIC int +xrep_xattr_recover( + struct xrep_xattr *rx) +{ + struct xfs_bmbt_irec got; + struct xfs_scrub *sc = rx->sc; + struct xfs_da_geometry *geo = sc->mp->m_attr_geo; + xfs_fileoff_t offset; + xfs_extlen_t len; + xfs_dablk_t dabno; + int nmap; + int error; + + /* + * Iterate each xattr leaf block in the attr fork to scan them for any + * attributes that we might salvage. + */ + for (offset = 0; + offset < XFS_MAX_FILEOFF; + offset = got.br_startoff + got.br_blockcount) { + nmap = 1; + error = xfs_bmapi_read(sc->ip, offset, XFS_MAX_FILEOFF - offset, + &got, &nmap, XFS_BMAPI_ATTRFORK); + if (error) + return error; + if (nmap != 1) + return -EFSCORRUPTED; + if (!xfs_bmap_is_written_extent(&got)) + continue; + + for (dabno = round_up(got.br_startoff, geo->fsbcount); + dabno < got.br_startoff + got.br_blockcount; + dabno += len) { + xfs_fileoff_t curr_offset = dabno - got.br_startoff; + xfs_extlen_t maxlen; + + if (xchk_should_terminate(rx->sc, &error)) + return error; + + maxlen = min_t(xfs_filblks_t, INT_MAX, + got.br_blockcount - curr_offset); + error = xrep_xattr_recover_block(rx, dabno, + curr_offset + got.br_startblock, + maxlen, &len); + if (error) + return error; + + if (xrep_xattr_want_flush_stashed(rx)) { + error = xrep_xattr_flush_stashed(rx); + if (error) + return error; + } + } + } + + return 0; +} + +/* + * Reset the extended attribute fork to a state where we can start re-adding + * the salvaged attributes. + */ +STATIC int +xrep_xattr_fork_remove( + struct xfs_scrub *sc, + struct xfs_inode *ip) +{ + struct xfs_attr_sf_hdr *hdr; + struct xfs_ifork *ifp = xfs_ifork_ptr(ip, XFS_ATTR_FORK); + + /* + * If the data fork is in btree format, we can't change di_forkoff + * because we could run afoul of the rule that the data fork isn't + * supposed to be in btree format if there's enough space in the fork + * that it could have used extents format. Instead, reinitialize the + * attr fork to have a shortform structure with zero attributes. + */ + if (ip->i_df.if_format == XFS_DINODE_FMT_BTREE) { + ifp->if_format = XFS_DINODE_FMT_LOCAL; + hdr = xfs_idata_realloc(ip, (int)sizeof(*hdr) - ifp->if_bytes, + XFS_ATTR_FORK); + hdr->count = 0; + hdr->totsize = cpu_to_be16(sizeof(*hdr)); + xfs_trans_log_inode(sc->tp, ip, + XFS_ILOG_CORE | XFS_ILOG_ADATA); + return 0; + } + + /* If we still have attr fork extents, something's wrong. */ + if (ifp->if_nextents != 0) { + struct xfs_iext_cursor icur; + struct xfs_bmbt_irec irec; + unsigned int i = 0; + + xfs_emerg(sc->mp, + "inode 0x%llx attr fork still has %llu attr extents, format %d?!", + ip->i_ino, ifp->if_nextents, ifp->if_format); + for_each_xfs_iext(ifp, &icur, &irec) { + xfs_err(sc->mp, + "[%u]: startoff %llu startblock %llu blockcount %llu state %u", + i++, irec.br_startoff, + irec.br_startblock, irec.br_blockcount, + irec.br_state); + } + ASSERT(0); + return -EFSCORRUPTED; + } + + xfs_attr_fork_remove(ip, sc->tp); + return 0; +} + +/* + * Free all the attribute fork blocks of the file being repaired and delete the + * fork. The caller must ILOCK the scrub file and join it to the transaction. + * This function returns with the inode joined to a clean transaction. + */ +int +xrep_xattr_reset_fork( + struct xfs_scrub *sc) +{ + int error; + + trace_xrep_xattr_reset_fork(sc->ip, sc->ip); + + /* Unmap all the attr blocks. */ + if (xfs_ifork_has_extents(&sc->ip->i_af)) { + error = xrep_reap_ifork(sc, sc->ip, XFS_ATTR_FORK); + if (error) + return error; + } + + error = xrep_xattr_fork_remove(sc, sc->ip); + if (error) + return error; + + return xfs_trans_roll_inode(&sc->tp, sc->ip); +} + +/* + * Free all the attribute fork blocks of the temporary file and delete the attr + * fork. The caller must ILOCK the tempfile and join it to the transaction. + * This function returns with the inode joined to a clean scrub transaction. + */ +STATIC int +xrep_xattr_reset_tempfile_fork( + struct xfs_scrub *sc) +{ + int error; + + trace_xrep_xattr_reset_fork(sc->ip, sc->tempip); + + /* + * Wipe out the attr fork of the temp file so that regular inode + * inactivation won't trip over the corrupt attr fork. + */ + if (xfs_ifork_has_extents(&sc->tempip->i_af)) { + error = xrep_reap_ifork(sc, sc->tempip, XFS_ATTR_FORK); + if (error) + return error; + } + + return xrep_xattr_fork_remove(sc, sc->tempip); +} + +/* + * Find all the extended attributes for this inode by scraping them out of the + * attribute key blocks by hand, and flushing them into the temp file. + * When we're done, free the staging memory before exchanging the xattr + * structures to reduce memory usage. + */ +STATIC int +xrep_xattr_salvage_attributes( + struct xrep_xattr *rx) +{ + struct xfs_inode *ip = rx->sc->ip; + int error; + + /* Short format xattrs are easy! */ + if (rx->sc->ip->i_af.if_format == XFS_DINODE_FMT_LOCAL) { + error = xrep_xattr_recover_sf(rx); + if (error) + return error; + + return xrep_xattr_flush_stashed(rx); + } + + /* + * For non-inline xattr structures, the salvage function scans the + * buffer cache looking for potential attr leaf blocks. The scan + * requires the ability to lock any buffer found and runs independently + * of any transaction <-> buffer item <-> buffer linkage. Therefore, + * roll the transaction to ensure there are no buffers joined. We hold + * the ILOCK independently of the transaction. + */ + error = xfs_trans_roll(&rx->sc->tp); + if (error) + return error; + + error = xfs_iread_extents(rx->sc->tp, ip, XFS_ATTR_FORK); + if (error) + return error; + + error = xrep_xattr_recover(rx); + if (error) + return error; + + return xrep_xattr_flush_stashed(rx); +} + +/* + * Prepare both inodes' attribute forks for an exchange. Promote the tempfile + * from short format to leaf format, and if the file being repaired has a short + * format attr fork, turn it into an empty extent list. + */ +STATIC int +xrep_xattr_swap_prep( + struct xfs_scrub *sc, + bool temp_local, + bool ip_local) +{ + int error; + + /* + * If the tempfile's attributes are in shortform format, convert that + * to a single leaf extent so that we can use the atomic mapping + * exchange. + */ + if (temp_local) { + struct xfs_da_args args = { + .dp = sc->tempip, + .geo = sc->mp->m_attr_geo, + .whichfork = XFS_ATTR_FORK, + .trans = sc->tp, + .total = 1, + .owner = sc->ip->i_ino, + }; + + error = xfs_attr_shortform_to_leaf(&args); + if (error) + return error; + + /* + * Roll the deferred log items to get us back to a clean + * transaction. + */ + error = xfs_defer_finish(&sc->tp); + if (error) + return error; + } + + /* + * If the file being repaired had a shortform attribute fork, convert + * that to an empty extent list in preparation for the atomic mapping + * exchange. + */ + if (ip_local) { + struct xfs_ifork *ifp; + + ifp = xfs_ifork_ptr(sc->ip, XFS_ATTR_FORK); + + xfs_idestroy_fork(ifp); + ifp->if_format = XFS_DINODE_FMT_EXTENTS; + ifp->if_nextents = 0; + ifp->if_bytes = 0; + ifp->if_data = NULL; + ifp->if_height = 0; + + xfs_trans_log_inode(sc->tp, sc->ip, + XFS_ILOG_CORE | XFS_ILOG_ADATA); + } + + return 0; +} + +/* Exchange the temporary file's attribute fork with the one being repaired. */ +STATIC int +xrep_xattr_swap( + struct xfs_scrub *sc, + struct xrep_tempexch *tx) +{ + bool ip_local, temp_local; + int error = 0; + + ip_local = sc->ip->i_af.if_format == XFS_DINODE_FMT_LOCAL; + temp_local = sc->tempip->i_af.if_format == XFS_DINODE_FMT_LOCAL; + + /* + * If the both files have a local format attr fork and the rebuilt + * xattr data would fit in the repaired file's attr fork, just copy + * the contents from the tempfile and declare ourselves done. + */ + if (ip_local && temp_local) { + int forkoff; + int newsize; + + newsize = xfs_attr_sf_totsize(sc->tempip); + forkoff = xfs_attr_shortform_bytesfit(sc->ip, newsize); + if (forkoff > 0) { + sc->ip->i_forkoff = forkoff; + xrep_tempfile_copyout_local(sc, XFS_ATTR_FORK); + return 0; + } + } + + /* Otherwise, make sure both attr forks are in block-mapping mode. */ + error = xrep_xattr_swap_prep(sc, temp_local, ip_local); + if (error) + return error; + + return xrep_tempexch_contents(sc, tx); +} + +/* + * Exchange the new extended attribute data (which we created in the tempfile) + * with the file being repaired. + */ +STATIC int +xrep_xattr_rebuild_tree( + struct xrep_xattr *rx) +{ + struct xfs_scrub *sc = rx->sc; + int error; + + /* + * If we didn't find any attributes to salvage, repair the file by + * zapping its attr fork. + */ + if (rx->attrs_found == 0) { + xfs_trans_ijoin(sc->tp, sc->ip, 0); + error = xrep_xattr_reset_fork(sc); + if (error) + return error; + + goto forget_acls; + } + + trace_xrep_xattr_rebuild_tree(sc->ip, sc->tempip); + + /* + * Commit the repair transaction and drop the ILOCKs so that we can use + * the atomic file content exchange helper functions to compute the + * correct resource reservations. + * + * We still hold IOLOCK_EXCL (aka i_rwsem) which will prevent xattr + * modifications, but there's nothing to prevent userspace from reading + * the attributes until we're ready for the exchange operation. Reads + * will return -EIO without shutting down the fs, so we're ok with + * that. + */ + error = xrep_trans_commit(sc); + if (error) + return error; + + xchk_iunlock(sc, XFS_ILOCK_EXCL); + + /* + * Take the IOLOCK on the temporary file so that we can run xattr + * operations with the same locks held as we would for a normal file. + * We still hold sc->ip's IOLOCK. + */ + error = xrep_tempfile_iolock_polled(rx->sc); + if (error) + return error; + + /* Allocate exchange transaction and lock both inodes. */ + error = xrep_tempexch_trans_alloc(rx->sc, XFS_ATTR_FORK, &rx->tx); + if (error) + return error; + + /* + * Exchange the blocks mapped by the tempfile's attr fork with the file + * being repaired. The old attr blocks will then be attached to the + * tempfile, so reap its attr fork. + */ + error = xrep_xattr_swap(sc, &rx->tx); + if (error) + return error; + + error = xrep_xattr_reset_tempfile_fork(sc); + if (error) + return error; + + /* + * Roll to get a transaction without any inodes joined to it. Then we + * can drop the tempfile's ILOCK and IOLOCK before doing more work on + * the scrub target file. + */ + error = xfs_trans_roll(&sc->tp); + if (error) + return error; + + xrep_tempfile_iunlock(sc); + xrep_tempfile_iounlock(sc); + +forget_acls: + /* Invalidate cached ACLs now that we've reloaded all the xattrs. */ + xfs_forget_acl(VFS_I(sc->ip), SGI_ACL_FILE); + xfs_forget_acl(VFS_I(sc->ip), SGI_ACL_DEFAULT); + return 0; +} + +/* Tear down all the incore scan stuff we created. */ +STATIC void +xrep_xattr_teardown( + struct xrep_xattr *rx) +{ + xfblob_destroy(rx->xattr_blobs); + xfarray_destroy(rx->xattr_records); + kfree(rx); +} + +/* Set up the filesystem scan so we can regenerate extended attributes. */ +STATIC int +xrep_xattr_setup_scan( + struct xfs_scrub *sc, + struct xrep_xattr **rxp) +{ + struct xrep_xattr *rx; + char *descr; + int max_len; + int error; + + rx = kzalloc(sizeof(struct xrep_xattr), XCHK_GFP_FLAGS); + if (!rx) + return -ENOMEM; + rx->sc = sc; + + /* + * Allocate enough memory to handle loading local attr values from the + * xfblob data while flushing stashed attrs to the temporary file. + * We only realloc the buffer when salvaging remote attr values. + */ + max_len = xfs_attr_leaf_entsize_local_max(sc->mp->m_attr_geo->blksize); + error = xchk_setup_xattr_buf(rx->sc, max_len); + if (error == -ENOMEM) + error = -EDEADLOCK; + if (error) + goto out_rx; + + /* Set up some staging for salvaged attribute keys and values */ + descr = xchk_xfile_ino_descr(sc, "xattr keys"); + error = xfarray_create(descr, 0, sizeof(struct xrep_xattr_key), + &rx->xattr_records); + kfree(descr); + if (error) + goto out_rx; + + descr = xchk_xfile_ino_descr(sc, "xattr names"); + error = xfblob_create(descr, &rx->xattr_blobs); + kfree(descr); + if (error) + goto out_keys; + + *rxp = rx; + return 0; +out_keys: + xfarray_destroy(rx->xattr_records); +out_rx: + kfree(rx); + return error; +} + +/* + * Repair the extended attribute metadata. + * + * XXX: Remote attribute value buffers encompass the entire (up to 64k) buffer. + * The buffer cache in XFS can't handle aliased multiblock buffers, so this + * might misbehave if the attr fork is crosslinked with other filesystem + * metadata. + */ +int +xrep_xattr( + struct xfs_scrub *sc) +{ + struct xrep_xattr *rx = NULL; + int error; + + if (!xfs_inode_hasattr(sc->ip)) + return -ENOENT; + + /* The rmapbt is required to reap the old attr fork. */ + if (!xfs_has_rmapbt(sc->mp)) + return -EOPNOTSUPP; + + error = xrep_xattr_setup_scan(sc, &rx); + if (error) + return error; + + ASSERT(sc->ilock_flags & XFS_ILOCK_EXCL); + + error = xrep_xattr_salvage_attributes(rx); + if (error) + goto out_scan; + + /* Last chance to abort before we start committing fixes. */ + if (xchk_should_terminate(sc, &error)) + goto out_scan; + + error = xrep_xattr_rebuild_tree(rx); + if (error) + goto out_scan; + +out_scan: + xrep_xattr_teardown(rx); + return error; +} diff --git a/fs/xfs/scrub/attr_repair.h b/fs/xfs/scrub/attr_repair.h new file mode 100644 index 000000000000..0a9ffa7cfa90 --- /dev/null +++ b/fs/xfs/scrub/attr_repair.h @@ -0,0 +1,11 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2018-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_ATTR_REPAIR_H__ +#define __XFS_SCRUB_ATTR_REPAIR_H__ + +int xrep_xattr_reset_fork(struct xfs_scrub *sc); + +#endif /* __XFS_SCRUB_ATTR_REPAIR_H__ */ diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index 443e62f72481..04aec0e9e4c3 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -32,6 +32,9 @@ #include "xfs_reflink.h" #include "xfs_health.h" #include "xfs_buf_mem.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_attr.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" @@ -39,6 +42,7 @@ #include "scrub/bitmap.h" #include "scrub/stats.h" #include "scrub/xfile.h" +#include "scrub/attr_repair.h" /* * Attempt to repair some metadata, if the metadata is corrupt and userspace @@ -1136,6 +1140,17 @@ xrep_metadata_inode_forks( return error; } + /* Clear the attr forks since metadata shouldn't have that. */ + if (xfs_inode_hasattr(sc->ip)) { + if (!dirty) { + dirty = true; + xfs_trans_ijoin(sc->tp, sc->ip, 0); + } + error = xrep_xattr_reset_fork(sc); + if (error) + return error; + } + /* * If we modified the inode, roll the transaction but don't rejoin the * inode to the new transaction because xrep_bmap_data can do that. @@ -1201,3 +1216,34 @@ xrep_trans_cancel_hook_dummy( current->journal_info = *cookiep; *cookiep = NULL; } + +/* + * See if this buffer can pass the given ->verify_struct() function. + * + * If the buffer already has ops attached and they're not the ones that were + * passed in, we reject the buffer. Otherwise, we perform the structure test + * (note that we do not check CRCs) and return the outcome of the test. The + * buffer ops and error state are left unchanged. + */ +bool +xrep_buf_verify_struct( + struct xfs_buf *bp, + const struct xfs_buf_ops *ops) +{ + const struct xfs_buf_ops *old_ops = bp->b_ops; + xfs_failaddr_t fa; + int old_error; + + if (old_ops) { + if (old_ops != ops) + return false; + } + + old_error = bp->b_error; + bp->b_ops = ops; + fa = bp->b_ops->verify_struct(bp); + bp->b_ops = old_ops; + bp->b_error = old_error; + + return fa == NULL; +} diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index 0e2b695ab8f6..9cbfd8da5620 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -90,6 +90,7 @@ int xrep_bmap(struct xfs_scrub *sc, int whichfork, bool allow_unwritten); int xrep_metadata_inode_forks(struct xfs_scrub *sc); int xrep_setup_ag_rmapbt(struct xfs_scrub *sc); int xrep_setup_ag_refcountbt(struct xfs_scrub *sc); +int xrep_setup_xattr(struct xfs_scrub *sc); /* Repair setup functions */ int xrep_setup_ag_allocbt(struct xfs_scrub *sc); @@ -123,6 +124,7 @@ int xrep_bmap_attr(struct xfs_scrub *sc); int xrep_bmap_cow(struct xfs_scrub *sc); int xrep_nlinks(struct xfs_scrub *sc); int xrep_fscounters(struct xfs_scrub *sc); +int xrep_xattr(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xrep_rtbitmap(struct xfs_scrub *sc); @@ -147,6 +149,8 @@ int xrep_trans_alloc_hook_dummy(struct xfs_mount *mp, void **cookiep, struct xfs_trans **tpp); void xrep_trans_cancel_hook_dummy(void **cookiep, struct xfs_trans *tp); +bool xrep_buf_verify_struct(struct xfs_buf *bp, const struct xfs_buf_ops *ops); + #else #define xrep_ino_dqattach(sc) (0) @@ -190,6 +194,7 @@ xrep_setup_nothing( #define xrep_setup_ag_allocbt xrep_setup_nothing #define xrep_setup_ag_rmapbt xrep_setup_nothing #define xrep_setup_ag_refcountbt xrep_setup_nothing +#define xrep_setup_xattr xrep_setup_nothing #define xrep_setup_inode(sc, imap) ((void)0) @@ -215,6 +220,7 @@ xrep_setup_nothing( #define xrep_nlinks xrep_notsupported #define xrep_fscounters xrep_notsupported #define xrep_rtsummary xrep_notsupported +#define xrep_xattr xrep_notsupported #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 62a064c1a5d3..547189a14b6b 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -331,7 +331,7 @@ static const struct xchk_meta_ops meta_scrub_ops[] = { .type = ST_INODE, .setup = xchk_setup_xattr, .scrub = xchk_xattr, - .repair = xrep_notsupported, + .repair = xrep_xattr, }, [XFS_SCRUB_TYPE_SYMLINK] = { /* symbolic link */ .type = ST_INODE, diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 7d07912d8f75..026813205b47 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2416,6 +2416,89 @@ TRACE_EVENT(xreap_bmapi_binval_scan, __entry->scan_blocks) ); +TRACE_EVENT(xrep_xattr_recover_leafblock, + TP_PROTO(struct xfs_inode *ip, xfs_dablk_t dabno, uint16_t magic), + TP_ARGS(ip, dabno, magic), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_dablk_t, dabno) + __field(uint16_t, magic) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->dabno = dabno; + __entry->magic = magic; + ), + TP_printk("dev %d:%d ino 0x%llx dablk 0x%x magic 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->dabno, + __entry->magic) +); + +DECLARE_EVENT_CLASS(xrep_xattr_salvage_class, + TP_PROTO(struct xfs_inode *ip, unsigned int flags, char *name, + unsigned int namelen, unsigned int valuelen), + TP_ARGS(ip, flags, name, namelen, valuelen), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(unsigned int, flags) + __field(unsigned int, namelen) + __dynamic_array(char, name, namelen) + __field(unsigned int, valuelen) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->flags = flags; + __entry->namelen = namelen; + memcpy(__get_str(name), name, namelen); + __entry->valuelen = valuelen; + ), + TP_printk("dev %d:%d ino 0x%llx flags %s name '%.*s' valuelen 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_flags(__entry->flags, "|", XFS_ATTR_NAMESPACE_STR), + __entry->namelen, + __get_str(name), + __entry->valuelen) +); +#define DEFINE_XREP_XATTR_SALVAGE_EVENT(name) \ +DEFINE_EVENT(xrep_xattr_salvage_class, name, \ + TP_PROTO(struct xfs_inode *ip, unsigned int flags, char *name, \ + unsigned int namelen, unsigned int valuelen), \ + TP_ARGS(ip, flags, name, namelen, valuelen)) +DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_xattr_salvage_rec); +DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_xattr_insert_rec); + +TRACE_EVENT(xrep_xattr_class, + TP_PROTO(struct xfs_inode *ip, struct xfs_inode *arg_ip), + TP_ARGS(ip, arg_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_ino_t, src_ino) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->src_ino = arg_ip->i_ino; + ), + TP_printk("dev %d:%d ino 0x%llx src 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->src_ino) +) +#define DEFINE_XREP_XATTR_EVENT(name) \ +DEFINE_EVENT(xrep_xattr_class, name, \ + TP_PROTO(struct xfs_inode *ip, struct xfs_inode *arg_ip), \ + TP_ARGS(ip, arg_ip)) +DEFINE_XREP_XATTR_EVENT(xrep_xattr_rebuild_tree); +DEFINE_XREP_XATTR_EVENT(xrep_xattr_reset_fork); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ diff --git a/fs/xfs/scrub/xfarray.c b/fs/xfs/scrub/xfarray.c index 17c982a4821d..b65cd3fc5ac9 100644 --- a/fs/xfs/scrub/xfarray.c +++ b/fs/xfs/scrub/xfarray.c @@ -1051,3 +1051,20 @@ out_free: kvfree(si); return error; } + +/* How many bytes is this array consuming? */ +unsigned long long +xfarray_bytes( + struct xfarray *array) +{ + return xfile_bytes(array->xfile); +} + +/* Empty the entire array. */ +void +xfarray_truncate( + struct xfarray *array) +{ + xfile_discard(array->xfile, 0, MAX_LFS_FILESIZE); + array->nr = 0; +} diff --git a/fs/xfs/scrub/xfarray.h b/fs/xfs/scrub/xfarray.h index acb2f94c56c1..3b10a58e9f14 100644 --- a/fs/xfs/scrub/xfarray.h +++ b/fs/xfs/scrub/xfarray.h @@ -44,6 +44,8 @@ int xfarray_unset(struct xfarray *array, xfarray_idx_t idx); int xfarray_store(struct xfarray *array, xfarray_idx_t idx, const void *ptr); int xfarray_store_anywhere(struct xfarray *array, const void *ptr); bool xfarray_element_is_null(struct xfarray *array, const void *ptr); +void xfarray_truncate(struct xfarray *array); +unsigned long long xfarray_bytes(struct xfarray *array); /* * Load an array element, but zero the buffer if there's no data because we diff --git a/fs/xfs/scrub/xfblob.c b/fs/xfs/scrub/xfblob.c index cec668debce5..6ef2a9637f16 100644 --- a/fs/xfs/scrub/xfblob.c +++ b/fs/xfs/scrub/xfblob.c @@ -149,3 +149,20 @@ xfblob_free( xfile_discard(blob->xfile, cookie, sizeof(key) + key.xb_size); return 0; } + +/* How many bytes is this blob storage object consuming? */ +unsigned long long +xfblob_bytes( + struct xfblob *blob) +{ + return xfile_bytes(blob->xfile); +} + +/* Drop all the blobs. */ +void +xfblob_truncate( + struct xfblob *blob) +{ + xfile_discard(blob->xfile, PAGE_SIZE, MAX_LFS_FILESIZE - PAGE_SIZE); + blob->last_offset = PAGE_SIZE; +} diff --git a/fs/xfs/scrub/xfblob.h b/fs/xfs/scrub/xfblob.h index bd98647407f1..78a67a06408f 100644 --- a/fs/xfs/scrub/xfblob.h +++ b/fs/xfs/scrub/xfblob.h @@ -20,5 +20,7 @@ int xfblob_load(struct xfblob *blob, xfblob_cookie cookie, void *ptr, int xfblob_store(struct xfblob *blob, xfblob_cookie *cookie, const void *ptr, uint32_t size); int xfblob_free(struct xfblob *blob, xfblob_cookie cookie); +unsigned long long xfblob_bytes(struct xfblob *blob); +void xfblob_truncate(struct xfblob *blob); #endif /* __XFS_SCRUB_XFBLOB_H__ */ diff --git a/fs/xfs/scrub/xfile.h b/fs/xfs/scrub/xfile.h index 8dfbae1fe33a..cc2cc1714cd4 100644 --- a/fs/xfs/scrub/xfile.h +++ b/fs/xfs/scrub/xfile.h @@ -27,4 +27,9 @@ struct folio *xfile_get_folio(struct xfile *xf, loff_t offset, size_t len, unsigned int flags); void xfile_put_folio(struct xfile *xf, struct folio *folio); +static inline unsigned long long xfile_bytes(struct xfile *xf) +{ + return file_inode(xf->file)->i_blocks << SECTOR_SHIFT; +} + #endif /* __XFS_SCRUB_XFILE_H__ */ diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index f0fa02264eda..8a0151e23f3d 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -494,6 +494,9 @@ _xfs_buf_obj_cmp( * it stale has not yet committed. i.e. we are * reallocating a busy extent. Skip this buffer and * continue searching for an exact match. + * + * Note: If we're scanning for incore buffers to stale, don't + * complain if we find non-stale buffers. */ if (!(map->bm_flags & XBM_LIVESCAN)) ASSERT(bp->b_flags & XBF_STALE); diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 939baf08331b..a80c3063a13f 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -31,6 +31,8 @@ * pos: file offset, in bytes * bytecount: number of bytes * + * dablk: directory or xattr block offset, in filesystem blocks + * * disize: ondisk file size, in bytes * isize: incore file size, in bytes * -- cgit v1.2.3-59-g8ed1b From 0ee230dec2626ef25dc96abd47b84494f9c251e3 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:46 -0700 Subject: xfs: scrub should set preen if attr leaf has holes If an attr block indicates that it could use compaction, set the preen flag to have the attr fork rebuilt, since the attr fork rebuilder can take care of that for us. Signed-off-by: Darrick J. Wong Reviewed-by: Dave Chinner Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr.c | 2 ++ fs/xfs/scrub/dabtree.c | 16 ++++++++++++++++ fs/xfs/scrub/dabtree.h | 1 + fs/xfs/scrub/trace.h | 1 + 4 files changed, 20 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 7621e548d730..ba06be86ac7d 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -428,6 +428,8 @@ xchk_xattr_block( xchk_da_set_corrupt(ds, level); if (!xchk_xattr_set_map(ds->sc, ab->usedmap, 0, hdrsize)) xchk_da_set_corrupt(ds, level); + if (leafhdr.holes) + xchk_da_set_preen(ds, level); if (ds->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) goto out; diff --git a/fs/xfs/scrub/dabtree.c b/fs/xfs/scrub/dabtree.c index c71254088dff..056de4819f86 100644 --- a/fs/xfs/scrub/dabtree.c +++ b/fs/xfs/scrub/dabtree.c @@ -78,6 +78,22 @@ xchk_da_set_corrupt( __return_address); } +/* Flag a da btree node in need of optimization. */ +void +xchk_da_set_preen( + struct xchk_da_btree *ds, + int level) +{ + struct xfs_scrub *sc = ds->sc; + + sc->sm->sm_flags |= XFS_SCRUB_OFLAG_PREEN; + trace_xchk_fblock_preen(sc, ds->dargs.whichfork, + xfs_dir2_da_to_db(ds->dargs.geo, + ds->state->path.blk[level].blkno), + __return_address); +} + +/* Find an entry at a certain level in a da btree. */ static struct xfs_da_node_entry * xchk_da_btree_node_entry( struct xchk_da_btree *ds, diff --git a/fs/xfs/scrub/dabtree.h b/fs/xfs/scrub/dabtree.h index 4f8c2138a1ec..d654c125feb4 100644 --- a/fs/xfs/scrub/dabtree.h +++ b/fs/xfs/scrub/dabtree.h @@ -35,6 +35,7 @@ bool xchk_da_process_error(struct xchk_da_btree *ds, int level, int *error); /* Check for da btree corruption. */ void xchk_da_set_corrupt(struct xchk_da_btree *ds, int level); +void xchk_da_set_preen(struct xchk_da_btree *ds, int level); int xchk_da_btree_hash(struct xchk_da_btree *ds, int level, __be32 *hashp); int xchk_da_btree(struct xfs_scrub *sc, int whichfork, diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 026813205b47..ffaff7722bf2 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -365,6 +365,7 @@ DEFINE_EVENT(xchk_fblock_error_class, name, \ DEFINE_SCRUB_FBLOCK_ERROR_EVENT(xchk_fblock_error); DEFINE_SCRUB_FBLOCK_ERROR_EVENT(xchk_fblock_warning); +DEFINE_SCRUB_FBLOCK_ERROR_EVENT(xchk_fblock_preen); #ifdef CONFIG_XFS_QUOTA DECLARE_EVENT_CLASS(xchk_dqiter_class, -- cgit v1.2.3-59-g8ed1b From 40190f9f918aaf3355c8f777e6e658ea3bf77870 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:47 -0700 Subject: xfs: flag empty xattr leaf blocks for optimization Empty xattr leaf blocks at offset zero are a waste of space but otherwise harmless. If we encounter one, flag it as an opportunity for optimization. If we encounter empty attr leaf blocks anywhere else in the attr fork, that's corruption. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr.c | 11 +++++++++++ fs/xfs/scrub/dabtree.h | 2 ++ 2 files changed, 13 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index ba06be86ac7d..696971204b87 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -420,6 +420,17 @@ xchk_xattr_block( xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &leafhdr, leaf); hdrsize = xfs_attr3_leaf_hdr_size(leaf); + /* + * Empty xattr leaf blocks mapped at block 0 are probably a byproduct + * of a race between setxattr and a log shutdown. Anywhere else in the + * attr fork is a corruption. + */ + if (leafhdr.count == 0) { + if (blk->blkno == 0) + xchk_da_set_preen(ds, level); + else + xchk_da_set_corrupt(ds, level); + } if (leafhdr.usedbytes > mp->m_attr_geo->blksize) xchk_da_set_corrupt(ds, level); if (leafhdr.firstused > mp->m_attr_geo->blksize) diff --git a/fs/xfs/scrub/dabtree.h b/fs/xfs/scrub/dabtree.h index d654c125feb4..de291e3b77dd 100644 --- a/fs/xfs/scrub/dabtree.h +++ b/fs/xfs/scrub/dabtree.h @@ -37,6 +37,8 @@ bool xchk_da_process_error(struct xchk_da_btree *ds, int level, int *error); void xchk_da_set_corrupt(struct xchk_da_btree *ds, int level); void xchk_da_set_preen(struct xchk_da_btree *ds, int level); +void xchk_da_set_preen(struct xchk_da_btree *ds, int level); + int xchk_da_btree_hash(struct xchk_da_btree *ds, int level, __be32 *hashp); int xchk_da_btree(struct xfs_scrub *sc, int whichfork, xchk_da_btree_rec_fn scrub_fn, void *private); -- cgit v1.2.3-59-g8ed1b From 6c631e79e73c7122c890ef943f8ca9aab9e1dec8 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:48 -0700 Subject: xfs: create an xattr iteration function for scrub Create a streamlined function to walk a file's xattrs, without all the cursor management stuff in the regular listxattr. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/attr.c | 125 +++++++------------ fs/xfs/scrub/dab_bitmap.h | 37 ++++++ fs/xfs/scrub/listxattr.c | 312 ++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/listxattr.h | 17 +++ 5 files changed, 414 insertions(+), 78 deletions(-) create mode 100644 fs/xfs/scrub/dab_bitmap.h create mode 100644 fs/xfs/scrub/listxattr.c create mode 100644 fs/xfs/scrub/listxattr.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 8647629ac7bf..7dbe6b3befb3 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -165,6 +165,7 @@ xfs-y += $(addprefix scrub/, \ ialloc.o \ inode.o \ iscan.o \ + listxattr.o \ nlinks.o \ parent.o \ readdir.o \ diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 696971204b87..8853e4d0eee3 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -21,6 +21,7 @@ #include "scrub/common.h" #include "scrub/dabtree.h" #include "scrub/attr.h" +#include "scrub/listxattr.h" #include "scrub/repair.h" /* Free the buffers linked from the xattr buffer. */ @@ -153,90 +154,81 @@ xchk_setup_xattr( /* Extended Attributes */ -struct xchk_xattr { - struct xfs_attr_list_context context; - struct xfs_scrub *sc; -}; - /* * Check that an extended attribute key can be looked up by hash. * - * We use the XFS attribute list iterator (i.e. xfs_attr_list_ilocked) - * to call this function for every attribute key in an inode. Once - * we're here, we load the attribute value to see if any errors happen, - * or if we get more or less data than we expected. + * We use the extended attribute walk helper to call this function for every + * attribute key in an inode. Once we're here, we load the attribute value to + * see if any errors happen, or if we get more or less data than we expected. */ -static void -xchk_xattr_listent( - struct xfs_attr_list_context *context, - int flags, - unsigned char *name, - int namelen, - int valuelen) +static int +xchk_xattr_actor( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) { struct xfs_da_args args = { .op_flags = XFS_DA_OP_NOTIME, - .attr_filter = flags & XFS_ATTR_NSP_ONDISK_MASK, - .geo = context->dp->i_mount->m_attr_geo, + .attr_filter = attr_flags & XFS_ATTR_NSP_ONDISK_MASK, + .geo = sc->mp->m_attr_geo, .whichfork = XFS_ATTR_FORK, - .dp = context->dp, + .dp = ip, .name = name, .namelen = namelen, .hashval = xfs_da_hashname(name, namelen), - .trans = context->tp, + .trans = sc->tp, .valuelen = valuelen, - .owner = context->dp->i_ino, + .owner = ip->i_ino, }; struct xchk_xattr_buf *ab; - struct xchk_xattr *sx; int error = 0; - sx = container_of(context, struct xchk_xattr, context); - ab = sx->sc->buf; + ab = sc->buf; - if (xchk_should_terminate(sx->sc, &error)) { - context->seen_enough = error; - return; - } + if (xchk_should_terminate(sc, &error)) + return error; - if (flags & XFS_ATTR_INCOMPLETE) { + if (attr_flags & XFS_ATTR_INCOMPLETE) { /* Incomplete attr key, just mark the inode for preening. */ - xchk_ino_set_preen(sx->sc, context->dp->i_ino); - return; + xchk_ino_set_preen(sc, ip->i_ino); + return 0; } /* Only one namespace bit allowed. */ - if (hweight32(flags & XFS_ATTR_NSP_ONDISK_MASK) > 1) { - xchk_fblock_set_corrupt(sx->sc, XFS_ATTR_FORK, args.blkno); - goto fail_xref; + if (hweight32(attr_flags & XFS_ATTR_NSP_ONDISK_MASK) > 1) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, args.blkno); + return -ECANCELED; } /* Does this name make sense? */ if (!xfs_attr_namecheck(name, namelen)) { - xchk_fblock_set_corrupt(sx->sc, XFS_ATTR_FORK, args.blkno); - goto fail_xref; + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, args.blkno); + return -ECANCELED; } /* - * Local xattr values are stored in the attr leaf block, so we don't - * need to retrieve the value from a remote block to detect corruption - * problems. + * Local and shortform xattr values are stored in the attr leaf block, + * so we don't need to retrieve the value from a remote block to detect + * corruption problems. */ - if (flags & XFS_ATTR_LOCAL) - goto fail_xref; + if (value) + return 0; /* - * Try to allocate enough memory to extrat the attr value. If that - * doesn't work, we overload the seen_enough variable to convey - * the error message back to the main scrub function. + * Try to allocate enough memory to extract the attr value. If that + * doesn't work, return -EDEADLOCK as a signal to try again with a + * maximally sized buffer. */ - error = xchk_setup_xattr_buf(sx->sc, valuelen); + error = xchk_setup_xattr_buf(sc, valuelen); if (error == -ENOMEM) error = -EDEADLOCK; - if (error) { - context->seen_enough = error; - return; - } + if (error) + return error; args.value = ab->value; @@ -244,16 +236,13 @@ xchk_xattr_listent( /* ENODATA means the hash lookup failed and the attr is bad */ if (error == -ENODATA) error = -EFSCORRUPTED; - if (!xchk_fblock_process_error(sx->sc, XFS_ATTR_FORK, args.blkno, + if (!xchk_fblock_process_error(sc, XFS_ATTR_FORK, args.blkno, &error)) - goto fail_xref; + return error; if (args.valuelen != valuelen) - xchk_fblock_set_corrupt(sx->sc, XFS_ATTR_FORK, - args.blkno); -fail_xref: - if (sx->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) - context->seen_enough = 1; - return; + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, args.blkno); + + return 0; } /* @@ -618,16 +607,6 @@ int xchk_xattr( struct xfs_scrub *sc) { - struct xchk_xattr sx = { - .sc = sc, - .context = { - .dp = sc->ip, - .tp = sc->tp, - .resynch = 1, - .put_listent = xchk_xattr_listent, - .allow_incomplete = true, - }, - }; xfs_dablk_t last_checked = -1U; int error = 0; @@ -656,12 +635,6 @@ xchk_xattr( /* * Look up every xattr in this file by name and hash. * - * Use the backend implementation of xfs_attr_list to call - * xchk_xattr_listent on every attribute key in this inode. - * In other words, we use the same iterator/callback mechanism - * that listattr uses to scrub extended attributes, though in our - * _listent function, we check the value of the attribute. - * * The VFS only locks i_rwsem when modifying attrs, so keep all * three locks held because that's the only way to ensure we're * the only thread poking into the da btree. We traverse the da @@ -669,13 +642,9 @@ xchk_xattr( * iteration, which doesn't really follow the usual buffer * locking order. */ - error = xfs_attr_list_ilocked(&sx.context); + error = xchk_xattr_walk(sc, sc->ip, xchk_xattr_actor, NULL); if (!xchk_fblock_process_error(sc, XFS_ATTR_FORK, 0, &error)) return error; - /* Did our listent function try to return any errors? */ - if (sx.context.seen_enough < 0) - return sx.context.seen_enough; - return 0; } diff --git a/fs/xfs/scrub/dab_bitmap.h b/fs/xfs/scrub/dab_bitmap.h new file mode 100644 index 000000000000..0c6e3aad4395 --- /dev/null +++ b/fs/xfs/scrub/dab_bitmap.h @@ -0,0 +1,37 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2022-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_DAB_BITMAP_H__ +#define __XFS_SCRUB_DAB_BITMAP_H__ + +/* Bitmaps, but for type-checked for xfs_dablk_t */ + +struct xdab_bitmap { + struct xbitmap32 dabitmap; +}; + +static inline void xdab_bitmap_init(struct xdab_bitmap *bitmap) +{ + xbitmap32_init(&bitmap->dabitmap); +} + +static inline void xdab_bitmap_destroy(struct xdab_bitmap *bitmap) +{ + xbitmap32_destroy(&bitmap->dabitmap); +} + +static inline int xdab_bitmap_set(struct xdab_bitmap *bitmap, + xfs_dablk_t dabno, xfs_extlen_t len) +{ + return xbitmap32_set(&bitmap->dabitmap, dabno, len); +} + +static inline bool xdab_bitmap_test(struct xdab_bitmap *bitmap, + xfs_dablk_t dabno, xfs_extlen_t *len) +{ + return xbitmap32_test(&bitmap->dabitmap, dabno, len); +} + +#endif /* __XFS_SCRUB_DAB_BITMAP_H__ */ diff --git a/fs/xfs/scrub/listxattr.c b/fs/xfs/scrub/listxattr.c new file mode 100644 index 000000000000..cbe5911ecbbc --- /dev/null +++ b/fs/xfs/scrub/listxattr.c @@ -0,0 +1,312 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2022-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_inode.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_attr.h" +#include "xfs_attr_leaf.h" +#include "xfs_attr_sf.h" +#include "xfs_trans.h" +#include "scrub/scrub.h" +#include "scrub/bitmap.h" +#include "scrub/dab_bitmap.h" +#include "scrub/listxattr.h" + +/* Call a function for every entry in a shortform xattr structure. */ +STATIC int +xchk_xattr_walk_sf( + struct xfs_scrub *sc, + struct xfs_inode *ip, + xchk_xattr_fn attr_fn, + void *priv) +{ + struct xfs_attr_sf_hdr *hdr = ip->i_af.if_data; + struct xfs_attr_sf_entry *sfe; + unsigned int i; + int error; + + sfe = xfs_attr_sf_firstentry(hdr); + for (i = 0; i < hdr->count; i++) { + error = attr_fn(sc, ip, sfe->flags, sfe->nameval, sfe->namelen, + &sfe->nameval[sfe->namelen], sfe->valuelen, + priv); + if (error) + return error; + + sfe = xfs_attr_sf_nextentry(sfe); + } + + return 0; +} + +/* Call a function for every entry in this xattr leaf block. */ +STATIC int +xchk_xattr_walk_leaf_entries( + struct xfs_scrub *sc, + struct xfs_inode *ip, + xchk_xattr_fn attr_fn, + struct xfs_buf *bp, + void *priv) +{ + struct xfs_attr3_icleaf_hdr ichdr; + struct xfs_mount *mp = sc->mp; + struct xfs_attr_leafblock *leaf = bp->b_addr; + struct xfs_attr_leaf_entry *entry; + unsigned int i; + int error; + + xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &ichdr, leaf); + entry = xfs_attr3_leaf_entryp(leaf); + + for (i = 0; i < ichdr.count; entry++, i++) { + void *value; + unsigned char *name; + unsigned int namelen, valuelen; + + if (entry->flags & XFS_ATTR_LOCAL) { + struct xfs_attr_leaf_name_local *name_loc; + + name_loc = xfs_attr3_leaf_name_local(leaf, i); + name = name_loc->nameval; + namelen = name_loc->namelen; + value = &name_loc->nameval[name_loc->namelen]; + valuelen = be16_to_cpu(name_loc->valuelen); + } else { + struct xfs_attr_leaf_name_remote *name_rmt; + + name_rmt = xfs_attr3_leaf_name_remote(leaf, i); + name = name_rmt->name; + namelen = name_rmt->namelen; + value = NULL; + valuelen = be32_to_cpu(name_rmt->valuelen); + } + + error = attr_fn(sc, ip, entry->flags, name, namelen, value, + valuelen, priv); + if (error) + return error; + + } + + return 0; +} + +/* + * Call a function for every entry in a leaf-format xattr structure. Avoid + * memory allocations for the loop detector since there's only one block. + */ +STATIC int +xchk_xattr_walk_leaf( + struct xfs_scrub *sc, + struct xfs_inode *ip, + xchk_xattr_fn attr_fn, + void *priv) +{ + struct xfs_buf *leaf_bp; + int error; + + error = xfs_attr3_leaf_read(sc->tp, ip, ip->i_ino, 0, &leaf_bp); + if (error) + return error; + + error = xchk_xattr_walk_leaf_entries(sc, ip, attr_fn, leaf_bp, priv); + xfs_trans_brelse(sc->tp, leaf_bp); + return error; +} + +/* Find the leftmost leaf in the xattr dabtree. */ +STATIC int +xchk_xattr_find_leftmost_leaf( + struct xfs_scrub *sc, + struct xfs_inode *ip, + struct xdab_bitmap *seen_dablks, + struct xfs_buf **leaf_bpp) +{ + struct xfs_da3_icnode_hdr nodehdr; + struct xfs_mount *mp = sc->mp; + struct xfs_trans *tp = sc->tp; + struct xfs_da_intnode *node; + struct xfs_da_node_entry *btree; + struct xfs_buf *bp; + xfs_failaddr_t fa; + xfs_dablk_t blkno = 0; + unsigned int expected_level = 0; + int error; + + for (;;) { + xfs_extlen_t len = 1; + uint16_t magic; + + /* Make sure we haven't seen this new block already. */ + if (xdab_bitmap_test(seen_dablks, blkno, &len)) + return -EFSCORRUPTED; + + error = xfs_da3_node_read(tp, ip, blkno, &bp, XFS_ATTR_FORK); + if (error) + return error; + + node = bp->b_addr; + magic = be16_to_cpu(node->hdr.info.magic); + if (magic == XFS_ATTR_LEAF_MAGIC || + magic == XFS_ATTR3_LEAF_MAGIC) + break; + + error = -EFSCORRUPTED; + if (magic != XFS_DA_NODE_MAGIC && + magic != XFS_DA3_NODE_MAGIC) + goto out_buf; + + fa = xfs_da3_node_header_check(bp, ip->i_ino); + if (fa) + goto out_buf; + + xfs_da3_node_hdr_from_disk(mp, &nodehdr, node); + + if (nodehdr.count == 0 || nodehdr.level >= XFS_DA_NODE_MAXDEPTH) + goto out_buf; + + /* Check the level from the root node. */ + if (blkno == 0) + expected_level = nodehdr.level - 1; + else if (expected_level != nodehdr.level) + goto out_buf; + else + expected_level--; + + /* Remember that we've seen this node. */ + error = xdab_bitmap_set(seen_dablks, blkno, 1); + if (error) + goto out_buf; + + /* Find the next level towards the leaves of the dabtree. */ + btree = nodehdr.btree; + blkno = be32_to_cpu(btree->before); + xfs_trans_brelse(tp, bp); + } + + error = -EFSCORRUPTED; + fa = xfs_attr3_leaf_header_check(bp, ip->i_ino); + if (fa) + goto out_buf; + + if (expected_level != 0) + goto out_buf; + + /* Remember that we've seen this leaf. */ + error = xdab_bitmap_set(seen_dablks, blkno, 1); + if (error) + goto out_buf; + + *leaf_bpp = bp; + return 0; + +out_buf: + xfs_trans_brelse(tp, bp); + return error; +} + +/* Call a function for every entry in a node-format xattr structure. */ +STATIC int +xchk_xattr_walk_node( + struct xfs_scrub *sc, + struct xfs_inode *ip, + xchk_xattr_fn attr_fn, + void *priv) +{ + struct xfs_attr3_icleaf_hdr leafhdr; + struct xdab_bitmap seen_dablks; + struct xfs_mount *mp = sc->mp; + struct xfs_attr_leafblock *leaf; + struct xfs_buf *leaf_bp; + int error; + + xdab_bitmap_init(&seen_dablks); + + error = xchk_xattr_find_leftmost_leaf(sc, ip, &seen_dablks, &leaf_bp); + if (error) + goto out_bitmap; + + for (;;) { + xfs_extlen_t len; + + error = xchk_xattr_walk_leaf_entries(sc, ip, attr_fn, leaf_bp, + priv); + if (error) + goto out_leaf; + + /* Find the right sibling of this leaf block. */ + leaf = leaf_bp->b_addr; + xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &leafhdr, leaf); + if (leafhdr.forw == 0) + goto out_leaf; + + xfs_trans_brelse(sc->tp, leaf_bp); + + /* Make sure we haven't seen this new leaf already. */ + len = 1; + if (xdab_bitmap_test(&seen_dablks, leafhdr.forw, &len)) { + error = -EFSCORRUPTED; + goto out_bitmap; + } + + error = xfs_attr3_leaf_read(sc->tp, ip, ip->i_ino, + leafhdr.forw, &leaf_bp); + if (error) + goto out_bitmap; + + /* Remember that we've seen this new leaf. */ + error = xdab_bitmap_set(&seen_dablks, leafhdr.forw, 1); + if (error) + goto out_leaf; + } + +out_leaf: + xfs_trans_brelse(sc->tp, leaf_bp); +out_bitmap: + xdab_bitmap_destroy(&seen_dablks); + return error; +} + +/* + * Call a function for every extended attribute in a file. + * + * Callers must hold the ILOCK. No validation or cursor restarts allowed. + * Returns -EFSCORRUPTED on any problem, including loops in the dabtree. + */ +int +xchk_xattr_walk( + struct xfs_scrub *sc, + struct xfs_inode *ip, + xchk_xattr_fn attr_fn, + void *priv) +{ + int error; + + xfs_assert_ilocked(ip, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL); + + if (!xfs_inode_hasattr(ip)) + return 0; + + if (ip->i_af.if_format == XFS_DINODE_FMT_LOCAL) + return xchk_xattr_walk_sf(sc, ip, attr_fn, priv); + + /* attr functions require that the attr fork is loaded */ + error = xfs_iread_extents(sc->tp, ip, XFS_ATTR_FORK); + if (error) + return error; + + if (xfs_attr_is_leaf(ip)) + return xchk_xattr_walk_leaf(sc, ip, attr_fn, priv); + + return xchk_xattr_walk_node(sc, ip, attr_fn, priv); +} diff --git a/fs/xfs/scrub/listxattr.h b/fs/xfs/scrub/listxattr.h new file mode 100644 index 000000000000..48fe89d05946 --- /dev/null +++ b/fs/xfs/scrub/listxattr.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2022-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_LISTXATTR_H__ +#define __XFS_SCRUB_LISTXATTR_H__ + +typedef int (*xchk_xattr_fn)(struct xfs_scrub *sc, struct xfs_inode *ip, + unsigned int attr_flags, const unsigned char *name, + unsigned int namelen, const void *value, unsigned int valuelen, + void *priv); + +int xchk_xattr_walk(struct xfs_scrub *sc, struct xfs_inode *ip, + xchk_xattr_fn attr_fn, void *priv); + +#endif /* __XFS_SCRUB_LISTXATTR_H__ */ -- cgit v1.2.3-59-g8ed1b From e921533ef1a610ae92319269bd8e41ab09bf09c4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:49 -0700 Subject: xfs: ensure unlinked list state is consistent with nlink during scrub Now that we have the means to tell if an inode is on an unlinked inode list or not, we can check that an inode with zero link count is on the unlinked list; and an inode that has nonzero link count is not on that list. Make repair clean things up too. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/inode.c | 19 +++++++++++++++++++ fs/xfs/scrub/inode_repair.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_inode.c | 5 +---- fs/xfs/xfs_inode.h | 2 ++ 4 files changed, 67 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/inode.c b/fs/xfs/scrub/inode.c index 6e2fe2d6250b..d32716fb2fec 100644 --- a/fs/xfs/scrub/inode.c +++ b/fs/xfs/scrub/inode.c @@ -739,6 +739,23 @@ xchk_inode_check_reflink_iflag( xchk_ino_set_corrupt(sc, ino); } +/* + * If this inode has zero link count, it must be on the unlinked list. If + * it has nonzero link count, it must not be on the unlinked list. + */ +STATIC void +xchk_inode_check_unlinked( + struct xfs_scrub *sc) +{ + if (VFS_I(sc->ip)->i_nlink == 0) { + if (!xfs_inode_on_unlinked_list(sc->ip)) + xchk_ino_set_corrupt(sc, sc->ip->i_ino); + } else { + if (xfs_inode_on_unlinked_list(sc->ip)) + xchk_ino_set_corrupt(sc, sc->ip->i_ino); + } +} + /* Scrub an inode. */ int xchk_inode( @@ -771,6 +788,8 @@ xchk_inode( if (S_ISREG(VFS_I(sc->ip)->i_mode)) xchk_inode_check_reflink_iflag(sc, sc->ip->i_ino); + xchk_inode_check_unlinked(sc); + xchk_inode_xref(sc, sc->ip->i_ino, &di); out: return error; diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c index 097afba3043f..c743772a523e 100644 --- a/fs/xfs/scrub/inode_repair.c +++ b/fs/xfs/scrub/inode_repair.c @@ -1745,6 +1745,46 @@ xrep_inode_problems( return xrep_roll_trans(sc); } +/* + * Make sure this inode's unlinked list pointers are consistent with its + * link count. + */ +STATIC int +xrep_inode_unlinked( + struct xfs_scrub *sc) +{ + unsigned int nlink = VFS_I(sc->ip)->i_nlink; + int error; + + /* + * If this inode is linked from the directory tree and on the unlinked + * list, remove it from the unlinked list. + */ + if (nlink > 0 && xfs_inode_on_unlinked_list(sc->ip)) { + struct xfs_perag *pag; + int error; + + pag = xfs_perag_get(sc->mp, + XFS_INO_TO_AGNO(sc->mp, sc->ip->i_ino)); + error = xfs_iunlink_remove(sc->tp, pag, sc->ip); + xfs_perag_put(pag); + if (error) + return error; + } + + /* + * If this inode is not linked from the directory tree yet not on the + * unlinked list, put it on the unlinked list. + */ + if (nlink == 0 && !xfs_inode_on_unlinked_list(sc->ip)) { + error = xfs_iunlink(sc->tp, sc->ip); + if (error) + return error; + } + + return 0; +} + /* Repair an inode's fields. */ int xrep_inode( @@ -1794,5 +1834,10 @@ xrep_inode( return error; } + /* Reconnect incore unlinked list */ + error = xrep_inode_unlinked(sc); + if (error) + return error; + return xrep_defer_finish(sc); } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index ac92c0525d9b..b24c0e23d37d 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -42,9 +42,6 @@ struct kmem_cache *xfs_inode_cache; -STATIC int xfs_iunlink_remove(struct xfs_trans *tp, struct xfs_perag *pag, - struct xfs_inode *); - /* * helper function to extract extent size hint from inode */ @@ -2252,7 +2249,7 @@ xfs_iunlink_remove_inode( /* * Pull the on-disk inode from the AGI unlinked list. */ -STATIC int +int xfs_iunlink_remove( struct xfs_trans *tp, struct xfs_perag *pag, diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 596eec715675..8157ae7f8e59 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -617,6 +617,8 @@ extern struct kmem_cache *xfs_inode_cache; bool xfs_inode_needs_inactive(struct xfs_inode *ip); int xfs_iunlink(struct xfs_trans *tp, struct xfs_inode *ip); +int xfs_iunlink_remove(struct xfs_trans *tp, struct xfs_perag *pag, + struct xfs_inode *ip); void xfs_end_io(struct work_struct *work); -- cgit v1.2.3-59-g8ed1b From 669dfe883c8e20231495f80a28ec7cc0b8fdddc4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:49 -0700 Subject: xfs: update the unlinked list when repairing link counts When we're repairing the link counts of a file, we must ensure either that the file has zero link count and is on the unlinked list; or that it has nonzero link count and is not on the unlinked list. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/nlinks_repair.c | 42 +++++++++++++++++++++++++++++++++--------- 1 file changed, 33 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/nlinks_repair.c b/fs/xfs/scrub/nlinks_repair.c index b87618322f55..58cacb8e94c1 100644 --- a/fs/xfs/scrub/nlinks_repair.c +++ b/fs/xfs/scrub/nlinks_repair.c @@ -17,6 +17,7 @@ #include "xfs_iwalk.h" #include "xfs_ialloc.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/repair.h" @@ -36,6 +37,20 @@ * inode is locked. */ +/* Remove an inode from the unlinked list. */ +STATIC int +xrep_nlinks_iunlink_remove( + struct xfs_scrub *sc) +{ + struct xfs_perag *pag; + int error; + + pag = xfs_perag_get(sc->mp, XFS_INO_TO_AGNO(sc->mp, sc->ip->i_ino)); + error = xfs_iunlink_remove(sc->tp, pag, sc->ip); + xfs_perag_put(pag); + return error; +} + /* * Correct the link count of the given inode. Because we have to grab locks * and resources in a certain order, it's possible that this will be a no-op. @@ -99,16 +114,25 @@ xrep_nlinks_repair_inode( } /* - * We did not find any links to this inode. If the inode agrees, we - * have nothing further to do. If not, the inode has a nonzero link - * count and we don't have anywhere to graft the child onto. Dropping - * a live inode's link count to zero can cause unexpected shutdowns in - * inactivation, so leave it alone. + * If this inode is linked from the directory tree and on the unlinked + * list, remove it from the unlinked list. */ - if (total_links == 0) { - if (actual_nlink != 0) - trace_xrep_nlinks_unfixable_inode(mp, ip, &obs); - goto out_trans; + if (total_links > 0 && xfs_inode_on_unlinked_list(ip)) { + error = xrep_nlinks_iunlink_remove(sc); + if (error) + goto out_trans; + dirty = true; + } + + /* + * If this inode is not linked from the directory tree yet not on the + * unlinked list, put it on the unlinked list. + */ + if (total_links == 0 && !xfs_inode_on_unlinked_list(ip)) { + error = xfs_iunlink(sc->tp, ip); + if (error) + goto out_trans; + dirty = true; } /* Commit the new link count if it changed. */ -- cgit v1.2.3-59-g8ed1b From 8d81082a8c9541bdf6164c4639dc1936209fe1c4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:50 -0700 Subject: xfs: inactivate directory data blocks Teach inode inactivation to delete all the incore buffers backing a directory. In normal runtime this should never happen because the VFS forbids rmdir on a non-empty directory. In the next patch, online directory repair stands up a new directory, exchanges it with the broken directory, and then drops the private temporary directory. If we cancel the repair just prior to exchanging the directory contents, the new directory will need to be torn down. Note: If we commit the repair, reaping will take care of all the ondisk space allocations and incore buffers for the old corrupt directory. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_inode.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index b24c0e23d37d..09d643a9e997 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -16,6 +16,7 @@ #include "xfs_inode.h" #include "xfs_dir2.h" #include "xfs_attr.h" +#include "xfs_bit.h" #include "xfs_trans_space.h" #include "xfs_trans.h" #include "xfs_buf_item.h" @@ -1551,6 +1552,51 @@ out_unlock: return error; } +/* + * Mark all the buffers attached to this directory stale. In theory we should + * never be freeing a directory with any blocks at all, but this covers the + * case where we've recovered a directory swap with a "temporary" directory + * created by online repair and now need to dump it. + */ +STATIC void +xfs_inactive_dir( + struct xfs_inode *dp) +{ + struct xfs_iext_cursor icur; + struct xfs_bmbt_irec got; + struct xfs_mount *mp = dp->i_mount; + struct xfs_da_geometry *geo = mp->m_dir_geo; + struct xfs_ifork *ifp = xfs_ifork_ptr(dp, XFS_DATA_FORK); + xfs_fileoff_t off; + + /* + * Invalidate each directory block. All directory blocks are of + * fsbcount length and alignment, so we only need to walk those same + * offsets. We hold the only reference to this inode, so we must wait + * for the buffer locks. + */ + for_each_xfs_iext(ifp, &icur, &got) { + for (off = round_up(got.br_startoff, geo->fsbcount); + off < got.br_startoff + got.br_blockcount; + off += geo->fsbcount) { + struct xfs_buf *bp = NULL; + xfs_fsblock_t fsbno; + int error; + + fsbno = (off - got.br_startoff) + got.br_startblock; + error = xfs_buf_incore(mp->m_ddev_targp, + XFS_FSB_TO_DADDR(mp, fsbno), + XFS_FSB_TO_BB(mp, geo->fsbcount), + XBF_LIVESCAN, &bp); + if (error) + continue; + + xfs_buf_stale(bp); + xfs_buf_relse(bp); + } + } +} + /* * xfs_inactive_truncate * @@ -1861,6 +1907,11 @@ xfs_inactive( goto out; } + if (S_ISDIR(VFS_I(ip)->i_mode) && ip->i_df.if_nextents > 0) { + xfs_inactive_dir(ip); + truncate = 1; + } + if (S_ISLNK(VFS_I(ip)->i_mode)) error = xfs_inactive_symlink(ip); else if (truncate) -- cgit v1.2.3-59-g8ed1b From b1991ee3e7cf852e95a3498801303cfbb4468681 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:51 -0700 Subject: xfs: online repair of directories If a directory looks like it's in bad shape, try to sift through the rubble to find whatever directory entries we can, scan the directory tree for the parent (if needed), stage the new directory contents in a temporary file and use the atomic extent swapping mechanism to commit the results in bulk. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/dir.c | 9 + fs/xfs/scrub/dir_repair.c | 1349 ++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/inode_repair.c | 5 + fs/xfs/scrub/nlinks.c | 23 + fs/xfs/scrub/nlinks_repair.c | 9 + fs/xfs/scrub/parent.c | 4 +- fs/xfs/scrub/readdir.c | 7 + fs/xfs/scrub/repair.c | 1 + fs/xfs/scrub/repair.h | 4 + fs/xfs/scrub/scrub.c | 2 +- fs/xfs/scrub/tempfile.c | 13 + fs/xfs/scrub/tempfile.h | 2 + fs/xfs/scrub/trace.h | 112 ++++ fs/xfs/scrub/xfblob.h | 24 + 15 files changed, 1563 insertions(+), 2 deletions(-) create mode 100644 fs/xfs/scrub/dir_repair.c (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 7dbe6b3befb3..5c9449e14f74 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -198,6 +198,7 @@ xfs-y += $(addprefix scrub/, \ attr_repair.o \ bmap_repair.o \ cow_repair.o \ + dir_repair.o \ fscounters_repair.o \ ialloc_repair.o \ inode_repair.o \ diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 7bac74621af7..3fe6ffcf9c06 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -21,12 +21,21 @@ #include "scrub/dabtree.h" #include "scrub/readdir.h" #include "scrub/health.h" +#include "scrub/repair.h" /* Set us up to scrub directories. */ int xchk_setup_directory( struct xfs_scrub *sc) { + int error; + + if (xchk_could_repair(sc)) { + error = xrep_setup_directory(sc); + if (error) + return error; + } + return xchk_setup_inode_contents(sc, 0); } diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c new file mode 100644 index 000000000000..48aa80d8c7dc --- /dev/null +++ b/fs/xfs/scrub/dir_repair.c @@ -0,0 +1,1349 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_bit.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_inode.h" +#include "xfs_icache.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_dir2.h" +#include "xfs_dir2_priv.h" +#include "xfs_bmap.h" +#include "xfs_quota.h" +#include "xfs_bmap_btree.h" +#include "xfs_trans_space.h" +#include "xfs_bmap_util.h" +#include "xfs_exchmaps.h" +#include "xfs_exchrange.h" +#include "xfs_ag.h" +#include "scrub/xfs_scrub.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/tempfile.h" +#include "scrub/tempexch.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" +#include "scrub/readdir.h" +#include "scrub/reap.h" + +/* + * Directory Repair + * ================ + * + * We repair directories by reading the directory data blocks looking for + * directory entries that look salvageable (name passes verifiers, entry points + * to a valid allocated inode, etc). Each entry worth salvaging is stashed in + * memory, and the stashed entries are periodically replayed into a temporary + * directory to constrain memory use. Batching the construction of the + * temporary directory in this fashion reduces lock cycling of the directory + * being repaired and the temporary directory, and will later become important + * for parent pointer scanning. + * + * Directory entries added to the temporary directory do not elevate the link + * counts of the inodes found. When salvaging completes, the remaining stashed + * entries are replayed to the temporary directory. An atomic mapping exchange + * is used to commit the new directory blocks to the directory being repaired. + * This will disrupt readdir cursors. + * + * Locking Issues + * -------------- + * + * If /a, /a/b, and /c are all directories, the VFS does not take i_rwsem on + * /a/b for a "mv /a/b /c/" operation. This means that only b's ILOCK protects + * b's dotdot update. This is in contrast to every other dotdot update (link, + * remove, mkdir). If the repair code drops the ILOCK, it must either + * revalidate the dotdot entry or use dirent hooks to capture updates from + * other threads. + */ + +/* Directory entry to be restored in the new directory. */ +struct xrep_dirent { + /* Cookie for retrieval of the dirent name. */ + xfblob_cookie name_cookie; + + /* Target inode number. */ + xfs_ino_t ino; + + /* Length of the dirent name. */ + uint8_t namelen; + + /* File type of the dirent. */ + uint8_t ftype; +}; + +/* + * Stash up to 8 pages of recovered dirent data in dir_entries and dir_names + * before we write them to the temp dir. + */ +#define XREP_DIR_MAX_STASH_BYTES (PAGE_SIZE * 8) + +struct xrep_dir { + struct xfs_scrub *sc; + + /* Fixed-size array of xrep_dirent structures. */ + struct xfarray *dir_entries; + + /* Blobs containing directory entry names. */ + struct xfblob *dir_names; + + /* Information for exchanging data forks at the end. */ + struct xrep_tempexch tx; + + /* Preallocated args struct for performing dir operations */ + struct xfs_da_args args; + + /* + * This is the parent that we're going to set on the reconstructed + * directory. + */ + xfs_ino_t parent_ino; + + /* How many subdirectories did we find? */ + uint64_t subdirs; + + /* How many dirents did we find? */ + unsigned int dirents; + + /* Directory entry name, plus the trailing null. */ + struct xfs_name xname; + unsigned char namebuf[MAXNAMELEN]; +}; + +/* Tear down all the incore stuff we created. */ +static void +xrep_dir_teardown( + struct xfs_scrub *sc) +{ + struct xrep_dir *rd = sc->buf; + + xfblob_destroy(rd->dir_names); + xfarray_destroy(rd->dir_entries); +} + +/* Set up for a directory repair. */ +int +xrep_setup_directory( + struct xfs_scrub *sc) +{ + struct xrep_dir *rd; + int error; + + error = xrep_tempfile_create(sc, S_IFDIR); + if (error) + return error; + + rd = kvzalloc(sizeof(struct xrep_dir), XCHK_GFP_FLAGS); + if (!rd) + return -ENOMEM; + rd->sc = sc; + rd->xname.name = rd->namebuf; + sc->buf = rd; + + return 0; +} + +/* + * If we're the root of a directory tree, we are our own parent. If we're an + * unlinked directory, the parent /won't/ have a link to us. Set the parent + * directory to the root for both cases. Returns NULLFSINO if we don't know + * what to do. + */ +static inline xfs_ino_t +xrep_dir_self_parent( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + + if (sc->ip->i_ino == sc->mp->m_sb.sb_rootino) + return sc->mp->m_sb.sb_rootino; + + if (VFS_I(sc->ip)->i_nlink == 0) + return sc->mp->m_sb.sb_rootino; + + return NULLFSINO; +} + +/* + * Look up the dotdot entry. Returns NULLFSINO if we don't know what to do. + * The next patch will check this more carefully. + */ +static inline xfs_ino_t +xrep_dir_lookup_parent( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + xfs_ino_t ino; + int error; + + error = xfs_dir_lookup(sc->tp, sc->ip, &xfs_name_dotdot, &ino, NULL); + if (error) + return NULLFSINO; + if (!xfs_verify_dir_ino(sc->mp, ino)) + return NULLFSINO; + + return ino; +} + +/* + * Try to find the parent of the directory being repaired. + * + * NOTE: This function will someday be augmented by the directory parent repair + * code, which will know how to check the parent and scan the filesystem if + * we cannot find anything. Inode scans will have to be done before we start + * salvaging directory entries, so we do this now. + */ +STATIC int +xrep_dir_find_parent( + struct xrep_dir *rd) +{ + xfs_ino_t ino; + + ino = xrep_dir_self_parent(rd); + if (ino != NULLFSINO) { + rd->parent_ino = ino; + return 0; + } + + ino = xrep_dir_lookup_parent(rd); + if (ino != NULLFSINO) { + rd->parent_ino = ino; + return 0; + } + + /* NOTE: A future patch will deal with moving orphans. */ + return -EFSCORRUPTED; +} + +/* + * Decide if we want to salvage this entry. We don't bother with oversized + * names or the dot entry. + */ +STATIC int +xrep_dir_want_salvage( + struct xrep_dir *rd, + const char *name, + int namelen, + xfs_ino_t ino) +{ + struct xfs_mount *mp = rd->sc->mp; + + /* No pointers to ourselves or to garbage. */ + if (ino == rd->sc->ip->i_ino) + return false; + if (!xfs_verify_dir_ino(mp, ino)) + return false; + + /* No weird looking names or dot entries. */ + if (namelen >= MAXNAMELEN || namelen <= 0) + return false; + if (namelen == 1 && name[0] == '.') + return false; + if (!xfs_dir2_namecheck(name, namelen)) + return false; + + return true; +} + +/* + * Remember that we want to create a dirent in the tempdir. These stashed + * actions will be replayed later. + */ +STATIC int +xrep_dir_stash_createname( + struct xrep_dir *rd, + const struct xfs_name *name, + xfs_ino_t ino) +{ + struct xrep_dirent dirent = { + .ino = ino, + .namelen = name->len, + .ftype = name->type, + }; + int error; + + trace_xrep_dir_stash_createname(rd->sc->tempip, name, ino); + + error = xfblob_storename(rd->dir_names, &dirent.name_cookie, name); + if (error) + return error; + + return xfarray_append(rd->dir_entries, &dirent); +} + +/* Allocate an in-core record to hold entries while we rebuild the dir data. */ +STATIC int +xrep_dir_salvage_entry( + struct xrep_dir *rd, + unsigned char *name, + unsigned int namelen, + xfs_ino_t ino) +{ + struct xfs_name xname = { + .name = name, + }; + struct xfs_scrub *sc = rd->sc; + struct xfs_inode *ip; + unsigned int i = 0; + int error = 0; + + if (xchk_should_terminate(sc, &error)) + return error; + + /* + * Truncate the name to the first character that would trip namecheck. + * If we no longer have a name after that, ignore this entry. + */ + while (i < namelen && name[i] != 0 && name[i] != '/') + i++; + if (i == 0) + return 0; + xname.len = i; + + /* Ignore '..' entries; we already picked the new parent. */ + if (xname.len == 2 && name[0] == '.' && name[1] == '.') { + trace_xrep_dir_salvaged_parent(sc->ip, ino); + return 0; + } + + trace_xrep_dir_salvage_entry(sc->ip, &xname, ino); + + /* + * Compute the ftype or dump the entry if we can't. We don't lock the + * inode because inodes can't change type while we have a reference. + */ + error = xchk_iget(sc, ino, &ip); + if (error) + return 0; + + xname.type = xfs_mode_to_ftype(VFS_I(ip)->i_mode); + xchk_irele(sc, ip); + + return xrep_dir_stash_createname(rd, &xname, ino); +} + +/* Record a shortform directory entry for later reinsertion. */ +STATIC int +xrep_dir_salvage_sf_entry( + struct xrep_dir *rd, + struct xfs_dir2_sf_hdr *sfp, + struct xfs_dir2_sf_entry *sfep) +{ + xfs_ino_t ino; + + ino = xfs_dir2_sf_get_ino(rd->sc->mp, sfp, sfep); + if (!xrep_dir_want_salvage(rd, sfep->name, sfep->namelen, ino)) + return 0; + + return xrep_dir_salvage_entry(rd, sfep->name, sfep->namelen, ino); +} + +/* Record a regular directory entry for later reinsertion. */ +STATIC int +xrep_dir_salvage_data_entry( + struct xrep_dir *rd, + struct xfs_dir2_data_entry *dep) +{ + xfs_ino_t ino; + + ino = be64_to_cpu(dep->inumber); + if (!xrep_dir_want_salvage(rd, dep->name, dep->namelen, ino)) + return 0; + + return xrep_dir_salvage_entry(rd, dep->name, dep->namelen, ino); +} + +/* Try to recover block/data format directory entries. */ +STATIC int +xrep_dir_recover_data( + struct xrep_dir *rd, + struct xfs_buf *bp) +{ + struct xfs_da_geometry *geo = rd->sc->mp->m_dir_geo; + unsigned int offset; + unsigned int end; + int error = 0; + + /* + * Loop over the data portion of the block. + * Each object is a real entry (dep) or an unused one (dup). + */ + offset = geo->data_entry_offset; + end = min_t(unsigned int, BBTOB(bp->b_length), + xfs_dir3_data_end_offset(geo, bp->b_addr)); + + while (offset < end) { + struct xfs_dir2_data_unused *dup = bp->b_addr + offset; + struct xfs_dir2_data_entry *dep = bp->b_addr + offset; + + if (xchk_should_terminate(rd->sc, &error)) + return error; + + /* Skip unused entries. */ + if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) { + offset += be16_to_cpu(dup->length); + continue; + } + + /* Don't walk off the end of the block. */ + offset += xfs_dir2_data_entsize(rd->sc->mp, dep->namelen); + if (offset > end) + break; + + /* Ok, let's save this entry. */ + error = xrep_dir_salvage_data_entry(rd, dep); + if (error) + return error; + + } + + return 0; +} + +/* Try to recover shortform directory entries. */ +STATIC int +xrep_dir_recover_sf( + struct xrep_dir *rd) +{ + struct xfs_dir2_sf_hdr *hdr; + struct xfs_dir2_sf_entry *sfep; + struct xfs_dir2_sf_entry *next; + struct xfs_ifork *ifp; + xfs_ino_t ino; + unsigned char *end; + int error = 0; + + ifp = xfs_ifork_ptr(rd->sc->ip, XFS_DATA_FORK); + hdr = ifp->if_data; + end = (unsigned char *)ifp->if_data + ifp->if_bytes; + + ino = xfs_dir2_sf_get_parent_ino(hdr); + trace_xrep_dir_salvaged_parent(rd->sc->ip, ino); + + sfep = xfs_dir2_sf_firstentry(hdr); + while ((unsigned char *)sfep < end) { + if (xchk_should_terminate(rd->sc, &error)) + return error; + + next = xfs_dir2_sf_nextentry(rd->sc->mp, hdr, sfep); + if ((unsigned char *)next > end) + break; + + /* Ok, let's save this entry. */ + error = xrep_dir_salvage_sf_entry(rd, hdr, sfep); + if (error) + return error; + + sfep = next; + } + + return 0; +} + +/* + * Try to figure out the format of this directory from the data fork mappings + * and the directory size. If we can be reasonably sure of format, we can be + * more aggressive in salvaging directory entries. On return, @magic_guess + * will be set to DIR3_BLOCK_MAGIC if we think this is a "block format" + * directory; DIR3_DATA_MAGIC if we think this is a "data format" directory, + * and 0 if we can't tell. + */ +STATIC void +xrep_dir_guess_format( + struct xrep_dir *rd, + __be32 *magic_guess) +{ + struct xfs_inode *dp = rd->sc->ip; + struct xfs_mount *mp = rd->sc->mp; + struct xfs_da_geometry *geo = mp->m_dir_geo; + xfs_fileoff_t last; + int error; + + ASSERT(xfs_has_crc(mp)); + + *magic_guess = 0; + + /* + * If there's a single directory block and the directory size is + * exactly one block, this has to be a single block format directory. + */ + error = xfs_bmap_last_offset(dp, &last, XFS_DATA_FORK); + if (!error && XFS_FSB_TO_B(mp, last) == geo->blksize && + dp->i_disk_size == geo->blksize) { + *magic_guess = cpu_to_be32(XFS_DIR3_BLOCK_MAGIC); + return; + } + + /* + * If the last extent before the leaf offset matches the directory + * size and the directory size is larger than 1 block, this is a + * data format directory. + */ + last = geo->leafblk; + error = xfs_bmap_last_before(rd->sc->tp, dp, &last, XFS_DATA_FORK); + if (!error && + XFS_FSB_TO_B(mp, last) > geo->blksize && + XFS_FSB_TO_B(mp, last) == dp->i_disk_size) { + *magic_guess = cpu_to_be32(XFS_DIR3_DATA_MAGIC); + return; + } +} + +/* Recover directory entries from a specific directory block. */ +STATIC int +xrep_dir_recover_dirblock( + struct xrep_dir *rd, + __be32 magic_guess, + xfs_dablk_t dabno) +{ + struct xfs_dir2_data_hdr *hdr; + struct xfs_buf *bp; + __be32 oldmagic; + int error; + + /* + * Try to read buffer. We invalidate them in the next step so we don't + * bother to set a buffer type or ops. + */ + error = xfs_da_read_buf(rd->sc->tp, rd->sc->ip, dabno, + XFS_DABUF_MAP_HOLE_OK, &bp, XFS_DATA_FORK, NULL); + if (error || !bp) + return error; + + hdr = bp->b_addr; + oldmagic = hdr->magic; + + trace_xrep_dir_recover_dirblock(rd->sc->ip, dabno, + be32_to_cpu(hdr->magic), be32_to_cpu(magic_guess)); + + /* + * If we're sure of the block's format, proceed with the salvage + * operation using the specified magic number. + */ + if (magic_guess) { + hdr->magic = magic_guess; + goto recover; + } + + /* + * If we couldn't guess what type of directory this is, then we will + * only salvage entries from directory blocks that match the magic + * number and pass verifiers. + */ + switch (hdr->magic) { + case cpu_to_be32(XFS_DIR2_BLOCK_MAGIC): + case cpu_to_be32(XFS_DIR3_BLOCK_MAGIC): + if (!xrep_buf_verify_struct(bp, &xfs_dir3_block_buf_ops)) + goto out; + if (xfs_dir3_block_header_check(bp, rd->sc->ip->i_ino) != NULL) + goto out; + break; + case cpu_to_be32(XFS_DIR2_DATA_MAGIC): + case cpu_to_be32(XFS_DIR3_DATA_MAGIC): + if (!xrep_buf_verify_struct(bp, &xfs_dir3_data_buf_ops)) + goto out; + if (xfs_dir3_data_header_check(bp, rd->sc->ip->i_ino) != NULL) + goto out; + break; + default: + goto out; + } + +recover: + error = xrep_dir_recover_data(rd, bp); + +out: + hdr->magic = oldmagic; + xfs_trans_brelse(rd->sc->tp, bp); + return error; +} + +static inline void +xrep_dir_init_args( + struct xrep_dir *rd, + struct xfs_inode *dp, + const struct xfs_name *name) +{ + memset(&rd->args, 0, sizeof(struct xfs_da_args)); + rd->args.geo = rd->sc->mp->m_dir_geo; + rd->args.whichfork = XFS_DATA_FORK; + rd->args.owner = rd->sc->ip->i_ino; + rd->args.trans = rd->sc->tp; + rd->args.dp = dp; + if (!name) + return; + rd->args.name = name->name; + rd->args.namelen = name->len; + rd->args.filetype = name->type; + rd->args.hashval = xfs_dir2_hashname(rd->sc->mp, name); +} + +/* Replay a stashed createname into the temporary directory. */ +STATIC int +xrep_dir_replay_createname( + struct xrep_dir *rd, + const struct xfs_name *name, + xfs_ino_t inum, + xfs_extlen_t total) +{ + struct xfs_scrub *sc = rd->sc; + struct xfs_inode *dp = rd->sc->tempip; + bool is_block, is_leaf; + int error; + + ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); + + error = xfs_dir_ino_validate(sc->mp, inum); + if (error) + return error; + + trace_xrep_dir_replay_createname(dp, name, inum); + + xrep_dir_init_args(rd, dp, name); + rd->args.inumber = inum; + rd->args.total = total; + rd->args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; + + if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return xfs_dir2_sf_addname(&rd->args); + + error = xfs_dir2_isblock(&rd->args, &is_block); + if (error) + return error; + if (is_block) + return xfs_dir2_block_addname(&rd->args); + + error = xfs_dir2_isleaf(&rd->args, &is_leaf); + if (error) + return error; + if (is_leaf) + return xfs_dir2_leaf_addname(&rd->args); + + return xfs_dir2_node_addname(&rd->args); +} + +/* + * Add this stashed incore directory entry to the temporary directory. + * The caller must hold the tempdir's IOLOCK, must not hold any ILOCKs, and + * must not be in transaction context. + */ +STATIC int +xrep_dir_replay_update( + struct xrep_dir *rd, + const struct xfs_name *xname, + const struct xrep_dirent *dirent) +{ + struct xfs_mount *mp = rd->sc->mp; +#ifdef DEBUG + xfs_ino_t ino; +#endif + uint resblks; + int error; + + resblks = XFS_LINK_SPACE_RES(mp, xname->len); + error = xchk_trans_alloc(rd->sc, resblks); + if (error) + return error; + + /* Lock the temporary directory and join it to the transaction */ + xrep_tempfile_ilock(rd->sc); + xfs_trans_ijoin(rd->sc->tp, rd->sc->tempip, 0); + + /* + * Create a replacement dirent in the temporary directory. Note that + * _createname doesn't check for existing entries. There shouldn't be + * any in the temporary dir, but we'll verify this in debug mode. + */ +#ifdef DEBUG + error = xchk_dir_lookup(rd->sc, rd->sc->tempip, xname, &ino); + if (error != -ENOENT) { + ASSERT(error != -ENOENT); + goto out_cancel; + } +#endif + + error = xrep_dir_replay_createname(rd, xname, dirent->ino, resblks); + if (error) + goto out_cancel; + + if (xname->type == XFS_DIR3_FT_DIR) + rd->subdirs++; + rd->dirents++; + + /* Commit and unlock. */ + error = xrep_trans_commit(rd->sc); + if (error) + return error; + + xrep_tempfile_iunlock(rd->sc); + return 0; +out_cancel: + xchk_trans_cancel(rd->sc); + xrep_tempfile_iunlock(rd->sc); + return error; +} + +/* + * Flush stashed incore dirent updates that have been recorded by the scanner. + * This is done to reduce the memory requirements of the directory rebuild, + * since directories can contain up to 32GB of directory data. + * + * Caller must not hold transactions or ILOCKs. Caller must hold the tempdir + * IOLOCK. + */ +STATIC int +xrep_dir_replay_updates( + struct xrep_dir *rd) +{ + xfarray_idx_t array_cur; + int error; + + /* Add all the salvaged dirents to the temporary directory. */ + foreach_xfarray_idx(rd->dir_entries, array_cur) { + struct xrep_dirent dirent; + + error = xfarray_load(rd->dir_entries, array_cur, &dirent); + if (error) + return error; + + error = xfblob_loadname(rd->dir_names, dirent.name_cookie, + &rd->xname, dirent.namelen); + if (error) + return error; + rd->xname.type = dirent.ftype; + + error = xrep_dir_replay_update(rd, &rd->xname, &dirent); + if (error) + return error; + } + + /* Empty out both arrays now that we've added the entries. */ + xfarray_truncate(rd->dir_entries); + xfblob_truncate(rd->dir_names); + return 0; +} + +/* + * Periodically flush stashed directory entries to the temporary dir. This + * is done to reduce the memory requirements of the directory rebuild, since + * directories can contain up to 32GB of directory data. + */ +STATIC int +xrep_dir_flush_stashed( + struct xrep_dir *rd) +{ + int error; + + /* + * Entering this function, the scrub context has a reference to the + * inode being repaired, the temporary file, and a scrub transaction + * that we use during dirent salvaging to avoid livelocking if there + * are cycles in the directory structures. We hold ILOCK_EXCL on both + * the inode being repaired and the temporary file, though they are + * not ijoined to the scrub transaction. + * + * To constrain kernel memory use, we occasionally write salvaged + * dirents from the xfarray and xfblob structures into the temporary + * directory in preparation for exchanging the directory structures at + * the end. Updating the temporary file requires a transaction, so we + * commit the scrub transaction and drop the two ILOCKs so that + * we can allocate whatever transaction we want. + * + * We still hold IOLOCK_EXCL on the inode being repaired, which + * prevents anyone from accessing the damaged directory data while we + * repair it. + */ + error = xrep_trans_commit(rd->sc); + if (error) + return error; + xchk_iunlock(rd->sc, XFS_ILOCK_EXCL); + + /* + * Take the IOLOCK of the temporary file while we modify dirents. This + * isn't strictly required because the temporary file is never revealed + * to userspace, but we follow the same locking rules. We still hold + * sc->ip's IOLOCK. + */ + error = xrep_tempfile_iolock_polled(rd->sc); + if (error) + return error; + + /* Write to the tempdir all the updates that we've stashed. */ + error = xrep_dir_replay_updates(rd); + xrep_tempfile_iounlock(rd->sc); + if (error) + return error; + + /* + * Recreate the salvage transaction and relock the dir we're salvaging. + */ + error = xchk_trans_alloc(rd->sc, 0); + if (error) + return error; + xchk_ilock(rd->sc, XFS_ILOCK_EXCL); + return 0; +} + +/* Decide if we've stashed too much dirent data in memory. */ +static inline bool +xrep_dir_want_flush_stashed( + struct xrep_dir *rd) +{ + unsigned long long bytes; + + bytes = xfarray_bytes(rd->dir_entries) + xfblob_bytes(rd->dir_names); + return bytes > XREP_DIR_MAX_STASH_BYTES; +} + +/* Extract as many directory entries as we can. */ +STATIC int +xrep_dir_recover( + struct xrep_dir *rd) +{ + struct xfs_bmbt_irec got; + struct xfs_scrub *sc = rd->sc; + struct xfs_da_geometry *geo = sc->mp->m_dir_geo; + xfs_fileoff_t offset; + xfs_dablk_t dabno; + __be32 magic_guess; + int nmap; + int error; + + xrep_dir_guess_format(rd, &magic_guess); + + /* Iterate each directory data block in the data fork. */ + for (offset = 0; + offset < geo->leafblk; + offset = got.br_startoff + got.br_blockcount) { + nmap = 1; + error = xfs_bmapi_read(sc->ip, offset, geo->leafblk - offset, + &got, &nmap, 0); + if (error) + return error; + if (nmap != 1) + return -EFSCORRUPTED; + if (!xfs_bmap_is_written_extent(&got)) + continue; + + for (dabno = round_up(got.br_startoff, geo->fsbcount); + dabno < got.br_startoff + got.br_blockcount; + dabno += geo->fsbcount) { + if (xchk_should_terminate(rd->sc, &error)) + return error; + + error = xrep_dir_recover_dirblock(rd, + magic_guess, dabno); + if (error) + return error; + + /* Flush dirents to constrain memory usage. */ + if (xrep_dir_want_flush_stashed(rd)) { + error = xrep_dir_flush_stashed(rd); + if (error) + return error; + } + } + } + + return 0; +} + +/* + * Find all the directory entries for this inode by scraping them out of the + * directory leaf blocks by hand, and flushing them into the temp dir. + */ +STATIC int +xrep_dir_find_entries( + struct xrep_dir *rd) +{ + struct xfs_inode *dp = rd->sc->ip; + int error; + + /* + * Salvage directory entries from the old directory, and write them to + * the temporary directory. + */ + if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { + error = xrep_dir_recover_sf(rd); + } else { + error = xfs_iread_extents(rd->sc->tp, dp, XFS_DATA_FORK); + if (error) + return error; + + error = xrep_dir_recover(rd); + } + if (error) + return error; + + return xrep_dir_flush_stashed(rd); +} + +/* Scan all files in the filesystem for dirents. */ +STATIC int +xrep_dir_salvage_entries( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + int error; + + /* + * Drop the ILOCK on this directory so that we can scan for this + * directory's parent. Figure out who is going to be the parent of + * this directory, then retake the ILOCK so that we can salvage + * directory entries. + */ + xchk_iunlock(sc, XFS_ILOCK_EXCL); + error = xrep_dir_find_parent(rd); + xchk_ilock(sc, XFS_ILOCK_EXCL); + if (error) + return error; + + /* + * Collect directory entries by parsing raw leaf blocks to salvage + * whatever we can. When we're done, free the staging memory before + * exchanging the directories to reduce memory usage. + */ + error = xrep_dir_find_entries(rd); + if (error) + return error; + + /* + * Cancel the repair transaction and drop the ILOCK so that we can + * (later) use the atomic mapping exchange functions to compute the + * correct block reservations and re-lock the inodes. + * + * We still hold IOLOCK_EXCL (aka i_rwsem) which will prevent directory + * modifications, but there's nothing to prevent userspace from reading + * the directory until we're ready for the exchange operation. Reads + * will return -EIO without shutting down the fs, so we're ok with + * that. + */ + error = xrep_trans_commit(sc); + if (error) + return error; + + xchk_iunlock(sc, XFS_ILOCK_EXCL); + return 0; +} + + +/* + * Free all the directory blocks and reset the data fork. The caller must + * join the inode to the transaction. This function returns with the inode + * joined to a clean scrub transaction. + */ +STATIC int +xrep_dir_reset_fork( + struct xrep_dir *rd, + xfs_ino_t parent_ino) +{ + struct xfs_scrub *sc = rd->sc; + struct xfs_ifork *ifp = xfs_ifork_ptr(sc->tempip, XFS_DATA_FORK); + int error; + + /* Unmap all the directory buffers. */ + if (xfs_ifork_has_extents(ifp)) { + error = xrep_reap_ifork(sc, sc->tempip, XFS_DATA_FORK); + if (error) + return error; + } + + trace_xrep_dir_reset_fork(sc->tempip, parent_ino); + + /* Reset the data fork to an empty data fork. */ + xfs_idestroy_fork(ifp); + ifp->if_bytes = 0; + sc->tempip->i_disk_size = 0; + + /* Reinitialize the short form directory. */ + xrep_dir_init_args(rd, sc->tempip, NULL); + return xfs_dir2_sf_create(&rd->args, parent_ino); +} + +/* + * Prepare both inodes' directory forks for exchanging mappings. Promote the + * tempfile from short format to leaf format, and if the file being repaired + * has a short format data fork, turn it into an empty extent list. + */ +STATIC int +xrep_dir_swap_prep( + struct xfs_scrub *sc, + bool temp_local, + bool ip_local) +{ + int error; + + /* + * If the tempfile's directory is in shortform format, convert that to + * a single leaf extent so that we can use the atomic mapping exchange. + */ + if (temp_local) { + struct xfs_da_args args = { + .dp = sc->tempip, + .geo = sc->mp->m_dir_geo, + .whichfork = XFS_DATA_FORK, + .trans = sc->tp, + .total = 1, + .owner = sc->ip->i_ino, + }; + + error = xfs_dir2_sf_to_block(&args); + if (error) + return error; + + /* + * Roll the deferred log items to get us back to a clean + * transaction. + */ + error = xfs_defer_finish(&sc->tp); + if (error) + return error; + } + + /* + * If the file being repaired had a shortform data fork, convert that + * to an empty extent list in preparation for the atomic mapping + * exchange. + */ + if (ip_local) { + struct xfs_ifork *ifp; + + ifp = xfs_ifork_ptr(sc->ip, XFS_DATA_FORK); + xfs_idestroy_fork(ifp); + ifp->if_format = XFS_DINODE_FMT_EXTENTS; + ifp->if_nextents = 0; + ifp->if_bytes = 0; + ifp->if_data = NULL; + ifp->if_height = 0; + + xfs_trans_log_inode(sc->tp, sc->ip, + XFS_ILOG_CORE | XFS_ILOG_DDATA); + } + + return 0; +} + +/* + * Replace the inode number of a directory entry. + */ +static int +xrep_dir_replace( + struct xrep_dir *rd, + struct xfs_inode *dp, + const struct xfs_name *name, + xfs_ino_t inum, + xfs_extlen_t total) +{ + struct xfs_scrub *sc = rd->sc; + bool is_block, is_leaf; + int error; + + ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); + + error = xfs_dir_ino_validate(sc->mp, inum); + if (error) + return error; + + xrep_dir_init_args(rd, dp, name); + rd->args.inumber = inum; + rd->args.total = total; + + if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return xfs_dir2_sf_replace(&rd->args); + + error = xfs_dir2_isblock(&rd->args, &is_block); + if (error) + return error; + if (is_block) + return xfs_dir2_block_replace(&rd->args); + + error = xfs_dir2_isleaf(&rd->args, &is_leaf); + if (error) + return error; + if (is_leaf) + return xfs_dir2_leaf_replace(&rd->args); + + return xfs_dir2_node_replace(&rd->args); +} + +/* + * Reset the link count of this directory and adjust the unlinked list pointers + * as needed. + */ +STATIC int +xrep_dir_set_nlink( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + struct xfs_inode *dp = sc->ip; + struct xfs_perag *pag; + unsigned int new_nlink = rd->subdirs + 2; + int error; + + /* + * The directory is not on the incore unlinked list, which means that + * it needs to be reachable via the directory tree. Update the nlink + * with our observed link count. + * + * XXX: A subsequent patch will handle parentless directories by moving + * them to the lost and found instead of aborting the repair. + */ + if (!xfs_inode_on_unlinked_list(dp)) + goto reset_nlink; + + /* + * The directory is on the unlinked list and we did not find any + * dirents. Set the link count to zero and let the directory + * inactivate when the last reference drops. + */ + if (rd->dirents == 0) { + new_nlink = 0; + goto reset_nlink; + } + + /* + * The directory is on the unlinked list and we found dirents. This + * directory needs to be reachable via the directory tree. Remove the + * dir from the unlinked list and update nlink with the observed link + * count. + */ + pag = xfs_perag_get(sc->mp, XFS_INO_TO_AGNO(sc->mp, dp->i_ino)); + if (!pag) { + ASSERT(0); + return -EFSCORRUPTED; + } + + error = xfs_iunlink_remove(sc->tp, pag, dp); + xfs_perag_put(pag); + if (error) + return error; + +reset_nlink: + if (VFS_I(dp)->i_nlink != new_nlink) + set_nlink(VFS_I(dp), new_nlink); + return 0; +} + +/* Exchange the temporary directory's data fork with the one being repaired. */ +STATIC int +xrep_dir_swap( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + bool ip_local, temp_local; + int error = 0; + + /* + * If we found enough subdirs to overflow this directory's link count, + * bail out to userspace before we modify anything. + */ + if (rd->subdirs + 2 > XFS_MAXLINK) + return -EFSCORRUPTED; + + /* + * Reset the temporary directory's '..' entry to point to the parent + * that we found. The temporary directory was created with the root + * directory as the parent, so we can skip this if repairing a + * subdirectory of the root. + * + * It's also possible that this replacement could also expand a sf + * tempdir into block format. + */ + if (rd->parent_ino != sc->mp->m_rootip->i_ino) { + error = xrep_dir_replace(rd, rd->sc->tempip, &xfs_name_dotdot, + rd->parent_ino, rd->tx.req.resblks); + if (error) + return error; + } + + /* + * Changing the dot and dotdot entries could have changed the shape of + * the directory, so we recompute these. + */ + ip_local = sc->ip->i_df.if_format == XFS_DINODE_FMT_LOCAL; + temp_local = sc->tempip->i_df.if_format == XFS_DINODE_FMT_LOCAL; + + /* + * If the both files have a local format data fork and the rebuilt + * directory data would fit in the repaired file's data fork, copy + * the contents from the tempfile and update the directory link count. + * We're done now. + */ + if (ip_local && temp_local && + sc->tempip->i_disk_size <= xfs_inode_data_fork_size(sc->ip)) { + xrep_tempfile_copyout_local(sc, XFS_DATA_FORK); + return xrep_dir_set_nlink(rd); + } + + /* + * Clean the transaction before we start working on exchanging + * directory contents. + */ + error = xrep_tempfile_roll_trans(rd->sc); + if (error) + return error; + + /* Otherwise, make sure both data forks are in block-mapping mode. */ + error = xrep_dir_swap_prep(sc, temp_local, ip_local); + if (error) + return error; + + /* + * Set nlink of the directory in the same transaction sequence that + * (atomically) commits the new directory data. + */ + error = xrep_dir_set_nlink(rd); + if (error) + return error; + + return xrep_tempexch_contents(sc, &rd->tx); +} + +/* + * Exchange the new directory contents (which we created in the tempfile) with + * the directory being repaired. + */ +STATIC int +xrep_dir_rebuild_tree( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + int error; + + trace_xrep_dir_rebuild_tree(sc->ip, rd->parent_ino); + + /* + * Take the IOLOCK on the temporary file so that we can run dir + * operations with the same locks held as we would for a normal file. + * We still hold sc->ip's IOLOCK. + */ + error = xrep_tempfile_iolock_polled(rd->sc); + if (error) + return error; + + /* Allocate transaction and ILOCK the scrub file and the temp file. */ + error = xrep_tempexch_trans_alloc(sc, XFS_DATA_FORK, &rd->tx); + if (error) + return error; + + /* + * Exchange the tempdir's data fork with the file being repaired. This + * recreates the transaction and re-takes the ILOCK in the scrub + * context. + */ + error = xrep_dir_swap(rd); + if (error) + return error; + + /* + * Release the old directory blocks and reset the data fork of the temp + * directory to an empty shortform directory because inactivation does + * nothing for directories. + */ + error = xrep_dir_reset_fork(rd, sc->mp->m_rootip->i_ino); + if (error) + return error; + + /* + * Roll to get a transaction without any inodes joined to it. Then we + * can drop the tempfile's ILOCK and IOLOCK before doing more work on + * the scrub target directory. + */ + error = xfs_trans_roll(&sc->tp); + if (error) + return error; + + xrep_tempfile_iunlock(sc); + xrep_tempfile_iounlock(sc); + return 0; +} + +/* Set up the filesystem scan so we can regenerate directory entries. */ +STATIC int +xrep_dir_setup_scan( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + char *descr; + int error; + + rd->parent_ino = NULLFSINO; + + /* Set up some staging memory for salvaging dirents. */ + descr = xchk_xfile_ino_descr(sc, "directory entries"); + error = xfarray_create(descr, 0, sizeof(struct xrep_dirent), + &rd->dir_entries); + kfree(descr); + if (error) + return error; + + descr = xchk_xfile_ino_descr(sc, "directory entry names"); + error = xfblob_create(descr, &rd->dir_names); + kfree(descr); + if (error) + goto out_xfarray; + + return 0; + +out_xfarray: + xfarray_destroy(rd->dir_entries); + rd->dir_entries = NULL; + return error; +} + +/* + * Repair the directory metadata. + * + * XXX: Directory entry buffers can be multiple fsblocks in size. The buffer + * cache in XFS can't handle aliased multiblock buffers, so this might + * misbehave if the directory blocks are crosslinked with other filesystem + * metadata. + * + * XXX: Is it necessary to check the dcache for this directory to make sure + * that we always recreate every cached entry? + */ +int +xrep_directory( + struct xfs_scrub *sc) +{ + struct xrep_dir *rd = sc->buf; + int error; + + /* The rmapbt is required to reap the old data fork. */ + if (!xfs_has_rmapbt(sc->mp)) + return -EOPNOTSUPP; + + error = xrep_dir_setup_scan(rd); + if (error) + return error; + + error = xrep_dir_salvage_entries(rd); + if (error) + goto out_teardown; + + /* Last chance to abort before we start committing fixes. */ + if (xchk_should_terminate(sc, &error)) + goto out_teardown; + + error = xrep_dir_rebuild_tree(rd); + if (error) + goto out_teardown; + +out_teardown: + xrep_dir_teardown(sc); + return error; +} diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c index c743772a523e..0dde5df2f8d3 100644 --- a/fs/xfs/scrub/inode_repair.c +++ b/fs/xfs/scrub/inode_repair.c @@ -46,6 +46,7 @@ #include "scrub/repair.h" #include "scrub/iscan.h" #include "scrub/readdir.h" +#include "scrub/tempfile.h" /* * Inode Record Repair @@ -340,6 +341,10 @@ xrep_dinode_findmode_walk_directory( unsigned int lock_mode; int error = 0; + /* Ignore temporary repair directories. */ + if (xrep_is_tempfile(dp)) + return 0; + /* * Scan the directory to see if there it contains an entry pointing to * the directory that we are repairing. diff --git a/fs/xfs/scrub/nlinks.c b/fs/xfs/scrub/nlinks.c index 8a7d9557897c..8b9aa73093d6 100644 --- a/fs/xfs/scrub/nlinks.c +++ b/fs/xfs/scrub/nlinks.c @@ -27,6 +27,7 @@ #include "scrub/nlinks.h" #include "scrub/trace.h" #include "scrub/readdir.h" +#include "scrub/tempfile.h" /* * Live Inode Link Count Checking @@ -152,6 +153,13 @@ xchk_nlinks_live_update( xnc = container_of(nb, struct xchk_nlink_ctrs, dhook.dirent_hook.nb); + /* + * Ignore temporary directories being used to stage dir repairs, since + * we don't bump the link counts of the children. + */ + if (xrep_is_tempfile(p->dp)) + return NOTIFY_DONE; + trace_xchk_nlinks_live_update(xnc->sc->mp, p->dp, action, p->ip->i_ino, p->delta, p->name->name, p->name->len); @@ -303,6 +311,13 @@ xchk_nlinks_collect_dir( unsigned int lock_mode; int error = 0; + /* + * Ignore temporary directories being used to stage dir repairs, since + * we don't bump the link counts of the children. + */ + if (xrep_is_tempfile(dp)) + return 0; + /* Prevent anyone from changing this directory while we walk it. */ xfs_ilock(dp, XFS_IOLOCK_SHARED); lock_mode = xfs_ilock_data_map_shared(dp); @@ -537,6 +552,14 @@ xchk_nlinks_compare_inode( unsigned int actual_nlink; int error; + /* + * Ignore temporary files being used to stage repairs, since we assume + * they're correct for non-directories, and the directory repair code + * doesn't bump the link counts for the children. + */ + if (xrep_is_tempfile(ip)) + return 0; + xfs_ilock(ip, XFS_ILOCK_SHARED); mutex_lock(&xnc->lock); diff --git a/fs/xfs/scrub/nlinks_repair.c b/fs/xfs/scrub/nlinks_repair.c index 58cacb8e94c1..23eb08c4b5ad 100644 --- a/fs/xfs/scrub/nlinks_repair.c +++ b/fs/xfs/scrub/nlinks_repair.c @@ -26,6 +26,7 @@ #include "scrub/iscan.h" #include "scrub/nlinks.h" #include "scrub/trace.h" +#include "scrub/tempfile.h" /* * Live Inode Link Count Repair @@ -68,6 +69,14 @@ xrep_nlinks_repair_inode( bool dirty = false; int error; + /* + * Ignore temporary files being used to stage repairs, since we assume + * they're correct for non-directories, and the directory repair code + * doesn't bump the link counts for the children. + */ + if (xrep_is_tempfile(ip)) + return 0; + xchk_ilock(sc, XFS_IOLOCK_EXCL); error = xfs_trans_alloc(mp, &M_RES(mp)->tr_link, 0, 0, 0, &sc->tp); diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index 5da10ed1fe8c..050a8e8914f6 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -17,6 +17,7 @@ #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/readdir.h" +#include "scrub/tempfile.h" /* Set us up to scrub parents. */ int @@ -143,7 +144,8 @@ xchk_parent_validate( } if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error)) return error; - if (dp == sc->ip || dp == sc->tempip || !S_ISDIR(VFS_I(dp)->i_mode)) { + if (dp == sc->ip || xrep_is_tempfile(dp) || + !S_ISDIR(VFS_I(dp)->i_mode)) { xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); goto out_rele; } diff --git a/fs/xfs/scrub/readdir.c b/fs/xfs/scrub/readdir.c index e94080469315..028690761c62 100644 --- a/fs/xfs/scrub/readdir.c +++ b/fs/xfs/scrub/readdir.c @@ -333,6 +333,13 @@ xchk_dir_lookup( if (xfs_is_shutdown(dp->i_mount)) return -EIO; + /* + * A temporary directory's block headers are written with the owner + * set to sc->ip, so we must switch the owner here for the lookup. + */ + if (dp == sc->tempip) + args.owner = sc->ip->i_ino; + ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); xfs_assert_ilocked(dp, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL); diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index 04aec0e9e4c3..369f0430e4ba 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -35,6 +35,7 @@ #include "xfs_da_format.h" #include "xfs_da_btree.h" #include "xfs_attr.h" +#include "xfs_dir2.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index 9cbfd8da5620..4e25aa95753a 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -91,6 +91,7 @@ int xrep_metadata_inode_forks(struct xfs_scrub *sc); int xrep_setup_ag_rmapbt(struct xfs_scrub *sc); int xrep_setup_ag_refcountbt(struct xfs_scrub *sc); int xrep_setup_xattr(struct xfs_scrub *sc); +int xrep_setup_directory(struct xfs_scrub *sc); /* Repair setup functions */ int xrep_setup_ag_allocbt(struct xfs_scrub *sc); @@ -125,6 +126,7 @@ int xrep_bmap_cow(struct xfs_scrub *sc); int xrep_nlinks(struct xfs_scrub *sc); int xrep_fscounters(struct xfs_scrub *sc); int xrep_xattr(struct xfs_scrub *sc); +int xrep_directory(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xrep_rtbitmap(struct xfs_scrub *sc); @@ -195,6 +197,7 @@ xrep_setup_nothing( #define xrep_setup_ag_rmapbt xrep_setup_nothing #define xrep_setup_ag_refcountbt xrep_setup_nothing #define xrep_setup_xattr xrep_setup_nothing +#define xrep_setup_directory xrep_setup_nothing #define xrep_setup_inode(sc, imap) ((void)0) @@ -221,6 +224,7 @@ xrep_setup_nothing( #define xrep_fscounters xrep_notsupported #define xrep_rtsummary xrep_notsupported #define xrep_xattr xrep_notsupported +#define xrep_directory xrep_notsupported #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 547189a14b6b..8e9e2bf121c2 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -325,7 +325,7 @@ static const struct xchk_meta_ops meta_scrub_ops[] = { .type = ST_INODE, .setup = xchk_setup_directory, .scrub = xchk_directory, - .repair = xrep_notsupported, + .repair = xrep_directory, }, [XFS_SCRUB_TYPE_XATTR] = { /* extended attributes */ .type = ST_INODE, diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index 0b3060be938f..4ca86a6a5be1 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -841,3 +841,16 @@ xrep_tempfile_copyout_local( ilog_flags |= xfs_ilog_fdata(whichfork); xfs_trans_log_inode(sc->tp, sc->ip, ilog_flags); } + +/* Decide if a given XFS inode is a temporary file for a repair. */ +bool +xrep_is_tempfile( + const struct xfs_inode *ip) +{ + const struct inode *inode = &ip->i_vnode; + + if (IS_PRIVATE(inode) && !(inode->i_opflags & IOP_XATTR)) + return true; + + return false; +} diff --git a/fs/xfs/scrub/tempfile.h b/fs/xfs/scrub/tempfile.h index d57e4f145a7c..e51399f595fe 100644 --- a/fs/xfs/scrub/tempfile.h +++ b/fs/xfs/scrub/tempfile.h @@ -35,11 +35,13 @@ int xrep_tempfile_set_isize(struct xfs_scrub *sc, unsigned long long isize); int xrep_tempfile_roll_trans(struct xfs_scrub *sc); void xrep_tempfile_copyout_local(struct xfs_scrub *sc, int whichfork); +bool xrep_is_tempfile(const struct xfs_inode *ip); #else static inline void xrep_tempfile_iolock_both(struct xfs_scrub *sc) { xchk_ilock(sc, XFS_IOLOCK_EXCL); } +# define xrep_is_tempfile(ip) (false) # define xrep_tempfile_rele(sc) #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index ffaff7722bf2..d6d9e8d6109c 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2500,6 +2500,118 @@ DEFINE_EVENT(xrep_xattr_class, name, \ DEFINE_XREP_XATTR_EVENT(xrep_xattr_rebuild_tree); DEFINE_XREP_XATTR_EVENT(xrep_xattr_reset_fork); +TRACE_EVENT(xrep_dir_recover_dirblock, + TP_PROTO(struct xfs_inode *dp, xfs_dablk_t dabno, uint32_t magic, + uint32_t magic_guess), + TP_ARGS(dp, dabno, magic, magic_guess), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, dir_ino) + __field(xfs_dablk_t, dabno) + __field(uint32_t, magic) + __field(uint32_t, magic_guess) + ), + TP_fast_assign( + __entry->dev = dp->i_mount->m_super->s_dev; + __entry->dir_ino = dp->i_ino; + __entry->dabno = dabno; + __entry->magic = magic; + __entry->magic_guess = magic_guess; + ), + TP_printk("dev %d:%d dir 0x%llx dablk 0x%x magic 0x%x magic_guess 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->dir_ino, + __entry->dabno, + __entry->magic, + __entry->magic_guess) +); + +DECLARE_EVENT_CLASS(xrep_dir_class, + TP_PROTO(struct xfs_inode *dp, xfs_ino_t parent_ino), + TP_ARGS(dp, parent_ino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, dir_ino) + __field(xfs_ino_t, parent_ino) + ), + TP_fast_assign( + __entry->dev = dp->i_mount->m_super->s_dev; + __entry->dir_ino = dp->i_ino; + __entry->parent_ino = parent_ino; + ), + TP_printk("dev %d:%d dir 0x%llx parent 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->dir_ino, + __entry->parent_ino) +) +#define DEFINE_XREP_DIR_EVENT(name) \ +DEFINE_EVENT(xrep_dir_class, name, \ + TP_PROTO(struct xfs_inode *dp, xfs_ino_t parent_ino), \ + TP_ARGS(dp, parent_ino)) +DEFINE_XREP_DIR_EVENT(xrep_dir_rebuild_tree); +DEFINE_XREP_DIR_EVENT(xrep_dir_reset_fork); + +DECLARE_EVENT_CLASS(xrep_dirent_class, + TP_PROTO(struct xfs_inode *dp, const struct xfs_name *name, + xfs_ino_t ino), + TP_ARGS(dp, name, ino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, dir_ino) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + __field(xfs_ino_t, ino) + __field(uint8_t, ftype) + ), + TP_fast_assign( + __entry->dev = dp->i_mount->m_super->s_dev; + __entry->dir_ino = dp->i_ino; + __entry->namelen = name->len; + memcpy(__get_str(name), name->name, name->len); + __entry->ino = ino; + __entry->ftype = name->type; + ), + TP_printk("dev %d:%d dir 0x%llx ftype %s name '%.*s' ino 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->dir_ino, + __print_symbolic(__entry->ftype, XFS_DIR3_FTYPE_STR), + __entry->namelen, + __get_str(name), + __entry->ino) +) +#define DEFINE_XREP_DIRENT_EVENT(name) \ +DEFINE_EVENT(xrep_dirent_class, name, \ + TP_PROTO(struct xfs_inode *dp, const struct xfs_name *name, \ + xfs_ino_t ino), \ + TP_ARGS(dp, name, ino)) +DEFINE_XREP_DIRENT_EVENT(xrep_dir_salvage_entry); +DEFINE_XREP_DIRENT_EVENT(xrep_dir_stash_createname); +DEFINE_XREP_DIRENT_EVENT(xrep_dir_replay_createname); + +DECLARE_EVENT_CLASS(xrep_parent_salvage_class, + TP_PROTO(struct xfs_inode *dp, xfs_ino_t ino), + TP_ARGS(dp, ino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, dir_ino) + __field(xfs_ino_t, ino) + ), + TP_fast_assign( + __entry->dev = dp->i_mount->m_super->s_dev; + __entry->dir_ino = dp->i_ino; + __entry->ino = ino; + ), + TP_printk("dev %d:%d dir 0x%llx parent 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->dir_ino, + __entry->ino) +) +#define DEFINE_XREP_PARENT_SALVAGE_EVENT(name) \ +DEFINE_EVENT(xrep_parent_salvage_class, name, \ + TP_PROTO(struct xfs_inode *dp, xfs_ino_t ino), \ + TP_ARGS(dp, ino)) +DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_dir_salvaged_parent); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ diff --git a/fs/xfs/scrub/xfblob.h b/fs/xfs/scrub/xfblob.h index 78a67a06408f..ae78322613ca 100644 --- a/fs/xfs/scrub/xfblob.h +++ b/fs/xfs/scrub/xfblob.h @@ -23,4 +23,28 @@ int xfblob_free(struct xfblob *blob, xfblob_cookie cookie); unsigned long long xfblob_bytes(struct xfblob *blob); void xfblob_truncate(struct xfblob *blob); +static inline int +xfblob_storename( + struct xfblob *blob, + xfblob_cookie *cookie, + const struct xfs_name *xname) +{ + return xfblob_store(blob, cookie, xname->name, xname->len); +} + +static inline int +xfblob_loadname( + struct xfblob *blob, + xfblob_cookie cookie, + struct xfs_name *xname, + uint32_t size) +{ + int ret = xfblob_load(blob, cookie, (void *)xname->name, size); + if (ret) + return ret; + + xname->len = size; + return 0; +} + #endif /* __XFS_SCRUB_XFBLOB_H__ */ -- cgit v1.2.3-59-g8ed1b From a07b45576264e77ea1a781b552873e76b8b0dacc Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:52 -0700 Subject: xfs: scan the filesystem to repair a directory dotdot entry Teach the online directory repair code to scan the filesystem so that we can set the dotdot entry when we're rebuilding a directory. This involves dropping ILOCK on the directory that we're repairing, which means that the VFS can sneak in and tell us to update dotdot at any time. Deal with these races by using a dirent hook to absorb dotdot updates, and be careful not to check the scan results until after we've retaken the ILOCK. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/dir_repair.c | 70 +++++--- fs/xfs/scrub/findparent.c | 412 ++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/findparent.h | 49 ++++++ fs/xfs/scrub/iscan.c | 18 +- fs/xfs/scrub/iscan.h | 1 + fs/xfs/scrub/trace.h | 1 + 7 files changed, 528 insertions(+), 24 deletions(-) create mode 100644 fs/xfs/scrub/findparent.c create mode 100644 fs/xfs/scrub/findparent.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 5c9449e14f74..3c754777ec28 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -199,6 +199,7 @@ xfs-y += $(addprefix scrub/, \ bmap_repair.o \ cow_repair.o \ dir_repair.o \ + findparent.o \ fscounters_repair.o \ ialloc_repair.o \ inode_repair.o \ diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 48aa80d8c7dc..b17de79207db 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -38,8 +38,10 @@ #include "scrub/xfile.h" #include "scrub/xfarray.h" #include "scrub/xfblob.h" +#include "scrub/iscan.h" #include "scrub/readdir.h" #include "scrub/reap.h" +#include "scrub/findparent.h" /* * Directory Repair @@ -108,10 +110,10 @@ struct xrep_dir { struct xfs_da_args args; /* - * This is the parent that we're going to set on the reconstructed - * directory. + * Information used to scan the filesystem to find the inumber of the + * dotdot entry for this directory. */ - xfs_ino_t parent_ino; + struct xrep_parent_scan_info pscan; /* How many subdirectories did we find? */ uint64_t subdirs; @@ -131,6 +133,7 @@ xrep_dir_teardown( { struct xrep_dir *rd = sc->buf; + xrep_findparent_scan_teardown(&rd->pscan); xfblob_destroy(rd->dir_names); xfarray_destroy(rd->dir_entries); } @@ -143,6 +146,8 @@ xrep_setup_directory( struct xrep_dir *rd; int error; + xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); + error = xrep_tempfile_create(sc, S_IFDIR); if (error) return error; @@ -179,8 +184,8 @@ xrep_dir_self_parent( } /* - * Look up the dotdot entry. Returns NULLFSINO if we don't know what to do. - * The next patch will check this more carefully. + * Look up the dotdot entry and confirm that it's really the parent. + * Returns NULLFSINO if we don't know what to do. */ static inline xfs_ino_t xrep_dir_lookup_parent( @@ -196,37 +201,39 @@ xrep_dir_lookup_parent( if (!xfs_verify_dir_ino(sc->mp, ino)) return NULLFSINO; + error = xrep_findparent_confirm(sc, &ino); + if (error) + return NULLFSINO; + return ino; } -/* - * Try to find the parent of the directory being repaired. - * - * NOTE: This function will someday be augmented by the directory parent repair - * code, which will know how to check the parent and scan the filesystem if - * we cannot find anything. Inode scans will have to be done before we start - * salvaging directory entries, so we do this now. - */ +/* Try to find the parent of the directory being repaired. */ STATIC int xrep_dir_find_parent( struct xrep_dir *rd) { xfs_ino_t ino; - ino = xrep_dir_self_parent(rd); + ino = xrep_findparent_self_reference(rd->sc); if (ino != NULLFSINO) { - rd->parent_ino = ino; + xrep_findparent_scan_finish_early(&rd->pscan, ino); return 0; } ino = xrep_dir_lookup_parent(rd); if (ino != NULLFSINO) { - rd->parent_ino = ino; + xrep_findparent_scan_finish_early(&rd->pscan, ino); return 0; } - /* NOTE: A future patch will deal with moving orphans. */ - return -EFSCORRUPTED; + /* + * A full filesystem scan is the last resort. On a busy filesystem, + * the scan can fail with -EBUSY if we cannot grab IOLOCKs. That means + * that we don't know what who the parent is, so we should return to + * userspace. + */ + return xrep_findparent_scan(&rd->pscan); } /* @@ -931,6 +938,10 @@ xrep_dir_salvage_entries( * the directory until we're ready for the exchange operation. Reads * will return -EIO without shutting down the fs, so we're ok with * that. + * + * The VFS can change dotdot on us, but the findparent scan will keep + * our incore parent inode up to date. See the note on locking issues + * for more details. */ error = xrep_trans_commit(sc); if (error) @@ -1154,6 +1165,14 @@ xrep_dir_swap( if (rd->subdirs + 2 > XFS_MAXLINK) return -EFSCORRUPTED; + /* + * If we never found the parent for this directory, we can't fix this + * directory. + */ + ASSERT(sc->ilock_flags & XFS_ILOCK_EXCL); + if (rd->pscan.parent_ino == NULLFSINO) + return -EFSCORRUPTED; + /* * Reset the temporary directory's '..' entry to point to the parent * that we found. The temporary directory was created with the root @@ -1163,9 +1182,9 @@ xrep_dir_swap( * It's also possible that this replacement could also expand a sf * tempdir into block format. */ - if (rd->parent_ino != sc->mp->m_rootip->i_ino) { + if (rd->pscan.parent_ino != sc->mp->m_rootip->i_ino) { error = xrep_dir_replace(rd, rd->sc->tempip, &xfs_name_dotdot, - rd->parent_ino, rd->tx.req.resblks); + rd->pscan.parent_ino, rd->tx.req.resblks); if (error) return error; } @@ -1224,7 +1243,7 @@ xrep_dir_rebuild_tree( struct xfs_scrub *sc = rd->sc; int error; - trace_xrep_dir_rebuild_tree(sc->ip, rd->parent_ino); + trace_xrep_dir_rebuild_tree(sc->ip, rd->pscan.parent_ino); /* * Take the IOLOCK on the temporary file so that we can run dir @@ -1281,8 +1300,6 @@ xrep_dir_setup_scan( char *descr; int error; - rd->parent_ino = NULLFSINO; - /* Set up some staging memory for salvaging dirents. */ descr = xchk_xfile_ino_descr(sc, "directory entries"); error = xfarray_create(descr, 0, sizeof(struct xrep_dirent), @@ -1297,8 +1314,15 @@ xrep_dir_setup_scan( if (error) goto out_xfarray; + error = xrep_findparent_scan_start(sc, &rd->pscan); + if (error) + goto out_xfblob; + return 0; +out_xfblob: + xfblob_destroy(rd->dir_names); + rd->dir_names = NULL; out_xfarray: xfarray_destroy(rd->dir_entries); rd->dir_entries = NULL; diff --git a/fs/xfs/scrub/findparent.c b/fs/xfs/scrub/findparent.c new file mode 100644 index 000000000000..7b3ec8d7d6cc --- /dev/null +++ b/fs/xfs/scrub/findparent.c @@ -0,0 +1,412 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_bit.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_inode.h" +#include "xfs_icache.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_dir2.h" +#include "xfs_bmap_btree.h" +#include "xfs_dir2_priv.h" +#include "xfs_trans_space.h" +#include "xfs_health.h" +#include "xfs_exchmaps.h" +#include "scrub/xfs_scrub.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/iscan.h" +#include "scrub/findparent.h" +#include "scrub/readdir.h" +#include "scrub/tempfile.h" + +/* + * Finding the Parent of a Directory + * ================================= + * + * Directories have parent pointers, in the sense that each directory contains + * a dotdot entry that points to the single allowed parent. The brute force + * way to find the parent of a given directory is to scan every directory in + * the filesystem looking for a child dirent that references this directory. + * + * This module wraps the process of scanning the directory tree. It requires + * that @sc->ip is the directory whose parent we want to find, and that the + * caller hold only the IOLOCK on that directory. The scan itself needs to + * take the ILOCK of each directory visited. + * + * Because we cannot hold @sc->ip's ILOCK during a scan of the whole fs, it is + * necessary to use dirent hook to update the parent scan results. Callers + * must not read the scan results without re-taking @sc->ip's ILOCK. + * + * There are a few shortcuts that we can take to avoid scanning the entire + * filesystem, such as noticing directory tree roots. + */ + +struct xrep_findparent_info { + /* The directory currently being scanned. */ + struct xfs_inode *dp; + + /* + * Scrub context. We're looking for a @dp containing a directory + * entry pointing to sc->ip->i_ino. + */ + struct xfs_scrub *sc; + + /* Optional scan information for a xrep_findparent_scan call. */ + struct xrep_parent_scan_info *parent_scan; + + /* + * Parent that we've found for sc->ip. If we're scanning the entire + * directory tree, we need this to ensure that we only find /one/ + * parent directory. + */ + xfs_ino_t found_parent; + + /* + * This is set to true if @found_parent was not observed directly from + * the directory scan but by noticing a change in dotdot entries after + * cycling the sc->ip IOLOCK. + */ + bool parent_tentative; +}; + +/* + * If this directory entry points to the scrub target inode, then the directory + * we're scanning is the parent of the scrub target inode. + */ +STATIC int +xrep_findparent_dirent( + struct xfs_scrub *sc, + struct xfs_inode *dp, + xfs_dir2_dataptr_t dapos, + const struct xfs_name *name, + xfs_ino_t ino, + void *priv) +{ + struct xrep_findparent_info *fpi = priv; + int error = 0; + + if (xchk_should_terminate(fpi->sc, &error)) + return error; + + if (ino != fpi->sc->ip->i_ino) + return 0; + + /* Ignore garbage directory entry names. */ + if (name->len == 0 || !xfs_dir2_namecheck(name->name, name->len)) + return -EFSCORRUPTED; + + /* + * Ignore dotdot and dot entries -- we're looking for parent -> child + * links only. + */ + if (name->name[0] == '.' && (name->len == 1 || + (name->len == 2 && name->name[1] == '.'))) + return 0; + + /* Uhoh, more than one parent for a dir? */ + if (fpi->found_parent != NULLFSINO && + !(fpi->parent_tentative && fpi->found_parent == fpi->dp->i_ino)) { + trace_xrep_findparent_dirent(fpi->sc->ip, 0); + return -EFSCORRUPTED; + } + + /* We found a potential parent; remember this. */ + trace_xrep_findparent_dirent(fpi->sc->ip, fpi->dp->i_ino); + fpi->found_parent = fpi->dp->i_ino; + fpi->parent_tentative = false; + + if (fpi->parent_scan) + xrep_findparent_scan_found(fpi->parent_scan, fpi->dp->i_ino); + + return 0; +} + +/* + * If this is a directory, walk the dirents looking for any that point to the + * scrub target inode. + */ +STATIC int +xrep_findparent_walk_directory( + struct xrep_findparent_info *fpi) +{ + struct xfs_scrub *sc = fpi->sc; + struct xfs_inode *dp = fpi->dp; + unsigned int lock_mode; + int error = 0; + + /* + * The inode being scanned cannot be its own parent, nor can any + * temporary directory we created to stage this repair. + */ + if (dp == sc->ip || dp == sc->tempip) + return 0; + + /* + * Similarly, temporary files created to stage a repair cannot be the + * parent of this inode. + */ + if (xrep_is_tempfile(dp)) + return 0; + + /* + * Scan the directory to see if there it contains an entry pointing to + * the directory that we are repairing. + */ + lock_mode = xfs_ilock_data_map_shared(dp); + + /* + * If this directory is known to be sick, we cannot scan it reliably + * and must abort. + */ + if (xfs_inode_has_sickness(dp, XFS_SICK_INO_CORE | + XFS_SICK_INO_BMBTD | + XFS_SICK_INO_DIR)) { + error = -EFSCORRUPTED; + goto out_unlock; + } + + /* + * We cannot complete our parent pointer scan if a directory looks as + * though it has been zapped by the inode record repair code. + */ + if (xchk_dir_looks_zapped(dp)) { + error = -EBUSY; + goto out_unlock; + } + + error = xchk_dir_walk(sc, dp, xrep_findparent_dirent, fpi); + if (error) + goto out_unlock; + +out_unlock: + xfs_iunlock(dp, lock_mode); + return error; +} + +/* + * Update this directory's dotdot pointer based on ongoing dirent updates. + */ +STATIC int +xrep_findparent_live_update( + struct notifier_block *nb, + unsigned long action, + void *data) +{ + struct xfs_dir_update_params *p = data; + struct xrep_parent_scan_info *pscan; + struct xfs_scrub *sc; + + pscan = container_of(nb, struct xrep_parent_scan_info, + dhook.dirent_hook.nb); + sc = pscan->sc; + + /* + * If @p->ip is the subdirectory that we're interested in and we've + * already scanned @p->dp, update the dotdot target inumber to the + * parent inode. + */ + if (p->ip->i_ino == sc->ip->i_ino && + xchk_iscan_want_live_update(&pscan->iscan, p->dp->i_ino)) { + if (p->delta > 0) { + xrep_findparent_scan_found(pscan, p->dp->i_ino); + } else { + xrep_findparent_scan_found(pscan, NULLFSINO); + } + } + + return NOTIFY_DONE; +} + +/* + * Set up a scan to find the parent of a directory. The provided dirent hook + * will be called when there is a dotdot update for the inode being repaired. + */ +int +xrep_findparent_scan_start( + struct xfs_scrub *sc, + struct xrep_parent_scan_info *pscan) +{ + int error; + + if (!(sc->flags & XCHK_FSGATES_DIRENTS)) { + ASSERT(sc->flags & XCHK_FSGATES_DIRENTS); + return -EINVAL; + } + + pscan->sc = sc; + pscan->parent_ino = NULLFSINO; + + mutex_init(&pscan->lock); + + xchk_iscan_start(sc, 30000, 100, &pscan->iscan); + + /* + * Hook into the dirent update code. The hook only operates on inodes + * that were already scanned, and the scanner thread takes each inode's + * ILOCK, which means that any in-progress inode updates will finish + * before we can scan the inode. + */ + xfs_dir_hook_setup(&pscan->dhook, xrep_findparent_live_update); + error = xfs_dir_hook_add(sc->mp, &pscan->dhook); + if (error) + goto out_iscan; + + return 0; +out_iscan: + xchk_iscan_teardown(&pscan->iscan); + mutex_destroy(&pscan->lock); + return error; +} + +/* + * Scan the entire filesystem looking for a parent inode for the inode being + * scrubbed. @sc->ip must not be the root of a directory tree. Callers must + * not hold a dirty transaction or any lock that would interfere with taking + * an ILOCK. + * + * Returns 0 with @pscan->parent_ino set to the parent that we found. + * Returns 0 with @pscan->parent_ino set to NULLFSINO if we found no parents. + * Returns the usual negative errno if something else happened. + */ +int +xrep_findparent_scan( + struct xrep_parent_scan_info *pscan) +{ + struct xrep_findparent_info fpi = { + .sc = pscan->sc, + .found_parent = NULLFSINO, + .parent_scan = pscan, + }; + struct xfs_scrub *sc = pscan->sc; + int ret; + + ASSERT(S_ISDIR(VFS_IC(sc->ip)->i_mode)); + + while ((ret = xchk_iscan_iter(&pscan->iscan, &fpi.dp)) == 1) { + if (S_ISDIR(VFS_I(fpi.dp)->i_mode)) + ret = xrep_findparent_walk_directory(&fpi); + else + ret = 0; + xchk_iscan_mark_visited(&pscan->iscan, fpi.dp); + xchk_irele(sc, fpi.dp); + if (ret) + break; + + if (xchk_should_terminate(sc, &ret)) + break; + } + xchk_iscan_iter_finish(&pscan->iscan); + + return ret; +} + +/* Tear down a parent scan. */ +void +xrep_findparent_scan_teardown( + struct xrep_parent_scan_info *pscan) +{ + xfs_dir_hook_del(pscan->sc->mp, &pscan->dhook); + xchk_iscan_teardown(&pscan->iscan); + mutex_destroy(&pscan->lock); +} + +/* Finish a parent scan early. */ +void +xrep_findparent_scan_finish_early( + struct xrep_parent_scan_info *pscan, + xfs_ino_t ino) +{ + xrep_findparent_scan_found(pscan, ino); + xchk_iscan_finish_early(&pscan->iscan); +} + +/* + * Confirm that the directory @parent_ino actually contains a directory entry + * pointing to the child @sc->ip->ino. This function returns one of several + * ways: + * + * Returns 0 with @parent_ino unchanged if the parent was confirmed. + * Returns 0 with @parent_ino set to NULLFSINO if the parent was not valid. + * Returns the usual negative errno if something else happened. + */ +int +xrep_findparent_confirm( + struct xfs_scrub *sc, + xfs_ino_t *parent_ino) +{ + struct xrep_findparent_info fpi = { + .sc = sc, + .found_parent = NULLFSINO, + }; + int error; + + /* + * The root directory always points to itself. Unlinked dirs can point + * anywhere, so we point them at the root dir too. + */ + if (sc->ip == sc->mp->m_rootip || VFS_I(sc->ip)->i_nlink == 0) { + *parent_ino = sc->mp->m_sb.sb_rootino; + return 0; + } + + /* Reject garbage parent inode numbers and self-referential parents. */ + if (*parent_ino == NULLFSINO) + return 0; + if (!xfs_verify_dir_ino(sc->mp, *parent_ino) || + *parent_ino == sc->ip->i_ino) { + *parent_ino = NULLFSINO; + return 0; + } + + error = xchk_iget(sc, *parent_ino, &fpi.dp); + if (error) + return error; + + if (!S_ISDIR(VFS_I(fpi.dp)->i_mode)) { + *parent_ino = NULLFSINO; + goto out_rele; + } + + error = xrep_findparent_walk_directory(&fpi); + if (error) + goto out_rele; + + *parent_ino = fpi.found_parent; +out_rele: + xchk_irele(sc, fpi.dp); + return error; +} + +/* + * If we're the root of a directory tree, we are our own parent. If we're an + * unlinked directory, the parent /won't/ have a link to us. Set the parent + * directory to the root for both cases. Returns NULLFSINO if we don't know + * what to do. + */ +xfs_ino_t +xrep_findparent_self_reference( + struct xfs_scrub *sc) +{ + if (sc->ip->i_ino == sc->mp->m_sb.sb_rootino) + return sc->mp->m_sb.sb_rootino; + + if (VFS_I(sc->ip)->i_nlink == 0) + return sc->mp->m_sb.sb_rootino; + + return NULLFSINO; +} diff --git a/fs/xfs/scrub/findparent.h b/fs/xfs/scrub/findparent.h new file mode 100644 index 000000000000..d946bc81f34e --- /dev/null +++ b/fs/xfs/scrub/findparent.h @@ -0,0 +1,49 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_FINDPARENT_H__ +#define __XFS_SCRUB_FINDPARENT_H__ + +struct xrep_parent_scan_info { + struct xfs_scrub *sc; + + /* Inode scan cursor. */ + struct xchk_iscan iscan; + + /* Hook to capture directory entry updates. */ + struct xfs_dir_hook dhook; + + /* Lock protecting parent_ino. */ + struct mutex lock; + + /* Parent inode that we've found. */ + xfs_ino_t parent_ino; + + bool lookup_parent; +}; + +int xrep_findparent_scan_start(struct xfs_scrub *sc, + struct xrep_parent_scan_info *pscan); +int xrep_findparent_scan(struct xrep_parent_scan_info *pscan); +void xrep_findparent_scan_teardown(struct xrep_parent_scan_info *pscan); + +static inline void +xrep_findparent_scan_found( + struct xrep_parent_scan_info *pscan, + xfs_ino_t ino) +{ + mutex_lock(&pscan->lock); + pscan->parent_ino = ino; + mutex_unlock(&pscan->lock); +} + +void xrep_findparent_scan_finish_early(struct xrep_parent_scan_info *pscan, + xfs_ino_t ino); + +int xrep_findparent_confirm(struct xfs_scrub *sc, xfs_ino_t *parent_ino); + +xfs_ino_t xrep_findparent_self_reference(struct xfs_scrub *sc); + +#endif /* __XFS_SCRUB_FINDPARENT_H__ */ diff --git a/fs/xfs/scrub/iscan.c b/fs/xfs/scrub/iscan.c index c643b7d79b60..c380207702e2 100644 --- a/fs/xfs/scrub/iscan.c +++ b/fs/xfs/scrub/iscan.c @@ -243,6 +243,17 @@ xchk_iscan_finish( mutex_unlock(&iscan->lock); } +/* Mark an inode scan finished before we actually scan anything. */ +void +xchk_iscan_finish_early( + struct xchk_iscan *iscan) +{ + ASSERT(iscan->cursor_ino == iscan->scan_start_ino); + ASSERT(iscan->__visited_ino == iscan->scan_start_ino); + + xchk_iscan_finish(iscan); +} + /* * Grab the AGI to advance the inode scan. Returns 0 if *agi_bpp is now set, * -ECANCELED if the live scan aborted, -EBUSY if the AGI could not be grabbed, @@ -436,8 +447,13 @@ xchk_iscan_iget( * It's possible that this inode has lost all of its links but * hasn't yet been inactivated. If we don't have a transaction * or it's not writable, flush the inodegc workers and wait. + * If we have a non-empty transaction, we must not block on + * inodegc, which allocates its own transactions. */ - xfs_inodegc_flush(mp); + if (sc->tp && !(sc->tp->t_flags & XFS_TRANS_NO_WRITECOUNT)) + xfs_inodegc_push(mp); + else + xfs_inodegc_flush(mp); return xchk_iscan_iget_retry(iscan, true); } diff --git a/fs/xfs/scrub/iscan.h b/fs/xfs/scrub/iscan.h index 5e0e4ed9dea6..f9f47fa01a9e 100644 --- a/fs/xfs/scrub/iscan.h +++ b/fs/xfs/scrub/iscan.h @@ -88,6 +88,7 @@ xchk_iscan_set_agi_trylock(struct xchk_iscan *iscan) void xchk_iscan_start(struct xfs_scrub *sc, unsigned int iget_timeout, unsigned int iget_retry_delay, struct xchk_iscan *iscan); +void xchk_iscan_finish_early(struct xchk_iscan *iscan); void xchk_iscan_teardown(struct xchk_iscan *iscan); int xchk_iscan_iter(struct xchk_iscan *iscan, struct xfs_inode **ipp); diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index d6d9e8d6109c..85537a87516e 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2611,6 +2611,7 @@ DEFINE_EVENT(xrep_parent_salvage_class, name, \ TP_PROTO(struct xfs_inode *dp, xfs_ino_t ino), \ TP_ARGS(dp, ino)) DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_dir_salvaged_parent); +DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_findparent_dirent); #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ -- cgit v1.2.3-59-g8ed1b From cc22edab9ea7f3ebcb61d41a417d4397e9b7b128 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:53 -0700 Subject: xfs: online repair of parent pointers Teach the online repair code to fix parent pointers for directories. For now, this means correcting the dotdot entry of an existing directory that is otherwise consistent. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/parent.c | 10 ++ fs/xfs/scrub/parent_repair.c | 221 +++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/repair.h | 4 + fs/xfs/scrub/scrub.c | 2 +- fs/xfs/scrub/trace.h | 1 + 6 files changed, 238 insertions(+), 1 deletion(-) create mode 100644 fs/xfs/scrub/parent_repair.c (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 3c754777ec28..d48646f86563 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -205,6 +205,7 @@ xfs-y += $(addprefix scrub/, \ inode_repair.o \ newbt.o \ nlinks_repair.o \ + parent_repair.o \ rcbag_btree.o \ rcbag.o \ reap.o \ diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index 050a8e8914f6..acb6282c3d14 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -10,6 +10,7 @@ #include "xfs_trans_resv.h" #include "xfs_mount.h" #include "xfs_log_format.h" +#include "xfs_trans.h" #include "xfs_inode.h" #include "xfs_icache.h" #include "xfs_dir2.h" @@ -18,12 +19,21 @@ #include "scrub/common.h" #include "scrub/readdir.h" #include "scrub/tempfile.h" +#include "scrub/repair.h" /* Set us up to scrub parents. */ int xchk_setup_parent( struct xfs_scrub *sc) { + int error; + + if (xchk_could_repair(sc)) { + error = xrep_setup_parent(sc); + if (error) + return error; + } + return xchk_setup_inode_contents(sc, 0); } diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c new file mode 100644 index 000000000000..0a9651bb0b05 --- /dev/null +++ b/fs/xfs/scrub/parent_repair.c @@ -0,0 +1,221 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2020-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_bit.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_inode.h" +#include "xfs_icache.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_dir2.h" +#include "xfs_bmap_btree.h" +#include "xfs_dir2_priv.h" +#include "xfs_trans_space.h" +#include "xfs_health.h" +#include "xfs_exchmaps.h" +#include "scrub/xfs_scrub.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/iscan.h" +#include "scrub/findparent.h" +#include "scrub/readdir.h" + +/* + * Repairing The Directory Parent Pointer + * ====================================== + * + * Currently, only directories support parent pointers (in the form of '..' + * entries), so we simply scan the filesystem and update the '..' entry. + * + * Note that because the only parent pointer is the dotdot entry, we won't + * touch an unhealthy directory, since the directory repair code is perfectly + * capable of rebuilding a directory with the proper parent inode. + * + * See the section on locking issues in dir_repair.c for more information about + * conflicts with the VFS. The findparent code wll keep our incore parent + * inode up to date. + */ + +struct xrep_parent { + struct xfs_scrub *sc; + + /* + * Information used to scan the filesystem to find the inumber of the + * dotdot entry for this directory. + */ + struct xrep_parent_scan_info pscan; +}; + +/* Tear down all the incore stuff we created. */ +static void +xrep_parent_teardown( + struct xrep_parent *rp) +{ + xrep_findparent_scan_teardown(&rp->pscan); +} + +/* Set up for a parent repair. */ +int +xrep_setup_parent( + struct xfs_scrub *sc) +{ + struct xrep_parent *rp; + + xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); + + rp = kvzalloc(sizeof(struct xrep_parent), XCHK_GFP_FLAGS); + if (!rp) + return -ENOMEM; + rp->sc = sc; + sc->buf = rp; + + return 0; +} + +/* + * Scan all files in the filesystem for a child dirent that we can turn into + * the dotdot entry for this directory. + */ +STATIC int +xrep_parent_find_dotdot( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + xfs_ino_t ino; + unsigned int sick, checked; + int error; + + /* + * Avoid sick directories. There shouldn't be anyone else clearing the + * directory's sick status. + */ + xfs_inode_measure_sickness(sc->ip, &sick, &checked); + if (sick & XFS_SICK_INO_DIR) + return -EFSCORRUPTED; + + ino = xrep_findparent_self_reference(sc); + if (ino != NULLFSINO) { + xrep_findparent_scan_finish_early(&rp->pscan, ino); + return 0; + } + + /* + * Drop the ILOCK on this directory so that we can scan for the dotdot + * entry. Figure out who is going to be the parent of this directory, + * then retake the ILOCK so that we can salvage directory entries. + */ + xchk_iunlock(sc, XFS_ILOCK_EXCL); + error = xrep_findparent_scan(&rp->pscan); + xchk_ilock(sc, XFS_ILOCK_EXCL); + + return error; +} + +/* Reset a directory's dotdot entry, if needed. */ +STATIC int +xrep_parent_reset_dotdot( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + xfs_ino_t ino; + unsigned int spaceres; + int error = 0; + + ASSERT(sc->ilock_flags & XFS_ILOCK_EXCL); + + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &ino); + if (error || ino == rp->pscan.parent_ino) + return error; + + xfs_trans_ijoin(sc->tp, sc->ip, 0); + + trace_xrep_parent_reset_dotdot(sc->ip, rp->pscan.parent_ino); + + /* + * Reserve more space just in case we have to expand the dir. We're + * allowed to exceed quota to repair inconsistent metadata. + */ + spaceres = XFS_RENAME_SPACE_RES(sc->mp, xfs_name_dotdot.len); + error = xfs_trans_reserve_more_inode(sc->tp, sc->ip, spaceres, 0, + true); + if (error) + return error; + + error = xfs_dir_replace(sc->tp, sc->ip, &xfs_name_dotdot, + rp->pscan.parent_ino, spaceres); + if (error) + return error; + + /* + * Roll transaction to detach the inode from the transaction but retain + * ILOCK_EXCL. + */ + return xfs_trans_roll(&sc->tp); +} + +/* + * Commit the new parent pointer structure (currently only the dotdot entry) to + * the file that we're repairing. + */ +STATIC int +xrep_parent_rebuild_tree( + struct xrep_parent *rp) +{ + if (rp->pscan.parent_ino == NULLFSINO) { + /* Cannot fix orphaned directories yet. */ + return -EFSCORRUPTED; + } + + return xrep_parent_reset_dotdot(rp); +} + +/* Set up the filesystem scan so we can look for parents. */ +STATIC int +xrep_parent_setup_scan( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + + return xrep_findparent_scan_start(sc, &rp->pscan); +} + +int +xrep_parent( + struct xfs_scrub *sc) +{ + struct xrep_parent *rp = sc->buf; + int error; + + error = xrep_parent_setup_scan(rp); + if (error) + return error; + + error = xrep_parent_find_dotdot(rp); + if (error) + goto out_teardown; + + /* Last chance to abort before we start committing fixes. */ + if (xchk_should_terminate(sc, &error)) + goto out_teardown; + + error = xrep_parent_rebuild_tree(rp); + if (error) + goto out_teardown; + +out_teardown: + xrep_parent_teardown(rp); + return error; +} diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index 4e25aa95753a..e53374fa5430 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -92,6 +92,7 @@ int xrep_setup_ag_rmapbt(struct xfs_scrub *sc); int xrep_setup_ag_refcountbt(struct xfs_scrub *sc); int xrep_setup_xattr(struct xfs_scrub *sc); int xrep_setup_directory(struct xfs_scrub *sc); +int xrep_setup_parent(struct xfs_scrub *sc); /* Repair setup functions */ int xrep_setup_ag_allocbt(struct xfs_scrub *sc); @@ -127,6 +128,7 @@ int xrep_nlinks(struct xfs_scrub *sc); int xrep_fscounters(struct xfs_scrub *sc); int xrep_xattr(struct xfs_scrub *sc); int xrep_directory(struct xfs_scrub *sc); +int xrep_parent(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xrep_rtbitmap(struct xfs_scrub *sc); @@ -198,6 +200,7 @@ xrep_setup_nothing( #define xrep_setup_ag_refcountbt xrep_setup_nothing #define xrep_setup_xattr xrep_setup_nothing #define xrep_setup_directory xrep_setup_nothing +#define xrep_setup_parent xrep_setup_nothing #define xrep_setup_inode(sc, imap) ((void)0) @@ -225,6 +228,7 @@ xrep_setup_nothing( #define xrep_rtsummary xrep_notsupported #define xrep_xattr xrep_notsupported #define xrep_directory xrep_notsupported +#define xrep_parent xrep_notsupported #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 8e9e2bf121c2..520d83db193c 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -343,7 +343,7 @@ static const struct xchk_meta_ops meta_scrub_ops[] = { .type = ST_INODE, .setup = xchk_setup_parent, .scrub = xchk_parent, - .repair = xrep_notsupported, + .repair = xrep_parent, }, [XFS_SCRUB_TYPE_RTBITMAP] = { /* realtime bitmap */ .type = ST_FS, diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 85537a87516e..e1755fe63e67 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2550,6 +2550,7 @@ DEFINE_EVENT(xrep_dir_class, name, \ TP_ARGS(dp, parent_ino)) DEFINE_XREP_DIR_EVENT(xrep_dir_rebuild_tree); DEFINE_XREP_DIR_EVENT(xrep_dir_reset_fork); +DEFINE_XREP_DIR_EVENT(xrep_parent_reset_dotdot); DECLARE_EVENT_CLASS(xrep_dirent_class, TP_PROTO(struct xfs_inode *dp, const struct xfs_name *name, -- cgit v1.2.3-59-g8ed1b From 34c9382c128270d0f4c8b36783b30f3c8085b2dd Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:54 -0700 Subject: xfs: ask the dentry cache if it knows the parent of a directory It's possible that the dentry cache can tell us the parent of a directory. Therefore, when repairing directory dot dot entries, query the dcache as a last resort before scanning the entire filesystem. A reviewer asks: "How high is the chance that we actually have a valid dcache entry for a file in a corrupted directory?" There's a decent chance of this actually working. Say you have a 1000-block directory foo, and block 980 gets corrupted. Let's further suppose that block 0 has a correct entry for ".." and "bar". If someone accesses /mnt/foo/bar, that will cause the dcache to create a dentry from /mnt to /mnt/foo whose d_parent points back to /mnt. If you then want to rebuild the directory, XFS can obtain the parent from the dcache without needing to wander into parent pointers or scan the filesystem to find /mnt's connection to foo. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/dir_repair.c | 29 +++++++++++++++++++++++++++++ fs/xfs/scrub/findparent.c | 38 +++++++++++++++++++++++++++++++++++++- fs/xfs/scrub/findparent.h | 1 + fs/xfs/scrub/parent_repair.c | 13 +++++++++++++ fs/xfs/scrub/trace.h | 1 + 5 files changed, 81 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index b17de79207db..34fe720fde0e 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -208,6 +208,29 @@ xrep_dir_lookup_parent( return ino; } +/* + * Look up '..' in the dentry cache and confirm that it's really the parent. + * Returns NULLFSINO if the dcache misses or if the hit is implausible. + */ +static inline xfs_ino_t +xrep_dir_dcache_parent( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + xfs_ino_t parent_ino; + int error; + + parent_ino = xrep_findparent_from_dcache(sc); + if (parent_ino == NULLFSINO) + return parent_ino; + + error = xrep_findparent_confirm(sc, &parent_ino); + if (error) + return NULLFSINO; + + return parent_ino; +} + /* Try to find the parent of the directory being repaired. */ STATIC int xrep_dir_find_parent( @@ -221,6 +244,12 @@ xrep_dir_find_parent( return 0; } + ino = xrep_dir_dcache_parent(rd); + if (ino != NULLFSINO) { + xrep_findparent_scan_finish_early(&rd->pscan, ino); + return 0; + } + ino = xrep_dir_lookup_parent(rd); if (ino != NULLFSINO) { xrep_findparent_scan_finish_early(&rd->pscan, ino); diff --git a/fs/xfs/scrub/findparent.c b/fs/xfs/scrub/findparent.c index 7b3ec8d7d6cc..712dd73e4789 100644 --- a/fs/xfs/scrub/findparent.c +++ b/fs/xfs/scrub/findparent.c @@ -53,7 +53,8 @@ * must not read the scan results without re-taking @sc->ip's ILOCK. * * There are a few shortcuts that we can take to avoid scanning the entire - * filesystem, such as noticing directory tree roots. + * filesystem, such as noticing directory tree roots and querying the dentry + * cache for parent information. */ struct xrep_findparent_info { @@ -410,3 +411,38 @@ xrep_findparent_self_reference( return NULLFSINO; } + +/* Check the dentry cache to see if knows of a parent for the scrub target. */ +xfs_ino_t +xrep_findparent_from_dcache( + struct xfs_scrub *sc) +{ + struct inode *pip = NULL; + struct dentry *dentry, *parent; + xfs_ino_t ret = NULLFSINO; + + dentry = d_find_alias(VFS_I(sc->ip)); + if (!dentry) + goto out; + + parent = dget_parent(dentry); + if (!parent) + goto out_dput; + + ASSERT(parent->d_sb == sc->ip->i_mount->m_super); + + pip = igrab(d_inode(parent)); + dput(parent); + + if (S_ISDIR(pip->i_mode)) { + trace_xrep_findparent_from_dcache(sc->ip, XFS_I(pip)->i_ino); + ret = XFS_I(pip)->i_ino; + } + + xchk_irele(sc, XFS_I(pip)); + +out_dput: + dput(dentry); +out: + return ret; +} diff --git a/fs/xfs/scrub/findparent.h b/fs/xfs/scrub/findparent.h index d946bc81f34e..501f99d3164e 100644 --- a/fs/xfs/scrub/findparent.h +++ b/fs/xfs/scrub/findparent.h @@ -45,5 +45,6 @@ void xrep_findparent_scan_finish_early(struct xrep_parent_scan_info *pscan, int xrep_findparent_confirm(struct xfs_scrub *sc, xfs_ino_t *parent_ino); xfs_ino_t xrep_findparent_self_reference(struct xfs_scrub *sc); +xfs_ino_t xrep_findparent_from_dcache(struct xfs_scrub *sc); #endif /* __XFS_SCRUB_FINDPARENT_H__ */ diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index 0a9651bb0b05..826926c2bb0d 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -118,7 +118,20 @@ xrep_parent_find_dotdot( * then retake the ILOCK so that we can salvage directory entries. */ xchk_iunlock(sc, XFS_ILOCK_EXCL); + + /* Does the VFS dcache have an answer for us? */ + ino = xrep_findparent_from_dcache(sc); + if (ino != NULLFSINO) { + error = xrep_findparent_confirm(sc, &ino); + if (!error && ino != NULLFSINO) { + xrep_findparent_scan_finish_early(&rp->pscan, ino); + goto out_relock; + } + } + + /* Scan the entire filesystem for a parent. */ error = xrep_findparent_scan(&rp->pscan); +out_relock: xchk_ilock(sc, XFS_ILOCK_EXCL); return error; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index e1755fe63e67..d68ec8e2781e 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2613,6 +2613,7 @@ DEFINE_EVENT(xrep_parent_salvage_class, name, \ TP_ARGS(dp, ino)) DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_dir_salvaged_parent); DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_findparent_dirent); +DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_findparent_from_dcache); #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ -- cgit v1.2.3-59-g8ed1b From 1e58a8ccf2597c9259a8e71a2bffac5e11e12ea0 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:55 -0700 Subject: xfs: move orphan files to the orphanage When we're repairing a directory structure or fixing the dotdot entry of a subdirectory, it's possible that we won't ever find a parent for the subdirectory. When this is the case, move it to the orphanage, aka /lost+found. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- .../filesystems/xfs/xfs-online-fsck-design.rst | 19 +- fs/xfs/Makefile | 1 + fs/xfs/scrub/dir_repair.c | 130 +++++- fs/xfs/scrub/orphanage.c | 498 +++++++++++++++++++++ fs/xfs/scrub/orphanage.h | 75 ++++ fs/xfs/scrub/parent_repair.c | 100 ++++- fs/xfs/scrub/scrub.c | 2 + fs/xfs/scrub/scrub.h | 4 + fs/xfs/scrub/trace.h | 28 ++ fs/xfs/xfs_inode.c | 6 +- fs/xfs/xfs_inode.h | 1 + 11 files changed, 844 insertions(+), 20 deletions(-) create mode 100644 fs/xfs/scrub/orphanage.c create mode 100644 fs/xfs/scrub/orphanage.h (limited to 'fs') diff --git a/Documentation/filesystems/xfs/xfs-online-fsck-design.rst b/Documentation/filesystems/xfs/xfs-online-fsck-design.rst index 3afa1bc5f47c..37dddaaeda50 100644 --- a/Documentation/filesystems/xfs/xfs-online-fsck-design.rst +++ b/Documentation/filesystems/xfs/xfs-online-fsck-design.rst @@ -4778,14 +4778,21 @@ Orphaned files are adopted by the orphanage as follows: The ``xrep_orphanage_iolock_two`` function follows the inode locking strategy discussed earlier. -3. Call ``xrep_orphanage_compute_blkres`` and ``xrep_orphanage_compute_name`` - to compute the new name in the orphanage and the block reservation required. - -4. Use ``xrep_orphanage_adoption_prep`` to reserve resources to the repair +3. Use ``xrep_adoption_trans_alloc`` to reserve resources to the repair transaction. -5. Call ``xrep_orphanage_adopt`` to reparent the orphaned file into the lost - and found, and update the kernel dentry cache. +4. Call ``xrep_orphanage_compute_name`` to compute the new name in the + orphanage. + +5. If the adoption is going to happen, call ``xrep_adoption_reparent`` to + reparent the orphaned file into the lost and found and invalidate the dentry + cache. + +6. Call ``xrep_adoption_finish`` to commit any filesystem updates, release the + orphanage ILOCK, and clean the scrub transaction. + +7. If a runtime error happens, call ``xrep_adoption_cancel`` to release all + resources. The proposed patches are in the `orphanage adoption diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index d48646f86563..1e23d1b3cd7b 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -205,6 +205,7 @@ xfs-y += $(addprefix scrub/, \ inode_repair.o \ newbt.o \ nlinks_repair.o \ + orphanage.o \ parent_repair.o \ rcbag_btree.o \ rcbag.o \ diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 34fe720fde0e..c150b2efa2c2 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -42,6 +42,7 @@ #include "scrub/readdir.h" #include "scrub/reap.h" #include "scrub/findparent.h" +#include "scrub/orphanage.h" /* * Directory Repair @@ -115,12 +116,21 @@ struct xrep_dir { */ struct xrep_parent_scan_info pscan; + /* + * Context information for attaching this directory to the lost+found + * if this directory does not have a parent. + */ + struct xrep_adoption adoption; + /* How many subdirectories did we find? */ uint64_t subdirs; /* How many dirents did we find? */ unsigned int dirents; + /* Should we move this directory to the orphanage? */ + bool needs_adoption; + /* Directory entry name, plus the trailing null. */ struct xfs_name xname; unsigned char namebuf[MAXNAMELEN]; @@ -148,6 +158,10 @@ xrep_setup_directory( xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); + error = xrep_orphanage_try_create(sc); + if (error) + return error; + error = xrep_tempfile_create(sc, S_IFDIR); if (error) return error; @@ -1137,10 +1151,8 @@ xrep_dir_set_nlink( /* * The directory is not on the incore unlinked list, which means that * it needs to be reachable via the directory tree. Update the nlink - * with our observed link count. - * - * XXX: A subsequent patch will handle parentless directories by moving - * them to the lost and found instead of aborting the repair. + * with our observed link count. If the directory has no parent, it + * will be moved to the orphanage. */ if (!xfs_inode_on_unlinked_list(dp)) goto reset_nlink; @@ -1151,6 +1163,7 @@ xrep_dir_set_nlink( * inactivate when the last reference drops. */ if (rd->dirents == 0) { + rd->needs_adoption = false; new_nlink = 0; goto reset_nlink; } @@ -1159,7 +1172,8 @@ xrep_dir_set_nlink( * The directory is on the unlinked list and we found dirents. This * directory needs to be reachable via the directory tree. Remove the * dir from the unlinked list and update nlink with the observed link - * count. + * count. If the directory has no parent, it will be moved to the + * orphanage. */ pag = xfs_perag_get(sc->mp, XFS_INO_TO_AGNO(sc->mp, dp->i_ino)); if (!pag) { @@ -1195,12 +1209,16 @@ xrep_dir_swap( return -EFSCORRUPTED; /* - * If we never found the parent for this directory, we can't fix this - * directory. + * If we never found the parent for this directory, temporarily assign + * the root dir as the parent; we'll move this to the orphanage after + * exchanging the dir contents. We hold the ILOCK of the dir being + * repaired, so we're not worried about racy updates of dotdot. */ ASSERT(sc->ilock_flags & XFS_ILOCK_EXCL); - if (rd->pscan.parent_ino == NULLFSINO) - return -EFSCORRUPTED; + if (rd->pscan.parent_ino == NULLFSINO) { + rd->needs_adoption = true; + rd->pscan.parent_ino = rd->sc->mp->m_sb.sb_rootino; + } /* * Reset the temporary directory's '..' entry to point to the parent @@ -1358,6 +1376,91 @@ out_xfarray: return error; } +/* + * Move the current file to the orphanage. + * + * Caller must hold IOLOCK_EXCL on @sc->ip, and no other inode locks. Upon + * successful return, the scrub transaction will have enough extra reservation + * to make the move; it will hold IOLOCK_EXCL and ILOCK_EXCL of @sc->ip and the + * orphanage; and both inodes will be ijoined. + */ +STATIC int +xrep_dir_move_to_orphanage( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + xfs_ino_t orig_parent, new_parent; + int error; + + /* + * We are about to drop the ILOCK on sc->ip to lock the orphanage and + * prepare for the adoption. Therefore, look up the old dotdot entry + * for sc->ip so that we can compare it after we re-lock sc->ip. + */ + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &orig_parent); + if (error) + return error; + + /* + * Drop the ILOCK on the scrub target and commit the transaction. + * Adoption computes its own resource requirements and gathers the + * necessary components. + */ + error = xrep_trans_commit(sc); + if (error) + return error; + xchk_iunlock(sc, XFS_ILOCK_EXCL); + + /* If we can take the orphanage's iolock then we're ready to move. */ + if (!xrep_orphanage_ilock_nowait(sc, XFS_IOLOCK_EXCL)) { + xchk_iunlock(sc, sc->ilock_flags); + error = xrep_orphanage_iolock_two(sc); + if (error) + return error; + } + + /* Grab transaction and ILOCK the two files. */ + error = xrep_adoption_trans_alloc(sc, &rd->adoption); + if (error) + return error; + + error = xrep_adoption_compute_name(&rd->adoption, &rd->xname); + if (error) + return error; + + /* + * Now that we've reacquired the ILOCK on sc->ip, look up the dotdot + * entry again. If the parent changed or the child was unlinked while + * the child directory was unlocked, we don't need to move the child to + * the orphanage after all. + */ + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &new_parent); + if (error) + return error; + + /* + * Attach to the orphanage if we still have a linked directory and it + * hasn't been moved. + */ + if (orig_parent == new_parent && VFS_I(sc->ip)->i_nlink > 0) { + error = xrep_adoption_move(&rd->adoption); + if (error) + return error; + } + + /* + * Launder the scrub transaction so we can drop the orphanage ILOCK + * and IOLOCK. Return holding the scrub target's ILOCK and IOLOCK. + */ + error = xrep_adoption_trans_roll(&rd->adoption); + if (error) + return error; + + xrep_orphanage_iunlock(sc, XFS_ILOCK_EXCL); + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + return 0; +} + /* * Repair the directory metadata. * @@ -1396,6 +1499,15 @@ xrep_directory( if (error) goto out_teardown; + if (rd->needs_adoption) { + if (!xrep_orphanage_can_adopt(rd->sc)) + error = -EFSCORRUPTED; + else + error = xrep_dir_move_to_orphanage(rd); + if (error) + goto out_teardown; + } + out_teardown: xrep_dir_teardown(sc); return error; diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c new file mode 100644 index 000000000000..41733be3ef45 --- /dev/null +++ b/fs/xfs/scrub/orphanage.c @@ -0,0 +1,498 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2021-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_quota.h" +#include "xfs_trans_space.h" +#include "xfs_dir2.h" +#include "xfs_icache.h" +#include "xfs_bmap.h" +#include "xfs_bmap_btree.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/repair.h" +#include "scrub/trace.h" +#include "scrub/orphanage.h" +#include "scrub/readdir.h" + +#include + +/* + * The Orphanage + * ============= + * + * If the directory tree is damaged, children of that directory become + * inaccessible via that file path. If a child has no other parents, the file + * is said to be orphaned. xfs_repair fixes this situation by creating a + * orphanage directory (specifically, /lost+found) and creating a directory + * entry pointing to the orphaned file. + * + * Online repair follows this tactic by creating a root-owned /lost+found + * directory if one does not exist. If an orphan is found, it will move that + * files into orphanage. + */ + +/* Make the orphanage owned by root. */ +STATIC int +xrep_chown_orphanage( + struct xfs_scrub *sc, + struct xfs_inode *dp) +{ + struct xfs_trans *tp; + struct xfs_mount *mp = sc->mp; + struct xfs_dquot *udqp = NULL, *gdqp = NULL, *pdqp = NULL; + struct xfs_dquot *oldu = NULL, *oldg = NULL, *oldp = NULL; + struct inode *inode = VFS_I(dp); + int error; + + error = xfs_qm_vop_dqalloc(dp, GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, 0, + XFS_QMOPT_QUOTALL, &udqp, &gdqp, &pdqp); + if (error) + return error; + + error = xfs_trans_alloc_ichange(dp, udqp, gdqp, pdqp, true, &tp); + if (error) + goto out_dqrele; + + /* + * Always clear setuid/setgid/sticky on the orphanage since we don't + * normally want that functionality on this directory and xfs_repair + * doesn't create it this way either. Leave the other access bits + * unchanged. + */ + inode->i_mode &= ~(S_ISUID | S_ISGID | S_ISVTX); + + /* + * Change the ownerships and register quota modifications + * in the transaction. + */ + if (!uid_eq(inode->i_uid, GLOBAL_ROOT_UID)) { + if (XFS_IS_UQUOTA_ON(mp)) + oldu = xfs_qm_vop_chown(tp, dp, &dp->i_udquot, udqp); + inode->i_uid = GLOBAL_ROOT_UID; + } + if (!gid_eq(inode->i_gid, GLOBAL_ROOT_GID)) { + if (XFS_IS_GQUOTA_ON(mp)) + oldg = xfs_qm_vop_chown(tp, dp, &dp->i_gdquot, gdqp); + inode->i_gid = GLOBAL_ROOT_GID; + } + if (dp->i_projid != 0) { + if (XFS_IS_PQUOTA_ON(mp)) + oldp = xfs_qm_vop_chown(tp, dp, &dp->i_pdquot, pdqp); + dp->i_projid = 0; + } + + dp->i_diflags &= ~(XFS_DIFLAG_REALTIME | XFS_DIFLAG_RTINHERIT); + xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); + + XFS_STATS_INC(mp, xs_ig_attrchg); + + if (xfs_has_wsync(mp)) + xfs_trans_set_sync(tp); + error = xfs_trans_commit(tp); + + xfs_qm_dqrele(oldu); + xfs_qm_dqrele(oldg); + xfs_qm_dqrele(oldp); + +out_dqrele: + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + xfs_qm_dqrele(pdqp); + return error; +} + +#define ORPHANAGE "lost+found" + +/* Create the orphanage directory, and set sc->orphanage to it. */ +int +xrep_orphanage_create( + struct xfs_scrub *sc) +{ + struct xfs_mount *mp = sc->mp; + struct dentry *root_dentry, *orphanage_dentry; + struct inode *root_inode = VFS_I(sc->mp->m_rootip); + struct inode *orphanage_inode; + int error; + + if (xfs_is_shutdown(mp)) + return -EIO; + if (xfs_is_readonly(mp)) { + sc->orphanage = NULL; + return 0; + } + + ASSERT(sc->tp == NULL); + ASSERT(sc->orphanage == NULL); + + /* Find the dentry for the root directory... */ + root_dentry = d_find_alias(root_inode); + if (!root_dentry) { + error = -EFSCORRUPTED; + goto out; + } + + /* ...which is a directory, right? */ + if (!d_is_dir(root_dentry)) { + error = -EFSCORRUPTED; + goto out_dput_root; + } + + /* Try to find the orphanage directory. */ + inode_lock_nested(root_inode, I_MUTEX_PARENT); + orphanage_dentry = lookup_one_len(ORPHANAGE, root_dentry, + strlen(ORPHANAGE)); + if (IS_ERR(orphanage_dentry)) { + error = PTR_ERR(orphanage_dentry); + goto out_unlock_root; + } + + /* + * Nothing found? Call mkdir to create the orphanage. Create the + * directory without other-user access because we're live and someone + * could have been relying partly on minimal access to a parent + * directory to control access to a file we put in here. + */ + if (d_really_is_negative(orphanage_dentry)) { + error = vfs_mkdir(&nop_mnt_idmap, root_inode, orphanage_dentry, + 0750); + if (error) + goto out_dput_orphanage; + } + + /* Not a directory? Bail out. */ + if (!d_is_dir(orphanage_dentry)) { + error = -ENOTDIR; + goto out_dput_orphanage; + } + + /* + * Grab a reference to the orphanage. This /should/ succeed since + * we hold the root directory locked and therefore nobody can delete + * the orphanage. + */ + orphanage_inode = igrab(d_inode(orphanage_dentry)); + if (!orphanage_inode) { + error = -ENOENT; + goto out_dput_orphanage; + } + + /* Make sure the orphanage is owned by root. */ + error = xrep_chown_orphanage(sc, XFS_I(orphanage_inode)); + if (error) + goto out_dput_orphanage; + + /* Stash the reference for later and bail out. */ + sc->orphanage = XFS_I(orphanage_inode); + sc->orphanage_ilock_flags = 0; + +out_dput_orphanage: + dput(orphanage_dentry); +out_unlock_root: + inode_unlock(VFS_I(sc->mp->m_rootip)); +out_dput_root: + dput(root_dentry); +out: + return error; +} + +void +xrep_orphanage_ilock( + struct xfs_scrub *sc, + unsigned int ilock_flags) +{ + sc->orphanage_ilock_flags |= ilock_flags; + xfs_ilock(sc->orphanage, ilock_flags); +} + +bool +xrep_orphanage_ilock_nowait( + struct xfs_scrub *sc, + unsigned int ilock_flags) +{ + if (xfs_ilock_nowait(sc->orphanage, ilock_flags)) { + sc->orphanage_ilock_flags |= ilock_flags; + return true; + } + + return false; +} + +void +xrep_orphanage_iunlock( + struct xfs_scrub *sc, + unsigned int ilock_flags) +{ + xfs_iunlock(sc->orphanage, ilock_flags); + sc->orphanage_ilock_flags &= ~ilock_flags; +} + +/* Grab the IOLOCK of the orphanage and sc->ip. */ +int +xrep_orphanage_iolock_two( + struct xfs_scrub *sc) +{ + int error = 0; + + while (true) { + if (xchk_should_terminate(sc, &error)) + return error; + + /* + * Normal XFS takes the IOLOCK before grabbing a transaction. + * Scrub holds a transaction, which means that we can't block + * on either IOLOCK. + */ + if (xrep_orphanage_ilock_nowait(sc, XFS_IOLOCK_EXCL)) { + if (xchk_ilock_nowait(sc, XFS_IOLOCK_EXCL)) + break; + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + } + delay(1); + } + + return 0; +} + +/* Release the orphanage. */ +void +xrep_orphanage_rele( + struct xfs_scrub *sc) +{ + if (!sc->orphanage) + return; + + if (sc->orphanage_ilock_flags) + xfs_iunlock(sc->orphanage, sc->orphanage_ilock_flags); + + xchk_irele(sc, sc->orphanage); + sc->orphanage = NULL; +} + +/* Adoption moves a file into /lost+found */ + +/* Can the orphanage adopt @sc->ip? */ +bool +xrep_orphanage_can_adopt( + struct xfs_scrub *sc) +{ + ASSERT(sc->ip != NULL); + + if (!sc->orphanage) + return false; + if (sc->ip == sc->orphanage) + return false; + if (xfs_internal_inum(sc->mp, sc->ip->i_ino)) + return false; + return true; +} + +/* + * Create a new transaction to send a child to the orphanage. + * + * Allocate a new transaction with sufficient disk space to handle the + * adoption, take ILOCK_EXCL of the orphanage and sc->ip, joins them to the + * transaction, and reserve quota to reparent the latter. Caller must hold the + * IOLOCK of the orphanage and sc->ip. + */ +int +xrep_adoption_trans_alloc( + struct xfs_scrub *sc, + struct xrep_adoption *adopt) +{ + struct xfs_mount *mp = sc->mp; + unsigned int child_blkres = 0; + int error; + + ASSERT(sc->tp == NULL); + ASSERT(sc->ip != NULL); + ASSERT(sc->orphanage != NULL); + ASSERT(sc->ilock_flags & XFS_IOLOCK_EXCL); + ASSERT(sc->orphanage_ilock_flags & XFS_IOLOCK_EXCL); + ASSERT(!(sc->ilock_flags & (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL))); + ASSERT(!(sc->orphanage_ilock_flags & + (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL))); + + /* Compute the worst case space reservation that we need. */ + adopt->sc = sc; + adopt->orphanage_blkres = XFS_LINK_SPACE_RES(mp, MAXNAMELEN); + if (S_ISDIR(VFS_I(sc->ip)->i_mode)) + child_blkres = XFS_RENAME_SPACE_RES(mp, xfs_name_dotdot.len); + adopt->child_blkres = child_blkres; + + /* + * Allocate a transaction to link the child into the parent, along with + * enough disk space to handle expansion of both the orphanage and the + * dotdot entry of a child directory. + */ + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_link, + adopt->orphanage_blkres + adopt->child_blkres, 0, 0, + &sc->tp); + if (error) + return error; + + xfs_lock_two_inodes(sc->orphanage, XFS_ILOCK_EXCL, + sc->ip, XFS_ILOCK_EXCL); + sc->ilock_flags |= XFS_ILOCK_EXCL; + sc->orphanage_ilock_flags |= XFS_ILOCK_EXCL; + + xfs_trans_ijoin(sc->tp, sc->orphanage, 0); + xfs_trans_ijoin(sc->tp, sc->ip, 0); + + /* + * Reserve enough quota in the orphan directory to add the new name. + * Normally the orphanage should have user/group/project ids of zero + * and hence is not subject to quota enforcement, but we're allowed to + * exceed quota to reattach disconnected parts of the directory tree. + */ + error = xfs_trans_reserve_quota_nblks(sc->tp, sc->orphanage, + adopt->orphanage_blkres, 0, true); + if (error) + goto out_cancel; + + /* + * Reserve enough quota in the child directory to change dotdot. + * Here we're also allowed to exceed file quota to repair inconsistent + * metadata. + */ + if (adopt->child_blkres) { + error = xfs_trans_reserve_quota_nblks(sc->tp, sc->ip, + adopt->child_blkres, 0, true); + if (error) + goto out_cancel; + } + + return 0; +out_cancel: + xchk_trans_cancel(sc); + xrep_orphanage_iunlock(sc, XFS_ILOCK_EXCL); + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + return error; +} + +/* + * Compute the xfs_name for the directory entry that we're adding to the + * orphanage. Caller must hold ILOCKs of sc->ip and the orphanage and must not + * reuse namebuf until the adoption completes or is dissolved. + */ +int +xrep_adoption_compute_name( + struct xrep_adoption *adopt, + struct xfs_name *xname) +{ + struct xfs_scrub *sc = adopt->sc; + char *namebuf = (void *)xname->name; + xfs_ino_t ino; + unsigned int incr = 0; + int error = 0; + + adopt->xname = xname; + xname->len = snprintf(namebuf, MAXNAMELEN, "%llu", sc->ip->i_ino); + xname->type = xfs_mode_to_ftype(VFS_I(sc->ip)->i_mode); + + /* Make sure the filename is unique in the lost+found. */ + error = xchk_dir_lookup(sc, sc->orphanage, xname, &ino); + while (error == 0 && incr < 10000) { + xname->len = snprintf(namebuf, MAXNAMELEN, "%llu.%u", + sc->ip->i_ino, ++incr); + error = xchk_dir_lookup(sc, sc->orphanage, xname, &ino); + } + if (error == 0) { + /* We already have 10,000 entries in the orphanage? */ + return -EFSCORRUPTED; + } + + if (error != -ENOENT) + return error; + return 0; +} + +/* + * Move the current file to the orphanage under the computed name. + * + * Returns with a dirty transaction so that the caller can handle any other + * work, such as fixing up unlinked lists or resetting link counts. + */ +int +xrep_adoption_move( + struct xrep_adoption *adopt) +{ + struct xfs_scrub *sc = adopt->sc; + bool isdir = S_ISDIR(VFS_I(sc->ip)->i_mode); + int error; + + trace_xrep_adoption_reparent(sc->orphanage, adopt->xname, + sc->ip->i_ino); + + /* Create the new name in the orphanage. */ + error = xfs_dir_createname(sc->tp, sc->orphanage, adopt->xname, + sc->ip->i_ino, adopt->orphanage_blkres); + if (error) + return error; + + /* + * Bump the link count of the orphanage if we just added a + * subdirectory, and update its timestamps. + */ + xfs_trans_ichgtime(sc->tp, sc->orphanage, + XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + if (isdir) + xfs_bumplink(sc->tp, sc->orphanage); + xfs_trans_log_inode(sc->tp, sc->orphanage, XFS_ILOG_CORE); + + /* Replace the dotdot entry if the child is a subdirectory. */ + if (isdir) { + error = xfs_dir_replace(sc->tp, sc->ip, &xfs_name_dotdot, + sc->orphanage->i_ino, adopt->child_blkres); + if (error) + return error; + } + + /* + * Notify dirent hooks that we moved the file to /lost+found, and + * finish all the deferred work so that we know the adoption is fully + * recorded in the log. + */ + xfs_dir_update_hook(sc->orphanage, sc->ip, 1, adopt->xname); + return 0; +} + +/* + * Roll to a clean scrub transaction so that we can release the orphanage, + * even if xrep_adoption_move was not called. + * + * Commits all the work and deferred ops attached to an adoption request and + * rolls to a clean scrub transaction. On success, returns 0 with the scrub + * context holding a clean transaction with no inodes joined. On failure, + * returns negative errno with no scrub transaction. All inode locks are + * still held after this function returns. + */ +int +xrep_adoption_trans_roll( + struct xrep_adoption *adopt) +{ + struct xfs_scrub *sc = adopt->sc; + int error; + + trace_xrep_adoption_trans_roll(sc->orphanage, sc->ip, + !!(sc->tp->t_flags & XFS_TRANS_DIRTY)); + + /* Finish all the deferred ops to commit all repairs. */ + error = xrep_defer_finish(sc); + if (error) + return error; + + /* Roll the transaction once more to detach the inodes. */ + return xfs_trans_roll(&sc->tp); +} diff --git a/fs/xfs/scrub/orphanage.h b/fs/xfs/scrub/orphanage.h new file mode 100644 index 000000000000..319179ab788d --- /dev/null +++ b/fs/xfs/scrub/orphanage.h @@ -0,0 +1,75 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2021-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_ORPHANAGE_H__ +#define __XFS_SCRUB_ORPHANAGE_H__ + +#ifdef CONFIG_XFS_ONLINE_REPAIR +int xrep_orphanage_create(struct xfs_scrub *sc); + +/* + * If we're doing a repair, ensure that the orphanage exists and attach it to + * the scrub context. + */ +static inline int +xrep_orphanage_try_create( + struct xfs_scrub *sc) +{ + int error; + + ASSERT(sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR); + + error = xrep_orphanage_create(sc); + switch (error) { + case 0: + case -ENOENT: + case -ENOTDIR: + case -ENOSPC: + /* + * If the orphanage can't be found or isn't a directory, we'll + * keep going, but we won't be able to attach the file to the + * orphanage if we can't find the parent. + */ + return 0; + } + + return error; +} + +int xrep_orphanage_iolock_two(struct xfs_scrub *sc); + +void xrep_orphanage_ilock(struct xfs_scrub *sc, unsigned int ilock_flags); +bool xrep_orphanage_ilock_nowait(struct xfs_scrub *sc, + unsigned int ilock_flags); +void xrep_orphanage_iunlock(struct xfs_scrub *sc, unsigned int ilock_flags); + +void xrep_orphanage_rele(struct xfs_scrub *sc); + +/* Information about a request to add a file to the orphanage. */ +struct xrep_adoption { + struct xfs_scrub *sc; + + /* Name used for the adoption. */ + struct xfs_name *xname; + + /* Block reservations for orphanage and child (if directory). */ + unsigned int orphanage_blkres; + unsigned int child_blkres; +}; + +bool xrep_orphanage_can_adopt(struct xfs_scrub *sc); + +int xrep_adoption_trans_alloc(struct xfs_scrub *sc, + struct xrep_adoption *adopt); +int xrep_adoption_compute_name(struct xrep_adoption *adopt, + struct xfs_name *xname); +int xrep_adoption_move(struct xrep_adoption *adopt); +int xrep_adoption_trans_roll(struct xrep_adoption *adopt); +#else +struct xrep_adoption { /* empty */ }; +# define xrep_orphanage_rele(sc) ((void)0) +#endif /* CONFIG_XFS_ONLINE_REPAIR */ + +#endif /* __XFS_SCRUB_ORPHANAGE_H__ */ diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index 826926c2bb0d..ebb5791bf839 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -32,6 +32,8 @@ #include "scrub/iscan.h" #include "scrub/findparent.h" #include "scrub/readdir.h" +#include "scrub/tempfile.h" +#include "scrub/orphanage.h" /* * Repairing The Directory Parent Pointer @@ -57,6 +59,13 @@ struct xrep_parent { * dotdot entry for this directory. */ struct xrep_parent_scan_info pscan; + + /* Orphanage reparenting request. */ + struct xrep_adoption adoption; + + /* Directory entry name, plus the trailing null. */ + struct xfs_name xname; + unsigned char namebuf[MAXNAMELEN]; }; /* Tear down all the incore stuff we created. */ @@ -80,9 +89,10 @@ xrep_setup_parent( if (!rp) return -ENOMEM; rp->sc = sc; + rp->xname.name = rp->namebuf; sc->buf = rp; - return 0; + return xrep_orphanage_try_create(sc); } /* @@ -179,6 +189,91 @@ xrep_parent_reset_dotdot( return xfs_trans_roll(&sc->tp); } +/* + * Move the current file to the orphanage. + * + * Caller must hold IOLOCK_EXCL on @sc->ip, and no other inode locks. Upon + * successful return, the scrub transaction will have enough extra reservation + * to make the move; it will hold IOLOCK_EXCL and ILOCK_EXCL of @sc->ip and the + * orphanage; and both inodes will be ijoined. + */ +STATIC int +xrep_parent_move_to_orphanage( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + xfs_ino_t orig_parent, new_parent; + int error; + + /* + * We are about to drop the ILOCK on sc->ip to lock the orphanage and + * prepare for the adoption. Therefore, look up the old dotdot entry + * for sc->ip so that we can compare it after we re-lock sc->ip. + */ + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &orig_parent); + if (error) + return error; + + /* + * Drop the ILOCK on the scrub target and commit the transaction. + * Adoption computes its own resource requirements and gathers the + * necessary components. + */ + error = xrep_trans_commit(sc); + if (error) + return error; + xchk_iunlock(sc, XFS_ILOCK_EXCL); + + /* If we can take the orphanage's iolock then we're ready to move. */ + if (!xrep_orphanage_ilock_nowait(sc, XFS_IOLOCK_EXCL)) { + xchk_iunlock(sc, sc->ilock_flags); + error = xrep_orphanage_iolock_two(sc); + if (error) + return error; + } + + /* Grab transaction and ILOCK the two files. */ + error = xrep_adoption_trans_alloc(sc, &rp->adoption); + if (error) + return error; + + error = xrep_adoption_compute_name(&rp->adoption, &rp->xname); + if (error) + return error; + + /* + * Now that we've reacquired the ILOCK on sc->ip, look up the dotdot + * entry again. If the parent changed or the child was unlinked while + * the child directory was unlocked, we don't need to move the child to + * the orphanage after all. + */ + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &new_parent); + if (error) + return error; + + /* + * Attach to the orphanage if we still have a linked directory and it + * hasn't been moved. + */ + if (orig_parent == new_parent && VFS_I(sc->ip)->i_nlink > 0) { + error = xrep_adoption_move(&rp->adoption); + if (error) + return error; + } + + /* + * Launder the scrub transaction so we can drop the orphanage ILOCK + * and IOLOCK. Return holding the scrub target's ILOCK and IOLOCK. + */ + error = xrep_adoption_trans_roll(&rp->adoption); + if (error) + return error; + + xrep_orphanage_iunlock(sc, XFS_ILOCK_EXCL); + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + return 0; +} + /* * Commit the new parent pointer structure (currently only the dotdot entry) to * the file that we're repairing. @@ -188,7 +283,8 @@ xrep_parent_rebuild_tree( struct xrep_parent *rp) { if (rp->pscan.parent_ino == NULLFSINO) { - /* Cannot fix orphaned directories yet. */ + if (xrep_orphanage_can_adopt(rp->sc)) + return xrep_parent_move_to_orphanage(rp); return -EFSCORRUPTED; } diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 520d83db193c..6417628ce26b 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -27,6 +27,7 @@ #include "scrub/stats.h" #include "scrub/xfile.h" #include "scrub/tempfile.h" +#include "scrub/orphanage.h" /* * Online Scrub and Repair @@ -217,6 +218,7 @@ xchk_teardown( } xrep_tempfile_rele(sc); + xrep_orphanage_rele(sc); xchk_fsgates_disable(sc); return error; } diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index d38f0b30416c..7abe498f7a46 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -105,6 +105,10 @@ struct xfs_scrub { /* Lock flags for @ip. */ uint ilock_flags; + /* The orphanage, for stashing files that have lost their parent. */ + uint orphanage_ilock_flags; + struct xfs_inode *orphanage; + /* A temporary file on this filesystem, for staging new metadata. */ struct xfs_inode *tempip; uint temp_ilock_flags; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index d68ec8e2781e..7c49aa6f6b8d 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2588,6 +2588,34 @@ DEFINE_EVENT(xrep_dirent_class, name, \ DEFINE_XREP_DIRENT_EVENT(xrep_dir_salvage_entry); DEFINE_XREP_DIRENT_EVENT(xrep_dir_stash_createname); DEFINE_XREP_DIRENT_EVENT(xrep_dir_replay_createname); +DEFINE_XREP_DIRENT_EVENT(xrep_adoption_reparent); + +DECLARE_EVENT_CLASS(xrep_adoption_class, + TP_PROTO(struct xfs_inode *dp, struct xfs_inode *ip, bool moved), + TP_ARGS(dp, ip, moved), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, dir_ino) + __field(xfs_ino_t, child_ino) + __field(bool, moved) + ), + TP_fast_assign( + __entry->dev = dp->i_mount->m_super->s_dev; + __entry->dir_ino = dp->i_ino; + __entry->child_ino = ip->i_ino; + __entry->moved = moved; + ), + TP_printk("dev %d:%d dir 0x%llx child 0x%llx moved? %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->dir_ino, + __entry->child_ino, + __entry->moved) +); +#define DEFINE_XREP_ADOPTION_EVENT(name) \ +DEFINE_EVENT(xrep_adoption_class, name, \ + TP_PROTO(struct xfs_inode *dp, struct xfs_inode *ip, bool moved), \ + TP_ARGS(dp, ip, moved)) +DEFINE_XREP_ADOPTION_EVENT(xrep_adoption_trans_roll); DECLARE_EVENT_CLASS(xrep_parent_salvage_class, TP_PROTO(struct xfs_inode *dp, xfs_ino_t ino), diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 09d643a9e997..803a64687014 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -914,10 +914,10 @@ xfs_droplink( /* * Increment the link count on an inode & log the change. */ -static void +void xfs_bumplink( - xfs_trans_t *tp, - xfs_inode_t *ip) + struct xfs_trans *tp, + struct xfs_inode *ip) { xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 8157ae7f8e59..18bc3d7750a0 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -625,6 +625,7 @@ void xfs_end_io(struct work_struct *work); int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2); void xfs_iunlock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2); void xfs_iunlock2_remapping(struct xfs_inode *ip1, struct xfs_inode *ip2); +void xfs_bumplink(struct xfs_trans *tp, struct xfs_inode *ip); static inline bool xfs_inode_unlinked_incomplete( -- cgit v1.2.3-59-g8ed1b From e6c9e75fbe792e1fb3bc7e7efce5c6bb015023c5 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:56 -0700 Subject: xfs: move files to orphanage instead of letting nlinks drop to zero If we encounter an inode with a nonzero link count but zero observed links, move it to the orphanage. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- .../filesystems/xfs/xfs-online-fsck-design.rst | 3 +- fs/xfs/scrub/nlinks.c | 20 +++- fs/xfs/scrub/nlinks.h | 7 ++ fs/xfs/scrub/nlinks_repair.c | 123 ++++++++++++++++++--- fs/xfs/scrub/repair.h | 2 + fs/xfs/scrub/trace.c | 1 + fs/xfs/scrub/trace.h | 26 +++++ 7 files changed, 163 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/Documentation/filesystems/xfs/xfs-online-fsck-design.rst b/Documentation/filesystems/xfs/xfs-online-fsck-design.rst index 37dddaaeda50..74a8e42c74bd 100644 --- a/Documentation/filesystems/xfs/xfs-online-fsck-design.rst +++ b/Documentation/filesystems/xfs/xfs-online-fsck-design.rst @@ -4789,7 +4789,8 @@ Orphaned files are adopted by the orphanage as follows: cache. 6. Call ``xrep_adoption_finish`` to commit any filesystem updates, release the - orphanage ILOCK, and clean the scrub transaction. + orphanage ILOCK, and clean the scrub transaction. Call + ``xrep_adoption_commit`` to commit the updates and the scrub transaction. 7. If a runtime error happens, call ``xrep_adoption_cancel`` to release all resources. diff --git a/fs/xfs/scrub/nlinks.c b/fs/xfs/scrub/nlinks.c index 8b9aa73093d6..c456523fac9c 100644 --- a/fs/xfs/scrub/nlinks.c +++ b/fs/xfs/scrub/nlinks.c @@ -24,6 +24,7 @@ #include "scrub/xfile.h" #include "scrub/xfarray.h" #include "scrub/iscan.h" +#include "scrub/orphanage.h" #include "scrub/nlinks.h" #include "scrub/trace.h" #include "scrub/readdir.h" @@ -44,11 +45,23 @@ int xchk_setup_nlinks( struct xfs_scrub *sc) { + struct xchk_nlink_ctrs *xnc; + int error; + xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); - sc->buf = kzalloc(sizeof(struct xchk_nlink_ctrs), XCHK_GFP_FLAGS); - if (!sc->buf) + if (xchk_could_repair(sc)) { + error = xrep_setup_nlinks(sc); + if (error) + return error; + } + + xnc = kvzalloc(sizeof(struct xchk_nlink_ctrs), XCHK_GFP_FLAGS); + if (!xnc) return -ENOMEM; + xnc->xname.name = xnc->namebuf; + xnc->sc = sc; + sc->buf = xnc; return xchk_setup_fs(sc); } @@ -873,9 +886,6 @@ xchk_nlinks_setup_scan( xfs_agino_t first_agino, last_agino; int error; - ASSERT(xnc->sc == NULL); - xnc->sc = sc; - mutex_init(&xnc->lock); /* Retry iget every tenth of a second for up to 30 seconds. */ diff --git a/fs/xfs/scrub/nlinks.h b/fs/xfs/scrub/nlinks.h index a950f3daf204..b820712bfd87 100644 --- a/fs/xfs/scrub/nlinks.h +++ b/fs/xfs/scrub/nlinks.h @@ -28,6 +28,13 @@ struct xchk_nlink_ctrs { * from other writer threads. */ struct xfs_dir_hook dhook; + + /* Orphanage reparenting request. */ + struct xrep_adoption adoption; + + /* Directory entry name, plus the trailing null. */ + struct xfs_name xname; + char namebuf[MAXNAMELEN]; }; /* diff --git a/fs/xfs/scrub/nlinks_repair.c b/fs/xfs/scrub/nlinks_repair.c index 23eb08c4b5ad..0cb67339eac8 100644 --- a/fs/xfs/scrub/nlinks_repair.c +++ b/fs/xfs/scrub/nlinks_repair.c @@ -24,6 +24,7 @@ #include "scrub/xfile.h" #include "scrub/xfarray.h" #include "scrub/iscan.h" +#include "scrub/orphanage.h" #include "scrub/nlinks.h" #include "scrub/trace.h" #include "scrub/tempfile.h" @@ -38,6 +39,34 @@ * inode is locked. */ +/* Set up to repair inode link counts. */ +int +xrep_setup_nlinks( + struct xfs_scrub *sc) +{ + return xrep_orphanage_try_create(sc); +} + +/* + * Inodes that aren't the root directory or the orphanage, have a nonzero link + * count, and no observed parents should be moved to the orphanage. + */ +static inline bool +xrep_nlinks_is_orphaned( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int actual_nlink, + const struct xchk_nlink *obs) +{ + struct xfs_mount *mp = ip->i_mount; + + if (obs->parents != 0) + return false; + if (ip == mp->m_rootip || ip == sc->orphanage) + return false; + return actual_nlink != 0; +} + /* Remove an inode from the unlinked list. */ STATIC int xrep_nlinks_iunlink_remove( @@ -66,6 +95,7 @@ xrep_nlinks_repair_inode( struct xfs_inode *ip = sc->ip; uint64_t total_links; uint64_t actual_nlink; + bool orphanage_available = false; bool dirty = false; int error; @@ -77,14 +107,41 @@ xrep_nlinks_repair_inode( if (xrep_is_tempfile(ip)) return 0; - xchk_ilock(sc, XFS_IOLOCK_EXCL); + /* + * If the filesystem has an orphanage attached to the scrub context, + * prepare for a link count repair that could involve @ip being adopted + * by the lost+found. + */ + if (xrep_orphanage_can_adopt(sc)) { + error = xrep_orphanage_iolock_two(sc); + if (error) + return error; + + error = xrep_adoption_trans_alloc(sc, &xnc->adoption); + if (error) { + xchk_iunlock(sc, XFS_IOLOCK_EXCL); + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + } else { + orphanage_available = true; + } + } - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_link, 0, 0, 0, &sc->tp); - if (error) - return error; + /* + * Either there is no orphanage or we couldn't allocate resources for + * that kind of update. Let's try again with only the resources we + * need for a simple link count update, since that's much more common. + */ + if (!orphanage_available) { + xchk_ilock(sc, XFS_IOLOCK_EXCL); + + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_link, 0, 0, 0, + &sc->tp); + if (error) + return error; - xchk_ilock(sc, XFS_ILOCK_EXCL); - xfs_trans_ijoin(sc->tp, ip, 0); + xchk_ilock(sc, XFS_ILOCK_EXCL); + xfs_trans_ijoin(sc->tp, ip, 0); + } mutex_lock(&xnc->lock); @@ -122,6 +179,41 @@ xrep_nlinks_repair_inode( goto out_trans; } + /* + * Decide if we're going to move this file to the orphanage, and fix + * up the incore link counts if we are. + */ + if (orphanage_available && + xrep_nlinks_is_orphaned(sc, ip, actual_nlink, &obs)) { + /* Figure out what name we're going to use here. */ + error = xrep_adoption_compute_name(&xnc->adoption, &xnc->xname); + if (error) + goto out_trans; + + /* + * Reattach this file to the directory tree by moving it to + * the orphanage per the adoption parameters that we already + * computed. + */ + error = xrep_adoption_move(&xnc->adoption); + if (error) + goto out_trans; + + /* + * Re-read the link counts since the reparenting will have + * updated our scan info. + */ + mutex_lock(&xnc->lock); + error = xfarray_load_sparse(xnc->nlinks, ip->i_ino, &obs); + mutex_unlock(&xnc->lock); + if (error) + goto out_trans; + + total_links = xchk_nlink_total(ip, &obs); + actual_nlink = VFS_I(ip)->i_nlink; + dirty = true; + } + /* * If this inode is linked from the directory tree and on the unlinked * list, remove it from the unlinked list. @@ -165,14 +257,19 @@ xrep_nlinks_repair_inode( xfs_trans_log_inode(sc->tp, ip, XFS_ILOG_CORE); error = xrep_trans_commit(sc); - xchk_iunlock(sc, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - return error; + goto out_unlock; out_scanlock: mutex_unlock(&xnc->lock); out_trans: xchk_trans_cancel(sc); - xchk_iunlock(sc, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); +out_unlock: + xchk_iunlock(sc, XFS_ILOCK_EXCL); + if (orphanage_available) { + xrep_orphanage_iunlock(sc, XFS_ILOCK_EXCL); + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + } + xchk_iunlock(sc, XFS_IOLOCK_EXCL); return error; } @@ -205,10 +302,10 @@ xrep_nlinks( /* * We need ftype for an accurate count of the number of child * subdirectory links. Child subdirectories with a back link (dotdot - * entry) but no forward link are unfixable, so we cannot repair the - * link count of the parent directory based on the back link count - * alone. Filesystems without ftype support are rare (old V4) so we - * just skip out here. + * entry) but no forward link are moved to the orphanage, so we cannot + * repair the link count of the parent directory based on the back link + * count alone. Filesystems without ftype support are rare (old V4) so + * we just skip out here. */ if (!xfs_has_ftype(sc->mp)) return -EOPNOTSUPP; diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index e53374fa5430..7e6aba7fe558 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -93,6 +93,7 @@ int xrep_setup_ag_refcountbt(struct xfs_scrub *sc); int xrep_setup_xattr(struct xfs_scrub *sc); int xrep_setup_directory(struct xfs_scrub *sc); int xrep_setup_parent(struct xfs_scrub *sc); +int xrep_setup_nlinks(struct xfs_scrub *sc); /* Repair setup functions */ int xrep_setup_ag_allocbt(struct xfs_scrub *sc); @@ -201,6 +202,7 @@ xrep_setup_nothing( #define xrep_setup_xattr xrep_setup_nothing #define xrep_setup_directory xrep_setup_nothing #define xrep_setup_parent xrep_setup_nothing +#define xrep_setup_nlinks xrep_setup_nothing #define xrep_setup_inode(sc, imap) ((void)0) diff --git a/fs/xfs/scrub/trace.c b/fs/xfs/scrub/trace.c index 3dd281d6d185..b2ce7b22cad3 100644 --- a/fs/xfs/scrub/trace.c +++ b/fs/xfs/scrub/trace.c @@ -24,6 +24,7 @@ #include "scrub/xfarray.h" #include "scrub/quota.h" #include "scrub/iscan.h" +#include "scrub/orphanage.h" #include "scrub/nlinks.h" #include "scrub/fscounters.h" diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 7c49aa6f6b8d..2a4c54f7992a 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2643,6 +2643,32 @@ DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_dir_salvaged_parent); DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_findparent_dirent); DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_findparent_from_dcache); +TRACE_EVENT(xrep_nlinks_set_record, + TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, + const struct xchk_nlink *obs), + TP_ARGS(mp, ino, obs), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_nlink_t, parents) + __field(xfs_nlink_t, backrefs) + __field(xfs_nlink_t, children) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->ino = ino; + __entry->parents = obs->parents; + __entry->backrefs = obs->backrefs; + __entry->children = obs->children; + ), + TP_printk("dev %d:%d ino 0x%llx parents %u backrefs %u children %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->parents, + __entry->backrefs, + __entry->children) +); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ -- cgit v1.2.3-59-g8ed1b From 73597e3e42b4a15030e6f93b71b53a04377ea419 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:57 -0700 Subject: xfs: ensure dentry consistency when the orphanage adopts a file When the orphanage adopts a file, that file becomes a child of the orphanage. The dentry cache may have entries for the orphanage directory and the name we've chosen, so (1) make sure we abort if the dcache has a positive entry because something's not right; and (2) invalidate and purge negative dentries if the adoption goes through. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/orphanage.c | 91 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/trace.h | 42 ++++++++++++++++++++++ 2 files changed, 133 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index 41733be3ef45..885b7d478a0a 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -418,6 +418,90 @@ xrep_adoption_compute_name( return 0; } +/* + * Make sure the dcache does not have a positive dentry for the name we've + * chosen. The caller should have checked with the ondisk directory, so any + * discrepancy is a sign that something is seriously wrong. + */ +static int +xrep_adoption_check_dcache( + struct xrep_adoption *adopt) +{ + struct qstr qname = QSTR_INIT(adopt->xname->name, + adopt->xname->len); + struct dentry *d_orphanage, *d_child; + int error = 0; + + d_orphanage = d_find_alias(VFS_I(adopt->sc->orphanage)); + if (!d_orphanage) + return 0; + + d_child = d_hash_and_lookup(d_orphanage, &qname); + if (d_child) { + trace_xrep_adoption_check_child(adopt->sc->mp, d_child); + + if (d_is_positive(d_child)) { + ASSERT(d_is_negative(d_child)); + error = -EFSCORRUPTED; + } + + dput(d_child); + } + + dput(d_orphanage); + if (error) + return error; + + /* + * Do we need to update d_parent of the dentry for the file being + * repaired? There shouldn't be a hashed dentry with a parent since + * the file had nonzero nlink but wasn't connected to any parent dir. + */ + d_child = d_find_alias(VFS_I(adopt->sc->ip)); + if (!d_child) + return 0; + + trace_xrep_adoption_check_alias(adopt->sc->mp, d_child); + + if (d_child->d_parent && !d_unhashed(d_child)) { + ASSERT(d_child->d_parent == NULL || d_unhashed(d_child)); + error = -EFSCORRUPTED; + } + + dput(d_child); + return error; +} + +/* + * Remove all negative dentries from the dcache. There should not be any + * positive entries, since we've maintained our lock on the orphanage + * directory. + */ +static void +xrep_adoption_zap_dcache( + struct xrep_adoption *adopt) +{ + struct qstr qname = QSTR_INIT(adopt->xname->name, + adopt->xname->len); + struct dentry *d_orphanage, *d_child; + + d_orphanage = d_find_alias(VFS_I(adopt->sc->orphanage)); + if (!d_orphanage) + return; + + d_child = d_hash_and_lookup(d_orphanage, &qname); + while (d_child != NULL) { + trace_xrep_adoption_invalidate_child(adopt->sc->mp, d_child); + + ASSERT(d_is_negative(d_child)); + d_invalidate(d_child); + dput(d_child); + d_child = d_lookup(d_orphanage, &qname); + } + + dput(d_orphanage); +} + /* * Move the current file to the orphanage under the computed name. * @@ -435,6 +519,10 @@ xrep_adoption_move( trace_xrep_adoption_reparent(sc->orphanage, adopt->xname, sc->ip->i_ino); + error = xrep_adoption_check_dcache(adopt); + if (error) + return error; + /* Create the new name in the orphanage. */ error = xfs_dir_createname(sc->tp, sc->orphanage, adopt->xname, sc->ip->i_ino, adopt->orphanage_blkres); @@ -465,6 +553,9 @@ xrep_adoption_move( * recorded in the log. */ xfs_dir_update_hook(sc->orphanage, sc->ip, 1, adopt->xname); + + /* Remove negative dentries from the lost+found's dcache */ + xrep_adoption_zap_dcache(adopt); return 0; } diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 2a4c54f7992a..668da6ff2ca2 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2669,6 +2669,48 @@ TRACE_EVENT(xrep_nlinks_set_record, __entry->children) ); +DECLARE_EVENT_CLASS(xrep_dentry_class, + TP_PROTO(struct xfs_mount *mp, const struct dentry *dentry), + TP_ARGS(mp, dentry), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned int, flags) + __field(unsigned long, ino) + __field(bool, positive) + __field(unsigned long, parent_ino) + __field(unsigned int, namelen) + __dynamic_array(char, name, dentry->d_name.len) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->flags = dentry->d_flags; + __entry->positive = d_is_positive(dentry); + if (dentry->d_parent && d_inode(dentry->d_parent)) + __entry->parent_ino = d_inode(dentry->d_parent)->i_ino; + else + __entry->parent_ino = -1UL; + __entry->ino = d_inode(dentry) ? d_inode(dentry)->i_ino : 0; + __entry->namelen = dentry->d_name.len; + memcpy(__get_str(name), dentry->d_name.name, dentry->d_name.len); + ), + TP_printk("dev %d:%d flags 0x%x positive? %d parent_ino 0x%lx ino 0x%lx name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->flags, + __entry->positive, + __entry->parent_ino, + __entry->ino, + __entry->namelen, + __get_str(name)) +); +#define DEFINE_REPAIR_DENTRY_EVENT(name) \ +DEFINE_EVENT(xrep_dentry_class, name, \ + TP_PROTO(struct xfs_mount *mp, const struct dentry *dentry), \ + TP_ARGS(mp, dentry)) +DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_child); +DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_alias); +DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_dentry); +DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_invalidate_child); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ -- cgit v1.2.3-59-g8ed1b From ef744be416b5c649d287604730400dfa728779fe Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:58 -0700 Subject: xfs: expose xfs_bmap_local_to_extents for online repair Allow online repair to call xfs_bmap_local_to_extents and add a void * argument at the end so that online repair can pass its own context. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_bmap.c | 11 ++++++----- fs/xfs/libxfs/xfs_bmap.h | 6 ++++++ fs/xfs/libxfs/xfs_symlink_remote.c | 3 ++- fs/xfs/libxfs/xfs_symlink_remote.h | 3 ++- 4 files changed, 16 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 46bbc9f0a117..59b8b9dc29cc 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -779,7 +779,7 @@ xfs_bmap_local_to_extents_empty( } -STATIC int /* error */ +int /* error */ xfs_bmap_local_to_extents( xfs_trans_t *tp, /* transaction pointer */ xfs_inode_t *ip, /* incore inode pointer */ @@ -789,7 +789,8 @@ xfs_bmap_local_to_extents( void (*init_fn)(struct xfs_trans *tp, struct xfs_buf *bp, struct xfs_inode *ip, - struct xfs_ifork *ifp)) + struct xfs_ifork *ifp, void *priv), + void *priv) { int error = 0; int flags; /* logging flags returned */ @@ -850,7 +851,7 @@ xfs_bmap_local_to_extents( * log here. Note that init_fn must also set the buffer log item type * correctly. */ - init_fn(tp, bp, ip, ifp); + init_fn(tp, bp, ip, ifp, priv); /* account for the change in fork size */ xfs_idata_realloc(ip, -ifp->if_bytes, whichfork); @@ -982,8 +983,8 @@ xfs_bmap_add_attrfork_local( if (S_ISLNK(VFS_I(ip)->i_mode)) return xfs_bmap_local_to_extents(tp, ip, 1, flags, - XFS_DATA_FORK, - xfs_symlink_local_to_remote); + XFS_DATA_FORK, xfs_symlink_local_to_remote, + NULL); /* should only be called for types that support local format data */ ASSERT(0); diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index b8bdbf1560e6..32fb2a455c29 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -179,6 +179,12 @@ unsigned int xfs_bmap_compute_attr_offset(struct xfs_mount *mp); int xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd); void xfs_bmap_local_to_extents_empty(struct xfs_trans *tp, struct xfs_inode *ip, int whichfork); +int xfs_bmap_local_to_extents(struct xfs_trans *tp, struct xfs_inode *ip, + xfs_extlen_t total, int *logflagsp, int whichfork, + void (*init_fn)(struct xfs_trans *tp, struct xfs_buf *bp, + struct xfs_inode *ip, struct xfs_ifork *ifp, + void *priv), + void *priv); void xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork); int xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip, xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork); diff --git a/fs/xfs/libxfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c index 8f0d5c584f46..d150576ddd0a 100644 --- a/fs/xfs/libxfs/xfs_symlink_remote.c +++ b/fs/xfs/libxfs/xfs_symlink_remote.c @@ -169,7 +169,8 @@ xfs_symlink_local_to_remote( struct xfs_trans *tp, struct xfs_buf *bp, struct xfs_inode *ip, - struct xfs_ifork *ifp) + struct xfs_ifork *ifp, + void *priv) { struct xfs_mount *mp = ip->i_mount; char *buf; diff --git a/fs/xfs/libxfs/xfs_symlink_remote.h b/fs/xfs/libxfs/xfs_symlink_remote.h index ac3dac8f617e..83b89a1deb9f 100644 --- a/fs/xfs/libxfs/xfs_symlink_remote.h +++ b/fs/xfs/libxfs/xfs_symlink_remote.h @@ -16,7 +16,8 @@ int xfs_symlink_hdr_set(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset, bool xfs_symlink_hdr_ok(xfs_ino_t ino, uint32_t offset, uint32_t size, struct xfs_buf *bp); void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp, - struct xfs_inode *ip, struct xfs_ifork *ifp); + struct xfs_inode *ip, struct xfs_ifork *ifp, + void *priv); xfs_failaddr_t xfs_symlink_shortform_verify(void *sfp, int64_t size); int xfs_symlink_remote_read(struct xfs_inode *ip, char *link); int xfs_symlink_write_target(struct xfs_trans *tp, struct xfs_inode *ip, -- cgit v1.2.3-59-g8ed1b From ea8214c3195c2ed3a205dea42bbe7746712fc461 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:59 -0700 Subject: xfs: pass the owner to xfs_symlink_write_target Require callers of xfs_symlink_write_target to pass the owner number explicitly. This sets us up for online repair to be able to write a remote symlink target to sc->tempip with sc->ip's inumber in the block heaader. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_symlink_remote.c | 4 ++-- fs/xfs/libxfs/xfs_symlink_remote.h | 4 ++-- fs/xfs/xfs_symlink.c | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c index d150576ddd0a..f228127a88ff 100644 --- a/fs/xfs/libxfs/xfs_symlink_remote.c +++ b/fs/xfs/libxfs/xfs_symlink_remote.c @@ -311,6 +311,7 @@ int xfs_symlink_write_target( struct xfs_trans *tp, struct xfs_inode *ip, + xfs_ino_t owner, const char *target_path, int pathlen, xfs_fsblock_t fs_blocks, @@ -365,8 +366,7 @@ xfs_symlink_write_target( byte_cnt = min(byte_cnt, pathlen); buf = bp->b_addr; - buf += xfs_symlink_hdr_set(mp, ip->i_ino, offset, byte_cnt, - bp); + buf += xfs_symlink_hdr_set(mp, owner, offset, byte_cnt, bp); memcpy(buf, cur_chunk, byte_cnt); diff --git a/fs/xfs/libxfs/xfs_symlink_remote.h b/fs/xfs/libxfs/xfs_symlink_remote.h index 83b89a1deb9f..c1672fe1f17b 100644 --- a/fs/xfs/libxfs/xfs_symlink_remote.h +++ b/fs/xfs/libxfs/xfs_symlink_remote.h @@ -21,8 +21,8 @@ void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp, xfs_failaddr_t xfs_symlink_shortform_verify(void *sfp, int64_t size); int xfs_symlink_remote_read(struct xfs_inode *ip, char *link); int xfs_symlink_write_target(struct xfs_trans *tp, struct xfs_inode *ip, - const char *target_path, int pathlen, xfs_fsblock_t fs_blocks, - uint resblks); + xfs_ino_t owner, const char *target_path, int pathlen, + xfs_fsblock_t fs_blocks, uint resblks); int xfs_symlink_remote_truncate(struct xfs_trans *tp, struct xfs_inode *ip); #endif /* __XFS_SYMLINK_REMOTE_H */ diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 3daeebff4bb4..fb060aaf6d40 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -181,8 +181,8 @@ xfs_symlink( xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp); resblks -= XFS_IALLOC_SPACE_RES(mp); - error = xfs_symlink_write_target(tp, ip, target_path, pathlen, - fs_blocks, resblks); + error = xfs_symlink_write_target(tp, ip, ip->i_ino, target_path, + pathlen, fs_blocks, resblks); if (error) goto out_trans_cancel; resblks -= fs_blocks; -- cgit v1.2.3-59-g8ed1b From 2651923d8d8db00a57665822f017fa7c76758044 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:54:59 -0700 Subject: xfs: online repair of symbolic links If a symbolic link target looks bad, try to sift through the rubble to find as much of the target buffer that we can, and stage a new target (short or remote format as needed) in a temporary file and use the atomic extent swapping mechanism to commit the results. In the worst case, we replace the target with an overly long filename that cannot possibly resolve. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/repair.h | 8 + fs/xfs/scrub/scrub.c | 2 +- fs/xfs/scrub/symlink.c | 13 +- fs/xfs/scrub/symlink_repair.c | 506 ++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/tempfile.c | 13 ++ fs/xfs/scrub/trace.h | 46 ++++ 7 files changed, 587 insertions(+), 2 deletions(-) create mode 100644 fs/xfs/scrub/symlink_repair.c (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 1e23d1b3cd7b..4e1eb3b6dbc4 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -213,6 +213,7 @@ xfs-y += $(addprefix scrub/, \ refcount_repair.o \ repair.o \ rmap_repair.o \ + symlink_repair.o \ tempfile.o \ xfblob.o \ ) diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index 7e6aba7fe558..622eb486a16f 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -94,6 +94,7 @@ int xrep_setup_xattr(struct xfs_scrub *sc); int xrep_setup_directory(struct xfs_scrub *sc); int xrep_setup_parent(struct xfs_scrub *sc); int xrep_setup_nlinks(struct xfs_scrub *sc); +int xrep_setup_symlink(struct xfs_scrub *sc, unsigned int *resblks); /* Repair setup functions */ int xrep_setup_ag_allocbt(struct xfs_scrub *sc); @@ -130,6 +131,7 @@ int xrep_fscounters(struct xfs_scrub *sc); int xrep_xattr(struct xfs_scrub *sc); int xrep_directory(struct xfs_scrub *sc); int xrep_parent(struct xfs_scrub *sc); +int xrep_symlink(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xrep_rtbitmap(struct xfs_scrub *sc); @@ -206,6 +208,11 @@ xrep_setup_nothing( #define xrep_setup_inode(sc, imap) ((void)0) +static inline int xrep_setup_symlink(struct xfs_scrub *sc, unsigned int *x) +{ + return 0; +} + #define xrep_revalidate_allocbt (NULL) #define xrep_revalidate_iallocbt (NULL) @@ -231,6 +238,7 @@ xrep_setup_nothing( #define xrep_xattr xrep_notsupported #define xrep_directory xrep_notsupported #define xrep_parent xrep_notsupported +#define xrep_symlink xrep_notsupported #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 6417628ce26b..301d5b753fdd 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -339,7 +339,7 @@ static const struct xchk_meta_ops meta_scrub_ops[] = { .type = ST_INODE, .setup = xchk_setup_symlink, .scrub = xchk_symlink, - .repair = xrep_notsupported, + .repair = xrep_symlink, }, [XFS_SCRUB_TYPE_PARENT] = { /* parent pointers */ .type = ST_INODE, diff --git a/fs/xfs/scrub/symlink.c b/fs/xfs/scrub/symlink.c index d77d8a9598f6..c848bcc07cd5 100644 --- a/fs/xfs/scrub/symlink.c +++ b/fs/xfs/scrub/symlink.c @@ -10,6 +10,7 @@ #include "xfs_trans_resv.h" #include "xfs_mount.h" #include "xfs_log_format.h" +#include "xfs_trans.h" #include "xfs_inode.h" #include "xfs_symlink.h" #include "xfs_health.h" @@ -17,18 +18,28 @@ #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/health.h" +#include "scrub/repair.h" /* Set us up to scrub a symbolic link. */ int xchk_setup_symlink( struct xfs_scrub *sc) { + unsigned int resblks = 0; + int error; + /* Allocate the buffer without the inode lock held. */ sc->buf = kvzalloc(XFS_SYMLINK_MAXLEN + 1, XCHK_GFP_FLAGS); if (!sc->buf) return -ENOMEM; - return xchk_setup_inode_contents(sc, 0); + if (xchk_could_repair(sc)) { + error = xrep_setup_symlink(sc, &resblks); + if (error) + return error; + } + + return xchk_setup_inode_contents(sc, resblks); } /* Symbolic links. */ diff --git a/fs/xfs/scrub/symlink_repair.c b/fs/xfs/scrub/symlink_repair.c new file mode 100644 index 000000000000..178304959535 --- /dev/null +++ b/fs/xfs/scrub/symlink_repair.c @@ -0,0 +1,506 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2018-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_btree.h" +#include "xfs_bit.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_inode.h" +#include "xfs_inode_fork.h" +#include "xfs_symlink.h" +#include "xfs_bmap.h" +#include "xfs_quota.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_trans_space.h" +#include "xfs_symlink_remote.h" +#include "xfs_exchmaps.h" +#include "xfs_exchrange.h" +#include "xfs_health.h" +#include "scrub/xfs_scrub.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/tempfile.h" +#include "scrub/tempexch.h" +#include "scrub/reap.h" + +/* + * Symbolic Link Repair + * ==================== + * + * We repair symbolic links by reading whatever target data we can find, up to + * the first NULL byte. If the recovered target strlen matches i_size, then + * we rewrite the target. In all other cases, we replace the target with an + * overly long string that cannot possibly resolve. The new target is written + * into a private hidden temporary file, and then a file contents exchange + * commits the new symlink target to the file being repaired. + */ + +/* Set us up to repair the symlink file. */ +int +xrep_setup_symlink( + struct xfs_scrub *sc, + unsigned int *resblks) +{ + struct xfs_mount *mp = sc->mp; + unsigned long long blocks; + int error; + + error = xrep_tempfile_create(sc, S_IFLNK); + if (error) + return error; + + /* + * If we're doing a repair, we reserve enough blocks to write out a + * completely new symlink file, plus twice as many blocks as we would + * need if we can only allocate one block per data fork mapping. This + * should cover the preallocation of the temporary file and exchanging + * the extent mappings. + * + * We cannot use xfs_exchmaps_estimate because we have not yet + * constructed the replacement symlink and therefore do not know how + * many extents it will use. By the time we do, we will have a dirty + * transaction (which we cannot drop because we cannot drop the + * symlink ILOCK) and cannot ask for more reservation. + */ + blocks = xfs_symlink_blocks(sc->mp, XFS_SYMLINK_MAXLEN); + blocks += xfs_bmbt_calc_size(mp, blocks) * 2; + if (blocks > UINT_MAX) + return -EOPNOTSUPP; + + *resblks += blocks; + return 0; +} + +/* + * Try to salvage the pathname from remote blocks. Returns the number of bytes + * salvaged or a negative errno. + */ +STATIC ssize_t +xrep_symlink_salvage_remote( + struct xfs_scrub *sc) +{ + struct xfs_bmbt_irec mval[XFS_SYMLINK_MAPS]; + struct xfs_inode *ip = sc->ip; + struct xfs_buf *bp; + char *target_buf = sc->buf; + xfs_failaddr_t fa; + xfs_filblks_t fsblocks; + xfs_daddr_t d; + loff_t len; + loff_t offset = 0; + unsigned int byte_cnt; + bool magic_ok; + bool hdr_ok; + int n; + int nmaps = XFS_SYMLINK_MAPS; + int error; + + /* We'll only read until the buffer is full. */ + len = min_t(loff_t, ip->i_disk_size, XFS_SYMLINK_MAXLEN); + fsblocks = xfs_symlink_blocks(sc->mp, len); + error = xfs_bmapi_read(ip, 0, fsblocks, mval, &nmaps, 0); + if (error) + return error; + + for (n = 0; n < nmaps; n++) { + struct xfs_dsymlink_hdr *dsl; + + d = XFS_FSB_TO_DADDR(sc->mp, mval[n].br_startblock); + + /* Read the rmt block. We'll run the verifiers manually. */ + error = xfs_trans_read_buf(sc->mp, sc->tp, sc->mp->m_ddev_targp, + d, XFS_FSB_TO_BB(sc->mp, mval[n].br_blockcount), + 0, &bp, NULL); + if (error) + return error; + bp->b_ops = &xfs_symlink_buf_ops; + + /* How many bytes do we expect to get out of this buffer? */ + byte_cnt = XFS_FSB_TO_B(sc->mp, mval[n].br_blockcount); + byte_cnt = XFS_SYMLINK_BUF_SPACE(sc->mp, byte_cnt); + byte_cnt = min_t(unsigned int, byte_cnt, len); + + /* + * See if the verifiers accept this block. We're willing to + * salvage if the if the offset/byte/ino are ok and either the + * verifier passed or the magic is ok. Anything else and we + * stop dead in our tracks. + */ + fa = bp->b_ops->verify_struct(bp); + dsl = bp->b_addr; + magic_ok = dsl->sl_magic == cpu_to_be32(XFS_SYMLINK_MAGIC); + hdr_ok = xfs_symlink_hdr_ok(ip->i_ino, offset, byte_cnt, bp); + if (!hdr_ok || (fa != NULL && !magic_ok)) + break; + + memcpy(target_buf + offset, dsl + 1, byte_cnt); + + len -= byte_cnt; + offset += byte_cnt; + } + return offset; +} + +/* + * Try to salvage an inline symlink's contents. Returns the number of bytes + * salvaged or a negative errno. + */ +STATIC ssize_t +xrep_symlink_salvage_inline( + struct xfs_scrub *sc) +{ + struct xfs_inode *ip = sc->ip; + char *target_buf = sc->buf; + char *old_target; + struct xfs_ifork *ifp; + unsigned int nr; + + ifp = xfs_ifork_ptr(ip, XFS_DATA_FORK); + if (!ifp->if_data) + return 0; + + /* + * If inode repair zapped the link target, pretend that we didn't find + * any bytes at all so that we can replace the (now totally lost) link + * target with a warning message. + */ + old_target = ifp->if_data; + if (xfs_inode_has_sickness(sc->ip, XFS_SICK_INO_SYMLINK_ZAPPED) && + sc->ip->i_disk_size == 1 && old_target[0] == '?') + return 0; + + nr = min(XFS_SYMLINK_MAXLEN, xfs_inode_data_fork_size(ip)); + strncpy(target_buf, ifp->if_data, nr); + return nr; +} + +#define DUMMY_TARGET \ + "The target of this symbolic link could not be recovered at all and " \ + "has been replaced with this explanatory message. To avoid " \ + "accidentally pointing to an existing file path, this message is " \ + "longer than the maximum supported file name length. That is an " \ + "acceptable length for a symlink target on XFS but will produce " \ + "File Name Too Long errors if resolved." + +/* Salvage whatever we can of the target. */ +STATIC int +xrep_symlink_salvage( + struct xfs_scrub *sc) +{ + char *target_buf = sc->buf; + ssize_t buflen = 0; + + BUILD_BUG_ON(sizeof(DUMMY_TARGET) - 1 <= NAME_MAX); + + /* + * Salvage the target if there weren't any corruption problems observed + * while scanning it. + */ + if (!(sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) { + if (sc->ip->i_df.if_format == XFS_DINODE_FMT_LOCAL) + buflen = xrep_symlink_salvage_inline(sc); + else + buflen = xrep_symlink_salvage_remote(sc); + if (buflen < 0) + return buflen; + + /* + * NULL-terminate the buffer because the ondisk target does not + * do that for us. If salvage didn't find the exact amount of + * data that we expected to find, don't salvage anything. + */ + target_buf[buflen] = 0; + if (strlen(target_buf) != sc->ip->i_disk_size) + buflen = 0; + } + + /* + * Change an empty target into a dummy target and clear the symlink + * target zapped flag. + */ + if (buflen == 0) { + sc->sick_mask |= XFS_SICK_INO_SYMLINK_ZAPPED; + sprintf(target_buf, DUMMY_TARGET); + } + + trace_xrep_symlink_salvage_target(sc->ip, target_buf, + strlen(target_buf)); + return 0; +} + +STATIC void +xrep_symlink_local_to_remote( + struct xfs_trans *tp, + struct xfs_buf *bp, + struct xfs_inode *ip, + struct xfs_ifork *ifp, + void *priv) +{ + struct xfs_scrub *sc = priv; + struct xfs_dsymlink_hdr *dsl = bp->b_addr; + + xfs_symlink_local_to_remote(tp, bp, ip, ifp, NULL); + + if (!xfs_has_crc(sc->mp)) + return; + + dsl->sl_owner = cpu_to_be64(sc->ip->i_ino); + xfs_trans_log_buf(tp, bp, 0, + sizeof(struct xfs_dsymlink_hdr) + ifp->if_bytes - 1); +} + +/* + * Prepare both links' data forks for an exchange. Promote the tempfile from + * local format to extents format, and if the file being repaired has a short + * format data fork, turn it into an empty extent list. + */ +STATIC int +xrep_symlink_swap_prep( + struct xfs_scrub *sc, + bool temp_local, + bool ip_local) +{ + int error; + + /* + * If the temp link is in shortform format, convert that to a remote + * target so that we can use the atomic mapping exchange. + */ + if (temp_local) { + int logflags = XFS_ILOG_CORE; + + error = xfs_bmap_local_to_extents(sc->tp, sc->tempip, 1, + &logflags, XFS_DATA_FORK, + xrep_symlink_local_to_remote, + sc); + if (error) + return error; + + xfs_trans_log_inode(sc->tp, sc->ip, 0); + + error = xfs_defer_finish(&sc->tp); + if (error) + return error; + } + + /* + * If the file being repaired had a shortform data fork, convert that + * to an empty extent list in preparation for the atomic mapping + * exchange. + */ + if (ip_local) { + struct xfs_ifork *ifp; + + ifp = xfs_ifork_ptr(sc->ip, XFS_DATA_FORK); + xfs_idestroy_fork(ifp); + ifp->if_format = XFS_DINODE_FMT_EXTENTS; + ifp->if_nextents = 0; + ifp->if_bytes = 0; + ifp->if_data = NULL; + ifp->if_height = 0; + + xfs_trans_log_inode(sc->tp, sc->ip, + XFS_ILOG_CORE | XFS_ILOG_DDATA); + } + + return 0; +} + +/* Exchange the temporary symlink's data fork with the one being repaired. */ +STATIC int +xrep_symlink_swap( + struct xfs_scrub *sc) +{ + struct xrep_tempexch *tx = sc->buf; + bool ip_local, temp_local; + int error; + + ip_local = sc->ip->i_df.if_format == XFS_DINODE_FMT_LOCAL; + temp_local = sc->tempip->i_df.if_format == XFS_DINODE_FMT_LOCAL; + + /* + * If the both links have a local format data fork and the rebuilt + * remote data would fit in the repaired file's data fork, copy the + * contents from the tempfile and declare ourselves done. + */ + if (ip_local && temp_local && + sc->tempip->i_disk_size <= xfs_inode_data_fork_size(sc->ip)) { + xrep_tempfile_copyout_local(sc, XFS_DATA_FORK); + return 0; + } + + /* Otherwise, make sure both data forks are in block-mapping mode. */ + error = xrep_symlink_swap_prep(sc, temp_local, ip_local); + if (error) + return error; + + return xrep_tempexch_contents(sc, tx); +} + +/* + * Free all the remote blocks and reset the data fork. The caller must join + * the inode to the transaction. This function returns with the inode joined + * to a clean scrub transaction. + */ +STATIC int +xrep_symlink_reset_fork( + struct xfs_scrub *sc) +{ + struct xfs_ifork *ifp = xfs_ifork_ptr(sc->tempip, XFS_DATA_FORK); + int error; + + /* Unmap all the remote target buffers. */ + if (xfs_ifork_has_extents(ifp)) { + error = xrep_reap_ifork(sc, sc->tempip, XFS_DATA_FORK); + if (error) + return error; + } + + trace_xrep_symlink_reset_fork(sc->tempip); + + /* Reset the temp symlink target to dummy content. */ + xfs_idestroy_fork(ifp); + return xfs_symlink_write_target(sc->tp, sc->tempip, sc->tempip->i_ino, + "?", 1, 0, 0); +} + +/* + * Reinitialize a link target. Caller must ensure the inode is joined to + * the transaction. + */ +STATIC int +xrep_symlink_rebuild( + struct xfs_scrub *sc) +{ + struct xrep_tempexch *tx; + char *target_buf = sc->buf; + xfs_fsblock_t fs_blocks; + unsigned int target_len; + unsigned int resblks; + int error; + + /* How many blocks do we need? */ + target_len = strlen(target_buf); + ASSERT(target_len != 0); + if (target_len == 0 || target_len > XFS_SYMLINK_MAXLEN) + return -EFSCORRUPTED; + + trace_xrep_symlink_rebuild(sc->ip); + + /* + * In preparation to write the new symlink target to the temporary + * file, drop the ILOCK of the file being repaired (it shouldn't be + * joined) and take the ILOCK of the temporary file. + * + * The VFS does not take the IOLOCK while reading a symlink (and new + * symlinks are hidden with INEW until they've been written) so it's + * possible that a readlink() could see the old corrupted contents + * while we're doing this. + */ + xchk_iunlock(sc, XFS_ILOCK_EXCL); + xrep_tempfile_ilock(sc); + xfs_trans_ijoin(sc->tp, sc->tempip, 0); + + /* + * Reserve resources to reinitialize the target. We're allowed to + * exceed file quota to repair inconsistent metadata, though this is + * unlikely. + */ + fs_blocks = xfs_symlink_blocks(sc->mp, target_len); + resblks = XFS_SYMLINK_SPACE_RES(sc->mp, target_len, fs_blocks); + error = xfs_trans_reserve_quota_nblks(sc->tp, sc->tempip, resblks, 0, + true); + if (error) + return error; + + /* Erase the dummy target set up by the tempfile initialization. */ + xfs_idestroy_fork(&sc->tempip->i_df); + sc->tempip->i_df.if_bytes = 0; + sc->tempip->i_df.if_format = XFS_DINODE_FMT_EXTENTS; + + /* Write the salvaged target to the temporary link. */ + error = xfs_symlink_write_target(sc->tp, sc->tempip, sc->ip->i_ino, + target_buf, target_len, fs_blocks, resblks); + if (error) + return error; + + /* + * Commit the repair transaction so that we can use the atomic mapping + * exchange functions to compute the correct block reservations and + * re-lock the inodes. + */ + target_buf = NULL; + error = xrep_trans_commit(sc); + if (error) + return error; + + /* Last chance to abort before we start committing fixes. */ + if (xchk_should_terminate(sc, &error)) + return error; + + xrep_tempfile_iunlock(sc); + + /* + * We're done with the temporary buffer, so we can reuse it for the + * tempfile contents exchange information. + */ + tx = sc->buf; + error = xrep_tempexch_trans_alloc(sc, XFS_DATA_FORK, tx); + if (error) + return error; + + /* + * Exchange the temp link's data fork with the file being repaired. + * This recreates the transaction and takes the ILOCKs of the file + * being repaired and the temporary file. + */ + error = xrep_symlink_swap(sc); + if (error) + return error; + + /* + * Release the old symlink blocks and reset the data fork of the temp + * link to an empty shortform link. This is the last repair action we + * perform on the symlink, so we don't need to clean the transaction. + */ + return xrep_symlink_reset_fork(sc); +} + +/* Repair a symbolic link. */ +int +xrep_symlink( + struct xfs_scrub *sc) +{ + int error; + + /* The rmapbt is required to reap the old data fork. */ + if (!xfs_has_rmapbt(sc->mp)) + return -EOPNOTSUPP; + + ASSERT(sc->ilock_flags & XFS_ILOCK_EXCL); + + error = xrep_symlink_salvage(sc); + if (error) + return error; + + /* Now reset the target. */ + error = xrep_symlink_rebuild(sc); + if (error) + return error; + + return xrep_trans_commit(sc); +} diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index 4ca86a6a5be1..c72e447eb8ec 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -21,6 +21,7 @@ #include "xfs_exchrange.h" #include "xfs_exchmaps.h" #include "xfs_defer.h" +#include "xfs_symlink_remote.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/repair.h" @@ -109,6 +110,18 @@ xrep_tempfile_create( error = xfs_dir_init(tp, sc->tempip, dp); if (error) goto out_trans_cancel; + } else if (S_ISLNK(VFS_I(sc->tempip)->i_mode)) { + /* + * Initialize the temporary symlink with a meaningless target + * that won't trip the verifiers. Repair must rewrite the + * target with meaningful content before swapping with the file + * being repaired. A single-byte target will not write a + * remote target block, so the owner is irrelevant. + */ + error = xfs_symlink_write_target(tp, sc->tempip, + sc->tempip->i_ino, ".", 1, 0, 0); + if (error) + goto out_trans_cancel; } /* diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 668da6ff2ca2..03cb095fc1a1 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2711,6 +2711,52 @@ DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_alias); DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_dentry); DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_invalidate_child); +TRACE_EVENT(xrep_symlink_salvage_target, + TP_PROTO(struct xfs_inode *ip, char *target, unsigned int targetlen), + TP_ARGS(ip, target, targetlen), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(unsigned int, targetlen) + __dynamic_array(char, target, targetlen + 1) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->targetlen = targetlen; + memcpy(__get_str(target), target, targetlen); + __get_str(target)[targetlen] = 0; + ), + TP_printk("dev %d:%d ip 0x%llx target '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->targetlen, + __get_str(target)) +); + +DECLARE_EVENT_CLASS(xrep_symlink_class, + TP_PROTO(struct xfs_inode *ip), + TP_ARGS(ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + ), + TP_printk("dev %d:%d ip 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino) +); + +#define DEFINE_XREP_SYMLINK_EVENT(name) \ +DEFINE_EVENT(xrep_symlink_class, name, \ + TP_PROTO(struct xfs_inode *ip), \ + TP_ARGS(ip)) +DEFINE_XREP_SYMLINK_EVENT(xrep_symlink_rebuild); +DEFINE_XREP_SYMLINK_EVENT(xrep_symlink_reset_fork); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ -- cgit v1.2.3-59-g8ed1b From 10d587ecb77f33d966cfa2e8a57d51296df986db Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:00 -0700 Subject: xfs: check AGI unlinked inode buckets Look for corruptions in the AGI unlinked bucket chains. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/agheader.c | 40 ++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_inode.c | 2 +- fs/xfs/xfs_inode.h | 1 + 3 files changed, 42 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/agheader.c b/fs/xfs/scrub/agheader.c index e954f07679dd..1528f14bd925 100644 --- a/fs/xfs/scrub/agheader.c +++ b/fs/xfs/scrub/agheader.c @@ -15,6 +15,7 @@ #include "xfs_ialloc.h" #include "xfs_rmap.h" #include "xfs_ag.h" +#include "xfs_inode.h" #include "scrub/scrub.h" #include "scrub/common.h" @@ -865,6 +866,43 @@ xchk_agi_xref( /* scrub teardown will take care of sc->sa for us */ } +/* + * Check the unlinked buckets for links to bad inodes. We hold the AGI, so + * there cannot be any threads updating unlinked list pointers in this AG. + */ +STATIC void +xchk_iunlink( + struct xfs_scrub *sc, + struct xfs_agi *agi) +{ + unsigned int i; + struct xfs_inode *ip; + + for (i = 0; i < XFS_AGI_UNLINKED_BUCKETS; i++) { + xfs_agino_t agino = be32_to_cpu(agi->agi_unlinked[i]); + + while (agino != NULLAGINO) { + if (agino % XFS_AGI_UNLINKED_BUCKETS != i) { + xchk_block_set_corrupt(sc, sc->sa.agi_bp); + return; + } + + ip = xfs_iunlink_lookup(sc->sa.pag, agino); + if (!ip) { + xchk_block_set_corrupt(sc, sc->sa.agi_bp); + return; + } + + if (!xfs_inode_on_unlinked_list(ip)) { + xchk_block_set_corrupt(sc, sc->sa.agi_bp); + return; + } + + agino = ip->i_next_unlinked; + } + } +} + /* Scrub the AGI. */ int xchk_agi( @@ -949,6 +987,8 @@ xchk_agi( if (pag->pagi_freecount != be32_to_cpu(agi->agi_freecount)) xchk_block_set_corrupt(sc, sc->sa.agi_bp); + xchk_iunlink(sc, agi); + xchk_agi_xref(sc); out: return error; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 803a64687014..fed0cd6bffdf 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1985,7 +1985,7 @@ out: * only unlinked, referenced inodes can be on the unlinked inode list. If we * don't find the inode in cache, then let the caller handle the situation. */ -static struct xfs_inode * +struct xfs_inode * xfs_iunlink_lookup( struct xfs_perag *pag, xfs_agino_t agino) diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 18bc3d7750a0..c74c48bc0945 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -619,6 +619,7 @@ bool xfs_inode_needs_inactive(struct xfs_inode *ip); int xfs_iunlink(struct xfs_trans *tp, struct xfs_inode *ip); int xfs_iunlink_remove(struct xfs_trans *tp, struct xfs_perag *pag, struct xfs_inode *ip); +struct xfs_inode *xfs_iunlink_lookup(struct xfs_perag *pag, xfs_agino_t agino); void xfs_end_io(struct work_struct *work); -- cgit v1.2.3-59-g8ed1b From 5b57257025f97c643d502253055651c81b0ffc66 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:01 -0700 Subject: xfs: hoist AGI repair context to a heap object Save ~460 bytes of stack space by moving all the repair context to a heap object. We're going to add even more context data in the next patch, which is why we really need to do this now. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/agheader_repair.c | 105 ++++++++++++++++++++++++----------------- 1 file changed, 63 insertions(+), 42 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c index 427054b65b23..d210bd7d5eb1 100644 --- a/fs/xfs/scrub/agheader_repair.c +++ b/fs/xfs/scrub/agheader_repair.c @@ -796,15 +796,29 @@ enum { XREP_AGI_MAX }; +struct xrep_agi { + struct xfs_scrub *sc; + + /* AGI buffer, tracked separately */ + struct xfs_buf *agi_bp; + + /* context for finding btree roots */ + struct xrep_find_ag_btree fab[XREP_AGI_MAX]; + + /* old AGI contents in case we have to revert */ + struct xfs_agi old_agi; +}; + /* * Given the inode btree roots described by *fab, find the roots, check them * for sanity, and pass the root data back out via *fab. */ STATIC int xrep_agi_find_btrees( - struct xfs_scrub *sc, - struct xrep_find_ag_btree *fab) + struct xrep_agi *ragi) { + struct xfs_scrub *sc = ragi->sc; + struct xrep_find_ag_btree *fab = ragi->fab; struct xfs_buf *agf_bp; struct xfs_mount *mp = sc->mp; int error; @@ -837,10 +851,11 @@ xrep_agi_find_btrees( */ STATIC void xrep_agi_init_header( - struct xfs_scrub *sc, - struct xfs_buf *agi_bp, - struct xfs_agi *old_agi) + struct xrep_agi *ragi) { + struct xfs_scrub *sc = ragi->sc; + struct xfs_buf *agi_bp = ragi->agi_bp; + struct xfs_agi *old_agi = &ragi->old_agi; struct xfs_agi *agi = agi_bp->b_addr; struct xfs_perag *pag = sc->sa.pag; struct xfs_mount *mp = sc->mp; @@ -868,10 +883,12 @@ xrep_agi_init_header( /* Set btree root information in an AGI. */ STATIC void xrep_agi_set_roots( - struct xfs_scrub *sc, - struct xfs_agi *agi, - struct xrep_find_ag_btree *fab) + struct xrep_agi *ragi) { + struct xfs_scrub *sc = ragi->sc; + struct xfs_agi *agi = ragi->agi_bp->b_addr; + struct xrep_find_ag_btree *fab = ragi->fab; + agi->agi_root = cpu_to_be32(fab[XREP_AGI_INOBT].root); agi->agi_level = cpu_to_be32(fab[XREP_AGI_INOBT].height); @@ -884,9 +901,10 @@ xrep_agi_set_roots( /* Update the AGI counters. */ STATIC int xrep_agi_calc_from_btrees( - struct xfs_scrub *sc, - struct xfs_buf *agi_bp) + struct xrep_agi *ragi) { + struct xfs_scrub *sc = ragi->sc; + struct xfs_buf *agi_bp = ragi->agi_bp; struct xfs_btree_cur *cur; struct xfs_agi *agi = agi_bp->b_addr; struct xfs_mount *mp = sc->mp; @@ -931,9 +949,10 @@ err: /* Trigger reinitialization of the in-core data. */ STATIC int xrep_agi_commit_new( - struct xfs_scrub *sc, - struct xfs_buf *agi_bp) + struct xrep_agi *ragi) { + struct xfs_scrub *sc = ragi->sc; + struct xfs_buf *agi_bp = ragi->agi_bp; struct xfs_perag *pag; struct xfs_agi *agi = agi_bp->b_addr; @@ -956,33 +975,36 @@ xrep_agi_commit_new( /* Repair the AGI. */ int xrep_agi( - struct xfs_scrub *sc) + struct xfs_scrub *sc) { - struct xrep_find_ag_btree fab[XREP_AGI_MAX] = { - [XREP_AGI_INOBT] = { - .rmap_owner = XFS_RMAP_OWN_INOBT, - .buf_ops = &xfs_inobt_buf_ops, - .maxlevels = M_IGEO(sc->mp)->inobt_maxlevels, - }, - [XREP_AGI_FINOBT] = { - .rmap_owner = XFS_RMAP_OWN_INOBT, - .buf_ops = &xfs_finobt_buf_ops, - .maxlevels = M_IGEO(sc->mp)->inobt_maxlevels, - }, - [XREP_AGI_END] = { - .buf_ops = NULL - }, - }; - struct xfs_agi old_agi; - struct xfs_mount *mp = sc->mp; - struct xfs_buf *agi_bp; - struct xfs_agi *agi; - int error; + struct xrep_agi *ragi; + struct xfs_mount *mp = sc->mp; + int error; /* We require the rmapbt to rebuild anything. */ if (!xfs_has_rmapbt(mp)) return -EOPNOTSUPP; + sc->buf = kzalloc(sizeof(struct xrep_agi), XCHK_GFP_FLAGS); + if (!sc->buf) + return -ENOMEM; + ragi = sc->buf; + ragi->sc = sc; + + ragi->fab[XREP_AGI_INOBT] = (struct xrep_find_ag_btree){ + .rmap_owner = XFS_RMAP_OWN_INOBT, + .buf_ops = &xfs_inobt_buf_ops, + .maxlevels = M_IGEO(sc->mp)->inobt_maxlevels, + }; + ragi->fab[XREP_AGI_FINOBT] = (struct xrep_find_ag_btree){ + .rmap_owner = XFS_RMAP_OWN_INOBT, + .buf_ops = &xfs_finobt_buf_ops, + .maxlevels = M_IGEO(sc->mp)->inobt_maxlevels, + }; + ragi->fab[XREP_AGI_END] = (struct xrep_find_ag_btree){ + .buf_ops = NULL, + }; + /* * Make sure we have the AGI buffer, as scrub might have decided it * was corrupt after xfs_ialloc_read_agi failed with -EFSCORRUPTED. @@ -990,14 +1012,13 @@ xrep_agi( error = xfs_trans_read_buf(mp, sc->tp, mp->m_ddev_targp, XFS_AG_DADDR(mp, sc->sa.pag->pag_agno, XFS_AGI_DADDR(mp)), - XFS_FSS_TO_BB(mp, 1), 0, &agi_bp, NULL); + XFS_FSS_TO_BB(mp, 1), 0, &ragi->agi_bp, NULL); if (error) return error; - agi_bp->b_ops = &xfs_agi_buf_ops; - agi = agi_bp->b_addr; + ragi->agi_bp->b_ops = &xfs_agi_buf_ops; /* Find the AGI btree roots. */ - error = xrep_agi_find_btrees(sc, fab); + error = xrep_agi_find_btrees(ragi); if (error) return error; @@ -1006,18 +1027,18 @@ xrep_agi( return error; /* Start rewriting the header and implant the btrees we found. */ - xrep_agi_init_header(sc, agi_bp, &old_agi); - xrep_agi_set_roots(sc, agi, fab); - error = xrep_agi_calc_from_btrees(sc, agi_bp); + xrep_agi_init_header(ragi); + xrep_agi_set_roots(ragi); + error = xrep_agi_calc_from_btrees(ragi); if (error) goto out_revert; /* Reinitialize in-core state. */ - return xrep_agi_commit_new(sc, agi_bp); + return xrep_agi_commit_new(ragi); out_revert: /* Mark the incore AGI state stale and revert the AGI. */ clear_bit(XFS_AGSTATE_AGI_INIT, &sc->sa.pag->pag_opstate); - memcpy(agi, &old_agi, sizeof(old_agi)); + memcpy(ragi->agi_bp->b_addr, &ragi->old_agi, sizeof(struct xfs_agi)); return error; } -- cgit v1.2.3-59-g8ed1b From ab97f4b1c030750f2475bf4da8a9554d02206640 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:02 -0700 Subject: xfs: repair AGI unlinked inode bucket lists Teach the AGI repair code to rebuild the unlinked buckets and lists. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/agheader_repair.c | 774 ++++++++++++++++++++++++++++++++++++++++- fs/xfs/scrub/agino_bitmap.h | 49 +++ fs/xfs/scrub/trace.h | 255 ++++++++++++++ 3 files changed, 1074 insertions(+), 4 deletions(-) create mode 100644 fs/xfs/scrub/agino_bitmap.h (limited to 'fs') diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c index d210bd7d5eb1..0dbc484b182f 100644 --- a/fs/xfs/scrub/agheader_repair.c +++ b/fs/xfs/scrub/agheader_repair.c @@ -21,13 +21,18 @@ #include "xfs_rmap_btree.h" #include "xfs_refcount_btree.h" #include "xfs_ag.h" +#include "xfs_inode.h" +#include "xfs_iunlink_item.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" #include "scrub/repair.h" #include "scrub/bitmap.h" #include "scrub/agb_bitmap.h" +#include "scrub/agino_bitmap.h" #include "scrub/reap.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" /* Superblock */ @@ -796,6 +801,8 @@ enum { XREP_AGI_MAX }; +#define XREP_AGI_LOOKUP_BATCH 32 + struct xrep_agi { struct xfs_scrub *sc; @@ -807,8 +814,34 @@ struct xrep_agi { /* old AGI contents in case we have to revert */ struct xfs_agi old_agi; + + /* bitmap of which inodes are unlinked */ + struct xagino_bitmap iunlink_bmp; + + /* heads of the unlinked inode bucket lists */ + xfs_agino_t iunlink_heads[XFS_AGI_UNLINKED_BUCKETS]; + + /* scratchpad for batched lookups of the radix tree */ + struct xfs_inode *lookup_batch[XREP_AGI_LOOKUP_BATCH]; + + /* Map of ino -> next_ino for unlinked inode processing. */ + struct xfarray *iunlink_next; + + /* Map of ino -> prev_ino for unlinked inode processing. */ + struct xfarray *iunlink_prev; }; +static void +xrep_agi_buf_cleanup( + void *buf) +{ + struct xrep_agi *ragi = buf; + + xfarray_destroy(ragi->iunlink_prev); + xfarray_destroy(ragi->iunlink_next); + xagino_bitmap_destroy(&ragi->iunlink_bmp); +} + /* * Given the inode btree roots described by *fab, find the roots, check them * for sanity, and pass the root data back out via *fab. @@ -871,10 +904,6 @@ xrep_agi_init_header( if (xfs_has_crc(mp)) uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_meta_uuid); - /* We don't know how to fix the unlinked list yet. */ - memcpy(&agi->agi_unlinked, &old_agi->agi_unlinked, - sizeof(agi->agi_unlinked)); - /* Mark the incore AGF data stale until we're done fixing things. */ ASSERT(xfs_perag_initialised_agi(pag)); clear_bit(XFS_AGSTATE_AGI_INIT, &pag->pag_opstate); @@ -946,6 +975,714 @@ err: return error; } +/* + * Record a forwards unlinked chain pointer from agino -> next_agino in our + * staging information. + */ +static inline int +xrep_iunlink_store_next( + struct xrep_agi *ragi, + xfs_agino_t agino, + xfs_agino_t next_agino) +{ + ASSERT(next_agino != 0); + + return xfarray_store(ragi->iunlink_next, agino, &next_agino); +} + +/* + * Record a backwards unlinked chain pointer from prev_ino <- agino in our + * staging information. + */ +static inline int +xrep_iunlink_store_prev( + struct xrep_agi *ragi, + xfs_agino_t agino, + xfs_agino_t prev_agino) +{ + ASSERT(prev_agino != 0); + + return xfarray_store(ragi->iunlink_prev, agino, &prev_agino); +} + +/* + * Given an @agino, look up the next inode in the iunlink bucket. Returns + * NULLAGINO if we're at the end of the chain, 0 if @agino is not in memory + * like it should be, or a per-AG inode number. + */ +static inline xfs_agino_t +xrep_iunlink_next( + struct xfs_scrub *sc, + xfs_agino_t agino) +{ + struct xfs_inode *ip; + + ip = xfs_iunlink_lookup(sc->sa.pag, agino); + if (!ip) + return 0; + + return ip->i_next_unlinked; +} + +/* + * Load the inode @agino into memory, set its i_prev_unlinked, and drop the + * inode so it can be inactivated. Returns NULLAGINO if we're at the end of + * the chain or if we should stop walking the chain due to corruption; or a + * per-AG inode number. + */ +STATIC xfs_agino_t +xrep_iunlink_reload_next( + struct xrep_agi *ragi, + xfs_agino_t prev_agino, + xfs_agino_t agino) +{ + struct xfs_scrub *sc = ragi->sc; + struct xfs_inode *ip; + xfs_ino_t ino; + xfs_agino_t ret = NULLAGINO; + int error; + + ino = XFS_AGINO_TO_INO(sc->mp, sc->sa.pag->pag_agno, agino); + error = xchk_iget(ragi->sc, ino, &ip); + if (error) + return ret; + + trace_xrep_iunlink_reload_next(ip, prev_agino); + + /* If this is a linked inode, stop processing the chain. */ + if (VFS_I(ip)->i_nlink != 0) { + xrep_iunlink_store_next(ragi, agino, NULLAGINO); + goto rele; + } + + ip->i_prev_unlinked = prev_agino; + ret = ip->i_next_unlinked; + + /* + * Drop the inode reference that we just took. We hold the AGI, so + * this inode cannot move off the unlinked list and hence cannot be + * reclaimed. + */ +rele: + xchk_irele(sc, ip); + return ret; +} + +/* + * Walk an AGI unlinked bucket's list to load incore any unlinked inodes that + * still existed at mount time. This can happen if iunlink processing fails + * during log recovery. + */ +STATIC int +xrep_iunlink_walk_ondisk_bucket( + struct xrep_agi *ragi, + unsigned int bucket) +{ + struct xfs_scrub *sc = ragi->sc; + struct xfs_agi *agi = sc->sa.agi_bp->b_addr; + xfs_agino_t prev_agino = NULLAGINO; + xfs_agino_t next_agino; + int error = 0; + + next_agino = be32_to_cpu(agi->agi_unlinked[bucket]); + while (next_agino != NULLAGINO) { + xfs_agino_t agino = next_agino; + + if (xchk_should_terminate(ragi->sc, &error)) + return error; + + trace_xrep_iunlink_walk_ondisk_bucket(sc->sa.pag, bucket, + prev_agino, agino); + + if (bucket != agino % XFS_AGI_UNLINKED_BUCKETS) + break; + + next_agino = xrep_iunlink_next(sc, agino); + if (!next_agino) + next_agino = xrep_iunlink_reload_next(ragi, prev_agino, + agino); + + prev_agino = agino; + } + + return 0; +} + +/* Decide if this is an unlinked inode in this AG. */ +STATIC bool +xrep_iunlink_igrab( + struct xfs_perag *pag, + struct xfs_inode *ip) +{ + struct xfs_mount *mp = pag->pag_mount; + + if (XFS_INO_TO_AGNO(mp, ip->i_ino) != pag->pag_agno) + return false; + + if (!xfs_inode_on_unlinked_list(ip)) + return false; + + return true; +} + +/* + * Mark the given inode in the lookup batch in our unlinked inode bitmap, and + * remember if this inode is the start of the unlinked chain. + */ +STATIC int +xrep_iunlink_visit( + struct xrep_agi *ragi, + unsigned int batch_idx) +{ + struct xfs_mount *mp = ragi->sc->mp; + struct xfs_inode *ip = ragi->lookup_batch[batch_idx]; + xfs_agino_t agino; + unsigned int bucket; + int error; + + ASSERT(XFS_INO_TO_AGNO(mp, ip->i_ino) == ragi->sc->sa.pag->pag_agno); + ASSERT(xfs_inode_on_unlinked_list(ip)); + + agino = XFS_INO_TO_AGINO(mp, ip->i_ino); + bucket = agino % XFS_AGI_UNLINKED_BUCKETS; + + trace_xrep_iunlink_visit(ragi->sc->sa.pag, bucket, + ragi->iunlink_heads[bucket], ip); + + error = xagino_bitmap_set(&ragi->iunlink_bmp, agino, 1); + if (error) + return error; + + if (ip->i_prev_unlinked == NULLAGINO) { + if (ragi->iunlink_heads[bucket] == NULLAGINO) + ragi->iunlink_heads[bucket] = agino; + } + + return 0; +} + +/* + * Find all incore unlinked inodes so that we can rebuild the unlinked buckets. + * We hold the AGI so there should not be any modifications to the unlinked + * list. + */ +STATIC int +xrep_iunlink_mark_incore( + struct xrep_agi *ragi) +{ + struct xfs_perag *pag = ragi->sc->sa.pag; + struct xfs_mount *mp = pag->pag_mount; + uint32_t first_index = 0; + bool done = false; + unsigned int nr_found = 0; + + do { + unsigned int i; + int error = 0; + + if (xchk_should_terminate(ragi->sc, &error)) + return error; + + rcu_read_lock(); + + nr_found = radix_tree_gang_lookup(&pag->pag_ici_root, + (void **)&ragi->lookup_batch, first_index, + XREP_AGI_LOOKUP_BATCH); + if (!nr_found) { + rcu_read_unlock(); + return 0; + } + + for (i = 0; i < nr_found; i++) { + struct xfs_inode *ip = ragi->lookup_batch[i]; + + if (done || !xrep_iunlink_igrab(pag, ip)) + ragi->lookup_batch[i] = NULL; + + /* + * Update the index for the next lookup. Catch + * overflows into the next AG range which can occur if + * we have inodes in the last block of the AG and we + * are currently pointing to the last inode. + * + * Because we may see inodes that are from the wrong AG + * due to RCU freeing and reallocation, only update the + * index if it lies in this AG. It was a race that lead + * us to see this inode, so another lookup from the + * same index will not find it again. + */ + if (XFS_INO_TO_AGNO(mp, ip->i_ino) != pag->pag_agno) + continue; + first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1); + if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) + done = true; + } + + /* unlock now we've grabbed the inodes. */ + rcu_read_unlock(); + + for (i = 0; i < nr_found; i++) { + if (!ragi->lookup_batch[i]) + continue; + error = xrep_iunlink_visit(ragi, i); + if (error) + return error; + } + } while (!done); + + return 0; +} + +/* Mark all the unlinked ondisk inodes in this inobt record in iunlink_bmp. */ +STATIC int +xrep_iunlink_mark_ondisk_rec( + struct xfs_btree_cur *cur, + const union xfs_btree_rec *rec, + void *priv) +{ + struct xfs_inobt_rec_incore irec; + struct xrep_agi *ragi = priv; + struct xfs_scrub *sc = ragi->sc; + struct xfs_mount *mp = cur->bc_mp; + xfs_agino_t agino; + unsigned int i; + int error = 0; + + xfs_inobt_btrec_to_irec(mp, rec, &irec); + + for (i = 0, agino = irec.ir_startino; + i < XFS_INODES_PER_CHUNK; + i++, agino++) { + struct xfs_inode *ip; + unsigned int len = 1; + + /* Skip free inodes */ + if (XFS_INOBT_MASK(i) & irec.ir_free) + continue; + /* Skip inodes we've seen before */ + if (xagino_bitmap_test(&ragi->iunlink_bmp, agino, &len)) + continue; + + /* + * Skip incore inodes; these were already picked up by + * the _mark_incore step. + */ + rcu_read_lock(); + ip = radix_tree_lookup(&sc->sa.pag->pag_ici_root, agino); + rcu_read_unlock(); + if (ip) + continue; + + /* + * Try to look up this inode. If we can't get it, just move + * on because we haven't actually scrubbed the inobt or the + * inodes yet. + */ + error = xchk_iget(ragi->sc, + XFS_AGINO_TO_INO(mp, sc->sa.pag->pag_agno, + agino), + &ip); + if (error) + continue; + + trace_xrep_iunlink_reload_ondisk(ip); + + if (VFS_I(ip)->i_nlink == 0) + error = xagino_bitmap_set(&ragi->iunlink_bmp, agino, 1); + xchk_irele(sc, ip); + if (error) + break; + } + + return error; +} + +/* + * Find ondisk inodes that are unlinked and not in cache, and mark them in + * iunlink_bmp. We haven't checked the inobt yet, so we don't error out if + * the btree is corrupt. + */ +STATIC void +xrep_iunlink_mark_ondisk( + struct xrep_agi *ragi) +{ + struct xfs_scrub *sc = ragi->sc; + struct xfs_buf *agi_bp = ragi->agi_bp; + struct xfs_btree_cur *cur; + int error; + + cur = xfs_inobt_init_cursor(sc->sa.pag, sc->tp, agi_bp); + error = xfs_btree_query_all(cur, xrep_iunlink_mark_ondisk_rec, ragi); + xfs_btree_del_cursor(cur, error); +} + +/* + * Walk an iunlink bucket's inode list. For each inode that should be on this + * chain, clear its entry in in iunlink_bmp because it's ok and we don't need + * to touch it further. + */ +STATIC int +xrep_iunlink_resolve_bucket( + struct xrep_agi *ragi, + unsigned int bucket) +{ + struct xfs_scrub *sc = ragi->sc; + struct xfs_inode *ip; + xfs_agino_t prev_agino = NULLAGINO; + xfs_agino_t next_agino = ragi->iunlink_heads[bucket]; + int error = 0; + + while (next_agino != NULLAGINO) { + if (xchk_should_terminate(ragi->sc, &error)) + return error; + + /* Find the next inode in the chain. */ + ip = xfs_iunlink_lookup(sc->sa.pag, next_agino); + if (!ip) { + /* Inode not incore? Terminate the chain. */ + trace_xrep_iunlink_resolve_uncached(sc->sa.pag, + bucket, prev_agino, next_agino); + + next_agino = NULLAGINO; + break; + } + + if (next_agino % XFS_AGI_UNLINKED_BUCKETS != bucket) { + /* + * Inode is in the wrong bucket. Advance the list, + * but pretend we didn't see this inode. + */ + trace_xrep_iunlink_resolve_wronglist(sc->sa.pag, + bucket, prev_agino, next_agino); + + next_agino = ip->i_next_unlinked; + continue; + } + + if (!xfs_inode_on_unlinked_list(ip)) { + /* + * Incore inode doesn't think this inode is on an + * unlinked list. This is probably because we reloaded + * it from disk. Advance the list, but pretend we + * didn't see this inode; we'll fix that later. + */ + trace_xrep_iunlink_resolve_nolist(sc->sa.pag, + bucket, prev_agino, next_agino); + next_agino = ip->i_next_unlinked; + continue; + } + + trace_xrep_iunlink_resolve_ok(sc->sa.pag, bucket, prev_agino, + next_agino); + + /* + * Otherwise, this inode's unlinked pointers are ok. Clear it + * from the unlinked bitmap since we're done with it, and make + * sure the chain is still correct. + */ + error = xagino_bitmap_clear(&ragi->iunlink_bmp, next_agino, 1); + if (error) + return error; + + /* Remember the previous inode's next pointer. */ + if (prev_agino != NULLAGINO) { + error = xrep_iunlink_store_next(ragi, prev_agino, + next_agino); + if (error) + return error; + } + + /* Remember this inode's previous pointer. */ + error = xrep_iunlink_store_prev(ragi, next_agino, prev_agino); + if (error) + return error; + + /* Advance the list and remember this inode. */ + prev_agino = next_agino; + next_agino = ip->i_next_unlinked; + } + + /* Update the previous inode's next pointer. */ + if (prev_agino != NULLAGINO) { + error = xrep_iunlink_store_next(ragi, prev_agino, next_agino); + if (error) + return error; + } + + return 0; +} + +/* Reinsert this unlinked inode into the head of the staged bucket list. */ +STATIC int +xrep_iunlink_add_to_bucket( + struct xrep_agi *ragi, + xfs_agino_t agino) +{ + xfs_agino_t current_head; + unsigned int bucket; + int error; + + bucket = agino % XFS_AGI_UNLINKED_BUCKETS; + + /* Point this inode at the current head of the bucket list. */ + current_head = ragi->iunlink_heads[bucket]; + + trace_xrep_iunlink_add_to_bucket(ragi->sc->sa.pag, bucket, agino, + current_head); + + error = xrep_iunlink_store_next(ragi, agino, current_head); + if (error) + return error; + + /* Remember the head inode's previous pointer. */ + if (current_head != NULLAGINO) { + error = xrep_iunlink_store_prev(ragi, current_head, agino); + if (error) + return error; + } + + ragi->iunlink_heads[bucket] = agino; + return 0; +} + +/* Reinsert unlinked inodes into the staged iunlink buckets. */ +STATIC int +xrep_iunlink_add_lost_inodes( + uint32_t start, + uint32_t len, + void *priv) +{ + struct xrep_agi *ragi = priv; + int error; + + for (; len > 0; start++, len--) { + error = xrep_iunlink_add_to_bucket(ragi, start); + if (error) + return error; + } + + return 0; +} + +/* + * Figure out the iunlink bucket values and find inodes that need to be + * reinserted into the list. + */ +STATIC int +xrep_iunlink_rebuild_buckets( + struct xrep_agi *ragi) +{ + unsigned int i; + int error; + + /* + * Walk the ondisk AGI unlinked list to find inodes that are on the + * list but aren't in memory. This can happen if a past log recovery + * tried to clear the iunlinked list but failed. Our scan rebuilds the + * unlinked list using incore inodes, so we must load and link them + * properly. + */ + for (i = 0; i < XFS_AGI_UNLINKED_BUCKETS; i++) { + error = xrep_iunlink_walk_ondisk_bucket(ragi, i); + if (error) + return error; + } + + /* + * Record all the incore unlinked inodes in iunlink_bmp that we didn't + * find by walking the ondisk iunlink buckets. This shouldn't happen, + * but we can't risk forgetting an inode somewhere. + */ + error = xrep_iunlink_mark_incore(ragi); + if (error) + return error; + + /* + * If there are ondisk inodes that are unlinked and are not been loaded + * into cache, record them in iunlink_bmp. + */ + xrep_iunlink_mark_ondisk(ragi); + + /* + * Walk each iunlink bucket to (re)construct as much of the incore list + * as would be correct. For each inode that survives this step, mark + * it clear in iunlink_bmp; we're done with those inodes. + */ + for (i = 0; i < XFS_AGI_UNLINKED_BUCKETS; i++) { + error = xrep_iunlink_resolve_bucket(ragi, i); + if (error) + return error; + } + + /* + * Any unlinked inodes that we didn't find through the bucket list + * walk (or was ignored by the walk) must be inserted into the bucket + * list. Stage this in memory for now. + */ + return xagino_bitmap_walk(&ragi->iunlink_bmp, + xrep_iunlink_add_lost_inodes, ragi); +} + +/* Update i_next_iunlinked for the inode @agino. */ +STATIC int +xrep_iunlink_relink_next( + struct xrep_agi *ragi, + xfarray_idx_t idx, + xfs_agino_t next_agino) +{ + struct xfs_scrub *sc = ragi->sc; + struct xfs_perag *pag = sc->sa.pag; + struct xfs_inode *ip; + xfarray_idx_t agino = idx - 1; + bool want_rele = false; + int error = 0; + + ip = xfs_iunlink_lookup(pag, agino); + if (!ip) { + xfs_ino_t ino; + xfs_agino_t prev_agino; + + /* + * No inode exists in cache. Load it off the disk so that we + * can reinsert it into the incore unlinked list. + */ + ino = XFS_AGINO_TO_INO(sc->mp, pag->pag_agno, agino); + error = xchk_iget(sc, ino, &ip); + if (error) + return -EFSCORRUPTED; + + want_rele = true; + + /* Set the backward pointer since this just came off disk. */ + error = xfarray_load(ragi->iunlink_prev, agino, &prev_agino); + if (error) + goto out_rele; + + trace_xrep_iunlink_relink_prev(ip, prev_agino); + ip->i_prev_unlinked = prev_agino; + } + + /* Update the forward pointer. */ + if (ip->i_next_unlinked != next_agino) { + error = xfs_iunlink_log_inode(sc->tp, ip, pag, next_agino); + if (error) + goto out_rele; + + trace_xrep_iunlink_relink_next(ip, next_agino); + ip->i_next_unlinked = next_agino; + } + +out_rele: + /* + * The iunlink lookup doesn't igrab because we hold the AGI buffer lock + * and the inode cannot be reclaimed. However, if we used iget to load + * a missing inode, we must irele it here. + */ + if (want_rele) + xchk_irele(sc, ip); + return error; +} + +/* Update i_prev_iunlinked for the inode @agino. */ +STATIC int +xrep_iunlink_relink_prev( + struct xrep_agi *ragi, + xfarray_idx_t idx, + xfs_agino_t prev_agino) +{ + struct xfs_scrub *sc = ragi->sc; + struct xfs_perag *pag = sc->sa.pag; + struct xfs_inode *ip; + xfarray_idx_t agino = idx - 1; + bool want_rele = false; + int error = 0; + + ASSERT(prev_agino != 0); + + ip = xfs_iunlink_lookup(pag, agino); + if (!ip) { + xfs_ino_t ino; + xfs_agino_t next_agino; + + /* + * No inode exists in cache. Load it off the disk so that we + * can reinsert it into the incore unlinked list. + */ + ino = XFS_AGINO_TO_INO(sc->mp, pag->pag_agno, agino); + error = xchk_iget(sc, ino, &ip); + if (error) + return -EFSCORRUPTED; + + want_rele = true; + + /* Set the forward pointer since this just came off disk. */ + error = xfarray_load(ragi->iunlink_prev, agino, &next_agino); + if (error) + goto out_rele; + + error = xfs_iunlink_log_inode(sc->tp, ip, pag, next_agino); + if (error) + goto out_rele; + + trace_xrep_iunlink_relink_next(ip, next_agino); + ip->i_next_unlinked = next_agino; + } + + /* Update the backward pointer. */ + if (ip->i_prev_unlinked != prev_agino) { + trace_xrep_iunlink_relink_prev(ip, prev_agino); + ip->i_prev_unlinked = prev_agino; + } + +out_rele: + /* + * The iunlink lookup doesn't igrab because we hold the AGI buffer lock + * and the inode cannot be reclaimed. However, if we used iget to load + * a missing inode, we must irele it here. + */ + if (want_rele) + xchk_irele(sc, ip); + return error; +} + +/* Log all the iunlink updates we need to finish regenerating the AGI. */ +STATIC int +xrep_iunlink_commit( + struct xrep_agi *ragi) +{ + struct xfs_agi *agi = ragi->agi_bp->b_addr; + xfarray_idx_t idx = XFARRAY_CURSOR_INIT; + xfs_agino_t agino; + unsigned int i; + int error; + + /* Fix all the forward links */ + while ((error = xfarray_iter(ragi->iunlink_next, &idx, &agino)) == 1) { + error = xrep_iunlink_relink_next(ragi, idx, agino); + if (error) + return error; + } + + /* Fix all the back links */ + idx = XFARRAY_CURSOR_INIT; + while ((error = xfarray_iter(ragi->iunlink_prev, &idx, &agino)) == 1) { + error = xrep_iunlink_relink_prev(ragi, idx, agino); + if (error) + return error; + } + + /* Copy the staged iunlink buckets to the new AGI. */ + for (i = 0; i < XFS_AGI_UNLINKED_BUCKETS; i++) { + trace_xrep_iunlink_commit_bucket(ragi->sc->sa.pag, i, + be32_to_cpu(ragi->old_agi.agi_unlinked[i]), + ragi->iunlink_heads[i]); + + agi->agi_unlinked[i] = cpu_to_be32(ragi->iunlink_heads[i]); + } + + return 0; +} + /* Trigger reinitialization of the in-core data. */ STATIC int xrep_agi_commit_new( @@ -979,6 +1716,8 @@ xrep_agi( { struct xrep_agi *ragi; struct xfs_mount *mp = sc->mp; + char *descr; + unsigned int i; int error; /* We require the rmapbt to rebuild anything. */ @@ -1005,6 +1744,26 @@ xrep_agi( .buf_ops = NULL, }; + for (i = 0; i < XFS_AGI_UNLINKED_BUCKETS; i++) + ragi->iunlink_heads[i] = NULLAGINO; + + xagino_bitmap_init(&ragi->iunlink_bmp); + sc->buf_cleanup = xrep_agi_buf_cleanup; + + descr = xchk_xfile_ag_descr(sc, "iunlinked next pointers"); + error = xfarray_create(descr, 0, sizeof(xfs_agino_t), + &ragi->iunlink_next); + kfree(descr); + if (error) + return error; + + descr = xchk_xfile_ag_descr(sc, "iunlinked prev pointers"); + error = xfarray_create(descr, 0, sizeof(xfs_agino_t), + &ragi->iunlink_prev); + kfree(descr); + if (error) + return error; + /* * Make sure we have the AGI buffer, as scrub might have decided it * was corrupt after xfs_ialloc_read_agi failed with -EFSCORRUPTED. @@ -1022,6 +1781,10 @@ xrep_agi( if (error) return error; + error = xrep_iunlink_rebuild_buckets(ragi); + if (error) + return error; + /* Last chance to abort before we start committing fixes. */ if (xchk_should_terminate(sc, &error)) return error; @@ -1030,6 +1793,9 @@ xrep_agi( xrep_agi_init_header(ragi); xrep_agi_set_roots(ragi); error = xrep_agi_calc_from_btrees(ragi); + if (error) + goto out_revert; + error = xrep_iunlink_commit(ragi); if (error) goto out_revert; diff --git a/fs/xfs/scrub/agino_bitmap.h b/fs/xfs/scrub/agino_bitmap.h new file mode 100644 index 000000000000..56d7db5f1699 --- /dev/null +++ b/fs/xfs/scrub/agino_bitmap.h @@ -0,0 +1,49 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2018-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_AGINO_BITMAP_H__ +#define __XFS_SCRUB_AGINO_BITMAP_H__ + +/* Bitmaps, but for type-checked for xfs_agino_t */ + +struct xagino_bitmap { + struct xbitmap32 aginobitmap; +}; + +static inline void xagino_bitmap_init(struct xagino_bitmap *bitmap) +{ + xbitmap32_init(&bitmap->aginobitmap); +} + +static inline void xagino_bitmap_destroy(struct xagino_bitmap *bitmap) +{ + xbitmap32_destroy(&bitmap->aginobitmap); +} + +static inline int xagino_bitmap_clear(struct xagino_bitmap *bitmap, + xfs_agino_t agino, unsigned int len) +{ + return xbitmap32_clear(&bitmap->aginobitmap, agino, len); +} + +static inline int xagino_bitmap_set(struct xagino_bitmap *bitmap, + xfs_agino_t agino, unsigned int len) +{ + return xbitmap32_set(&bitmap->aginobitmap, agino, len); +} + +static inline bool xagino_bitmap_test(struct xagino_bitmap *bitmap, + xfs_agino_t agino, unsigned int *len) +{ + return xbitmap32_test(&bitmap->aginobitmap, agino, len); +} + +static inline int xagino_bitmap_walk(struct xagino_bitmap *bitmap, + xbitmap32_walk_fn fn, void *priv) +{ + return xbitmap32_walk(&bitmap->aginobitmap, fn, priv); +} + +#endif /* __XFS_SCRUB_AGINO_BITMAP_H__ */ diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 03cb095fc1a1..814db1d1747a 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2757,6 +2757,261 @@ DEFINE_EVENT(xrep_symlink_class, name, \ DEFINE_XREP_SYMLINK_EVENT(xrep_symlink_rebuild); DEFINE_XREP_SYMLINK_EVENT(xrep_symlink_reset_fork); +TRACE_EVENT(xrep_iunlink_visit, + TP_PROTO(struct xfs_perag *pag, unsigned int bucket, + xfs_agino_t bucket_agino, struct xfs_inode *ip), + TP_ARGS(pag, bucket, bucket_agino, ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agino_t, agino) + __field(unsigned int, bucket) + __field(xfs_agino_t, bucket_agino) + __field(xfs_agino_t, prev_agino) + __field(xfs_agino_t, next_agino) + ), + TP_fast_assign( + __entry->dev = pag->pag_mount->m_super->s_dev; + __entry->agno = pag->pag_agno; + __entry->agino = XFS_INO_TO_AGINO(pag->pag_mount, ip->i_ino); + __entry->bucket = bucket; + __entry->bucket_agino = bucket_agino; + __entry->prev_agino = ip->i_prev_unlinked; + __entry->next_agino = ip->i_next_unlinked; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u agino 0x%x bucket_agino 0x%x prev_agino 0x%x next_agino 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->bucket, + __entry->agino, + __entry->bucket_agino, + __entry->prev_agino, + __entry->next_agino) +); + +TRACE_EVENT(xrep_iunlink_reload_next, + TP_PROTO(struct xfs_inode *ip, xfs_agino_t prev_agino), + TP_ARGS(ip, prev_agino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agino_t, agino) + __field(xfs_agino_t, old_prev_agino) + __field(xfs_agino_t, prev_agino) + __field(xfs_agino_t, next_agino) + __field(unsigned int, nlink) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->agno = XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino); + __entry->agino = XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino); + __entry->old_prev_agino = ip->i_prev_unlinked; + __entry->prev_agino = prev_agino; + __entry->next_agino = ip->i_next_unlinked; + __entry->nlink = VFS_I(ip)->i_nlink; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u agino 0x%x nlink %u old_prev_agino %u prev_agino 0x%x next_agino 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agino % XFS_AGI_UNLINKED_BUCKETS, + __entry->agino, + __entry->nlink, + __entry->old_prev_agino, + __entry->prev_agino, + __entry->next_agino) +); + +TRACE_EVENT(xrep_iunlink_reload_ondisk, + TP_PROTO(struct xfs_inode *ip), + TP_ARGS(ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agino_t, agino) + __field(unsigned int, nlink) + __field(xfs_agino_t, next_agino) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->agno = XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino); + __entry->agino = XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino); + __entry->nlink = VFS_I(ip)->i_nlink; + __entry->next_agino = ip->i_next_unlinked; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u agino 0x%x nlink %u next_agino 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agino % XFS_AGI_UNLINKED_BUCKETS, + __entry->agino, + __entry->nlink, + __entry->next_agino) +); + +TRACE_EVENT(xrep_iunlink_walk_ondisk_bucket, + TP_PROTO(struct xfs_perag *pag, unsigned int bucket, + xfs_agino_t prev_agino, xfs_agino_t next_agino), + TP_ARGS(pag, bucket, prev_agino, next_agino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(unsigned int, bucket) + __field(xfs_agino_t, prev_agino) + __field(xfs_agino_t, next_agino) + ), + TP_fast_assign( + __entry->dev = pag->pag_mount->m_super->s_dev; + __entry->agno = pag->pag_agno; + __entry->bucket = bucket; + __entry->prev_agino = prev_agino; + __entry->next_agino = next_agino; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u prev_agino 0x%x next_agino 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->bucket, + __entry->prev_agino, + __entry->next_agino) +); + +DECLARE_EVENT_CLASS(xrep_iunlink_resolve_class, + TP_PROTO(struct xfs_perag *pag, unsigned int bucket, + xfs_agino_t prev_agino, xfs_agino_t next_agino), + TP_ARGS(pag, bucket, prev_agino, next_agino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(unsigned int, bucket) + __field(xfs_agino_t, prev_agino) + __field(xfs_agino_t, next_agino) + ), + TP_fast_assign( + __entry->dev = pag->pag_mount->m_super->s_dev; + __entry->agno = pag->pag_agno; + __entry->bucket = bucket; + __entry->prev_agino = prev_agino; + __entry->next_agino = next_agino; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u prev_agino 0x%x next_agino 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->bucket, + __entry->prev_agino, + __entry->next_agino) +); +#define DEFINE_REPAIR_IUNLINK_RESOLVE_EVENT(name) \ +DEFINE_EVENT(xrep_iunlink_resolve_class, name, \ + TP_PROTO(struct xfs_perag *pag, unsigned int bucket, \ + xfs_agino_t prev_agino, xfs_agino_t next_agino), \ + TP_ARGS(pag, bucket, prev_agino, next_agino)) +DEFINE_REPAIR_IUNLINK_RESOLVE_EVENT(xrep_iunlink_resolve_uncached); +DEFINE_REPAIR_IUNLINK_RESOLVE_EVENT(xrep_iunlink_resolve_wronglist); +DEFINE_REPAIR_IUNLINK_RESOLVE_EVENT(xrep_iunlink_resolve_nolist); +DEFINE_REPAIR_IUNLINK_RESOLVE_EVENT(xrep_iunlink_resolve_ok); + +TRACE_EVENT(xrep_iunlink_relink_next, + TP_PROTO(struct xfs_inode *ip, xfs_agino_t next_agino), + TP_ARGS(ip, next_agino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agino_t, agino) + __field(xfs_agino_t, next_agino) + __field(xfs_agino_t, new_next_agino) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->agno = XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino); + __entry->agino = XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino); + __entry->next_agino = ip->i_next_unlinked; + __entry->new_next_agino = next_agino; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u agino 0x%x next_agino 0x%x -> 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agino % XFS_AGI_UNLINKED_BUCKETS, + __entry->agino, + __entry->next_agino, + __entry->new_next_agino) +); + +TRACE_EVENT(xrep_iunlink_relink_prev, + TP_PROTO(struct xfs_inode *ip, xfs_agino_t prev_agino), + TP_ARGS(ip, prev_agino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agino_t, agino) + __field(xfs_agino_t, prev_agino) + __field(xfs_agino_t, new_prev_agino) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->agno = XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino); + __entry->agino = XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino); + __entry->prev_agino = ip->i_prev_unlinked; + __entry->new_prev_agino = prev_agino; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u agino 0x%x prev_agino 0x%x -> 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agino % XFS_AGI_UNLINKED_BUCKETS, + __entry->agino, + __entry->prev_agino, + __entry->new_prev_agino) +); + +TRACE_EVENT(xrep_iunlink_add_to_bucket, + TP_PROTO(struct xfs_perag *pag, unsigned int bucket, + xfs_agino_t agino, xfs_agino_t curr_head), + TP_ARGS(pag, bucket, agino, curr_head), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(unsigned int, bucket) + __field(xfs_agino_t, agino) + __field(xfs_agino_t, next_agino) + ), + TP_fast_assign( + __entry->dev = pag->pag_mount->m_super->s_dev; + __entry->agno = pag->pag_agno; + __entry->bucket = bucket; + __entry->agino = agino; + __entry->next_agino = curr_head; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u agino 0x%x next_agino 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->bucket, + __entry->agino, + __entry->next_agino) +); + +TRACE_EVENT(xrep_iunlink_commit_bucket, + TP_PROTO(struct xfs_perag *pag, unsigned int bucket, + xfs_agino_t old_agino, xfs_agino_t agino), + TP_ARGS(pag, bucket, old_agino, agino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(unsigned int, bucket) + __field(xfs_agino_t, old_agino) + __field(xfs_agino_t, agino) + ), + TP_fast_assign( + __entry->dev = pag->pag_mount->m_super->s_dev; + __entry->agno = pag->pag_agno; + __entry->bucket = bucket; + __entry->old_agino = old_agino; + __entry->agino = agino; + ), + TP_printk("dev %d:%d agno 0x%x bucket %u agino 0x%x -> 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->bucket, + __entry->old_agino, + __entry->agino) +); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ -- cgit v1.2.3-59-g8ed1b From 40cb8613d6122ca80a9e42e4cecc4d308c3b80fb Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:03 -0700 Subject: xfs: check unused nlink fields in the ondisk inode v2/v3 inodes use di_nlink and not di_onlink; and v1 inodes use di_onlink and not di_nlink. Whichever field is not in use, make sure its contents are zero, and teach xfs_scrub to fix that if it is. This clears a bunch of missing scrub failure errors in xfs/385 for core.onlink. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_inode_buf.c | 8 ++++++++ fs/xfs/scrub/inode_repair.c | 12 ++++++++++++ 2 files changed, 20 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c index d0dcce462bf4..d79002343d0b 100644 --- a/fs/xfs/libxfs/xfs_inode_buf.c +++ b/fs/xfs/libxfs/xfs_inode_buf.c @@ -491,6 +491,14 @@ xfs_dinode_verify( return __this_address; } + if (dip->di_version > 1) { + if (dip->di_onlink) + return __this_address; + } else { + if (dip->di_nlink) + return __this_address; + } + /* don't allow invalid i_size */ di_size = be64_to_cpu(dip->di_size); if (di_size & (1ULL << 63)) diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c index 0dde5df2f8d3..e3b74ea50fde 100644 --- a/fs/xfs/scrub/inode_repair.c +++ b/fs/xfs/scrub/inode_repair.c @@ -516,6 +516,17 @@ xrep_dinode_mode( return 0; } +/* Fix unused link count fields having nonzero values. */ +STATIC void +xrep_dinode_nlinks( + struct xfs_dinode *dip) +{ + if (dip->di_version > 1) + dip->di_onlink = 0; + else + dip->di_nlink = 0; +} + /* Fix any conflicting flags that the verifiers complain about. */ STATIC void xrep_dinode_flags( @@ -1377,6 +1388,7 @@ xrep_dinode_core( iget_error = xrep_dinode_mode(ri, dip); if (iget_error) goto write; + xrep_dinode_nlinks(dip); xrep_dinode_flags(sc, dip, ri->rt_extents > 0); xrep_dinode_size(ri, dip); xrep_dinode_extsize_hints(sc, dip); -- cgit v1.2.3-59-g8ed1b From 2935213a6831a0087442d406301c2cdcc87b0f61 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:04 -0700 Subject: xfs: try to avoid allocating from sick inode clusters I noticed that xfs/413 and xfs/375 occasionally failed while fuzzing core.mode of an inode. The root cause of these problems is that the field we fuzzed (core.mode or core.magic, typically) causes the entire inode cluster buffer verification to fail, which affects several inodes at once. The repair process tries to create either a /lost+found or a temporary repair file, but regrettably it picks the same inode cluster that we just corrupted, with the result that repair triggers the demise of the filesystem. Try avoid this by making the inode allocation path detect when the perag health status indicates that someone has found bad inode cluster buffers, and try to read the inode cluster buffer. If the cluster buffer fails the verifiers, try another AG. This isn't foolproof and can result in premature ENOSPC, but that might be better than shutting down. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_ialloc.c | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index cb37f0007731..14c81f227c5b 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -1057,6 +1057,33 @@ xfs_inobt_first_free_inode( return xfs_lowbit64(realfree); } +/* + * If this AG has corrupt inodes, check if allocating this inode would fail + * with corruption errors. Returns 0 if we're clear, or EAGAIN to try again + * somewhere else. + */ +static int +xfs_dialloc_check_ino( + struct xfs_perag *pag, + struct xfs_trans *tp, + xfs_ino_t ino) +{ + struct xfs_imap imap; + struct xfs_buf *bp; + int error; + + error = xfs_imap(pag, tp, ino, &imap, 0); + if (error) + return -EAGAIN; + + error = xfs_imap_to_bp(pag->pag_mount, tp, &imap, &bp); + if (error) + return -EAGAIN; + + xfs_trans_brelse(tp, bp); + return 0; +} + /* * Allocate an inode using the inobt-only algorithm. */ @@ -1309,6 +1336,13 @@ alloc_inode: ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) % XFS_INODES_PER_CHUNK) == 0); ino = XFS_AGINO_TO_INO(mp, pag->pag_agno, rec.ir_startino + offset); + + if (xfs_ag_has_sickness(pag, XFS_SICK_AG_INODES)) { + error = xfs_dialloc_check_ino(pag, tp, ino); + if (error) + goto error0; + } + rec.ir_free &= ~XFS_INOBT_MASK(offset); rec.ir_freecount--; error = xfs_inobt_update(cur, &rec); @@ -1584,6 +1618,12 @@ xfs_dialloc_ag( XFS_INODES_PER_CHUNK) == 0); ino = XFS_AGINO_TO_INO(mp, pag->pag_agno, rec.ir_startino + offset); + if (xfs_ag_has_sickness(pag, XFS_SICK_AG_INODES)) { + error = xfs_dialloc_check_ino(pag, tp, ino); + if (error) + goto error_cur; + } + /* * Modify or remove the finobt record. */ -- cgit v1.2.3-59-g8ed1b From 5f204051d998ec3d7306db0d749bddcbf4c97693 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:05 -0700 Subject: xfs: pin inodes that would otherwise overflow link count The VFS inc_nlink function does not explicitly check for integer overflows in the i_nlink field. Instead, it checks the link count against s_max_links in the vfs_{link,create,rename} functions. XFS sets the maximum link count to 2.1 billion, so integer overflows should not be a problem. However. It's possible that online repair could find that a file has more than four billion links, particularly if the link count got corrupted while creating hardlinks to the file. The di_nlinkv2 field is not large enough to store a value larger than 2^32, so we ought to define a magic pin value of ~0U which means that the inode never gets deleted. This will prevent a UAF error if the repair finds this situation and users begin deleting links to the file. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_format.h | 6 ++++++ fs/xfs/scrub/dir_repair.c | 11 +++-------- fs/xfs/scrub/nlinks.c | 4 +++- fs/xfs/scrub/nlinks_repair.c | 8 ++------ fs/xfs/xfs_inode.c | 33 ++++++++++++++++++++++----------- 5 files changed, 36 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index 10153ce116d4..f1818c54af6f 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -899,6 +899,12 @@ static inline uint xfs_dinode_size(int version) */ #define XFS_MAXLINK ((1U << 31) - 1U) +/* + * Any file that hits the maximum ondisk link count should be pinned to avoid + * a use-after-free situation. + */ +#define XFS_NLINK_PINNED (~0U) + /* * Values for di_format * diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index c150b2efa2c2..38957da26b94 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -1145,7 +1145,9 @@ xrep_dir_set_nlink( struct xfs_scrub *sc = rd->sc; struct xfs_inode *dp = sc->ip; struct xfs_perag *pag; - unsigned int new_nlink = rd->subdirs + 2; + unsigned int new_nlink = min_t(unsigned long long, + rd->subdirs + 2, + XFS_NLINK_PINNED); int error; /* @@ -1201,13 +1203,6 @@ xrep_dir_swap( bool ip_local, temp_local; int error = 0; - /* - * If we found enough subdirs to overflow this directory's link count, - * bail out to userspace before we modify anything. - */ - if (rd->subdirs + 2 > XFS_MAXLINK) - return -EFSCORRUPTED; - /* * If we never found the parent for this directory, temporarily assign * the root dir as the parent; we'll move this to the orphanage after diff --git a/fs/xfs/scrub/nlinks.c b/fs/xfs/scrub/nlinks.c index c456523fac9c..fcb9c473f372 100644 --- a/fs/xfs/scrub/nlinks.c +++ b/fs/xfs/scrub/nlinks.c @@ -607,9 +607,11 @@ xchk_nlinks_compare_inode( * this as a corruption. The VFS won't let users increase the link * count, but it will let them decrease it. */ - if (total_links > XFS_MAXLINK) { + if (total_links > XFS_NLINK_PINNED) { xchk_ino_set_corrupt(sc, ip->i_ino); goto out_corrupt; + } else if (total_links > XFS_MAXLINK) { + xchk_ino_set_warning(sc, ip->i_ino); } /* Link counts should match. */ diff --git a/fs/xfs/scrub/nlinks_repair.c b/fs/xfs/scrub/nlinks_repair.c index 0cb67339eac8..83f8637bb08f 100644 --- a/fs/xfs/scrub/nlinks_repair.c +++ b/fs/xfs/scrub/nlinks_repair.c @@ -238,14 +238,10 @@ xrep_nlinks_repair_inode( /* Commit the new link count if it changed. */ if (total_links != actual_nlink) { - if (total_links > XFS_MAXLINK) { - trace_xrep_nlinks_unfixable_inode(mp, ip, &obs); - goto out_trans; - } - trace_xrep_nlinks_update_inode(mp, ip, &obs); - set_nlink(VFS_I(ip), total_links); + set_nlink(VFS_I(ip), min_t(unsigned long long, total_links, + XFS_NLINK_PINNED)); dirty = true; } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index fed0cd6bffdf..03dcb4ac0431 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -890,22 +890,25 @@ xfs_init_new_inode( */ static int /* error */ xfs_droplink( - xfs_trans_t *tp, - xfs_inode_t *ip) + struct xfs_trans *tp, + struct xfs_inode *ip) { - if (VFS_I(ip)->i_nlink == 0) { - xfs_alert(ip->i_mount, - "%s: Attempt to drop inode (%llu) with nlink zero.", - __func__, ip->i_ino); - return -EFSCORRUPTED; - } + struct inode *inode = VFS_I(ip); xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); - drop_nlink(VFS_I(ip)); + if (inode->i_nlink == 0) { + xfs_info_ratelimited(tp->t_mountp, + "Inode 0x%llx link count dropped below zero. Pinning link count.", + ip->i_ino); + set_nlink(inode, XFS_NLINK_PINNED); + } + if (inode->i_nlink != XFS_NLINK_PINNED) + drop_nlink(inode); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - if (VFS_I(ip)->i_nlink) + if (inode->i_nlink) return 0; return xfs_iunlink(tp, ip); @@ -919,9 +922,17 @@ xfs_bumplink( struct xfs_trans *tp, struct xfs_inode *ip) { + struct inode *inode = VFS_I(ip); + xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); - inc_nlink(VFS_I(ip)); + if (inode->i_nlink == XFS_NLINK_PINNED - 1) + xfs_info_ratelimited(tp->t_mountp, + "Inode 0x%llx link count exceeded maximum. Pinning link count.", + ip->i_ino); + if (inode->i_nlink != XFS_NLINK_PINNED) + inc_nlink(inode); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); } -- cgit v1.2.3-59-g8ed1b From 1a5f6e08d4e379a23da5be974aee50b26a20c5b0 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:06 -0700 Subject: xfs: create subordinate scrub contexts for xchk_metadata_inode_subtype When a file-based metadata structure is being scrubbed in xchk_metadata_inode_subtype, we should create an entirely new scrub context so that each scrubber doesn't trip over another's buffers. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/common.c | 23 +++--------------- fs/xfs/scrub/repair.c | 67 ++++++++++----------------------------------------- fs/xfs/scrub/scrub.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/scrub.h | 11 +++++++++ 4 files changed, 91 insertions(+), 73 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index a2da2bef509a..48302532d10d 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -1203,27 +1203,12 @@ xchk_metadata_inode_subtype( struct xfs_scrub *sc, unsigned int scrub_type) { - __u32 smtype = sc->sm->sm_type; - unsigned int sick_mask = sc->sick_mask; + struct xfs_scrub_subord *sub; int error; - sc->sm->sm_type = scrub_type; - - switch (scrub_type) { - case XFS_SCRUB_TYPE_INODE: - error = xchk_inode(sc); - break; - case XFS_SCRUB_TYPE_BMBTD: - error = xchk_bmap_data(sc); - break; - default: - ASSERT(0); - error = -EFSCORRUPTED; - break; - } - - sc->sick_mask = sick_mask; - sc->sm->sm_type = smtype; + sub = xchk_scrub_create_subord(sc, scrub_type); + error = sub->sc.ops->scrub(&sub->sc); + xchk_scrub_free_subord(sub); return error; } diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index 369f0430e4ba..b6aff89679d5 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -1009,55 +1009,27 @@ xrep_metadata_inode_subtype( struct xfs_scrub *sc, unsigned int scrub_type) { - __u32 smtype = sc->sm->sm_type; - __u32 smflags = sc->sm->sm_flags; - unsigned int sick_mask = sc->sick_mask; + struct xfs_scrub_subord *sub; int error; /* - * Let's see if the inode needs repair. We're going to open-code calls - * to the scrub and repair functions so that we can hang on to the + * Let's see if the inode needs repair. Use a subordinate scrub context + * to call the scrub and repair functions so that we can hang on to the * resources that we already acquired instead of using the standard * setup/teardown routines. */ - sc->sm->sm_flags &= ~XFS_SCRUB_FLAGS_OUT; - sc->sm->sm_type = scrub_type; - - switch (scrub_type) { - case XFS_SCRUB_TYPE_INODE: - error = xchk_inode(sc); - break; - case XFS_SCRUB_TYPE_BMBTD: - error = xchk_bmap_data(sc); - break; - case XFS_SCRUB_TYPE_BMBTA: - error = xchk_bmap_attr(sc); - break; - default: - ASSERT(0); - error = -EFSCORRUPTED; - } + sub = xchk_scrub_create_subord(sc, scrub_type); + error = sub->sc.ops->scrub(&sub->sc); if (error) goto out; - - if (!xrep_will_attempt(sc)) + if (!xrep_will_attempt(&sub->sc)) goto out; /* * Repair some part of the inode. This will potentially join the inode * to the transaction. */ - switch (scrub_type) { - case XFS_SCRUB_TYPE_INODE: - error = xrep_inode(sc); - break; - case XFS_SCRUB_TYPE_BMBTD: - error = xrep_bmap(sc, XFS_DATA_FORK, false); - break; - case XFS_SCRUB_TYPE_BMBTA: - error = xrep_bmap(sc, XFS_ATTR_FORK, false); - break; - } + error = sub->sc.ops->repair(&sub->sc); if (error) goto out; @@ -1066,10 +1038,10 @@ xrep_metadata_inode_subtype( * that the inode will not be joined to the transaction when we exit * the function. */ - error = xfs_defer_finish(&sc->tp); + error = xfs_defer_finish(&sub->sc.tp); if (error) goto out; - error = xfs_trans_roll(&sc->tp); + error = xfs_trans_roll(&sub->sc.tp); if (error) goto out; @@ -1077,31 +1049,18 @@ xrep_metadata_inode_subtype( * Clear the corruption flags and re-check the metadata that we just * repaired. */ - sc->sm->sm_flags &= ~XFS_SCRUB_FLAGS_OUT; - - switch (scrub_type) { - case XFS_SCRUB_TYPE_INODE: - error = xchk_inode(sc); - break; - case XFS_SCRUB_TYPE_BMBTD: - error = xchk_bmap_data(sc); - break; - case XFS_SCRUB_TYPE_BMBTA: - error = xchk_bmap_attr(sc); - break; - } + sub->sc.sm->sm_flags &= ~XFS_SCRUB_FLAGS_OUT; + error = sub->sc.ops->scrub(&sub->sc); if (error) goto out; /* If corruption persists, the repair has failed. */ - if (xchk_needs_repair(sc->sm)) { + if (xchk_needs_repair(sub->sc.sm)) { error = -EFSCORRUPTED; goto out; } out: - sc->sick_mask = sick_mask; - sc->sm->sm_type = smtype; - sc->sm->sm_flags = smflags; + xchk_scrub_free_subord(sub); return error; } diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 301d5b753fdd..ebb06838c31b 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -177,6 +177,39 @@ xchk_fsgates_disable( } #undef FSGATES_MASK +/* Free the resources associated with a scrub subtype. */ +void +xchk_scrub_free_subord( + struct xfs_scrub_subord *sub) +{ + struct xfs_scrub *sc = sub->parent_sc; + + ASSERT(sc->ip == sub->sc.ip); + ASSERT(sc->orphanage == sub->sc.orphanage); + ASSERT(sc->tempip == sub->sc.tempip); + + sc->sm->sm_type = sub->old_smtype; + sc->sm->sm_flags = sub->old_smflags | + (sc->sm->sm_flags & XFS_SCRUB_FLAGS_OUT); + sc->tp = sub->sc.tp; + + if (sub->sc.buf) { + if (sub->sc.buf_cleanup) + sub->sc.buf_cleanup(sub->sc.buf); + kvfree(sub->sc.buf); + } + if (sub->sc.xmbtp) + xmbuf_free(sub->sc.xmbtp); + if (sub->sc.xfile) + xfile_destroy(sub->sc.xfile); + + sc->ilock_flags = sub->sc.ilock_flags; + sc->orphanage_ilock_flags = sub->sc.orphanage_ilock_flags; + sc->temp_ilock_flags = sub->sc.temp_ilock_flags; + + kfree(sub); +} + /* Free all the resources and finish the transactions. */ STATIC int xchk_teardown( @@ -505,6 +538,36 @@ static inline void xchk_postmortem(struct xfs_scrub *sc) } #endif /* CONFIG_XFS_ONLINE_REPAIR */ +/* + * Create a new scrub context from an existing one, but with a different scrub + * type. + */ +struct xfs_scrub_subord * +xchk_scrub_create_subord( + struct xfs_scrub *sc, + unsigned int subtype) +{ + struct xfs_scrub_subord *sub; + + sub = kzalloc(sizeof(*sub), XCHK_GFP_FLAGS); + if (!sub) + return ERR_PTR(-ENOMEM); + + sub->old_smtype = sc->sm->sm_type; + sub->old_smflags = sc->sm->sm_flags; + sub->parent_sc = sc; + memcpy(&sub->sc, sc, sizeof(struct xfs_scrub)); + sub->sc.ops = &meta_scrub_ops[subtype]; + sub->sc.sm->sm_type = subtype; + sub->sc.sm->sm_flags &= ~XFS_SCRUB_FLAGS_OUT; + sub->sc.buf = NULL; + sub->sc.buf_cleanup = NULL; + sub->sc.xfile = NULL; + sub->sc.xmbtp = NULL; + + return sub; +} + /* Dispatch metadata scrubbing. */ int xfs_scrub_metadata( diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index 7abe498f7a46..54a4242bc79c 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -156,6 +156,17 @@ struct xfs_scrub { */ #define XREP_FSGATES_ALL (XREP_FSGATES_EXCHANGE_RANGE) +struct xfs_scrub_subord { + struct xfs_scrub sc; + struct xfs_scrub *parent_sc; + unsigned int old_smtype; + unsigned int old_smflags; +}; + +struct xfs_scrub_subord *xchk_scrub_create_subord(struct xfs_scrub *sc, + unsigned int subtype); +void xchk_scrub_free_subord(struct xfs_scrub_subord *sub); + /* Metadata scrubbers */ int xchk_tester(struct xfs_scrub *sc); int xchk_superblock(struct xfs_scrub *sc); -- cgit v1.2.3-59-g8ed1b From b0ffe661fab4b939e4472ef96b8dac3c74e0e03e Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:06 -0700 Subject: xfs: fix performance problems when fstrimming a subset of a fragmented AG On a 10TB filesystem where the free space in each AG is heavily fragmented, I noticed some very high runtimes on a FITRIM call for the entire filesystem. xfs_scrub likes to report progress information on each phase of the scrub, which means that a strace for the entire filesystem: ioctl(3, FITRIM, {start=0x0, len=10995116277760, minlen=0}) = 0 <686.209839> shows that scrub is uncommunicative for the entire duration. Reducing the size of the FITRIM requests to a single AG at a time produces lower times for each individual call, but even this isn't quite acceptable, because the time between progress reports are still very high: Strace for the first 4x 1TB AGs looks like (2): ioctl(3, FITRIM, {start=0x0, len=1099511627776, minlen=0}) = 0 <68.352033> ioctl(3, FITRIM, {start=0x10000000000, len=1099511627776, minlen=0}) = 0 <68.760323> ioctl(3, FITRIM, {start=0x20000000000, len=1099511627776, minlen=0}) = 0 <67.235226> ioctl(3, FITRIM, {start=0x30000000000, len=1099511627776, minlen=0}) = 0 <69.465744> I then had the idea to limit the length parameter of each call to a smallish amount (~11GB) so that we could report progress relatively quickly, but much to my surprise, each FITRIM call still took ~68 seconds! Unfortunately, the by-length fstrim implementation handles this poorly because it walks the entire free space by length index (cntbt), which is a very inefficient way to walk a subset of the blocks of an AG. Therefore, create a second implementation that will walk the bnobt and perform the trims in block number order. This implementation avoids the worst problems of the original code, though it lacks the desirable attribute of freeing the biggest chunks first. On the other hand, this second implementation will be much easier to constrain the system call latency, and makes it much easier to report fstrim progress to anyone who's running xfs_scrub. Signed-off-by: Darrick J. Wong Reviewed-by: Dave Chinner ar_startblock == NULLAGBLOCK) - error = xfs_alloc_lookup_ge(cur, 0, tcur->ar_blockcount, &i); - else - error = xfs_alloc_lookup_le(cur, tcur->ar_startblock, - tcur->ar_blockcount, &i); + if (tcur->by_bno) { + /* sub-AG discard request always starts at tcur->start */ + cur = xfs_bnobt_init_cursor(mp, tp, agbp, pag); + error = xfs_alloc_lookup_le(cur, tcur->start, 0, &i); + if (!error && !i) + error = xfs_alloc_lookup_ge(cur, tcur->start, 0, &i); + } else if (tcur->start == 0) { + /* first time through a by-len starts with max length */ + cur = xfs_cntbt_init_cursor(mp, tp, agbp, pag); + error = xfs_alloc_lookup_ge(cur, 0, tcur->count, &i); + } else { + /* nth time through a by-len starts where we left off */ + cur = xfs_cntbt_init_cursor(mp, tp, agbp, pag); + error = xfs_alloc_lookup_le(cur, tcur->start, tcur->count, &i); + } if (error) goto out_del_cursor; if (i == 0) { /* nothing of that length left in the AG, we are done */ - tcur->ar_blockcount = 0; + tcur->count = 0; goto out_del_cursor; } @@ -204,8 +213,6 @@ xfs_trim_gather_extents( while (i) { xfs_agblock_t fbno; xfs_extlen_t flen; - xfs_daddr_t dbno; - xfs_extlen_t dlen; error = xfs_alloc_get_rec(cur, &fbno, &flen, &i); if (error) @@ -221,38 +228,46 @@ xfs_trim_gather_extents( * Update the cursor to point at this extent so we * restart the next batch from this extent. */ - tcur->ar_startblock = fbno; - tcur->ar_blockcount = flen; - break; - } - - /* - * use daddr format for all range/len calculations as that is - * the format the range/len variables are supplied in by - * userspace. - */ - dbno = XFS_AGB_TO_DADDR(mp, pag->pag_agno, fbno); - dlen = XFS_FSB_TO_BB(mp, flen); - - /* - * Too small? Give up. - */ - if (dlen < minlen) { - trace_xfs_discard_toosmall(mp, pag->pag_agno, fbno, flen); - tcur->ar_blockcount = 0; + tcur->start = fbno; + tcur->count = flen; break; } /* * If the extent is entirely outside of the range we are - * supposed to discard skip it. Do not bother to trim - * down partially overlapping ranges for now. + * supposed to skip it. Do not bother to trim down partially + * overlapping ranges for now. */ - if (dbno + dlen < start || dbno > end) { + if (fbno + flen < tcur->start) { + trace_xfs_discard_exclude(mp, pag->pag_agno, fbno, flen); + goto next_extent; + } + if (fbno > tcur->end) { trace_xfs_discard_exclude(mp, pag->pag_agno, fbno, flen); + if (tcur->by_bno) { + tcur->count = 0; + break; + } goto next_extent; } + /* Trim the extent returned to the range we want. */ + if (fbno < tcur->start) { + flen -= tcur->start - fbno; + fbno = tcur->start; + } + if (fbno + flen > tcur->end + 1) + flen = tcur->end - fbno + 1; + + /* Too small? Give up. */ + if (flen < tcur->minlen) { + trace_xfs_discard_toosmall(mp, pag->pag_agno, fbno, flen); + if (tcur->by_bno) + goto next_extent; + tcur->count = 0; + break; + } + /* * If any blocks in the range are still busy, skip the * discard and try again the next time. @@ -266,7 +281,10 @@ xfs_trim_gather_extents( &extents->extent_list); *blocks_trimmed += flen; next_extent: - error = xfs_btree_decrement(cur, 0, &i); + if (tcur->by_bno) + error = xfs_btree_increment(cur, 0, &i); + else + error = xfs_btree_decrement(cur, 0, &i); if (error) break; @@ -276,7 +294,7 @@ next_extent: * is no more extents to search. */ if (i == 0) - tcur->ar_blockcount = 0; + tcur->count = 0; } /* @@ -306,17 +324,22 @@ xfs_trim_should_stop(void) static int xfs_trim_extents( struct xfs_perag *pag, - xfs_daddr_t start, - xfs_daddr_t end, - xfs_daddr_t minlen, + xfs_agblock_t start, + xfs_agblock_t end, + xfs_extlen_t minlen, uint64_t *blocks_trimmed) { - struct xfs_alloc_rec_incore tcur = { - .ar_blockcount = pag->pagf_longest, - .ar_startblock = NULLAGBLOCK, + struct xfs_trim_cur tcur = { + .start = start, + .count = pag->pagf_longest, + .end = end, + .minlen = minlen, }; int error = 0; + if (start != 0 || end != pag->block_count) + tcur.by_bno = true; + do { struct xfs_busy_extents *extents; @@ -330,8 +353,8 @@ xfs_trim_extents( extents->owner = extents; INIT_LIST_HEAD(&extents->extent_list); - error = xfs_trim_gather_extents(pag, start, end, minlen, - &tcur, extents, blocks_trimmed); + error = xfs_trim_gather_extents(pag, &tcur, extents, + blocks_trimmed); if (error) { kfree(extents); break; @@ -354,7 +377,7 @@ xfs_trim_extents( if (xfs_trim_should_stop()) break; - } while (tcur.ar_blockcount != 0); + } while (tcur.count != 0); return error; @@ -378,8 +401,10 @@ xfs_ioc_trim( unsigned int granularity = bdev_discard_granularity(mp->m_ddev_targp->bt_bdev); struct fstrim_range range; - xfs_daddr_t start, end, minlen; - xfs_agnumber_t agno; + xfs_daddr_t start, end; + xfs_extlen_t minlen; + xfs_agnumber_t start_agno, end_agno; + xfs_agblock_t start_agbno, end_agbno; uint64_t blocks_trimmed = 0; int error, last_error = 0; @@ -399,7 +424,8 @@ xfs_ioc_trim( return -EFAULT; range.minlen = max_t(u64, granularity, range.minlen); - minlen = BTOBB(range.minlen); + minlen = XFS_B_TO_FSB(mp, range.minlen); + /* * Truncating down the len isn't actually quite correct, but using * BBTOB would mean we trivially get overflows for values @@ -413,15 +439,21 @@ xfs_ioc_trim( return -EINVAL; start = BTOBB(range.start); - end = start + BTOBBT(range.len) - 1; + end = min_t(xfs_daddr_t, start + BTOBBT(range.len), + XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)) - 1; + + start_agno = xfs_daddr_to_agno(mp, start); + start_agbno = xfs_daddr_to_agbno(mp, start); + end_agno = xfs_daddr_to_agno(mp, end); + end_agbno = xfs_daddr_to_agbno(mp, end); - if (end > XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1) - end = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1; + for_each_perag_range(mp, start_agno, end_agno, pag) { + xfs_agblock_t agend = pag->block_count; - agno = xfs_daddr_to_agno(mp, start); - for_each_perag_range(mp, agno, xfs_daddr_to_agno(mp, end), pag) { - error = xfs_trim_extents(pag, start, end, minlen, - &blocks_trimmed); + if (start_agno == end_agno) + agend = end_agbno; + error = xfs_trim_extents(pag, start_agbno, agend, minlen, + &blocks_trimmed); if (error) last_error = error; @@ -429,6 +461,7 @@ xfs_ioc_trim( xfs_perag_rele(pag); break; } + start_agbno = 0; } if (last_error) -- cgit v1.2.3-59-g8ed1b From 7560c937b4b5a3c671053be86ff00156a6fdd399 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 15 Apr 2024 14:55:11 -0700 Subject: xfs: Increase XFS_DEFER_OPS_NR_INODES to 5 Renames that generate parent pointer updates can join up to 5 inodes locked in sorted order. So we need to increase the number of defer ops inodes and relock them in the same way. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Reviewed-by: Catherine Hoang [djwong: have one sorting function] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_defer.c | 6 +++++- fs/xfs/libxfs/xfs_defer.h | 8 +++++++- fs/xfs/xfs_inode.c | 27 ++++++++++++++++++--------- fs/xfs/xfs_inode.h | 2 ++ 4 files changed, 32 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c index 061cc01245a9..4a078e07e1a0 100644 --- a/fs/xfs/libxfs/xfs_defer.c +++ b/fs/xfs/libxfs/xfs_defer.c @@ -1092,7 +1092,11 @@ xfs_defer_ops_continue( ASSERT(!(tp->t_flags & XFS_TRANS_DIRTY)); /* Lock the captured resources to the new transaction. */ - if (dfc->dfc_held.dr_inos == 2) + if (dfc->dfc_held.dr_inos > 2) { + xfs_sort_inodes(dfc->dfc_held.dr_ip, dfc->dfc_held.dr_inos); + xfs_lock_inodes(dfc->dfc_held.dr_ip, dfc->dfc_held.dr_inos, + XFS_ILOCK_EXCL); + } else if (dfc->dfc_held.dr_inos == 2) xfs_lock_two_inodes(dfc->dfc_held.dr_ip[0], XFS_ILOCK_EXCL, dfc->dfc_held.dr_ip[1], XFS_ILOCK_EXCL); else if (dfc->dfc_held.dr_inos == 1) diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h index 81cca60d70a3..8b338031e487 100644 --- a/fs/xfs/libxfs/xfs_defer.h +++ b/fs/xfs/libxfs/xfs_defer.h @@ -77,7 +77,13 @@ extern const struct xfs_defer_op_type xfs_exchmaps_defer_type; /* * Deferred operation item relogging limits. */ -#define XFS_DEFER_OPS_NR_INODES 2 /* join up to two inodes */ + +/* + * Rename w/ parent pointers can require up to 5 inodes with deferred ops to + * be joined to the transaction: src_dp, target_dp, src_ip, target_ip, and wip. + * These inodes are locked in sorted order by their inode numbers + */ +#define XFS_DEFER_OPS_NR_INODES 5 #define XFS_DEFER_OPS_NR_BUFS 2 /* join up to two buffers */ /* Resources that must be held across a transaction roll. */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 03dcb4ac0431..efd040094753 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -418,7 +418,7 @@ xfs_lock_inumorder( * lock more than one at a time, lockdep will report false positives saying we * have violated locking orders. */ -static void +void xfs_lock_inodes( struct xfs_inode **ips, int inodes, @@ -2802,7 +2802,7 @@ xfs_sort_for_rename( struct xfs_inode **i_tab,/* out: sorted array of inodes */ int *num_inodes) /* in/out: inodes in array */ { - int i, j; + int i; ASSERT(*num_inodes == __XFS_SORT_INODES); memset(i_tab, 0, *num_inodes * sizeof(struct xfs_inode *)); @@ -2824,17 +2824,26 @@ xfs_sort_for_rename( i_tab[i++] = wip; *num_inodes = i; + xfs_sort_inodes(i_tab, *num_inodes); +} + +void +xfs_sort_inodes( + struct xfs_inode **i_tab, + unsigned int num_inodes) +{ + int i, j; + + ASSERT(num_inodes <= __XFS_SORT_INODES); + /* * Sort the elements via bubble sort. (Remember, there are at * most 5 elements to sort, so this is adequate.) */ - for (i = 0; i < *num_inodes; i++) { - for (j = 1; j < *num_inodes; j++) { - if (i_tab[j]->i_ino < i_tab[j-1]->i_ino) { - struct xfs_inode *temp = i_tab[j]; - i_tab[j] = i_tab[j-1]; - i_tab[j-1] = temp; - } + for (i = 0; i < num_inodes; i++) { + for (j = 1; j < num_inodes; j++) { + if (i_tab[j]->i_ino < i_tab[j-1]->i_ino) + swap(i_tab[j], i_tab[j - 1]); } } } diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index c74c48bc0945..a6da1ab8ab13 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -627,6 +627,8 @@ int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2); void xfs_iunlock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2); void xfs_iunlock2_remapping(struct xfs_inode *ip1, struct xfs_inode *ip2); void xfs_bumplink(struct xfs_trans *tp, struct xfs_inode *ip); +void xfs_lock_inodes(struct xfs_inode **ips, int inodes, uint lock_mode); +void xfs_sort_inodes(struct xfs_inode **i_tab, unsigned int num_inodes); static inline bool xfs_inode_unlinked_incomplete( -- cgit v1.2.3-59-g8ed1b From f103df763563ad6849307ed5985d1513acc586dd Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 15 Apr 2024 14:55:12 -0700 Subject: xfs: Increase XFS_QM_TRANS_MAXDQS to 5 With parent pointers enabled, a rename operation can update up to 5 inodes: src_dp, target_dp, src_ip, target_ip and wip. This causes their dquots to a be attached to the transaction chain, so we need to increase XFS_QM_TRANS_MAXDQS. This patch also add a helper function xfs_dqlockn to lock an arbitrary number of dquots. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_dquot.c | 41 +++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_dquot.h | 1 + fs/xfs/xfs_qm.h | 2 +- fs/xfs/xfs_trans_dquot.c | 15 ++++++++++----- 4 files changed, 53 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index c98cb468c357..13aba84bd64a 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -1371,6 +1371,47 @@ xfs_dqlock2( } } +static int +xfs_dqtrx_cmp( + const void *a, + const void *b) +{ + const struct xfs_dqtrx *qa = a; + const struct xfs_dqtrx *qb = b; + + if (qa->qt_dquot->q_id > qb->qt_dquot->q_id) + return 1; + if (qa->qt_dquot->q_id < qb->qt_dquot->q_id) + return -1; + return 0; +} + +void +xfs_dqlockn( + struct xfs_dqtrx *q) +{ + unsigned int i; + + BUILD_BUG_ON(XFS_QM_TRANS_MAXDQS > MAX_LOCKDEP_SUBCLASSES); + + /* Sort in order of dquot id, do not allow duplicates */ + for (i = 0; i < XFS_QM_TRANS_MAXDQS && q[i].qt_dquot != NULL; i++) { + unsigned int j; + + for (j = 0; j < i; j++) + ASSERT(q[i].qt_dquot != q[j].qt_dquot); + } + if (i == 0) + return; + + sort(q, i, sizeof(struct xfs_dqtrx), xfs_dqtrx_cmp, NULL); + + mutex_lock(&q[0].qt_dquot->q_qlock); + for (i = 1; i < XFS_QM_TRANS_MAXDQS && q[i].qt_dquot != NULL; i++) + mutex_lock_nested(&q[i].qt_dquot->q_qlock, + XFS_QLOCK_NESTED + i - 1); +} + int __init xfs_qm_init(void) { diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h index 956272d9b302..677bb2dc9ac9 100644 --- a/fs/xfs/xfs_dquot.h +++ b/fs/xfs/xfs_dquot.h @@ -223,6 +223,7 @@ int xfs_qm_dqget_uncached(struct xfs_mount *mp, void xfs_qm_dqput(struct xfs_dquot *dqp); void xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *); +void xfs_dqlockn(struct xfs_dqtrx *q); void xfs_dquot_set_prealloc_limits(struct xfs_dquot *); diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h index f5993012bf98..6e09dfcd13e2 100644 --- a/fs/xfs/xfs_qm.h +++ b/fs/xfs/xfs_qm.h @@ -136,7 +136,7 @@ enum { XFS_QM_TRANS_PRJ, XFS_QM_TRANS_DQTYPES }; -#define XFS_QM_TRANS_MAXDQS 2 +#define XFS_QM_TRANS_MAXDQS 5 struct xfs_dquot_acct { struct xfs_dqtrx dqs[XFS_QM_TRANS_DQTYPES][XFS_QM_TRANS_MAXDQS]; }; diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index 577b535a595c..b368e13424c4 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -379,24 +379,29 @@ xfs_trans_mod_dquot( /* * Given an array of dqtrx structures, lock all the dquots associated and join - * them to the transaction, provided they have been modified. We know that the - * highest number of dquots of one type - usr, grp and prj - involved in a - * transaction is 3 so we don't need to make this very generic. + * them to the transaction, provided they have been modified. */ STATIC void xfs_trans_dqlockedjoin( struct xfs_trans *tp, struct xfs_dqtrx *q) { + unsigned int i; ASSERT(q[0].qt_dquot != NULL); if (q[1].qt_dquot == NULL) { xfs_dqlock(q[0].qt_dquot); xfs_trans_dqjoin(tp, q[0].qt_dquot); - } else { - ASSERT(XFS_QM_TRANS_MAXDQS == 2); + } else if (q[2].qt_dquot == NULL) { xfs_dqlock2(q[0].qt_dquot, q[1].qt_dquot); xfs_trans_dqjoin(tp, q[0].qt_dquot); xfs_trans_dqjoin(tp, q[1].qt_dquot); + } else { + xfs_dqlockn(q); + for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { + if (q[i].qt_dquot == NULL) + break; + xfs_trans_dqjoin(tp, q[i].qt_dquot); + } } } -- cgit v1.2.3-59-g8ed1b From 267979b4ce75767a9caf69072e1b92416e286a29 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 15 Apr 2024 14:55:12 -0700 Subject: xfs: Hold inode locks in xfs_ialloc Modify xfs_ialloc to hold locks after return. Caller will be responsible for manual unlock. We will need this later to hold locks across parent pointer operations Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Reviewed-by: Catherine Hoang [djwong: hold the parent ilocked across transaction rolls too] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_inode.c | 12 +++++++++--- fs/xfs/xfs_qm.c | 4 +++- fs/xfs/xfs_symlink.c | 6 ++++-- 3 files changed, 16 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index efd040094753..2ec005e6c1da 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -747,6 +747,8 @@ xfs_inode_inherit_flags2( /* * Initialise a newly allocated inode and return the in-core inode to the * caller locked exclusively. + * + * Caller is responsible for unlocking the inode manually upon return */ int xfs_init_new_inode( @@ -873,7 +875,7 @@ xfs_init_new_inode( /* * Log the new values stuffed into the inode. */ - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, 0); xfs_trans_log_inode(tp, ip, flags); /* now that we have an i_mode we can setup the inode structure */ @@ -1101,8 +1103,7 @@ xfs_create( * the transaction cancel unlocking dp so don't do it explicitly in the * error path. */ - xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - unlock_dp_on_error = false; + xfs_trans_ijoin(tp, dp, 0); error = xfs_dir_createname(tp, dp, name, ip->i_ino, resblks - XFS_IALLOC_SPACE_RES(mp)); @@ -1151,6 +1152,8 @@ xfs_create( xfs_qm_dqrele(pdqp); *ipp = ip; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(dp, XFS_ILOCK_EXCL); return 0; out_trans_cancel: @@ -1162,6 +1165,7 @@ xfs_create( * transactions and deadlocks from xfs_inactive. */ if (ip) { + xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_finish_inode_setup(ip); xfs_irele(ip); } @@ -1247,6 +1251,7 @@ xfs_create_tmpfile( xfs_qm_dqrele(pdqp); *ipp = ip; + xfs_iunlock(ip, XFS_ILOCK_EXCL); return 0; out_trans_cancel: @@ -1258,6 +1263,7 @@ xfs_create_tmpfile( * transactions and deadlocks from xfs_inactive. */ if (ip) { + xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_finish_inode_setup(ip); xfs_irele(ip); } diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index 0f4cf4170c35..47120b745c47 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -836,8 +836,10 @@ xfs_qm_qino_alloc( ASSERT(xfs_is_shutdown(mp)); xfs_alert(mp, "%s failed (error %d)!", __func__, error); } - if (need_alloc) + if (need_alloc) { + xfs_iunlock(*ipp, XFS_ILOCK_EXCL); xfs_finish_inode_setup(*ipp); + } return error; } diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index fb060aaf6d40..85ef56fdd7df 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -172,8 +172,7 @@ xfs_symlink( * the transaction cancel unlocking dp so don't do it explicitly in the * error path. */ - xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - unlock_dp_on_error = false; + xfs_trans_ijoin(tp, dp, 0); /* * Also attach the dquot(s) to it, if applicable. @@ -215,6 +214,8 @@ xfs_symlink( xfs_qm_dqrele(pdqp); *ipp = ip; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(dp, XFS_ILOCK_EXCL); return 0; out_trans_cancel: @@ -226,6 +227,7 @@ out_release_inode: * transactions and deadlocks from xfs_inactive. */ if (ip) { + xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_finish_inode_setup(ip); xfs_irele(ip); } -- cgit v1.2.3-59-g8ed1b From bd5562111d58392298a3c3b93caad71dff681b4b Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 15 Apr 2024 14:55:13 -0700 Subject: xfs: Hold inode locks in xfs_trans_alloc_dir Modify xfs_trans_alloc_dir to hold locks after return. Caller will be responsible for manual unlock. We will need this later to hold locks across parent pointer operations Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Reviewed-by: Catherine Hoang Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_inode.c | 14 ++++++++++++-- fs/xfs/xfs_trans.c | 9 +++++++-- 2 files changed, 19 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 2ec005e6c1da..36e1012e156a 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1368,10 +1368,15 @@ xfs_link( if (xfs_has_wsync(mp) || xfs_has_dirsync(mp)) xfs_trans_set_sync(tp); - return xfs_trans_commit(tp); + error = xfs_trans_commit(tp); + xfs_iunlock(tdp, XFS_ILOCK_EXCL); + xfs_iunlock(sip, XFS_ILOCK_EXCL); + return error; error_return: xfs_trans_cancel(tp); + xfs_iunlock(tdp, XFS_ILOCK_EXCL); + xfs_iunlock(sip, XFS_ILOCK_EXCL); std_return: if (error == -ENOSPC && nospace_error) error = nospace_error; @@ -2781,15 +2786,20 @@ xfs_remove( error = xfs_trans_commit(tp); if (error) - goto std_return; + goto out_unlock; if (is_dir && xfs_inode_is_filestream(ip)) xfs_filestream_deassociate(ip); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(dp, XFS_ILOCK_EXCL); return 0; out_trans_cancel: xfs_trans_cancel(tp); + out_unlock: + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(dp, XFS_ILOCK_EXCL); std_return: return error; } diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 7350640059cc..50d878d78a5e 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -1430,6 +1430,8 @@ out_cancel: * The caller must ensure that the on-disk dquots attached to this inode have * already been allocated and initialized. The ILOCKs will be dropped when the * transaction is committed or cancelled. + * + * Caller is responsible for unlocking the inodes manually upon return */ int xfs_trans_alloc_dir( @@ -1460,8 +1462,8 @@ retry: xfs_lock_two_inodes(dp, XFS_ILOCK_EXCL, ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, dp, 0); + xfs_trans_ijoin(tp, ip, 0); error = xfs_qm_dqattach_locked(dp, false); if (error) { @@ -1484,6 +1486,9 @@ retry: if (error == -EDQUOT || error == -ENOSPC) { if (!retried) { xfs_trans_cancel(tp); + xfs_iunlock(dp, XFS_ILOCK_EXCL); + if (dp != ip) + xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_blockgc_free_quota(dp, 0); retried = true; goto retry; -- cgit v1.2.3-59-g8ed1b From 69291726caf12b28e1854cbffb73bd8354579fe7 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 15 Apr 2024 14:55:14 -0700 Subject: xfs: Hold inode locks in xfs_rename Modify xfs_rename to hold all inode locks across a rename operation We will need this later when we add parent pointers Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Reviewed-by: Catherine Hoang Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_inode.c | 45 +++++++++++++++++++++++++++++++++------------ 1 file changed, 33 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 36e1012e156a..2aec7ab59aeb 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2804,6 +2804,21 @@ xfs_remove( return error; } +static inline void +xfs_iunlock_rename( + struct xfs_inode **i_tab, + int num_inodes) +{ + int i; + + for (i = num_inodes - 1; i >= 0; i--) { + /* Skip duplicate inodes if src and target dps are the same */ + if (!i_tab[i] || (i > 0 && i_tab[i] == i_tab[i - 1])) + continue; + xfs_iunlock(i_tab[i], XFS_ILOCK_EXCL); + } +} + /* * Enter all inodes for a rename transaction into a sorted array. */ @@ -3113,8 +3128,10 @@ retry: * Attach the dquots to the inodes */ error = xfs_qm_vop_rename_dqattach(inodes); - if (error) - goto out_trans_cancel; + if (error) { + xfs_trans_cancel(tp); + goto out_release_wip; + } /* * Lock all the participating inodes. Depending upon whether @@ -3125,18 +3142,16 @@ retry: xfs_lock_inodes(inodes, num_inodes, XFS_ILOCK_EXCL); /* - * Join all the inodes to the transaction. From this point on, - * we can rely on either trans_commit or trans_cancel to unlock - * them. + * Join all the inodes to the transaction. */ - xfs_trans_ijoin(tp, src_dp, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, src_dp, 0); if (new_parent) - xfs_trans_ijoin(tp, target_dp, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, target_dp, 0); + xfs_trans_ijoin(tp, src_ip, 0); if (target_ip) - xfs_trans_ijoin(tp, target_ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, target_ip, 0); if (wip) - xfs_trans_ijoin(tp, wip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, wip, 0); /* * If we are using project inheritance, we only allow renames @@ -3150,10 +3165,13 @@ retry: } /* RENAME_EXCHANGE is unique from here on. */ - if (flags & RENAME_EXCHANGE) - return xfs_cross_rename(tp, src_dp, src_name, src_ip, + if (flags & RENAME_EXCHANGE) { + error = xfs_cross_rename(tp, src_dp, src_name, src_ip, target_dp, target_name, target_ip, spaceres); + xfs_iunlock_rename(inodes, num_inodes); + return error; + } /* * Try to reserve quota to handle an expansion of the target directory. @@ -3167,6 +3185,7 @@ retry: if (error == -EDQUOT || error == -ENOSPC) { if (!retried) { xfs_trans_cancel(tp); + xfs_iunlock_rename(inodes, num_inodes); xfs_blockgc_free_quota(target_dp, 0); retried = true; goto retry; @@ -3393,12 +3412,14 @@ retry: xfs_dir_update_hook(src_dp, wip, 1, src_name); error = xfs_finish_rename(tp); + xfs_iunlock_rename(inodes, num_inodes); if (wip) xfs_irele(wip); return error; out_trans_cancel: xfs_trans_cancel(tp); + xfs_iunlock_rename(inodes, num_inodes); out_release_wip: if (wip) xfs_irele(wip); -- cgit v1.2.3-59-g8ed1b From 34ef5e17d5fd62a49bd9790834560261ebdec607 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:15 -0700 Subject: xfs: don't pick up IOLOCK during rmapbt repair scan Now that we've fixed the directory operations to hold the ILOCK until they're finished with rmapbt updates for directory shape changes, we no longer need to take this lock when scanning directories for rmapbt records. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/rmap_repair.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/rmap_repair.c b/fs/xfs/scrub/rmap_repair.c index e8e07b683eab..25acd69614c2 100644 --- a/fs/xfs/scrub/rmap_repair.c +++ b/fs/xfs/scrub/rmap_repair.c @@ -578,23 +578,9 @@ xrep_rmap_scan_inode( struct xrep_rmap *rr, struct xfs_inode *ip) { - unsigned int lock_mode = 0; + unsigned int lock_mode = xrep_rmap_scan_ilock(ip); int error; - /* - * Directory updates (create/link/unlink/rename) drop the directory's - * ILOCK before finishing any rmapbt updates associated with directory - * shape changes. For this scan to coordinate correctly with the live - * update hook, we must take the only lock (i_rwsem) that is held all - * the way to dir op completion. This will get fixed by the parent - * pointer patchset. - */ - if (S_ISDIR(VFS_I(ip)->i_mode)) { - lock_mode = XFS_IOLOCK_SHARED; - xfs_ilock(ip, lock_mode); - } - lock_mode |= xrep_rmap_scan_ilock(ip); - /* Check the data fork. */ error = xrep_rmap_scan_ifork(rr, ip, XFS_DATA_FORK); if (error) -- cgit v1.2.3-59-g8ed1b From df760471477400ccd3ddcea85d2d6d92f4dad28c Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 15 Apr 2024 14:55:16 -0700 Subject: xfs: unlock new repair tempfiles after creation After creation, drop the ILOCK on temporary files that have been created to stage a repair. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/tempfile.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index c72e447eb8ec..6f39504a216e 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -153,6 +153,7 @@ xrep_tempfile_create( xfs_qm_dqrele(pdqp); /* Finish setting up the incore / vfs context. */ + xfs_iunlock(sc->tempip, XFS_ILOCK_EXCL); xfs_setup_iops(sc->tempip); xfs_finish_inode_setup(sc->tempip); @@ -168,6 +169,7 @@ out_release_inode: * transactions and deadlocks from xfs_inactive. */ if (sc->tempip) { + xfs_iunlock(sc->tempip, XFS_ILOCK_EXCL); xfs_finish_inode_setup(sc->tempip); xchk_irele(sc, sc->tempip); } -- cgit v1.2.3-59-g8ed1b From 74871791ffa9562d43567c5ff2ae93def3f39f65 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Mon, 25 Mar 2024 09:34:36 +0100 Subject: ntfs3: serve as alias for the legacy ntfs driver Johan Hovold reported that removing the legacy ntfs driver broke boot for him since his fstab uses the legacy ntfs driver to access firmware from the original Windows partition. Use ntfs3 as an alias for legacy ntfs if CONFIG_NTFS_FS is selected. This is similar to how ext3 is treated. Link: https://lore.kernel.org/r/Zf2zPf5TO5oYt3I3@hovoldconsulting.com Link: https://lore.kernel.org/r/20240325-hinkriegen-zuziehen-d7e2c490427a@brauner Fixes: 7ffa8f3d3023 ("fs: Remove NTFS classic") Tested-by: Johan Hovold Cc: Matthew Wilcox (Oracle) Cc: Johan Hovold Signed-off-by: Christian Brauner --- fs/ntfs3/Kconfig | 9 +++++++++ fs/ntfs3/super.c | 31 +++++++++++++++++++++++++++++++ 2 files changed, 40 insertions(+) (limited to 'fs') diff --git a/fs/ntfs3/Kconfig b/fs/ntfs3/Kconfig index cdfdf51e55d7..7bc31d69f680 100644 --- a/fs/ntfs3/Kconfig +++ b/fs/ntfs3/Kconfig @@ -46,3 +46,12 @@ config NTFS3_FS_POSIX_ACL NOTE: this is linux only feature. Windows will ignore these ACLs. If you don't know what Access Control Lists are, say N. + +config NTFS_FS + tristate "NTFS file system support" + select NTFS3_FS + select BUFFER_HEAD + select NLS + help + This config option is here only for backward compatibility. NTFS + filesystem is now handled by the NTFS3 driver. diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c index 9df7c20d066f..8d2e51bae2cb 100644 --- a/fs/ntfs3/super.c +++ b/fs/ntfs3/super.c @@ -1798,6 +1798,35 @@ static struct file_system_type ntfs_fs_type = { .kill_sb = ntfs3_kill_sb, .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, }; + +#if IS_ENABLED(CONFIG_NTFS_FS) +static struct file_system_type ntfs_legacy_fs_type = { + .owner = THIS_MODULE, + .name = "ntfs", + .init_fs_context = ntfs_init_fs_context, + .parameters = ntfs_fs_parameters, + .kill_sb = ntfs3_kill_sb, + .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, +}; +MODULE_ALIAS_FS("ntfs"); + +static inline void register_as_ntfs_legacy(void) +{ + int err = register_filesystem(&ntfs_legacy_fs_type); + if (err) + pr_warn("ntfs3: Failed to register legacy ntfs filesystem driver: %d\n", err); +} + +static inline void unregister_as_ntfs_legacy(void) +{ + unregister_filesystem(&ntfs_legacy_fs_type); +} +#else +static inline void register_as_ntfs_legacy(void) {} +static inline void unregister_as_ntfs_legacy(void) {} +#endif + + // clang-format on static int __init init_ntfs_fs(void) @@ -1832,6 +1861,7 @@ static int __init init_ntfs_fs(void) goto out1; } + register_as_ntfs_legacy(); err = register_filesystem(&ntfs_fs_type); if (err) goto out; @@ -1849,6 +1879,7 @@ static void __exit exit_ntfs_fs(void) rcu_barrier(); kmem_cache_destroy(ntfs_inode_cachep); unregister_filesystem(&ntfs_fs_type); + unregister_as_ntfs_legacy(); ntfs3_exit_bitmap(); #ifdef CONFIG_PROC_FS -- cgit v1.2.3-59-g8ed1b From 700b04808fad2eac24abf050f234f059199fa3fe Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:35 -0400 Subject: dlm: increment ls_count for dlm_scand Increment the ls_count value while dlm_scand is processing a lockspace so that release_lockspace()/remove_lockspace() will wait for dlm_scand to finish. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lockspace.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index c3681a50decb..731c48371a27 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -255,6 +255,7 @@ static struct dlm_ls *find_ls_to_scan(void) list_for_each_entry(ls, &lslist, ls_list) { if (time_after_eq(jiffies, ls->ls_scan_time + dlm_config.ci_scan_secs * HZ)) { + atomic_inc(&ls->ls_count); spin_unlock_bh(&lslist_lock); return ls; } @@ -277,6 +278,8 @@ static int dlm_scand(void *data) } else { ls->ls_scan_time += HZ; } + + dlm_put_lockspace(ls); continue; } schedule_timeout_interruptible(dlm_config.ci_scan_secs * HZ); -- cgit v1.2.3-59-g8ed1b From dcdaad05ca15150ae076299ba827867f243c0623 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:36 -0400 Subject: dlm: change to single hashtable lock Prepare to replace our own hash table with rhashtable by replacing the per-bucket locks in our own hash table with a single lock. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/debug_fs.c | 24 ++++++++-------- fs/dlm/dir.c | 4 +-- fs/dlm/dlm_internal.h | 2 +- fs/dlm/lock.c | 77 +++++++++++++++++++++++++-------------------------- fs/dlm/lockspace.c | 2 +- fs/dlm/recover.c | 4 +-- fs/dlm/recoverd.c | 8 +++--- 7 files changed, 60 insertions(+), 61 deletions(-) (limited to 'fs') diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index cba5514688ee..b8234eba5e34 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -452,7 +452,7 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; - spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); if (!RB_EMPTY_ROOT(tree)) { for (node = rb_first(tree); node; node = rb_next(node)) { r = rb_entry(node, struct dlm_rsb, res_hashnode); @@ -460,12 +460,12 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return ri; } } } - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); /* * move to the first rsb in the next non-empty bucket @@ -484,18 +484,18 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) } tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; - spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); if (!RB_EMPTY_ROOT(tree)) { node = rb_first(tree); r = rb_entry(node, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); *pos = n; return ri; } - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); } } @@ -516,7 +516,7 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) * move to the next rsb in the same bucket */ - spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); rp = ri->rsb; next = rb_next(&rp->res_hashnode); @@ -524,12 +524,12 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) r = rb_entry(next, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); dlm_put_rsb(rp); ++*pos; return ri; } - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); dlm_put_rsb(rp); /* @@ -550,18 +550,18 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) } tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; - spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); if (!RB_EMPTY_ROOT(tree)) { next = rb_first(tree); r = rb_entry(next, struct dlm_rsb, res_hashnode); dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); *pos = n; return ri; } - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); } } diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index ff3a51c759b5..5315f4f46cc7 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -204,12 +204,12 @@ static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, const char *name, hash = jhash(name, len, 0); bucket = hash & (ls->ls_rsbtbl_size - 1); - spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].keep, name, len, &r); if (rv) rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].toss, name, len, &r); - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); if (!rv) return r; diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 269c12e0824f..2c961db53b27 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -105,7 +105,6 @@ do { \ struct dlm_rsbtable { struct rb_root keep; struct rb_root toss; - spinlock_t lock; unsigned long flags; }; @@ -593,6 +592,7 @@ struct dlm_ls { spinlock_t ls_lkbidr_spin; struct dlm_rsbtable *ls_rsbtbl; + spinlock_t ls_rsbtbl_lock; uint32_t ls_rsbtbl_size; spinlock_t ls_waiters_lock; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 4ff4ef2a5f87..af57d9d12434 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -369,13 +369,12 @@ static inline int dlm_kref_put_lock_bh(struct kref *kref, static void put_rsb(struct dlm_rsb *r) { struct dlm_ls *ls = r->res_ls; - uint32_t bucket = r->res_bucket; int rv; rv = dlm_kref_put_lock_bh(&r->res_ref, toss_rsb, - &ls->ls_rsbtbl[bucket].lock); + &ls->ls_rsbtbl_lock); if (rv) - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); } void dlm_put_rsb(struct dlm_rsb *r) @@ -615,7 +614,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, goto out; } - spin_lock_bh(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (error) @@ -685,7 +684,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, error = get_rsb_struct(ls, name, len, &r); if (error == -EAGAIN) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); goto retry; } if (error) @@ -734,7 +733,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, out_add: error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); out_unlock: - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); out: *r_ret = r; return error; @@ -759,7 +758,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, if (error < 0) goto out; - spin_lock_bh(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (error) @@ -817,7 +816,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, error = get_rsb_struct(ls, name, len, &r); if (error == -EAGAIN) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); goto retry; } if (error) @@ -832,7 +831,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); out_unlock: - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); out: *r_ret = r; return error; @@ -1049,7 +1048,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, if (error < 0) return error; - spin_lock_bh(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (!error) { /* because the rsb is active, we need to lock_rsb before @@ -1057,7 +1056,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, */ hold_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); lock_rsb(r); __dlm_master_lookup(ls, r, our_nodeid, from_nodeid, false, @@ -1083,14 +1082,14 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, r->res_toss_time = jiffies; /* the rsb was inactive (on toss list) */ - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return 0; not_found: error = get_rsb_struct(ls, name, len, &r); if (error == -EAGAIN) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); goto retry; } if (error) @@ -1108,7 +1107,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, if (error) { /* should never happen */ dlm_free_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); goto retry; } @@ -1116,7 +1115,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, *result = DLM_LU_ADD; *r_nodeid = from_nodeid; out_unlock: - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return error; } @@ -1126,15 +1125,15 @@ static void dlm_dump_rsb_hash(struct dlm_ls *ls, uint32_t hash) struct dlm_rsb *r; int i; + spin_lock_bh(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - spin_lock_bh(&ls->ls_rsbtbl[i].lock); for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); if (r->res_hash == hash) dlm_dump_rsb(r); } - spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } + spin_unlock_bh(&ls->ls_rsbtbl_lock); } void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) @@ -1146,7 +1145,7 @@ void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) hash = jhash(name, len, 0); b = hash & (ls->ls_rsbtbl_size - 1); - spin_lock_bh(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); if (!error) goto out_dump; @@ -1157,7 +1156,7 @@ void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) out_dump: dlm_dump_rsb(r); out: - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); } static void toss_rsb(struct kref *kref) @@ -1621,10 +1620,10 @@ static void shrink_bucket(struct dlm_ls *ls, int b) memset(&ls->ls_remove_lens, 0, sizeof(int) * DLM_REMOVE_NAMES_MAX); - spin_lock_bh(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); if (!test_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags)) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return; } @@ -1681,7 +1680,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags); else clear_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); /* * While searching for rsb's to free, we found some that require @@ -1696,16 +1695,16 @@ static void shrink_bucket(struct dlm_ls *ls, int b) name = ls->ls_remove_names[i]; len = ls->ls_remove_lens[i]; - spin_lock_bh(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); if (rv) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); log_debug(ls, "remove_name not toss %s", name); continue; } if (r->res_master_nodeid != our_nodeid) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); log_debug(ls, "remove_name master %d dir %d our %d %s", r->res_master_nodeid, r->res_dir_nodeid, our_nodeid, name); @@ -1714,7 +1713,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) if (r->res_dir_nodeid == our_nodeid) { /* should never happen */ - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); log_error(ls, "remove_name dir %d master %d our %d %s", r->res_dir_nodeid, r->res_master_nodeid, our_nodeid, name); @@ -1723,21 +1722,21 @@ static void shrink_bucket(struct dlm_ls *ls, int b) if (!time_after_eq(jiffies, r->res_toss_time + dlm_config.ci_toss_secs * HZ)) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); log_debug(ls, "remove_name toss_time %lu now %lu %s", r->res_toss_time, jiffies, name); continue; } if (!kref_put(&r->res_ref, kill_rsb)) { - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); log_error(ls, "remove_name in use %s", name); continue; } rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); send_remove(r); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); dlm_free_rsb(r); } @@ -4201,7 +4200,7 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) hash = jhash(name, len, 0); b = hash & (ls->ls_rsbtbl_size - 1); - spin_lock_bh(&ls->ls_rsbtbl[b].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); if (rv) { @@ -4211,7 +4210,7 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) /* should not happen */ log_error(ls, "receive_remove from %d not found %s", from_nodeid, name); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return; } if (r->res_master_nodeid != from_nodeid) { @@ -4219,14 +4218,14 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) log_error(ls, "receive_remove keep from %d master %d", from_nodeid, r->res_master_nodeid); dlm_print_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return; } log_debug(ls, "receive_remove from %d master %d first %x %s", from_nodeid, r->res_master_nodeid, r->res_first_lkid, name); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return; } @@ -4234,19 +4233,19 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) log_error(ls, "receive_remove toss from %d master %d", from_nodeid, r->res_master_nodeid); dlm_print_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return; } if (kref_put(&r->res_ref, kill_rsb)) { rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); dlm_free_rsb(r); } else { log_error(ls, "receive_remove from %d rsb ref error", from_nodeid); dlm_print_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl[b].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); } } @@ -5314,7 +5313,7 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket) struct rb_node *n; struct dlm_rsb *r; - spin_lock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_lock_bh(&ls->ls_rsbtbl_lock); for (n = rb_first(&ls->ls_rsbtbl[bucket].keep); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); @@ -5325,10 +5324,10 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket) continue; } hold_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return r; } - spin_unlock_bh(&ls->ls_rsbtbl[bucket].lock); + spin_unlock_bh(&ls->ls_rsbtbl_lock); return NULL; } diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 731c48371a27..d33dbcd5f4a1 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -495,6 +495,7 @@ static int new_lockspace(const char *name, const char *cluster, */ ls->ls_exflags = (flags & ~(DLM_LSFL_FS | DLM_LSFL_NEWEXCL)); + spin_lock_init(&ls->ls_rsbtbl_lock); size = READ_ONCE(dlm_config.ci_rsbtbl_size); ls->ls_rsbtbl_size = size; @@ -504,7 +505,6 @@ static int new_lockspace(const char *name, const char *cluster, for (i = 0; i < size; i++) { ls->ls_rsbtbl[i].keep.rb_node = NULL; ls->ls_rsbtbl[i].toss.rb_node = NULL; - spin_lock_init(&ls->ls_rsbtbl[i].lock); } for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) { diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 13bc845fa305..9a4c8e4b2442 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -886,8 +886,8 @@ void dlm_clear_toss(struct dlm_ls *ls) unsigned int count = 0; int i; + spin_lock(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - spin_lock_bh(&ls->ls_rsbtbl[i].lock); for (n = rb_first(&ls->ls_rsbtbl[i].toss); n; n = next) { next = rb_next(n); r = rb_entry(n, struct dlm_rsb, res_hashnode); @@ -895,8 +895,8 @@ void dlm_clear_toss(struct dlm_ls *ls) dlm_free_rsb(r); count++; } - spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } + spin_unlock_bh(&ls->ls_rsbtbl_lock); if (count) log_rinfo(ls, "dlm_clear_toss %u done", count); diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index c82cc48988c6..fa6608363302 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -33,8 +33,8 @@ static int dlm_create_masters_list(struct dlm_ls *ls) goto out; } + spin_lock_bh(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - spin_lock_bh(&ls->ls_rsbtbl[i].lock); for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); if (r->res_nodeid) @@ -43,8 +43,8 @@ static int dlm_create_masters_list(struct dlm_ls *ls) list_add(&r->res_masters_list, &ls->ls_masters_list); dlm_hold_rsb(r); } - spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } + spin_unlock_bh(&ls->ls_rsbtbl_lock); out: write_unlock_bh(&ls->ls_masters_lock); return error; @@ -68,8 +68,8 @@ static void dlm_create_root_list(struct dlm_ls *ls, struct list_head *root_list) struct dlm_rsb *r; int i; + spin_lock_bh(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - spin_lock_bh(&ls->ls_rsbtbl[i].lock); for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); list_add(&r->res_root_list, root_list); @@ -78,8 +78,8 @@ static void dlm_create_root_list(struct dlm_ls *ls, struct list_head *root_list) if (!RB_EMPTY_ROOT(&ls->ls_rsbtbl[i].toss)) log_error(ls, "%s toss not empty", __func__); - spin_unlock_bh(&ls->ls_rsbtbl[i].lock); } + spin_unlock_bh(&ls->ls_rsbtbl_lock); } static void dlm_release_root_list(struct list_head *root_list) -- cgit v1.2.3-59-g8ed1b From 2d90354027ad2011c0c5a2a404fe81afc745c2a7 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:37 -0400 Subject: dlm: merge toss and keep hash table lists into one list There are several places where lock processing can perform two hash table lookups, first in the "keep" list, and if not found, in the "toss" list. This patch introduces a new rsb state flag "RSB_TOSS" to represent the difference between the state of being on keep vs toss list, so that the two lists can be combined. This avoids cases of two lookups. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/debug_fs.c | 29 ++++++++++++-- fs/dlm/dir.c | 6 +-- fs/dlm/dlm_internal.h | 4 +- fs/dlm/lock.c | 103 +++++++++++++++++++++++++++----------------------- fs/dlm/lockspace.c | 13 ++----- fs/dlm/recover.c | 7 +++- fs/dlm/recoverd.c | 12 +++--- 7 files changed, 98 insertions(+), 76 deletions(-) (limited to 'fs') diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index b8234eba5e34..37f4dfca5e44 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -450,12 +450,20 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) if (seq->op == &format4_seq_ops) ri->format = 4; - tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; + tree = &ls->ls_rsbtbl[bucket].r; spin_lock_bh(&ls->ls_rsbtbl_lock); if (!RB_EMPTY_ROOT(tree)) { for (node = rb_first(tree); node; node = rb_next(node)) { r = rb_entry(node, struct dlm_rsb, res_hashnode); + if (toss) { + if (!rsb_flag(r, RSB_TOSS)) + continue; + } else { + if (rsb_flag(r, RSB_TOSS)) + continue; + } + if (!entry--) { dlm_hold_rsb(r); ri->rsb = r; @@ -482,12 +490,20 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) kfree(ri); return NULL; } - tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; + tree = &ls->ls_rsbtbl[bucket].r; spin_lock_bh(&ls->ls_rsbtbl_lock); if (!RB_EMPTY_ROOT(tree)) { node = rb_first(tree); r = rb_entry(node, struct dlm_rsb, res_hashnode); + if (toss) { + if (!rsb_flag(r, RSB_TOSS)) + continue; + } else { + if (rsb_flag(r, RSB_TOSS)) + continue; + } + dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; @@ -548,12 +564,19 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) ++*pos; return NULL; } - tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep; + tree = &ls->ls_rsbtbl[bucket].r; spin_lock_bh(&ls->ls_rsbtbl_lock); if (!RB_EMPTY_ROOT(tree)) { next = rb_first(tree); r = rb_entry(next, struct dlm_rsb, res_hashnode); + if (toss) { + if (!rsb_flag(r, RSB_TOSS)) + continue; + } else { + if (rsb_flag(r, RSB_TOSS)) + continue; + } dlm_hold_rsb(r); ri->rsb = r; ri->bucket = bucket; diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index 5315f4f46cc7..f8039f3ee2d1 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -205,12 +205,8 @@ static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, const char *name, bucket = hash & (ls->ls_rsbtbl_size - 1); spin_lock_bh(&ls->ls_rsbtbl_lock); - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].keep, name, len, &r); - if (rv) - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].toss, - name, len, &r); + rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].r, name, len, &r); spin_unlock_bh(&ls->ls_rsbtbl_lock); - if (!rv) return r; diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 2c961db53b27..af88fc2f978c 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -103,8 +103,7 @@ do { \ #define DLM_RTF_SHRINK_BIT 0 struct dlm_rsbtable { - struct rb_root keep; - struct rb_root toss; + struct rb_root r; unsigned long flags; }; @@ -376,6 +375,7 @@ enum rsb_flags { RSB_RECOVER_CONVERT, RSB_RECOVER_GRANT, RSB_RECOVER_LVB_INVAL, + RSB_TOSS, }; static inline void rsb_set_flag(struct dlm_rsb *r, enum rsb_flags flag) diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index af57d9d12434..08ec1a04476a 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -616,23 +616,22 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); if (error) - goto do_toss; + goto do_new; /* * rsb is active, so we can't check master_nodeid without lock_rsb. */ + if (rsb_flag(r, RSB_TOSS)) + goto do_toss; + kref_get(&r->res_ref); goto out_unlock; do_toss: - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); - if (error) - goto do_new; - /* * rsb found inactive (master_nodeid may be out of date unless * we are the dir_nodeid or were the master) No other thread @@ -669,8 +668,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, r->res_first_lkid = 0; } - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); - error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); + rsb_clear_flag(r, RSB_TOSS); goto out_unlock; @@ -731,7 +729,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, } out_add: - error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); + error = rsb_insert(r, &ls->ls_rsbtbl[b].r); out_unlock: spin_unlock_bh(&ls->ls_rsbtbl_lock); out: @@ -760,8 +758,11 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); if (error) + goto do_new; + + if (rsb_flag(r, RSB_TOSS)) goto do_toss; /* @@ -773,10 +774,6 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, do_toss: - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); - if (error) - goto do_new; - /* * rsb found inactive. No other thread is using this rsb because * it's on the toss list, so we can look at or update @@ -804,8 +801,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, r->res_nodeid = 0; } - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); - error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); + rsb_clear_flag(r, RSB_TOSS); goto out_unlock; @@ -829,7 +825,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, r->res_nodeid = (dir_nodeid == our_nodeid) ? 0 : dir_nodeid; kref_init(&r->res_ref); - error = rsb_insert(r, &ls->ls_rsbtbl[b].keep); + error = rsb_insert(r, &ls->ls_rsbtbl[b].r); out_unlock: spin_unlock_bh(&ls->ls_rsbtbl_lock); out: @@ -1049,8 +1045,11 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, return error; spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); if (!error) { + if (rsb_flag(r, RSB_TOSS)) + goto do_toss; + /* because the rsb is active, we need to lock_rsb before * checking/changing re_master_nodeid */ @@ -1067,12 +1066,11 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, put_rsb(r); return 0; - } - - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); - if (error) + } else { goto not_found; + } + do_toss: /* because the rsb is inactive (on toss list), it's not refcounted * and lock_rsb is not used, but is protected by the rsbtbl lock */ @@ -1102,8 +1100,9 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, r->res_nodeid = from_nodeid; kref_init(&r->res_ref); r->res_toss_time = jiffies; + rsb_set_flag(r, RSB_TOSS); - error = rsb_insert(r, &ls->ls_rsbtbl[b].toss); + error = rsb_insert(r, &ls->ls_rsbtbl[b].r); if (error) { /* should never happen */ dlm_free_rsb(r); @@ -1127,8 +1126,11 @@ static void dlm_dump_rsb_hash(struct dlm_ls *ls, uint32_t hash) spin_lock_bh(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { + for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); + if (rsb_flag(r, RSB_TOSS)) + continue; + if (r->res_hash == hash) dlm_dump_rsb(r); } @@ -1146,14 +1148,10 @@ void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) b = hash & (ls->ls_rsbtbl_size - 1); spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); if (!error) - goto out_dump; - - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); - if (error) goto out; - out_dump: + dlm_dump_rsb(r); out: spin_unlock_bh(&ls->ls_rsbtbl_lock); @@ -1166,8 +1164,8 @@ static void toss_rsb(struct kref *kref) DLM_ASSERT(list_empty(&r->res_root_list), dlm_print_rsb(r);); kref_init(&r->res_ref); - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[r->res_bucket].keep); - rsb_insert(r, &ls->ls_rsbtbl[r->res_bucket].toss); + WARN_ON(rsb_flag(r, RSB_TOSS)); + rsb_set_flag(r, RSB_TOSS); r->res_toss_time = jiffies; set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[r->res_bucket].flags); if (r->res_lvbptr) { @@ -1627,9 +1625,11 @@ static void shrink_bucket(struct dlm_ls *ls, int b) return; } - for (n = rb_first(&ls->ls_rsbtbl[b].toss); n; n = next) { + for (n = rb_first(&ls->ls_rsbtbl[b].r); n; n = next) { next = rb_next(n); r = rb_entry(n, struct dlm_rsb, res_hashnode); + if (!rsb_flag(r, RSB_TOSS)) + continue; /* If we're the directory record for this rsb, and we're not the master of it, then we need to wait @@ -1672,7 +1672,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) continue; } - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); + rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); dlm_free_rsb(r); } @@ -1696,8 +1696,14 @@ static void shrink_bucket(struct dlm_ls *ls, int b) len = ls->ls_remove_lens[i]; spin_lock_bh(&ls->ls_rsbtbl_lock); - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); + rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); if (rv) { + spin_unlock_bh(&ls->ls_rsbtbl_lock); + log_error(ls, "remove_name not found %s", name); + continue; + } + + if (!rsb_flag(r, RSB_TOSS)) { spin_unlock_bh(&ls->ls_rsbtbl_lock); log_debug(ls, "remove_name not toss %s", name); continue; @@ -1734,7 +1740,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) continue; } - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); + rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); send_remove(r); spin_unlock_bh(&ls->ls_rsbtbl_lock); @@ -4202,17 +4208,16 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) spin_lock_bh(&ls->ls_rsbtbl_lock); - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].toss, name, len, &r); + rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); if (rv) { - /* verify the rsb is on keep list per comment above */ - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].keep, name, len, &r); - if (rv) { - /* should not happen */ - log_error(ls, "receive_remove from %d not found %s", - from_nodeid, name); - spin_unlock_bh(&ls->ls_rsbtbl_lock); - return; - } + /* should not happen */ + log_error(ls, "%s from %d not found %s", __func__, + from_nodeid, name); + spin_unlock_bh(&ls->ls_rsbtbl_lock); + return; + } + + if (!rsb_flag(r, RSB_TOSS)) { if (r->res_master_nodeid != from_nodeid) { /* should not happen */ log_error(ls, "receive_remove keep from %d master %d", @@ -4238,7 +4243,7 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) } if (kref_put(&r->res_ref, kill_rsb)) { - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].toss); + rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); spin_unlock_bh(&ls->ls_rsbtbl_lock); dlm_free_rsb(r); } else { @@ -5314,8 +5319,10 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket) struct dlm_rsb *r; spin_lock_bh(&ls->ls_rsbtbl_lock); - for (n = rb_first(&ls->ls_rsbtbl[bucket].keep); n; n = rb_next(n)) { + for (n = rb_first(&ls->ls_rsbtbl[bucket].r); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); + if (rsb_flag(r, RSB_TOSS)) + continue; if (!rsb_flag(r, RSB_RECOVER_GRANT)) continue; diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index d33dbcd5f4a1..b5184ad550fa 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -503,8 +503,7 @@ static int new_lockspace(const char *name, const char *cluster, if (!ls->ls_rsbtbl) goto out_lsfree; for (i = 0; i < size; i++) { - ls->ls_rsbtbl[i].keep.rb_node = NULL; - ls->ls_rsbtbl[i].toss.rb_node = NULL; + ls->ls_rsbtbl[i].r.rb_node = NULL; } for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) { @@ -837,15 +836,9 @@ static int release_lockspace(struct dlm_ls *ls, int force) */ for (i = 0; i < ls->ls_rsbtbl_size; i++) { - while ((n = rb_first(&ls->ls_rsbtbl[i].keep))) { + while ((n = rb_first(&ls->ls_rsbtbl[i].r))) { rsb = rb_entry(n, struct dlm_rsb, res_hashnode); - rb_erase(n, &ls->ls_rsbtbl[i].keep); - dlm_free_rsb(rsb); - } - - while ((n = rb_first(&ls->ls_rsbtbl[i].toss))) { - rsb = rb_entry(n, struct dlm_rsb, res_hashnode); - rb_erase(n, &ls->ls_rsbtbl[i].toss); + rb_erase(n, &ls->ls_rsbtbl[i].r); dlm_free_rsb(rsb); } } diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 9a4c8e4b2442..e53d88e4ec93 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -888,10 +888,13 @@ void dlm_clear_toss(struct dlm_ls *ls) spin_lock(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].toss); n; n = next) { + for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = next) { next = rb_next(n); r = rb_entry(n, struct dlm_rsb, res_hashnode); - rb_erase(n, &ls->ls_rsbtbl[i].toss); + if (!rsb_flag(r, RSB_TOSS)) + continue; + + rb_erase(n, &ls->ls_rsbtbl[i].r); dlm_free_rsb(r); count++; } diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index fa6608363302..ad696528ebe7 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -35,9 +35,9 @@ static int dlm_create_masters_list(struct dlm_ls *ls) spin_lock_bh(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { + for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); - if (r->res_nodeid) + if (rsb_flag(r, RSB_TOSS) || r->res_nodeid) continue; list_add(&r->res_masters_list, &ls->ls_masters_list); @@ -70,14 +70,14 @@ static void dlm_create_root_list(struct dlm_ls *ls, struct list_head *root_list) spin_lock_bh(&ls->ls_rsbtbl_lock); for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].keep); n; n = rb_next(n)) { + for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = rb_next(n)) { r = rb_entry(n, struct dlm_rsb, res_hashnode); + if (WARN_ON_ONCE(rsb_flag(r, RSB_TOSS))) + continue; + list_add(&r->res_root_list, root_list); dlm_hold_rsb(r); } - - if (!RB_EMPTY_ROOT(&ls->ls_rsbtbl[i].toss)) - log_error(ls, "%s toss not empty", __func__); } spin_unlock_bh(&ls->ls_rsbtbl_lock); } -- cgit v1.2.3-59-g8ed1b From 93a693d19d2a4aeaa5aede5354cc0f749a780374 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:38 -0400 Subject: dlm: add rsb lists for iteration To prepare for using rhashtable, add two rsb lists for iterating through rsb's in two uncommon cases where this is necesssary: - when dumping rsb state from debugfs, now using seq_list. - when looking at all rsb's during recovery. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/debug_fs.c | 233 ++++++++------------------------------------------ fs/dlm/dlm_internal.h | 4 + fs/dlm/lock.c | 47 +++++----- fs/dlm/lockspace.c | 3 + fs/dlm/recover.c | 24 ++---- fs/dlm/recoverd.c | 34 +++----- 6 files changed, 84 insertions(+), 261 deletions(-) (limited to 'fs') diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 37f4dfca5e44..70567919f1b7 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -366,12 +366,10 @@ static void print_format4(struct dlm_rsb *r, struct seq_file *s) unlock_rsb(r); } -struct rsbtbl_iter { - struct dlm_rsb *rsb; - unsigned bucket; - int format; - int header; -}; +static const struct seq_operations format1_seq_ops; +static const struct seq_operations format2_seq_ops; +static const struct seq_operations format3_seq_ops; +static const struct seq_operations format4_seq_ops; /* * If the buffer is full, seq_printf can be called again, but it @@ -382,220 +380,61 @@ struct rsbtbl_iter { static int table_seq_show(struct seq_file *seq, void *iter_ptr) { - struct rsbtbl_iter *ri = iter_ptr; - - switch (ri->format) { - case 1: - print_format1(ri->rsb, seq); - break; - case 2: - if (ri->header) { - seq_puts(seq, "id nodeid remid pid xid exflags flags sts grmode rqmode time_ms r_nodeid r_len r_name\n"); - ri->header = 0; - } - print_format2(ri->rsb, seq); - break; - case 3: - if (ri->header) { - seq_puts(seq, "rsb ptr nodeid first_lkid flags !root_list_empty !recover_list_empty recover_locks_count len\n"); - ri->header = 0; - } - print_format3(ri->rsb, seq); - break; - case 4: - if (ri->header) { - seq_puts(seq, "rsb ptr nodeid master_nodeid dir_nodeid our_nodeid toss_time flags len str|hex name\n"); - ri->header = 0; - } - print_format4(ri->rsb, seq); - break; - } + struct dlm_rsb *rsb = list_entry(iter_ptr, struct dlm_rsb, res_rsbs_list); + + if (seq->op == &format1_seq_ops) + print_format1(rsb, seq); + else if (seq->op == &format2_seq_ops) + print_format2(rsb, seq); + else if (seq->op == &format3_seq_ops) + print_format3(rsb, seq); + else if (seq->op == &format4_seq_ops) + print_format4(rsb, seq); return 0; } -static const struct seq_operations format1_seq_ops; -static const struct seq_operations format2_seq_ops; -static const struct seq_operations format3_seq_ops; -static const struct seq_operations format4_seq_ops; - static void *table_seq_start(struct seq_file *seq, loff_t *pos) { - struct rb_root *tree; - struct rb_node *node; struct dlm_ls *ls = seq->private; - struct rsbtbl_iter *ri; - struct dlm_rsb *r; - loff_t n = *pos; - unsigned bucket, entry; - int toss = (seq->op == &format4_seq_ops); - - bucket = n >> 32; - entry = n & ((1LL << 32) - 1); - - if (bucket >= ls->ls_rsbtbl_size) - return NULL; - - ri = kzalloc(sizeof(*ri), GFP_NOFS); - if (!ri) - return NULL; - if (n == 0) - ri->header = 1; - if (seq->op == &format1_seq_ops) - ri->format = 1; - if (seq->op == &format2_seq_ops) - ri->format = 2; - if (seq->op == &format3_seq_ops) - ri->format = 3; - if (seq->op == &format4_seq_ops) - ri->format = 4; - - tree = &ls->ls_rsbtbl[bucket].r; + struct list_head *list; - spin_lock_bh(&ls->ls_rsbtbl_lock); - if (!RB_EMPTY_ROOT(tree)) { - for (node = rb_first(tree); node; node = rb_next(node)) { - r = rb_entry(node, struct dlm_rsb, res_hashnode); - if (toss) { - if (!rsb_flag(r, RSB_TOSS)) - continue; - } else { - if (rsb_flag(r, RSB_TOSS)) - continue; - } - - if (!entry--) { - dlm_hold_rsb(r); - ri->rsb = r; - ri->bucket = bucket; - spin_unlock_bh(&ls->ls_rsbtbl_lock); - return ri; - } - } + if (!*pos) { + if (seq->op == &format2_seq_ops) + seq_puts(seq, "id nodeid remid pid xid exflags flags sts grmode rqmode time_ms r_nodeid r_len r_name\n"); + else if (seq->op == &format3_seq_ops) + seq_puts(seq, "rsb ptr nodeid first_lkid flags !root_list_empty !recover_list_empty recover_locks_count len\n"); + else if (seq->op == &format4_seq_ops) + seq_puts(seq, "rsb ptr nodeid master_nodeid dir_nodeid our_nodeid toss_time flags len str|hex name\n"); } - spin_unlock_bh(&ls->ls_rsbtbl_lock); - - /* - * move to the first rsb in the next non-empty bucket - */ - - /* zero the entry */ - n &= ~((1LL << 32) - 1); - while (1) { - bucket++; - n += 1LL << 32; + if (seq->op == &format4_seq_ops) + list = &ls->ls_toss; + else + list = &ls->ls_keep; - if (bucket >= ls->ls_rsbtbl_size) { - kfree(ri); - return NULL; - } - tree = &ls->ls_rsbtbl[bucket].r; - - spin_lock_bh(&ls->ls_rsbtbl_lock); - if (!RB_EMPTY_ROOT(tree)) { - node = rb_first(tree); - r = rb_entry(node, struct dlm_rsb, res_hashnode); - if (toss) { - if (!rsb_flag(r, RSB_TOSS)) - continue; - } else { - if (rsb_flag(r, RSB_TOSS)) - continue; - } - - dlm_hold_rsb(r); - ri->rsb = r; - ri->bucket = bucket; - spin_unlock_bh(&ls->ls_rsbtbl_lock); - *pos = n; - return ri; - } - spin_unlock_bh(&ls->ls_rsbtbl_lock); - } + spin_lock_bh(&ls->ls_rsbtbl_lock); + return seq_list_start(list, *pos); } static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos) { struct dlm_ls *ls = seq->private; - struct rsbtbl_iter *ri = iter_ptr; - struct rb_root *tree; - struct rb_node *next; - struct dlm_rsb *r, *rp; - loff_t n = *pos; - unsigned bucket; - int toss = (seq->op == &format4_seq_ops); - - bucket = n >> 32; - - /* - * move to the next rsb in the same bucket - */ - - spin_lock_bh(&ls->ls_rsbtbl_lock); - rp = ri->rsb; - next = rb_next(&rp->res_hashnode); - - if (next) { - r = rb_entry(next, struct dlm_rsb, res_hashnode); - dlm_hold_rsb(r); - ri->rsb = r; - spin_unlock_bh(&ls->ls_rsbtbl_lock); - dlm_put_rsb(rp); - ++*pos; - return ri; - } - spin_unlock_bh(&ls->ls_rsbtbl_lock); - dlm_put_rsb(rp); - - /* - * move to the first rsb in the next non-empty bucket - */ + struct list_head *list; - /* zero the entry */ - n &= ~((1LL << 32) - 1); - - while (1) { - bucket++; - n += 1LL << 32; + if (seq->op == &format4_seq_ops) + list = &ls->ls_toss; + else + list = &ls->ls_keep; - if (bucket >= ls->ls_rsbtbl_size) { - kfree(ri); - ++*pos; - return NULL; - } - tree = &ls->ls_rsbtbl[bucket].r; - - spin_lock_bh(&ls->ls_rsbtbl_lock); - if (!RB_EMPTY_ROOT(tree)) { - next = rb_first(tree); - r = rb_entry(next, struct dlm_rsb, res_hashnode); - if (toss) { - if (!rsb_flag(r, RSB_TOSS)) - continue; - } else { - if (rsb_flag(r, RSB_TOSS)) - continue; - } - dlm_hold_rsb(r); - ri->rsb = r; - ri->bucket = bucket; - spin_unlock_bh(&ls->ls_rsbtbl_lock); - *pos = n; - return ri; - } - spin_unlock_bh(&ls->ls_rsbtbl_lock); - } + return seq_list_next(iter_ptr, list, pos); } static void table_seq_stop(struct seq_file *seq, void *iter_ptr) { - struct rsbtbl_iter *ri = iter_ptr; + struct dlm_ls *ls = seq->private; - if (ri) { - dlm_put_rsb(ri->rsb); - kfree(ri); - } + spin_unlock_bh(&ls->ls_rsbtbl_lock); } static const struct seq_operations format1_seq_ops = { diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index af88fc2f978c..6d06840029c3 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -339,6 +339,7 @@ struct dlm_rsb { struct list_head res_convertqueue; struct list_head res_waitqueue; + struct list_head res_rsbs_list; struct list_head res_root_list; /* used for recovery */ struct list_head res_masters_list; /* used for recovery */ struct list_head res_recover_list; /* used for recovery */ @@ -595,6 +596,9 @@ struct dlm_ls { spinlock_t ls_rsbtbl_lock; uint32_t ls_rsbtbl_size; + struct list_head ls_toss; + struct list_head ls_keep; + spinlock_t ls_waiters_lock; struct list_head ls_waiters; /* lkbs needing a reply */ diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 08ec1a04476a..a70b8edb5d3f 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -668,6 +668,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, r->res_first_lkid = 0; } + list_move(&r->res_rsbs_list, &ls->ls_keep); rsb_clear_flag(r, RSB_TOSS); goto out_unlock; @@ -730,6 +731,8 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, out_add: error = rsb_insert(r, &ls->ls_rsbtbl[b].r); + if (!error) + list_add(&r->res_rsbs_list, &ls->ls_keep); out_unlock: spin_unlock_bh(&ls->ls_rsbtbl_lock); out: @@ -801,6 +804,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, r->res_nodeid = 0; } + list_move(&r->res_rsbs_list, &ls->ls_keep); rsb_clear_flag(r, RSB_TOSS); goto out_unlock; @@ -826,6 +830,8 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, kref_init(&r->res_ref); error = rsb_insert(r, &ls->ls_rsbtbl[b].r); + if (!error) + list_add(&r->res_rsbs_list, &ls->ls_keep); out_unlock: spin_unlock_bh(&ls->ls_rsbtbl_lock); out: @@ -1110,6 +1116,8 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, goto retry; } + list_add(&r->res_rsbs_list, &ls->ls_toss); + if (result) *result = DLM_LU_ADD; *r_nodeid = from_nodeid; @@ -1120,20 +1128,12 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, static void dlm_dump_rsb_hash(struct dlm_ls *ls, uint32_t hash) { - struct rb_node *n; struct dlm_rsb *r; - int i; spin_lock_bh(&ls->ls_rsbtbl_lock); - for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = rb_next(n)) { - r = rb_entry(n, struct dlm_rsb, res_hashnode); - if (rsb_flag(r, RSB_TOSS)) - continue; - - if (r->res_hash == hash) - dlm_dump_rsb(r); - } + list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { + if (r->res_hash == hash) + dlm_dump_rsb(r); } spin_unlock_bh(&ls->ls_rsbtbl_lock); } @@ -1166,6 +1166,7 @@ static void toss_rsb(struct kref *kref) kref_init(&r->res_ref); WARN_ON(rsb_flag(r, RSB_TOSS)); rsb_set_flag(r, RSB_TOSS); + list_move(&r->res_rsbs_list, &ls->ls_toss); r->res_toss_time = jiffies; set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[r->res_bucket].flags); if (r->res_lvbptr) { @@ -1672,6 +1673,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) continue; } + list_del(&r->res_rsbs_list); rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); dlm_free_rsb(r); } @@ -1740,6 +1742,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) continue; } + list_del(&r->res_rsbs_list); rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); send_remove(r); spin_unlock_bh(&ls->ls_rsbtbl_lock); @@ -4243,6 +4246,7 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) } if (kref_put(&r->res_ref, kill_rsb)) { + list_del(&r->res_rsbs_list); rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); spin_unlock_bh(&ls->ls_rsbtbl_lock); dlm_free_rsb(r); @@ -5313,17 +5317,12 @@ void dlm_recover_purge(struct dlm_ls *ls, const struct list_head *root_list) lkb_count, nodes_count); } -static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket) +static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls) { - struct rb_node *n; struct dlm_rsb *r; spin_lock_bh(&ls->ls_rsbtbl_lock); - for (n = rb_first(&ls->ls_rsbtbl[bucket].r); n; n = rb_next(n)) { - r = rb_entry(n, struct dlm_rsb, res_hashnode); - if (rsb_flag(r, RSB_TOSS)) - continue; - + list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { if (!rsb_flag(r, RSB_RECOVER_GRANT)) continue; if (!is_master(r)) { @@ -5358,19 +5357,15 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls, int bucket) void dlm_recover_grant(struct dlm_ls *ls) { struct dlm_rsb *r; - int bucket = 0; unsigned int count = 0; unsigned int rsb_count = 0; unsigned int lkb_count = 0; while (1) { - r = find_grant_rsb(ls, bucket); - if (!r) { - if (bucket == ls->ls_rsbtbl_size - 1) - break; - bucket++; - continue; - } + r = find_grant_rsb(ls); + if (!r) + break; + rsb_count++; count = 0; lock_rsb(r); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index b5184ad550fa..2b5771a7bf31 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -495,6 +495,8 @@ static int new_lockspace(const char *name, const char *cluster, */ ls->ls_exflags = (flags & ~(DLM_LSFL_FS | DLM_LSFL_NEWEXCL)); + INIT_LIST_HEAD(&ls->ls_toss); + INIT_LIST_HEAD(&ls->ls_keep); spin_lock_init(&ls->ls_rsbtbl_lock); size = READ_ONCE(dlm_config.ci_rsbtbl_size); ls->ls_rsbtbl_size = size; @@ -838,6 +840,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) for (i = 0; i < ls->ls_rsbtbl_size; i++) { while ((n = rb_first(&ls->ls_rsbtbl[i].r))) { rsb = rb_entry(n, struct dlm_rsb, res_hashnode); + list_del(&rsb->res_rsbs_list); rb_erase(n, &ls->ls_rsbtbl[i].r); dlm_free_rsb(rsb); } diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index e53d88e4ec93..512c1ae81a96 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -881,23 +881,15 @@ void dlm_recover_rsbs(struct dlm_ls *ls, const struct list_head *root_list) void dlm_clear_toss(struct dlm_ls *ls) { - struct rb_node *n, *next; - struct dlm_rsb *r; + struct dlm_rsb *r, *safe; unsigned int count = 0; - int i; - - spin_lock(&ls->ls_rsbtbl_lock); - for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = next) { - next = rb_next(n); - r = rb_entry(n, struct dlm_rsb, res_hashnode); - if (!rsb_flag(r, RSB_TOSS)) - continue; - - rb_erase(n, &ls->ls_rsbtbl[i].r); - dlm_free_rsb(r); - count++; - } + + spin_lock_bh(&ls->ls_rsbtbl_lock); + list_for_each_entry_safe(r, safe, &ls->ls_toss, res_rsbs_list) { + list_del(&r->res_rsbs_list); + rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[r->res_bucket].r); + dlm_free_rsb(r); + count++; } spin_unlock_bh(&ls->ls_rsbtbl_lock); diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index ad696528ebe7..5e8e10030b74 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -22,9 +22,8 @@ static int dlm_create_masters_list(struct dlm_ls *ls) { - struct rb_node *n; struct dlm_rsb *r; - int i, error = 0; + int error = 0; write_lock_bh(&ls->ls_masters_lock); if (!list_empty(&ls->ls_masters_list)) { @@ -34,15 +33,12 @@ static int dlm_create_masters_list(struct dlm_ls *ls) } spin_lock_bh(&ls->ls_rsbtbl_lock); - for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = rb_next(n)) { - r = rb_entry(n, struct dlm_rsb, res_hashnode); - if (rsb_flag(r, RSB_TOSS) || r->res_nodeid) - continue; - - list_add(&r->res_masters_list, &ls->ls_masters_list); - dlm_hold_rsb(r); - } + list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { + if (r->res_nodeid) + continue; + + list_add(&r->res_masters_list, &ls->ls_masters_list); + dlm_hold_rsb(r); } spin_unlock_bh(&ls->ls_rsbtbl_lock); out: @@ -64,21 +60,15 @@ static void dlm_release_masters_list(struct dlm_ls *ls) static void dlm_create_root_list(struct dlm_ls *ls, struct list_head *root_list) { - struct rb_node *n; struct dlm_rsb *r; - int i; spin_lock_bh(&ls->ls_rsbtbl_lock); - for (i = 0; i < ls->ls_rsbtbl_size; i++) { - for (n = rb_first(&ls->ls_rsbtbl[i].r); n; n = rb_next(n)) { - r = rb_entry(n, struct dlm_rsb, res_hashnode); - if (WARN_ON_ONCE(rsb_flag(r, RSB_TOSS))) - continue; - - list_add(&r->res_root_list, root_list); - dlm_hold_rsb(r); - } + list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { + list_add(&r->res_root_list, root_list); + dlm_hold_rsb(r); } + + WARN_ON_ONCE(!list_empty(&ls->ls_toss)); spin_unlock_bh(&ls->ls_rsbtbl_lock); } -- cgit v1.2.3-59-g8ed1b From 6c648035cbe75d78836f6d7d2fdd9d996048a66b Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:39 -0400 Subject: dlm: switch to use rhashtable for rsbs Replace our own hash table with the more advanced rhashtable for keeping rsb structs. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/config.c | 8 +++ fs/dlm/config.h | 2 + fs/dlm/dir.c | 6 +- fs/dlm/dlm_internal.h | 18 ++---- fs/dlm/lock.c | 172 ++++++++++++++++---------------------------------- fs/dlm/lock.h | 2 +- fs/dlm/lockspace.c | 35 ++++------ fs/dlm/recover.c | 3 +- 8 files changed, 86 insertions(+), 160 deletions(-) (limited to 'fs') diff --git a/fs/dlm/config.c b/fs/dlm/config.c index e55e0a2cd2e8..517fa975dc5a 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -63,6 +63,14 @@ static void release_node(struct config_item *); static struct configfs_attribute *comm_attrs[]; static struct configfs_attribute *node_attrs[]; +const struct rhashtable_params dlm_rhash_rsb_params = { + .nelem_hint = 3, /* start small */ + .key_len = DLM_RESNAME_MAXLEN, + .key_offset = offsetof(struct dlm_rsb, res_name), + .head_offset = offsetof(struct dlm_rsb, res_node), + .automatic_shrinking = true, +}; + struct dlm_cluster { struct config_group group; unsigned int cl_tcp_port; diff --git a/fs/dlm/config.h b/fs/dlm/config.h index 4c91fcca0fd4..ed237d910208 100644 --- a/fs/dlm/config.h +++ b/fs/dlm/config.h @@ -21,6 +21,8 @@ struct dlm_config_node { uint32_t comm_seq; }; +extern const struct rhashtable_params dlm_rhash_rsb_params; + #define DLM_MAX_ADDR_COUNT 3 #define DLM_PROTO_TCP 0 diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index f8039f3ee2d1..9687f908476b 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -198,14 +198,10 @@ static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, const char *name, int len) { struct dlm_rsb *r; - uint32_t hash, bucket; int rv; - hash = jhash(name, len, 0); - bucket = hash & (ls->ls_rsbtbl_size - 1); - spin_lock_bh(&ls->ls_rsbtbl_lock); - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[bucket].r, name, len, &r); + rv = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); spin_unlock_bh(&ls->ls_rsbtbl_lock); if (!rv) return r; diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 6d06840029c3..cf43b97cf3e5 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -34,6 +34,7 @@ #include #include #include +#include #include #include #include @@ -99,15 +100,6 @@ do { \ } \ } - -#define DLM_RTF_SHRINK_BIT 0 - -struct dlm_rsbtable { - struct rb_root r; - unsigned long flags; -}; - - /* * Lockspace member (per node in a ls) */ @@ -327,13 +319,12 @@ struct dlm_rsb { int res_id; /* for ls_recover_idr */ uint32_t res_lvbseq; uint32_t res_hash; - uint32_t res_bucket; /* rsbtbl */ unsigned long res_toss_time; uint32_t res_first_lkid; struct list_head res_lookup; /* lkbs waiting on first */ union { struct list_head res_hashchain; - struct rb_node res_hashnode; /* rsbtbl */ + struct rhash_head res_node; /* rsbtbl */ }; struct list_head res_grantqueue; struct list_head res_convertqueue; @@ -592,9 +583,10 @@ struct dlm_ls { struct idr ls_lkbidr; spinlock_t ls_lkbidr_spin; - struct dlm_rsbtable *ls_rsbtbl; + struct rhashtable ls_rsbtbl; +#define DLM_RTF_SHRINK_BIT 0 + unsigned long ls_rsbtbl_flags; spinlock_t ls_rsbtbl_lock; - uint32_t ls_rsbtbl_size; struct list_head ls_toss; struct list_head ls_keep; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index a70b8edb5d3f..defb90b56b72 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -436,8 +436,6 @@ static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len, r = list_first_entry(&ls->ls_new_rsb, struct dlm_rsb, res_hashchain); list_del(&r->res_hashchain); - /* Convert the empty list_head to a NULL rb_node for tree usage: */ - memset(&r->res_hashnode, 0, sizeof(struct rb_node)); ls->ls_new_rsb_count--; spin_unlock_bh(&ls->ls_new_rsb_spin); @@ -458,67 +456,31 @@ static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len, return 0; } -static int rsb_cmp(struct dlm_rsb *r, const char *name, int nlen) +int dlm_search_rsb_tree(struct rhashtable *rhash, const void *name, int len, + struct dlm_rsb **r_ret) { - char maxname[DLM_RESNAME_MAXLEN]; + char key[DLM_RESNAME_MAXLEN] = {}; - memset(maxname, 0, DLM_RESNAME_MAXLEN); - memcpy(maxname, name, nlen); - return memcmp(r->res_name, maxname, DLM_RESNAME_MAXLEN); -} + memcpy(key, name, len); + *r_ret = rhashtable_lookup_fast(rhash, &key, dlm_rhash_rsb_params); + if (*r_ret) + return 0; -int dlm_search_rsb_tree(struct rb_root *tree, const void *name, int len, - struct dlm_rsb **r_ret) -{ - struct rb_node *node = tree->rb_node; - struct dlm_rsb *r; - int rc; - - while (node) { - r = rb_entry(node, struct dlm_rsb, res_hashnode); - rc = rsb_cmp(r, name, len); - if (rc < 0) - node = node->rb_left; - else if (rc > 0) - node = node->rb_right; - else - goto found; - } - *r_ret = NULL; return -EBADR; - - found: - *r_ret = r; - return 0; } -static int rsb_insert(struct dlm_rsb *rsb, struct rb_root *tree) +static int rsb_insert(struct dlm_rsb *rsb, struct rhashtable *rhash) { - struct rb_node **newn = &tree->rb_node; - struct rb_node *parent = NULL; - int rc; - - while (*newn) { - struct dlm_rsb *cur = rb_entry(*newn, struct dlm_rsb, - res_hashnode); + int rv; - parent = *newn; - rc = rsb_cmp(cur, rsb->res_name, rsb->res_length); - if (rc < 0) - newn = &parent->rb_left; - else if (rc > 0) - newn = &parent->rb_right; - else { - log_print("rsb_insert match"); - dlm_dump_rsb(rsb); - dlm_dump_rsb(cur); - return -EEXIST; - } + rv = rhashtable_insert_fast(rhash, &rsb->res_node, + dlm_rhash_rsb_params); + if (rv == -EEXIST) { + log_print("%s match", __func__); + dlm_dump_rsb(rsb); } - rb_link_node(&rsb->res_hashnode, parent, newn); - rb_insert_color(&rsb->res_hashnode, tree); - return 0; + return rv; } /* @@ -566,8 +528,7 @@ static int rsb_insert(struct dlm_rsb *rsb, struct rb_root *tree) */ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, - uint32_t hash, uint32_t b, - int dir_nodeid, int from_nodeid, + uint32_t hash, int dir_nodeid, int from_nodeid, unsigned int flags, struct dlm_rsb **r_ret) { struct dlm_rsb *r = NULL; @@ -616,7 +577,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (error) goto do_new; @@ -690,7 +651,6 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, goto out_unlock; r->res_hash = hash; - r->res_bucket = b; r->res_dir_nodeid = dir_nodeid; kref_init(&r->res_ref); @@ -730,7 +690,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, } out_add: - error = rsb_insert(r, &ls->ls_rsbtbl[b].r); + error = rsb_insert(r, &ls->ls_rsbtbl); if (!error) list_add(&r->res_rsbs_list, &ls->ls_keep); out_unlock: @@ -745,8 +705,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, dlm_recover_masters). */ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, - uint32_t hash, uint32_t b, - int dir_nodeid, int from_nodeid, + uint32_t hash, int dir_nodeid, int from_nodeid, unsigned int flags, struct dlm_rsb **r_ret) { struct dlm_rsb *r = NULL; @@ -761,7 +720,7 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (error) goto do_new; @@ -823,13 +782,12 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, goto out_unlock; r->res_hash = hash; - r->res_bucket = b; r->res_dir_nodeid = dir_nodeid; r->res_master_nodeid = dir_nodeid; r->res_nodeid = (dir_nodeid == our_nodeid) ? 0 : dir_nodeid; kref_init(&r->res_ref); - error = rsb_insert(r, &ls->ls_rsbtbl[b].r); + error = rsb_insert(r, &ls->ls_rsbtbl); if (!error) list_add(&r->res_rsbs_list, &ls->ls_keep); out_unlock: @@ -843,23 +801,21 @@ static int find_rsb(struct dlm_ls *ls, const void *name, int len, int from_nodeid, unsigned int flags, struct dlm_rsb **r_ret) { - uint32_t hash, b; int dir_nodeid; + uint32_t hash; if (len > DLM_RESNAME_MAXLEN) return -EINVAL; hash = jhash(name, len, 0); - b = hash & (ls->ls_rsbtbl_size - 1); - dir_nodeid = dlm_hash2nodeid(ls, hash); if (dlm_no_directory(ls)) - return find_rsb_nodir(ls, name, len, hash, b, dir_nodeid, + return find_rsb_nodir(ls, name, len, hash, dir_nodeid, from_nodeid, flags, r_ret); else - return find_rsb_dir(ls, name, len, hash, b, dir_nodeid, - from_nodeid, flags, r_ret); + return find_rsb_dir(ls, name, len, hash, dir_nodeid, + from_nodeid, flags, r_ret); } /* we have received a request and found that res_master_nodeid != our_nodeid, @@ -1020,7 +976,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, int len, unsigned int flags, int *r_nodeid, int *result) { struct dlm_rsb *r = NULL; - uint32_t hash, b; + uint32_t hash; int our_nodeid = dlm_our_nodeid(); int dir_nodeid, error; @@ -1034,8 +990,6 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, } hash = jhash(name, len, 0); - b = hash & (ls->ls_rsbtbl_size - 1); - dir_nodeid = dlm_hash2nodeid(ls, hash); if (dir_nodeid != our_nodeid) { log_error(ls, "dlm_master_lookup from %d dir %d our %d h %x %d", @@ -1051,7 +1005,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, return error; spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (!error) { if (rsb_flag(r, RSB_TOSS)) goto do_toss; @@ -1100,7 +1054,6 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, goto out_unlock; r->res_hash = hash; - r->res_bucket = b; r->res_dir_nodeid = our_nodeid; r->res_master_nodeid = from_nodeid; r->res_nodeid = from_nodeid; @@ -1108,7 +1061,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, r->res_toss_time = jiffies; rsb_set_flag(r, RSB_TOSS); - error = rsb_insert(r, &ls->ls_rsbtbl[b].r); + error = rsb_insert(r, &ls->ls_rsbtbl); if (error) { /* should never happen */ dlm_free_rsb(r); @@ -1141,14 +1094,10 @@ static void dlm_dump_rsb_hash(struct dlm_ls *ls, uint32_t hash) void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) { struct dlm_rsb *r = NULL; - uint32_t hash, b; int error; - hash = jhash(name, len, 0); - b = hash & (ls->ls_rsbtbl_size - 1); - spin_lock_bh(&ls->ls_rsbtbl_lock); - error = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); + error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (!error) goto out; @@ -1168,7 +1117,7 @@ static void toss_rsb(struct kref *kref) rsb_set_flag(r, RSB_TOSS); list_move(&r->res_rsbs_list, &ls->ls_toss); r->res_toss_time = jiffies; - set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[r->res_bucket].flags); + set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags); if (r->res_lvbptr) { dlm_free_lvb(r->res_lvbptr); r->res_lvbptr = NULL; @@ -1607,10 +1556,9 @@ static int remove_from_waiters_ms(struct dlm_lkb *lkb, return error; } -static void shrink_bucket(struct dlm_ls *ls, int b) +static void shrink_bucket(struct dlm_ls *ls) { - struct rb_node *n, *next; - struct dlm_rsb *r; + struct dlm_rsb *r, *safe; char *name; int our_nodeid = dlm_our_nodeid(); int remote_count = 0; @@ -1621,17 +1569,12 @@ static void shrink_bucket(struct dlm_ls *ls, int b) spin_lock_bh(&ls->ls_rsbtbl_lock); - if (!test_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags)) { + if (!test_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags)) { spin_unlock_bh(&ls->ls_rsbtbl_lock); return; } - for (n = rb_first(&ls->ls_rsbtbl[b].r); n; n = next) { - next = rb_next(n); - r = rb_entry(n, struct dlm_rsb, res_hashnode); - if (!rsb_flag(r, RSB_TOSS)) - continue; - + list_for_each_entry_safe(r, safe, &ls->ls_toss, res_rsbs_list) { /* If we're the directory record for this rsb, and we're not the master of it, then we need to wait for the master node to send us a dir remove for @@ -1674,14 +1617,15 @@ static void shrink_bucket(struct dlm_ls *ls, int b) } list_del(&r->res_rsbs_list); - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); + rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, + dlm_rhash_rsb_params); dlm_free_rsb(r); } if (need_shrink) - set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags); + set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags); else - clear_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl[b].flags); + clear_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags); spin_unlock_bh(&ls->ls_rsbtbl_lock); /* @@ -1698,7 +1642,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) len = ls->ls_remove_lens[i]; spin_lock_bh(&ls->ls_rsbtbl_lock); - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); + rv = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (rv) { spin_unlock_bh(&ls->ls_rsbtbl_lock); log_error(ls, "remove_name not found %s", name); @@ -1743,7 +1687,8 @@ static void shrink_bucket(struct dlm_ls *ls, int b) } list_del(&r->res_rsbs_list); - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); + rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, + dlm_rhash_rsb_params); send_remove(r); spin_unlock_bh(&ls->ls_rsbtbl_lock); @@ -1753,14 +1698,7 @@ static void shrink_bucket(struct dlm_ls *ls, int b) void dlm_scan_rsbs(struct dlm_ls *ls) { - int i; - - for (i = 0; i < ls->ls_rsbtbl_size; i++) { - shrink_bucket(ls, i); - if (dlm_locking_stopped(ls)) - break; - cond_resched(); - } + shrink_bucket(ls); } /* lkb is master or local copy */ @@ -4174,7 +4112,6 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) { char name[DLM_RESNAME_MAXLEN+1]; struct dlm_rsb *r; - uint32_t hash, b; int rv, len, dir_nodeid, from_nodeid; from_nodeid = le32_to_cpu(ms->m_header.h_nodeid); @@ -4194,24 +4131,22 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) return; } - /* Look for name on rsbtbl.toss, if it's there, kill it. - If it's on rsbtbl.keep, it's being used, and we should ignore this - message. This is an expected race between the dir node sending a - request to the master node at the same time as the master node sends - a remove to the dir node. The resolution to that race is for the - dir node to ignore the remove message, and the master node to - recreate the master rsb when it gets a request from the dir node for - an rsb it doesn't have. */ + /* Look for name in rsb toss state, if it's there, kill it. + * If it's in non toss state, it's being used, and we should ignore this + * message. This is an expected race between the dir node sending a + * request to the master node at the same time as the master node sends + * a remove to the dir node. The resolution to that race is for the + * dir node to ignore the remove message, and the master node to + * recreate the master rsb when it gets a request from the dir node for + * an rsb it doesn't have. + */ memset(name, 0, sizeof(name)); memcpy(name, ms->m_extra, len); - hash = jhash(name, len, 0); - b = hash & (ls->ls_rsbtbl_size - 1); - spin_lock_bh(&ls->ls_rsbtbl_lock); - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl[b].r, name, len, &r); + rv = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (rv) { /* should not happen */ log_error(ls, "%s from %d not found %s", __func__, @@ -4247,7 +4182,8 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) if (kref_put(&r->res_ref, kill_rsb)) { list_del(&r->res_rsbs_list); - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[b].r); + rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, + dlm_rhash_rsb_params); spin_unlock_bh(&ls->ls_rsbtbl_lock); dlm_free_rsb(r); } else { diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index 45a74869810a..33616d4b0cdb 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -29,7 +29,7 @@ void dlm_unlock_recovery(struct dlm_ls *ls); int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, int len, unsigned int flags, int *r_nodeid, int *result); -int dlm_search_rsb_tree(struct rb_root *tree, const void *name, int len, +int dlm_search_rsb_tree(struct rhashtable *rhash, const void *name, int len, struct dlm_rsb **r_ret); void dlm_recover_purge(struct dlm_ls *ls, const struct list_head *root_list); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 2b5771a7bf31..890e1a4cf787 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -410,9 +410,9 @@ static int new_lockspace(const char *name, const char *cluster, int *ops_result, dlm_lockspace_t **lockspace) { struct dlm_ls *ls; - int i, size, error; int do_unreg = 0; int namelen = strlen(name); + int i, error; if (namelen > DLM_LOCKSPACE_LEN || namelen == 0) return -EINVAL; @@ -498,15 +498,10 @@ static int new_lockspace(const char *name, const char *cluster, INIT_LIST_HEAD(&ls->ls_toss); INIT_LIST_HEAD(&ls->ls_keep); spin_lock_init(&ls->ls_rsbtbl_lock); - size = READ_ONCE(dlm_config.ci_rsbtbl_size); - ls->ls_rsbtbl_size = size; - ls->ls_rsbtbl = vmalloc(array_size(size, sizeof(struct dlm_rsbtable))); - if (!ls->ls_rsbtbl) + error = rhashtable_init(&ls->ls_rsbtbl, &dlm_rhash_rsb_params); + if (error) goto out_lsfree; - for (i = 0; i < size; i++) { - ls->ls_rsbtbl[i].r.rb_node = NULL; - } for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) { ls->ls_remove_names[i] = kzalloc(DLM_RESNAME_MAXLEN+1, @@ -669,7 +664,7 @@ static int new_lockspace(const char *name, const char *cluster, out_rsbtbl: for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) kfree(ls->ls_remove_names[i]); - vfree(ls->ls_rsbtbl); + rhashtable_destroy(&ls->ls_rsbtbl); out_lsfree: if (do_unreg) kobject_put(&ls->ls_kobj); @@ -772,10 +767,16 @@ static int lockspace_busy(struct dlm_ls *ls, int force) return rv; } +static void rhash_free_rsb(void *ptr, void *arg) +{ + struct dlm_rsb *rsb = ptr; + + dlm_free_rsb(rsb); +} + static int release_lockspace(struct dlm_ls *ls, int force) { struct dlm_rsb *rsb; - struct rb_node *n; int i, busy, rv; busy = lockspace_busy(ls, force); @@ -834,19 +835,9 @@ static int release_lockspace(struct dlm_ls *ls, int force) idr_destroy(&ls->ls_lkbidr); /* - * Free all rsb's on rsbtbl[] lists + * Free all rsb's on rsbtbl */ - - for (i = 0; i < ls->ls_rsbtbl_size; i++) { - while ((n = rb_first(&ls->ls_rsbtbl[i].r))) { - rsb = rb_entry(n, struct dlm_rsb, res_hashnode); - list_del(&rsb->res_rsbs_list); - rb_erase(n, &ls->ls_rsbtbl[i].r); - dlm_free_rsb(rsb); - } - } - - vfree(ls->ls_rsbtbl); + rhashtable_free_and_destroy(&ls->ls_rsbtbl, rhash_free_rsb, NULL); for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) kfree(ls->ls_remove_names[i]); diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 512c1ae81a96..c21ef115123b 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -887,7 +887,8 @@ void dlm_clear_toss(struct dlm_ls *ls) spin_lock_bh(&ls->ls_rsbtbl_lock); list_for_each_entry_safe(r, safe, &ls->ls_toss, res_rsbs_list) { list_del(&r->res_rsbs_list); - rb_erase(&r->res_hashnode, &ls->ls_rsbtbl[r->res_bucket].r); + rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, + dlm_rhash_rsb_params); dlm_free_rsb(r); count++; } -- cgit v1.2.3-59-g8ed1b From 6644925abf056030cd9efc73fc05ea5a5df4f59f Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:40 -0400 Subject: dlm: do not use ref counts for rsb in the toss state In the past we had problems when an rsb had a reference counter greater than one while in the toss state. An rsb in the toss state is not actively used for locking, and should not have any other references apart from the single ref keeping it on the rsb hash. Shift to freeing rsb's directly rather than using kref_put to free them, since the ref counting is not meant to be used in this state. Add warnings if ref counting is seen while an rsb is in the toss state. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lock.c | 61 ++++++++++++++++++++++++++++---------------------------- fs/dlm/lock.h | 1 + fs/dlm/recover.c | 2 +- 3 files changed, 32 insertions(+), 32 deletions(-) (limited to 'fs') diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index defb90b56b72..fee1a4164fc1 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -325,6 +325,8 @@ static void queue_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rqmode) static inline void hold_rsb(struct dlm_rsb *r) { + /* rsbs in toss state never get referenced */ + WARN_ON(rsb_flag(r, RSB_TOSS)); kref_get(&r->res_ref); } @@ -631,6 +633,11 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, list_move(&r->res_rsbs_list, &ls->ls_keep); rsb_clear_flag(r, RSB_TOSS); + /* rsb got out of toss state, it becomes alive again + * and we reinit the reference counter that is only + * valid for keep state rsbs + */ + kref_init(&r->res_ref); goto out_unlock; @@ -765,6 +772,11 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, list_move(&r->res_rsbs_list, &ls->ls_keep); rsb_clear_flag(r, RSB_TOSS); + /* rsb got out of toss state, it becomes alive again + * and we reinit the reference counter that is only + * valid for keep state rsbs + */ + kref_init(&r->res_ref); goto out_unlock; @@ -1112,8 +1124,6 @@ static void toss_rsb(struct kref *kref) struct dlm_ls *ls = r->res_ls; DLM_ASSERT(list_empty(&r->res_root_list), dlm_print_rsb(r);); - kref_init(&r->res_ref); - WARN_ON(rsb_flag(r, RSB_TOSS)); rsb_set_flag(r, RSB_TOSS); list_move(&r->res_rsbs_list, &ls->ls_toss); r->res_toss_time = jiffies; @@ -1129,16 +1139,20 @@ static void toss_rsb(struct kref *kref) static void unhold_rsb(struct dlm_rsb *r) { int rv; + + /* rsbs in toss state never get referenced */ + WARN_ON(rsb_flag(r, RSB_TOSS)); rv = kref_put(&r->res_ref, toss_rsb); DLM_ASSERT(!rv, dlm_dump_rsb(r);); } -static void kill_rsb(struct kref *kref) +void free_toss_rsb(struct dlm_rsb *r) { - struct dlm_rsb *r = container_of(kref, struct dlm_rsb, res_ref); + WARN_ON_ONCE(!rsb_flag(r, RSB_TOSS)); - /* All work is done after the return from kref_put() so we - can release the write_lock before the remove and free. */ + /* check if all work is done after the rsb is on toss list + * and it can be freed. + */ DLM_ASSERT(list_empty(&r->res_lookup), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_grantqueue), dlm_dump_rsb(r);); @@ -1147,6 +1161,8 @@ static void kill_rsb(struct kref *kref) DLM_ASSERT(list_empty(&r->res_root_list), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_recover_list), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_masters_list), dlm_dump_rsb(r);); + + dlm_free_rsb(r); } /* Attaching/detaching lkb's from rsb's is for rsb reference counting. @@ -1611,15 +1627,10 @@ static void shrink_bucket(struct dlm_ls *ls) continue; } - if (!kref_put(&r->res_ref, kill_rsb)) { - log_error(ls, "tossed rsb in use %s", r->res_name); - continue; - } - list_del(&r->res_rsbs_list); rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, dlm_rhash_rsb_params); - dlm_free_rsb(r); + free_toss_rsb(r); } if (need_shrink) @@ -1680,19 +1691,13 @@ static void shrink_bucket(struct dlm_ls *ls) continue; } - if (!kref_put(&r->res_ref, kill_rsb)) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); - log_error(ls, "remove_name in use %s", name); - continue; - } - list_del(&r->res_rsbs_list); rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, dlm_rhash_rsb_params); send_remove(r); spin_unlock_bh(&ls->ls_rsbtbl_lock); - dlm_free_rsb(r); + free_toss_rsb(r); } } @@ -4180,18 +4185,12 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) return; } - if (kref_put(&r->res_ref, kill_rsb)) { - list_del(&r->res_rsbs_list); - rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, - dlm_rhash_rsb_params); - spin_unlock_bh(&ls->ls_rsbtbl_lock); - dlm_free_rsb(r); - } else { - log_error(ls, "receive_remove from %d rsb ref error", - from_nodeid); - dlm_print_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); - } + list_del(&r->res_rsbs_list); + rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, + dlm_rhash_rsb_params); + spin_unlock_bh(&ls->ls_rsbtbl_lock); + + free_toss_rsb(r); } static void receive_purge(struct dlm_ls *ls, const struct dlm_message *ms) diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index 33616d4b0cdb..b56a34802762 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -18,6 +18,7 @@ void dlm_receive_message_saved(struct dlm_ls *ls, const struct dlm_message *ms, uint32_t saved_seq); void dlm_receive_buffer(const union dlm_packet *p, int nodeid); int dlm_modes_compat(int mode1, int mode2); +void free_toss_rsb(struct dlm_rsb *r); void dlm_put_rsb(struct dlm_rsb *r); void dlm_hold_rsb(struct dlm_rsb *r); int dlm_put_lkb(struct dlm_lkb *lkb); diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index c21ef115123b..d43189532b14 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -889,7 +889,7 @@ void dlm_clear_toss(struct dlm_ls *ls) list_del(&r->res_rsbs_list); rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, dlm_rhash_rsb_params); - dlm_free_rsb(r); + free_toss_rsb(r); count++; } spin_unlock_bh(&ls->ls_rsbtbl_lock); -- cgit v1.2.3-59-g8ed1b From b1f2381c1a8d52b973944090ed8b42c750152533 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:41 -0400 Subject: dlm: drop dlm_scand kthread and use timers Currently the scand kthread acts like a garbage collection for expired rsbs on toss list, to clean them up after a certain timeout. It triggers every couple of seconds and iterates over the toss list while holding ls_rsbtbl_lock for the whole hash bucket iteration. To reduce the amount of time holding ls_rsbtbl_lock, we now handle the disposal of expired rsbs using a per-lockspace timer that expires for the earliest tossed rsb on the lockspace toss queue. This toss queue is ordered according to the rsb res_toss_time with the earliest tossed rsb as the first entry. The toss timer will only trylock() necessary locks, since it is low priority garbage collection, and will rearm the timer if trylock() fails. If the timer function does not find any expired rsb's, it rearms the timer with the next earliest expired rsb. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dlm_internal.h | 16 ++- fs/dlm/lock.c | 382 +++++++++++++++++++++++++++++++------------------- fs/dlm/lock.h | 2 + fs/dlm/lockspace.c | 105 ++------------ fs/dlm/member.c | 2 + fs/dlm/recover.c | 5 + fs/dlm/recoverd.c | 10 ++ 7 files changed, 283 insertions(+), 239 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index cf43b97cf3e5..98a0ac511bc8 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -334,6 +334,7 @@ struct dlm_rsb { struct list_head res_root_list; /* used for recovery */ struct list_head res_masters_list; /* used for recovery */ struct list_head res_recover_list; /* used for recovery */ + struct list_head res_toss_q_list; int res_recover_locks_count; char *res_lvbptr; @@ -584,13 +585,20 @@ struct dlm_ls { spinlock_t ls_lkbidr_spin; struct rhashtable ls_rsbtbl; -#define DLM_RTF_SHRINK_BIT 0 - unsigned long ls_rsbtbl_flags; spinlock_t ls_rsbtbl_lock; struct list_head ls_toss; struct list_head ls_keep; + struct timer_list ls_timer; + /* this queue is ordered according the + * absolute res_toss_time jiffies time + * to mod_timer() with the first element + * if necessary. + */ + struct list_head ls_toss_q; + spinlock_t ls_toss_q_lock; + spinlock_t ls_waiters_lock; struct list_head ls_waiters; /* lkbs needing a reply */ @@ -601,9 +609,6 @@ struct dlm_ls { int ls_new_rsb_count; struct list_head ls_new_rsb; /* new rsb structs */ - char *ls_remove_names[DLM_REMOVE_NAMES_MAX]; - int ls_remove_lens[DLM_REMOVE_NAMES_MAX]; - struct list_head ls_nodes; /* current nodes in ls */ struct list_head ls_nodes_gone; /* dead node list, recovery */ int ls_num_nodes; /* number of nodes in ls */ @@ -640,7 +645,6 @@ struct dlm_ls { spinlock_t ls_cb_lock; struct list_head ls_cb_delay; /* save for queue_work later */ - struct timer_list ls_timer; struct task_struct *ls_recoverd_task; struct mutex ls_recoverd_active; spinlock_t ls_recover_lock; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index fee1a4164fc1..7c97181a04fe 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -320,6 +320,11 @@ static void queue_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rqmode) * Basic operations on rsb's and lkb's */ +static inline unsigned long rsb_toss_jiffies(void) +{ + return jiffies + (READ_ONCE(dlm_config.ci_toss_secs) * HZ); +} + /* This is only called to add a reference when the code already holds a valid reference to the rsb, so there's no need for locking. */ @@ -416,6 +421,229 @@ static int pre_rsb_struct(struct dlm_ls *ls) return 0; } +/* connected with timer_delete_sync() in dlm_ls_stop() to stop + * new timers when recovery is triggered and don't run them + * again until a dlm_timer_resume() tries it again. + */ +static void __rsb_mod_timer(struct dlm_ls *ls, unsigned long jiffies) +{ + if (!dlm_locking_stopped(ls)) + mod_timer(&ls->ls_timer, jiffies); +} + +/* This function tries to resume the timer callback if a rsb + * is on the toss list and no timer is pending. It might that + * the first entry is on currently executed as timer callback + * but we don't care if a timer queued up again and does + * nothing. Should be a rare case. + */ +void dlm_timer_resume(struct dlm_ls *ls) +{ + struct dlm_rsb *r; + + spin_lock_bh(&ls->ls_toss_q_lock); + r = list_first_entry_or_null(&ls->ls_toss_q, struct dlm_rsb, + res_toss_q_list); + if (r && !timer_pending(&ls->ls_timer)) + __rsb_mod_timer(ls, r->res_toss_time); + spin_unlock_bh(&ls->ls_toss_q_lock); +} + +/* ls_rsbtbl_lock must be held and being sure the rsb is in toss state */ +static void rsb_delete_toss_timer(struct dlm_ls *ls, struct dlm_rsb *r) +{ + struct dlm_rsb *first; + + spin_lock_bh(&ls->ls_toss_q_lock); + r->res_toss_time = 0; + + /* if the rsb is not queued do nothing */ + if (list_empty(&r->res_toss_q_list)) + goto out; + + /* get the first element before delete */ + first = list_first_entry(&ls->ls_toss_q, struct dlm_rsb, + res_toss_q_list); + list_del_init(&r->res_toss_q_list); + /* check if the first element was the rsb we deleted */ + if (first == r) { + /* try to get the new first element, if the list + * is empty now try to delete the timer, if we are + * too late we don't care. + * + * if the list isn't empty and a new first element got + * in place, set the new timer expire time. + */ + first = list_first_entry_or_null(&ls->ls_toss_q, struct dlm_rsb, + res_toss_q_list); + if (!first) + timer_delete(&ls->ls_timer); + else + __rsb_mod_timer(ls, first->res_toss_time); + } + +out: + spin_unlock_bh(&ls->ls_toss_q_lock); +} + +/* Caller must held ls_rsbtbl_lock and need to be called every time + * when either the rsb enters toss state or the toss state changes + * the dir/master nodeid. + */ +static void rsb_mod_timer(struct dlm_ls *ls, struct dlm_rsb *r) +{ + int our_nodeid = dlm_our_nodeid(); + struct dlm_rsb *first; + + /* If we're the directory record for this rsb, and + * we're not the master of it, then we need to wait + * for the master node to send us a dir remove for + * before removing the dir record. + */ + if (!dlm_no_directory(ls) && + (r->res_master_nodeid != our_nodeid) && + (dlm_dir_nodeid(r) == our_nodeid)) { + rsb_delete_toss_timer(ls, r); + return; + } + + spin_lock_bh(&ls->ls_toss_q_lock); + /* set the new rsb absolute expire time in the rsb */ + r->res_toss_time = rsb_toss_jiffies(); + if (list_empty(&ls->ls_toss_q)) { + /* if the queue is empty add the element and it's + * our new expire time + */ + list_add_tail(&r->res_toss_q_list, &ls->ls_toss_q); + __rsb_mod_timer(ls, r->res_toss_time); + } else { + /* check if the rsb was already queued, if so delete + * it from the toss queue + */ + if (!list_empty(&r->res_toss_q_list)) + list_del(&r->res_toss_q_list); + + /* try to get the maybe new first element and then add + * to this rsb with the oldest expire time to the end + * of the queue. If the list was empty before this + * rsb expire time is our next expiration if it wasn't + * the now new first elemet is our new expiration time + */ + first = list_first_entry_or_null(&ls->ls_toss_q, struct dlm_rsb, + res_toss_q_list); + list_add_tail(&r->res_toss_q_list, &ls->ls_toss_q); + if (!first) + __rsb_mod_timer(ls, r->res_toss_time); + else + __rsb_mod_timer(ls, first->res_toss_time); + } + spin_unlock_bh(&ls->ls_toss_q_lock); +} + +/* if we hit contention we do in 250 ms a retry to trylock. + * if there is any other mod_timer in between we don't care + * about that it expires earlier again this is only for the + * unlikely case nothing happened in this time. + */ +#define DLM_TOSS_TIMER_RETRY (jiffies + msecs_to_jiffies(250)) + +void dlm_rsb_toss_timer(struct timer_list *timer) +{ + struct dlm_ls *ls = from_timer(ls, timer, ls_timer); + int our_nodeid = dlm_our_nodeid(); + struct dlm_rsb *r; + int rv; + + while (1) { + /* interrupting point to leave iteration when + * recovery waits for timer_delete_sync(), recovery + * will take care to delete everything in toss queue. + */ + if (dlm_locking_stopped(ls)) + break; + + rv = spin_trylock(&ls->ls_toss_q_lock); + if (!rv) { + /* rearm again try timer */ + __rsb_mod_timer(ls, DLM_TOSS_TIMER_RETRY); + break; + } + + r = list_first_entry_or_null(&ls->ls_toss_q, struct dlm_rsb, + res_toss_q_list); + if (!r) { + /* nothing to do anymore next rsb queue will + * set next mod_timer() expire. + */ + spin_unlock(&ls->ls_toss_q_lock); + break; + } + + /* test if the first rsb isn't expired yet, if + * so we stop freeing rsb from toss queue as + * the order in queue is ascending to the + * absolute res_toss_time jiffies + */ + if (time_before(jiffies, r->res_toss_time)) { + /* rearm with the next rsb to expire in the future */ + __rsb_mod_timer(ls, r->res_toss_time); + spin_unlock(&ls->ls_toss_q_lock); + break; + } + + /* in find_rsb_dir/nodir there is a reverse order of this + * lock, however this is only a trylock if we hit some + * possible contention we try it again. + * + * This lock synchronized while holding ls_toss_q_lock + * synchronize everything that rsb_delete_toss_timer() + * or rsb_mod_timer() can't run after this timer callback + * deletes the rsb from the ls_toss_q. Whereas the other + * holders have always a priority to run as this is only + * a caching handling and the other holders might to put + * this rsb out of the toss state. + */ + rv = spin_trylock(&ls->ls_rsbtbl_lock); + if (!rv) { + spin_unlock(&ls->ls_toss_q_lock); + /* rearm again try timer */ + __rsb_mod_timer(ls, DLM_TOSS_TIMER_RETRY); + break; + } + + list_del(&r->res_rsbs_list); + rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, + dlm_rhash_rsb_params); + + /* not necessary to held the ls_rsbtbl_lock when + * calling send_remove() + */ + spin_unlock(&ls->ls_rsbtbl_lock); + + /* remove the rsb out of the toss queue its gone + * drom DLM now + */ + list_del_init(&r->res_toss_q_list); + spin_unlock(&ls->ls_toss_q_lock); + + /* no rsb in this state should ever run a timer */ + WARN_ON(!dlm_no_directory(ls) && + (r->res_master_nodeid != our_nodeid) && + (dlm_dir_nodeid(r) == our_nodeid)); + + /* We're the master of this rsb but we're not + * the directory record, so we need to tell the + * dir node to remove the dir record + */ + if (!dlm_no_directory(ls) && + (r->res_master_nodeid == our_nodeid) && + (dlm_dir_nodeid(r) != our_nodeid)) + send_remove(r); + + free_toss_rsb(r); + } +} + /* If ls->ls_new_rsb is empty, return -EAGAIN, so the caller can unlock any spinlocks, go back and call pre_rsb_struct again. Otherwise, take an rsb off the list and return it. */ @@ -451,6 +679,7 @@ static int get_rsb_struct(struct dlm_ls *ls, const void *name, int len, INIT_LIST_HEAD(&r->res_convertqueue); INIT_LIST_HEAD(&r->res_waitqueue); INIT_LIST_HEAD(&r->res_root_list); + INIT_LIST_HEAD(&r->res_toss_q_list); INIT_LIST_HEAD(&r->res_recover_list); INIT_LIST_HEAD(&r->res_masters_list); @@ -638,6 +867,9 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, * valid for keep state rsbs */ kref_init(&r->res_ref); + rsb_delete_toss_timer(ls, r); + spin_unlock_bh(&ls->ls_rsbtbl_lock); + goto out_unlock; @@ -777,6 +1009,9 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, * valid for keep state rsbs */ kref_init(&r->res_ref); + rsb_delete_toss_timer(ls, r); + spin_unlock_bh(&ls->ls_rsbtbl_lock); + goto out_unlock; @@ -1050,7 +1285,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, __dlm_master_lookup(ls, r, our_nodeid, from_nodeid, true, flags, r_nodeid, result); - r->res_toss_time = jiffies; + rsb_mod_timer(ls, r); /* the rsb was inactive (on toss list) */ spin_unlock_bh(&ls->ls_rsbtbl_lock); @@ -1070,7 +1305,6 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, r->res_master_nodeid = from_nodeid; r->res_nodeid = from_nodeid; kref_init(&r->res_ref); - r->res_toss_time = jiffies; rsb_set_flag(r, RSB_TOSS); error = rsb_insert(r, &ls->ls_rsbtbl); @@ -1082,6 +1316,7 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, } list_add(&r->res_rsbs_list, &ls->ls_toss); + rsb_mod_timer(ls, r); if (result) *result = DLM_LU_ADD; @@ -1126,8 +1361,8 @@ static void toss_rsb(struct kref *kref) DLM_ASSERT(list_empty(&r->res_root_list), dlm_print_rsb(r);); rsb_set_flag(r, RSB_TOSS); list_move(&r->res_rsbs_list, &ls->ls_toss); - r->res_toss_time = jiffies; - set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags); + rsb_mod_timer(ls, r); + if (r->res_lvbptr) { dlm_free_lvb(r->res_lvbptr); r->res_lvbptr = NULL; @@ -1150,15 +1385,12 @@ void free_toss_rsb(struct dlm_rsb *r) { WARN_ON_ONCE(!rsb_flag(r, RSB_TOSS)); - /* check if all work is done after the rsb is on toss list - * and it can be freed. - */ - DLM_ASSERT(list_empty(&r->res_lookup), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_grantqueue), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_convertqueue), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_waitqueue), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_root_list), dlm_dump_rsb(r);); + DLM_ASSERT(list_empty(&r->res_toss_q_list), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_recover_list), dlm_dump_rsb(r);); DLM_ASSERT(list_empty(&r->res_masters_list), dlm_dump_rsb(r);); @@ -1572,140 +1804,6 @@ static int remove_from_waiters_ms(struct dlm_lkb *lkb, return error; } -static void shrink_bucket(struct dlm_ls *ls) -{ - struct dlm_rsb *r, *safe; - char *name; - int our_nodeid = dlm_our_nodeid(); - int remote_count = 0; - int need_shrink = 0; - int i, len, rv; - - memset(&ls->ls_remove_lens, 0, sizeof(int) * DLM_REMOVE_NAMES_MAX); - - spin_lock_bh(&ls->ls_rsbtbl_lock); - - if (!test_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags)) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); - return; - } - - list_for_each_entry_safe(r, safe, &ls->ls_toss, res_rsbs_list) { - /* If we're the directory record for this rsb, and - we're not the master of it, then we need to wait - for the master node to send us a dir remove for - before removing the dir record. */ - - if (!dlm_no_directory(ls) && - (r->res_master_nodeid != our_nodeid) && - (dlm_dir_nodeid(r) == our_nodeid)) { - continue; - } - - need_shrink = 1; - - if (!time_after_eq(jiffies, r->res_toss_time + - dlm_config.ci_toss_secs * HZ)) { - continue; - } - - if (!dlm_no_directory(ls) && - (r->res_master_nodeid == our_nodeid) && - (dlm_dir_nodeid(r) != our_nodeid)) { - - /* We're the master of this rsb but we're not - the directory record, so we need to tell the - dir node to remove the dir record. */ - - ls->ls_remove_lens[remote_count] = r->res_length; - memcpy(ls->ls_remove_names[remote_count], r->res_name, - DLM_RESNAME_MAXLEN); - remote_count++; - - if (remote_count >= DLM_REMOVE_NAMES_MAX) - break; - continue; - } - - list_del(&r->res_rsbs_list); - rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, - dlm_rhash_rsb_params); - free_toss_rsb(r); - } - - if (need_shrink) - set_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags); - else - clear_bit(DLM_RTF_SHRINK_BIT, &ls->ls_rsbtbl_flags); - spin_unlock_bh(&ls->ls_rsbtbl_lock); - - /* - * While searching for rsb's to free, we found some that require - * remote removal. We leave them in place and find them again here - * so there is a very small gap between removing them from the toss - * list and sending the removal. Keeping this gap small is - * important to keep us (the master node) from being out of sync - * with the remote dir node for very long. - */ - - for (i = 0; i < remote_count; i++) { - name = ls->ls_remove_names[i]; - len = ls->ls_remove_lens[i]; - - spin_lock_bh(&ls->ls_rsbtbl_lock); - rv = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); - if (rv) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); - log_error(ls, "remove_name not found %s", name); - continue; - } - - if (!rsb_flag(r, RSB_TOSS)) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); - log_debug(ls, "remove_name not toss %s", name); - continue; - } - - if (r->res_master_nodeid != our_nodeid) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); - log_debug(ls, "remove_name master %d dir %d our %d %s", - r->res_master_nodeid, r->res_dir_nodeid, - our_nodeid, name); - continue; - } - - if (r->res_dir_nodeid == our_nodeid) { - /* should never happen */ - spin_unlock_bh(&ls->ls_rsbtbl_lock); - log_error(ls, "remove_name dir %d master %d our %d %s", - r->res_dir_nodeid, r->res_master_nodeid, - our_nodeid, name); - continue; - } - - if (!time_after_eq(jiffies, r->res_toss_time + - dlm_config.ci_toss_secs * HZ)) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); - log_debug(ls, "remove_name toss_time %lu now %lu %s", - r->res_toss_time, jiffies, name); - continue; - } - - list_del(&r->res_rsbs_list); - rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, - dlm_rhash_rsb_params); - send_remove(r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); - - free_toss_rsb(r); - } -} - -void dlm_scan_rsbs(struct dlm_ls *ls) -{ - shrink_bucket(ls); -} - /* lkb is master or local copy */ static void set_lvb_lock(struct dlm_rsb *r, struct dlm_lkb *lkb) diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index b56a34802762..8de9dee4c058 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -11,6 +11,7 @@ #ifndef __LOCK_DOT_H__ #define __LOCK_DOT_H__ +void dlm_rsb_toss_timer(struct timer_list *timer); void dlm_dump_rsb(struct dlm_rsb *r); void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len); void dlm_print_lkb(struct dlm_lkb *lkb); @@ -26,6 +27,7 @@ void dlm_scan_rsbs(struct dlm_ls *ls); int dlm_lock_recovery_try(struct dlm_ls *ls); void dlm_lock_recovery(struct dlm_ls *ls); void dlm_unlock_recovery(struct dlm_ls *ls); +void dlm_timer_resume(struct dlm_ls *ls); int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, int len, unsigned int flags, int *r_nodeid, int *result); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 890e1a4cf787..931eb3f22ec6 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -29,8 +29,6 @@ static int ls_count; static struct mutex ls_lock; static struct list_head lslist; static spinlock_t lslist_lock; -static struct task_struct * scand_task; - static ssize_t dlm_control_store(struct dlm_ls *ls, const char *buf, size_t len) { @@ -247,64 +245,6 @@ void dlm_lockspace_exit(void) kset_unregister(dlm_kset); } -static struct dlm_ls *find_ls_to_scan(void) -{ - struct dlm_ls *ls; - - spin_lock_bh(&lslist_lock); - list_for_each_entry(ls, &lslist, ls_list) { - if (time_after_eq(jiffies, ls->ls_scan_time + - dlm_config.ci_scan_secs * HZ)) { - atomic_inc(&ls->ls_count); - spin_unlock_bh(&lslist_lock); - return ls; - } - } - spin_unlock_bh(&lslist_lock); - return NULL; -} - -static int dlm_scand(void *data) -{ - struct dlm_ls *ls; - - while (!kthread_should_stop()) { - ls = find_ls_to_scan(); - if (ls) { - if (dlm_lock_recovery_try(ls)) { - ls->ls_scan_time = jiffies; - dlm_scan_rsbs(ls); - dlm_unlock_recovery(ls); - } else { - ls->ls_scan_time += HZ; - } - - dlm_put_lockspace(ls); - continue; - } - schedule_timeout_interruptible(dlm_config.ci_scan_secs * HZ); - } - return 0; -} - -static int dlm_scand_start(void) -{ - struct task_struct *p; - int error = 0; - - p = kthread_run(dlm_scand, NULL, "dlm_scand"); - if (IS_ERR(p)) - error = PTR_ERR(p); - else - scand_task = p; - return error; -} - -static void dlm_scand_stop(void) -{ - kthread_stop(scand_task); -} - struct dlm_ls *dlm_find_lockspace_global(uint32_t id) { struct dlm_ls *ls; @@ -385,22 +325,9 @@ static int threads_start(void) /* Thread for sending/receiving messages for all lockspace's */ error = dlm_midcomms_start(); - if (error) { + if (error) log_print("cannot start dlm midcomms %d", error); - goto fail; - } - error = dlm_scand_start(); - if (error) { - log_print("cannot start dlm_scand thread %d", error); - goto midcomms_fail; - } - - return 0; - - midcomms_fail: - dlm_midcomms_stop(); - fail: return error; } @@ -412,7 +339,7 @@ static int new_lockspace(const char *name, const char *cluster, struct dlm_ls *ls; int do_unreg = 0; int namelen = strlen(name); - int i, error; + int error; if (namelen > DLM_LOCKSPACE_LEN || namelen == 0) return -EINVAL; @@ -503,13 +430,6 @@ static int new_lockspace(const char *name, const char *cluster, if (error) goto out_lsfree; - for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) { - ls->ls_remove_names[i] = kzalloc(DLM_RESNAME_MAXLEN+1, - GFP_KERNEL); - if (!ls->ls_remove_names[i]) - goto out_rsbtbl; - } - idr_init(&ls->ls_lkbidr); spin_lock_init(&ls->ls_lkbidr_spin); @@ -582,6 +502,11 @@ static int new_lockspace(const char *name, const char *cluster, INIT_LIST_HEAD(&ls->ls_dir_dump_list); rwlock_init(&ls->ls_dir_dump_lock); + INIT_LIST_HEAD(&ls->ls_toss_q); + spin_lock_init(&ls->ls_toss_q_lock); + timer_setup(&ls->ls_timer, dlm_rsb_toss_timer, + TIMER_DEFERRABLE); + spin_lock_bh(&lslist_lock); ls->ls_create_count = 1; list_add(&ls->ls_list, &lslist); @@ -661,9 +586,6 @@ static int new_lockspace(const char *name, const char *cluster, kfree(ls->ls_recover_buf); out_lkbidr: idr_destroy(&ls->ls_lkbidr); - out_rsbtbl: - for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) - kfree(ls->ls_remove_names[i]); rhashtable_destroy(&ls->ls_rsbtbl); out_lsfree: if (do_unreg) @@ -696,7 +618,6 @@ static int __dlm_new_lockspace(const char *name, const char *cluster, if (error > 0) error = 0; if (!ls_count) { - dlm_scand_stop(); dlm_midcomms_shutdown(); dlm_midcomms_stop(); } @@ -777,7 +698,7 @@ static void rhash_free_rsb(void *ptr, void *arg) static int release_lockspace(struct dlm_ls *ls, int force) { struct dlm_rsb *rsb; - int i, busy, rv; + int busy, rv; busy = lockspace_busy(ls, force); @@ -812,8 +733,13 @@ static int release_lockspace(struct dlm_ls *ls, int force) dlm_recoverd_stop(ls); + /* clear the LSFL_RUNNING flag to fast up + * time_shutdown_sync(), we don't care anymore + */ + clear_bit(LSFL_RUNNING, &ls->ls_flags); + timer_shutdown_sync(&ls->ls_timer); + if (ls_count == 1) { - dlm_scand_stop(); dlm_clear_members(ls); dlm_midcomms_shutdown(); } @@ -839,9 +765,6 @@ static int release_lockspace(struct dlm_ls *ls, int force) */ rhashtable_free_and_destroy(&ls->ls_rsbtbl, rhash_free_rsb, NULL); - for (i = 0; i < DLM_REMOVE_NAMES_MAX; i++) - kfree(ls->ls_remove_names[i]); - while (!list_empty(&ls->ls_new_rsb)) { rsb = list_first_entry(&ls->ls_new_rsb, struct dlm_rsb, res_hashchain); diff --git a/fs/dlm/member.c b/fs/dlm/member.c index 6401916a97ef..c46e306f2e5c 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -641,6 +641,8 @@ int dlm_ls_stop(struct dlm_ls *ls) spin_lock_bh(&ls->ls_recover_lock); set_bit(LSFL_RECOVER_STOP, &ls->ls_flags); new = test_and_clear_bit(LSFL_RUNNING, &ls->ls_flags); + if (new) + timer_delete_sync(&ls->ls_timer); ls->ls_recover_seq++; /* activate requestqueue and stop processing */ diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index d43189532b14..960a14b95605 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -889,6 +889,11 @@ void dlm_clear_toss(struct dlm_ls *ls) list_del(&r->res_rsbs_list); rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, dlm_rhash_rsb_params); + + /* remove it from the toss queue if its part of it */ + if (!list_empty(&r->res_toss_q_list)) + list_del_init(&r->res_toss_q_list); + free_toss_rsb(r); count++; } diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index 5e8e10030b74..c831e0275912 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -98,6 +98,16 @@ static int enable_locking(struct dlm_ls *ls, uint64_t seq) spin_lock_bh(&ls->ls_recover_lock); if (ls->ls_recover_seq == seq) { set_bit(LSFL_RUNNING, &ls->ls_flags); + /* Schedule next timer if recovery put something on toss. + * + * The rsbs that was queued while recovery on toss hasn't + * started yet because LSFL_RUNNING was set everything + * else recovery hasn't started as well because ls_in_recovery + * is still hold. So we should not run into the case that + * dlm_timer_resume() queues a timer that can occur in + * a no op. + */ + dlm_timer_resume(ls); /* unblocks processes waiting to enter the dlm */ up_write(&ls->ls_in_recovery); clear_bit(LSFL_RECOVER_LOCK, &ls->ls_flags); -- cgit v1.2.3-59-g8ed1b From e91313591b29ce724fe2f1bdf29f2482878fc275 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:42 -0400 Subject: dlm: use rwlock for rsb hash table The conversion to rhashtable introduced a hash table lock per lockspace, in place of per bucket locks. To make this more scalable, switch to using a rwlock for hash table access. The common case fast path uses it as a read lock. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/debug_fs.c | 4 +- fs/dlm/dir.c | 4 +- fs/dlm/dlm_internal.h | 2 +- fs/dlm/lock.c | 269 ++++++++++++++++++++++++++++++++++++-------------- fs/dlm/lockspace.c | 2 +- fs/dlm/recover.c | 4 +- fs/dlm/recoverd.c | 8 +- 7 files changed, 206 insertions(+), 87 deletions(-) (limited to 'fs') diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c index 70567919f1b7..6ab3ed4074c6 100644 --- a/fs/dlm/debug_fs.c +++ b/fs/dlm/debug_fs.c @@ -413,7 +413,7 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos) else list = &ls->ls_keep; - spin_lock_bh(&ls->ls_rsbtbl_lock); + read_lock_bh(&ls->ls_rsbtbl_lock); return seq_list_start(list, *pos); } @@ -434,7 +434,7 @@ static void table_seq_stop(struct seq_file *seq, void *iter_ptr) { struct dlm_ls *ls = seq->private; - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); } static const struct seq_operations format1_seq_ops = { diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c index 9687f908476b..b1ab0adbd9d0 100644 --- a/fs/dlm/dir.c +++ b/fs/dlm/dir.c @@ -200,9 +200,9 @@ static struct dlm_rsb *find_rsb_root(struct dlm_ls *ls, const char *name, struct dlm_rsb *r; int rv; - spin_lock_bh(&ls->ls_rsbtbl_lock); + read_lock_bh(&ls->ls_rsbtbl_lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); if (!rv) return r; diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 98a0ac511bc8..b675bffb61ae 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -585,7 +585,7 @@ struct dlm_ls { spinlock_t ls_lkbidr_spin; struct rhashtable ls_rsbtbl; - spinlock_t ls_rsbtbl_lock; + rwlock_t ls_rsbtbl_lock; struct list_head ls_toss; struct list_head ls_keep; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 7c97181a04fe..790d0fd76bbe 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -342,15 +342,15 @@ void dlm_hold_rsb(struct dlm_rsb *r) /* TODO move this to lib/refcount.c */ static __must_check bool -dlm_refcount_dec_and_lock_bh(refcount_t *r, spinlock_t *lock) +dlm_refcount_dec_and_write_lock_bh(refcount_t *r, rwlock_t *lock) __cond_acquires(lock) { if (refcount_dec_not_one(r)) return false; - spin_lock_bh(lock); + write_lock_bh(lock); if (!refcount_dec_and_test(r)) { - spin_unlock_bh(lock); + write_unlock_bh(lock); return false; } @@ -358,11 +358,11 @@ __cond_acquires(lock) } /* TODO move this to include/linux/kref.h */ -static inline int dlm_kref_put_lock_bh(struct kref *kref, - void (*release)(struct kref *kref), - spinlock_t *lock) +static inline int dlm_kref_put_write_lock_bh(struct kref *kref, + void (*release)(struct kref *kref), + rwlock_t *lock) { - if (dlm_refcount_dec_and_lock_bh(&kref->refcount, lock)) { + if (dlm_refcount_dec_and_write_lock_bh(&kref->refcount, lock)) { release(kref); return 1; } @@ -378,10 +378,10 @@ static void put_rsb(struct dlm_rsb *r) struct dlm_ls *ls = r->res_ls; int rv; - rv = dlm_kref_put_lock_bh(&r->res_ref, toss_rsb, - &ls->ls_rsbtbl_lock); + rv = dlm_kref_put_write_lock_bh(&r->res_ref, toss_rsb, + &ls->ls_rsbtbl_lock); if (rv) - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); } void dlm_put_rsb(struct dlm_rsb *r) @@ -603,7 +603,7 @@ void dlm_rsb_toss_timer(struct timer_list *timer) * a caching handling and the other holders might to put * this rsb out of the toss state. */ - rv = spin_trylock(&ls->ls_rsbtbl_lock); + rv = write_trylock(&ls->ls_rsbtbl_lock); if (!rv) { spin_unlock(&ls->ls_toss_q_lock); /* rearm again try timer */ @@ -618,7 +618,7 @@ void dlm_rsb_toss_timer(struct timer_list *timer) /* not necessary to held the ls_rsbtbl_lock when * calling send_remove() */ - spin_unlock(&ls->ls_rsbtbl_lock); + write_unlock(&ls->ls_rsbtbl_lock); /* remove the rsb out of the toss queue its gone * drom DLM now @@ -702,16 +702,8 @@ int dlm_search_rsb_tree(struct rhashtable *rhash, const void *name, int len, static int rsb_insert(struct dlm_rsb *rsb, struct rhashtable *rhash) { - int rv; - - rv = rhashtable_insert_fast(rhash, &rsb->res_node, - dlm_rhash_rsb_params); - if (rv == -EEXIST) { - log_print("%s match", __func__); - dlm_dump_rsb(rsb); - } - - return rv; + return rhashtable_insert_fast(rhash, &rsb->res_node, + dlm_rhash_rsb_params); } /* @@ -806,24 +798,47 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, goto out; } - spin_lock_bh(&ls->ls_rsbtbl_lock); + retry_lookup: + /* check if the rsb is in keep state under read lock - likely path */ + read_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); - if (error) + if (error) { + read_unlock_bh(&ls->ls_rsbtbl_lock); goto do_new; + } /* * rsb is active, so we can't check master_nodeid without lock_rsb. */ - if (rsb_flag(r, RSB_TOSS)) + if (rsb_flag(r, RSB_TOSS)) { + read_unlock_bh(&ls->ls_rsbtbl_lock); goto do_toss; + } kref_get(&r->res_ref); - goto out_unlock; + read_unlock_bh(&ls->ls_rsbtbl_lock); + goto out; do_toss: + write_lock_bh(&ls->ls_rsbtbl_lock); + + /* retry lookup under write lock to see if its still in toss state + * if not it's in keep state and we relookup - unlikely path. + */ + error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); + if (!error) { + if (!rsb_flag(r, RSB_TOSS)) { + write_unlock_bh(&ls->ls_rsbtbl_lock); + goto retry_lookup; + } + } else { + write_unlock_bh(&ls->ls_rsbtbl_lock); + goto do_new; + } + /* * rsb found inactive (master_nodeid may be out of date unless * we are the dir_nodeid or were the master) No other thread @@ -837,8 +852,9 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, log_debug(ls, "find_rsb toss from_other %d master %d dir %d %s", from_nodeid, r->res_master_nodeid, dir_nodeid, r->res_name); + write_unlock_bh(&ls->ls_rsbtbl_lock); error = -ENOTBLK; - goto out_unlock; + goto out; } if ((r->res_master_nodeid != our_nodeid) && from_dir) { @@ -868,9 +884,9 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, */ kref_init(&r->res_ref); rsb_delete_toss_timer(ls, r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); - goto out_unlock; + goto out; do_new: @@ -879,15 +895,13 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, */ if (error == -EBADR && !create) - goto out_unlock; + goto out; error = get_rsb_struct(ls, name, len, &r); - if (error == -EAGAIN) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); + if (error == -EAGAIN) goto retry; - } if (error) - goto out_unlock; + goto out; r->res_hash = hash; r->res_dir_nodeid = dir_nodeid; @@ -909,7 +923,7 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, dlm_free_rsb(r); r = NULL; error = -ENOTBLK; - goto out_unlock; + goto out; } if (from_other) { @@ -929,11 +943,20 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, } out_add: + + write_lock_bh(&ls->ls_rsbtbl_lock); error = rsb_insert(r, &ls->ls_rsbtbl); - if (!error) + if (error == -EEXIST) { + /* somebody else was faster and it seems the + * rsb exists now, we do a whole relookup + */ + write_unlock_bh(&ls->ls_rsbtbl_lock); + dlm_free_rsb(r); + goto retry_lookup; + } else if (!error) { list_add(&r->res_rsbs_list, &ls->ls_keep); - out_unlock: - spin_unlock_bh(&ls->ls_rsbtbl_lock); + } + write_unlock_bh(&ls->ls_rsbtbl_lock); out: *r_ret = r; return error; @@ -957,24 +980,49 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, if (error < 0) goto out; - spin_lock_bh(&ls->ls_rsbtbl_lock); + retry_lookup: + /* check if the rsb is in keep state under read lock - likely path */ + read_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); - if (error) + if (error) { + read_unlock_bh(&ls->ls_rsbtbl_lock); goto do_new; + } - if (rsb_flag(r, RSB_TOSS)) + if (rsb_flag(r, RSB_TOSS)) { + read_unlock_bh(&ls->ls_rsbtbl_lock); goto do_toss; + } /* * rsb is active, so we can't check master_nodeid without lock_rsb. */ kref_get(&r->res_ref); - goto out_unlock; + read_unlock_bh(&ls->ls_rsbtbl_lock); + + goto out; do_toss: + write_lock_bh(&ls->ls_rsbtbl_lock); + + /* retry lookup under write lock to see if its still in toss state + * if not it's in keep state and we relookup - unlikely path. + */ + error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); + if (!error) { + if (!rsb_flag(r, RSB_TOSS)) { + write_unlock_bh(&ls->ls_rsbtbl_lock); + goto retry_lookup; + } + } else { + write_unlock_bh(&ls->ls_rsbtbl_lock); + goto do_new; + } + + /* * rsb found inactive. No other thread is using this rsb because * it's on the toss list, so we can look at or update @@ -987,8 +1035,9 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, log_error(ls, "find_rsb toss from_nodeid %d master %d dir %d", from_nodeid, r->res_master_nodeid, dir_nodeid); dlm_print_rsb(r); + write_unlock_bh(&ls->ls_rsbtbl_lock); error = -ENOTBLK; - goto out_unlock; + goto out; } if (!recover && (r->res_master_nodeid != our_nodeid) && @@ -1010,9 +1059,9 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, */ kref_init(&r->res_ref); rsb_delete_toss_timer(ls, r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); - goto out_unlock; + goto out; do_new: @@ -1022,11 +1071,10 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, error = get_rsb_struct(ls, name, len, &r); if (error == -EAGAIN) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); goto retry; } if (error) - goto out_unlock; + goto out; r->res_hash = hash; r->res_dir_nodeid = dir_nodeid; @@ -1034,11 +1082,20 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, r->res_nodeid = (dir_nodeid == our_nodeid) ? 0 : dir_nodeid; kref_init(&r->res_ref); + write_lock_bh(&ls->ls_rsbtbl_lock); error = rsb_insert(r, &ls->ls_rsbtbl); - if (!error) + if (error == -EEXIST) { + /* somebody else was faster and it seems the + * rsb exists now, we do a whole relookup + */ + write_unlock_bh(&ls->ls_rsbtbl_lock); + dlm_free_rsb(r); + goto retry_lookup; + } else if (!error) { list_add(&r->res_rsbs_list, &ls->ls_keep); - out_unlock: - spin_unlock_bh(&ls->ls_rsbtbl_lock); + } + write_unlock_bh(&ls->ls_rsbtbl_lock); + out: *r_ret = r; return error; @@ -1251,18 +1308,23 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, if (error < 0) return error; - spin_lock_bh(&ls->ls_rsbtbl_lock); + retry_lookup: + + /* check if the rsb is in keep state under read lock - likely path */ + read_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (!error) { - if (rsb_flag(r, RSB_TOSS)) + if (rsb_flag(r, RSB_TOSS)) { + read_unlock_bh(&ls->ls_rsbtbl_lock); goto do_toss; + } /* because the rsb is active, we need to lock_rsb before * checking/changing re_master_nodeid */ hold_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); lock_rsb(r); __dlm_master_lookup(ls, r, our_nodeid, from_nodeid, false, @@ -1274,10 +1336,31 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, return 0; } else { + read_unlock_bh(&ls->ls_rsbtbl_lock); goto not_found; } do_toss: + /* unlikely path - relookup under write */ + write_lock_bh(&ls->ls_rsbtbl_lock); + + /* rsb_mod_timer() requires to held ls_rsbtbl_lock in write lock + * check if the rsb is still in toss state, if not relookup + */ + error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); + if (!error) { + if (!rsb_flag(r, RSB_TOSS)) { + write_unlock_bh(&ls->ls_rsbtbl_lock); + /* something as changed, very unlikely but + * try again + */ + goto retry_lookup; + } + } else { + write_unlock_bh(&ls->ls_rsbtbl_lock); + goto not_found; + } + /* because the rsb is inactive (on toss list), it's not refcounted * and lock_rsb is not used, but is protected by the rsbtbl lock */ @@ -1287,18 +1370,16 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, rsb_mod_timer(ls, r); /* the rsb was inactive (on toss list) */ - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); return 0; not_found: error = get_rsb_struct(ls, name, len, &r); - if (error == -EAGAIN) { - spin_unlock_bh(&ls->ls_rsbtbl_lock); + if (error == -EAGAIN) goto retry; - } if (error) - goto out_unlock; + goto out; r->res_hash = hash; r->res_dir_nodeid = our_nodeid; @@ -1307,22 +1388,30 @@ int dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *name, kref_init(&r->res_ref); rsb_set_flag(r, RSB_TOSS); + write_lock_bh(&ls->ls_rsbtbl_lock); error = rsb_insert(r, &ls->ls_rsbtbl); - if (error) { + if (error == -EEXIST) { + /* somebody else was faster and it seems the + * rsb exists now, we do a whole relookup + */ + write_unlock_bh(&ls->ls_rsbtbl_lock); + dlm_free_rsb(r); + goto retry_lookup; + } else if (error) { + write_unlock_bh(&ls->ls_rsbtbl_lock); /* should never happen */ dlm_free_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); goto retry; } list_add(&r->res_rsbs_list, &ls->ls_toss); rsb_mod_timer(ls, r); + write_unlock_bh(&ls->ls_rsbtbl_lock); if (result) *result = DLM_LU_ADD; *r_nodeid = from_nodeid; - out_unlock: - spin_unlock_bh(&ls->ls_rsbtbl_lock); + out: return error; } @@ -1330,12 +1419,12 @@ static void dlm_dump_rsb_hash(struct dlm_ls *ls, uint32_t hash) { struct dlm_rsb *r; - spin_lock_bh(&ls->ls_rsbtbl_lock); + read_lock_bh(&ls->ls_rsbtbl_lock); list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { if (r->res_hash == hash) dlm_dump_rsb(r); } - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); } void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) @@ -1343,14 +1432,14 @@ void dlm_dump_rsb_name(struct dlm_ls *ls, const char *name, int len) struct dlm_rsb *r = NULL; int error; - spin_lock_bh(&ls->ls_rsbtbl_lock); + read_lock_bh(&ls->ls_rsbtbl_lock); error = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (!error) goto out; dlm_dump_rsb(r); out: - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); } static void toss_rsb(struct kref *kref) @@ -1478,6 +1567,36 @@ static void kill_lkb(struct kref *kref) DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb);); } +/* TODO move this to lib/refcount.c */ +static __must_check bool +dlm_refcount_dec_and_lock_bh(refcount_t *r, spinlock_t *lock) +__cond_acquires(lock) +{ + if (refcount_dec_not_one(r)) + return false; + + spin_lock_bh(lock); + if (!refcount_dec_and_test(r)) { + spin_unlock_bh(lock); + return false; + } + + return true; +} + +/* TODO move this to include/linux/kref.h */ +static inline int dlm_kref_put_lock_bh(struct kref *kref, + void (*release)(struct kref *kref), + spinlock_t *lock) +{ + if (dlm_refcount_dec_and_lock_bh(&kref->refcount, lock)) { + release(kref); + return 1; + } + + return 0; +} + /* __put_lkb() is used when an lkb may not have an rsb attached to it so we need to provide the lockspace explicitly */ @@ -4247,14 +4366,14 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) memset(name, 0, sizeof(name)); memcpy(name, ms->m_extra, len); - spin_lock_bh(&ls->ls_rsbtbl_lock); + write_lock_bh(&ls->ls_rsbtbl_lock); rv = dlm_search_rsb_tree(&ls->ls_rsbtbl, name, len, &r); if (rv) { /* should not happen */ log_error(ls, "%s from %d not found %s", __func__, from_nodeid, name); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); return; } @@ -4264,14 +4383,14 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) log_error(ls, "receive_remove keep from %d master %d", from_nodeid, r->res_master_nodeid); dlm_print_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); return; } log_debug(ls, "receive_remove from %d master %d first %x %s", from_nodeid, r->res_master_nodeid, r->res_first_lkid, name); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); return; } @@ -4279,14 +4398,14 @@ static void receive_remove(struct dlm_ls *ls, const struct dlm_message *ms) log_error(ls, "receive_remove toss from %d master %d", from_nodeid, r->res_master_nodeid); dlm_print_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); return; } list_del(&r->res_rsbs_list); rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, dlm_rhash_rsb_params); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); free_toss_rsb(r); } @@ -5354,7 +5473,7 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls) { struct dlm_rsb *r; - spin_lock_bh(&ls->ls_rsbtbl_lock); + read_lock_bh(&ls->ls_rsbtbl_lock); list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { if (!rsb_flag(r, RSB_RECOVER_GRANT)) continue; @@ -5363,10 +5482,10 @@ static struct dlm_rsb *find_grant_rsb(struct dlm_ls *ls) continue; } hold_rsb(r); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); return r; } - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); return NULL; } diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 931eb3f22ec6..04f4c74831ce 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -424,7 +424,7 @@ static int new_lockspace(const char *name, const char *cluster, INIT_LIST_HEAD(&ls->ls_toss); INIT_LIST_HEAD(&ls->ls_keep); - spin_lock_init(&ls->ls_rsbtbl_lock); + rwlock_init(&ls->ls_rsbtbl_lock); error = rhashtable_init(&ls->ls_rsbtbl, &dlm_rhash_rsb_params); if (error) diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index 960a14b95605..f493d5f30c58 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -884,7 +884,7 @@ void dlm_clear_toss(struct dlm_ls *ls) struct dlm_rsb *r, *safe; unsigned int count = 0; - spin_lock_bh(&ls->ls_rsbtbl_lock); + write_lock_bh(&ls->ls_rsbtbl_lock); list_for_each_entry_safe(r, safe, &ls->ls_toss, res_rsbs_list) { list_del(&r->res_rsbs_list); rhashtable_remove_fast(&ls->ls_rsbtbl, &r->res_node, @@ -897,7 +897,7 @@ void dlm_clear_toss(struct dlm_ls *ls) free_toss_rsb(r); count++; } - spin_unlock_bh(&ls->ls_rsbtbl_lock); + write_unlock_bh(&ls->ls_rsbtbl_lock); if (count) log_rinfo(ls, "dlm_clear_toss %u done", count); diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index c831e0275912..17a40d1e6036 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -32,7 +32,7 @@ static int dlm_create_masters_list(struct dlm_ls *ls) goto out; } - spin_lock_bh(&ls->ls_rsbtbl_lock); + read_lock_bh(&ls->ls_rsbtbl_lock); list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { if (r->res_nodeid) continue; @@ -40,7 +40,7 @@ static int dlm_create_masters_list(struct dlm_ls *ls) list_add(&r->res_masters_list, &ls->ls_masters_list); dlm_hold_rsb(r); } - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); out: write_unlock_bh(&ls->ls_masters_lock); return error; @@ -62,14 +62,14 @@ static void dlm_create_root_list(struct dlm_ls *ls, struct list_head *root_list) { struct dlm_rsb *r; - spin_lock_bh(&ls->ls_rsbtbl_lock); + read_lock_bh(&ls->ls_rsbtbl_lock); list_for_each_entry(r, &ls->ls_keep, res_rsbs_list) { list_add(&r->res_root_list, root_list); dlm_hold_rsb(r); } WARN_ON_ONCE(!list_empty(&ls->ls_toss)); - spin_unlock_bh(&ls->ls_rsbtbl_lock); + read_unlock_bh(&ls->ls_rsbtbl_lock); } static void dlm_release_root_list(struct list_head *root_list) -- cgit v1.2.3-59-g8ed1b From 15fd7e55177a8d6243096bb6b9536cd15f56d547 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 15 Apr 2024 14:39:43 -0400 Subject: dlm: use rwlock for lkbidr Convert the lock for lkbidr to an rwlock. Most idr lookups will use the read lock. Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dlm_internal.h | 2 +- fs/dlm/lock.c | 44 +++++++------------------------------------- fs/dlm/lockspace.c | 6 +++--- 3 files changed, 11 insertions(+), 41 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index b675bffb61ae..19e57cbd5b13 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -582,7 +582,7 @@ struct dlm_ls { struct kobject ls_kobj; struct idr ls_lkbidr; - spinlock_t ls_lkbidr_spin; + rwlock_t ls_lkbidr_lock; struct rhashtable ls_rsbtbl; rwlock_t ls_rsbtbl_lock; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 790d0fd76bbe..bbbc9593a64e 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -1522,11 +1522,11 @@ static int _create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret, INIT_LIST_HEAD(&lkb->lkb_ownqueue); INIT_LIST_HEAD(&lkb->lkb_rsb_lookup); - spin_lock_bh(&ls->ls_lkbidr_spin); + write_lock_bh(&ls->ls_lkbidr_lock); rv = idr_alloc(&ls->ls_lkbidr, lkb, start, end, GFP_NOWAIT); if (rv >= 0) lkb->lkb_id = rv; - spin_unlock_bh(&ls->ls_lkbidr_spin); + write_unlock_bh(&ls->ls_lkbidr_lock); if (rv < 0) { log_error(ls, "create_lkb idr error %d", rv); @@ -1547,11 +1547,11 @@ static int find_lkb(struct dlm_ls *ls, uint32_t lkid, struct dlm_lkb **lkb_ret) { struct dlm_lkb *lkb; - spin_lock_bh(&ls->ls_lkbidr_spin); + read_lock_bh(&ls->ls_lkbidr_lock); lkb = idr_find(&ls->ls_lkbidr, lkid); if (lkb) kref_get(&lkb->lkb_ref); - spin_unlock_bh(&ls->ls_lkbidr_spin); + read_unlock_bh(&ls->ls_lkbidr_lock); *lkb_ret = lkb; return lkb ? 0 : -ENOENT; @@ -1567,36 +1567,6 @@ static void kill_lkb(struct kref *kref) DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb);); } -/* TODO move this to lib/refcount.c */ -static __must_check bool -dlm_refcount_dec_and_lock_bh(refcount_t *r, spinlock_t *lock) -__cond_acquires(lock) -{ - if (refcount_dec_not_one(r)) - return false; - - spin_lock_bh(lock); - if (!refcount_dec_and_test(r)) { - spin_unlock_bh(lock); - return false; - } - - return true; -} - -/* TODO move this to include/linux/kref.h */ -static inline int dlm_kref_put_lock_bh(struct kref *kref, - void (*release)(struct kref *kref), - spinlock_t *lock) -{ - if (dlm_refcount_dec_and_lock_bh(&kref->refcount, lock)) { - release(kref); - return 1; - } - - return 0; -} - /* __put_lkb() is used when an lkb may not have an rsb attached to it so we need to provide the lockspace explicitly */ @@ -1605,11 +1575,11 @@ static int __put_lkb(struct dlm_ls *ls, struct dlm_lkb *lkb) uint32_t lkid = lkb->lkb_id; int rv; - rv = dlm_kref_put_lock_bh(&lkb->lkb_ref, kill_lkb, - &ls->ls_lkbidr_spin); + rv = dlm_kref_put_write_lock_bh(&lkb->lkb_ref, kill_lkb, + &ls->ls_lkbidr_lock); if (rv) { idr_remove(&ls->ls_lkbidr, lkid); - spin_unlock_bh(&ls->ls_lkbidr_spin); + write_unlock_bh(&ls->ls_lkbidr_lock); detach_lkb(lkb); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 04f4c74831ce..5ce26882159e 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -431,7 +431,7 @@ static int new_lockspace(const char *name, const char *cluster, goto out_lsfree; idr_init(&ls->ls_lkbidr); - spin_lock_init(&ls->ls_lkbidr_spin); + rwlock_init(&ls->ls_lkbidr_lock); INIT_LIST_HEAD(&ls->ls_waiters); spin_lock_init(&ls->ls_waiters_lock); @@ -676,7 +676,7 @@ static int lockspace_busy(struct dlm_ls *ls, int force) { int rv; - spin_lock_bh(&ls->ls_lkbidr_spin); + read_lock_bh(&ls->ls_lkbidr_lock); if (force == 0) { rv = idr_for_each(&ls->ls_lkbidr, lkb_idr_is_any, ls); } else if (force == 1) { @@ -684,7 +684,7 @@ static int lockspace_busy(struct dlm_ls *ls, int force) } else { rv = 0; } - spin_unlock_bh(&ls->ls_lkbidr_spin); + read_unlock_bh(&ls->ls_lkbidr_lock); return rv; } -- cgit v1.2.3-59-g8ed1b From 9253c54e01b6505d348afbc02abaa4d9f8a01395 Mon Sep 17 00:00:00 2001 From: Phillip Lougher Date: Mon, 8 Apr 2024 23:02:06 +0100 Subject: Squashfs: check the inode number is not the invalid value of zero Syskiller has produced an out of bounds access in fill_meta_index(). That out of bounds access is ultimately caused because the inode has an inode number with the invalid value of zero, which was not checked. The reason this causes the out of bounds access is due to following sequence of events: 1. Fill_meta_index() is called to allocate (via empty_meta_index()) and fill a metadata index. It however suffers a data read error and aborts, invalidating the newly returned empty metadata index. It does this by setting the inode number of the index to zero, which means unused (zero is not a valid inode number). 2. When fill_meta_index() is subsequently called again on another read operation, locate_meta_index() returns the previous index because it matches the inode number of 0. Because this index has been returned it is expected to have been filled, and because it hasn't been, an out of bounds access is performed. This patch adds a sanity check which checks that the inode number is not zero when the inode is created and returns -EINVAL if it is. [phillip@squashfs.org.uk: whitespace fix] Link: https://lkml.kernel.org/r/20240409204723.446925-1-phillip@squashfs.org.uk Link: https://lkml.kernel.org/r/20240408220206.435788-1-phillip@squashfs.org.uk Signed-off-by: Phillip Lougher Reported-by: "Ubisectech Sirius" Closes: https://lore.kernel.org/lkml/87f5c007-b8a5-41ae-8b57-431e924c5915.bugreport@ubisectech.com/ Cc: Christian Brauner Cc: Signed-off-by: Andrew Morton --- fs/squashfs/inode.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c index aa3411354e66..16bd693d0b3a 100644 --- a/fs/squashfs/inode.c +++ b/fs/squashfs/inode.c @@ -48,6 +48,10 @@ static int squashfs_new_inode(struct super_block *sb, struct inode *inode, gid_t i_gid; int err; + inode->i_ino = le32_to_cpu(sqsh_ino->inode_number); + if (inode->i_ino == 0) + return -EINVAL; + err = squashfs_get_id(sb, le16_to_cpu(sqsh_ino->uid), &i_uid); if (err) return err; @@ -58,7 +62,6 @@ static int squashfs_new_inode(struct super_block *sb, struct inode *inode, i_uid_write(inode, i_uid); i_gid_write(inode, i_gid); - inode->i_ino = le32_to_cpu(sqsh_ino->inode_number); inode_set_mtime(inode, le32_to_cpu(sqsh_ino->mtime), 0); inode_set_atime(inode, inode_get_mtime_sec(inode), 0); inode_set_ctime(inode, inode_get_mtime_sec(inode), 0); -- cgit v1.2.3-59-g8ed1b From c4a7dc9523b59b3e73fd522c73e95e072f876b16 Mon Sep 17 00:00:00 2001 From: Jeongjun Park Date: Tue, 16 Apr 2024 03:20:48 +0900 Subject: nilfs2: fix OOB in nilfs_set_de_type The size of the nilfs_type_by_mode array in the fs/nilfs2/dir.c file is defined as "S_IFMT >> S_SHIFT", but the nilfs_set_de_type() function, which uses this array, specifies the index to read from the array in the same way as "(mode & S_IFMT) >> S_SHIFT". static void nilfs_set_de_type(struct nilfs_dir_entry *de, struct inode *inode) { umode_t mode = inode->i_mode; de->file_type = nilfs_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; // oob } However, when the index is determined this way, an out-of-bounds (OOB) error occurs by referring to an index that is 1 larger than the array size when the condition "mode & S_IFMT == S_IFMT" is satisfied. Therefore, a patch to resize the nilfs_type_by_mode array should be applied to prevent OOB errors. Link: https://lkml.kernel.org/r/20240415182048.7144-1-konishi.ryusuke@gmail.com Reported-by: syzbot+2e22057de05b9f3b30d8@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=2e22057de05b9f3b30d8 Fixes: 2ba466d74ed7 ("nilfs2: directory entry operations") Signed-off-by: Jeongjun Park Signed-off-by: Ryusuke Konishi Tested-by: Ryusuke Konishi Cc: Signed-off-by: Andrew Morton --- fs/nilfs2/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c index bc846b904b68..aee40db7a036 100644 --- a/fs/nilfs2/dir.c +++ b/fs/nilfs2/dir.c @@ -240,7 +240,7 @@ nilfs_filetype_table[NILFS_FT_MAX] = { #define S_SHIFT 12 static unsigned char -nilfs_type_by_mode[S_IFMT >> S_SHIFT] = { +nilfs_type_by_mode[(S_IFMT >> S_SHIFT) + 1] = { [S_IFREG >> S_SHIFT] = NILFS_FT_REG_FILE, [S_IFDIR >> S_SHIFT] = NILFS_FT_DIR, [S_IFCHR >> S_SHIFT] = NILFS_FT_CHRDEV, -- cgit v1.2.3-59-g8ed1b From 02bed83d59e37da30b745e30129511b1cc595c92 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 16 Apr 2024 17:55:02 -0400 Subject: bcachefs: Fix null ptr deref in twf from BCH_IOCTL_FSCK_OFFLINE We need to initialize the stdio redirects before they're used. Signed-off-by: Kent Overstreet --- fs/bcachefs/chardev.c | 4 +++- fs/bcachefs/thread_with_file.c | 15 +++++++++++++-- fs/bcachefs/thread_with_file.h | 3 +++ 3 files changed, 19 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index 72781aad6ba7..4d14f19f5185 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -232,13 +232,15 @@ static long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_a /* We need request_key() to be called before we punt to kthread: */ opt_set(thr->opts, nostart, true); + bch2_thread_with_stdio_init(&thr->thr, &bch2_offline_fsck_ops); + thr->c = bch2_fs_open(devs.data, arg.nr_devs, thr->opts); if (!IS_ERR(thr->c) && thr->c->opts.errors == BCH_ON_ERROR_panic) thr->c->opts.errors = BCH_ON_ERROR_ro; - ret = bch2_run_thread_with_stdio(&thr->thr, &bch2_offline_fsck_ops); + ret = __bch2_run_thread_with_stdio(&thr->thr); out: darray_for_each(devs, i) kfree(*i); diff --git a/fs/bcachefs/thread_with_file.c b/fs/bcachefs/thread_with_file.c index 940db15d6a93..b1af7ac430f6 100644 --- a/fs/bcachefs/thread_with_file.c +++ b/fs/bcachefs/thread_with_file.c @@ -294,16 +294,27 @@ static int thread_with_stdio_fn(void *arg) return 0; } -int bch2_run_thread_with_stdio(struct thread_with_stdio *thr, - const struct thread_with_stdio_ops *ops) +void bch2_thread_with_stdio_init(struct thread_with_stdio *thr, + const struct thread_with_stdio_ops *ops) { stdio_buf_init(&thr->stdio.input); stdio_buf_init(&thr->stdio.output); thr->ops = ops; +} +int __bch2_run_thread_with_stdio(struct thread_with_stdio *thr) +{ return bch2_run_thread_with_file(&thr->thr, &thread_with_stdio_fops, thread_with_stdio_fn); } +int bch2_run_thread_with_stdio(struct thread_with_stdio *thr, + const struct thread_with_stdio_ops *ops) +{ + bch2_thread_with_stdio_init(thr, ops); + + return __bch2_run_thread_with_stdio(thr); +} + int bch2_run_thread_with_stdout(struct thread_with_stdio *thr, const struct thread_with_stdio_ops *ops) { diff --git a/fs/bcachefs/thread_with_file.h b/fs/bcachefs/thread_with_file.h index af54ea8f5b0f..1d63d14d7dca 100644 --- a/fs/bcachefs/thread_with_file.h +++ b/fs/bcachefs/thread_with_file.h @@ -63,6 +63,9 @@ struct thread_with_stdio { const struct thread_with_stdio_ops *ops; }; +void bch2_thread_with_stdio_init(struct thread_with_stdio *, + const struct thread_with_stdio_ops *); +int __bch2_run_thread_with_stdio(struct thread_with_stdio *); int bch2_run_thread_with_stdio(struct thread_with_stdio *, const struct thread_with_stdio_ops *); int bch2_run_thread_with_stdout(struct thread_with_stdio *, -- cgit v1.2.3-59-g8ed1b From 9fd5a48a1e3ff79ed54922668279ccb4d7190a62 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 16 Apr 2024 08:16:02 -0700 Subject: bcachefs: Fix format specifier in validate_bset_keys() When building for 32-bit platforms, for which size_t is 'unsigned int', there is a warning from a format string in validate_bset_keys(): fs/bcachefs/btree_io.c: In function 'validate_bset_keys': fs/bcachefs/btree_io.c:891:34: error: format '%lu' expects argument of type 'long unsigned int', but argument 12 has type 'unsigned int' [-Werror=format=] 891 | "bad k->u64s %u (min %u max %lu)", k->u64s, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ fs/bcachefs/btree_io.c:603:32: note: in definition of macro 'btree_err' 603 | msg, ##__VA_ARGS__); \ | ^~~ fs/bcachefs/btree_io.c:887:21: note: in expansion of macro 'btree_err_on' 887 | if (btree_err_on(!bkeyp_u64s_valid(&b->format, k), | ^~~~~~~~~~~~ fs/bcachefs/btree_io.c:891:64: note: format string is defined here 891 | "bad k->u64s %u (min %u max %lu)", k->u64s, | ~~^ | | | long unsigned int | %u cc1: all warnings being treated as errors BKEY_U64s is size_t so the entire expression is promoted to size_t. Use the '%zu' specifier so that there is no warning regardless of the width of size_t. Fixes: 031ad9e7dbd1 ("bcachefs: Check for packed bkeys that are too big") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202404130747.wH6Dd23p-lkp@intel.com/ Closes: https://lore.kernel.org/oe-kbuild-all/202404131536.HdAMBOVc-lkp@intel.com/ Signed-off-by: Nathan Chancellor Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 9678b2375bed..debb0edc3455 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -888,7 +888,7 @@ static int validate_bset_keys(struct bch_fs *c, struct btree *b, -BCH_ERR_btree_node_read_err_fixable, c, NULL, b, i, btree_node_bkey_bad_u64s, - "bad k->u64s %u (min %u max %lu)", k->u64s, + "bad k->u64s %u (min %u max %zu)", k->u64s, bkeyp_key_u64s(&b->format, k), U8_MAX - BKEY_U64s + bkeyp_key_u64s(&b->format, k))) goto drop_this_key; -- cgit v1.2.3-59-g8ed1b From fabb4d49854281027454b0fa305d33f6c9ec4b47 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 15 Apr 2024 22:54:10 -0400 Subject: bcachefs: node scan: ignore multiple nodes with same seq if interior Interior nodes are not really needed, when we have to scan - but if this pops up for leaf nodes we'll need a real heuristic. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_node_scan.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index 866bd278439f..c60794264da2 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -302,6 +302,8 @@ again: start->max_key = bpos_predecessor(n->min_key); start->range_updated = true; + } else if (n->level) { + n->overwritten = true; } else { struct printbuf buf = PRINTBUF; -- cgit v1.2.3-59-g8ed1b From 79055f50a65fe5eb58e9da1f79fb0a4f4bc82fff Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 15 Apr 2024 23:53:12 -0400 Subject: bcachefs: make sure to release last journal pin in replay This fixes a deadlock when journal replay has many keys to insert that were from fsck, not the journal. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 0f328aba9760..be5b47619327 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -249,7 +249,10 @@ int bch2_journal_replay(struct bch_fs *c) struct journal_key *k = *kp; - replay_now_at(j, k->journal_seq); + if (k->journal_seq) + replay_now_at(j, k->journal_seq); + else + replay_now_at(j, j->replay_journal_seq_end); ret = commit_do(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc| -- cgit v1.2.3-59-g8ed1b From c6854e5a267c28300ff045480b5a7ee7f6f1d913 Mon Sep 17 00:00:00 2001 From: Ilya Denisyev Date: Fri, 12 Apr 2024 18:53:54 +0300 Subject: jffs2: prevent xattr node from overflowing the eraseblock Add a check to make sure that the requested xattr node size is no larger than the eraseblock minus the cleanmarker. Unlike the usual inode nodes, the xattr nodes aren't split into parts and spread across multiple eraseblocks, which means that a xattr node must not occupy more than one eraseblock. If the requested xattr value is too large, the xattr node can spill onto the next eraseblock, overwriting the nodes and causing errors such as: jffs2: argh. node added in wrong place at 0x0000b050(2) jffs2: nextblock 0x0000a000, expected at 0000b00c jffs2: error: (823) do_verify_xattr_datum: node CRC failed at 0x01e050, read=0xfc892c93, calc=0x000000 jffs2: notice: (823) jffs2_get_inode_nodes: Node header CRC failed at 0x01e00c. {848f,2fc4,0fef511f,59a3d171} jffs2: Node at 0x0000000c with length 0x00001044 would run over the end of the erase block jffs2: Perhaps the file system was created with the wrong erase size? jffs2: jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000010: 0x1044 instead This breaks the filesystem and can lead to KASAN crashes such as: BUG: KASAN: slab-out-of-bounds in jffs2_sum_add_kvec+0x125e/0x15d0 Read of size 4 at addr ffff88802c31e914 by task repro/830 CPU: 0 PID: 830 Comm: repro Not tainted 6.9.0-rc3+ #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Arch Linux 1.16.3-1-1 04/01/2014 Call Trace: dump_stack_lvl+0xc6/0x120 print_report+0xc4/0x620 ? __virt_addr_valid+0x308/0x5b0 kasan_report+0xc1/0xf0 ? jffs2_sum_add_kvec+0x125e/0x15d0 ? jffs2_sum_add_kvec+0x125e/0x15d0 jffs2_sum_add_kvec+0x125e/0x15d0 jffs2_flash_direct_writev+0xa8/0xd0 jffs2_flash_writev+0x9c9/0xef0 ? __x64_sys_setxattr+0xc4/0x160 ? do_syscall_64+0x69/0x140 ? entry_SYSCALL_64_after_hwframe+0x76/0x7e [...] Found by Linux Verification Center (linuxtesting.org) with Syzkaller. Fixes: aa98d7cf59b5 ("[JFFS2][XATTR] XATTR support on JFFS2 (version. 5)") Signed-off-by: Ilya Denisyev Link: https://lore.kernel.org/r/20240412155357.237803-1-dev@elkcl.ru Signed-off-by: Christian Brauner --- fs/jffs2/xattr.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c index 00224f3a8d6e..defb4162c3d5 100644 --- a/fs/jffs2/xattr.c +++ b/fs/jffs2/xattr.c @@ -1110,6 +1110,9 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, return rc; request = PAD(sizeof(struct jffs2_raw_xattr) + strlen(xname) + 1 + size); + if (request > c->sector_size - c->cleanmarker_size) + return -ERANGE; + rc = jffs2_reserve_space(c, request, &length, ALLOC_NORMAL, JFFS2_SUMMARY_XATTR_SIZE); if (rc) { -- cgit v1.2.3-59-g8ed1b From 23cdd0eed3f1fff3af323092b0b88945a7950d8e Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 15 Apr 2024 11:20:54 -0400 Subject: libfs: Fix simple_offset_rename_exchange() User space expects the replacement (old) directory entry to have the same directory offset after the rename. Suggested-by: Christian Brauner Fixes: a2e459555c5f ("shmem: stable directory offsets") Signed-off-by: Chuck Lever Link: https://lore.kernel.org/r/20240415152057.4605-2-cel@kernel.org Signed-off-by: Christian Brauner --- fs/libfs.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/libfs.c b/fs/libfs.c index 3a6f2cb364f8..ab61fae92cde 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -295,6 +295,18 @@ int simple_offset_add(struct offset_ctx *octx, struct dentry *dentry) return 0; } +static int simple_offset_replace(struct offset_ctx *octx, struct dentry *dentry, + long offset) +{ + int ret; + + ret = mtree_store(&octx->mt, offset, dentry, GFP_KERNEL); + if (ret) + return ret; + offset_set(dentry, offset); + return 0; +} + /** * simple_offset_remove - Remove an entry to a directory's offset map * @octx: directory offset ctx to be updated @@ -352,6 +364,9 @@ int simple_offset_empty(struct dentry *dentry) * @new_dir: destination parent * @new_dentry: destination dentry * + * This API preserves the directory offset values. Caller provides + * appropriate serialization. + * * Returns zero on success. Otherwise a negative errno is returned and the * rename is rolled back. */ @@ -369,11 +384,11 @@ int simple_offset_rename_exchange(struct inode *old_dir, simple_offset_remove(old_ctx, old_dentry); simple_offset_remove(new_ctx, new_dentry); - ret = simple_offset_add(new_ctx, old_dentry); + ret = simple_offset_replace(new_ctx, old_dentry, new_index); if (ret) goto out_restore; - ret = simple_offset_add(old_ctx, new_dentry); + ret = simple_offset_replace(old_ctx, new_dentry, old_index); if (ret) { simple_offset_remove(new_ctx, old_dentry); goto out_restore; @@ -388,10 +403,8 @@ int simple_offset_rename_exchange(struct inode *old_dir, return 0; out_restore: - offset_set(old_dentry, old_index); - mtree_store(&old_ctx->mt, old_index, old_dentry, GFP_KERNEL); - offset_set(new_dentry, new_index); - mtree_store(&new_ctx->mt, new_index, new_dentry, GFP_KERNEL); + (void)simple_offset_replace(old_ctx, old_dentry, old_index); + (void)simple_offset_replace(new_ctx, new_dentry, new_index); return ret; } -- cgit v1.2.3-59-g8ed1b From 5a1a25be995e1014abd01600479915683e356f5c Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 15 Apr 2024 11:20:55 -0400 Subject: libfs: Add simple_offset_rename() API I'm about to fix a tmpfs rename bug that requires the use of internal simple_offset helpers that are not available in mm/shmem.c Signed-off-by: Chuck Lever Link: https://lore.kernel.org/r/20240415152057.4605-3-cel@kernel.org Signed-off-by: Christian Brauner --- fs/libfs.c | 21 +++++++++++++++++++++ include/linux/fs.h | 2 ++ mm/shmem.c | 3 +-- 3 files changed, 24 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/libfs.c b/fs/libfs.c index ab61fae92cde..c392a6edd393 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -357,6 +357,27 @@ int simple_offset_empty(struct dentry *dentry) return ret; } +/** + * simple_offset_rename - handle directory offsets for rename + * @old_dir: parent directory of source entry + * @old_dentry: dentry of source entry + * @new_dir: parent_directory of destination entry + * @new_dentry: dentry of destination + * + * Caller provides appropriate serialization. + * + * Returns zero on success, a negative errno value on failure. + */ +int simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) +{ + struct offset_ctx *old_ctx = old_dir->i_op->get_offset_ctx(old_dir); + struct offset_ctx *new_ctx = new_dir->i_op->get_offset_ctx(new_dir); + + simple_offset_remove(old_ctx, old_dentry); + return simple_offset_add(new_ctx, old_dentry); +} + /** * simple_offset_rename_exchange - exchange rename with directory offsets * @old_dir: parent of dentry being moved diff --git a/include/linux/fs.h b/include/linux/fs.h index 00fc429b0af0..82c707e40fea 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3338,6 +3338,8 @@ void simple_offset_init(struct offset_ctx *octx); int simple_offset_add(struct offset_ctx *octx, struct dentry *dentry); void simple_offset_remove(struct offset_ctx *octx, struct dentry *dentry); int simple_offset_empty(struct dentry *dentry); +int simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry); int simple_offset_rename_exchange(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, diff --git a/mm/shmem.c b/mm/shmem.c index 0aad0d9a621b..c0fb65223963 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -3473,8 +3473,7 @@ static int shmem_rename2(struct mnt_idmap *idmap, return error; } - simple_offset_remove(shmem_get_offset_ctx(old_dir), old_dentry); - error = simple_offset_add(shmem_get_offset_ctx(new_dir), old_dentry); + error = simple_offset_rename(old_dir, old_dentry, new_dir, new_dentry); if (error) return error; -- cgit v1.2.3-59-g8ed1b From ad191eb6d6942bb835a0b20b647f7c53c1d99ca4 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 15 Apr 2024 11:20:56 -0400 Subject: shmem: Fix shmem_rename2() When renaming onto an existing directory entry, user space expects the replacement entry to have the same directory offset as the original one. Link: https://gitlab.alpinelinux.org/alpine/aports/-/issues/15966 Fixes: a2e459555c5f ("shmem: stable directory offsets") Signed-off-by: Chuck Lever Link: https://lore.kernel.org/r/20240415152057.4605-4-cel@kernel.org Signed-off-by: Christian Brauner --- fs/libfs.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'fs') diff --git a/fs/libfs.c b/fs/libfs.c index c392a6edd393..b635ee5adbcc 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -366,6 +366,9 @@ int simple_offset_empty(struct dentry *dentry) * * Caller provides appropriate serialization. * + * User space expects the directory offset value of the replaced + * (new) directory entry to be unchanged after a rename. + * * Returns zero on success, a negative errno value on failure. */ int simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry, @@ -373,8 +376,14 @@ int simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry, { struct offset_ctx *old_ctx = old_dir->i_op->get_offset_ctx(old_dir); struct offset_ctx *new_ctx = new_dir->i_op->get_offset_ctx(new_dir); + long new_offset = dentry2offset(new_dentry); simple_offset_remove(old_ctx, old_dentry); + + if (new_offset) { + offset_set(new_dentry, 0); + return simple_offset_replace(new_ctx, old_dentry, new_offset); + } return simple_offset_add(new_ctx, old_dentry); } -- cgit v1.2.3-59-g8ed1b From cef359074cad6456aa060f4d701bd519a5c7baff Mon Sep 17 00:00:00 2001 From: Lenko Donchev Date: Sat, 10 Feb 2024 19:53:14 -0600 Subject: fs/ntfs3: use kcalloc() instead of kzalloc() We are trying to get rid of all multiplications from allocation functions to prevent integer overflows[1]. Here the multiplication is obviously safe, but using kcalloc() is more appropriate and improves readability. This patch has no effect on runtime behavior. Link: https://github.com/KSPP/linux/issues/162 [1] Link: https://www.kernel.org/doc/html/next/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments [2] Signed-off-by: Lenko Donchev Signed-off-by: Konstantin Komarov --- fs/ntfs3/frecord.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ntfs3/frecord.c b/fs/ntfs3/frecord.c index 7f27382e0ce2..0008670939a4 100644 --- a/fs/ntfs3/frecord.c +++ b/fs/ntfs3/frecord.c @@ -2636,7 +2636,7 @@ int ni_read_frame(struct ntfs_inode *ni, u64 frame_vbo, struct page **pages, goto out1; } - pages_disk = kzalloc(npages_disk * sizeof(struct page *), GFP_NOFS); + pages_disk = kcalloc(npages_disk, sizeof(*pages_disk), GFP_NOFS); if (!pages_disk) { err = -ENOMEM; goto out2; -- cgit v1.2.3-59-g8ed1b From 93b4d70f6a4129fcfad18deae248bf89c129d1b5 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 18 Mar 2024 14:28:50 -0400 Subject: fs/ntfs3: remove atomic_open atomic_open is an optional VFS operation, and is primarily for network filesystems. NFS (for instance) can just send an open call for the last path component rather than doing a lookup and then having to follow that up with an open when it doesn't have a dentry in cache. ntfs3 is a local filesystem however, and its atomic_open just does a typical lookup + open, but in a convoluted way. atomic_open will also make directory leases more difficult to implement on the filesystem. Remove ntfs_atomic_open. Signed-off-by: Jeff Layton Signed-off-by: Konstantin Komarov --- fs/ntfs3/namei.c | 90 -------------------------------------------------------- 1 file changed, 90 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/namei.c b/fs/ntfs3/namei.c index 084d19d78397..edb6a7141246 100644 --- a/fs/ntfs3/namei.c +++ b/fs/ntfs3/namei.c @@ -358,95 +358,6 @@ out: return err; } -/* - * ntfs_atomic_open - * - * inode_operations::atomic_open - */ -static int ntfs_atomic_open(struct inode *dir, struct dentry *dentry, - struct file *file, u32 flags, umode_t mode) -{ - int err; - struct inode *inode; - struct ntfs_fnd *fnd = NULL; - struct ntfs_inode *ni = ntfs_i(dir); - struct dentry *d = NULL; - struct cpu_str *uni = __getname(); - bool locked = false; - - if (!uni) - return -ENOMEM; - - err = ntfs_nls_to_utf16(ni->mi.sbi, dentry->d_name.name, - dentry->d_name.len, uni, NTFS_NAME_LEN, - UTF16_HOST_ENDIAN); - if (err < 0) - goto out; - -#ifdef CONFIG_NTFS3_FS_POSIX_ACL - if (IS_POSIXACL(dir)) { - /* - * Load in cache current acl to avoid ni_lock(dir): - * ntfs_create_inode -> ntfs_init_acl -> posix_acl_create -> - * ntfs_get_acl -> ntfs_get_acl_ex -> ni_lock - */ - struct posix_acl *p = get_inode_acl(dir, ACL_TYPE_DEFAULT); - - if (IS_ERR(p)) { - err = PTR_ERR(p); - goto out; - } - posix_acl_release(p); - } -#endif - - if (d_in_lookup(dentry)) { - ni_lock_dir(ni); - locked = true; - fnd = fnd_get(); - if (!fnd) { - err = -ENOMEM; - goto out1; - } - - d = d_splice_alias(dir_search_u(dir, uni, fnd), dentry); - if (IS_ERR(d)) { - err = PTR_ERR(d); - d = NULL; - goto out2; - } - - if (d) - dentry = d; - } - - if (!(flags & O_CREAT) || d_really_is_positive(dentry)) { - err = finish_no_open(file, d); - goto out2; - } - - file->f_mode |= FMODE_CREATED; - - /* - * fnd contains tree's path to insert to. - * If fnd is not NULL then dir is locked. - */ - inode = ntfs_create_inode(file_mnt_idmap(file), dir, dentry, uni, - mode, 0, NULL, 0, fnd); - err = IS_ERR(inode) ? PTR_ERR(inode) : - finish_open(file, dentry, ntfs_file_open); - dput(d); - -out2: - fnd_put(fnd); -out1: - if (locked) - ni_unlock(ni); -out: - __putname(uni); - return err; -} - struct dentry *ntfs3_get_parent(struct dentry *child) { struct inode *inode = d_inode(child); @@ -612,7 +523,6 @@ const struct inode_operations ntfs_dir_inode_operations = { .setattr = ntfs3_setattr, .getattr = ntfs_getattr, .listxattr = ntfs_listxattr, - .atomic_open = ntfs_atomic_open, .fiemap = ntfs_fiemap, }; -- cgit v1.2.3-59-g8ed1b From 795bb82d12a16a4cee42845b0e4c7e3276574e5d Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Tue, 16 Apr 2024 21:14:52 +0300 Subject: fsnotify: fix UAF from FS_ERROR event on a shutting down filesystem Protect against use after free when filesystem calls fsnotify_sb_error() during fs shutdown. Move freeing of sb->s_fsnotify_info to destroy_super_work(), because it may be accessed from fs shutdown context. Reported-by: syzbot+5e3f9b2a67b45f16d4e6@syzkaller.appspotmail.com Suggested-by: Jan Kara Link: https://lore.kernel.org/linux-fsdevel/20240416173211.4lnmgctyo4jn5fha@quack3/ Fixes: 07a3b8d0bf72 ("fsnotify: lazy attach fsnotify_sb_info state to sb") Reviewed-by: Christian Brauner Signed-off-by: Amir Goldstein Signed-off-by: Jan Kara Message-Id: <20240416181452.567070-1-amir73il@gmail.com> --- fs/notify/fsnotify.c | 6 +++++- fs/super.c | 1 + include/linux/fsnotify_backend.h | 4 ++++ 3 files changed, 10 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 2ae965ef37e8..ff69ae24c4e8 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -103,7 +103,11 @@ void fsnotify_sb_delete(struct super_block *sb) WARN_ON(fsnotify_sb_has_priority_watchers(sb, FSNOTIFY_PRIO_CONTENT)); WARN_ON(fsnotify_sb_has_priority_watchers(sb, FSNOTIFY_PRIO_PRE_CONTENT)); - kfree(sbinfo); +} + +void fsnotify_sb_free(struct super_block *sb) +{ + kfree(sb->s_fsnotify_info); } /* diff --git a/fs/super.c b/fs/super.c index 71d9779c42b1..81656c7807db 100644 --- a/fs/super.c +++ b/fs/super.c @@ -274,6 +274,7 @@ static void destroy_super_work(struct work_struct *work) { struct super_block *s = container_of(work, struct super_block, destroy_work); + fsnotify_sb_free(s); security_sb_free(s); put_user_ns(s->s_user_ns); kfree(s->s_subtype); diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h index 7f1ab8264e41..4dd6143db271 100644 --- a/include/linux/fsnotify_backend.h +++ b/include/linux/fsnotify_backend.h @@ -576,6 +576,7 @@ extern int __fsnotify_parent(struct dentry *dentry, __u32 mask, const void *data extern void __fsnotify_inode_delete(struct inode *inode); extern void __fsnotify_vfsmount_delete(struct vfsmount *mnt); extern void fsnotify_sb_delete(struct super_block *sb); +extern void fsnotify_sb_free(struct super_block *sb); extern u32 fsnotify_get_cookie(void); static inline __u32 fsnotify_parent_needed_mask(__u32 mask) @@ -880,6 +881,9 @@ static inline void __fsnotify_vfsmount_delete(struct vfsmount *mnt) static inline void fsnotify_sb_delete(struct super_block *sb) {} +static inline void fsnotify_sb_free(struct super_block *sb) +{} + static inline void fsnotify_update_flags(struct dentry *dentry) {} -- cgit v1.2.3-59-g8ed1b From 7b012732d005fba22912dda038dd253e3b9a1bfc Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 17 Apr 2024 15:13:22 -0400 Subject: dlm: fix sleep in atomic context This patch changes the orphans mutex to a spinlock since commit c288745f1d4a ("dlm: avoid blocking receive at the end of recovery") is using a rwlock_t to lock the DLM message receive path and do_purge() can be called while this lock is held that forbids to sleep. We need to use spin_lock_bh() because also a user context that calls dlm_user_purge() can call do_purge() and since commit 92d59adfaf71 ("dlm: do message processing in softirq context") the DLM message receive path is done under softirq context. Fixes: c288745f1d4a ("dlm: avoid blocking receive at the end of recovery") Reported-by: Dan Carpenter Closes: https://lore.kernel.org/gfs2/9ad928eb-2ece-4ad9-a79c-d2bce228e4bc@moroto.mountain/ Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/dlm_internal.h | 2 +- fs/dlm/lock.c | 12 ++++++------ fs/dlm/lockspace.c | 2 +- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 19e57cbd5b13..9085ba3b2f20 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -602,7 +602,7 @@ struct dlm_ls { spinlock_t ls_waiters_lock; struct list_head ls_waiters; /* lkbs needing a reply */ - struct mutex ls_orphans_mutex; + spinlock_t ls_orphans_lock; struct list_head ls_orphans; spinlock_t ls_new_rsb_spin; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index bbbc9593a64e..f103b8c30592 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -5880,7 +5880,7 @@ int dlm_user_adopt_orphan(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, int found_other_mode = 0; int rv = 0; - mutex_lock(&ls->ls_orphans_mutex); + spin_lock_bh(&ls->ls_orphans_lock); list_for_each_entry(iter, &ls->ls_orphans, lkb_ownqueue) { if (iter->lkb_resource->res_length != namelen) continue; @@ -5897,7 +5897,7 @@ int dlm_user_adopt_orphan(struct dlm_ls *ls, struct dlm_user_args *ua_tmp, *lkid = iter->lkb_id; break; } - mutex_unlock(&ls->ls_orphans_mutex); + spin_unlock_bh(&ls->ls_orphans_lock); if (!lkb && found_other_mode) { rv = -EAGAIN; @@ -6089,9 +6089,9 @@ static int orphan_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb) int error; hold_lkb(lkb); /* reference for the ls_orphans list */ - mutex_lock(&ls->ls_orphans_mutex); + spin_lock_bh(&ls->ls_orphans_lock); list_add_tail(&lkb->lkb_ownqueue, &ls->ls_orphans); - mutex_unlock(&ls->ls_orphans_mutex); + spin_unlock_bh(&ls->ls_orphans_lock); set_unlock_args(0, lkb->lkb_ua, &args); @@ -6241,7 +6241,7 @@ static void do_purge(struct dlm_ls *ls, int nodeid, int pid) { struct dlm_lkb *lkb, *safe; - mutex_lock(&ls->ls_orphans_mutex); + spin_lock_bh(&ls->ls_orphans_lock); list_for_each_entry_safe(lkb, safe, &ls->ls_orphans, lkb_ownqueue) { if (pid && lkb->lkb_ownpid != pid) continue; @@ -6249,7 +6249,7 @@ static void do_purge(struct dlm_ls *ls, int nodeid, int pid) list_del_init(&lkb->lkb_ownqueue); dlm_put_lkb(lkb); } - mutex_unlock(&ls->ls_orphans_mutex); + spin_unlock_bh(&ls->ls_orphans_lock); } static int send_purge(struct dlm_ls *ls, int nodeid, int pid) diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 5ce26882159e..ed23787271b1 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -436,7 +436,7 @@ static int new_lockspace(const char *name, const char *cluster, INIT_LIST_HEAD(&ls->ls_waiters); spin_lock_init(&ls->ls_waiters_lock); INIT_LIST_HEAD(&ls->ls_orphans); - mutex_init(&ls->ls_orphans_mutex); + spin_lock_init(&ls->ls_orphans_lock); INIT_LIST_HEAD(&ls->ls_new_rsb); spin_lock_init(&ls->ls_new_rsb_spin); -- cgit v1.2.3-59-g8ed1b From fa845c73497f5e9d2f6f1cf48c3aad05c2fdacb8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 16 Apr 2024 19:16:45 -0400 Subject: bcachefs: Fix bch2_dev_btree_bitmap_marked_sectors() shift Fixes: 27c15ed297cb bcachefs: bch_member.btree_allocated_bitmap Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-members.c | 4 ++-- fs/bcachefs/sb-members.h | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 522a969345e5..5b8e621ac5eb 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -463,8 +463,8 @@ static void __bch2_dev_btree_bitmap_mark(struct bch_sb_field_members_v2 *mi, uns m->btree_bitmap_shift += resize; } - for (unsigned bit = sectors >> m->btree_bitmap_shift; - bit << m->btree_bitmap_shift < end; + for (unsigned bit = start >> m->btree_bitmap_shift; + (u64) bit << m->btree_bitmap_shift < end; bit++) bitmap |= BIT_ULL(bit); diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index b27c3e4467cf..5efa64eca5f8 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -235,11 +235,11 @@ static inline bool bch2_dev_btree_bitmap_marked_sectors(struct bch_dev *ca, u64 { u64 end = start + sectors; - if (end > 64 << ca->mi.btree_bitmap_shift) + if (end > 64ULL << ca->mi.btree_bitmap_shift) return false; - for (unsigned bit = sectors >> ca->mi.btree_bitmap_shift; - bit << ca->mi.btree_bitmap_shift < end; + for (unsigned bit = start >> ca->mi.btree_bitmap_shift; + (u64) bit << ca->mi.btree_bitmap_shift < end; bit++) if (!(ca->mi.btree_allocated_bitmap & BIT_ULL(bit))) return false; -- cgit v1.2.3-59-g8ed1b From 605109ff5e43addefdf92d1cfa2a693114430024 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 17 Apr 2024 02:04:23 -0400 Subject: bcachefs: KEY_TYPE_error is allowed for reflink KEY_TYPE_error is left behind when we have to delete all pointers in an extent in fsck; it allows errors to be correctly returned by reads later. Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 085987435a5e..f7fbfccd2b1e 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -1504,7 +1504,8 @@ enum btree_id_flags { BIT_ULL(KEY_TYPE_stripe)) \ x(reflink, 7, BTREE_ID_EXTENTS|BTREE_ID_DATA, \ BIT_ULL(KEY_TYPE_reflink_v)| \ - BIT_ULL(KEY_TYPE_indirect_inline_data)) \ + BIT_ULL(KEY_TYPE_indirect_inline_data)| \ + BIT_ULL(KEY_TYPE_error)) \ x(subvolumes, 8, 0, \ BIT_ULL(KEY_TYPE_subvolume)) \ x(snapshots, 9, 0, \ -- cgit v1.2.3-59-g8ed1b From 719aec84b106ba3bd3639eddb2be46c510ef683a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 17 Apr 2024 02:17:21 -0400 Subject: bcachefs: fix leak in bch2_gc_write_reflink_key Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index ecbd9598f69f..791470b0c654 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -1587,7 +1587,7 @@ static int bch2_gc_write_reflink_key(struct btree_trans *trans, struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, k); ret = PTR_ERR_OR_ZERO(new); if (ret) - return ret; + goto out; if (!r->refcount) new->k.type = KEY_TYPE_deleted; @@ -1595,6 +1595,7 @@ static int bch2_gc_write_reflink_key(struct btree_trans *trans, *bkey_refcount(bkey_i_to_s(new)) = cpu_to_le64(r->refcount); ret = bch2_trans_update(trans, iter, new, 0); } +out: fsck_err: printbuf_exit(&buf); return ret; -- cgit v1.2.3-59-g8ed1b From 0389c09b2fb702ca7924ddf550ce0c8af708b8be Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 17 Apr 2024 17:27:43 -0400 Subject: bcachefs: Fix bio alloc in check_extent_checksum() if the buffer is virtually mapped it won't be a single bvec Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index fadb1078903d..a20044201002 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -470,7 +470,7 @@ found: goto err; } - bio = bio_alloc(ca->disk_sb.bdev, 1, REQ_OP_READ, GFP_KERNEL); + bio = bio_alloc(ca->disk_sb.bdev, buf_pages(data_buf, bytes), REQ_OP_READ, GFP_KERNEL); bio->bi_iter.bi_sector = p.ptr.offset; bch2_bio_map(bio, data_buf, bytes); ret = submit_bio_wait(bio); -- cgit v1.2.3-59-g8ed1b From e5a78fdec0114266d3c47df413d2d7955807fad9 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Tue, 24 Jan 2023 20:44:38 +0100 Subject: btrfs: remove colon from messages with state The message format in syslog is usually made of two parts: prefix ":" message Various tools parse the prefix up to the first ":". When there's an additional status of a btrfs filesystem like [5.199782] BTRFS info (device nvme1n1p1: state M): use zstd compression, level 9 where 'M' is for remount, there's one more ":" that does not conform to the format. Remove it. Reviewed-by: Anand Jain Signed-off-by: David Sterba --- fs/btrfs/messages.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/messages.c b/fs/btrfs/messages.c index c96dd66fd0f7..210d9c82e2ae 100644 --- a/fs/btrfs/messages.c +++ b/fs/btrfs/messages.c @@ -7,7 +7,7 @@ #ifdef CONFIG_PRINTK -#define STATE_STRING_PREFACE ": state " +#define STATE_STRING_PREFACE " state " #define STATE_STRING_BUF_LEN (sizeof(STATE_STRING_PREFACE) + BTRFS_FS_STATE_COUNT + 1) /* -- cgit v1.2.3-59-g8ed1b From 7192833c4e55b26e8f15ef58577867a1bc808036 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Tue, 9 Apr 2024 23:18:52 +0900 Subject: btrfs: scrub: run relocation repair when/only needed When btrfs scrub finds an error, it reads mirrors to find correct data. If all the errors are fixed, sctx->error_bitmap is cleared for the stripe range. However, in the zoned mode, it runs relocation to repair scrub errors when the bitmap is *not* empty, which is a flipped condition. Also, it runs the relocation even if the scrub is read-only. This was missed by a fix in commit 1f2030ff6e49 ("btrfs: scrub: respect the read-only flag during repair"). The repair is only necessary when there is a repaired sector and should be done on read-write scrub. So, tweak the condition for both regular and zoned case. Fixes: 54765392a1b9 ("btrfs: scrub: introduce helper to queue a stripe for scrub") Fixes: 1f2030ff6e49 ("btrfs: scrub: respect the read-only flag during repair") CC: stable@vger.kernel.org # 6.6+ Reviewed-by: Qu Wenruo Reviewed-by: Johannes Thumshirn Signed-off-by: Naohiro Aota Signed-off-by: David Sterba --- fs/btrfs/scrub.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index fa25004ab04e..4b22cfe9a98c 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -1012,6 +1012,7 @@ static void scrub_stripe_read_repair_worker(struct work_struct *work) struct btrfs_fs_info *fs_info = sctx->fs_info; int num_copies = btrfs_num_copies(fs_info, stripe->bg->start, stripe->bg->length); + unsigned long repaired; int mirror; int i; @@ -1078,16 +1079,15 @@ out: * Submit the repaired sectors. For zoned case, we cannot do repair * in-place, but queue the bg to be relocated. */ - if (btrfs_is_zoned(fs_info)) { - if (!bitmap_empty(&stripe->error_bitmap, stripe->nr_sectors)) + bitmap_andnot(&repaired, &stripe->init_error_bitmap, &stripe->error_bitmap, + stripe->nr_sectors); + if (!sctx->readonly && !bitmap_empty(&repaired, stripe->nr_sectors)) { + if (btrfs_is_zoned(fs_info)) { btrfs_repair_one_zone(fs_info, sctx->stripes[0].bg->start); - } else if (!sctx->readonly) { - unsigned long repaired; - - bitmap_andnot(&repaired, &stripe->init_error_bitmap, - &stripe->error_bitmap, stripe->nr_sectors); - scrub_write_sectors(sctx, stripe, repaired, false); - wait_scrub_stripe_io(stripe); + } else { + scrub_write_sectors(sctx, stripe, repaired, false); + wait_scrub_stripe_io(stripe); + } } scrub_stripe_report_errors(sctx, stripe); -- cgit v1.2.3-59-g8ed1b From 131a821a243f89be312ced9e62ccc37b2cf3846c Mon Sep 17 00:00:00 2001 From: Sweet Tea Dorminy Date: Sat, 6 Apr 2024 04:45:02 -0400 Subject: btrfs: fallback if compressed IO fails for ENOSPC In commit b4ccace878f4 ("btrfs: refactor submit_compressed_extents()"), if an async extent compressed but failed to find enough space, we changed from falling back to an uncompressed write to just failing the write altogether. The principle was that if there's not enough space to write the compressed version of the data, there can't possibly be enough space to write the larger, uncompressed version of the data. However, this isn't necessarily true: due to fragmentation, there could be enough discontiguous free blocks to write the uncompressed version, but not enough contiguous free blocks to write the smaller but unsplittable compressed version. This has occurred to an internal workload which relied on write()'s return value indicating there was space. While rare, it has happened a few times. Thus, in order to prevent early ENOSPC, re-add a fallback to uncompressed writing. Fixes: b4ccace878f4 ("btrfs: refactor submit_compressed_extents()") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Qu Wenruo Co-developed-by: Neal Gompa Signed-off-by: Neal Gompa Signed-off-by: Sweet Tea Dorminy Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c65fe5de4022..7fed887e700c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1145,13 +1145,13 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, 0, *alloc_hint, &ins, 1, 1); if (ret) { /* - * Here we used to try again by going back to non-compressed - * path for ENOSPC. But we can't reserve space even for - * compressed size, how could it work for uncompressed size - * which requires larger size? So here we directly go error - * path. + * We can't reserve contiguous space for the compressed size. + * Unlikely, but it's possible that we could have enough + * non-contiguous space for the uncompressed size instead. So + * fall back to uncompressed. */ - goto out_free; + submit_uncompressed_range(inode, async_extent, locked_page); + goto done; } /* Here we're doing allocation and writeback of the compressed pages */ @@ -1203,7 +1203,6 @@ done: out_free_reserve: btrfs_dec_block_group_reservations(fs_info, ins.objectid); btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1); -out_free: mapping_set_error(inode->vfs_inode.i_mapping, -EIO); extent_clear_unlock_delalloc(inode, start, end, NULL, EXTENT_LOCKED | EXTENT_DELALLOC | -- cgit v1.2.3-59-g8ed1b From 2f7ef5bb4a2f3e481ef05fab946edb97c84f67cf Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Wed, 17 Apr 2024 10:45:47 +0200 Subject: btrfs: fix information leak in btrfs_ioctl_logical_to_ino() Syzbot reported the following information leak for in btrfs_ioctl_logical_to_ino(): BUG: KMSAN: kernel-infoleak in instrument_copy_to_user include/linux/instrumented.h:114 [inline] BUG: KMSAN: kernel-infoleak in _copy_to_user+0xbc/0x110 lib/usercopy.c:40 instrument_copy_to_user include/linux/instrumented.h:114 [inline] _copy_to_user+0xbc/0x110 lib/usercopy.c:40 copy_to_user include/linux/uaccess.h:191 [inline] btrfs_ioctl_logical_to_ino+0x440/0x750 fs/btrfs/ioctl.c:3499 btrfs_ioctl+0x714/0x1260 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:904 [inline] __se_sys_ioctl+0x261/0x450 fs/ioctl.c:890 __x64_sys_ioctl+0x96/0xe0 fs/ioctl.c:890 x64_sys_call+0x1883/0x3b50 arch/x86/include/generated/asm/syscalls_64.h:17 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Uninit was created at: __kmalloc_large_node+0x231/0x370 mm/slub.c:3921 __do_kmalloc_node mm/slub.c:3954 [inline] __kmalloc_node+0xb07/0x1060 mm/slub.c:3973 kmalloc_node include/linux/slab.h:648 [inline] kvmalloc_node+0xc0/0x2d0 mm/util.c:634 kvmalloc include/linux/slab.h:766 [inline] init_data_container+0x49/0x1e0 fs/btrfs/backref.c:2779 btrfs_ioctl_logical_to_ino+0x17c/0x750 fs/btrfs/ioctl.c:3480 btrfs_ioctl+0x714/0x1260 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:904 [inline] __se_sys_ioctl+0x261/0x450 fs/ioctl.c:890 __x64_sys_ioctl+0x96/0xe0 fs/ioctl.c:890 x64_sys_call+0x1883/0x3b50 arch/x86/include/generated/asm/syscalls_64.h:17 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Bytes 40-65535 of 65536 are uninitialized Memory access of size 65536 starts at ffff888045a40000 This happens, because we're copying a 'struct btrfs_data_container' back to user-space. This btrfs_data_container is allocated in 'init_data_container()' via kvmalloc(), which does not zero-fill the memory. Fix this by using kvzalloc() which zeroes out the memory on allocation. CC: stable@vger.kernel.org # 4.14+ Reported-by: Reviewed-by: Qu Wenruo Reviewed-by: Filipe Manana Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/backref.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index c1e6a5bbeeaf..58110c968667 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -2776,20 +2776,14 @@ struct btrfs_data_container *init_data_container(u32 total_bytes) size_t alloc_bytes; alloc_bytes = max_t(size_t, total_bytes, sizeof(*data)); - data = kvmalloc(alloc_bytes, GFP_KERNEL); + data = kvzalloc(alloc_bytes, GFP_KERNEL); if (!data) return ERR_PTR(-ENOMEM); - if (total_bytes >= sizeof(*data)) { + if (total_bytes >= sizeof(*data)) data->bytes_left = total_bytes - sizeof(*data); - data->bytes_missing = 0; - } else { + else data->bytes_missing = sizeof(*data) - total_bytes; - data->bytes_left = 0; - } - - data->elem_cnt = 0; - data->elem_missed = 0; return data; } -- cgit v1.2.3-59-g8ed1b From fe1c6c7acce10baf9521d6dccc17268d91ee2305 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Tue, 9 Apr 2024 20:32:34 +0930 Subject: btrfs: fix wrong block_start calculation for btrfs_drop_extent_map_range() [BUG] During my extent_map cleanup/refactor, with extra sanity checks, extent-map-tests::test_case_7() would not pass the checks. The problem is, after btrfs_drop_extent_map_range(), the resulted extent_map has a @block_start way too large. Meanwhile my btrfs_file_extent_item based members are returning a correct @disk_bytenr/@offset combination. The extent map layout looks like this: 0 16K 32K 48K | PINNED | | Regular | The regular em at [32K, 48K) also has 32K @block_start. Then drop range [0, 36K), which should shrink the regular one to be [36K, 48K). However the @block_start is incorrect, we expect 32K + 4K, but got 52K. [CAUSE] Inside btrfs_drop_extent_map_range() function, if we hit an extent_map that covers the target range but is still beyond it, we need to split that extent map into half: |<-- drop range -->| |<----- existing extent_map --->| And if the extent map is not compressed, we need to forward extent_map::block_start by the difference between the end of drop range and the extent map start. However in that particular case, the difference is calculated using (start + len - em->start). The problem is @start can be modified if the drop range covers any pinned extent. This leads to wrong calculation, and would be caught by my later extent_map sanity checks, which checks the em::block_start against btrfs_file_extent_item::disk_bytenr + btrfs_file_extent_item::offset. This is a regression caused by commit c962098ca4af ("btrfs: fix incorrect splitting in btrfs_drop_extent_map_range"), which removed the @len update for pinned extents. [FIX] Fix it by avoiding using @start completely, and use @end - em->start instead, which @end is exclusive bytenr number. And update the test case to verify the @block_start to prevent such problem from happening. Thankfully this is not going to lead to any data corruption, as IO path does not utilize btrfs_drop_extent_map_range() with @skip_pinned set. So this fix is only here for the sake of consistency/correctness. CC: stable@vger.kernel.org # 6.5+ Fixes: c962098ca4af ("btrfs: fix incorrect splitting in btrfs_drop_extent_map_range") Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 2 +- fs/btrfs/tests/extent-map-tests.c | 5 +++++ 2 files changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 445f7716f1e2..24a048210b15 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -817,7 +817,7 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, split->block_len = em->block_len; split->orig_start = em->orig_start; } else { - const u64 diff = start + len - em->start; + const u64 diff = end - em->start; split->block_len = split->len; split->block_start += diff; diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c index 253cce7ffecf..47b5d301038e 100644 --- a/fs/btrfs/tests/extent-map-tests.c +++ b/fs/btrfs/tests/extent-map-tests.c @@ -847,6 +847,11 @@ static int test_case_7(struct btrfs_fs_info *fs_info) goto out; } + if (em->block_start != SZ_32K + SZ_4K) { + test_err("em->block_start is %llu, expected 36K", em->block_start); + goto out; + } + free_extent_map(em); read_lock(&em_tree->lock); -- cgit v1.2.3-59-g8ed1b From 7643f3fe27729b20e6fcf6b314b00b8b93504356 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Wed, 17 Apr 2024 20:01:55 +0000 Subject: f2fs: assign the write hint per stream by default This reverts commit 930e2607638d ("f2fs: remove obsolete whint_mode"), as we decide to pass write hints to the disk. Cc: Hyunchul Lee Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- Documentation/filesystems/f2fs.rst | 29 +++++++++++++++ fs/f2fs/data.c | 2 + fs/f2fs/f2fs.h | 3 ++ fs/f2fs/file.c | 15 +++++++- fs/f2fs/segment.c | 76 +++++++++++++++++++++++++++++++++++--- 5 files changed, 118 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/Documentation/filesystems/f2fs.rst b/Documentation/filesystems/f2fs.rst index efc3493fd6f8..68a0885fb5e6 100644 --- a/Documentation/filesystems/f2fs.rst +++ b/Documentation/filesystems/f2fs.rst @@ -774,6 +774,35 @@ In order to identify whether the data in the victim segment are valid or not, F2FS manages a bitmap. Each bit represents the validity of a block, and the bitmap is composed of a bit stream covering whole blocks in main area. +Write-hint Policy +----------------- + +F2FS sets the whint all the time with the below policy. + +===================== ======================== =================== +User F2FS Block +===================== ======================== =================== +N/A META WRITE_LIFE_NONE|REQ_META +N/A HOT_NODE WRITE_LIFE_NONE +N/A WARM_NODE WRITE_LIFE_MEDIUM +N/A COLD_NODE WRITE_LIFE_LONG +ioctl(COLD) COLD_DATA WRITE_LIFE_EXTREME +extension list " " + +-- buffered io +N/A COLD_DATA WRITE_LIFE_EXTREME +N/A HOT_DATA WRITE_LIFE_SHORT +N/A WARM_DATA WRITE_LIFE_NOT_SET + +-- direct io +WRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME +WRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT +WRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET +WRITE_LIFE_NONE " WRITE_LIFE_NONE +WRITE_LIFE_MEDIUM " WRITE_LIFE_MEDIUM +WRITE_LIFE_LONG " WRITE_LIFE_LONG +===================== ======================== =================== + Fallocate(2) Policy ------------------- diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 5d641fac02ba..ed7d08785fcf 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -465,6 +465,8 @@ static struct bio *__bio_alloc(struct f2fs_io_info *fio, int npages) } else { bio->bi_end_io = f2fs_write_end_io; bio->bi_private = sbi; + bio->bi_write_hint = f2fs_io_type_to_rw_hint(sbi, + fio->type, fio->temp); } iostat_alloc_and_bind_ctx(sbi, bio, NULL); diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index dd530dc70005..3f7196122574 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3722,6 +3722,7 @@ void f2fs_replace_block(struct f2fs_sb_info *sbi, struct dnode_of_data *dn, block_t old_addr, block_t new_addr, unsigned char version, bool recover_curseg, bool recover_newaddr); +int f2fs_get_segment_temp(int seg_type); int f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, block_t old_blkaddr, block_t *new_blkaddr, struct f2fs_summary *sum, int type, @@ -3745,6 +3746,8 @@ void f2fs_destroy_segment_manager(struct f2fs_sb_info *sbi); int __init f2fs_create_segment_manager_caches(void); void f2fs_destroy_segment_manager_caches(void); int f2fs_rw_hint_to_seg_type(enum rw_hint hint); +enum rw_hint f2fs_io_type_to_rw_hint(struct f2fs_sb_info *sbi, + enum page_type type, enum temp_type temp); unsigned int f2fs_usable_segs_in_sec(struct f2fs_sb_info *sbi, unsigned int segno); unsigned int f2fs_usable_blks_in_seg(struct f2fs_sb_info *sbi, diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index ac1ae85f3cc3..d382f8bc2fbe 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4685,8 +4685,21 @@ static int f2fs_dio_write_end_io(struct kiocb *iocb, ssize_t size, int error, return 0; } +static void f2fs_dio_write_submit_io(const struct iomap_iter *iter, + struct bio *bio, loff_t file_offset) +{ + struct inode *inode = iter->inode; + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + int seg_type = f2fs_rw_hint_to_seg_type(inode->i_write_hint); + enum temp_type temp = f2fs_get_segment_temp(seg_type); + + bio->bi_write_hint = f2fs_io_type_to_rw_hint(sbi, DATA, temp); + submit_bio(bio); +} + static const struct iomap_dio_ops f2fs_iomap_dio_write_ops = { - .end_io = f2fs_dio_write_end_io, + .end_io = f2fs_dio_write_end_io, + .submit_io = f2fs_dio_write_submit_io, }; static void f2fs_flush_buffered_write(struct address_space *mapping, diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index f0da516ba8dc..2206199e8099 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3364,6 +3364,65 @@ int f2fs_rw_hint_to_seg_type(enum rw_hint hint) } } +/* + * This returns write hints for each segment type. This hints will be + * passed down to block layer as below by default. + * + * User F2FS Block + * ---- ---- ----- + * META WRITE_LIFE_NONE|REQ_META + * HOT_NODE WRITE_LIFE_NONE + * WARM_NODE WRITE_LIFE_MEDIUM + * COLD_NODE WRITE_LIFE_LONG + * ioctl(COLD) COLD_DATA WRITE_LIFE_EXTREME + * extension list " " + * + * -- buffered io + * COLD_DATA WRITE_LIFE_EXTREME + * HOT_DATA WRITE_LIFE_SHORT + * WARM_DATA WRITE_LIFE_NOT_SET + * + * -- direct io + * WRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME + * WRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT + * WRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET + * WRITE_LIFE_NONE " WRITE_LIFE_NONE + * WRITE_LIFE_MEDIUM " WRITE_LIFE_MEDIUM + * WRITE_LIFE_LONG " WRITE_LIFE_LONG + */ +enum rw_hint f2fs_io_type_to_rw_hint(struct f2fs_sb_info *sbi, + enum page_type type, enum temp_type temp) +{ + switch (type) { + case DATA: + switch (temp) { + case WARM: + return WRITE_LIFE_NOT_SET; + case HOT: + return WRITE_LIFE_SHORT; + case COLD: + return WRITE_LIFE_EXTREME; + default: + return WRITE_LIFE_NONE; + } + case NODE: + switch (temp) { + case WARM: + return WRITE_LIFE_MEDIUM; + case HOT: + return WRITE_LIFE_NONE; + case COLD: + return WRITE_LIFE_LONG; + default: + return WRITE_LIFE_NONE; + } + case META: + return WRITE_LIFE_NONE; + default: + return WRITE_LIFE_NONE; + } +} + static int __get_segment_type_2(struct f2fs_io_info *fio) { if (fio->type == DATA) @@ -3443,6 +3502,15 @@ static int __get_segment_type_6(struct f2fs_io_info *fio) } } +int f2fs_get_segment_temp(int seg_type) +{ + if (IS_HOT(seg_type)) + return HOT; + else if (IS_WARM(seg_type)) + return WARM; + return COLD; +} + static int __get_segment_type(struct f2fs_io_info *fio) { int type = 0; @@ -3461,12 +3529,8 @@ static int __get_segment_type(struct f2fs_io_info *fio) f2fs_bug_on(fio->sbi, true); } - if (IS_HOT(type)) - fio->temp = HOT; - else if (IS_WARM(type)) - fio->temp = WARM; - else - fio->temp = COLD; + fio->temp = f2fs_get_segment_temp(type); + return type; } -- cgit v1.2.3-59-g8ed1b From db92e6c729d87e6f5b0f467f01a96dbf1e452106 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 16 Apr 2024 15:23:15 +0800 Subject: f2fs: convert f2fs_mpage_readpages() to use folio Convert f2fs_mpage_readpages() to use folio and related functionality. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 80 +++++++++++++++++++++++++++++----------------------------- 1 file changed, 40 insertions(+), 40 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index ed7d08785fcf..2fd86cff81e2 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2345,7 +2345,7 @@ out: * Major change was from block_size == page_size in f2fs by default. */ static int f2fs_mpage_readpages(struct inode *inode, - struct readahead_control *rac, struct page *page) + struct readahead_control *rac, struct folio *folio) { struct bio *bio = NULL; sector_t last_block_in_bio = 0; @@ -2365,6 +2365,7 @@ static int f2fs_mpage_readpages(struct inode *inode, #endif unsigned nr_pages = rac ? readahead_count(rac) : 1; unsigned max_nr_pages = nr_pages; + pgoff_t index; int ret = 0; map.m_pblk = 0; @@ -2378,64 +2379,63 @@ static int f2fs_mpage_readpages(struct inode *inode, for (; nr_pages; nr_pages--) { if (rac) { - page = readahead_page(rac); - prefetchw(&page->flags); + folio = readahead_folio(rac); + prefetchw(&folio->flags); } -#ifdef CONFIG_F2FS_FS_COMPRESSION - if (f2fs_compressed_file(inode)) { - /* there are remained compressed pages, submit them */ - if (!f2fs_cluster_can_merge_page(&cc, page->index)) { - ret = f2fs_read_multi_pages(&cc, &bio, - max_nr_pages, - &last_block_in_bio, - rac != NULL, false); - f2fs_destroy_compress_ctx(&cc, false); - if (ret) - goto set_error_page; - } - if (cc.cluster_idx == NULL_CLUSTER) { - if (nc_cluster_idx == - page->index >> cc.log_cluster_size) { - goto read_single_page; - } - - ret = f2fs_is_compressed_cluster(inode, page->index); - if (ret < 0) - goto set_error_page; - else if (!ret) { - nc_cluster_idx = - page->index >> cc.log_cluster_size; - goto read_single_page; - } + index = folio_index(folio); - nc_cluster_idx = NULL_CLUSTER; - } - ret = f2fs_init_compress_ctx(&cc); +#ifdef CONFIG_F2FS_FS_COMPRESSION + if (!f2fs_compressed_file(inode)) + goto read_single_page; + + /* there are remained compressed pages, submit them */ + if (!f2fs_cluster_can_merge_page(&cc, index)) { + ret = f2fs_read_multi_pages(&cc, &bio, + max_nr_pages, + &last_block_in_bio, + rac != NULL, false); + f2fs_destroy_compress_ctx(&cc, false); if (ret) goto set_error_page; + } + if (cc.cluster_idx == NULL_CLUSTER) { + if (nc_cluster_idx == index >> cc.log_cluster_size) + goto read_single_page; - f2fs_compress_ctx_add_page(&cc, page); + ret = f2fs_is_compressed_cluster(inode, index); + if (ret < 0) + goto set_error_page; + else if (!ret) { + nc_cluster_idx = + index >> cc.log_cluster_size; + goto read_single_page; + } - goto next_page; + nc_cluster_idx = NULL_CLUSTER; } + ret = f2fs_init_compress_ctx(&cc); + if (ret) + goto set_error_page; + + f2fs_compress_ctx_add_page(&cc, &folio->page); + + goto next_page; read_single_page: #endif - ret = f2fs_read_single_page(inode, page, max_nr_pages, &map, + ret = f2fs_read_single_page(inode, &folio->page, max_nr_pages, &map, &bio, &last_block_in_bio, rac); if (ret) { #ifdef CONFIG_F2FS_FS_COMPRESSION set_error_page: #endif - zero_user_segment(page, 0, PAGE_SIZE); - unlock_page(page); + folio_zero_segment(folio, 0, folio_size(folio)); + folio_unlock(folio); } #ifdef CONFIG_F2FS_FS_COMPRESSION next_page: #endif - if (rac) - put_page(page); #ifdef CONFIG_F2FS_FS_COMPRESSION if (f2fs_compressed_file(inode)) { @@ -2472,7 +2472,7 @@ static int f2fs_read_data_folio(struct file *file, struct folio *folio) if (f2fs_has_inline_data(inode)) ret = f2fs_read_inline_data(inode, page); if (ret == -EAGAIN) - ret = f2fs_mpage_readpages(inode, NULL, page); + ret = f2fs_mpage_readpages(inode, NULL, folio); return ret; } -- cgit v1.2.3-59-g8ed1b From ed54eed355675f79d9322c07d92f38037fc6b514 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 16 Apr 2024 15:23:16 +0800 Subject: f2fs: convert f2fs_read_single_page() to use folio Convert f2fs_read_single_page() to use folio and related functionality. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 2fd86cff81e2..8874ca3a32e2 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2063,7 +2063,7 @@ static inline loff_t f2fs_readpage_limit(struct inode *inode) return i_size_read(inode); } -static int f2fs_read_single_page(struct inode *inode, struct page *page, +static int f2fs_read_single_page(struct inode *inode, struct folio *folio, unsigned nr_pages, struct f2fs_map_blocks *map, struct bio **bio_ret, @@ -2076,9 +2076,10 @@ static int f2fs_read_single_page(struct inode *inode, struct page *page, sector_t last_block; sector_t last_block_in_file; sector_t block_nr; + pgoff_t index = folio_index(folio); int ret = 0; - block_in_file = (sector_t)page_index(page); + block_in_file = (sector_t)index; last_block = block_in_file + nr_pages; last_block_in_file = bytes_to_blks(inode, f2fs_readpage_limit(inode) + blocksize - 1); @@ -2109,7 +2110,7 @@ static int f2fs_read_single_page(struct inode *inode, struct page *page, got_it: if ((map->m_flags & F2FS_MAP_MAPPED)) { block_nr = map->m_pblk + block_in_file - map->m_lblk; - SetPageMappedToDisk(page); + folio_set_mappedtodisk(folio); if (!f2fs_is_valid_blkaddr(F2FS_I_SB(inode), block_nr, DATA_GENERIC_ENHANCE_READ)) { @@ -2118,15 +2119,15 @@ got_it: } } else { zero_out: - zero_user_segment(page, 0, PAGE_SIZE); - if (f2fs_need_verity(inode, page->index) && - !fsverity_verify_page(page)) { + folio_zero_segment(folio, 0, folio_size(folio)); + if (f2fs_need_verity(inode, index) && + !fsverity_verify_folio(folio)) { ret = -EIO; goto out; } - if (!PageUptodate(page)) - SetPageUptodate(page); - unlock_page(page); + if (!folio_test_uptodate(folio)) + folio_mark_uptodate(folio); + folio_unlock(folio); goto out; } @@ -2136,14 +2137,14 @@ zero_out: */ if (bio && (!page_is_mergeable(F2FS_I_SB(inode), bio, *last_block_in_bio, block_nr) || - !f2fs_crypt_mergeable_bio(bio, inode, page->index, NULL))) { + !f2fs_crypt_mergeable_bio(bio, inode, index, NULL))) { submit_and_realloc: f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); bio = NULL; } if (bio == NULL) { bio = f2fs_grab_read_bio(inode, block_nr, nr_pages, - is_readahead ? REQ_RAHEAD : 0, page->index, + is_readahead ? REQ_RAHEAD : 0, index, false); if (IS_ERR(bio)) { ret = PTR_ERR(bio); @@ -2158,7 +2159,7 @@ submit_and_realloc: */ f2fs_wait_on_block_writeback(inode, block_nr); - if (bio_add_page(bio, page, blocksize, 0) < blocksize) + if (!bio_add_folio(bio, folio, blocksize, 0)) goto submit_and_realloc; inc_page_count(F2FS_I_SB(inode), F2FS_RD_DATA); @@ -2424,7 +2425,7 @@ static int f2fs_mpage_readpages(struct inode *inode, read_single_page: #endif - ret = f2fs_read_single_page(inode, &folio->page, max_nr_pages, &map, + ret = f2fs_read_single_page(inode, folio, max_nr_pages, &map, &bio, &last_block_in_bio, rac); if (ret) { #ifdef CONFIG_F2FS_FS_COMPRESSION -- cgit v1.2.3-59-g8ed1b From 96ea46f30b2657375fc7695f78ddb2cb50413509 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 16 Apr 2024 15:23:17 +0800 Subject: f2fs: convert f2fs_read_inline_data() to use folio Convert f2fs_read_inline_data() to use folio and related functionality, and also convert its caller to use folio. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 11 +++++------ fs/f2fs/f2fs.h | 4 ++-- fs/f2fs/inline.c | 34 +++++++++++++++++----------------- 3 files changed, 24 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 8874ca3a32e2..763f0952efdd 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2458,20 +2458,19 @@ next_page: static int f2fs_read_data_folio(struct file *file, struct folio *folio) { - struct page *page = &folio->page; - struct inode *inode = page_file_mapping(page)->host; + struct inode *inode = folio_file_mapping(folio)->host; int ret = -EAGAIN; - trace_f2fs_readpage(page, DATA); + trace_f2fs_readpage(&folio->page, DATA); if (!f2fs_is_compress_backend_ready(inode)) { - unlock_page(page); + folio_unlock(folio); return -EOPNOTSUPP; } /* If the file has inline data, try to read it directly */ if (f2fs_has_inline_data(inode)) - ret = f2fs_read_inline_data(inode, page); + ret = f2fs_read_inline_data(inode, folio); if (ret == -EAGAIN) ret = f2fs_mpage_readpages(inode, NULL, folio); return ret; @@ -3400,7 +3399,7 @@ restart: if (f2fs_has_inline_data(inode)) { if (pos + len <= MAX_INLINE_DATA(inode)) { - f2fs_do_read_inline_data(page, ipage); + f2fs_do_read_inline_data(page_folio(page), ipage); set_inode_flag(inode, FI_DATA_EXIST); if (inode->i_nlink) set_page_private_inline(ipage); diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 3f7196122574..a0ae99bcca39 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -4154,10 +4154,10 @@ extern struct kmem_cache *f2fs_inode_entry_slab; bool f2fs_may_inline_data(struct inode *inode); bool f2fs_sanity_check_inline_data(struct inode *inode); bool f2fs_may_inline_dentry(struct inode *inode); -void f2fs_do_read_inline_data(struct page *page, struct page *ipage); +void f2fs_do_read_inline_data(struct folio *folio, struct page *ipage); void f2fs_truncate_inline_inode(struct inode *inode, struct page *ipage, u64 from); -int f2fs_read_inline_data(struct inode *inode, struct page *page); +int f2fs_read_inline_data(struct inode *inode, struct folio *folio); int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page); int f2fs_convert_inline_inode(struct inode *inode); int f2fs_try_convert_inline_dir(struct inode *dir, struct dentry *dentry); diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index 3d3218a4b29d..7638d0d7b7ee 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -61,22 +61,22 @@ bool f2fs_may_inline_dentry(struct inode *inode) return true; } -void f2fs_do_read_inline_data(struct page *page, struct page *ipage) +void f2fs_do_read_inline_data(struct folio *folio, struct page *ipage) { - struct inode *inode = page->mapping->host; + struct inode *inode = folio_file_mapping(folio)->host; - if (PageUptodate(page)) + if (folio_test_uptodate(folio)) return; - f2fs_bug_on(F2FS_P_SB(page), page->index); + f2fs_bug_on(F2FS_I_SB(inode), folio_index(folio)); - zero_user_segment(page, MAX_INLINE_DATA(inode), PAGE_SIZE); + folio_zero_segment(folio, MAX_INLINE_DATA(inode), folio_size(folio)); /* Copy the whole inline data block */ - memcpy_to_page(page, 0, inline_data_addr(inode, ipage), + memcpy_to_folio(folio, 0, inline_data_addr(inode, ipage), MAX_INLINE_DATA(inode)); - if (!PageUptodate(page)) - SetPageUptodate(page); + if (!folio_test_uptodate(folio)) + folio_mark_uptodate(folio); } void f2fs_truncate_inline_inode(struct inode *inode, @@ -97,13 +97,13 @@ void f2fs_truncate_inline_inode(struct inode *inode, clear_inode_flag(inode, FI_DATA_EXIST); } -int f2fs_read_inline_data(struct inode *inode, struct page *page) +int f2fs_read_inline_data(struct inode *inode, struct folio *folio) { struct page *ipage; ipage = f2fs_get_node_page(F2FS_I_SB(inode), inode->i_ino); if (IS_ERR(ipage)) { - unlock_page(page); + folio_unlock(folio); return PTR_ERR(ipage); } @@ -112,15 +112,15 @@ int f2fs_read_inline_data(struct inode *inode, struct page *page) return -EAGAIN; } - if (page->index) - zero_user_segment(page, 0, PAGE_SIZE); + if (folio_index(folio)) + folio_zero_segment(folio, 0, folio_size(folio)); else - f2fs_do_read_inline_data(page, ipage); + f2fs_do_read_inline_data(folio, ipage); - if (!PageUptodate(page)) - SetPageUptodate(page); + if (!folio_test_uptodate(folio)) + folio_mark_uptodate(folio); f2fs_put_page(ipage, 1); - unlock_page(page); + folio_unlock(folio); return 0; } @@ -166,7 +166,7 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page) f2fs_bug_on(F2FS_P_SB(page), folio_test_writeback(page_folio(page))); - f2fs_do_read_inline_data(page, dn->inode_page); + f2fs_do_read_inline_data(page_folio(page), dn->inode_page); set_page_dirty(page); /* clear dirty state */ -- cgit v1.2.3-59-g8ed1b From 92f750d847c997ff4b0f04d83443af00fb729ba3 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 16 Apr 2024 15:23:18 +0800 Subject: f2fs: convert f2fs__page tracepoint class to use folio Convert f2fs__page tracepoint class() and its instances to use folio and related functionality, and rename it to f2fs__folio(). Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 4 ++-- fs/f2fs/data.c | 10 +++++----- fs/f2fs/node.c | 4 ++-- include/trace/events/f2fs.h | 42 +++++++++++++++++++++--------------------- 4 files changed, 30 insertions(+), 30 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index eac698b8dd38..5d05a413f451 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -345,7 +345,7 @@ static int __f2fs_write_meta_page(struct page *page, { struct f2fs_sb_info *sbi = F2FS_P_SB(page); - trace_f2fs_writepage(page, META); + trace_f2fs_writepage(page_folio(page), META); if (unlikely(f2fs_cp_error(sbi))) { if (is_sbi_flag_set(sbi, SBI_IS_CLOSE)) { @@ -492,7 +492,7 @@ stop: static bool f2fs_dirty_meta_folio(struct address_space *mapping, struct folio *folio) { - trace_f2fs_set_page_dirty(&folio->page, META); + trace_f2fs_set_page_dirty(folio, META); if (!folio_test_uptodate(folio)) folio_mark_uptodate(folio); diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 763f0952efdd..bee1e45f76b8 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2461,7 +2461,7 @@ static int f2fs_read_data_folio(struct file *file, struct folio *folio) struct inode *inode = folio_file_mapping(folio)->host; int ret = -EAGAIN; - trace_f2fs_readpage(&folio->page, DATA); + trace_f2fs_readpage(folio, DATA); if (!f2fs_is_compress_backend_ready(inode)) { folio_unlock(folio); @@ -2710,7 +2710,7 @@ got_it: } else { set_inode_flag(inode, FI_UPDATE_WRITE); } - trace_f2fs_do_write_data_page(fio->page, IPU); + trace_f2fs_do_write_data_page(page_folio(page), IPU); return err; } @@ -2739,7 +2739,7 @@ got_it: /* LFS mode write path */ f2fs_outplace_write_data(&dn, fio); - trace_f2fs_do_write_data_page(page, OPU); + trace_f2fs_do_write_data_page(page_folio(page), OPU); set_inode_flag(inode, FI_APPEND_WRITE); out_writepage: f2fs_put_dnode(&dn); @@ -2786,7 +2786,7 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, .last_block = last_block, }; - trace_f2fs_writepage(page, DATA); + trace_f2fs_writepage(page_folio(page), DATA); /* we should bypass data pages to proceed the kworker jobs */ if (unlikely(f2fs_cp_error(sbi))) { @@ -3760,7 +3760,7 @@ static bool f2fs_dirty_data_folio(struct address_space *mapping, { struct inode *inode = mapping->host; - trace_f2fs_set_page_dirty(&folio->page, DATA); + trace_f2fs_set_page_dirty(folio, DATA); if (!folio_test_uptodate(folio)) folio_mark_uptodate(folio); diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 3b9eb5693683..95cecf08cb37 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -1624,7 +1624,7 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted, }; unsigned int seq; - trace_f2fs_writepage(page, NODE); + trace_f2fs_writepage(page_folio(page), NODE); if (unlikely(f2fs_cp_error(sbi))) { /* keep node pages in remount-ro mode */ @@ -2171,7 +2171,7 @@ skip_write: static bool f2fs_dirty_node_folio(struct address_space *mapping, struct folio *folio) { - trace_f2fs_set_page_dirty(&folio->page, NODE); + trace_f2fs_set_page_dirty(folio, NODE); if (!folio_test_uptodate(folio)) folio_mark_uptodate(folio); diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 7ed0fc430dc6..371ba28415f5 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -1304,11 +1304,11 @@ TRACE_EVENT(f2fs_write_end, __entry->copied) ); -DECLARE_EVENT_CLASS(f2fs__page, +DECLARE_EVENT_CLASS(f2fs__folio, - TP_PROTO(struct page *page, int type), + TP_PROTO(struct folio *folio, int type), - TP_ARGS(page, type), + TP_ARGS(folio, type), TP_STRUCT__entry( __field(dev_t, dev) @@ -1321,14 +1321,14 @@ DECLARE_EVENT_CLASS(f2fs__page, ), TP_fast_assign( - __entry->dev = page_file_mapping(page)->host->i_sb->s_dev; - __entry->ino = page_file_mapping(page)->host->i_ino; + __entry->dev = folio_file_mapping(folio)->host->i_sb->s_dev; + __entry->ino = folio_file_mapping(folio)->host->i_ino; __entry->type = type; __entry->dir = - S_ISDIR(page_file_mapping(page)->host->i_mode); - __entry->index = page->index; - __entry->dirty = PageDirty(page); - __entry->uptodate = PageUptodate(page); + S_ISDIR(folio_file_mapping(folio)->host->i_mode); + __entry->index = folio_index(folio); + __entry->dirty = folio_test_dirty(folio); + __entry->uptodate = folio_test_uptodate(folio); ), TP_printk("dev = (%d,%d), ino = %lu, %s, %s, index = %lu, " @@ -1341,32 +1341,32 @@ DECLARE_EVENT_CLASS(f2fs__page, __entry->uptodate) ); -DEFINE_EVENT(f2fs__page, f2fs_writepage, +DEFINE_EVENT(f2fs__folio, f2fs_writepage, - TP_PROTO(struct page *page, int type), + TP_PROTO(struct folio *folio, int type), - TP_ARGS(page, type) + TP_ARGS(folio, type) ); -DEFINE_EVENT(f2fs__page, f2fs_do_write_data_page, +DEFINE_EVENT(f2fs__folio, f2fs_do_write_data_page, - TP_PROTO(struct page *page, int type), + TP_PROTO(struct folio *folio, int type), - TP_ARGS(page, type) + TP_ARGS(folio, type) ); -DEFINE_EVENT(f2fs__page, f2fs_readpage, +DEFINE_EVENT(f2fs__folio, f2fs_readpage, - TP_PROTO(struct page *page, int type), + TP_PROTO(struct folio *folio, int type), - TP_ARGS(page, type) + TP_ARGS(folio, type) ); -DEFINE_EVENT(f2fs__page, f2fs_set_page_dirty, +DEFINE_EVENT(f2fs__folio, f2fs_set_page_dirty, - TP_PROTO(struct page *page, int type), + TP_PROTO(struct folio *folio, int type), - TP_ARGS(page, type) + TP_ARGS(folio, type) ); TRACE_EVENT(f2fs_replace_atomic_write_block, -- cgit v1.2.3-59-g8ed1b From 5bf624c0122960ebeeb544e2bc1a2e531ac11392 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 16 Apr 2024 15:21:07 +0800 Subject: f2fs: fix comment in sanity_check_raw_super() Commit d7e9a9037de2 ("f2fs: Support Block Size == Page Size") missed to adjust comment in sanity_check_raw_super(), fix it. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index df32573d1f62..5f4262f06302 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -3456,7 +3456,7 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi, } } - /* Currently, support only 4KB block size */ + /* only support block_size equals to PAGE_SIZE */ if (le32_to_cpu(raw_super->log_blocksize) != F2FS_BLKSIZE_BITS) { f2fs_info(sbi, "Invalid log_blocksize (%u), supports only %u", le32_to_cpu(raw_super->log_blocksize), -- cgit v1.2.3-59-g8ed1b From 06b206d9e2b4c3e142d60d21912a7b46988dea29 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 16 Apr 2024 15:21:08 +0800 Subject: f2fs: remove unnecessary block size check in init_f2fs_fs() After commit d7e9a9037de2 ("f2fs: Support Block Size == Page Size"), F2FS_BLKSIZE equals to PAGE_SIZE, remove unnecessary check condition. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/super.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 5f4262f06302..ef673f853366 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -4931,12 +4931,6 @@ static int __init init_f2fs_fs(void) { int err; - if (PAGE_SIZE != F2FS_BLKSIZE) { - printk("F2FS not supported on PAGE_SIZE(%lu) != BLOCK_SIZE(%lu)\n", - PAGE_SIZE, F2FS_BLKSIZE); - return -EINVAL; - } - err = init_inodecache(); if (err) goto fail; -- cgit v1.2.3-59-g8ed1b From dad80c6bff770d25f67ec25fe011730e4a463008 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 4 Apr 2024 16:05:19 +0100 Subject: cifs: Fix reacquisition of volume cookie on still-live connection During mount, cifs_mount_get_tcon() gets a tcon resource connection record and then attaches an fscache volume cookie to it. However, it does this irrespective of whether or not the tcon returned from cifs_get_tcon() is a new record or one that's already in use. This leads to a warning about a volume cookie collision and a leaked volume cookie because tcon->fscache gets reset. Fix this be adding a mutex and a "we've already tried this" flag and only doing it once for the lifetime of the tcon. [!] Note: Looking at cifs_mount_get_tcon(), a more general solution may actually be required. Reacquiring the volume cookie isn't the only thing that function does: it also partially reinitialises the tcon record without any locking - which may cause live filesystem ops already using the tcon through a previous mount to malfunction. This can be reproduced simply by something like: mount //example.com/test /xfstest.test -o user=shares,pass=xxx,fsc mount //example.com/test /mnt -o user=shares,pass=xxx,fsc Fixes: 70431bfd825d ("cifs: Support fscache indexing rewrite") Signed-off-by: David Howells Acked-by: Paulo Alcantara (Red Hat) cc: Shyam Prasad N cc: linux-cifs@vger.kernel.org cc: linux-fsdevel@vger.kernel.org Signed-off-by: Steve French --- fs/smb/client/cifsglob.h | 2 ++ fs/smb/client/fscache.c | 13 +++++++++++++ fs/smb/client/misc.c | 3 +++ 3 files changed, 18 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index d6669ce4ae87..fc09d1c0ee07 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1276,7 +1276,9 @@ struct cifs_tcon { __u32 max_cached_dirs; #ifdef CONFIG_CIFS_FSCACHE u64 resource_id; /* server resource id */ + bool fscache_acquired; /* T if we've tried acquiring a cookie */ struct fscache_volume *fscache; /* cookie for share */ + struct mutex fscache_lock; /* Prevent regetting a cookie */ #endif struct list_head pending_opens; /* list of incomplete opens */ struct cached_fids *cfids; diff --git a/fs/smb/client/fscache.c b/fs/smb/client/fscache.c index 340efce8f052..113bde8f1e61 100644 --- a/fs/smb/client/fscache.c +++ b/fs/smb/client/fscache.c @@ -43,12 +43,23 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) char *key; int ret = -ENOMEM; + if (tcon->fscache_acquired) + return 0; + + mutex_lock(&tcon->fscache_lock); + if (tcon->fscache_acquired) { + mutex_unlock(&tcon->fscache_lock); + return 0; + } + tcon->fscache_acquired = true; + tcon->fscache = NULL; switch (sa->sa_family) { case AF_INET: case AF_INET6: break; default: + mutex_unlock(&tcon->fscache_lock); cifs_dbg(VFS, "Unknown network family '%d'\n", sa->sa_family); return -EINVAL; } @@ -57,6 +68,7 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) sharename = extract_sharename(tcon->tree_name); if (IS_ERR(sharename)) { + mutex_unlock(&tcon->fscache_lock); cifs_dbg(FYI, "%s: couldn't extract sharename\n", __func__); return PTR_ERR(sharename); } @@ -90,6 +102,7 @@ out_2: kfree(key); out: kfree(sharename); + mutex_unlock(&tcon->fscache_lock); return ret; } diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c index 7d15a1969b81..ad44f8d66b37 100644 --- a/fs/smb/client/misc.c +++ b/fs/smb/client/misc.c @@ -139,6 +139,9 @@ tcon_info_alloc(bool dir_leases_enabled) atomic_set(&ret_buf->num_local_opens, 0); atomic_set(&ret_buf->num_remote_opens, 0); ret_buf->stats_from_time = ktime_get_real_seconds(); +#ifdef CONFIG_CIFS_FSCACHE + mutex_init(&ret_buf->fscache_lock); +#endif return ret_buf; } -- cgit v1.2.3-59-g8ed1b From afc23febd51c7e24361e3a9c09f3e892eb0a41ea Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 4 Apr 2024 13:51:36 +0100 Subject: cifs: Add tracing for the cifs_tcon struct refcounting Add tracing for the refcounting/lifecycle of the cifs_tcon struct, marking different events with different labels and giving each tcon its own debug ID so that the tracelines corresponding to individual tcons can be distinguished. This can be enabled with: echo 1 >/sys/kernel/debug/tracing/events/cifs/smb3_tcon_ref/enable Signed-off-by: David Howells Acked-by: Paulo Alcantara (Red Hat) cc: Shyam Prasad N cc: linux-cifs@vger.kernel.org cc: linux-fsdevel@vger.kernel.org Signed-off-by: Steve French --- fs/smb/client/cifsfs.c | 2 + fs/smb/client/cifsglob.h | 1 + fs/smb/client/cifsproto.h | 9 ++--- fs/smb/client/connect.c | 21 +++++----- fs/smb/client/fscache.c | 7 ++++ fs/smb/client/misc.c | 10 +++-- fs/smb/client/smb2misc.c | 10 +++-- fs/smb/client/smb2ops.c | 7 +++- fs/smb/client/smb2pdu.c | 8 ++-- fs/smb/client/smb2transport.c | 2 + fs/smb/client/trace.h | 92 ++++++++++++++++++++++++++++++++++++++++++- 11 files changed, 143 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index d41eedbff674..30781789dfd9 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -739,6 +739,8 @@ static void cifs_umount_begin(struct super_block *sb) spin_lock(&cifs_tcp_ses_lock); spin_lock(&tcon->tc_lock); + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_see_umount); if ((tcon->tc_count > 1) || (tcon->status == TID_EXITING)) { /* we have other mounts to same share or we have already tried to umount this and woken up diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index fc09d1c0ee07..6ff35570db81 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1190,6 +1190,7 @@ struct cifs_fattr { */ struct cifs_tcon { struct list_head tcon_list; + int debug_id; /* Debugging for tracing */ int tc_count; struct list_head rlist; /* reconnect list */ spinlock_t tc_lock; /* protect anything here that is not protected */ diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index 8e0a348f1f66..fbc358c09da3 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -303,7 +303,7 @@ cifs_get_tcp_session(struct smb3_fs_context *ctx, struct TCP_Server_Info *primary_server); extern void cifs_put_tcp_session(struct TCP_Server_Info *server, int from_reconnect); -extern void cifs_put_tcon(struct cifs_tcon *tcon); +extern void cifs_put_tcon(struct cifs_tcon *tcon, enum smb3_tcon_ref_trace trace); extern void cifs_release_automount_timer(void); @@ -530,8 +530,9 @@ extern int CIFSSMBLogoff(const unsigned int xid, struct cifs_ses *ses); extern struct cifs_ses *sesInfoAlloc(void); extern void sesInfoFree(struct cifs_ses *); -extern struct cifs_tcon *tcon_info_alloc(bool dir_leases_enabled); -extern void tconInfoFree(struct cifs_tcon *); +extern struct cifs_tcon *tcon_info_alloc(bool dir_leases_enabled, + enum smb3_tcon_ref_trace trace); +extern void tconInfoFree(struct cifs_tcon *tcon, enum smb3_tcon_ref_trace trace); extern int cifs_sign_rqst(struct smb_rqst *rqst, struct TCP_Server_Info *server, __u32 *pexpected_response_sequence_number); @@ -721,8 +722,6 @@ static inline int cifs_create_options(struct cifs_sb_info *cifs_sb, int options) return options; } -struct super_block *cifs_get_tcon_super(struct cifs_tcon *tcon); -void cifs_put_tcon_super(struct super_block *sb); int cifs_wait_for_server_reconnect(struct TCP_Server_Info *server, bool retry); /* Put references of @ses and its children */ diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 4e35970681bf..7a16e12f5da8 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -1943,7 +1943,7 @@ cifs_setup_ipc(struct cifs_ses *ses, struct smb3_fs_context *ctx) } /* no need to setup directory caching on IPC share, so pass in false */ - tcon = tcon_info_alloc(false); + tcon = tcon_info_alloc(false, netfs_trace_tcon_ref_new_ipc); if (tcon == NULL) return -ENOMEM; @@ -1960,7 +1960,7 @@ cifs_setup_ipc(struct cifs_ses *ses, struct smb3_fs_context *ctx) if (rc) { cifs_server_dbg(VFS, "failed to connect to IPC (rc=%d)\n", rc); - tconInfoFree(tcon); + tconInfoFree(tcon, netfs_trace_tcon_ref_free_ipc_fail); goto out; } @@ -2043,7 +2043,7 @@ void __cifs_put_smb_ses(struct cifs_ses *ses) * files on session close, as specified in MS-SMB2 3.3.5.6 Receiving an * SMB2 LOGOFF Request. */ - tconInfoFree(tcon); + tconInfoFree(tcon, netfs_trace_tcon_ref_free_ipc); if (do_logoff) { xid = get_xid(); rc = server->ops->logoff(xid, ses); @@ -2432,6 +2432,8 @@ cifs_find_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) continue; } ++tcon->tc_count; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_get_find); spin_unlock(&tcon->tc_lock); spin_unlock(&cifs_tcp_ses_lock); return tcon; @@ -2441,7 +2443,7 @@ cifs_find_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) } void -cifs_put_tcon(struct cifs_tcon *tcon) +cifs_put_tcon(struct cifs_tcon *tcon, enum smb3_tcon_ref_trace trace) { unsigned int xid; struct cifs_ses *ses; @@ -2457,6 +2459,7 @@ cifs_put_tcon(struct cifs_tcon *tcon) cifs_dbg(FYI, "%s: tc_count=%d\n", __func__, tcon->tc_count); spin_lock(&cifs_tcp_ses_lock); spin_lock(&tcon->tc_lock); + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count - 1, trace); if (--tcon->tc_count > 0) { spin_unlock(&tcon->tc_lock); spin_unlock(&cifs_tcp_ses_lock); @@ -2493,7 +2496,7 @@ cifs_put_tcon(struct cifs_tcon *tcon) _free_xid(xid); cifs_fscache_release_super_cookie(tcon); - tconInfoFree(tcon); + tconInfoFree(tcon, netfs_trace_tcon_ref_free); cifs_put_smb_ses(ses); } @@ -2547,7 +2550,7 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) nohandlecache = ctx->nohandlecache; else nohandlecache = true; - tcon = tcon_info_alloc(!nohandlecache); + tcon = tcon_info_alloc(!nohandlecache, netfs_trace_tcon_ref_new); if (tcon == NULL) { rc = -ENOMEM; goto out_fail; @@ -2737,7 +2740,7 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) return tcon; out_fail: - tconInfoFree(tcon); + tconInfoFree(tcon, netfs_trace_tcon_ref_free_fail); return ERR_PTR(rc); } @@ -2754,7 +2757,7 @@ cifs_put_tlink(struct tcon_link *tlink) } if (!IS_ERR(tlink_tcon(tlink))) - cifs_put_tcon(tlink_tcon(tlink)); + cifs_put_tcon(tlink_tcon(tlink), netfs_trace_tcon_ref_put_tlink); kfree(tlink); } @@ -3319,7 +3322,7 @@ void cifs_mount_put_conns(struct cifs_mount_ctx *mnt_ctx) int rc = 0; if (mnt_ctx->tcon) - cifs_put_tcon(mnt_ctx->tcon); + cifs_put_tcon(mnt_ctx->tcon, netfs_trace_tcon_ref_put_mnt_ctx); else if (mnt_ctx->ses) cifs_put_smb_ses(mnt_ctx->ses); else if (mnt_ctx->server) diff --git a/fs/smb/client/fscache.c b/fs/smb/client/fscache.c index 113bde8f1e61..1a895e6243ee 100644 --- a/fs/smb/client/fscache.c +++ b/fs/smb/client/fscache.c @@ -94,6 +94,11 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) } pr_err("Cache volume key already in use (%s)\n", key); vcookie = NULL; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_see_fscache_collision); + } else { + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_see_fscache_okay); } tcon->fscache = vcookie; @@ -115,6 +120,8 @@ void cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) cifs_fscache_fill_volume_coherency(tcon, &cd); fscache_relinquish_volume(tcon->fscache, &cd, false); tcon->fscache = NULL; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_see_fscache_relinq); } void cifs_fscache_get_inode_cookie(struct inode *inode) diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c index ad44f8d66b37..07c468ddb88a 100644 --- a/fs/smb/client/misc.c +++ b/fs/smb/client/misc.c @@ -111,9 +111,10 @@ sesInfoFree(struct cifs_ses *buf_to_free) } struct cifs_tcon * -tcon_info_alloc(bool dir_leases_enabled) +tcon_info_alloc(bool dir_leases_enabled, enum smb3_tcon_ref_trace trace) { struct cifs_tcon *ret_buf; + static atomic_t tcon_debug_id; ret_buf = kzalloc(sizeof(*ret_buf), GFP_KERNEL); if (!ret_buf) @@ -130,7 +131,8 @@ tcon_info_alloc(bool dir_leases_enabled) atomic_inc(&tconInfoAllocCount); ret_buf->status = TID_NEW; - ++ret_buf->tc_count; + ret_buf->debug_id = atomic_inc_return(&tcon_debug_id); + ret_buf->tc_count = 1; spin_lock_init(&ret_buf->tc_lock); INIT_LIST_HEAD(&ret_buf->openFileList); INIT_LIST_HEAD(&ret_buf->tcon_list); @@ -142,17 +144,19 @@ tcon_info_alloc(bool dir_leases_enabled) #ifdef CONFIG_CIFS_FSCACHE mutex_init(&ret_buf->fscache_lock); #endif + trace_smb3_tcon_ref(ret_buf->debug_id, ret_buf->tc_count, trace); return ret_buf; } void -tconInfoFree(struct cifs_tcon *tcon) +tconInfoFree(struct cifs_tcon *tcon, enum smb3_tcon_ref_trace trace) { if (tcon == NULL) { cifs_dbg(FYI, "Null buffer passed to tconInfoFree\n"); return; } + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, trace); free_cached_dirs(tcon->cfids); atomic_dec(&tconInfoAllocCount); kfree(tcon->nativeFileSystem); diff --git a/fs/smb/client/smb2misc.c b/fs/smb/client/smb2misc.c index cc72be5a93a9..677ef6f99a5b 100644 --- a/fs/smb/client/smb2misc.c +++ b/fs/smb/client/smb2misc.c @@ -767,7 +767,7 @@ smb2_cancelled_close_fid(struct work_struct *work) if (rc) cifs_tcon_dbg(VFS, "Close cancelled mid failed rc:%d\n", rc); - cifs_put_tcon(tcon); + cifs_put_tcon(tcon, netfs_trace_tcon_ref_put_cancelled_close_fid); kfree(cancelled); } @@ -811,6 +811,8 @@ smb2_handle_cancelled_close(struct cifs_tcon *tcon, __u64 persistent_fid, if (tcon->tc_count <= 0) { struct TCP_Server_Info *server = NULL; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_see_cancelled_close); WARN_ONCE(tcon->tc_count < 0, "tcon refcount is negative"); spin_unlock(&cifs_tcp_ses_lock); @@ -823,12 +825,14 @@ smb2_handle_cancelled_close(struct cifs_tcon *tcon, __u64 persistent_fid, return 0; } tcon->tc_count++; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_get_cancelled_close); spin_unlock(&cifs_tcp_ses_lock); rc = __smb2_handle_cancelled_cmd(tcon, SMB2_CLOSE_HE, 0, persistent_fid, volatile_fid); if (rc) - cifs_put_tcon(tcon); + cifs_put_tcon(tcon, netfs_trace_tcon_ref_put_cancelled_close); return rc; } @@ -856,7 +860,7 @@ smb2_handle_cancelled_mid(struct mid_q_entry *mid, struct TCP_Server_Info *serve rsp->PersistentFileId, rsp->VolatileFileId); if (rc) - cifs_put_tcon(tcon); + cifs_put_tcon(tcon, netfs_trace_tcon_ref_put_cancelled_mid); return rc; } diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 78c94d0350fe..28f0b7d19d53 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -2915,8 +2915,11 @@ smb2_get_dfs_refer(const unsigned int xid, struct cifs_ses *ses, tcon = list_first_entry_or_null(&ses->tcon_list, struct cifs_tcon, tcon_list); - if (tcon) + if (tcon) { tcon->tc_count++; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_get_dfs_refer); + } spin_unlock(&cifs_tcp_ses_lock); } @@ -2980,6 +2983,8 @@ smb2_get_dfs_refer(const unsigned int xid, struct cifs_ses *ses, /* ipc tcons are not refcounted */ spin_lock(&cifs_tcp_ses_lock); tcon->tc_count--; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_dec_dfs_refer); /* tc_count can never go negative */ WARN_ON(tcon->tc_count < 0); spin_unlock(&cifs_tcp_ses_lock); diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 86c647a947cc..a5efce03cb58 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -4138,6 +4138,8 @@ void smb2_reconnect_server(struct work_struct *work) list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { if (tcon->need_reconnect || tcon->need_reopen_files) { tcon->tc_count++; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_get_reconnect_server); list_add_tail(&tcon->rlist, &tmp_list); tcon_selected = true; } @@ -4176,14 +4178,14 @@ void smb2_reconnect_server(struct work_struct *work) if (tcon->ipc) cifs_put_smb_ses(tcon->ses); else - cifs_put_tcon(tcon); + cifs_put_tcon(tcon, netfs_trace_tcon_ref_put_reconnect_server); } if (!ses_exist) goto done; /* allocate a dummy tcon struct used for reconnect */ - tcon = tcon_info_alloc(false); + tcon = tcon_info_alloc(false, netfs_trace_tcon_ref_new_reconnect_server); if (!tcon) { resched = true; list_for_each_entry_safe(ses, ses2, &tmp_ses_list, rlist) { @@ -4206,7 +4208,7 @@ void smb2_reconnect_server(struct work_struct *work) list_del_init(&ses->rlist); cifs_put_smb_ses(ses); } - tconInfoFree(tcon); + tconInfoFree(tcon, netfs_trace_tcon_ref_free_reconnect_server); done: cifs_dbg(FYI, "Reconnecting tcons and channels finished\n"); diff --git a/fs/smb/client/smb2transport.c b/fs/smb/client/smb2transport.c index 1d6e54f7879e..02135a605305 100644 --- a/fs/smb/client/smb2transport.c +++ b/fs/smb/client/smb2transport.c @@ -189,6 +189,8 @@ smb2_find_smb_sess_tcon_unlocked(struct cifs_ses *ses, __u32 tid) if (tcon->tid != tid) continue; ++tcon->tc_count; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_get_find_sess_tcon); return tcon; } diff --git a/fs/smb/client/trace.h b/fs/smb/client/trace.h index 5e83cb9da902..604e52876cd2 100644 --- a/fs/smb/client/trace.h +++ b/fs/smb/client/trace.h @@ -3,6 +3,9 @@ * Copyright (C) 2018, Microsoft Corporation. * * Author(s): Steve French + * + * Please use this 3-part article as a reference for writing new tracepoints: + * https://lwn.net/Articles/379903/ */ #undef TRACE_SYSTEM #define TRACE_SYSTEM cifs @@ -15,9 +18,70 @@ #include /* - * Please use this 3-part article as a reference for writing new tracepoints: - * https://lwn.net/Articles/379903/ + * Specify enums for tracing information. + */ +#define smb3_tcon_ref_traces \ + EM(netfs_trace_tcon_ref_dec_dfs_refer, "DEC DfsRef") \ + EM(netfs_trace_tcon_ref_free, "FRE ") \ + EM(netfs_trace_tcon_ref_free_fail, "FRE Fail ") \ + EM(netfs_trace_tcon_ref_free_ipc, "FRE Ipc ") \ + EM(netfs_trace_tcon_ref_free_ipc_fail, "FRE Ipc-F ") \ + EM(netfs_trace_tcon_ref_free_reconnect_server, "FRE Reconn") \ + EM(netfs_trace_tcon_ref_get_cancelled_close, "GET Cn-Cls") \ + EM(netfs_trace_tcon_ref_get_dfs_refer, "GET DfsRef") \ + EM(netfs_trace_tcon_ref_get_find, "GET Find ") \ + EM(netfs_trace_tcon_ref_get_find_sess_tcon, "GET FndSes") \ + EM(netfs_trace_tcon_ref_get_reconnect_server, "GET Reconn") \ + EM(netfs_trace_tcon_ref_new, "NEW ") \ + EM(netfs_trace_tcon_ref_new_ipc, "NEW Ipc ") \ + EM(netfs_trace_tcon_ref_new_reconnect_server, "NEW Reconn") \ + EM(netfs_trace_tcon_ref_put_cancelled_close, "PUT Cn-Cls") \ + EM(netfs_trace_tcon_ref_put_cancelled_close_fid, "PUT Cn-Fid") \ + EM(netfs_trace_tcon_ref_put_cancelled_mid, "PUT Cn-Mid") \ + EM(netfs_trace_tcon_ref_put_mnt_ctx, "PUT MntCtx") \ + EM(netfs_trace_tcon_ref_put_reconnect_server, "PUT Reconn") \ + EM(netfs_trace_tcon_ref_put_tlink, "PUT Tlink ") \ + EM(netfs_trace_tcon_ref_see_cancelled_close, "SEE Cn-Cls") \ + EM(netfs_trace_tcon_ref_see_fscache_collision, "SEE FV-CO!") \ + EM(netfs_trace_tcon_ref_see_fscache_okay, "SEE FV-Ok ") \ + EM(netfs_trace_tcon_ref_see_fscache_relinq, "SEE FV-Rlq") \ + E_(netfs_trace_tcon_ref_see_umount, "SEE Umount") + +#undef EM +#undef E_ + +/* + * Define those tracing enums. + */ +#ifndef __SMB3_DECLARE_TRACE_ENUMS_ONCE_ONLY +#define __SMB3_DECLARE_TRACE_ENUMS_ONCE_ONLY + +#define EM(a, b) a, +#define E_(a, b) a + +enum smb3_tcon_ref_trace { smb3_tcon_ref_traces } __mode(byte); + +#undef EM +#undef E_ +#endif + +/* + * Export enum symbols via userspace. + */ +#define EM(a, b) TRACE_DEFINE_ENUM(a); +#define E_(a, b) TRACE_DEFINE_ENUM(a); + +smb3_tcon_ref_traces; + +#undef EM +#undef E_ + +/* + * Now redefine the EM() and E_() macros to map the enums to the strings that + * will be printed in the output. */ +#define EM(a, b) { a, b }, +#define E_(a, b) { a, b } /* For logging errors in read or write */ DECLARE_EVENT_CLASS(smb3_rw_err_class, @@ -1125,6 +1189,30 @@ DEFINE_SMB3_CREDIT_EVENT(waitff_credits); DEFINE_SMB3_CREDIT_EVENT(overflow_credits); DEFINE_SMB3_CREDIT_EVENT(set_credits); + +TRACE_EVENT(smb3_tcon_ref, + TP_PROTO(unsigned int tcon_debug_id, int ref, + enum smb3_tcon_ref_trace trace), + TP_ARGS(tcon_debug_id, ref, trace), + TP_STRUCT__entry( + __field(unsigned int, tcon) + __field(int, ref) + __field(enum smb3_tcon_ref_trace, trace) + ), + TP_fast_assign( + __entry->tcon = tcon_debug_id; + __entry->ref = ref; + __entry->trace = trace; + ), + TP_printk("TC=%08x %s r=%u", + __entry->tcon, + __print_symbolic(__entry->trace, smb3_tcon_ref_traces), + __entry->ref) + ); + + +#undef EM +#undef E_ #endif /* _CIFS_TRACE_H */ #undef TRACE_INCLUDE_PATH -- cgit v1.2.3-59-g8ed1b From 18d86965e31f9be4d477da0744a7cdc9815858de Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Fri, 19 Apr 2024 12:05:07 -0300 Subject: smb: client: fix rename(2) regression against samba After commit 2c7d399e551c ("smb: client: reuse file lease key in compound operations") the client started reusing lease keys for rename, unlink and set path size operations to prevent it from breaking its own leases and thus causing unnecessary lease breaks to same connection. The implementation relies on positive dentries and cifsInodeInfo::lease_granted to decide whether reusing lease keys for the compound requests. cifsInodeInfo::lease_granted was introduced by commit 0ab95c2510b6 ("Defer close only when lease is enabled.") to indicate whether lease caching is granted for a specific file, but that can only happen until file is open, so cifsInodeInfo::lease_granted was left uninitialised in ->alloc_inode and then client started sending random lease keys for files that hadn't any leases. This fixes the following test case against samba: mount.cifs //srv/share /mnt/1 -o ...,nosharesock mount.cifs //srv/share /mnt/2 -o ...,nosharesock touch /mnt/1/foo; tail -f /mnt/1/foo & pid=$! mv /mnt/2/foo /mnt/2/bar # fails with -EIO kill $pid Fixes: 0ab95c2510b6 ("Defer close only when lease is enabled.") Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/cifsfs.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 30781789dfd9..39277c37185c 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -389,6 +389,7 @@ cifs_alloc_inode(struct super_block *sb) * server, can not assume caching of file data or metadata. */ cifs_set_oplock_level(cifs_inode, 0); + cifs_inode->lease_granted = false; cifs_inode->flags = 0; spin_lock_init(&cifs_inode->writers_lock); cifs_inode->writers = 0; -- cgit v1.2.3-59-g8ed1b From c119f4ede3fa90a9463f50831761c28f989bfb20 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Thu, 11 Apr 2024 23:02:15 +0900 Subject: ksmbd: fix slab-out-of-bounds in smb2_allocate_rsp_buf If ->ProtocolId is SMB2_TRANSFORM_PROTO_NUM, smb2 request size validation could be skipped. if request size is smaller than sizeof(struct smb2_query_info_req), slab-out-of-bounds read can happen in smb2_allocate_rsp_buf(). This patch allocate response buffer after decrypting transform request. smb3_decrypt_req() will validate transform request size and avoid slab-out-of-bound in smb2_allocate_rsp_buf(). Reported-by: Norbert Szetei Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/server.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/server.c b/fs/smb/server/server.c index c0788188aa82..c67fbc8d6683 100644 --- a/fs/smb/server/server.c +++ b/fs/smb/server/server.c @@ -167,20 +167,17 @@ static void __handle_ksmbd_work(struct ksmbd_work *work, int rc; bool is_chained = false; - if (conn->ops->allocate_rsp_buf(work)) - return; - if (conn->ops->is_transform_hdr && conn->ops->is_transform_hdr(work->request_buf)) { rc = conn->ops->decrypt_req(work); - if (rc < 0) { - conn->ops->set_rsp_status(work, STATUS_DATA_ERROR); - goto send; - } - + if (rc < 0) + return; work->encrypted = true; } + if (conn->ops->allocate_rsp_buf(work)) + return; + rc = conn->ops->init_rsp_hdr(work); if (rc) { /* either uid or tid is not correct */ -- cgit v1.2.3-59-g8ed1b From 17cf0c2794bdb6f39671265aa18aea5c22ee8c4a Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Fri, 12 Apr 2024 09:45:00 +0900 Subject: ksmbd: validate request buffer size in smb2_allocate_rsp_buf() The response buffer should be allocated in smb2_allocate_rsp_buf before validating request. But the fields in payload as well as smb2 header is used in smb2_allocate_rsp_buf(). This patch add simple buffer size validation to avoid potencial out-of-bounds in request buffer. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 5723bbf372d7..ee4b2875a021 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -535,6 +535,10 @@ int smb2_allocate_rsp_buf(struct ksmbd_work *work) if (cmd == SMB2_QUERY_INFO_HE) { struct smb2_query_info_req *req; + if (get_rfc1002_len(work->request_buf) < + offsetof(struct smb2_query_info_req, OutputBufferLength)) + return -EINVAL; + req = smb2_get_msg(work->request_buf); if ((req->InfoType == SMB2_O_INFO_FILE && (req->FileInfoClass == FILE_FULL_EA_INFORMATION || -- cgit v1.2.3-59-g8ed1b From 4973b04d3ea577db80c501c5f14e68ec69fe1794 Mon Sep 17 00:00:00 2001 From: Marios Makassikis Date: Mon, 15 Apr 2024 15:12:48 +0200 Subject: ksmbd: clear RENAME_NOREPLACE before calling vfs_rename File overwrite case is explicitly handled, so it is not necessary to pass RENAME_NOREPLACE to vfs_rename. Clearing the flag fixes rename operations when the share is a ntfs-3g mount. The latter uses an older version of fuse with no support for flags in the ->rename op. Cc: stable@vger.kernel.org Signed-off-by: Marios Makassikis Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/vfs.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'fs') diff --git a/fs/smb/server/vfs.c b/fs/smb/server/vfs.c index 22f0f3db3ac9..51b1b0bed616 100644 --- a/fs/smb/server/vfs.c +++ b/fs/smb/server/vfs.c @@ -754,10 +754,15 @@ retry: goto out4; } + /* + * explicitly handle file overwrite case, for compatibility with + * filesystems that may not support rename flags (e.g: fuse) + */ if ((flags & RENAME_NOREPLACE) && d_is_positive(new_dentry)) { err = -EEXIST; goto out4; } + flags &= ~(RENAME_NOREPLACE); if (old_child == trap) { err = -EINVAL; -- cgit v1.2.3-59-g8ed1b From 0268a7cc7fdc47d90b6c18859de7718d5059f6f1 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Fri, 19 Apr 2024 23:46:34 +0900 Subject: ksmbd: common: use struct_group_attr instead of struct_group for network_open_info 4byte padding cause the connection issue with the applications of MacOS. smb2_close response size increases by 4 bytes by padding, And the smb client of MacOS check it and stop the connection. This patch use struct_group_attr instead of struct_group for network_open_info to use __packed to avoid padding. Fixes: 0015eb6e1238 ("smb: client, common: fix fortify warnings") Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/common/smb2pdu.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/common/smb2pdu.h b/fs/smb/common/smb2pdu.h index 1b594307c9d5..202ff9128156 100644 --- a/fs/smb/common/smb2pdu.h +++ b/fs/smb/common/smb2pdu.h @@ -711,7 +711,7 @@ struct smb2_close_rsp { __le16 StructureSize; /* 60 */ __le16 Flags; __le32 Reserved; - struct_group(network_open_info, + struct_group_attr(network_open_info, __packed, __le64 CreationTime; __le64 LastAccessTime; __le64 LastWriteTime; -- cgit v1.2.3-59-g8ed1b From e9d8c2f95ab8acaf3f4d4a53682a4afa3c263692 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Sat, 20 Apr 2024 09:17:58 +0900 Subject: ksmbd: add continuous availability share parameter If capabilities of the share is not SMB2_SHARE_CAP_CONTINUOUS_AVAILABILITY, ksmbd should not grant a persistent handle to the client. This patch add continuous availability share parameter to control it. Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/ksmbd_netlink.h | 35 ++++++++++++++++++----------------- fs/smb/server/smb2pdu.c | 11 +++++++++-- 2 files changed, 27 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/ksmbd_netlink.h b/fs/smb/server/ksmbd_netlink.h index 686b321c5a8b..f4e55199938d 100644 --- a/fs/smb/server/ksmbd_netlink.h +++ b/fs/smb/server/ksmbd_netlink.h @@ -340,23 +340,24 @@ enum KSMBD_TREE_CONN_STATUS { /* * Share config flags. */ -#define KSMBD_SHARE_FLAG_INVALID (0) -#define KSMBD_SHARE_FLAG_AVAILABLE BIT(0) -#define KSMBD_SHARE_FLAG_BROWSEABLE BIT(1) -#define KSMBD_SHARE_FLAG_WRITEABLE BIT(2) -#define KSMBD_SHARE_FLAG_READONLY BIT(3) -#define KSMBD_SHARE_FLAG_GUEST_OK BIT(4) -#define KSMBD_SHARE_FLAG_GUEST_ONLY BIT(5) -#define KSMBD_SHARE_FLAG_STORE_DOS_ATTRS BIT(6) -#define KSMBD_SHARE_FLAG_OPLOCKS BIT(7) -#define KSMBD_SHARE_FLAG_PIPE BIT(8) -#define KSMBD_SHARE_FLAG_HIDE_DOT_FILES BIT(9) -#define KSMBD_SHARE_FLAG_INHERIT_OWNER BIT(10) -#define KSMBD_SHARE_FLAG_STREAMS BIT(11) -#define KSMBD_SHARE_FLAG_FOLLOW_SYMLINKS BIT(12) -#define KSMBD_SHARE_FLAG_ACL_XATTR BIT(13) -#define KSMBD_SHARE_FLAG_UPDATE BIT(14) -#define KSMBD_SHARE_FLAG_CROSSMNT BIT(15) +#define KSMBD_SHARE_FLAG_INVALID (0) +#define KSMBD_SHARE_FLAG_AVAILABLE BIT(0) +#define KSMBD_SHARE_FLAG_BROWSEABLE BIT(1) +#define KSMBD_SHARE_FLAG_WRITEABLE BIT(2) +#define KSMBD_SHARE_FLAG_READONLY BIT(3) +#define KSMBD_SHARE_FLAG_GUEST_OK BIT(4) +#define KSMBD_SHARE_FLAG_GUEST_ONLY BIT(5) +#define KSMBD_SHARE_FLAG_STORE_DOS_ATTRS BIT(6) +#define KSMBD_SHARE_FLAG_OPLOCKS BIT(7) +#define KSMBD_SHARE_FLAG_PIPE BIT(8) +#define KSMBD_SHARE_FLAG_HIDE_DOT_FILES BIT(9) +#define KSMBD_SHARE_FLAG_INHERIT_OWNER BIT(10) +#define KSMBD_SHARE_FLAG_STREAMS BIT(11) +#define KSMBD_SHARE_FLAG_FOLLOW_SYMLINKS BIT(12) +#define KSMBD_SHARE_FLAG_ACL_XATTR BIT(13) +#define KSMBD_SHARE_FLAG_UPDATE BIT(14) +#define KSMBD_SHARE_FLAG_CROSSMNT BIT(15) +#define KSMBD_SHARE_FLAG_CONTINUOUS_AVAILABILITY BIT(16) /* * Tree connect request flags. diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index ee4b2875a021..355824151c2d 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -1988,7 +1988,12 @@ int smb2_tree_connect(struct ksmbd_work *work) write_unlock(&sess->tree_conns_lock); rsp->StructureSize = cpu_to_le16(16); out_err1: - rsp->Capabilities = 0; + if (server_conf.flags & KSMBD_GLOBAL_FLAG_DURABLE_HANDLE && + test_share_config_flag(share, + KSMBD_SHARE_FLAG_CONTINUOUS_AVAILABILITY)) + rsp->Capabilities = SMB2_SHARE_CAP_CONTINUOUS_AVAILABILITY; + else + rsp->Capabilities = 0; rsp->Reserved = 0; /* default manual caching */ rsp->ShareFlags = SMB2_SHAREFLAG_MANUAL_CACHING; @@ -3502,7 +3507,9 @@ int smb2_open(struct ksmbd_work *work) memcpy(fp->client_guid, conn->ClientGUID, SMB2_CLIENT_GUID_SIZE); if (dh_info.type == DURABLE_REQ_V2 || dh_info.type == DURABLE_REQ) { - if (dh_info.type == DURABLE_REQ_V2 && dh_info.persistent) + if (dh_info.type == DURABLE_REQ_V2 && dh_info.persistent && + test_share_config_flag(work->tcon->share_conf, + KSMBD_SHARE_FLAG_CONTINUOUS_AVAILABILITY)) fp->is_persistent = true; else fp->is_durable = true; -- cgit v1.2.3-59-g8ed1b From fcdbc1d7a4b638e5d5668de461f320386f3002aa Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 17 Apr 2024 15:19:50 -0400 Subject: bcachefs: Check for journal entries overruning end of sb clean section Fix a missing bounds check in superblock validation. Note that we don't yet have repair code for this case - repair code for individual items is generally low priority, since the whole superblock is checksummed, validated prior to write, and we have backups. Reported-by: lei lu Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-clean.c | 8 ++++++++ fs/bcachefs/sb-errors_types.h | 3 ++- 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index 5980ba2563fe..35ca3f138de6 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -29,6 +29,14 @@ int bch2_sb_clean_validate_late(struct bch_fs *c, struct bch_sb_field_clean *cle for (entry = clean->start; entry < (struct jset_entry *) vstruct_end(&clean->field); entry = vstruct_next(entry)) { + if (vstruct_end(entry) > vstruct_end(&clean->field)) { + bch_err(c, "journal entry (u64s %u) overran end of superblock clean section (u64s %u) by %zu", + le16_to_cpu(entry->u64s), le32_to_cpu(clean->field.u64s), + (u64 *) vstruct_end(entry) - (u64 *) vstruct_end(&clean->field)); + bch2_sb_error_count(c, BCH_FSCK_ERR_sb_clean_entry_overrun); + return -BCH_ERR_fsck_repair_unimplemented; + } + ret = bch2_journal_entry_validate(c, NULL, entry, le16_to_cpu(c->disk_sb.sb->version), BCH_SB_BIG_ENDIAN(c->disk_sb.sb), diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 4ca6e7b0d8aa..06c7a644f4a4 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -271,7 +271,8 @@ x(btree_root_unreadable_and_scan_found_nothing, 263) \ x(snapshot_node_missing, 264) \ x(dup_backpointer_to_bad_csum_extent, 265) \ - x(btree_bitmap_not_marked, 266) + x(btree_bitmap_not_marked, 266) \ + x(sb_clean_entry_overrun, 267) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, -- cgit v1.2.3-59-g8ed1b From ec438ac59d7a8bd7e76d3e1201d55071be484626 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 00:31:32 -0400 Subject: bcachefs: Fix missing call to bch2_fs_allocator_background_exit() Signed-off-by: Kent Overstreet --- fs/bcachefs/super.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 8daf80a38d60..88e214c609bb 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -544,6 +544,7 @@ static void __bch2_fs_free(struct bch_fs *c) bch2_find_btree_nodes_exit(&c->found_btree_nodes); bch2_free_pending_node_rewrites(c); + bch2_fs_allocator_background_exit(c); bch2_fs_sb_errors_exit(c); bch2_fs_counters_exit(c); bch2_fs_snapshots_exit(c); -- cgit v1.2.3-59-g8ed1b From 2c03d9560ecebf2865b963e457ba89299c5f1966 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 3 Apr 2024 08:28:28 +1100 Subject: xfs: fix CIL sparse lock context warnings Sparse reports: fs/xfs/xfs_log_cil.c:1127:1: warning: context imbalance in 'xlog_cil_push_work' - different lock contexts for basic block fs/xfs/xfs_log_cil.c:1380:1: warning: context imbalance in 'xlog_cil_push_background' - wrong count at exit fs/xfs/xfs_log_cil.c:1623:9: warning: context imbalance in 'xlog_cil_commit' - unexpected unlock xlog_cil_push_background() has a locking annotations for an rw_sem. Sparse does not track lock contexts for rw_sems, so the annotation generates false warnings. Remove the annotation. xlog_wait_on_iclog() drops the log->l_ic_loglock. The function has a sparse annotation, but the prototype in xfs_log_priv.h does not. Hence the warning from xlog_cil_push_work() which calls xlog_wait_on_iclog(). Add the missing annotation. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Signed-off-by: Chandan Babu R --- fs/xfs/xfs_log_cil.c | 2 +- fs/xfs/xfs_log_priv.h | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index 73f5b7f628f4..f51cbc6405c1 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -1378,7 +1378,7 @@ out_abort_free_ticket: */ static void xlog_cil_push_background( - struct xlog *log) __releases(cil->xc_ctx_lock) + struct xlog *log) { struct xfs_cil *cil = log->l_cilp; int space_used = atomic_read(&cil->xc_ctx->space_used); diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 43881575cd49..a78c4bfaad44 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -620,7 +620,8 @@ xlog_wait( remove_wait_queue(wq, &wait); } -int xlog_wait_on_iclog(struct xlog_in_core *iclog); +int xlog_wait_on_iclog(struct xlog_in_core *iclog) + __releases(iclog->ic_log->l_icloglock); /* * The LSN is valid so long as it is behind the current LSN. If it isn't, this -- cgit v1.2.3-59-g8ed1b From 6e4d9bd110e293513c3c2a3ff2dfa0a0735699e0 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 15:13:20 -0400 Subject: bcachefs: bkey_cached.btree_trans_barrier_seq needs to be a ulong this stores the SRCU sequence number, which we use to check if an SRCU barrier has elapsed; this is a partial fix for the key cache shrinker not actually freeing. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index e0c982a4195c..c69b233c41bb 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -321,9 +321,9 @@ struct bkey_cached { struct btree_bkey_cached_common c; unsigned long flags; + unsigned long btree_trans_barrier_seq; u16 u64s; bool valid; - u32 btree_trans_barrier_seq; struct bkey_cached_key key; struct rhash_head hash; -- cgit v1.2.3-59-g8ed1b From adfe9357c39e251ffe22ceaa1edb4b7662ed76e6 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 15:35:40 -0400 Subject: bcachefs: Tweak btree key cache shrinker so it actually frees Freeing key cache items is a multi stage process; we need to wait for an SRCU grace period to elapse, and we handle this ourselves - partially to avoid callback overhead, but primarily so that when allocating we can first allocate from the freed items waiting for an SRCU grace period. Previously, the shrinker was counting the items on the 'waiting for SRCU grace period' lists as items being scanned, but this meant that too many items waiting for an SRCU grace period could prevent it from doing any work at all. After this, we're seeing that items skipped due to the accessed bit are the main cause of the shrinker not making any progress, and we actually want the key cache shrinker to run quite aggressively because reclaimed items will still generally be found (more compactly) in the btree node cache - so we also tweak the shrinker to not count those against nr_to_scan. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 19 ++++--------------- 1 file changed, 4 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 88a3582a3275..e8c1c530cd95 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -842,8 +842,6 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, * Newest freed entries are at the end of the list - once we hit one * that's too new to be freed, we can bail out: */ - scanned += bc->nr_freed_nonpcpu; - list_for_each_entry_safe(ck, t, &bc->freed_nonpcpu, list) { if (!poll_state_synchronize_srcu(&c->btree_trans_barrier, ck->btree_trans_barrier_seq)) @@ -857,11 +855,6 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, bc->nr_freed_nonpcpu--; } - if (scanned >= nr) - goto out; - - scanned += bc->nr_freed_pcpu; - list_for_each_entry_safe(ck, t, &bc->freed_pcpu, list) { if (!poll_state_synchronize_srcu(&c->btree_trans_barrier, ck->btree_trans_barrier_seq)) @@ -875,9 +868,6 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, bc->nr_freed_pcpu--; } - if (scanned >= nr) - goto out; - rcu_read_lock(); tbl = rht_dereference_rcu(bc->table.tbl, &bc->table); if (bc->shrink_iter >= tbl->size) @@ -893,12 +883,12 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, next = rht_dereference_bucket_rcu(pos->next, tbl, bc->shrink_iter); ck = container_of(pos, struct bkey_cached, hash); - if (test_bit(BKEY_CACHED_DIRTY, &ck->flags)) + if (test_bit(BKEY_CACHED_DIRTY, &ck->flags)) { goto next; - - if (test_bit(BKEY_CACHED_ACCESSED, &ck->flags)) + } else if (test_bit(BKEY_CACHED_ACCESSED, &ck->flags)) { clear_bit(BKEY_CACHED_ACCESSED, &ck->flags); - else if (bkey_cached_lock_for_evict(ck)) { + goto next; + } else if (bkey_cached_lock_for_evict(ck)) { bkey_cached_evict(bc, ck); bkey_cached_free(bc, ck); } @@ -916,7 +906,6 @@ next: } while (scanned < nr && bc->shrink_iter != start); rcu_read_unlock(); -out: memalloc_nofs_restore(flags); srcu_read_unlock(&c->btree_trans_barrier, srcu_idx); mutex_unlock(&bc->lock); -- cgit v1.2.3-59-g8ed1b From 85ab365f7cdf2b2a713823a93e7e5e94f0529627 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 19 Apr 2024 21:54:32 -0400 Subject: bcachefs: Fix deadlock in journal write path bch2_journal_write() was incorrectly waiting on earlier journal writes synchronously; this usually worked because most of the time we'd be running in the context of a thread that did a journal_buf_put(), but sometimes we'd be running out of the same workqueue that completes those prior journal writes. Additionally, this makes sure to punt to a workqueue before submitting preflushes - we really don't want to be calling submit_bio() in the main transaction commit path. Signed-off-by: Kent Overstreet --- fs/bcachefs/journal_io.c | 60 +++++++++++++++++++++++++++++++++--------------- 1 file changed, 42 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 9aa28b52ab92..eb1f9d6f5a19 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1723,7 +1723,7 @@ static void journal_write_endio(struct bio *bio) percpu_ref_put(&ca->io_ref); } -static CLOSURE_CALLBACK(do_journal_write) +static CLOSURE_CALLBACK(journal_write_submit) { closure_type(w, struct journal_buf, io); struct journal *j = container_of(w, struct journal, buf[w->idx]); @@ -1768,6 +1768,44 @@ static CLOSURE_CALLBACK(do_journal_write) continue_at(cl, journal_write_done, j->wq); } +static CLOSURE_CALLBACK(journal_write_preflush) +{ + closure_type(w, struct journal_buf, io); + struct journal *j = container_of(w, struct journal, buf[w->idx]); + struct bch_fs *c = container_of(j, struct bch_fs, journal); + + if (j->seq_ondisk + 1 != le64_to_cpu(w->data->seq)) { + spin_lock(&j->lock); + closure_wait(&j->async_wait, cl); + spin_unlock(&j->lock); + + continue_at(cl, journal_write_preflush, j->wq); + return; + } + + if (w->separate_flush) { + for_each_rw_member(c, ca) { + percpu_ref_get(&ca->io_ref); + + struct journal_device *ja = &ca->journal; + struct bio *bio = &ja->bio[w->idx]->bio; + bio_reset(bio, ca->disk_sb.bdev, + REQ_OP_WRITE|REQ_SYNC|REQ_META|REQ_PREFLUSH); + bio->bi_end_io = journal_write_endio; + bio->bi_private = ca; + closure_bio_submit(bio, cl); + } + + continue_at(cl, journal_write_submit, j->wq); + } else { + /* + * no need to punt to another work item if we're not waiting on + * preflushes + */ + journal_write_submit(&cl->work); + } +} + static int bch2_journal_write_prep(struct journal *j, struct journal_buf *w) { struct bch_fs *c = container_of(j, struct bch_fs, journal); @@ -2033,23 +2071,9 @@ CLOSURE_CALLBACK(bch2_journal_write) goto err; if (!JSET_NO_FLUSH(w->data)) - closure_wait_event(&j->async_wait, j->seq_ondisk + 1 == le64_to_cpu(w->data->seq)); - - if (!JSET_NO_FLUSH(w->data) && w->separate_flush) { - for_each_rw_member(c, ca) { - percpu_ref_get(&ca->io_ref); - - struct journal_device *ja = &ca->journal; - struct bio *bio = &ja->bio[w->idx]->bio; - bio_reset(bio, ca->disk_sb.bdev, - REQ_OP_WRITE|REQ_SYNC|REQ_META|REQ_PREFLUSH); - bio->bi_end_io = journal_write_endio; - bio->bi_private = ca; - closure_bio_submit(bio, cl); - } - } - - continue_at(cl, do_journal_write, j->wq); + continue_at(cl, journal_write_preflush, j->wq); + else + continue_at(cl, journal_write_submit, j->wq); return; no_io: continue_at(cl, journal_write_done, j->wq); -- cgit v1.2.3-59-g8ed1b From 0e42f381193d7f9b47922f1c4308e7729a45ba13 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 22:26:47 -0400 Subject: bcachefs: Fix inode early destruction path discard_new_inode() is the wrong interface to use when we need to free an inode that was never inserted into the inode hash table; we can bypass the whole iput() -> evict() path and replace it with __destroy_inode(); kmem_cache_free() - this fixes a WARN_ON() about I_NEW. Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index b5ea9fa1259d..fce690007edf 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -188,7 +188,8 @@ static struct bch_inode_info *bch2_inode_insert(struct bch_fs *c, struct bch_ino BUG_ON(!old); if (unlikely(old != inode)) { - discard_new_inode(&inode->v); + __destroy_inode(&inode->v); + kmem_cache_free(bch2_inode_cache, inode); inode = old; } else { mutex_lock(&c->vfs_inodes_lock); @@ -225,8 +226,10 @@ static struct bch_inode_info *bch2_new_inode(struct btree_trans *trans) if (unlikely(!inode)) { int ret = drop_locks_do(trans, (inode = to_bch_ei(new_inode(c->vfs_sb))) ? 0 : -ENOMEM); - if (ret && inode) - discard_new_inode(&inode->v); + if (ret && inode) { + __destroy_inode(&inode->v); + kmem_cache_free(bch2_inode_cache, inode); + } if (ret) return ERR_PTR(ret); } -- cgit v1.2.3-59-g8ed1b From e858beeddfa3a400844c0e22d2118b3b52f1ea5e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 21 Apr 2024 23:32:18 -0400 Subject: bcachefs: If we run merges at a lower watermark, they must be nonblocking Fix another deadlock related to the merge path; previously, we switched to always running merges at a lower watermark (because they are noncritical); but when we run at a lower watermark we also need to run nonblocking or we've introduced a new deadlock. Signed-off-by: Kent Overstreet Reported-and-tested-by: s@m-h.ug --- fs/bcachefs/btree_update_interior.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 6030c396754f..b4efd8cc4d1a 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1960,7 +1960,11 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, if ((flags & BCH_WATERMARK_MASK) == BCH_WATERMARK_interior_updates) return 0; - flags &= ~BCH_WATERMARK_MASK; + if ((flags & BCH_WATERMARK_MASK) <= BCH_WATERMARK_reclaim) { + flags &= ~BCH_WATERMARK_MASK; + flags |= BCH_WATERMARK_btree; + flags |= BCH_TRANS_COMMIT_journal_reclaim; + } b = trans->paths[path].l[level].b; -- cgit v1.2.3-59-g8ed1b From 27a7a9d903a0473cb763afa662c38debf6f65bfa Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 3 Apr 2024 08:28:30 +1100 Subject: xfs: silence sparse warning when checking version number Scrub checks the superblock version number against the known good feature bits that can be set in the version mask. It calculates the version mask to compare like so: vernum_mask = cpu_to_be16(~XFS_SB_VERSION_OKBITS | XFS_SB_VERSION_NUMBITS | XFS_SB_VERSION_ALIGNBIT | XFS_SB_VERSION_DALIGNBIT | XFS_SB_VERSION_SHAREDBIT | XFS_SB_VERSION_LOGV2BIT | XFS_SB_VERSION_SECTORBIT | XFS_SB_VERSION_EXTFLGBIT | XFS_SB_VERSION_DIRV2BIT); This generates a sparse warning: fs/xfs/scrub/agheader.c:168:23: warning: cast truncates bits from constant value (ffff3f8f becomes 3f8f) This is because '~XFS_SB_VERSION_OKBITS' is considered a 32 bit constant, even though it's value is always under 16 bits. This is a kinda silly thing to do, because: /* * Supported feature bit list is just all bits in the versionnum field because * we've used them all up and understand them all. Except, of course, for the * shared superblock bit, which nobody knows what it does and so is unsupported. */ #define XFS_SB_VERSION_OKBITS \ ((XFS_SB_VERSION_NUMBITS | XFS_SB_VERSION_ALLFBITS) & \ ~XFS_SB_VERSION_SHAREDBIT) #define XFS_SB_VERSION_NUMBITS 0x000f #define XFS_SB_VERSION_ALLFBITS 0xfff0 #define XFS_SB_VERSION_SHAREDBIT 0x0200 XFS_SB_VERSION_OKBITS has a value of 0xfdff, and so ~XFS_SB_VERSION_OKBITS == XFS_SB_VERSION_SHAREDBIT. The calculated mask already sets XFS_SB_VERSION_SHAREDBIT, so starting with ~XFS_SB_VERSION_OKBITS is completely redundant.... Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/scrub/agheader.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/agheader.c b/fs/xfs/scrub/agheader.c index 1528f14bd925..f8e5b67128d2 100644 --- a/fs/xfs/scrub/agheader.c +++ b/fs/xfs/scrub/agheader.c @@ -166,8 +166,7 @@ xchk_superblock( xchk_block_set_corrupt(sc, bp); /* Check sb_versionnum bits that are set at mkfs time. */ - vernum_mask = cpu_to_be16(~XFS_SB_VERSION_OKBITS | - XFS_SB_VERSION_NUMBITS | + vernum_mask = cpu_to_be16(XFS_SB_VERSION_NUMBITS | XFS_SB_VERSION_ALIGNBIT | XFS_SB_VERSION_DALIGNBIT | XFS_SB_VERSION_SHAREDBIT | -- cgit v1.2.3-59-g8ed1b From 76f011f7e659305daf8beba0986fd4a03b3e56f6 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 3 Apr 2024 08:28:32 +1100 Subject: xfs: fix sparse warnings about unused interval tree functions Sparse throws warnings about the interval tree functions that are defined and then not used in the scrub bitmap code: fs/xfs/scrub/bitmap.c:57:1: warning: unused function 'xbitmap64_tree_iter_next' [-Wunused-function] INTERVAL_TREE_DEFINE(struct xbitmap64_node, bn_rbnode, uint64_t, ^ ./include/linux/interval_tree_generic.h:151:33: note: expanded from macro 'INTERVAL_TREE_DEFINE' ITSTATIC ITSTRUCT * \ ^ :3:1: note: expanded from here xbitmap64_tree_iter_next ^ fs/xfs/scrub/bitmap.c:331:1: warning: unused function 'xbitmap32_tree_iter_next' [-Wunused-function] INTERVAL_TREE_DEFINE(struct xbitmap32_node, bn_rbnode, uint32_t, ^ ./include/linux/interval_tree_generic.h:151:33: note: expanded from macro 'INTERVAL_TREE_DEFINE' ITSTATIC ITSTRUCT * \ ^ :59:1: note: expanded from here xbitmap32_tree_iter_next Fix these by marking the functions created by the interval tree creation macro as __maybe_unused to suppress this warning. Signed-off-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/scrub/bitmap.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/bitmap.c b/fs/xfs/scrub/bitmap.c index 0cb8d43912a8..7ba35a7a7920 100644 --- a/fs/xfs/scrub/bitmap.c +++ b/fs/xfs/scrub/bitmap.c @@ -40,22 +40,23 @@ struct xbitmap64_node { * These functions are defined by the INTERVAL_TREE_DEFINE macro, but we'll * forward-declare them anyway for clarity. */ -static inline void +static inline __maybe_unused void xbitmap64_tree_insert(struct xbitmap64_node *node, struct rb_root_cached *root); -static inline void +static inline __maybe_unused void xbitmap64_tree_remove(struct xbitmap64_node *node, struct rb_root_cached *root); -static inline struct xbitmap64_node * +static inline __maybe_unused struct xbitmap64_node * xbitmap64_tree_iter_first(struct rb_root_cached *root, uint64_t start, uint64_t last); -static inline struct xbitmap64_node * +static inline __maybe_unused struct xbitmap64_node * xbitmap64_tree_iter_next(struct xbitmap64_node *node, uint64_t start, uint64_t last); INTERVAL_TREE_DEFINE(struct xbitmap64_node, bn_rbnode, uint64_t, - __bn_subtree_last, START, LAST, static inline, xbitmap64_tree) + __bn_subtree_last, START, LAST, static inline __maybe_unused, + xbitmap64_tree) /* Iterate each interval of a bitmap. Do not change the bitmap. */ #define for_each_xbitmap64_extent(bn, bitmap) \ @@ -314,22 +315,23 @@ struct xbitmap32_node { * These functions are defined by the INTERVAL_TREE_DEFINE macro, but we'll * forward-declare them anyway for clarity. */ -static inline void +static inline __maybe_unused void xbitmap32_tree_insert(struct xbitmap32_node *node, struct rb_root_cached *root); -static inline void +static inline __maybe_unused void xbitmap32_tree_remove(struct xbitmap32_node *node, struct rb_root_cached *root); -static inline struct xbitmap32_node * +static inline __maybe_unused struct xbitmap32_node * xbitmap32_tree_iter_first(struct rb_root_cached *root, uint32_t start, uint32_t last); -static inline struct xbitmap32_node * +static inline __maybe_unused struct xbitmap32_node * xbitmap32_tree_iter_next(struct xbitmap32_node *node, uint32_t start, uint32_t last); INTERVAL_TREE_DEFINE(struct xbitmap32_node, bn_rbnode, uint32_t, - __bn_subtree_last, START, LAST, static inline, xbitmap32_tree) + __bn_subtree_last, START, LAST, static inline __maybe_unused, + xbitmap32_tree) /* Iterate each interval of a bitmap. Do not change the bitmap. */ #define for_each_xbitmap32_extent(bn, bitmap) \ -- cgit v1.2.3-59-g8ed1b From 7d7c82a04d3d93c8b6f8ed71aa5bc39cdc96377c Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Sat, 13 Apr 2024 00:59:27 +0200 Subject: xfs: Fix typo in comment s/somethign/something/ Signed-off-by: Thorsten Blum Reviewed-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig Signed-off-by: Chandan Babu R --- fs/xfs/xfs_log_priv.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index a78c4bfaad44..40e22ec0fbe6 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -681,7 +681,7 @@ xlog_valid_lsn( * flags to control the kmalloc() behaviour within kvmalloc(). Hence kmalloc() * will do direct reclaim and compaction in the slow path, both of which are * horrendously expensive. We just want kmalloc to fail fast and fall back to - * vmalloc if it can't get somethign straight away from the free lists or + * vmalloc if it can't get something straight away from the free lists or * buddy allocator. Hence we have to open code kvmalloc outselves here. * * This assumes that the caller uses memalloc_nofs_save task context here, so -- cgit v1.2.3-59-g8ed1b From d983ff63af6068a6773283660222fff10f66e184 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 7 Mar 2024 11:39:18 +0300 Subject: xfs: small cleanup in xrep_update_qflags() The "mp" pointer is the same as "sc->mp" so this change doesn't affect runtime at all. However, it's nicer to use same name for both the lock and the unlock. Signed-off-by: Dan Carpenter Reviewed-by: Christoph Hellwig Signed-off-by: Chandan Babu R --- fs/xfs/scrub/repair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index b6aff89679d5..92f81e374bc0 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -729,7 +729,7 @@ xrep_update_qflags( xfs_trans_log_buf(sc->tp, bp, 0, sizeof(struct xfs_dsb) - 1); no_update: - mutex_unlock(&sc->mp->m_quotainfo->qi_quotaofflock); + mutex_unlock(&mp->m_quotainfo->qi_quotaofflock); } /* Force a quotacheck the next time we mount. */ -- cgit v1.2.3-59-g8ed1b From 05150d46a33fdb567d9849b831aa139e1693e39c Mon Sep 17 00:00:00 2001 From: Jiapeng Chong Date: Mon, 4 Mar 2024 10:24:08 +0800 Subject: xfs: Remove unused function is_rt_data_fork The function are defined in the rmap_repair.c file, but not called elsewhere, so delete the unused function. fs/xfs/scrub/rmap_repair.c:436:1: warning: unused function 'is_rt_data_fork'. Reported-by: Abaci Robot Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=8425 Signed-off-by: Jiapeng Chong Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/scrub/rmap_repair.c | 8 -------- 1 file changed, 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/rmap_repair.c b/fs/xfs/scrub/rmap_repair.c index 25acd69614c2..e8080eba37d2 100644 --- a/fs/xfs/scrub/rmap_repair.c +++ b/fs/xfs/scrub/rmap_repair.c @@ -432,14 +432,6 @@ out: return error; } -static inline bool -is_rt_data_fork( - struct xfs_inode *ip, - int whichfork) -{ - return XFS_IS_REALTIME_INODE(ip) && whichfork == XFS_DATA_FORK; -} - /* * Iterate the block mapping btree to collect rmap records for anything in this * fork that matches the AG. Sets @mappings_done to true if we've scanned the -- cgit v1.2.3-59-g8ed1b From e78a3ce28331583cc0b57b0602528c4d7628fc19 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 5 Apr 2024 08:07:08 +0200 Subject: xfs: move more logic into xfs_extent_busy_clear_one Move the handling of discarded entries into xfs_extent_busy_clear_one to reuse the length check and tidy up the logic in the caller. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_extent_busy.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c index 56cfa1498571..6fbffa46e5e9 100644 --- a/fs/xfs/xfs_extent_busy.c +++ b/fs/xfs/xfs_extent_busy.c @@ -518,20 +518,26 @@ fail: goto out; } -STATIC void +static bool xfs_extent_busy_clear_one( - struct xfs_mount *mp, struct xfs_perag *pag, - struct xfs_extent_busy *busyp) + struct xfs_extent_busy *busyp, + bool do_discard) { if (busyp->length) { - trace_xfs_extent_busy_clear(mp, busyp->agno, busyp->bno, - busyp->length); + if (do_discard && + !(busyp->flags & XFS_EXTENT_BUSY_SKIP_DISCARD)) { + busyp->flags = XFS_EXTENT_BUSY_DISCARDED; + return false; + } + trace_xfs_extent_busy_clear(pag->pag_mount, busyp->agno, + busyp->bno, busyp->length); rb_erase(&busyp->rb_node, &pag->pagb_tree); } list_del_init(&busyp->list); kfree(busyp); + return true; } static void @@ -575,13 +581,8 @@ xfs_extent_busy_clear( wakeup = false; } - if (do_discard && busyp->length && - !(busyp->flags & XFS_EXTENT_BUSY_SKIP_DISCARD)) { - busyp->flags = XFS_EXTENT_BUSY_DISCARDED; - } else { - xfs_extent_busy_clear_one(mp, pag, busyp); + if (xfs_extent_busy_clear_one(pag, busyp, do_discard)) wakeup = true; - } } if (pag) -- cgit v1.2.3-59-g8ed1b From c37d6ed87462751bc979f133a570716089fe40de Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 5 Apr 2024 08:07:09 +0200 Subject: xfs: unwind xfs_extent_busy_clear The current structure of xfs_extent_busy_clear that locks the first busy extent in each AG and unlocks when switching to a new AG makes sparse unhappy as the lock critical section tracking can't cope with taking the lock conditionally and inside a loop. Rewrite xfs_extent_busy_clear so that it has an outer loop only advancing when moving to a new AG, and an inner loop that consumes busy extents for the given AG to make life easier for sparse and to also make this logic more obvious to humans. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_extent_busy.c | 59 ++++++++++++++++++++---------------------------- 1 file changed, 25 insertions(+), 34 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c index 6fbffa46e5e9..a73e7c73b664 100644 --- a/fs/xfs/xfs_extent_busy.c +++ b/fs/xfs/xfs_extent_busy.c @@ -540,21 +540,6 @@ xfs_extent_busy_clear_one( return true; } -static void -xfs_extent_busy_put_pag( - struct xfs_perag *pag, - bool wakeup) - __releases(pag->pagb_lock) -{ - if (wakeup) { - pag->pagb_gen++; - wake_up_all(&pag->pagb_wait); - } - - spin_unlock(&pag->pagb_lock); - xfs_perag_put(pag); -} - /* * Remove all extents on the passed in list from the busy extents tree. * If do_discard is set skip extents that need to be discarded, and mark @@ -566,27 +551,33 @@ xfs_extent_busy_clear( struct list_head *list, bool do_discard) { - struct xfs_extent_busy *busyp, *n; - struct xfs_perag *pag = NULL; - xfs_agnumber_t agno = NULLAGNUMBER; - bool wakeup = false; - - list_for_each_entry_safe(busyp, n, list, list) { - if (busyp->agno != agno) { - if (pag) - xfs_extent_busy_put_pag(pag, wakeup); - agno = busyp->agno; - pag = xfs_perag_get(mp, agno); - spin_lock(&pag->pagb_lock); - wakeup = false; - } + struct xfs_extent_busy *busyp, *next; - if (xfs_extent_busy_clear_one(pag, busyp, do_discard)) - wakeup = true; - } + busyp = list_first_entry_or_null(list, typeof(*busyp), list); + if (!busyp) + return; + + do { + bool wakeup = false; + struct xfs_perag *pag; - if (pag) - xfs_extent_busy_put_pag(pag, wakeup); + pag = xfs_perag_get(mp, busyp->agno); + spin_lock(&pag->pagb_lock); + do { + next = list_next_entry(busyp, list); + if (xfs_extent_busy_clear_one(pag, busyp, do_discard)) + wakeup = true; + busyp = next; + } while (!list_entry_is_head(busyp, list, list) && + busyp->agno == pag->pag_agno); + + if (wakeup) { + pag->pagb_gen++; + wake_up_all(&pag->pagb_wait); + } + spin_unlock(&pag->pagb_lock); + xfs_perag_put(pag); + } while (!list_entry_is_head(busyp, list, list)); } /* -- cgit v1.2.3-59-g8ed1b From c0ac6cb251bdc3bcc04b2070296788e0b93652d7 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 5 Apr 2024 08:07:10 +0200 Subject: xfs: remove the unused xfs_extent_busy_enomem trace event Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_trace.h | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index a80c3063a13f..0f844396754d 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -1659,7 +1659,6 @@ DEFINE_EVENT(xfs_extent_busy_class, name, \ xfs_agblock_t agbno, xfs_extlen_t len), \ TP_ARGS(mp, agno, agbno, len)) DEFINE_BUSY_EVENT(xfs_extent_busy); -DEFINE_BUSY_EVENT(xfs_extent_busy_enomem); DEFINE_BUSY_EVENT(xfs_extent_busy_force); DEFINE_BUSY_EVENT(xfs_extent_busy_reuse); DEFINE_BUSY_EVENT(xfs_extent_busy_clear); -- cgit v1.2.3-59-g8ed1b From 4887e53163825189c70a0db4f13b42eae8798625 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 08:13:16 +0200 Subject: xfs: compile out v4 support if disabled Add a few strategic IS_ENABLED statements to let the compiler eliminate unused code when CONFIG_XFS_SUPPORT_V4 is disabled. This saves multiple kilobytes of .text in my .config: $ size xfs.o.* text data bss dec hex filename 1363633 294836 592 1659061 1950b5 xfs.o.new 1371453 294868 592 1666913 196f61 xfs.o.old Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_mount.h | 40 +++++++++++++++++++++++++++++++--------- fs/xfs/xfs_super.c | 22 +++++++++++++--------- 2 files changed, 44 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index b022e5120dc4..283a5a26db31 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -332,19 +332,10 @@ static inline void xfs_add_ ## name (struct xfs_mount *mp) \ __XFS_ADD_FEAT(attr, ATTR) __XFS_HAS_FEAT(nlink, NLINK) __XFS_ADD_FEAT(quota, QUOTA) -__XFS_HAS_FEAT(align, ALIGN) __XFS_HAS_FEAT(dalign, DALIGN) -__XFS_HAS_FEAT(logv2, LOGV2) __XFS_HAS_FEAT(sector, SECTOR) -__XFS_HAS_FEAT(extflg, EXTFLG) __XFS_HAS_FEAT(asciici, ASCIICI) -__XFS_HAS_FEAT(lazysbcount, LAZYSBCOUNT) -__XFS_ADD_FEAT(attr2, ATTR2) __XFS_HAS_FEAT(parent, PARENT) -__XFS_ADD_FEAT(projid32, PROJID32) -__XFS_HAS_FEAT(crc, CRC) -__XFS_HAS_FEAT(v3inodes, V3INODES) -__XFS_HAS_FEAT(pquotino, PQUOTINO) __XFS_HAS_FEAT(ftype, FTYPE) __XFS_HAS_FEAT(finobt, FINOBT) __XFS_HAS_FEAT(rmapbt, RMAPBT) @@ -358,6 +349,37 @@ __XFS_HAS_FEAT(needsrepair, NEEDSREPAIR) __XFS_HAS_FEAT(large_extent_counts, NREXT64) __XFS_HAS_FEAT(exchange_range, EXCHANGE_RANGE) +/* + * Some features are always on for v5 file systems, allow the compiler to + * eliminiate dead code when building without v4 support. + */ +#define __XFS_HAS_V4_FEAT(name, NAME) \ +static inline bool xfs_has_ ## name (struct xfs_mount *mp) \ +{ \ + return !IS_ENABLED(CONFIG_XFS_SUPPORT_V4) || \ + (mp->m_features & XFS_FEAT_ ## NAME); \ +} + +#define __XFS_ADD_V4_FEAT(name, NAME) \ + __XFS_HAS_V4_FEAT(name, NAME); \ +static inline void xfs_add_ ## name (struct xfs_mount *mp) \ +{ \ + if (IS_ENABLED(CONFIG_XFS_SUPPORT_V4)) { \ + mp->m_features |= XFS_FEAT_ ## NAME; \ + xfs_sb_version_add ## name(&mp->m_sb); \ + } \ +} + +__XFS_HAS_V4_FEAT(align, ALIGN) +__XFS_HAS_V4_FEAT(logv2, LOGV2) +__XFS_HAS_V4_FEAT(extflg, EXTFLG) +__XFS_HAS_V4_FEAT(lazysbcount, LAZYSBCOUNT) +__XFS_ADD_V4_FEAT(attr2, ATTR2) +__XFS_ADD_V4_FEAT(projid32, PROJID32) +__XFS_HAS_V4_FEAT(v3inodes, V3INODES) +__XFS_HAS_V4_FEAT(crc, CRC) +__XFS_HAS_V4_FEAT(pquotino, PQUOTINO) + /* * Mount features * diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 5c9ba974252d..02c1a4aeaa4c 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1580,17 +1580,21 @@ xfs_fs_fill_super( if (error) goto out_free_sb; - /* V4 support is undergoing deprecation. */ - if (!xfs_has_crc(mp)) { -#ifdef CONFIG_XFS_SUPPORT_V4 + /* + * V4 support is undergoing deprecation. + * + * Note: this has to use an open coded m_features check as xfs_has_crc + * always returns false for !CONFIG_XFS_SUPPORT_V4. + */ + if (!(mp->m_features & XFS_FEAT_CRC)) { + if (!IS_ENABLED(CONFIG_XFS_SUPPORT_V4)) { + xfs_warn(mp, + "Deprecated V4 format (crc=0) not supported by kernel."); + error = -EINVAL; + goto out_free_sb; + } xfs_warn_once(mp, "Deprecated V4 format (crc=0) will not be supported after September 2030."); -#else - xfs_warn(mp, - "Deprecated V4 format (crc=0) not supported by kernel."); - error = -EINVAL; - goto out_free_sb; -#endif } /* ASCII case insensitivity is undergoing deprecation. */ -- cgit v1.2.3-59-g8ed1b From 330c4f94b0d73e440de3f738a625e38defe1bc15 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:07 +0200 Subject: xfs: make XFS_TRANS_LOWMODE match the other XFS_TRANS_ definitions Commit bb7b1c9c5dd3 ("xfs: tag transactions that contain intent done items") switched the XFS_TRANS_ definitions to be bit based, and using comments above the definitions. As XFS_TRANS_LOWMODE was last and has a big fat comment it was missed. Switch it to the same style. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_shared.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h index dfd61fa8332e..f35640ad3e7f 100644 --- a/fs/xfs/libxfs/xfs_shared.h +++ b/fs/xfs/libxfs/xfs_shared.h @@ -124,7 +124,6 @@ void xfs_log_get_max_trans_res(struct xfs_mount *mp, #define XFS_TRANS_RES_FDBLKS (1u << 6) /* Transaction contains an intent done log item */ #define XFS_TRANS_HAS_INTENT_DONE (1u << 7) - /* * LOWMODE is used by the allocator to activate the lowspace algorithm - when * free space is running low the extent allocator may choose to allocate an @@ -136,7 +135,7 @@ void xfs_log_get_max_trans_res(struct xfs_mount *mp, * for free space from AG 0. If the correct transaction reservations have been * made then this algorithm will eventually find all the space it needs. */ -#define XFS_TRANS_LOWMODE 0x100 /* allocate in low space mode */ +#define XFS_TRANS_LOWMODE (1u << 8) /* * Field values for xfs_trans_mod_sb. -- cgit v1.2.3-59-g8ed1b From b7e23c0e2e3b1c520a3370f058870b914071a470 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:08 +0200 Subject: xfs: refactor realtime inode locking Create helper functions to deal with locking realtime metadata inodes. This enables us to maintain correct locking order once we start adding the realtime rmap and refcount btree inodes. Signed-off-by: "Darrick J. Wong" Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 7 ++---- fs/xfs/libxfs/xfs_rtbitmap.c | 57 ++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_rtbitmap.h | 17 +++++++++++++ fs/xfs/scrub/common.c | 1 + fs/xfs/scrub/fscounters.c | 4 ++-- fs/xfs/xfs_fsmap.c | 4 ++-- fs/xfs/xfs_rtalloc.c | 20 +++++----------- 7 files changed, 87 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 59b8b9dc29cc..3e56173a8fe4 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -5418,12 +5418,9 @@ __xfs_bunmapi( if (isrt) { /* - * Synchronize by locking the bitmap inode. + * Synchronize by locking the realtime bitmap. */ - xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL|XFS_ILOCK_RTBITMAP); - xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL); - xfs_ilock(mp->m_rsumip, XFS_ILOCK_EXCL|XFS_ILOCK_RTSUM); - xfs_trans_ijoin(tp, mp->m_rsumip, XFS_ILOCK_EXCL); + xfs_rtbitmap_lock(tp, mp); } extno = 0; diff --git a/fs/xfs/libxfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c index f246d6dbf4ec..386b672c5058 100644 --- a/fs/xfs/libxfs/xfs_rtbitmap.c +++ b/fs/xfs/libxfs/xfs_rtbitmap.c @@ -1168,3 +1168,60 @@ xfs_rtsummary_wordcount( blocks = xfs_rtsummary_blockcount(mp, rsumlevels, rbmblocks); return XFS_FSB_TO_B(mp, blocks) >> XFS_WORDLOG; } + +/* + * Lock both realtime free space metadata inodes for a freespace update. If a + * transaction is given, the inodes will be joined to the transaction and the + * ILOCKs will be released on transaction commit. + */ +void +xfs_rtbitmap_lock( + struct xfs_trans *tp, + struct xfs_mount *mp) +{ + xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL | XFS_ILOCK_RTBITMAP); + if (tp) + xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL); + + xfs_ilock(mp->m_rsumip, XFS_ILOCK_EXCL | XFS_ILOCK_RTSUM); + if (tp) + xfs_trans_ijoin(tp, mp->m_rsumip, XFS_ILOCK_EXCL); +} + +/* Unlock both realtime free space metadata inodes after a freespace update. */ +void +xfs_rtbitmap_unlock( + struct xfs_mount *mp) +{ + xfs_iunlock(mp->m_rsumip, XFS_ILOCK_EXCL | XFS_ILOCK_RTSUM); + xfs_iunlock(mp->m_rbmip, XFS_ILOCK_EXCL | XFS_ILOCK_RTBITMAP); +} + +/* + * Lock the realtime free space metadata inodes for a freespace scan. Callers + * must walk metadata blocks in order of increasing file offset. + */ +void +xfs_rtbitmap_lock_shared( + struct xfs_mount *mp, + unsigned int rbmlock_flags) +{ + if (rbmlock_flags & XFS_RBMLOCK_BITMAP) + xfs_ilock(mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); + + if (rbmlock_flags & XFS_RBMLOCK_SUMMARY) + xfs_ilock(mp->m_rsumip, XFS_ILOCK_SHARED | XFS_ILOCK_RTSUM); +} + +/* Unlock the realtime free space metadata inodes after a freespace scan. */ +void +xfs_rtbitmap_unlock_shared( + struct xfs_mount *mp, + unsigned int rbmlock_flags) +{ + if (rbmlock_flags & XFS_RBMLOCK_SUMMARY) + xfs_iunlock(mp->m_rsumip, XFS_ILOCK_SHARED | XFS_ILOCK_RTSUM); + + if (rbmlock_flags & XFS_RBMLOCK_BITMAP) + xfs_iunlock(mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); +} diff --git a/fs/xfs/libxfs/xfs_rtbitmap.h b/fs/xfs/libxfs/xfs_rtbitmap.h index 152a66750af5..6186585f2c37 100644 --- a/fs/xfs/libxfs/xfs_rtbitmap.h +++ b/fs/xfs/libxfs/xfs_rtbitmap.h @@ -360,6 +360,19 @@ xfs_filblks_t xfs_rtsummary_blockcount(struct xfs_mount *mp, unsigned int rsumlevels, xfs_extlen_t rbmblocks); unsigned long long xfs_rtsummary_wordcount(struct xfs_mount *mp, unsigned int rsumlevels, xfs_extlen_t rbmblocks); + +void xfs_rtbitmap_lock(struct xfs_trans *tp, struct xfs_mount *mp); +void xfs_rtbitmap_unlock(struct xfs_mount *mp); + +/* Lock the rt bitmap inode in shared mode */ +#define XFS_RBMLOCK_BITMAP (1U << 0) +/* Lock the rt summary inode in shared mode */ +#define XFS_RBMLOCK_SUMMARY (1U << 1) + +void xfs_rtbitmap_lock_shared(struct xfs_mount *mp, + unsigned int rbmlock_flags); +void xfs_rtbitmap_unlock_shared(struct xfs_mount *mp, + unsigned int rbmlock_flags); #else /* CONFIG_XFS_RT */ # define xfs_rtfree_extent(t,b,l) (-ENOSYS) # define xfs_rtfree_blocks(t,rb,rl) (-ENOSYS) @@ -378,6 +391,10 @@ xfs_rtbitmap_blockcount(struct xfs_mount *mp, xfs_rtbxlen_t rtextents) # define xfs_rtbitmap_wordcount(mp, r) (0) # define xfs_rtsummary_blockcount(mp, l, b) (0) # define xfs_rtsummary_wordcount(mp, l, b) (0) +# define xfs_rtbitmap_lock(tp, mp) do { } while (0) +# define xfs_rtbitmap_unlock(mp) do { } while (0) +# define xfs_rtbitmap_lock_shared(mp, lf) do { } while (0) +# define xfs_rtbitmap_unlock_shared(mp, lf) do { } while (0) #endif /* CONFIG_XFS_RT */ #endif /* __XFS_RTBITMAP_H__ */ diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index 48302532d10d..a7d3a2279662 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -32,6 +32,7 @@ #include "xfs_error.h" #include "xfs_quota.h" #include "xfs_exchmaps.h" +#include "xfs_rtbitmap.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c index da2f6729699d..67ac870052b1 100644 --- a/fs/xfs/scrub/fscounters.c +++ b/fs/xfs/scrub/fscounters.c @@ -415,7 +415,7 @@ xchk_fscount_count_frextents( if (!xfs_has_realtime(mp)) return 0; - xfs_ilock(sc->mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); + xfs_rtbitmap_lock_shared(sc->mp, XFS_RBMLOCK_BITMAP); error = xfs_rtalloc_query_all(sc->mp, sc->tp, xchk_fscount_add_frextent, fsc); if (error) { @@ -424,7 +424,7 @@ xchk_fscount_count_frextents( } out_unlock: - xfs_iunlock(sc->mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); + xfs_rtbitmap_unlock_shared(sc->mp, XFS_RBMLOCK_BITMAP); return error; } #else diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c index de59eec74765..85dbb46452ca 100644 --- a/fs/xfs/xfs_fsmap.c +++ b/fs/xfs/xfs_fsmap.c @@ -533,7 +533,7 @@ xfs_getfsmap_rtdev_rtbitmap( trace_xfs_fsmap_low_key_linear(mp, info->dev, start_rtb); trace_xfs_fsmap_high_key_linear(mp, info->dev, end_rtb); - xfs_ilock(mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); + xfs_rtbitmap_lock_shared(mp, XFS_RBMLOCK_BITMAP); /* * Set up query parameters to return free rtextents covering the range @@ -557,7 +557,7 @@ xfs_getfsmap_rtdev_rtbitmap( if (error) goto err; err: - xfs_iunlock(mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); + xfs_rtbitmap_unlock_shared(mp, XFS_RBMLOCK_BITMAP); return error; } #endif /* CONFIG_XFS_RT */ diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index e66f9bd5de5c..86f928d30fed 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -957,10 +957,10 @@ xfs_growfs_rt( nargs.tp = tp; /* - * Lock out other callers by grabbing the bitmap inode lock. + * Lock out other callers by grabbing the bitmap and summary + * inode locks and joining them to the transaction. */ - xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL | XFS_ILOCK_RTBITMAP); - xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL); + xfs_rtbitmap_lock(tp, mp); /* * Update the bitmap inode's size ondisk and incore. We need * to update the incore size so that inode inactivation won't @@ -970,11 +970,6 @@ xfs_growfs_rt( nsbp->sb_rbmblocks * nsbp->sb_blocksize; i_size_write(VFS_I(mp->m_rbmip), mp->m_rbmip->i_disk_size); xfs_trans_log_inode(tp, mp->m_rbmip, XFS_ILOG_CORE); - /* - * Get the summary inode into the transaction. - */ - xfs_ilock(mp->m_rsumip, XFS_ILOCK_EXCL | XFS_ILOCK_RTSUM); - xfs_trans_ijoin(tp, mp->m_rsumip, XFS_ILOCK_EXCL); /* * Update the summary inode's size. We need to update the * incore size so that inode inactivation won't punch what it @@ -1142,10 +1137,10 @@ xfs_rtalloc_reinit_frextents( uint64_t val = 0; int error; - xfs_ilock(mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); + xfs_rtbitmap_lock_shared(mp, XFS_RBMLOCK_BITMAP); error = xfs_rtalloc_query_all(mp, NULL, xfs_rtalloc_count_frextent, &val); - xfs_iunlock(mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); + xfs_rtbitmap_unlock_shared(mp, XFS_RBMLOCK_BITMAP); if (error) return error; @@ -1382,10 +1377,7 @@ retry: * Lock out modifications to both the RT bitmap and summary inodes */ if (!rtlocked) { - xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL|XFS_ILOCK_RTBITMAP); - xfs_trans_ijoin(ap->tp, mp->m_rbmip, XFS_ILOCK_EXCL); - xfs_ilock(mp->m_rsumip, XFS_ILOCK_EXCL|XFS_ILOCK_RTSUM); - xfs_trans_ijoin(ap->tp, mp->m_rsumip, XFS_ILOCK_EXCL); + xfs_rtbitmap_lock(ap->tp, mp); rtlocked = true; } -- cgit v1.2.3-59-g8ed1b From 9871d0963751293bf0587759a9b6b8f808e35c7c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:09 +0200 Subject: xfs: free RT extents after updating the bmap btree Currently xfs_bmap_del_extent_real frees RT extents before updating the bmap btree, while it frees regular blocks after performing the bmap btree update for convoluted historic reasons. Switch to free the RT blocks in the same place as the regular data blocks instead to simply the code and fix a very theoretical bug. A short history of this code researched by Dave Chiner below: The truncate for data device extents was originally a two-phase operation. First it removed the bmapbt record, but because this can free BMBT extents, it can use up all the free space tree reservation space. So the transaction gets rolled to commit the BMBT change and the xfs_bmap_finish() call that frees the data extent runs with a new transaction reservation that allows different free space btrees to be logged without overrun. However, on crash, this could lose the free space because there was nothing to tell recovery about the extents removed from the BMBT, hence EFIs were introduced. They tie the extent free operation to the bmapbt record removal commit for recovery of the second phase of the extent removal process. Then RT extents came along. RT extent freeing does not require a free space btree reservation because the free space metadata is static and transaction size is bound. Hence we don't need to care if the BMBT record removal modifies the per-ag free space trees and we don't need a two-phase extent remove transaction. The only thing we have to care about is not losing space on crash. Hence instead of recording the extent for freeing in the bmap list for xfs_bmap_finish() to process in a new transaction, it simply freed the rtextent directly. So the original code (from 1994) simply replaced the "free AG extent later" queueing with a direct free. This code was originally at the start of xfs_dmap_del_extent(), but the xfs_bmap_add_free() got moved to the end of the function via the "do_fx" flag (the current code logic) in 1997 (commit c4fac74eaa58 in the historic xfs-import tree) because there was a shutdown occurring because of a case where splitting the extent record failed because the BMBT split and the filesystem didn't have enough space for the split to be done. (FWIW, I'm not sure this can happen anymore.) The commit backed out the BMBT change on ENOSPC error, and in doing so I think this actually breaks RT free space tracking. However, it then returns an ENOSPC error, and we have a dirty transaction in the RT case so this will shut down the filesysetm when the transaction is cancelled. Hence the corrupted "bmbt now points at freed rt dev space" condition never make it to disk, but it's still the wrong way to handle the issue. IOWs, this proposed change fixes that "shutdown at ENOSPC on rt devices" situation that was introduced by the above commit back in 1997. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 26 +++++++++----------------- 1 file changed, 9 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 3e56173a8fe4..f529aa407109 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -5109,8 +5109,7 @@ xfs_bmap_del_extent_real( { xfs_fsblock_t del_endblock=0; /* first block past del */ xfs_fileoff_t del_endoff; /* first offset past del */ - int do_fx; /* free extent at end of routine */ - int error; /* error return value */ + int error = 0; /* error return value */ struct xfs_bmbt_irec got; /* current extent entry */ xfs_fileoff_t got_endoff; /* first offset past got */ int i; /* temp state */ @@ -5153,20 +5152,10 @@ xfs_bmap_del_extent_real( return -ENOSPC; *logflagsp = XFS_ILOG_CORE; - if (xfs_ifork_is_realtime(ip, whichfork)) { - if (!(bflags & XFS_BMAPI_REMAP)) { - error = xfs_rtfree_blocks(tp, del->br_startblock, - del->br_blockcount); - if (error) - return error; - } - - do_fx = 0; + if (xfs_ifork_is_realtime(ip, whichfork)) qfield = XFS_TRANS_DQ_RTBCOUNT; - } else { - do_fx = 1; + else qfield = XFS_TRANS_DQ_BCOUNT; - } nblks = del->br_blockcount; del_endblock = del->br_startblock + del->br_blockcount; @@ -5314,18 +5303,21 @@ xfs_bmap_del_extent_real( /* * If we need to, add to list of extents to delete. */ - if (do_fx && !(bflags & XFS_BMAPI_REMAP)) { + if (!(bflags & XFS_BMAPI_REMAP)) { if (xfs_is_reflink_inode(ip) && whichfork == XFS_DATA_FORK) { xfs_refcount_decrease_extent(tp, del); + } else if (xfs_ifork_is_realtime(ip, whichfork)) { + error = xfs_rtfree_blocks(tp, del->br_startblock, + del->br_blockcount); } else { error = xfs_free_extent_later(tp, del->br_startblock, del->br_blockcount, NULL, XFS_AG_RESV_NONE, ((bflags & XFS_BMAPI_NODISCARD) || del->br_state == XFS_EXT_UNWRITTEN)); - if (error) - return error; } + if (error) + return error; } /* -- cgit v1.2.3-59-g8ed1b From de37dbd0ccc6933fbf4bd7b3ccbc5ac640e80b28 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:10 +0200 Subject: xfs: move RT inode locking out of __xfs_bunmapi __xfs_bunmapi is a bit of an odd place to lock the rtbitmap and rtsummary inodes given that it is very high level code. While this only looks ugly right now, it will become a problem when supporting delayed allocations for RT inodes as __xfs_bunmapi might end up deleting only delalloc extents and thus never unlock the rt inodes. Move the locking into xfs_bmap_del_extent_real just before the call to xfs_rtfree_blocks instead and use a new flag in the transaction to ensure that the locking happens only once. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 15 ++++++++------- fs/xfs/libxfs/xfs_shared.h | 3 +++ 2 files changed, 11 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index f529aa407109..22d44627ec59 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -5307,6 +5307,14 @@ xfs_bmap_del_extent_real( if (xfs_is_reflink_inode(ip) && whichfork == XFS_DATA_FORK) { xfs_refcount_decrease_extent(tp, del); } else if (xfs_ifork_is_realtime(ip, whichfork)) { + /* + * Ensure the bitmap and summary inodes are locked + * and joined to the transaction before modifying them. + */ + if (!(tp->t_flags & XFS_TRANS_RTBITMAP_LOCKED)) { + tp->t_flags |= XFS_TRANS_RTBITMAP_LOCKED; + xfs_rtbitmap_lock(tp, mp); + } error = xfs_rtfree_blocks(tp, del->br_startblock, del->br_blockcount); } else { @@ -5408,13 +5416,6 @@ __xfs_bunmapi( } else cur = NULL; - if (isrt) { - /* - * Synchronize by locking the realtime bitmap. - */ - xfs_rtbitmap_lock(tp, mp); - } - extno = 0; while (end != (xfs_fileoff_t)-1 && end >= start && (nexts == 0 || extno < nexts)) { diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h index f35640ad3e7f..34f104ed372c 100644 --- a/fs/xfs/libxfs/xfs_shared.h +++ b/fs/xfs/libxfs/xfs_shared.h @@ -137,6 +137,9 @@ void xfs_log_get_max_trans_res(struct xfs_mount *mp, */ #define XFS_TRANS_LOWMODE (1u << 8) +/* Transaction has locked the rtbitmap and rtsum inodes */ +#define XFS_TRANS_RTBITMAP_LOCKED (1u << 9) + /* * Field values for xfs_trans_mod_sb. */ -- cgit v1.2.3-59-g8ed1b From 5e1e4d4fc79c6e5f80864860208ceae27dead8a2 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:11 +0200 Subject: xfs: block deltas in xfs_trans_unreserve_and_mod_sb must be positive And to make that more clear, rearrange the code a bit and add asserts and a comment. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_trans.c | 38 ++++++++++++++++++++++++-------------- 1 file changed, 24 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 50d878d78a5e..95921ad93547 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -594,28 +594,38 @@ xfs_trans_unreserve_and_mod_sb( { struct xfs_mount *mp = tp->t_mountp; bool rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0; - int64_t blkdelta = 0; - int64_t rtxdelta = 0; + int64_t blkdelta = tp->t_blk_res; + int64_t rtxdelta = tp->t_rtx_res; int64_t idelta = 0; int64_t ifreedelta = 0; int error; - /* calculate deltas */ - if (tp->t_blk_res > 0) - blkdelta = tp->t_blk_res; - if ((tp->t_fdblocks_delta != 0) && - (xfs_has_lazysbcount(mp) || - (tp->t_flags & XFS_TRANS_SB_DIRTY))) + /* + * Calculate the deltas. + * + * t_fdblocks_delta and t_frextents_delta can be positive or negative: + * + * - positive values indicate blocks freed in the transaction. + * - negative values indicate blocks allocated in the transaction + * + * Negative values can only happen if the transaction has a block + * reservation that covers the allocated block. The end result is + * that the calculated delta values must always be positive and we + * can only put back previous allocated or reserved blocks here. + */ + ASSERT(tp->t_blk_res || tp->t_fdblocks_delta >= 0); + if (xfs_has_lazysbcount(mp) || (tp->t_flags & XFS_TRANS_SB_DIRTY)) { blkdelta += tp->t_fdblocks_delta; + ASSERT(blkdelta >= 0); + } - if (tp->t_rtx_res > 0) - rtxdelta = tp->t_rtx_res; - if ((tp->t_frextents_delta != 0) && - (tp->t_flags & XFS_TRANS_SB_DIRTY)) + ASSERT(tp->t_rtx_res || tp->t_frextents_delta >= 0); + if (tp->t_flags & XFS_TRANS_SB_DIRTY) { rtxdelta += tp->t_frextents_delta; + ASSERT(rtxdelta >= 0); + } - if (xfs_has_lazysbcount(mp) || - (tp->t_flags & XFS_TRANS_SB_DIRTY)) { + if (xfs_has_lazysbcount(mp) || (tp->t_flags & XFS_TRANS_SB_DIRTY)) { idelta = tp->t_icount_delta; ifreedelta = tp->t_ifree_delta; } -- cgit v1.2.3-59-g8ed1b From f30f656e25eb72c4309e76b16fa45062e183a2ee Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:12 +0200 Subject: xfs: split xfs_mod_freecounter xfs_mod_freecounter has two entirely separate code paths for adding or subtracting from the free counters. Only the subtract case looks at the rsvd flag and can return an error. Split xfs_mod_freecounter into separate helpers for subtracting or adding the freecounter, and remove all the impossible to reach error handling for the addition case. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_ag.c | 4 +-- fs/xfs/libxfs/xfs_ag_resv.c | 24 ++++------------ fs/xfs/libxfs/xfs_ag_resv.h | 2 +- fs/xfs/libxfs/xfs_alloc.c | 4 +-- fs/xfs/libxfs/xfs_bmap.c | 23 ++++++++-------- fs/xfs/scrub/fscounters.c | 2 +- fs/xfs/scrub/repair.c | 5 +--- fs/xfs/xfs_fsops.c | 29 ++++++-------------- fs/xfs/xfs_fsops.h | 2 +- fs/xfs/xfs_mount.c | 67 +++++++++++++++++++++++---------------------- fs/xfs/xfs_mount.h | 27 ++++++++++++------ fs/xfs/xfs_super.c | 6 +--- fs/xfs/xfs_trace.h | 1 - fs/xfs/xfs_trans.c | 25 ++++++----------- 14 files changed, 97 insertions(+), 124 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index 09fe9412eab4..240e079cb3fb 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -963,9 +963,7 @@ xfs_ag_shrink_space( * Disable perag reservations so it doesn't cause the allocation request * to fail. We'll reestablish reservation before we return. */ - error = xfs_ag_resv_free(pag); - if (error) - return error; + xfs_ag_resv_free(pag); /* internal log shouldn't also show up in the free space btrees */ error = xfs_alloc_vextent_exact_bno(&args, diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c index da1057bd0e60..216423df939e 100644 --- a/fs/xfs/libxfs/xfs_ag_resv.c +++ b/fs/xfs/libxfs/xfs_ag_resv.c @@ -126,14 +126,13 @@ xfs_ag_resv_needed( } /* Clean out a reservation */ -static int +static void __xfs_ag_resv_free( struct xfs_perag *pag, enum xfs_ag_resv_type type) { struct xfs_ag_resv *resv; xfs_extlen_t oldresv; - int error; trace_xfs_ag_resv_free(pag, type, 0); @@ -149,30 +148,19 @@ __xfs_ag_resv_free( oldresv = resv->ar_orig_reserved; else oldresv = resv->ar_reserved; - error = xfs_mod_fdblocks(pag->pag_mount, oldresv, true); + xfs_add_fdblocks(pag->pag_mount, oldresv); resv->ar_reserved = 0; resv->ar_asked = 0; resv->ar_orig_reserved = 0; - - if (error) - trace_xfs_ag_resv_free_error(pag->pag_mount, pag->pag_agno, - error, _RET_IP_); - return error; } /* Free a per-AG reservation. */ -int +void xfs_ag_resv_free( struct xfs_perag *pag) { - int error; - int err2; - - error = __xfs_ag_resv_free(pag, XFS_AG_RESV_RMAPBT); - err2 = __xfs_ag_resv_free(pag, XFS_AG_RESV_METADATA); - if (err2 && !error) - error = err2; - return error; + __xfs_ag_resv_free(pag, XFS_AG_RESV_RMAPBT); + __xfs_ag_resv_free(pag, XFS_AG_RESV_METADATA); } static int @@ -216,7 +204,7 @@ __xfs_ag_resv_init( if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_AG_RESV_FAIL)) error = -ENOSPC; else - error = xfs_mod_fdblocks(mp, -(int64_t)hidden_space, true); + error = xfs_dec_fdblocks(mp, hidden_space, true); if (error) { trace_xfs_ag_resv_init_error(pag->pag_mount, pag->pag_agno, error, _RET_IP_); diff --git a/fs/xfs/libxfs/xfs_ag_resv.h b/fs/xfs/libxfs/xfs_ag_resv.h index b74b210008ea..ff20ed93de77 100644 --- a/fs/xfs/libxfs/xfs_ag_resv.h +++ b/fs/xfs/libxfs/xfs_ag_resv.h @@ -6,7 +6,7 @@ #ifndef __XFS_AG_RESV_H__ #define __XFS_AG_RESV_H__ -int xfs_ag_resv_free(struct xfs_perag *pag); +void xfs_ag_resv_free(struct xfs_perag *pag); int xfs_ag_resv_init(struct xfs_perag *pag, struct xfs_trans *tp); bool xfs_ag_resv_critical(struct xfs_perag *pag, enum xfs_ag_resv_type type); diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 9da52e92172a..6cb8b2ddc541 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -79,7 +79,7 @@ xfs_prealloc_blocks( } /* - * The number of blocks per AG that we withhold from xfs_mod_fdblocks to + * The number of blocks per AG that we withhold from xfs_dec_fdblocks to * guarantee that we can refill the AGFL prior to allocating space in a nearly * full AG. Although the space described by the free space btrees, the * blocks used by the freesp btrees themselves, and the blocks owned by the @@ -89,7 +89,7 @@ xfs_prealloc_blocks( * until the fs goes down, we subtract this many AG blocks from the incore * fdblocks to ensure user allocation does not overcommit the space the * filesystem needs for the AGFLs. The rmap btree uses a per-AG reservation to - * withhold space from xfs_mod_fdblocks, so we do not account for that here. + * withhold space from xfs_dec_fdblocks, so we do not account for that here. */ #define XFS_ALLOCBT_AGFL_RESERVE 4 diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 22d44627ec59..0311a98df1c5 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -1985,10 +1985,11 @@ xfs_bmap_add_extent_delay_real( } /* adjust for changes in reserved delayed indirect blocks */ - if (da_new != da_old) { - ASSERT(state == 0 || da_new < da_old); - error = xfs_mod_fdblocks(mp, (int64_t)(da_old - da_new), - false); + if (da_new < da_old) { + xfs_add_fdblocks(mp, da_old - da_new); + } else if (da_new > da_old) { + ASSERT(state == 0); + error = xfs_dec_fdblocks(mp, da_new - da_old, false); } xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork); @@ -2690,8 +2691,8 @@ xfs_bmap_add_extent_hole_delay( } if (oldlen != newlen) { ASSERT(oldlen > newlen); - xfs_mod_fdblocks(ip->i_mount, (int64_t)(oldlen - newlen), - false); + xfs_add_fdblocks(ip->i_mount, oldlen - newlen); + /* * Nothing to do for disk quota accounting here. */ @@ -4110,11 +4111,11 @@ xfs_bmapi_reserve_delalloc( indlen = (xfs_extlen_t)xfs_bmap_worst_indlen(ip, alen); ASSERT(indlen > 0); - error = xfs_mod_fdblocks(mp, -((int64_t)alen), false); + error = xfs_dec_fdblocks(mp, alen, false); if (error) goto out_unreserve_quota; - error = xfs_mod_fdblocks(mp, -((int64_t)indlen), false); + error = xfs_dec_fdblocks(mp, indlen, false); if (error) goto out_unreserve_blocks; @@ -4142,7 +4143,7 @@ xfs_bmapi_reserve_delalloc( return 0; out_unreserve_blocks: - xfs_mod_fdblocks(mp, alen, false); + xfs_add_fdblocks(mp, alen); out_unreserve_quota: if (XFS_IS_QUOTA_ON(mp)) xfs_quota_unreserve_blkres(ip, alen); @@ -4928,7 +4929,7 @@ xfs_bmap_del_extent_delay( ASSERT(got_endoff >= del_endoff); if (isrt) - xfs_mod_frextents(mp, xfs_rtb_to_rtx(mp, del->br_blockcount)); + xfs_add_frextents(mp, xfs_rtb_to_rtx(mp, del->br_blockcount)); /* * Update the inode delalloc counter now and wait to update the @@ -5015,7 +5016,7 @@ xfs_bmap_del_extent_delay( if (!isrt) da_diff += del->br_blockcount; if (da_diff) { - xfs_mod_fdblocks(mp, da_diff, false); + xfs_add_fdblocks(mp, da_diff); xfs_mod_delalloc(mp, -da_diff); } return error; diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c index 67ac870052b1..b662e52fe695 100644 --- a/fs/xfs/scrub/fscounters.c +++ b/fs/xfs/scrub/fscounters.c @@ -517,7 +517,7 @@ xchk_fscounters( /* * If the filesystem is not frozen, the counter summation calls above - * can race with xfs_mod_freecounter, which subtracts a requested space + * can race with xfs_dec_freecounter, which subtracts a requested space * reservation from the counter and undoes the subtraction if that made * the counter go negative. Therefore, it's possible to see negative * values here, and we should only flag that as a corruption if we diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c index 92f81e374bc0..67478294f11a 100644 --- a/fs/xfs/scrub/repair.c +++ b/fs/xfs/scrub/repair.c @@ -968,9 +968,7 @@ xrep_reset_perag_resv( ASSERT(sc->tp); sc->flags &= ~XREP_RESET_PERAG_RESV; - error = xfs_ag_resv_free(sc->sa.pag); - if (error) - goto out; + xfs_ag_resv_free(sc->sa.pag); error = xfs_ag_resv_init(sc->sa.pag, sc->tp); if (error == -ENOSPC) { xfs_err(sc->mp, @@ -979,7 +977,6 @@ xrep_reset_perag_resv( error = 0; } -out: return error; } diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 83f708f62ed9..c211ea2b63c4 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -213,10 +213,8 @@ xfs_growfs_data_private( struct xfs_perag *pag; pag = xfs_perag_get(mp, id.agno); - error = xfs_ag_resv_free(pag); + xfs_ag_resv_free(pag); xfs_perag_put(pag); - if (error) - return error; } /* * Reserve AG metadata blocks. ENOSPC here does not mean there @@ -385,14 +383,14 @@ xfs_reserve_blocks( */ if (mp->m_resblks > request) { lcounter = mp->m_resblks_avail - request; - if (lcounter > 0) { /* release unused blocks */ + if (lcounter > 0) { /* release unused blocks */ fdblks_delta = lcounter; mp->m_resblks_avail -= lcounter; } mp->m_resblks = request; if (fdblks_delta) { spin_unlock(&mp->m_sb_lock); - error = xfs_mod_fdblocks(mp, fdblks_delta, 0); + xfs_add_fdblocks(mp, fdblks_delta); spin_lock(&mp->m_sb_lock); } @@ -428,9 +426,9 @@ xfs_reserve_blocks( */ fdblks_delta = min(free, delta); spin_unlock(&mp->m_sb_lock); - error = xfs_mod_fdblocks(mp, -fdblks_delta, 0); + error = xfs_dec_fdblocks(mp, fdblks_delta, 0); if (!error) - xfs_mod_fdblocks(mp, fdblks_delta, 0); + xfs_add_fdblocks(mp, fdblks_delta); spin_lock(&mp->m_sb_lock); } out: @@ -556,24 +554,13 @@ xfs_fs_reserve_ag_blocks( /* * Free space reserved for per-AG metadata. */ -int +void xfs_fs_unreserve_ag_blocks( struct xfs_mount *mp) { xfs_agnumber_t agno; struct xfs_perag *pag; - int error = 0; - int err2; - for_each_perag(mp, agno, pag) { - err2 = xfs_ag_resv_free(pag); - if (err2 && !error) - error = err2; - } - - if (error) - xfs_warn(mp, - "Error %d freeing per-AG metadata reserve pool.", error); - - return error; + for_each_perag(mp, agno, pag) + xfs_ag_resv_free(pag); } diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h index 44457b0a0593..3e2f73bcf831 100644 --- a/fs/xfs/xfs_fsops.h +++ b/fs/xfs/xfs_fsops.h @@ -12,6 +12,6 @@ int xfs_reserve_blocks(struct xfs_mount *mp, uint64_t request); int xfs_fs_goingdown(struct xfs_mount *mp, uint32_t inflags); int xfs_fs_reserve_ag_blocks(struct xfs_mount *mp); -int xfs_fs_unreserve_ag_blocks(struct xfs_mount *mp); +void xfs_fs_unreserve_ag_blocks(struct xfs_mount *mp); #endif /* __XFS_FSOPS_H__ */ diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index d37ba10f5fa3..b9df7cc9c473 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1136,16 +1136,44 @@ xfs_fs_writable( return true; } -/* Adjust m_fdblocks or m_frextents. */ +void +xfs_add_freecounter( + struct xfs_mount *mp, + struct percpu_counter *counter, + uint64_t delta) +{ + bool has_resv_pool = (counter == &mp->m_fdblocks); + uint64_t res_used; + + /* + * If the reserve pool is depleted, put blocks back into it first. + * Most of the time the pool is full. + */ + if (!has_resv_pool || mp->m_resblks == mp->m_resblks_avail) { + percpu_counter_add(counter, delta); + return; + } + + spin_lock(&mp->m_sb_lock); + res_used = mp->m_resblks - mp->m_resblks_avail; + if (res_used > delta) { + mp->m_resblks_avail += delta; + } else { + delta -= res_used; + mp->m_resblks_avail = mp->m_resblks; + percpu_counter_add(counter, delta); + } + spin_unlock(&mp->m_sb_lock); +} + int -xfs_mod_freecounter( +xfs_dec_freecounter( struct xfs_mount *mp, struct percpu_counter *counter, - int64_t delta, + uint64_t delta, bool rsvd) { int64_t lcounter; - long long res_used; uint64_t set_aside = 0; s32 batch; bool has_resv_pool; @@ -1155,31 +1183,6 @@ xfs_mod_freecounter( if (rsvd) ASSERT(has_resv_pool); - if (delta > 0) { - /* - * If the reserve pool is depleted, put blocks back into it - * first. Most of the time the pool is full. - */ - if (likely(!has_resv_pool || - mp->m_resblks == mp->m_resblks_avail)) { - percpu_counter_add(counter, delta); - return 0; - } - - spin_lock(&mp->m_sb_lock); - res_used = (long long)(mp->m_resblks - mp->m_resblks_avail); - - if (res_used > delta) { - mp->m_resblks_avail += delta; - } else { - delta -= res_used; - mp->m_resblks_avail = mp->m_resblks; - percpu_counter_add(counter, delta); - } - spin_unlock(&mp->m_sb_lock); - return 0; - } - /* * Taking blocks away, need to be more accurate the closer we * are to zero. @@ -1207,7 +1210,7 @@ xfs_mod_freecounter( */ if (has_resv_pool) set_aside = xfs_fdblocks_unavailable(mp); - percpu_counter_add_batch(counter, delta, batch); + percpu_counter_add_batch(counter, -((int64_t)delta), batch); if (__percpu_counter_compare(counter, set_aside, XFS_FDBLOCKS_BATCH) >= 0) { /* we had space! */ @@ -1219,11 +1222,11 @@ xfs_mod_freecounter( * that took us to ENOSPC. */ spin_lock(&mp->m_sb_lock); - percpu_counter_add(counter, -delta); + percpu_counter_add(counter, delta); if (!has_resv_pool || !rsvd) goto fdblocks_enospc; - lcounter = (long long)mp->m_resblks_avail + delta; + lcounter = (long long)mp->m_resblks_avail - delta; if (lcounter >= 0) { mp->m_resblks_avail = lcounter; spin_unlock(&mp->m_sb_lock); diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 283a5a26db31..3a985c2ff06f 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -562,19 +562,30 @@ xfs_fdblocks_unavailable( return mp->m_alloc_set_aside + atomic64_read(&mp->m_allocbt_blks); } -int xfs_mod_freecounter(struct xfs_mount *mp, struct percpu_counter *counter, - int64_t delta, bool rsvd); +int xfs_dec_freecounter(struct xfs_mount *mp, struct percpu_counter *counter, + uint64_t delta, bool rsvd); +void xfs_add_freecounter(struct xfs_mount *mp, struct percpu_counter *counter, + uint64_t delta); -static inline int -xfs_mod_fdblocks(struct xfs_mount *mp, int64_t delta, bool reserved) +static inline int xfs_dec_fdblocks(struct xfs_mount *mp, uint64_t delta, + bool reserved) { - return xfs_mod_freecounter(mp, &mp->m_fdblocks, delta, reserved); + return xfs_dec_freecounter(mp, &mp->m_fdblocks, delta, reserved); } -static inline int -xfs_mod_frextents(struct xfs_mount *mp, int64_t delta) +static inline void xfs_add_fdblocks(struct xfs_mount *mp, uint64_t delta) { - return xfs_mod_freecounter(mp, &mp->m_frextents, delta, false); + xfs_add_freecounter(mp, &mp->m_fdblocks, delta); +} + +static inline int xfs_dec_frextents(struct xfs_mount *mp, uint64_t delta) +{ + return xfs_dec_freecounter(mp, &mp->m_frextents, delta, false); +} + +static inline void xfs_add_frextents(struct xfs_mount *mp, uint64_t delta) +{ + xfs_add_freecounter(mp, &mp->m_frextents, delta); } extern int xfs_readsb(xfs_mount_t *, int); diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 02c1a4aeaa4c..ed4a2f8c1936 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1882,11 +1882,7 @@ xfs_remount_ro( xfs_inodegc_stop(mp); /* Free the per-AG metadata reservation pool. */ - error = xfs_fs_unreserve_ag_blocks(mp); - if (error) { - xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); - return error; - } + xfs_fs_unreserve_ag_blocks(mp); /* * Before we sync the metadata, we need to free up the reserve block diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 0f844396754d..08df5bb70a6c 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -3069,7 +3069,6 @@ DEFINE_AG_RESV_EVENT(xfs_ag_resv_free_extent); DEFINE_AG_RESV_EVENT(xfs_ag_resv_critical); DEFINE_AG_RESV_EVENT(xfs_ag_resv_needed); -DEFINE_AG_ERROR_EVENT(xfs_ag_resv_free_error); DEFINE_AG_ERROR_EVENT(xfs_ag_resv_init_error); /* refcount tracepoint classes */ diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 95921ad93547..828da4ac4316 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -163,7 +163,7 @@ xfs_trans_reserve( * fail if the count would go below zero. */ if (blocks > 0) { - error = xfs_mod_fdblocks(mp, -((int64_t)blocks), rsvd); + error = xfs_dec_fdblocks(mp, blocks, rsvd); if (error != 0) return -ENOSPC; tp->t_blk_res += blocks; @@ -210,7 +210,7 @@ xfs_trans_reserve( * fail if the count would go below zero. */ if (rtextents > 0) { - error = xfs_mod_frextents(mp, -((int64_t)rtextents)); + error = xfs_dec_frextents(mp, rtextents); if (error) { error = -ENOSPC; goto undo_log; @@ -234,7 +234,7 @@ undo_log: undo_blocks: if (blocks > 0) { - xfs_mod_fdblocks(mp, (int64_t)blocks, rsvd); + xfs_add_fdblocks(mp, blocks); tp->t_blk_res = 0; } return error; @@ -593,12 +593,10 @@ xfs_trans_unreserve_and_mod_sb( struct xfs_trans *tp) { struct xfs_mount *mp = tp->t_mountp; - bool rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0; int64_t blkdelta = tp->t_blk_res; int64_t rtxdelta = tp->t_rtx_res; int64_t idelta = 0; int64_t ifreedelta = 0; - int error; /* * Calculate the deltas. @@ -631,10 +629,8 @@ xfs_trans_unreserve_and_mod_sb( } /* apply the per-cpu counters */ - if (blkdelta) { - error = xfs_mod_fdblocks(mp, blkdelta, rsvd); - ASSERT(!error); - } + if (blkdelta) + xfs_add_fdblocks(mp, blkdelta); if (idelta) percpu_counter_add_batch(&mp->m_icount, idelta, @@ -643,10 +639,8 @@ xfs_trans_unreserve_and_mod_sb( if (ifreedelta) percpu_counter_add(&mp->m_ifree, ifreedelta); - if (rtxdelta) { - error = xfs_mod_frextents(mp, rtxdelta); - ASSERT(!error); - } + if (rtxdelta) + xfs_add_frextents(mp, rtxdelta); if (!(tp->t_flags & XFS_TRANS_SB_DIRTY)) return; @@ -682,7 +676,6 @@ xfs_trans_unreserve_and_mod_sb( */ ASSERT(mp->m_sb.sb_imax_pct >= 0); ASSERT(mp->m_sb.sb_rextslog >= 0); - return; } /* Add the given log item to the transaction's list of log items. */ @@ -1301,9 +1294,9 @@ xfs_trans_reserve_more_inode( return 0; /* Quota failed, give back the new reservation. */ - xfs_mod_fdblocks(mp, dblocks, tp->t_flags & XFS_TRANS_RESERVE); + xfs_add_fdblocks(mp, dblocks); tp->t_blk_res -= dblocks; - xfs_mod_frextents(mp, rtx); + xfs_add_frextents(mp, rtx); tp->t_rtx_res -= rtx; return error; } -- cgit v1.2.3-59-g8ed1b From dc1b17a25c321c8f1b4f90f9d6f8afb1d132b69c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:13 +0200 Subject: xfs: reinstate RT support in xfs_bmapi_reserve_delalloc Allocate data blocks for RT inodes using xfs_dec_frextents. While at it optimize the data device case by doing only a single xfs_dec_fdblocks call for the extent itself and the indirect blocks. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 0311a98df1c5..f14224ff4d56 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4069,6 +4069,7 @@ xfs_bmapi_reserve_delalloc( struct xfs_ifork *ifp = xfs_ifork_ptr(ip, whichfork); xfs_extlen_t alen; xfs_extlen_t indlen; + uint64_t fdblocks; int error; xfs_fileoff_t aoff = off; @@ -4111,14 +4112,18 @@ xfs_bmapi_reserve_delalloc( indlen = (xfs_extlen_t)xfs_bmap_worst_indlen(ip, alen); ASSERT(indlen > 0); - error = xfs_dec_fdblocks(mp, alen, false); - if (error) - goto out_unreserve_quota; + fdblocks = indlen; + if (XFS_IS_REALTIME_INODE(ip)) { + error = xfs_dec_frextents(mp, xfs_rtb_to_rtx(mp, alen)); + if (error) + goto out_unreserve_quota; + } else { + fdblocks += alen; + } - error = xfs_dec_fdblocks(mp, indlen, false); + error = xfs_dec_fdblocks(mp, fdblocks, false); if (error) - goto out_unreserve_blocks; - + goto out_unreserve_frextents; ip->i_delayed_blks += alen; xfs_mod_delalloc(ip->i_mount, alen + indlen); @@ -4142,8 +4147,9 @@ xfs_bmapi_reserve_delalloc( return 0; -out_unreserve_blocks: - xfs_add_fdblocks(mp, alen); +out_unreserve_frextents: + if (XFS_IS_REALTIME_INODE(ip)) + xfs_add_frextents(mp, xfs_rtb_to_rtx(mp, alen)); out_unreserve_quota: if (XFS_IS_QUOTA_ON(mp)) xfs_quota_unreserve_blkres(ip, alen); -- cgit v1.2.3-59-g8ed1b From 7e77d57a1fea5f6bfe166210385ba9f227a606d1 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:14 +0200 Subject: xfs: cleanup fdblock/frextent accounting in xfs_bmap_del_extent_delay The code to account fdblocks and frextents in xfs_bmap_del_extent_delay is a bit weird in that it accounts frextents before the iext tree manipulations and fdblocks after it. Given that the iext tree manipulations cannot fail currently that's not really a problem, but still odd. Move the frextent manipulation to the end, and use a fdblocks variable to account of the unconditional indirect blocks and the data blocks only freed for !RT. This prepares for following updates in the area and already makes the code more readable. Also remove the !isrt assert given that this code clearly handles rt extents correctly, and we'll soon reinstate delalloc support for RT inodes. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index f14224ff4d56..d7a5ce2e4305 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4919,6 +4919,7 @@ xfs_bmap_del_extent_delay( xfs_fileoff_t del_endoff, got_endoff; xfs_filblks_t got_indlen, new_indlen, stolen; uint32_t state = xfs_bmap_fork_to_state(whichfork); + uint64_t fdblocks; int error = 0; bool isrt; @@ -4934,15 +4935,11 @@ xfs_bmap_del_extent_delay( ASSERT(got->br_startoff <= del->br_startoff); ASSERT(got_endoff >= del_endoff); - if (isrt) - xfs_add_frextents(mp, xfs_rtb_to_rtx(mp, del->br_blockcount)); - /* * Update the inode delalloc counter now and wait to update the * sb counters as we might have to borrow some blocks for the * indirect block accounting. */ - ASSERT(!isrt); error = xfs_quota_unreserve_blkres(ip, del->br_blockcount); if (error) return error; @@ -5019,12 +5016,15 @@ xfs_bmap_del_extent_delay( ASSERT(da_old >= da_new); da_diff = da_old - da_new; - if (!isrt) - da_diff += del->br_blockcount; - if (da_diff) { - xfs_add_fdblocks(mp, da_diff); - xfs_mod_delalloc(mp, -da_diff); - } + fdblocks = da_diff; + + if (isrt) + xfs_add_frextents(mp, xfs_rtb_to_rtx(mp, del->br_blockcount)); + else + fdblocks += del->br_blockcount; + + xfs_add_fdblocks(mp, fdblocks); + xfs_mod_delalloc(mp, -(int64_t)fdblocks); return error; } -- cgit v1.2.3-59-g8ed1b From 7099bd0f243fa7511de6e95b0b8807ba7d3e5204 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:15 +0200 Subject: xfs: support RT inodes in xfs_mod_delalloc To prepare for re-enabling delalloc on RT devices, track the data blocks (which use the RT device when the inode sits on it) and the indirect blocks (which don't) separately to xfs_mod_delalloc, and add a new percpu counter to also track the RT delalloc blocks. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 12 ++++++------ fs/xfs/scrub/fscounters.c | 6 +++++- fs/xfs/scrub/fscounters.h | 1 + fs/xfs/scrub/fscounters_repair.c | 12 +++++++++++- fs/xfs/xfs_mount.c | 18 +++++++++++++++--- fs/xfs/xfs_mount.h | 9 ++++++++- fs/xfs/xfs_super.c | 11 ++++++++++- 7 files changed, 56 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index d7a5ce2e4305..a2b715450ec2 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -1977,7 +1977,7 @@ xfs_bmap_add_extent_delay_real( } if (da_new != da_old) - xfs_mod_delalloc(mp, (int64_t)da_new - da_old); + xfs_mod_delalloc(bma->ip, 0, (int64_t)da_new - da_old); if (bma->cur) { da_new += bma->cur->bc_bmap.allocated; @@ -2696,7 +2696,7 @@ xfs_bmap_add_extent_hole_delay( /* * Nothing to do for disk quota accounting here. */ - xfs_mod_delalloc(ip->i_mount, (int64_t)newlen - oldlen); + xfs_mod_delalloc(ip, 0, (int64_t)newlen - oldlen); } } @@ -3373,7 +3373,7 @@ xfs_bmap_alloc_account( * yet. */ if (ap->wasdel) { - xfs_mod_delalloc(ap->ip->i_mount, -(int64_t)ap->length); + xfs_mod_delalloc(ap->ip, -(int64_t)ap->length, 0); return; } @@ -3397,7 +3397,7 @@ xfs_bmap_alloc_account( xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE); if (ap->wasdel) { ap->ip->i_delayed_blks -= ap->length; - xfs_mod_delalloc(ap->ip->i_mount, -(int64_t)ap->length); + xfs_mod_delalloc(ap->ip, -(int64_t)ap->length, 0); fld = isrt ? XFS_TRANS_DQ_DELRTBCOUNT : XFS_TRANS_DQ_DELBCOUNT; } else { fld = isrt ? XFS_TRANS_DQ_RTBCOUNT : XFS_TRANS_DQ_BCOUNT; @@ -4126,7 +4126,7 @@ xfs_bmapi_reserve_delalloc( goto out_unreserve_frextents; ip->i_delayed_blks += alen; - xfs_mod_delalloc(ip->i_mount, alen + indlen); + xfs_mod_delalloc(ip, alen, indlen); got->br_startoff = aoff; got->br_startblock = nullstartblock(indlen); @@ -5024,7 +5024,7 @@ xfs_bmap_del_extent_delay( fdblocks += del->br_blockcount; xfs_add_fdblocks(mp, fdblocks); - xfs_mod_delalloc(mp, -(int64_t)fdblocks); + xfs_mod_delalloc(ip, -(int64_t)del->br_blockcount, -da_diff); return error; } diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c index b662e52fe695..1d3e98346933 100644 --- a/fs/xfs/scrub/fscounters.c +++ b/fs/xfs/scrub/fscounters.c @@ -412,6 +412,7 @@ xchk_fscount_count_frextents( int error; fsc->frextents = 0; + fsc->frextents_delayed = 0; if (!xfs_has_realtime(mp)) return 0; @@ -423,6 +424,8 @@ xchk_fscount_count_frextents( goto out_unlock; } + fsc->frextents_delayed = percpu_counter_sum(&mp->m_delalloc_rtextents); + out_unlock: xfs_rtbitmap_unlock_shared(sc->mp, XFS_RBMLOCK_BITMAP); return error; @@ -434,6 +437,7 @@ xchk_fscount_count_frextents( struct xchk_fscounters *fsc) { fsc->frextents = 0; + fsc->frextents_delayed = 0; return 0; } #endif /* CONFIG_XFS_RT */ @@ -593,7 +597,7 @@ xchk_fscounters( } if (!xchk_fscount_within_range(sc, frextents, &mp->m_frextents, - fsc->frextents)) { + fsc->frextents - fsc->frextents_delayed)) { if (fsc->frozen) xchk_set_corrupt(sc); else diff --git a/fs/xfs/scrub/fscounters.h b/fs/xfs/scrub/fscounters.h index 461a13d25f4b..bcf56e1c36f9 100644 --- a/fs/xfs/scrub/fscounters.h +++ b/fs/xfs/scrub/fscounters.h @@ -12,6 +12,7 @@ struct xchk_fscounters { uint64_t ifree; uint64_t fdblocks; uint64_t frextents; + uint64_t frextents_delayed; unsigned long long icount_min; unsigned long long icount_max; bool frozen; diff --git a/fs/xfs/scrub/fscounters_repair.c b/fs/xfs/scrub/fscounters_repair.c index 94cdb852bee4..469bf645dbea 100644 --- a/fs/xfs/scrub/fscounters_repair.c +++ b/fs/xfs/scrub/fscounters_repair.c @@ -65,7 +65,17 @@ xrep_fscounters( percpu_counter_set(&mp->m_icount, fsc->icount); percpu_counter_set(&mp->m_ifree, fsc->ifree); percpu_counter_set(&mp->m_fdblocks, fsc->fdblocks); - percpu_counter_set(&mp->m_frextents, fsc->frextents); + + /* + * Online repair is only supported on v5 file systems, which require + * lazy sb counters and thus no update of sb_fdblocks here. But as of + * now we don't support lazy counting sb_frextents yet, and thus need + * to also update it directly here. And for that we need to keep + * track of the delalloc reservations separately, as they are are + * subtracted from m_frextents, but not included in sb_frextents. + */ + percpu_counter_set(&mp->m_frextents, + fsc->frextents - fsc->frextents_delayed); mp->m_sb.sb_frextents = fsc->frextents; return 0; diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index b9df7cc9c473..b2e5653b5200 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -34,6 +34,7 @@ #include "xfs_health.h" #include "xfs_trace.h" #include "xfs_ag.h" +#include "xfs_rtbitmap.h" #include "scrub/stats.h" static DEFINE_MUTEX(xfs_uuid_table_mutex); @@ -1408,9 +1409,20 @@ xfs_clear_incompat_log_features( #define XFS_DELALLOC_BATCH (4096) void xfs_mod_delalloc( - struct xfs_mount *mp, - int64_t delta) + struct xfs_inode *ip, + int64_t data_delta, + int64_t ind_delta) { - percpu_counter_add_batch(&mp->m_delalloc_blks, delta, + struct xfs_mount *mp = ip->i_mount; + + if (XFS_IS_REALTIME_INODE(ip)) { + percpu_counter_add_batch(&mp->m_delalloc_rtextents, + xfs_rtb_to_rtx(mp, data_delta), + XFS_DELALLOC_BATCH); + if (!ind_delta) + return; + data_delta = 0; + } + percpu_counter_add_batch(&mp->m_delalloc_blks, data_delta + ind_delta, XFS_DELALLOC_BATCH); } diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 3a985c2ff06f..ca6f105990a2 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -195,6 +195,12 @@ typedef struct xfs_mount { * extents or anything related to the rt device. */ struct percpu_counter m_delalloc_blks; + + /* + * RT version of the above. + */ + struct percpu_counter m_delalloc_rtextents; + /* * Global count of allocation btree blocks in use across all AGs. Only * used when perag reservation is enabled. Helps prevent block @@ -605,6 +611,7 @@ struct xfs_error_cfg * xfs_error_get_cfg(struct xfs_mount *mp, void xfs_force_summary_recalc(struct xfs_mount *mp); int xfs_add_incompat_log_feature(struct xfs_mount *mp, uint32_t feature); bool xfs_clear_incompat_log_features(struct xfs_mount *mp); -void xfs_mod_delalloc(struct xfs_mount *mp, int64_t delta); +void xfs_mod_delalloc(struct xfs_inode *ip, int64_t data_delta, + int64_t ind_delta); #endif /* __XFS_MOUNT_H__ */ diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index ed4a2f8c1936..e525a6c477ff 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1052,12 +1052,18 @@ xfs_init_percpu_counters( if (error) goto free_fdblocks; - error = percpu_counter_init(&mp->m_frextents, 0, GFP_KERNEL); + error = percpu_counter_init(&mp->m_delalloc_rtextents, 0, GFP_KERNEL); if (error) goto free_delalloc; + error = percpu_counter_init(&mp->m_frextents, 0, GFP_KERNEL); + if (error) + goto free_delalloc_rt; + return 0; +free_delalloc_rt: + percpu_counter_destroy(&mp->m_delalloc_rtextents); free_delalloc: percpu_counter_destroy(&mp->m_delalloc_blks); free_fdblocks: @@ -1086,6 +1092,9 @@ xfs_destroy_percpu_counters( percpu_counter_destroy(&mp->m_icount); percpu_counter_destroy(&mp->m_ifree); percpu_counter_destroy(&mp->m_fdblocks); + ASSERT(xfs_is_shutdown(mp) || + percpu_counter_sum(&mp->m_delalloc_rtextents) == 0); + percpu_counter_destroy(&mp->m_delalloc_rtextents); ASSERT(xfs_is_shutdown(mp) || percpu_counter_sum(&mp->m_delalloc_blks) == 0); percpu_counter_destroy(&mp->m_delalloc_blks); -- cgit v1.2.3-59-g8ed1b From 727f8431638fdb5dc9ce9c81bdcc33fb416d45ee Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:16 +0200 Subject: xfs: look at m_frextents in xfs_iomap_prealloc_size for RT allocations Add a check for files on the RT subvolume and use m_frextents instead of m_fdblocks to adjust the preallocation size. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_iomap.c | 43 +++++++++++++++++++++++++++++++------------ 1 file changed, 31 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 4087af7f3c9f..bba5a0d87d03 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -28,6 +28,7 @@ #include "xfs_dquot.h" #include "xfs_reflink.h" #include "xfs_health.h" +#include "xfs_rtbitmap.h" #define XFS_ALLOC_ALIGN(mp, off) \ (((off) >> mp->m_allocsize_log) << mp->m_allocsize_log) @@ -404,6 +405,29 @@ xfs_quota_calc_throttle( } } +static int64_t +xfs_iomap_freesp( + struct percpu_counter *counter, + uint64_t low_space[XFS_LOWSP_MAX], + int *shift) +{ + int64_t freesp; + + freesp = percpu_counter_read_positive(counter); + if (freesp < low_space[XFS_LOWSP_5_PCNT]) { + *shift = 2; + if (freesp < low_space[XFS_LOWSP_4_PCNT]) + (*shift)++; + if (freesp < low_space[XFS_LOWSP_3_PCNT]) + (*shift)++; + if (freesp < low_space[XFS_LOWSP_2_PCNT]) + (*shift)++; + if (freesp < low_space[XFS_LOWSP_1_PCNT]) + (*shift)++; + } + return freesp; +} + /* * If we don't have a user specified preallocation size, dynamically increase * the preallocation size as the size of the file grows. Cap the maximum size @@ -486,18 +510,13 @@ xfs_iomap_prealloc_size( alloc_blocks = XFS_FILEOFF_MIN(roundup_pow_of_two(XFS_MAX_BMBT_EXTLEN), alloc_blocks); - freesp = percpu_counter_read_positive(&mp->m_fdblocks); - if (freesp < mp->m_low_space[XFS_LOWSP_5_PCNT]) { - shift = 2; - if (freesp < mp->m_low_space[XFS_LOWSP_4_PCNT]) - shift++; - if (freesp < mp->m_low_space[XFS_LOWSP_3_PCNT]) - shift++; - if (freesp < mp->m_low_space[XFS_LOWSP_2_PCNT]) - shift++; - if (freesp < mp->m_low_space[XFS_LOWSP_1_PCNT]) - shift++; - } + if (unlikely(XFS_IS_REALTIME_INODE(ip))) + freesp = xfs_rtx_to_rtb(mp, + xfs_iomap_freesp(&mp->m_frextents, + mp->m_low_rtexts, &shift)); + else + freesp = xfs_iomap_freesp(&mp->m_fdblocks, mp->m_low_space, + &shift); /* * Check each quota to cap the prealloc size, provide a shift value to -- cgit v1.2.3-59-g8ed1b From da2b9c3a8d2cbdeec3f13cebf4c6c86c13e1077e Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:17 +0200 Subject: xfs: rework splitting of indirect block reservations Move the check if we have enough indirect blocks and the stealing of the deleted extent blocks out of xfs_bmap_split_indlen and into the caller to prepare for handling delayed allocation of RT extents that can't easily be stolen. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 38 ++++++++++++++++---------------------- 1 file changed, 16 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index a2b715450ec2..fe33254ca390 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4831,31 +4831,17 @@ error0: * ores == 1). The number of stolen blocks is returned. The availability and * subsequent accounting of stolen blocks is the responsibility of the caller. */ -static xfs_filblks_t +static void xfs_bmap_split_indlen( xfs_filblks_t ores, /* original res. */ xfs_filblks_t *indlen1, /* ext1 worst indlen */ - xfs_filblks_t *indlen2, /* ext2 worst indlen */ - xfs_filblks_t avail) /* stealable blocks */ + xfs_filblks_t *indlen2) /* ext2 worst indlen */ { xfs_filblks_t len1 = *indlen1; xfs_filblks_t len2 = *indlen2; xfs_filblks_t nres = len1 + len2; /* new total res. */ - xfs_filblks_t stolen = 0; xfs_filblks_t resfactor; - /* - * Steal as many blocks as we can to try and satisfy the worst case - * indlen for both new extents. - */ - if (ores < nres && avail) - stolen = XFS_FILBLKS_MIN(nres - ores, avail); - ores += stolen; - - /* nothing else to do if we've satisfied the new reservation */ - if (ores >= nres) - return stolen; - /* * We can't meet the total required reservation for the two extents. * Calculate the percent of the overall shortage between both extents @@ -4900,8 +4886,6 @@ xfs_bmap_split_indlen( *indlen1 = len1; *indlen2 = len2; - - return stolen; } int @@ -4917,7 +4901,7 @@ xfs_bmap_del_extent_delay( struct xfs_bmbt_irec new; int64_t da_old, da_new, da_diff = 0; xfs_fileoff_t del_endoff, got_endoff; - xfs_filblks_t got_indlen, new_indlen, stolen; + xfs_filblks_t got_indlen, new_indlen, stolen = 0; uint32_t state = xfs_bmap_fork_to_state(whichfork); uint64_t fdblocks; int error = 0; @@ -4996,8 +4980,19 @@ xfs_bmap_del_extent_delay( new_indlen = xfs_bmap_worst_indlen(ip, new.br_blockcount); WARN_ON_ONCE(!got_indlen || !new_indlen); - stolen = xfs_bmap_split_indlen(da_old, &got_indlen, &new_indlen, - del->br_blockcount); + /* + * Steal as many blocks as we can to try and satisfy the worst + * case indlen for both new extents. + */ + da_new = got_indlen + new_indlen; + if (da_new > da_old) { + stolen = XFS_FILBLKS_MIN(da_new - da_old, + del->br_blockcount); + da_old += stolen; + } + if (da_new > da_old) + xfs_bmap_split_indlen(da_old, &got_indlen, &new_indlen); + da_new = got_indlen + new_indlen; got->br_startblock = nullstartblock((int)got_indlen); @@ -5009,7 +5004,6 @@ xfs_bmap_del_extent_delay( xfs_iext_next(ifp, icur); xfs_iext_insert(ip, icur, &new, state); - da_new = got_indlen + new_indlen - stolen; del->br_blockcount -= stolen; break; } -- cgit v1.2.3-59-g8ed1b From bd1753d8c42b6bd5d9a81c81d1ce6e3affe3a59f Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:18 +0200 Subject: xfs: stop the steal (of data blocks for RT indirect blocks) When xfs_bmap_del_extent_delay has to split an indirect block it tries to steal blocks from the the part that gets unmapped to increase the indirect block reservation that now needs to cover for two extents instead of one. This works perfectly fine on the data device, where the data and indirect blocks come from the same pool. It has no chance of working when the inode sits on the RT device. To support re-enabling delalloc for inodes on the RT device, make this behavior conditional on not being for rt extents. Note that split of delalloc extents should only happen on writeback failure, as for other kinds of hole punching we first write back all data and thus convert the delalloc reservations covering the hole to a real allocation. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index fe33254ca390..8a1446e025e0 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4983,9 +4983,14 @@ xfs_bmap_del_extent_delay( /* * Steal as many blocks as we can to try and satisfy the worst * case indlen for both new extents. + * + * However, we can't just steal reservations from the data + * blocks if this is an RT inodes as the data and metadata + * blocks come from different pools. We'll have to live with + * under-filled indirect reservation in this case. */ da_new = got_indlen + new_indlen; - if (da_new > da_old) { + if (da_new > da_old && !isrt) { stolen = XFS_FILBLKS_MIN(da_new - da_old, del->br_blockcount); da_old += stolen; -- cgit v1.2.3-59-g8ed1b From 6a94b1acda7e7262418e23f906c12a2b08b69d12 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 13:20:19 +0200 Subject: xfs: reinstate delalloc for RT inodes (if sb_rextsize == 1) Commit aff3a9edb708 ("xfs: Use preallocation for inodes with extsz hints") disabled delayed allocation for all inodes with extent size hints due a data exposure problem. It turns out we fixed this data exposure problem since by always creating unwritten extents for delalloc conversions due to more data exposure problems, but the writeback path doesn't actually support extent size hints when converting delalloc these days, which probably isn't a problem given that people using the hints know what they get. However due to the way how xfs_get_extsz_hint is implemented, it always claims an extent size hint for RT inodes even if the RT extent size is a single FSB. Due to that the above commit effectively disabled delalloc support for RT inodes. Switch xfs_get_extsz_hint to return 0 for this case and work around that in a few places to reinstate delalloc support for RT inodes on file systems with an sb_rextsize of 1. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_inode.c | 3 ++- fs/xfs/xfs_iomap.c | 2 -- fs/xfs/xfs_iops.c | 2 +- fs/xfs/xfs_rtalloc.c | 2 ++ 4 files changed, 5 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 2aec7ab59aeb..3c843223b4ed 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -58,7 +58,8 @@ xfs_get_extsz_hint( return 0; if ((ip->i_diflags & XFS_DIFLAG_EXTSIZE) && ip->i_extsize) return ip->i_extsize; - if (XFS_IS_REALTIME_INODE(ip)) + if (XFS_IS_REALTIME_INODE(ip) && + ip->i_mount->m_sb.sb_rextsize > 1) return ip->i_mount->m_sb.sb_rextsize; return 0; } diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index bba5a0d87d03..9ce0f6b9df93 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -1001,8 +1001,6 @@ xfs_buffered_write_iomap_begin( return xfs_direct_write_iomap_begin(inode, offset, count, flags, iomap, srcmap); - ASSERT(!XFS_IS_REALTIME_INODE(ip)); - error = xfs_qm_dqattach(ip); if (error) return error; diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 7f0c840f0fd2..ad76704ab133 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -525,7 +525,7 @@ xfs_stat_blksize( * always return the realtime extent size. */ if (XFS_IS_REALTIME_INODE(ip)) - return XFS_FSB_TO_B(mp, xfs_get_extsz_hint(ip)); + return XFS_FSB_TO_B(mp, xfs_get_extsz_hint(ip) ? : 1); /* * Allow large block sizes to be reported to userspace programs if the diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 86f928d30fed..b476a876478d 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -1341,6 +1341,8 @@ xfs_bmap_rtalloc( int error; align = xfs_get_extsz_hint(ap->ip); + if (!align) + align = 1; retry: error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev, align, 1, ap->eof, 0, -- cgit v1.2.3-59-g8ed1b From aef8acd79f363ced098cd3bcde0a5978a52607ad Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Mon, 22 Apr 2024 09:40:08 +0300 Subject: fuse: verify zero padding in fuse_backing_map To allow us extending the interface in the future. Fixes: 44350256ab94 ("fuse: implement ioctls to manage backing files") Signed-off-by: Amir Goldstein Signed-off-by: Miklos Szeredi --- fs/fuse/passthrough.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fuse/passthrough.c b/fs/fuse/passthrough.c index 1567f0323858..9666d13884ce 100644 --- a/fs/fuse/passthrough.c +++ b/fs/fuse/passthrough.c @@ -225,7 +225,7 @@ int fuse_backing_open(struct fuse_conn *fc, struct fuse_backing_map *map) goto out; res = -EINVAL; - if (map->flags) + if (map->flags || map->padding) goto out; file = fget(map->fd); -- cgit v1.2.3-59-g8ed1b From d05dcfdf5e1659b2949d13060284eff3888b644e Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Mon, 15 Apr 2024 20:24:37 +0000 Subject: fs/9p: mitigate inode collisions Detect and mitigate inode collsions that now occur since we fixed 9p generating duplicate inode structures. Underlying cause of these appears to be a race condition between reuse of inode numbers in underlying file system and cleanup of inode numbers in the client. Enabling caching makes this much more likely to happen as it increases cleanup latency due to writebacks. Reported-by: Kent Overstreet Signed-off-by: Eric Van Hensbergen --- fs/9p/v9fs.h | 11 ++++++----- fs/9p/vfs_inode.c | 37 +++++++++++++++++++++++++++++-------- fs/9p/vfs_inode_dotl.c | 28 ++++++++++++++++++++-------- fs/9p/vfs_super.c | 2 +- 4 files changed, 56 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h index 9defa12208f9..1775fcc7f0e8 100644 --- a/fs/9p/v9fs.h +++ b/fs/9p/v9fs.h @@ -179,13 +179,14 @@ extern int v9fs_vfs_rename(struct mnt_idmap *idmap, struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags); -extern struct inode *v9fs_fid_iget(struct super_block *sb, struct p9_fid *fid); +extern struct inode *v9fs_fid_iget(struct super_block *sb, struct p9_fid *fid, + bool new); extern const struct inode_operations v9fs_dir_inode_operations_dotl; extern const struct inode_operations v9fs_file_inode_operations_dotl; extern const struct inode_operations v9fs_symlink_inode_operations_dotl; extern const struct netfs_request_ops v9fs_req_ops; extern struct inode *v9fs_fid_iget_dotl(struct super_block *sb, - struct p9_fid *fid); + struct p9_fid *fid, bool new); /* other default globals */ #define V9FS_PORT 564 @@ -224,12 +225,12 @@ static inline int v9fs_proto_dotl(struct v9fs_session_info *v9ses) */ static inline struct inode * v9fs_get_inode_from_fid(struct v9fs_session_info *v9ses, struct p9_fid *fid, - struct super_block *sb) + struct super_block *sb, bool new) { if (v9fs_proto_dotl(v9ses)) - return v9fs_fid_iget_dotl(sb, fid); + return v9fs_fid_iget_dotl(sb, fid, new); else - return v9fs_fid_iget(sb, fid); + return v9fs_fid_iget(sb, fid, new); } #endif diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 47bd77199e20..7a3308d77606 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -364,7 +364,8 @@ void v9fs_evict_inode(struct inode *inode) clear_inode(inode); } -struct inode *v9fs_fid_iget(struct super_block *sb, struct p9_fid *fid) +struct inode * +v9fs_fid_iget(struct super_block *sb, struct p9_fid *fid, bool new) { dev_t rdev; int retval; @@ -376,8 +377,18 @@ struct inode *v9fs_fid_iget(struct super_block *sb, struct p9_fid *fid) inode = iget_locked(sb, QID2INO(&fid->qid)); if (unlikely(!inode)) return ERR_PTR(-ENOMEM); - if (!(inode->i_state & I_NEW)) - return inode; + if (!(inode->i_state & I_NEW)) { + if (!new) { + goto done; + } else { + p9_debug(P9_DEBUG_VFS, "WARNING: Inode collision %ld\n", + inode->i_ino); + iput(inode); + remove_inode_hash(inode); + inode = iget_locked(sb, QID2INO(&fid->qid)); + WARN_ON(!(inode->i_state & I_NEW)); + } + } /* * initialize the inode with the stat info @@ -401,11 +412,11 @@ struct inode *v9fs_fid_iget(struct super_block *sb, struct p9_fid *fid) v9fs_set_netfs_context(inode); v9fs_cache_inode_get_cookie(inode); unlock_new_inode(inode); +done: return inode; error: iget_failed(inode); return ERR_PTR(retval); - } /** @@ -437,8 +448,15 @@ static int v9fs_at_to_dotl_flags(int flags) */ static void v9fs_dec_count(struct inode *inode) { - if (!S_ISDIR(inode->i_mode) || inode->i_nlink > 2) - drop_nlink(inode); + if (!S_ISDIR(inode->i_mode) || inode->i_nlink > 2) { + if (inode->i_nlink) { + drop_nlink(inode); + } else { + p9_debug(P9_DEBUG_VFS, + "WARNING: unexpected i_nlink zero %d inode %ld\n", + inode->i_nlink, inode->i_ino); + } + } } /** @@ -489,6 +507,9 @@ static int v9fs_remove(struct inode *dir, struct dentry *dentry, int flags) } else v9fs_dec_count(inode); + if (inode->i_nlink <= 0) /* no more refs unhash it */ + remove_inode_hash(inode); + v9fs_invalidate_inode_attr(inode); v9fs_invalidate_inode_attr(dir); @@ -554,7 +575,7 @@ v9fs_create(struct v9fs_session_info *v9ses, struct inode *dir, /* * instantiate inode and assign the unopened fid to the dentry */ - inode = v9fs_get_inode_from_fid(v9ses, fid, dir->i_sb); + inode = v9fs_get_inode_from_fid(v9ses, fid, dir->i_sb, true); if (IS_ERR(inode)) { err = PTR_ERR(inode); p9_debug(P9_DEBUG_VFS, @@ -683,7 +704,7 @@ struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, else if (IS_ERR(fid)) inode = ERR_CAST(fid); else - inode = v9fs_get_inode_from_fid(v9ses, fid, dir->i_sb); + inode = v9fs_get_inode_from_fid(v9ses, fid, dir->i_sb, false); /* * If we had a rename on the server and a parallel lookup * for the new name, then make sure we instantiate with diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 55dde186041a..c61b97bd13b9 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -52,7 +52,10 @@ static kgid_t v9fs_get_fsgid_for_create(struct inode *dir_inode) return current_fsgid(); } -struct inode *v9fs_fid_iget_dotl(struct super_block *sb, struct p9_fid *fid) + + +struct inode * +v9fs_fid_iget_dotl(struct super_block *sb, struct p9_fid *fid, bool new) { int retval; struct inode *inode; @@ -62,8 +65,18 @@ struct inode *v9fs_fid_iget_dotl(struct super_block *sb, struct p9_fid *fid) inode = iget_locked(sb, QID2INO(&fid->qid)); if (unlikely(!inode)) return ERR_PTR(-ENOMEM); - if (!(inode->i_state & I_NEW)) - return inode; + if (!(inode->i_state & I_NEW)) { + if (!new) { + goto done; + } else { /* deal with race condition in inode number reuse */ + p9_debug(P9_DEBUG_ERROR, "WARNING: Inode collision %lx\n", + inode->i_ino); + iput(inode); + remove_inode_hash(inode); + inode = iget_locked(sb, QID2INO(&fid->qid)); + WARN_ON(!(inode->i_state & I_NEW)); + } + } /* * initialize the inode with the stat info @@ -90,12 +103,11 @@ struct inode *v9fs_fid_iget_dotl(struct super_block *sb, struct p9_fid *fid) goto error; unlock_new_inode(inode); - +done: return inode; error: iget_failed(inode); return ERR_PTR(retval); - } struct dotl_openflag_map { @@ -247,7 +259,7 @@ v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry, p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); goto out; } - inode = v9fs_fid_iget_dotl(dir->i_sb, fid); + inode = v9fs_fid_iget_dotl(dir->i_sb, fid, true); if (IS_ERR(inode)) { err = PTR_ERR(inode); p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", err); @@ -340,7 +352,7 @@ static int v9fs_vfs_mkdir_dotl(struct mnt_idmap *idmap, } /* instantiate inode and assign the unopened fid to the dentry */ - inode = v9fs_fid_iget_dotl(dir->i_sb, fid); + inode = v9fs_fid_iget_dotl(dir->i_sb, fid, true); if (IS_ERR(inode)) { err = PTR_ERR(inode); p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", @@ -776,7 +788,7 @@ v9fs_vfs_mknod_dotl(struct mnt_idmap *idmap, struct inode *dir, err); goto error; } - inode = v9fs_fid_iget_dotl(dir->i_sb, fid); + inode = v9fs_fid_iget_dotl(dir->i_sb, fid, true); if (IS_ERR(inode)) { err = PTR_ERR(inode); p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index 55e67e36ae68..f52fdf42945c 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -139,7 +139,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, else sb->s_d_op = &v9fs_dentry_operations; - inode = v9fs_get_inode_from_fid(v9ses, fid, sb); + inode = v9fs_get_inode_from_fid(v9ses, fid, sb, true); if (IS_ERR(inode)) { retval = PTR_ERR(inode); goto release_sb; -- cgit v1.2.3-59-g8ed1b From 77d8aa79ecfb209308e0644c02f655122b31def7 Mon Sep 17 00:00:00 2001 From: Takayuki Nagata Date: Mon, 15 Apr 2024 16:47:49 +0900 Subject: cifs: reinstate original behavior again for forceuid/forcegid forceuid/forcegid should be enabled by default when uid=/gid= options are specified, but commit 24e0a1eff9e2 ("cifs: switch to new mount api") changed the behavior. Due to the change, a mounted share does not show intentional uid/gid for files and directories even though uid=/gid= options are specified since forceuid/forcegid are not enabled. This patch reinstates original behavior that overrides uid/gid with specified uid/gid by the options. Fixes: 24e0a1eff9e2 ("cifs: switch to new mount api") Signed-off-by: Takayuki Nagata Acked-by: Paulo Alcantara (Red Hat) Acked-by: Ronnie Sahlberg Acked-by: Tom Talpey Signed-off-by: Steve French --- fs/smb/client/fs_context.c | 12 ++++++++++++ fs/smb/client/fs_context.h | 2 ++ 2 files changed, 14 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c index 6c727d8c31e8..3bbac925d076 100644 --- a/fs/smb/client/fs_context.c +++ b/fs/smb/client/fs_context.c @@ -748,6 +748,16 @@ static int smb3_fs_context_validate(struct fs_context *fc) /* set the port that we got earlier */ cifs_set_port((struct sockaddr *)&ctx->dstaddr, ctx->port); + if (ctx->uid_specified && !ctx->forceuid_specified) { + ctx->override_uid = 1; + pr_notice("enabling forceuid mount option implicitly because uid= option is specified\n"); + } + + if (ctx->gid_specified && !ctx->forcegid_specified) { + ctx->override_gid = 1; + pr_notice("enabling forcegid mount option implicitly because gid= option is specified\n"); + } + if (ctx->override_uid && !ctx->uid_specified) { ctx->override_uid = 0; pr_notice("ignoring forceuid mount option specified with no uid= option\n"); @@ -1019,12 +1029,14 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, ctx->override_uid = 0; else ctx->override_uid = 1; + ctx->forceuid_specified = true; break; case Opt_forcegid: if (result.negated) ctx->override_gid = 0; else ctx->override_gid = 1; + ctx->forcegid_specified = true; break; case Opt_perm: if (result.negated) diff --git a/fs/smb/client/fs_context.h b/fs/smb/client/fs_context.h index a947bddeba27..cf577ec0dd0a 100644 --- a/fs/smb/client/fs_context.h +++ b/fs/smb/client/fs_context.h @@ -165,6 +165,8 @@ enum cifs_param { }; struct smb3_fs_context { + bool forceuid_specified; + bool forcegid_specified; bool uid_specified; bool cruid_specified; bool gid_specified; -- cgit v1.2.3-59-g8ed1b From 110b24eb1a749bea3440f3ca2ff890a26179050a Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Wed, 17 Apr 2024 10:33:06 +0300 Subject: fs/ntfs3: Taking DOS names into account during link counting When counting and checking hard links in an ntfs file record, struct MFT_REC { struct NTFS_RECORD_HEADER rhdr; // 'FILE' __le16 seq; // 0x10: Sequence number for this record. >> __le16 hard_links; // 0x12: The number of hard links to record. __le16 attr_off; // 0x14: Offset to attributes. ... the ntfs3 driver ignored short names (DOS names), causing the link count to be reduced by 1 and messages to be output to dmesg. For Windows, such a situation is a minor error, meaning chkdsk does not report errors on such a volume, and in the case of using the /f switch, it silently corrects them, reporting that no errors were found. This does not affect the consistency of the file system. Nevertheless, the behavior in the ntfs3 driver is incorrect and changes the content of the file system. This patch should fix that. PS: most likely, there has been a confusion of concepts MFT_REC::hard_links and inode::__i_nlink. Fixes: 82cae269cfa95 ("fs/ntfs3: Add initialization of super block") Signed-off-by: Konstantin Komarov Cc: stable@vger.kernel.org --- fs/ntfs3/inode.c | 7 ++++--- fs/ntfs3/record.c | 11 ++--------- 2 files changed, 6 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c index eb7a8c9fba01..05f169018c4e 100644 --- a/fs/ntfs3/inode.c +++ b/fs/ntfs3/inode.c @@ -37,7 +37,7 @@ static struct inode *ntfs_read_mft(struct inode *inode, bool is_dir; unsigned long ino = inode->i_ino; u32 rp_fa = 0, asize, t32; - u16 roff, rsize, names = 0; + u16 roff, rsize, names = 0, links = 0; const struct ATTR_FILE_NAME *fname = NULL; const struct INDEX_ROOT *root; struct REPARSE_DATA_BUFFER rp; // 0x18 bytes @@ -200,11 +200,12 @@ next_attr: rsize < SIZEOF_ATTRIBUTE_FILENAME) goto out; + names += 1; fname = Add2Ptr(attr, roff); if (fname->type == FILE_NAME_DOS) goto next_attr; - names += 1; + links += 1; if (name && name->len == fname->name_len && !ntfs_cmp_names_cpu(name, (struct le_str *)&fname->name_len, NULL, false)) @@ -429,7 +430,7 @@ end_enum: ni->mi.dirty = true; } - set_nlink(inode, names); + set_nlink(inode, links); if (S_ISDIR(mode)) { ni->std_fa |= FILE_ATTRIBUTE_DIRECTORY; diff --git a/fs/ntfs3/record.c b/fs/ntfs3/record.c index 6aa3a9d44df1..6c76503edc20 100644 --- a/fs/ntfs3/record.c +++ b/fs/ntfs3/record.c @@ -534,16 +534,9 @@ bool mi_remove_attr(struct ntfs_inode *ni, struct mft_inode *mi, if (aoff + asize > used) return false; - if (ni && is_attr_indexed(attr)) { + if (ni && is_attr_indexed(attr) && attr->type == ATTR_NAME) { u16 links = le16_to_cpu(ni->mi.mrec->hard_links); - struct ATTR_FILE_NAME *fname = - attr->type != ATTR_NAME ? - NULL : - resident_data_ex(attr, - SIZEOF_ATTRIBUTE_FILENAME); - if (fname && fname->type == FILE_NAME_DOS) { - /* Do not decrease links count deleting DOS name. */ - } else if (!links) { + if (!links) { /* minor error. Not critical. */ } else { ni->mi.mrec->hard_links = cpu_to_le16(links - 1); -- cgit v1.2.3-59-g8ed1b From a8948b5450e7c65a3a34ebf4ccfcebc19335d4fb Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Wed, 3 Apr 2024 10:08:04 +0300 Subject: fs/ntfs3: Remove max link count info display during driver init Removes the output of this purely informational message from the kernel buffer: "ntfs3: Max link count 4000" Signed-off-by: Konstantin Komarov Cc: stable@vger.kernel.org --- fs/ntfs3/super.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c index 9df7c20d066f..ac4722011140 100644 --- a/fs/ntfs3/super.c +++ b/fs/ntfs3/super.c @@ -1804,8 +1804,6 @@ static int __init init_ntfs_fs(void) { int err; - pr_info("ntfs3: Max link count %u\n", NTFS_LINK_MAX); - if (IS_ENABLED(CONFIG_NTFS3_FS_POSIX_ACL)) pr_info("ntfs3: Enabled Linux POSIX ACLs support\n"); if (IS_ENABLED(CONFIG_NTFS3_64BIT_CLUSTER)) -- cgit v1.2.3-59-g8ed1b From b0a5ddee56a3683bdd6400f763158efea78cee3c Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 16 Apr 2024 09:39:48 +0300 Subject: fs/ntfs3: Missed le32_to_cpu conversion NTFS data structure fields are stored in little-endian, it is necessary to take this into account when working on big-endian architectures. Fixes: 1b7dd28e14c47("fs/ntfs3: Correct function is_rst_area_valid") Signed-off-by: Konstantin Komarov --- fs/ntfs3/fslog.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ntfs3/fslog.c b/fs/ntfs3/fslog.c index 855519713bf7..d9d08823de62 100644 --- a/fs/ntfs3/fslog.c +++ b/fs/ntfs3/fslog.c @@ -517,7 +517,7 @@ static inline bool is_rst_area_valid(const struct RESTART_HDR *rhdr) seq_bits -= 1; } - if (seq_bits != ra->seq_num_bits) + if (seq_bits != le32_to_cpu(ra->seq_num_bits)) return false; /* The log page data offset and record header length must be quad-aligned. */ -- cgit v1.2.3-59-g8ed1b From 1cd6c96219c429ebcfa8e79a865277376c563803 Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 16 Apr 2024 09:54:34 +0300 Subject: fs/ntfs3: Check 'folio' pointer for NULL It can be NULL if bmap is called. Fixes: 82cae269cfa95 ("fs/ntfs3: Add initialization of super block") Signed-off-by: Konstantin Komarov --- fs/ntfs3/inode.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c index 05f169018c4e..502a527e51cd 100644 --- a/fs/ntfs3/inode.c +++ b/fs/ntfs3/inode.c @@ -571,13 +571,18 @@ static noinline int ntfs_get_block_vbo(struct inode *inode, u64 vbo, clear_buffer_uptodate(bh); if (is_resident(ni)) { - ni_lock(ni); - err = attr_data_read_resident(ni, &folio->page); - ni_unlock(ni); - - if (!err) - set_buffer_uptodate(bh); + bh->b_blocknr = RESIDENT_LCN; bh->b_size = block_size; + if (!folio) { + err = 0; + } else { + ni_lock(ni); + err = attr_data_read_resident(ni, &folio->page); + ni_unlock(ni); + + if (!err) + set_buffer_uptodate(bh); + } return err; } -- cgit v1.2.3-59-g8ed1b From e931f6b630ffb22d66caab202a52aa8cbb10c649 Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 16 Apr 2024 09:45:09 +0300 Subject: fs/ntfs3: Use 64 bit variable to avoid 32 bit overflow For example, in the expression: vbo = 2 * vbo + skip Fixes: b46acd6a6a627 ("fs/ntfs3: Add NTFS journal") Signed-off-by: Konstantin Komarov --- fs/ntfs3/fslog.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ntfs3/fslog.c b/fs/ntfs3/fslog.c index d9d08823de62..d7807d255dfe 100644 --- a/fs/ntfs3/fslog.c +++ b/fs/ntfs3/fslog.c @@ -1184,7 +1184,8 @@ out: static int log_read_rst(struct ntfs_log *log, bool first, struct restart_info *info) { - u32 skip, vbo; + u32 skip; + u64 vbo; struct RESTART_HDR *r_page = NULL; /* Determine which restart area we are looking for. */ -- cgit v1.2.3-59-g8ed1b From 1997cdc3e727526aa5d84b32f7cbb3f56459b7ef Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 16 Apr 2024 09:43:58 +0300 Subject: fs/ntfs3: Use variable length array instead of fixed size Should fix smatch warning: ntfs_set_label() error: __builtin_memcpy() 'uni->name' too small (20 vs 256) Fixes: 4534a70b7056f ("fs/ntfs3: Add headers and misc files") Reported-by: kernel test robot Reported-by: Dan Carpenter Closes: https://lore.kernel.org/r/202401091421.3RJ24Mn3-lkp@intel.com/ Signed-off-by: Konstantin Komarov --- fs/ntfs3/ntfs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ntfs3/ntfs.h b/fs/ntfs3/ntfs.h index 9c7478150a03..3d6143c7abc0 100644 --- a/fs/ntfs3/ntfs.h +++ b/fs/ntfs3/ntfs.h @@ -59,7 +59,7 @@ struct GUID { struct cpu_str { u8 len; u8 unused; - u16 name[10]; + u16 name[]; }; struct le_str { -- cgit v1.2.3-59-g8ed1b From c935c66878867dc87c36c36b21d35d7e7f08adec Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 16 Apr 2024 09:52:47 +0300 Subject: fs/ntfs3: Redesign ntfs_create_inode to return error code instead of inode As Al Viro correctly pointed out, there is no need to return the whole structure to check the error. https://lore.kernel.org/ntfs3/20240322023515.GK538574@ZenIV/ Acked-by: Al Viro Signed-off-by: Konstantin Komarov --- fs/ntfs3/inode.c | 22 +++++++++++----------- fs/ntfs3/namei.c | 31 ++++++++----------------------- fs/ntfs3/ntfs_fs.h | 9 ++++----- 3 files changed, 23 insertions(+), 39 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c index 502a527e51cd..8fdcf37b3186 100644 --- a/fs/ntfs3/inode.c +++ b/fs/ntfs3/inode.c @@ -1216,11 +1216,10 @@ out: * * NOTE: if fnd != NULL (ntfs_atomic_open) then @dir is locked */ -struct inode *ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, - struct dentry *dentry, - const struct cpu_str *uni, umode_t mode, - dev_t dev, const char *symname, u32 size, - struct ntfs_fnd *fnd) +int ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const struct cpu_str *uni, + umode_t mode, dev_t dev, const char *symname, u32 size, + struct ntfs_fnd *fnd) { int err; struct super_block *sb = dir->i_sb; @@ -1245,6 +1244,9 @@ struct inode *ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, struct REPARSE_DATA_BUFFER *rp = NULL; bool rp_inserted = false; + /* New file will be resident or non resident. */ + const bool new_file_resident = 1; + if (!fnd) ni_lock_dir(dir_ni); @@ -1484,7 +1486,7 @@ struct inode *ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, attr->size = cpu_to_le32(SIZEOF_RESIDENT); attr->name_off = SIZEOF_RESIDENT_LE; attr->res.data_off = SIZEOF_RESIDENT_LE; - } else if (S_ISREG(mode)) { + } else if (!new_file_resident && S_ISREG(mode)) { /* * Regular file. Create empty non resident data attribute. */ @@ -1721,12 +1723,10 @@ out1: if (!fnd) ni_unlock(dir_ni); - if (err) - return ERR_PTR(err); - - unlock_new_inode(inode); + if (!err) + unlock_new_inode(inode); - return inode; + return err; } int ntfs_link_inode(struct inode *inode, struct dentry *dentry) diff --git a/fs/ntfs3/namei.c b/fs/ntfs3/namei.c index edb6a7141246..71498421ce60 100644 --- a/fs/ntfs3/namei.c +++ b/fs/ntfs3/namei.c @@ -107,12 +107,8 @@ static struct dentry *ntfs_lookup(struct inode *dir, struct dentry *dentry, static int ntfs_create(struct mnt_idmap *idmap, struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) { - struct inode *inode; - - inode = ntfs_create_inode(idmap, dir, dentry, NULL, S_IFREG | mode, 0, - NULL, 0, NULL); - - return IS_ERR(inode) ? PTR_ERR(inode) : 0; + return ntfs_create_inode(idmap, dir, dentry, NULL, S_IFREG | mode, 0, + NULL, 0, NULL); } /* @@ -123,12 +119,8 @@ static int ntfs_create(struct mnt_idmap *idmap, struct inode *dir, static int ntfs_mknod(struct mnt_idmap *idmap, struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { - struct inode *inode; - - inode = ntfs_create_inode(idmap, dir, dentry, NULL, mode, rdev, NULL, 0, - NULL); - - return IS_ERR(inode) ? PTR_ERR(inode) : 0; + return ntfs_create_inode(idmap, dir, dentry, NULL, mode, rdev, NULL, 0, + NULL); } /* @@ -200,15 +192,12 @@ static int ntfs_symlink(struct mnt_idmap *idmap, struct inode *dir, struct dentry *dentry, const char *symname) { u32 size = strlen(symname); - struct inode *inode; if (unlikely(ntfs3_forced_shutdown(dir->i_sb))) return -EIO; - inode = ntfs_create_inode(idmap, dir, dentry, NULL, S_IFLNK | 0777, 0, - symname, size, NULL); - - return IS_ERR(inode) ? PTR_ERR(inode) : 0; + return ntfs_create_inode(idmap, dir, dentry, NULL, S_IFLNK | 0777, 0, + symname, size, NULL); } /* @@ -217,12 +206,8 @@ static int ntfs_symlink(struct mnt_idmap *idmap, struct inode *dir, static int ntfs_mkdir(struct mnt_idmap *idmap, struct inode *dir, struct dentry *dentry, umode_t mode) { - struct inode *inode; - - inode = ntfs_create_inode(idmap, dir, dentry, NULL, S_IFDIR | mode, 0, - NULL, 0, NULL); - - return IS_ERR(inode) ? PTR_ERR(inode) : 0; + return ntfs_create_inode(idmap, dir, dentry, NULL, S_IFDIR | mode, 0, + NULL, 0, NULL); } /* diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h index 79356fd29a14..3db6a61f61dc 100644 --- a/fs/ntfs3/ntfs_fs.h +++ b/fs/ntfs3/ntfs_fs.h @@ -714,11 +714,10 @@ int ntfs_sync_inode(struct inode *inode); int ntfs_flush_inodes(struct super_block *sb, struct inode *i1, struct inode *i2); int inode_write_data(struct inode *inode, const void *data, size_t bytes); -struct inode *ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, - struct dentry *dentry, - const struct cpu_str *uni, umode_t mode, - dev_t dev, const char *symname, u32 size, - struct ntfs_fnd *fnd); +int ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const struct cpu_str *uni, + umode_t mode, dev_t dev, const char *symname, u32 size, + struct ntfs_fnd *fnd); int ntfs_link_inode(struct inode *inode, struct dentry *dentry); int ntfs_unlink_inode(struct inode *dir, const struct dentry *dentry); void ntfs_evict_inode(struct inode *inode); -- cgit v1.2.3-59-g8ed1b From 40bb3c590582f488ec1ff8c31b7fc806e5732f42 Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 16 Apr 2024 10:08:18 +0300 Subject: fs/ntfs3: Always make file nonresident on fallocate call xfstest 438 is starting to pass with this change. Signed-off-by: Konstantin Komarov --- fs/ntfs3/attrib.c | 32 ++++++++++++++++++++++++++++++++ fs/ntfs3/file.c | 9 +++++++++ fs/ntfs3/ntfs_fs.h | 1 + 3 files changed, 42 insertions(+) (limited to 'fs') diff --git a/fs/ntfs3/attrib.c b/fs/ntfs3/attrib.c index 7aadf5010999..8e6bcdf99770 100644 --- a/fs/ntfs3/attrib.c +++ b/fs/ntfs3/attrib.c @@ -2558,3 +2558,35 @@ undo_insert_range: goto out; } + +/* + * attr_force_nonresident + * + * Convert default data attribute into non resident form. + */ +int attr_force_nonresident(struct ntfs_inode *ni) +{ + int err; + struct ATTRIB *attr; + struct ATTR_LIST_ENTRY *le = NULL; + struct mft_inode *mi; + + attr = ni_find_attr(ni, NULL, &le, ATTR_DATA, NULL, 0, NULL, &mi); + if (!attr) { + ntfs_bad_inode(&ni->vfs_inode, "no data attribute"); + return -ENOENT; + } + + if (attr->non_res) { + /* Already non resident. */ + return 0; + } + + down_write(&ni->file.run_lock); + err = attr_make_nonresident(ni, attr, le, mi, + le32_to_cpu(attr->res.data_size), + &ni->file.run, &attr, NULL); + up_write(&ni->file.run_lock); + + return err; +} diff --git a/fs/ntfs3/file.c b/fs/ntfs3/file.c index 5418662c80d8..fce8ea098d60 100644 --- a/fs/ntfs3/file.c +++ b/fs/ntfs3/file.c @@ -578,6 +578,15 @@ static long ntfs_fallocate(struct file *file, int mode, loff_t vbo, loff_t len) /* Check new size. */ u8 cluster_bits = sbi->cluster_bits; + /* Be sure file is non resident. */ + if (is_resident(ni)) { + ni_lock(ni); + err = attr_force_nonresident(ni); + ni_unlock(ni); + if (err) + goto out; + } + /* generic/213: expected -ENOSPC instead of -EFBIG. */ if (!is_supported_holes) { loff_t to_alloc = new_size - inode_get_bytes(inode); diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h index 3db6a61f61dc..00dec0ec5648 100644 --- a/fs/ntfs3/ntfs_fs.h +++ b/fs/ntfs3/ntfs_fs.h @@ -452,6 +452,7 @@ int attr_allocate_frame(struct ntfs_inode *ni, CLST frame, size_t compr_size, int attr_collapse_range(struct ntfs_inode *ni, u64 vbo, u64 bytes); int attr_insert_range(struct ntfs_inode *ni, u64 vbo, u64 bytes); int attr_punch_hole(struct ntfs_inode *ni, u64 vbo, u64 bytes, u32 *frame_size); +int attr_force_nonresident(struct ntfs_inode *ni); /* Functions from attrlist.c */ void al_destroy(struct ntfs_inode *ni); -- cgit v1.2.3-59-g8ed1b From 24f6f5020b0b2c89c2cba5ec224547be95f753ee Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Mon, 22 Apr 2024 17:18:51 +0300 Subject: fs/ntfs3: Mark volume as dirty if xattr is broken Mark a volume as corrupted if the name length exceeds the space occupied by ea. Signed-off-by: Konstantin Komarov --- fs/ntfs3/xattr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ntfs3/xattr.c b/fs/ntfs3/xattr.c index 53e7d1fa036a..73785dece7a7 100644 --- a/fs/ntfs3/xattr.c +++ b/fs/ntfs3/xattr.c @@ -219,8 +219,11 @@ static ssize_t ntfs_list_ea(struct ntfs_inode *ni, char *buffer, if (!ea->name_len) break; - if (ea->name_len > ea_size) + if (ea->name_len > ea_size) { + ntfs_set_state(ni->mi.sbi, NTFS_DIRTY_ERROR); + err = -EINVAL; /* corrupted fs */ break; + } if (buffer) { /* Check if we can use field ea->name */ -- cgit v1.2.3-59-g8ed1b From d55f90e9b243faa5bcd5c8a323a8f43040500106 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Tue, 16 Apr 2024 12:08:51 +0200 Subject: ntfs3: enforce read-only when used as legacy ntfs driver Ensure that ntfs3 is mounted read-only when it is used to provide the legacy ntfs driver. Signed-off-by: Christian Brauner --- fs/ntfs3/ntfs_fs.h | 2 ++ fs/ntfs3/super.c | 36 ++++++++++++++++++++++++++++++++---- 2 files changed, 34 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h index 79356fd29a14..184c8bc76b92 100644 --- a/fs/ntfs3/ntfs_fs.h +++ b/fs/ntfs3/ntfs_fs.h @@ -1154,4 +1154,6 @@ static inline void le64_sub_cpu(__le64 *var, u64 val) *var = cpu_to_le64(le64_to_cpu(*var) - val); } +bool is_legacy_ntfs(struct super_block *sb); + #endif /* _LINUX_NTFS3_NTFS_FS_H */ diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c index 8d2e51bae2cb..b26d95a8d327 100644 --- a/fs/ntfs3/super.c +++ b/fs/ntfs3/super.c @@ -408,6 +408,12 @@ static int ntfs_fs_reconfigure(struct fs_context *fc) struct ntfs_mount_options *new_opts = fc->fs_private; int ro_rw; + /* If ntfs3 is used as legacy ntfs enforce read-only mode. */ + if (is_legacy_ntfs(sb)) { + fc->sb_flags |= SB_RDONLY; + goto out; + } + ro_rw = sb_rdonly(sb) && !(fc->sb_flags & SB_RDONLY); if (ro_rw && (sbi->flags & NTFS_FLAGS_NEED_REPLAY)) { errorf(fc, @@ -427,8 +433,6 @@ static int ntfs_fs_reconfigure(struct fs_context *fc) fc, "ntfs3: Cannot use different iocharset when remounting!"); - sync_filesystem(sb); - if (ro_rw && (sbi->volume.flags & VOLUME_FLAG_DIRTY) && !new_opts->force) { errorf(fc, @@ -436,6 +440,8 @@ static int ntfs_fs_reconfigure(struct fs_context *fc) return -EINVAL; } +out: + sync_filesystem(sb); swap(sbi->options, fc->fs_private); return 0; @@ -1613,6 +1619,8 @@ load_root: } #endif + if (is_legacy_ntfs(sb)) + sb->s_flags |= SB_RDONLY; return 0; put_inode_out: @@ -1730,7 +1738,7 @@ static const struct fs_context_operations ntfs_context_ops = { * This will called when mount/remount. We will first initialize * options so that if remount we can use just that. */ -static int ntfs_init_fs_context(struct fs_context *fc) +static int __ntfs_init_fs_context(struct fs_context *fc) { struct ntfs_mount_options *opts; struct ntfs_sb_info *sbi; @@ -1778,6 +1786,11 @@ free_opts: return -ENOMEM; } +static int ntfs_init_fs_context(struct fs_context *fc) +{ + return __ntfs_init_fs_context(fc); +} + static void ntfs3_kill_sb(struct super_block *sb) { struct ntfs_sb_info *sbi = sb->s_fs_info; @@ -1800,10 +1813,20 @@ static struct file_system_type ntfs_fs_type = { }; #if IS_ENABLED(CONFIG_NTFS_FS) +static int ntfs_legacy_init_fs_context(struct fs_context *fc) +{ + int ret; + + ret = __ntfs_init_fs_context(fc); + /* If ntfs3 is used as legacy ntfs enforce read-only mode. */ + fc->sb_flags |= SB_RDONLY; + return ret; +} + static struct file_system_type ntfs_legacy_fs_type = { .owner = THIS_MODULE, .name = "ntfs", - .init_fs_context = ntfs_init_fs_context, + .init_fs_context = ntfs_legacy_init_fs_context, .parameters = ntfs_fs_parameters, .kill_sb = ntfs3_kill_sb, .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, @@ -1821,9 +1844,14 @@ static inline void unregister_as_ntfs_legacy(void) { unregister_filesystem(&ntfs_legacy_fs_type); } +bool is_legacy_ntfs(struct super_block *sb) +{ + return sb->s_type == &ntfs_legacy_fs_type; +} #else static inline void register_as_ntfs_legacy(void) {} static inline void unregister_as_ntfs_legacy(void) {} +bool is_legacy_ntfs(struct super_block *sb) { return false; } #endif -- cgit v1.2.3-59-g8ed1b From 9b872cc50daa7d1cb07d5bfd27ee9fa3f4e7eda9 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Tue, 16 Apr 2024 12:20:50 +0200 Subject: ntfs3: add legacy ntfs file operations To ensure that ioctl()s can't be used to circumvent write restrictions. Signed-off-by: Christian Brauner --- fs/ntfs3/dir.c | 7 +++++++ fs/ntfs3/file.c | 8 ++++++++ fs/ntfs3/inode.c | 20 ++++++++++++++++---- fs/ntfs3/ntfs_fs.h | 2 ++ 4 files changed, 33 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/ntfs3/dir.c b/fs/ntfs3/dir.c index 5cf3d9decf64..263635199b60 100644 --- a/fs/ntfs3/dir.c +++ b/fs/ntfs3/dir.c @@ -616,4 +616,11 @@ const struct file_operations ntfs_dir_operations = { .compat_ioctl = ntfs_compat_ioctl, #endif }; + +const struct file_operations ntfs_legacy_dir_operations = { + .llseek = generic_file_llseek, + .read = generic_read_dir, + .iterate_shared = ntfs_readdir, + .open = ntfs_file_open, +}; // clang-format on diff --git a/fs/ntfs3/file.c b/fs/ntfs3/file.c index 5418662c80d8..b73969e05052 100644 --- a/fs/ntfs3/file.c +++ b/fs/ntfs3/file.c @@ -1236,4 +1236,12 @@ const struct file_operations ntfs_file_operations = { .fallocate = ntfs_fallocate, .release = ntfs_file_release, }; + +const struct file_operations ntfs_legacy_file_operations = { + .llseek = generic_file_llseek, + .read_iter = ntfs_file_read_iter, + .splice_read = ntfs_file_splice_read, + .open = ntfs_file_open, + .release = ntfs_file_release, +}; // clang-format on diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c index eb7a8c9fba01..d273eda1cf45 100644 --- a/fs/ntfs3/inode.c +++ b/fs/ntfs3/inode.c @@ -440,7 +440,10 @@ end_enum: * Usually a hard links to directories are disabled. */ inode->i_op = &ntfs_dir_inode_operations; - inode->i_fop = &ntfs_dir_operations; + if (is_legacy_ntfs(inode->i_sb)) + inode->i_fop = &ntfs_legacy_dir_operations; + else + inode->i_fop = &ntfs_dir_operations; ni->i_valid = 0; } else if (S_ISLNK(mode)) { ni->std_fa &= ~FILE_ATTRIBUTE_DIRECTORY; @@ -450,7 +453,10 @@ end_enum: } else if (S_ISREG(mode)) { ni->std_fa &= ~FILE_ATTRIBUTE_DIRECTORY; inode->i_op = &ntfs_file_inode_operations; - inode->i_fop = &ntfs_file_operations; + if (is_legacy_ntfs(inode->i_sb)) + inode->i_fop = &ntfs_legacy_file_operations; + else + inode->i_fop = &ntfs_file_operations; inode->i_mapping->a_ops = is_compressed(ni) ? &ntfs_aops_cmpr : &ntfs_aops; if (ino != MFT_REC_MFT) @@ -1614,7 +1620,10 @@ struct inode *ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, if (S_ISDIR(mode)) { inode->i_op = &ntfs_dir_inode_operations; - inode->i_fop = &ntfs_dir_operations; + if (is_legacy_ntfs(inode->i_sb)) + inode->i_fop = &ntfs_legacy_dir_operations; + else + inode->i_fop = &ntfs_dir_operations; } else if (S_ISLNK(mode)) { inode->i_op = &ntfs_link_inode_operations; inode->i_fop = NULL; @@ -1623,7 +1632,10 @@ struct inode *ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir, inode_nohighmem(inode); } else if (S_ISREG(mode)) { inode->i_op = &ntfs_file_inode_operations; - inode->i_fop = &ntfs_file_operations; + if (is_legacy_ntfs(inode->i_sb)) + inode->i_fop = &ntfs_legacy_file_operations; + else + inode->i_fop = &ntfs_file_operations; inode->i_mapping->a_ops = is_compressed(ni) ? &ntfs_aops_cmpr : &ntfs_aops; init_rwsem(&ni->file.run_lock); diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h index 184c8bc76b92..5f4d288c6adf 100644 --- a/fs/ntfs3/ntfs_fs.h +++ b/fs/ntfs3/ntfs_fs.h @@ -493,6 +493,7 @@ struct inode *dir_search_u(struct inode *dir, const struct cpu_str *uni, struct ntfs_fnd *fnd); bool dir_is_empty(struct inode *dir); extern const struct file_operations ntfs_dir_operations; +extern const struct file_operations ntfs_legacy_dir_operations; /* Globals from file.c */ int ntfs_getattr(struct mnt_idmap *idmap, const struct path *path, @@ -507,6 +508,7 @@ long ntfs_compat_ioctl(struct file *filp, u32 cmd, unsigned long arg); extern const struct inode_operations ntfs_special_inode_operations; extern const struct inode_operations ntfs_file_inode_operations; extern const struct file_operations ntfs_file_operations; +extern const struct file_operations ntfs_legacy_file_operations; /* Globals from frecord.c */ void ni_remove_mi(struct ntfs_inode *ni, struct mft_inode *mi); -- cgit v1.2.3-59-g8ed1b From 619606a7b8d5e54b71578ecc988d3f8e1896bbc6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 17 Apr 2024 09:47:19 +0100 Subject: netfs: Fix writethrough-mode error handling Fix the error return in netfs_perform_write() acting in writethrough-mode to return any cached error in the case that netfs_end_writethrough() returns 0. This can affect the use of O_SYNC/O_DSYNC/RWF_SYNC/RWF_DSYNC in 9p and afs. Fixes: 41d8e7673a77 ("netfs: Implement a write-through caching option") Signed-off-by: David Howells Link: https://lore.kernel.org/r/6736.1713343639@warthog.procyon.org.uk Reviewed-by: Jeff Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: Marc Dionne cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org Signed-off-by: Christian Brauner --- fs/netfs/buffered_write.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 9a0d32e4b422..8f13ca8fbc74 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -164,7 +164,7 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, enum netfs_how_to_modify howto; enum netfs_folio_trace trace; unsigned int bdp_flags = (iocb->ki_flags & IOCB_SYNC) ? 0: BDP_ASYNC; - ssize_t written = 0, ret; + ssize_t written = 0, ret, ret2; loff_t i_size, pos = iocb->ki_pos, from, to; size_t max_chunk = PAGE_SIZE << MAX_PAGECACHE_ORDER; bool maybe_trouble = false; @@ -395,10 +395,12 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, out: if (unlikely(wreq)) { - ret = netfs_end_writethrough(wreq, iocb); + ret2 = netfs_end_writethrough(wreq, iocb); wbc_detach_inode(&wbc); - if (ret == -EIOCBQUEUED) - return ret; + if (ret2 == -EIOCBQUEUED) + return ret2; + if (ret == 0) + ret = ret2; } iocb->ki_pos += written; -- cgit v1.2.3-59-g8ed1b From b591dfb8330e30da0dcc8f5466c6f18173cdab32 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 17 Apr 2024 16:04:07 +0100 Subject: udf: Convert udf_symlink_filler() to use a folio Remove the conversion to struct page and use folio APIs throughout. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jan Kara Message-Id: <20240417150416.752929-2-willy@infradead.org> --- fs/udf/symlink.c | 22 ++++++---------------- 1 file changed, 6 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c index f7eaf7b14594..0105e7e2ba3d 100644 --- a/fs/udf/symlink.c +++ b/fs/udf/symlink.c @@ -99,18 +99,17 @@ static int udf_pc_to_char(struct super_block *sb, unsigned char *from, static int udf_symlink_filler(struct file *file, struct folio *folio) { - struct page *page = &folio->page; - struct inode *inode = page->mapping->host; + struct inode *inode = folio->mapping->host; struct buffer_head *bh = NULL; unsigned char *symlink; int err = 0; - unsigned char *p = page_address(page); + unsigned char *p = folio_address(folio); struct udf_inode_info *iinfo = UDF_I(inode); /* We don't support symlinks longer than one block */ if (inode->i_size > inode->i_sb->s_blocksize) { err = -ENAMETOOLONG; - goto out_unlock; + goto out; } if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { @@ -120,24 +119,15 @@ static int udf_symlink_filler(struct file *file, struct folio *folio) if (!bh) { if (!err) err = -EFSCORRUPTED; - goto out_err; + goto out; } symlink = bh->b_data; } err = udf_pc_to_char(inode->i_sb, symlink, inode->i_size, p, PAGE_SIZE); brelse(bh); - if (err) - goto out_err; - - SetPageUptodate(page); - unlock_page(page); - return 0; - -out_err: - SetPageError(page); -out_unlock: - unlock_page(page); +out: + folio_end_read(folio, err == 0); return err; } -- cgit v1.2.3-59-g8ed1b From d08f069cc2dafddf81b23db2ead45539b12cd0c2 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 17 Apr 2024 16:04:08 +0100 Subject: udf: Convert udf_write_begin() to use a folio Use the folio APIs throughout instead of the deprecated page APIs. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jan Kara Message-Id: <20240417150416.752929-3-willy@infradead.org> --- fs/udf/inode.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 2f831a3a91af..5146b9d7aba3 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -254,7 +254,7 @@ static int udf_write_begin(struct file *file, struct address_space *mapping, struct page **pagep, void **fsdata) { struct udf_inode_info *iinfo = UDF_I(file_inode(file)); - struct page *page; + struct folio *folio; int ret; if (iinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) { @@ -266,12 +266,13 @@ static int udf_write_begin(struct file *file, struct address_space *mapping, } if (WARN_ON_ONCE(pos >= PAGE_SIZE)) return -EIO; - page = grab_cache_page_write_begin(mapping, 0); - if (!page) - return -ENOMEM; - *pagep = page; - if (!PageUptodate(page)) - udf_adinicb_readpage(page); + folio = __filemap_get_folio(mapping, 0, FGP_WRITEBEGIN, + mapping_gfp_mask(mapping)); + if (IS_ERR(folio)) + return PTR_ERR(folio); + *pagep = &folio->page; + if (!folio_test_uptodate(folio)) + udf_adinicb_readpage(&folio->page); return 0; } -- cgit v1.2.3-59-g8ed1b From db6754090a4f99c67e05ae6b87343ba6e013531f Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 17 Apr 2024 16:04:09 +0100 Subject: udf: Convert udf_expand_file_adinicb() to use a folio Use the folio APIs throughout this function. Signed-off-by: Matthew Wilcox (Oracle) Fixes: 1eeceaec794e ("udf: Convert udf_expand_file_adinicb() to avoid kmap_atomic()") Signed-off-by: Jan Kara Message-Id: <20240417150416.752929-4-willy@infradead.org> --- fs/udf/inode.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 5146b9d7aba3..59215494e6f6 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -342,7 +342,7 @@ const struct address_space_operations udf_aops = { */ int udf_expand_file_adinicb(struct inode *inode) { - struct page *page; + struct folio *folio; struct udf_inode_info *iinfo = UDF_I(inode); int err; @@ -358,12 +358,13 @@ int udf_expand_file_adinicb(struct inode *inode) return 0; } - page = find_or_create_page(inode->i_mapping, 0, GFP_KERNEL); - if (!page) - return -ENOMEM; + folio = __filemap_get_folio(inode->i_mapping, 0, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, GFP_KERNEL); + if (IS_ERR(folio)) + return PTR_ERR(folio); - if (!PageUptodate(page)) - udf_adinicb_readpage(page); + if (!folio_test_uptodate(folio)) + udf_adinicb_readpage(&folio->page); down_write(&iinfo->i_data_sem); memset(iinfo->i_data + iinfo->i_lenEAttr, 0x00, iinfo->i_lenAlloc); @@ -372,22 +373,22 @@ int udf_expand_file_adinicb(struct inode *inode) iinfo->i_alloc_type = ICBTAG_FLAG_AD_SHORT; else iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG; - set_page_dirty(page); - unlock_page(page); + folio_mark_dirty(folio); + folio_unlock(folio); up_write(&iinfo->i_data_sem); err = filemap_fdatawrite(inode->i_mapping); if (err) { /* Restore everything back so that we don't lose data... */ - lock_page(page); + folio_lock(folio); down_write(&iinfo->i_data_sem); - memcpy_to_page(page, 0, iinfo->i_data + iinfo->i_lenEAttr, - inode->i_size); - unlock_page(page); + memcpy_from_folio(iinfo->i_data + iinfo->i_lenEAttr, + folio, 0, inode->i_size); + folio_unlock(folio); iinfo->i_alloc_type = ICBTAG_FLAG_AD_IN_ICB; iinfo->i_lenAlloc = inode->i_size; up_write(&iinfo->i_data_sem); } - put_page(page); + folio_put(folio); mark_inode_dirty(inode); return err; -- cgit v1.2.3-59-g8ed1b From d257d924a3fc6fe26594fba221c8be62f043b8d0 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 17 Apr 2024 16:04:10 +0100 Subject: udf: Convert udf_adinicb_readpage() to udf_adinicb_read_folio() Now that all three callers have a folio, convert this function to take a folio, and use the folio APIs. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jan Kara Message-Id: <20240417150416.752929-5-willy@infradead.org> --- fs/udf/inode.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 59215494e6f6..d34562156522 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -208,19 +208,14 @@ static int udf_writepages(struct address_space *mapping, return write_cache_pages(mapping, wbc, udf_adinicb_writepage, NULL); } -static void udf_adinicb_readpage(struct page *page) +static void udf_adinicb_read_folio(struct folio *folio) { - struct inode *inode = page->mapping->host; - char *kaddr; + struct inode *inode = folio->mapping->host; struct udf_inode_info *iinfo = UDF_I(inode); loff_t isize = i_size_read(inode); - kaddr = kmap_local_page(page); - memcpy(kaddr, iinfo->i_data + iinfo->i_lenEAttr, isize); - memset(kaddr + isize, 0, PAGE_SIZE - isize); - flush_dcache_page(page); - SetPageUptodate(page); - kunmap_local(kaddr); + folio_fill_tail(folio, 0, iinfo->i_data + iinfo->i_lenEAttr, isize); + folio_mark_uptodate(folio); } static int udf_read_folio(struct file *file, struct folio *folio) @@ -228,7 +223,7 @@ static int udf_read_folio(struct file *file, struct folio *folio) struct udf_inode_info *iinfo = UDF_I(file_inode(file)); if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { - udf_adinicb_readpage(&folio->page); + udf_adinicb_read_folio(folio); folio_unlock(folio); return 0; } @@ -272,7 +267,7 @@ static int udf_write_begin(struct file *file, struct address_space *mapping, return PTR_ERR(folio); *pagep = &folio->page; if (!folio_test_uptodate(folio)) - udf_adinicb_readpage(&folio->page); + udf_adinicb_read_folio(folio); return 0; } @@ -364,7 +359,7 @@ int udf_expand_file_adinicb(struct inode *inode) return PTR_ERR(folio); if (!folio_test_uptodate(folio)) - udf_adinicb_readpage(&folio->page); + udf_adinicb_read_folio(folio); down_write(&iinfo->i_data_sem); memset(iinfo->i_data + iinfo->i_lenEAttr, 0x00, iinfo->i_lenAlloc); -- cgit v1.2.3-59-g8ed1b From 2f1c1bd7b18768377ff4a84974d77e6e8b371a67 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 17 Apr 2024 16:04:11 +0100 Subject: udf: Convert udf_symlink_getattr() to use a folio We're getting this from the page cache, so it's definitely a folio. Saves a call to compound_head() hidden in put_page(). Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jan Kara Message-Id: <20240417150416.752929-6-willy@infradead.org> --- fs/udf/symlink.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c index 0105e7e2ba3d..fe03745d09b1 100644 --- a/fs/udf/symlink.c +++ b/fs/udf/symlink.c @@ -137,12 +137,12 @@ static int udf_symlink_getattr(struct mnt_idmap *idmap, { struct dentry *dentry = path->dentry; struct inode *inode = d_backing_inode(dentry); - struct page *page; + struct folio *folio; generic_fillattr(&nop_mnt_idmap, request_mask, inode, stat); - page = read_mapping_page(inode->i_mapping, 0, NULL); - if (IS_ERR(page)) - return PTR_ERR(page); + folio = read_mapping_folio(inode->i_mapping, 0, NULL); + if (IS_ERR(folio)) + return PTR_ERR(folio); /* * UDF uses non-trivial encoding of symlinks so i_size does not match * number of characters reported by readlink(2) which apparently some @@ -152,8 +152,8 @@ static int udf_symlink_getattr(struct mnt_idmap *idmap, * let's report the length of string returned by readlink(2) for * st_size. */ - stat->size = strlen(page_address(page)); - put_page(page); + stat->size = strlen(folio_address(folio)); + folio_put(folio); return 0; } -- cgit v1.2.3-59-g8ed1b From f5985ef281f9efff3e8291123fdb8a748b506951 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 17 Apr 2024 16:04:12 +0100 Subject: udf: Convert udf_page_mkwrite() to use a folio Convert the vm_fault page to a folio, then use it throughout. Replaces five calls to compound_head() with one. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jan Kara Message-Id: <20240417150416.752929-7-willy@infradead.org> --- fs/udf/file.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/udf/file.c b/fs/udf/file.c index 0ceac4b5937c..97c59585208c 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -39,7 +39,7 @@ static vm_fault_t udf_page_mkwrite(struct vm_fault *vmf) struct vm_area_struct *vma = vmf->vma; struct inode *inode = file_inode(vma->vm_file); struct address_space *mapping = inode->i_mapping; - struct page *page = vmf->page; + struct folio *folio = page_folio(vmf->page); loff_t size; unsigned int end; vm_fault_t ret = VM_FAULT_LOCKED; @@ -48,31 +48,31 @@ static vm_fault_t udf_page_mkwrite(struct vm_fault *vmf) sb_start_pagefault(inode->i_sb); file_update_time(vma->vm_file); filemap_invalidate_lock_shared(mapping); - lock_page(page); + folio_lock(folio); size = i_size_read(inode); - if (page->mapping != inode->i_mapping || page_offset(page) >= size) { - unlock_page(page); + if (folio->mapping != inode->i_mapping || folio_pos(folio) >= size) { + folio_unlock(folio); ret = VM_FAULT_NOPAGE; goto out_unlock; } /* Space is already allocated for in-ICB file */ if (UDF_I(inode)->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) goto out_dirty; - if (page->index == size >> PAGE_SHIFT) + if (folio->index == size >> PAGE_SHIFT) end = size & ~PAGE_MASK; else end = PAGE_SIZE; - err = __block_write_begin(page, 0, end, udf_get_block); + err = __block_write_begin(&folio->page, 0, end, udf_get_block); if (err) { - unlock_page(page); + folio_unlock(folio); ret = vmf_fs_error(err); goto out_unlock; } - block_commit_write(page, 0, end); + block_commit_write(&folio->page, 0, end); out_dirty: - set_page_dirty(page); - wait_for_stable_page(page); + folio_mark_dirty(folio); + folio_wait_stable(folio); out_unlock: filemap_invalidate_unlock_shared(mapping); sb_end_pagefault(inode->i_sb); -- cgit v1.2.3-59-g8ed1b From e29741676fac5d4430b3d2b799a5ff671ea9f023 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 17 Apr 2024 16:04:13 +0100 Subject: udf: Use a folio in udf_write_end() Convert the page to a folio and use the folio APIs. Replaces three calls to compound_head() with one. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jan Kara Message-Id: <20240417150416.752929-8-willy@infradead.org> --- fs/udf/inode.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/udf/inode.c b/fs/udf/inode.c index d34562156522..2fb21c5ffccf 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -276,17 +276,19 @@ static int udf_write_end(struct file *file, struct address_space *mapping, struct page *page, void *fsdata) { struct inode *inode = file_inode(file); + struct folio *folio; loff_t last_pos; if (UDF_I(inode)->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) return generic_write_end(file, mapping, pos, len, copied, page, fsdata); + folio = page_folio(page); last_pos = pos + copied; if (last_pos > inode->i_size) i_size_write(inode, last_pos); - set_page_dirty(page); - unlock_page(page); - put_page(page); + folio_mark_dirty(folio); + folio_unlock(folio); + folio_put(folio); return copied; } -- cgit v1.2.3-59-g8ed1b From f566d5b9fb7136d39d4e9c54d84c82835b539b4e Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:19 -0700 Subject: xfs: remove XFS_DA_OP_REMOVE Nobody checks this flag, so get rid of it. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.h | 1 - fs/xfs/libxfs/xfs_da_btree.h | 6 ++---- 2 files changed, 2 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index e4f55008552b..670ab2a613fc 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -590,7 +590,6 @@ xfs_attr_init_add_state(struct xfs_da_args *args) static inline enum xfs_delattr_state xfs_attr_init_remove_state(struct xfs_da_args *args) { - args->op_flags |= XFS_DA_OP_REMOVE; if (xfs_attr_is_shortform(args->dp)) return XFS_DAS_SF_REMOVE; if (xfs_attr_is_leaf(args->dp)) diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 7a004786ee0a..76e764080d99 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -91,9 +91,8 @@ typedef struct xfs_da_args { #define XFS_DA_OP_OKNOENT (1u << 3) /* lookup op, ENOENT ok, else die */ #define XFS_DA_OP_CILOOKUP (1u << 4) /* lookup returns CI name if found */ #define XFS_DA_OP_NOTIME (1u << 5) /* don't update inode timestamps */ -#define XFS_DA_OP_REMOVE (1u << 6) /* this is a remove operation */ -#define XFS_DA_OP_RECOVERY (1u << 7) /* Log recovery operation */ -#define XFS_DA_OP_LOGGED (1u << 8) /* Use intent items to track op */ +#define XFS_DA_OP_RECOVERY (1u << 6) /* Log recovery operation */ +#define XFS_DA_OP_LOGGED (1u << 7) /* Use intent items to track op */ #define XFS_DA_OP_FLAGS \ { XFS_DA_OP_JUSTCHECK, "JUSTCHECK" }, \ @@ -102,7 +101,6 @@ typedef struct xfs_da_args { { XFS_DA_OP_OKNOENT, "OKNOENT" }, \ { XFS_DA_OP_CILOOKUP, "CILOOKUP" }, \ { XFS_DA_OP_NOTIME, "NOTIME" }, \ - { XFS_DA_OP_REMOVE, "REMOVE" }, \ { XFS_DA_OP_RECOVERY, "RECOVERY" }, \ { XFS_DA_OP_LOGGED, "LOGGED" } -- cgit v1.2.3-59-g8ed1b From 779a4b606c7678343e3603398fe07de38b817ef4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:20 -0700 Subject: xfs: remove XFS_DA_OP_NOTIME The only user of this flag sets it prior to an xfs_attr_get_ilocked call, which doesn't update anything. Get rid of the flag. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 5 ++--- fs/xfs/libxfs/xfs_da_btree.h | 6 ++---- fs/xfs/scrub/attr.c | 1 - 3 files changed, 4 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 05d22c5e3885..30e6084122d8 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -365,7 +365,7 @@ xfs_attr_try_sf_addname( * Commit the shortform mods, and we're done. * NOTE: this is also the error path (EEXIST, etc). */ - if (!error && !(args->op_flags & XFS_DA_OP_NOTIME)) + if (!error) xfs_trans_ichgtime(args->trans, dp, XFS_ICHGTIME_CHG); if (xfs_has_wsync(dp->i_mount)) @@ -1033,8 +1033,7 @@ xfs_attr_set( if (xfs_has_wsync(mp)) xfs_trans_set_sync(args->trans); - if (!(args->op_flags & XFS_DA_OP_NOTIME)) - xfs_trans_ichgtime(args->trans, dp, XFS_ICHGTIME_CHG); + xfs_trans_ichgtime(args->trans, dp, XFS_ICHGTIME_CHG); /* * Commit the last in the sequence of transactions. diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 76e764080d99..b04a3290ffac 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -90,9 +90,8 @@ typedef struct xfs_da_args { #define XFS_DA_OP_ADDNAME (1u << 2) /* this is an add operation */ #define XFS_DA_OP_OKNOENT (1u << 3) /* lookup op, ENOENT ok, else die */ #define XFS_DA_OP_CILOOKUP (1u << 4) /* lookup returns CI name if found */ -#define XFS_DA_OP_NOTIME (1u << 5) /* don't update inode timestamps */ -#define XFS_DA_OP_RECOVERY (1u << 6) /* Log recovery operation */ -#define XFS_DA_OP_LOGGED (1u << 7) /* Use intent items to track op */ +#define XFS_DA_OP_RECOVERY (1u << 5) /* Log recovery operation */ +#define XFS_DA_OP_LOGGED (1u << 6) /* Use intent items to track op */ #define XFS_DA_OP_FLAGS \ { XFS_DA_OP_JUSTCHECK, "JUSTCHECK" }, \ @@ -100,7 +99,6 @@ typedef struct xfs_da_args { { XFS_DA_OP_ADDNAME, "ADDNAME" }, \ { XFS_DA_OP_OKNOENT, "OKNOENT" }, \ { XFS_DA_OP_CILOOKUP, "CILOOKUP" }, \ - { XFS_DA_OP_NOTIME, "NOTIME" }, \ { XFS_DA_OP_RECOVERY, "RECOVERY" }, \ { XFS_DA_OP_LOGGED, "LOGGED" } diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 8853e4d0eee3..5b855d7c9821 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -173,7 +173,6 @@ xchk_xattr_actor( void *priv) { struct xfs_da_args args = { - .op_flags = XFS_DA_OP_NOTIME, .attr_filter = attr_flags & XFS_ATTR_NSP_ONDISK_MASK, .geo = sc->mp->m_attr_geo, .whichfork = XFS_ATTR_FORK, -- cgit v1.2.3-59-g8ed1b From 54275d8496f3e4764302cebc0e9517d950ba6589 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:21 -0700 Subject: xfs: remove xfs_da_args.attr_flags This field only ever contains XATTR_{CREATE,REPLACE}, and it only goes as deep as xfs_attr_set. Remove the field from the structure and replace it with an enum specifying exactly what kind of change we want to make to the xattr structure. Upsert is the name that we'll give to the flags==0 operation, because we're either updating an existing value or inserting it, and the caller doesn't care. Note: The "UPSERTR" name created here is to make userspace porting easier. It will be removed in the next patch. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 7 ++++--- fs/xfs/libxfs/xfs_attr.h | 9 ++++++++- fs/xfs/libxfs/xfs_da_btree.h | 1 - fs/xfs/scrub/attr_repair.c | 3 +-- fs/xfs/xfs_acl.c | 2 +- fs/xfs/xfs_ioctl.c | 14 ++++++-------- fs/xfs/xfs_iops.c | 2 +- fs/xfs/xfs_trace.h | 7 +------ fs/xfs/xfs_xattr.c | 19 +++++++++++++++---- fs/xfs/xfs_xattr.h | 3 ++- 10 files changed, 39 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 30e6084122d8..b04e09143869 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -922,7 +922,8 @@ xfs_attr_defer_add( */ int xfs_attr_set( - struct xfs_da_args *args) + struct xfs_da_args *args, + enum xfs_attr_update op) { struct xfs_inode *dp = args->dp; struct xfs_mount *mp = dp->i_mount; @@ -1008,7 +1009,7 @@ xfs_attr_set( } /* Pure create fails if the attr already exists */ - if (args->attr_flags & XATTR_CREATE) + if (op == XFS_ATTRUPDATE_CREATE) goto out_trans_cancel; xfs_attr_defer_add(args, XFS_ATTRI_OP_FLAGS_REPLACE); break; @@ -1018,7 +1019,7 @@ xfs_attr_set( goto out_trans_cancel; /* Pure replace fails if no existing attr to replace. */ - if (args->attr_flags & XATTR_REPLACE) + if (op == XFS_ATTRUPDATE_REPLACE) goto out_trans_cancel; xfs_attr_defer_add(args, XFS_ATTRI_OP_FLAGS_SET); break; diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index 670ab2a613fc..228360f7c85c 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -544,7 +544,14 @@ int xfs_inode_hasattr(struct xfs_inode *ip); bool xfs_attr_is_leaf(struct xfs_inode *ip); int xfs_attr_get_ilocked(struct xfs_da_args *args); int xfs_attr_get(struct xfs_da_args *args); -int xfs_attr_set(struct xfs_da_args *args); + +enum xfs_attr_update { + XFS_ATTRUPDATE_UPSERTR, /* set/remove value, replace any existing attr */ + XFS_ATTRUPDATE_CREATE, /* set value, fail if attr already exists */ + XFS_ATTRUPDATE_REPLACE, /* set value, fail if attr does not exist */ +}; + +int xfs_attr_set(struct xfs_da_args *args, enum xfs_attr_update op); int xfs_attr_set_iter(struct xfs_attr_intent *attr); int xfs_attr_remove_iter(struct xfs_attr_intent *attr); bool xfs_attr_namecheck(const void *name, size_t length); diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index b04a3290ffac..706b529a81fe 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -60,7 +60,6 @@ typedef struct xfs_da_args { void *value; /* set of bytes (maybe contain NULLs) */ int valuelen; /* length of value */ unsigned int attr_filter; /* XFS_ATTR_{ROOT,SECURE,INCOMPLETE} */ - unsigned int attr_flags; /* XATTR_{CREATE,REPLACE} */ xfs_dahash_t hashval; /* hash value of name */ xfs_ino_t inumber; /* input/output inode number */ struct xfs_inode *dp; /* directory inode to manipulate */ diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index 7b4318764d03..3066d662ea13 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -557,7 +557,6 @@ xrep_xattr_insert_rec( struct xfs_da_args args = { .dp = rx->sc->tempip, .attr_filter = key->flags, - .attr_flags = XATTR_CREATE, .namelen = key->namelen, .valuelen = key->valuelen, .owner = rx->sc->ip->i_ino, @@ -605,7 +604,7 @@ xrep_xattr_insert_rec( * xfs_attr_set creates and commits its own transaction. If the attr * already exists, we'll just drop it during the rebuild. */ - error = xfs_attr_set(&args); + error = xfs_attr_set(&args, XFS_ATTRUPDATE_CREATE); if (error == -EEXIST) error = 0; diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 4bf69c9c088e..12f98af9e709 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -203,7 +203,7 @@ __xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) xfs_acl_to_disk(args.value, acl); } - error = xfs_attr_change(&args); + error = xfs_attr_change(&args, XFS_ATTRUPDATE_UPSERTR); kvfree(args.value); /* diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index efa95892655d..0eca7a9096e2 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -361,15 +361,15 @@ xfs_attr_filter( return 0; } -static unsigned int -xfs_attr_flags( +static inline enum xfs_attr_update +xfs_xattr_flags( u32 ioc_flags) { if (ioc_flags & XFS_IOC_ATTR_CREATE) - return XATTR_CREATE; + return XFS_ATTRUPDATE_CREATE; if (ioc_flags & XFS_IOC_ATTR_REPLACE) - return XATTR_REPLACE; - return 0; + return XFS_ATTRUPDATE_REPLACE; + return XFS_ATTRUPDATE_UPSERTR; } int @@ -476,7 +476,6 @@ xfs_attrmulti_attr_get( struct xfs_da_args args = { .dp = XFS_I(inode), .attr_filter = xfs_attr_filter(flags), - .attr_flags = xfs_attr_flags(flags), .name = name, .namelen = strlen(name), .valuelen = *len, @@ -510,7 +509,6 @@ xfs_attrmulti_attr_set( struct xfs_da_args args = { .dp = XFS_I(inode), .attr_filter = xfs_attr_filter(flags), - .attr_flags = xfs_attr_flags(flags), .name = name, .namelen = strlen(name), }; @@ -528,7 +526,7 @@ xfs_attrmulti_attr_set( args.valuelen = len; } - error = xfs_attr_change(&args); + error = xfs_attr_change(&args, xfs_xattr_flags(flags)); if (!error && (flags & XFS_IOC_ATTR_ROOT)) xfs_forget_acl(inode, name); kfree(args.value); diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index ad76704ab133..d02ca2248bbc 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -63,7 +63,7 @@ xfs_initxattrs( .value = xattr->value, .valuelen = xattr->value_len, }; - error = xfs_attr_change(&args); + error = xfs_attr_change(&args, XFS_ATTRUPDATE_UPSERTR); if (error < 0) break; } diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 08df5bb70a6c..57f225ba7e8a 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -1999,7 +1999,6 @@ DECLARE_EVENT_CLASS(xfs_attr_class, __field(int, valuelen) __field(xfs_dahash_t, hashval) __field(unsigned int, attr_filter) - __field(unsigned int, attr_flags) __field(uint32_t, op_flags) ), TP_fast_assign( @@ -2011,11 +2010,10 @@ DECLARE_EVENT_CLASS(xfs_attr_class, __entry->valuelen = args->valuelen; __entry->hashval = args->hashval; __entry->attr_filter = args->attr_filter; - __entry->attr_flags = args->attr_flags; __entry->op_flags = args->op_flags; ), TP_printk("dev %d:%d ino 0x%llx name %.*s namelen %d valuelen %d " - "hashval 0x%x filter %s flags %s op_flags %s", + "hashval 0x%x filter %s op_flags %s", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->namelen, @@ -2025,9 +2023,6 @@ DECLARE_EVENT_CLASS(xfs_attr_class, __entry->hashval, __print_flags(__entry->attr_filter, "|", XFS_ATTR_FILTER_FLAGS), - __print_flags(__entry->attr_flags, "|", - { XATTR_CREATE, "CREATE" }, - { XATTR_REPLACE, "REPLACE" }), __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS)) ) diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 4ebf7052eb67..6ce1e06f650a 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -73,7 +73,8 @@ xfs_attr_want_log_assist( */ int xfs_attr_change( - struct xfs_da_args *args) + struct xfs_da_args *args, + enum xfs_attr_update op) { struct xfs_mount *mp = args->dp->i_mount; int error; @@ -88,7 +89,7 @@ xfs_attr_change( args->op_flags |= XFS_DA_OP_LOGGED; } - return xfs_attr_set(args); + return xfs_attr_set(args, op); } @@ -115,6 +116,17 @@ xfs_xattr_get(const struct xattr_handler *handler, struct dentry *unused, return args.valuelen; } +static inline enum xfs_attr_update +xfs_xattr_flags_to_op( + int flags) +{ + if (flags & XATTR_CREATE) + return XFS_ATTRUPDATE_CREATE; + if (flags & XATTR_REPLACE) + return XFS_ATTRUPDATE_REPLACE; + return XFS_ATTRUPDATE_UPSERTR; +} + static int xfs_xattr_set(const struct xattr_handler *handler, struct mnt_idmap *idmap, struct dentry *unused, @@ -124,7 +136,6 @@ xfs_xattr_set(const struct xattr_handler *handler, struct xfs_da_args args = { .dp = XFS_I(inode), .attr_filter = handler->flags, - .attr_flags = flags, .name = name, .namelen = strlen(name), .value = (void *)value, @@ -132,7 +143,7 @@ xfs_xattr_set(const struct xattr_handler *handler, }; int error; - error = xfs_attr_change(&args); + error = xfs_attr_change(&args, xfs_xattr_flags_to_op(flags)); if (!error && (handler->flags & XFS_ATTR_ROOT)) xfs_forget_acl(inode, name); return error; diff --git a/fs/xfs/xfs_xattr.h b/fs/xfs/xfs_xattr.h index cec766cad26c..c3eb858fb59e 100644 --- a/fs/xfs/xfs_xattr.h +++ b/fs/xfs/xfs_xattr.h @@ -6,7 +6,8 @@ #ifndef __XFS_XATTR_H__ #define __XFS_XATTR_H__ -int xfs_attr_change(struct xfs_da_args *args); +enum xfs_attr_update; +int xfs_attr_change(struct xfs_da_args *args, enum xfs_attr_update op); extern const struct xattr_handler * const xfs_xattr_handlers[]; -- cgit v1.2.3-59-g8ed1b From c27411d4c640037d70e2fa5751616e175e52ca5e Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:22 -0700 Subject: xfs: make attr removal an explicit operation Parent pointers match attrs on name+value, unlike everything else which matches on only the name. Therefore, we cannot keep using the heuristic that !value means remove. Make this an explicit operation code. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 19 ++++++++++--------- fs/xfs/libxfs/xfs_attr.h | 3 ++- fs/xfs/xfs_acl.c | 17 +++++++++-------- fs/xfs/xfs_ioctl.c | 9 ++++++--- fs/xfs/xfs_iops.c | 2 +- fs/xfs/xfs_xattr.c | 9 ++++++--- 6 files changed, 34 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index b04e09143869..f8f7445b063c 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -916,10 +916,6 @@ xfs_attr_defer_add( trace_xfs_attr_defer_add(new->xattri_dela_state, args->dp); } -/* - * Note: If args->value is NULL the attribute will be removed, just like the - * Linux ->setattr API. - */ int xfs_attr_set( struct xfs_da_args *args, @@ -955,7 +951,10 @@ xfs_attr_set( args->op_flags = XFS_DA_OP_OKNOENT | (args->op_flags & XFS_DA_OP_LOGGED); - if (args->value) { + switch (op) { + case XFS_ATTRUPDATE_UPSERT: + case XFS_ATTRUPDATE_CREATE: + case XFS_ATTRUPDATE_REPLACE: XFS_STATS_INC(mp, xs_attr_set); args->total = xfs_attr_calc_size(args, &local); @@ -975,9 +974,11 @@ xfs_attr_set( if (!local) rmt_blks = xfs_attr3_rmt_blocks(mp, args->valuelen); - } else { + break; + case XFS_ATTRUPDATE_REMOVE: XFS_STATS_INC(mp, xs_attr_remove); rmt_blks = xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); + break; } /* @@ -989,7 +990,7 @@ xfs_attr_set( if (error) return error; - if (args->value || xfs_inode_hasattr(dp)) { + if (op != XFS_ATTRUPDATE_REMOVE || xfs_inode_hasattr(dp)) { error = xfs_iext_count_may_overflow(dp, XFS_ATTR_FORK, XFS_IEXT_ATTR_MANIP_CNT(rmt_blks)); if (error == -EFBIG) @@ -1002,7 +1003,7 @@ xfs_attr_set( error = xfs_attr_lookup(args); switch (error) { case -EEXIST: - if (!args->value) { + if (op == XFS_ATTRUPDATE_REMOVE) { /* if no value, we are performing a remove operation */ xfs_attr_defer_add(args, XFS_ATTRI_OP_FLAGS_REMOVE); break; @@ -1015,7 +1016,7 @@ xfs_attr_set( break; case -ENOATTR: /* Can't remove what isn't there. */ - if (!args->value) + if (op == XFS_ATTRUPDATE_REMOVE) goto out_trans_cancel; /* Pure replace fails if no existing attr to replace. */ diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index 228360f7c85c..c8005f52102a 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -546,7 +546,8 @@ int xfs_attr_get_ilocked(struct xfs_da_args *args); int xfs_attr_get(struct xfs_da_args *args); enum xfs_attr_update { - XFS_ATTRUPDATE_UPSERTR, /* set/remove value, replace any existing attr */ + XFS_ATTRUPDATE_REMOVE, /* remove attr */ + XFS_ATTRUPDATE_UPSERT, /* set value, replace any existing attr */ XFS_ATTRUPDATE_CREATE, /* set value, fail if attr already exists */ XFS_ATTRUPDATE_REPLACE, /* set value, fail if attr does not exist */ }; diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 12f98af9e709..c7c3dcfa2718 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -201,16 +201,17 @@ __xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) if (!args.value) return -ENOMEM; xfs_acl_to_disk(args.value, acl); + error = xfs_attr_change(&args, XFS_ATTRUPDATE_UPSERT); + kvfree(args.value); + } else { + error = xfs_attr_change(&args, XFS_ATTRUPDATE_REMOVE); + /* + * If the attribute didn't exist to start with that's fine. + */ + if (error == -ENOATTR) + error = 0; } - error = xfs_attr_change(&args, XFS_ATTRUPDATE_UPSERTR); - kvfree(args.value); - - /* - * If the attribute didn't exist to start with that's fine. - */ - if (!acl && error == -ENOATTR) - error = 0; if (!error) set_cached_acl(inode, type, acl); return error; diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 0eca7a9096e2..e30f9f40f086 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -363,13 +363,16 @@ xfs_attr_filter( static inline enum xfs_attr_update xfs_xattr_flags( - u32 ioc_flags) + u32 ioc_flags, + void *value) { + if (!value) + return XFS_ATTRUPDATE_REMOVE; if (ioc_flags & XFS_IOC_ATTR_CREATE) return XFS_ATTRUPDATE_CREATE; if (ioc_flags & XFS_IOC_ATTR_REPLACE) return XFS_ATTRUPDATE_REPLACE; - return XFS_ATTRUPDATE_UPSERTR; + return XFS_ATTRUPDATE_UPSERT; } int @@ -526,7 +529,7 @@ xfs_attrmulti_attr_set( args.valuelen = len; } - error = xfs_attr_change(&args, xfs_xattr_flags(flags)); + error = xfs_attr_change(&args, xfs_xattr_flags(flags, args.value)); if (!error && (flags & XFS_IOC_ATTR_ROOT)) xfs_forget_acl(inode, name); kfree(args.value); diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index d02ca2248bbc..659fd10c0cda 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -63,7 +63,7 @@ xfs_initxattrs( .value = xattr->value, .valuelen = xattr->value_len, }; - error = xfs_attr_change(&args, XFS_ATTRUPDATE_UPSERTR); + error = xfs_attr_change(&args, XFS_ATTRUPDATE_UPSERT); if (error < 0) break; } diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 6ce1e06f650a..0cbb93cf2869 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -118,13 +118,16 @@ xfs_xattr_get(const struct xattr_handler *handler, struct dentry *unused, static inline enum xfs_attr_update xfs_xattr_flags_to_op( - int flags) + int flags, + const void *value) { + if (!value) + return XFS_ATTRUPDATE_REMOVE; if (flags & XATTR_CREATE) return XFS_ATTRUPDATE_CREATE; if (flags & XATTR_REPLACE) return XFS_ATTRUPDATE_REPLACE; - return XFS_ATTRUPDATE_UPSERTR; + return XFS_ATTRUPDATE_UPSERT; } static int @@ -143,7 +146,7 @@ xfs_xattr_set(const struct xattr_handler *handler, }; int error; - error = xfs_attr_change(&args, xfs_xattr_flags_to_op(flags)); + error = xfs_attr_change(&args, xfs_xattr_flags_to_op(flags, value)); if (!error && (handler->flags & XFS_ATTR_ROOT)) xfs_forget_acl(inode, name); return error; -- cgit v1.2.3-59-g8ed1b From cda60317ac57add7a0a2865aa29afbc6caad3e9a Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:23 -0700 Subject: xfs: rearrange xfs_da_args a bit to use less space A few notes about struct xfs_da_args: The XFS_ATTR_* flags only go up as far as XFS_ATTR_INCOMPLETE, which means that attr_filter could be a u8 field. I've reduced the number of XFS_DA_OP_* flags down to the point where op_flags would also fit into a u8. filetype has 7 bytes of slack after it, which is wasteful. namelen will never be greater than MAXNAMELEN, which is 256. This field could be reduced to a short. Rearrange the fields in xfs_da_args to waste less space. This reduces the structure size from 136 bytes to 128. Later when we add extra fields to support parent pointer replacement, this will only bloat the structure to 144 bytes, instead of 168. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_da_btree.h | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 706b529a81fe..17cef594b5bb 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -54,16 +54,20 @@ enum xfs_dacmp { */ typedef struct xfs_da_args { struct xfs_da_geometry *geo; /* da block geometry */ - const uint8_t *name; /* string (maybe not NULL terminated) */ - int namelen; /* length of string (maybe no NULL) */ - uint8_t filetype; /* filetype of inode for directories */ + const uint8_t *name; /* string (maybe not NULL terminated) */ void *value; /* set of bytes (maybe contain NULLs) */ - int valuelen; /* length of value */ - unsigned int attr_filter; /* XFS_ATTR_{ROOT,SECURE,INCOMPLETE} */ - xfs_dahash_t hashval; /* hash value of name */ - xfs_ino_t inumber; /* input/output inode number */ struct xfs_inode *dp; /* directory inode to manipulate */ struct xfs_trans *trans; /* current trans (changes over time) */ + + xfs_ino_t inumber; /* input/output inode number */ + xfs_ino_t owner; /* inode that owns the dir/attr data */ + + int valuelen; /* length of value */ + uint8_t filetype; /* filetype of inode for directories */ + uint8_t op_flags; /* operation flags */ + uint8_t attr_filter; /* XFS_ATTR_{ROOT,SECURE,INCOMPLETE} */ + short namelen; /* length of string (maybe no NULL) */ + xfs_dahash_t hashval; /* hash value of name */ xfs_extlen_t total; /* total blocks needed, for 1st bmap */ int whichfork; /* data or attribute fork */ xfs_dablk_t blkno; /* blkno of attr leaf of interest */ @@ -76,9 +80,7 @@ typedef struct xfs_da_args { xfs_dablk_t rmtblkno2; /* remote attr value starting blkno */ int rmtblkcnt2; /* remote attr value block count */ int rmtvaluelen2; /* remote attr value length in bytes */ - uint32_t op_flags; /* operation flags */ enum xfs_dacmp cmpresult; /* name compare result for lookups */ - xfs_ino_t owner; /* inode that owns the dir/attr data */ } xfs_da_args_t; /* -- cgit v1.2.3-59-g8ed1b From ef80de940a6344da1d4f12c948a0ad4d6ff6e841 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:24 -0700 Subject: xfs: attr fork iext must be loaded before calling xfs_attr_is_leaf Christoph noticed that the xfs_attr_is_leaf in xfs_attr_get_ilocked can access the incore extent tree of the attr fork, but nothing in the xfs_attr_get path guarantees that the incore tree is actually loaded. Most of the time it is, but seeing as xfs_attr_is_leaf ignores the return value of xfs_iext_get_extent I guess we've been making choices based on random stack contents and nobody's complained? Reported-by: Christoph Hellwig Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 17 +++++++++++++++++ fs/xfs/xfs_attr_item.c | 42 ++++++++++++++++++++++++++++++++++++------ fs/xfs/xfs_attr_list.c | 7 +++++++ 3 files changed, 60 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index f8f7445b063c..54edc690ac1e 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -87,6 +87,8 @@ xfs_attr_is_leaf( struct xfs_iext_cursor icur; struct xfs_bmbt_irec imap; + ASSERT(!xfs_need_iread_extents(ifp)); + if (ifp->if_nextents != 1 || ifp->if_format != XFS_DINODE_FMT_EXTENTS) return false; @@ -224,11 +226,21 @@ int xfs_attr_get_ilocked( struct xfs_da_args *args) { + int error; + xfs_assert_ilocked(args->dp, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL); if (!xfs_inode_hasattr(args->dp)) return -ENOATTR; + /* + * The incore attr fork iext tree must be loaded for xfs_attr_is_leaf + * to work correctly. + */ + error = xfs_iread_extents(args->trans, args->dp, XFS_ATTR_FORK); + if (error) + return error; + if (args->dp->i_af.if_format == XFS_DINODE_FMT_LOCAL) return xfs_attr_shortform_getvalue(args); if (xfs_attr_is_leaf(args->dp)) @@ -870,6 +882,11 @@ xfs_attr_lookup( return -ENOATTR; } + /* Prerequisite for xfs_attr_is_leaf */ + error = xfs_iread_extents(args->trans, args->dp, XFS_ATTR_FORK); + if (error) + return error; + if (xfs_attr_is_leaf(dp)) { error = xfs_attr_leaf_hasname(args, &bp); diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index d46034705694..541455731618 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -498,6 +498,25 @@ xfs_attri_validate( return xfs_verify_ino(mp, attrp->alfi_ino); } +static int +xfs_attri_iread_extents( + struct xfs_inode *ip) +{ + struct xfs_trans *tp; + int error; + + error = xfs_trans_alloc_empty(ip->i_mount, &tp); + if (error) + return error; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + error = xfs_iread_extents(tp, ip, XFS_ATTR_FORK); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_trans_cancel(tp); + + return error; +} + static inline struct xfs_attr_intent * xfs_attri_recover_work( struct xfs_mount *mp, @@ -508,13 +527,22 @@ xfs_attri_recover_work( { struct xfs_attr_intent *attr; struct xfs_da_args *args; + struct xfs_inode *ip; int local; int error; - error = xlog_recover_iget(mp, attrp->alfi_ino, ipp); + error = xlog_recover_iget(mp, attrp->alfi_ino, &ip); if (error) return ERR_PTR(error); + if (xfs_inode_has_attr_fork(ip)) { + error = xfs_attri_iread_extents(ip); + if (error) { + xfs_irele(ip); + return ERR_PTR(error); + } + } + attr = kzalloc(sizeof(struct xfs_attr_intent) + sizeof(struct xfs_da_args), GFP_KERNEL | __GFP_NOFAIL); args = (struct xfs_da_args *)(attr + 1); @@ -531,7 +559,7 @@ xfs_attri_recover_work( attr->xattri_nameval = xfs_attri_log_nameval_get(nv); ASSERT(attr->xattri_nameval); - args->dp = *ipp; + args->dp = ip; args->geo = mp->m_attr_geo; args->whichfork = XFS_ATTR_FORK; args->name = nv->name.i_addr; @@ -561,6 +589,7 @@ xfs_attri_recover_work( } xfs_defer_add_item(dfp, &attr->xattri_list); + *ipp = ip; return attr; } @@ -615,16 +644,17 @@ xfs_attr_recover_work( XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, &attrip->attri_format, sizeof(attrip->attri_format)); - if (error) { - xfs_trans_cancel(tp); - goto out_unlock; - } + if (error) + goto out_cancel; error = xfs_defer_ops_capture_and_commit(tp, capture_list); out_unlock: xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_irele(ip); return error; +out_cancel: + xfs_trans_cancel(tp); + goto out_unlock; } /* Re-log an intent item to push the log tail forward. */ diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index 6a621f016f04..97c8f3dcfb89 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -544,6 +544,7 @@ xfs_attr_list_ilocked( struct xfs_attr_list_context *context) { struct xfs_inode *dp = context->dp; + int error; xfs_assert_ilocked(dp, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL); @@ -554,6 +555,12 @@ xfs_attr_list_ilocked( return 0; if (dp->i_af.if_format == XFS_DINODE_FMT_LOCAL) return xfs_attr_shortform_list(context); + + /* Prerequisite for xfs_attr_is_leaf */ + error = xfs_iread_extents(NULL, dp, XFS_ATTR_FORK); + if (error) + return error; + if (xfs_attr_is_leaf(dp)) return xfs_attr_leaf_list(context); return xfs_attr_node_list(context); -- cgit v1.2.3-59-g8ed1b From 8ef1d96a985e4dc07ffbd71bd7fc5604a80cc644 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:24 -0700 Subject: xfs: require XFS_SB_FEAT_INCOMPAT_LOG_XATTRS for attr log intent item recovery The XFS_SB_FEAT_INCOMPAT_LOG_XATTRS feature bit protects a filesystem from old kernels that do not know how to recover extended attribute log intent items. Make this check mandatory instead of a debugging assert. Fixes: fd920008784ea ("xfs: Set up infrastructure for log attribute replay") Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 541455731618..dfe7039dac98 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -469,6 +469,9 @@ xfs_attri_validate( unsigned int op = attrp->alfi_op_flags & XFS_ATTRI_OP_FLAGS_TYPE_MASK; + if (!xfs_sb_version_haslogxattrs(&mp->m_sb)) + return false; + if (attrp->__pad != 0) return false; @@ -570,8 +573,6 @@ xfs_attri_recover_work( XFS_DA_OP_LOGGED; args->owner = args->dp->i_ino; - ASSERT(xfs_sb_version_haslogxattrs(&mp->m_sb)); - switch (attr->xattri_op_flags) { case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: -- cgit v1.2.3-59-g8ed1b From f759784cb61ceb77604326cd53cc2da88d24842f Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:25 -0700 Subject: xfs: use an XFS_OPSTATE_ flag for detecting if logged xattrs are available Per reviewer request, use an OPSTATE flag (+ helpers) to decide if logged xattrs are enabled, instead of querying the xfs_sb. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 2 +- fs/xfs/xfs_mount.c | 16 ++++++++++++++++ fs/xfs/xfs_mount.h | 6 +++++- fs/xfs/xfs_xattr.c | 3 ++- 4 files changed, 24 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index dfe7039dac98..e5e7ddbc594b 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -469,7 +469,7 @@ xfs_attri_validate( unsigned int op = attrp->alfi_op_flags & XFS_ATTRI_OP_FLAGS_TYPE_MASK; - if (!xfs_sb_version_haslogxattrs(&mp->m_sb)) + if (!xfs_is_using_logged_xattrs(mp)) return false; if (attrp->__pad != 0) diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index b2e5653b5200..09eef1721ef4 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -231,6 +231,13 @@ reread: mp->m_features |= xfs_sb_version_to_features(sbp); xfs_reinit_percpu_counters(mp); + /* + * If logged xattrs are enabled after log recovery finishes, then set + * the opstate so that log recovery will work properly. + */ + if (xfs_sb_version_haslogxattrs(&mp->m_sb)) + xfs_set_using_logged_xattrs(mp); + /* no need to be quiet anymore, so reset the buf ops */ bp->b_ops = &xfs_sb_buf_ops; @@ -829,6 +836,15 @@ xfs_mountfs( goto out_inodegc_shrinker; } + /* + * If logged xattrs are still enabled after log recovery finishes, then + * they'll be available until unmount. Otherwise, turn them off. + */ + if (xfs_sb_version_haslogxattrs(&mp->m_sb)) + xfs_set_using_logged_xattrs(mp); + else + xfs_clear_using_logged_xattrs(mp); + /* Enable background inode inactivation workers. */ xfs_inodegc_start(mp); xfs_blockgc_start(mp); diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index ca6f105990a2..d0567dfbc036 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -444,6 +444,8 @@ __XFS_HAS_FEAT(nouuid, NOUUID) #define XFS_OPSTATE_QUOTACHECK_RUNNING 10 /* Do we want to clear log incompat flags? */ #define XFS_OPSTATE_UNSET_LOG_INCOMPAT 11 +/* Filesystem can use logged extended attributes */ +#define XFS_OPSTATE_USE_LARP 12 #define __XFS_IS_OPSTATE(name, NAME) \ static inline bool xfs_is_ ## name (struct xfs_mount *mp) \ @@ -472,6 +474,7 @@ __XFS_IS_OPSTATE(quotacheck_running, QUOTACHECK_RUNNING) # define xfs_is_quotacheck_running(mp) (false) #endif __XFS_IS_OPSTATE(done_with_log_incompat, UNSET_LOG_INCOMPAT) +__XFS_IS_OPSTATE(using_logged_xattrs, USE_LARP) static inline bool xfs_should_warn(struct xfs_mount *mp, long nr) @@ -491,7 +494,8 @@ xfs_should_warn(struct xfs_mount *mp, long nr) { (1UL << XFS_OPSTATE_WARNED_SHRINK), "wshrink" }, \ { (1UL << XFS_OPSTATE_WARNED_LARP), "wlarp" }, \ { (1UL << XFS_OPSTATE_QUOTACHECK_RUNNING), "quotacheck" }, \ - { (1UL << XFS_OPSTATE_UNSET_LOG_INCOMPAT), "unset_log_incompat" } + { (1UL << XFS_OPSTATE_UNSET_LOG_INCOMPAT), "unset_log_incompat" }, \ + { (1UL << XFS_OPSTATE_USE_LARP), "logged_xattrs" } /* * Max and min values for mount-option defined I/O diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 0cbb93cf2869..ba56a9e73144 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -31,7 +31,7 @@ xfs_attr_grab_log_assist( int error = 0; /* xattr update log intent items are already enabled */ - if (xfs_sb_version_haslogxattrs(&mp->m_sb)) + if (xfs_is_using_logged_xattrs(mp)) return 0; /* @@ -48,6 +48,7 @@ xfs_attr_grab_log_assist( XFS_SB_FEAT_INCOMPAT_LOG_XATTRS); if (error) return error; + xfs_set_using_logged_xattrs(mp); xfs_warn_mount(mp, XFS_OPSTATE_WARNED_LARP, "EXPERIMENTAL logged extended attributes feature in use. Use at your own risk!"); -- cgit v1.2.3-59-g8ed1b From ad206ae50eca62836c5460ab5bbf2a6c59a268e7 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:26 -0700 Subject: xfs: check opcode and iovec count match in xlog_recover_attri_commit_pass2 Check that the number of recovered log iovecs is what is expected for the xattri opcode is expecting. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index e5e7ddbc594b..d3559e6b24b7 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -737,6 +737,7 @@ xlog_recover_attri_commit_pass2( const void *attr_value = NULL; const void *attr_name; size_t len; + unsigned int op; attri_formatp = item->ri_buf[0].i_addr; attr_name = item->ri_buf[1].i_addr; @@ -755,6 +756,32 @@ xlog_recover_attri_commit_pass2( return -EFSCORRUPTED; } + /* Check the number of log iovecs makes sense for the op code. */ + op = attri_formatp->alfi_op_flags & XFS_ATTRI_OP_FLAGS_TYPE_MASK; + switch (op) { + case XFS_ATTRI_OP_FLAGS_SET: + case XFS_ATTRI_OP_FLAGS_REPLACE: + /* Log item, attr name, attr value */ + if (item->ri_total != 3) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + break; + case XFS_ATTRI_OP_FLAGS_REMOVE: + /* Log item, attr name */ + if (item->ri_total != 2) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + break; + default: + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + /* Validate the attr name */ if (item->ri_buf[1].i_len != xlog_calc_iovec_len(attri_formatp->alfi_name_len)) { -- cgit v1.2.3-59-g8ed1b From f660ec8eaeb50d0317c29601aacabdb15e5f2203 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:27 -0700 Subject: xfs: fix missing check for invalid attr flags The xattr scrubber doesn't check for undefined flags in shortform attr entries. Therefore, define a mask XFS_ATTR_ONDISK_MASK that has all possible XFS_ATTR_* flags in it, and use that to check for unknown bits in xchk_xattr_actor. Refactor the check in the dabtree scanner function to use the new mask as well. The redundant checks need to be in place because the dabtree check examines the hash mappings and therefore needs to decode the attr leaf entries to compute the namehash. This happens before the walk of the xattr entries themselves. Fixes: ae0506eba78fd ("xfs: check used space of shortform xattr structures") Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_da_format.h | 5 +++++ fs/xfs/scrub/attr.c | 13 +++++++++---- 2 files changed, 14 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h index aac3fe039614..ecd0616f5776 100644 --- a/fs/xfs/libxfs/xfs_da_format.h +++ b/fs/xfs/libxfs/xfs_da_format.h @@ -719,8 +719,13 @@ struct xfs_attr3_leafblock { #define XFS_ATTR_ROOT (1u << XFS_ATTR_ROOT_BIT) #define XFS_ATTR_SECURE (1u << XFS_ATTR_SECURE_BIT) #define XFS_ATTR_INCOMPLETE (1u << XFS_ATTR_INCOMPLETE_BIT) + #define XFS_ATTR_NSP_ONDISK_MASK (XFS_ATTR_ROOT | XFS_ATTR_SECURE) +#define XFS_ATTR_ONDISK_MASK (XFS_ATTR_NSP_ONDISK_MASK | \ + XFS_ATTR_LOCAL | \ + XFS_ATTR_INCOMPLETE) + #define XFS_ATTR_NAMESPACE_STR \ { XFS_ATTR_LOCAL, "local" }, \ { XFS_ATTR_ROOT, "root" }, \ diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 5b855d7c9821..5ca79af47e81 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -192,6 +192,11 @@ xchk_xattr_actor( if (xchk_should_terminate(sc, &error)) return error; + if (attr_flags & ~XFS_ATTR_ONDISK_MASK) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, args.blkno); + return -ECANCELED; + } + if (attr_flags & XFS_ATTR_INCOMPLETE) { /* Incomplete attr key, just mark the inode for preening. */ xchk_ino_set_preen(sc, ip->i_ino); @@ -481,7 +486,6 @@ xchk_xattr_rec( xfs_dahash_t hash; int nameidx; int hdrsize; - unsigned int badflags; int error; ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC); @@ -511,10 +515,11 @@ xchk_xattr_rec( /* Retrieve the entry and check it. */ hash = be32_to_cpu(ent->hashval); - badflags = ~(XFS_ATTR_LOCAL | XFS_ATTR_ROOT | XFS_ATTR_SECURE | - XFS_ATTR_INCOMPLETE); - if ((ent->flags & badflags) != 0) + if (ent->flags & ~XFS_ATTR_ONDISK_MASK) { xchk_da_set_corrupt(ds, level); + return 0; + } + if (ent->flags & XFS_ATTR_LOCAL) { lentry = (struct xfs_attr_leaf_name_local *) (((char *)bp->b_addr) + nameidx); -- cgit v1.2.3-59-g8ed1b From 309dc9cbbb4379241bcc9b5a6a42c04279a0e5a7 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:28 -0700 Subject: xfs: check shortform attr entry flags specifically While reviewing flag checking in the attr scrub functions, we noticed that the shortform attr scanner didn't catch entries that have the LOCAL or INCOMPLETE bits set. Neither of these flags can ever be set on a shortform attr, so we need to check this narrower set of valid flags. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 5ca79af47e81..fd22d652a63a 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -579,6 +579,15 @@ xchk_xattr_check_sf( break; } + /* + * Shortform entries do not set LOCAL or INCOMPLETE, so the + * only valid flag bits here are for namespaces. + */ + if (sfe->flags & ~XFS_ATTR_NSP_ONDISK_MASK) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); + break; + } + if (!xchk_xattr_set_map(sc, ab->usedmap, (char *)sfe - (char *)sf, sizeof(struct xfs_attr_sf_entry))) { -- cgit v1.2.3-59-g8ed1b From 992c3b5c3fe6f42778436649ddae2b7a2984b7aa Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:29 -0700 Subject: xfs: restructure xfs_attr_complete_op a bit Eliminate the local variable from this function so that we can streamline things a bit later when we add the PPTR_REPLACE op code. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 54edc690ac1e..ba59dab6c56d 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -432,14 +432,13 @@ xfs_attr_complete_op( enum xfs_delattr_state replace_state) { struct xfs_da_args *args = attr->xattri_da_args; - bool do_replace = args->op_flags & XFS_DA_OP_REPLACE; + + if (!(args->op_flags & XFS_DA_OP_REPLACE)) + replace_state = XFS_DAS_DONE; args->op_flags &= ~XFS_DA_OP_REPLACE; args->attr_filter &= ~XFS_ATTR_INCOMPLETE; - if (do_replace) - return replace_state; - - return XFS_DAS_DONE; + return replace_state; } static int -- cgit v1.2.3-59-g8ed1b From 2a2c05d013d0562076ec475a6deb0991ce1942ca Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:29 -0700 Subject: xfs: use helpers to extract xattr op from opflags Create helper functions to extract the xattr op from the ondisk xattri log item and the incore attr intent item. These will get more use in the patches that follow. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.h | 5 +++++ fs/xfs/xfs_attr_item.c | 16 ++++++++++------ 2 files changed, 15 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index c8005f52102a..79b457adb7bd 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -529,6 +529,11 @@ struct xfs_attr_intent { struct xfs_bmbt_irec xattri_map; }; +static inline unsigned int +xfs_attr_intent_op(const struct xfs_attr_intent *attr) +{ + return attr->xattri_op_flags & XFS_ATTRI_OP_FLAGS_TYPE_MASK; +} /*======================================================================== * Function prototypes for the kernel. diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index d3559e6b24b7..b4c2dcb4581b 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -308,6 +308,12 @@ xfs_attrd_item_intent( return &ATTRD_ITEM(lip)->attrd_attrip->attri_item; } +static inline unsigned int +xfs_attr_log_item_op(const struct xfs_attri_log_format *attrp) +{ + return attrp->alfi_op_flags & XFS_ATTRI_OP_FLAGS_TYPE_MASK; +} + /* Log an attr to the intent item. */ STATIC void xfs_attr_log_item( @@ -466,8 +472,7 @@ xfs_attri_validate( struct xfs_mount *mp, struct xfs_attri_log_format *attrp) { - unsigned int op = attrp->alfi_op_flags & - XFS_ATTRI_OP_FLAGS_TYPE_MASK; + unsigned int op = xfs_attr_log_item_op(attrp); if (!xfs_is_using_logged_xattrs(mp)) return false; @@ -551,8 +556,7 @@ xfs_attri_recover_work( args = (struct xfs_da_args *)(attr + 1); attr->xattri_da_args = args; - attr->xattri_op_flags = attrp->alfi_op_flags & - XFS_ATTRI_OP_FLAGS_TYPE_MASK; + attr->xattri_op_flags = xfs_attr_log_item_op(attrp); /* * We're reconstructing the deferred work state structure from the @@ -573,7 +577,7 @@ xfs_attri_recover_work( XFS_DA_OP_LOGGED; args->owner = args->dp->i_ino; - switch (attr->xattri_op_flags) { + switch (xfs_attr_intent_op(attr)) { case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: args->value = nv->value.i_addr; @@ -757,7 +761,7 @@ xlog_recover_attri_commit_pass2( } /* Check the number of log iovecs makes sense for the op code. */ - op = attri_formatp->alfi_op_flags & XFS_ATTRI_OP_FLAGS_TYPE_MASK; + op = xfs_attr_log_item_op(attri_formatp); switch (op) { case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: -- cgit v1.2.3-59-g8ed1b From 1c7f09d210aba2f2bb206e2e8c97c9f11a3fd880 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:30 -0700 Subject: xfs: validate recovered name buffers when recovering xattr items Strengthen the xattri log item recovery code by checking that we actually have the required name and newname buffers for whatever operation we're replaying. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 58 ++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 47 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index b4c2dcb4581b..ebd6e98d9c66 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -741,22 +741,20 @@ xlog_recover_attri_commit_pass2( const void *attr_value = NULL; const void *attr_name; size_t len; - unsigned int op; - - attri_formatp = item->ri_buf[0].i_addr; - attr_name = item->ri_buf[1].i_addr; + unsigned int op, i = 0; /* Validate xfs_attri_log_format before the large memory allocation */ len = sizeof(struct xfs_attri_log_format); - if (item->ri_buf[0].i_len != len) { + if (item->ri_buf[i].i_len != len) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, item->ri_buf[0].i_addr, item->ri_buf[0].i_len); return -EFSCORRUPTED; } + attri_formatp = item->ri_buf[i].i_addr; if (!xfs_attri_validate(mp, attri_formatp)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, - item->ri_buf[0].i_addr, item->ri_buf[0].i_len); + attri_formatp, len); return -EFSCORRUPTED; } @@ -785,31 +783,69 @@ xlog_recover_attri_commit_pass2( attri_formatp, len); return -EFSCORRUPTED; } + i++; /* Validate the attr name */ - if (item->ri_buf[1].i_len != + if (item->ri_buf[i].i_len != xlog_calc_iovec_len(attri_formatp->alfi_name_len)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, - item->ri_buf[0].i_addr, item->ri_buf[0].i_len); + attri_formatp, len); return -EFSCORRUPTED; } + attr_name = item->ri_buf[i].i_addr; if (!xfs_attr_namecheck(attr_name, attri_formatp->alfi_name_len)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, - item->ri_buf[1].i_addr, item->ri_buf[1].i_len); + attri_formatp, len); return -EFSCORRUPTED; } + i++; /* Validate the attr value, if present */ if (attri_formatp->alfi_value_len != 0) { - if (item->ri_buf[2].i_len != xlog_calc_iovec_len(attri_formatp->alfi_value_len)) { + if (item->ri_buf[i].i_len != xlog_calc_iovec_len(attri_formatp->alfi_value_len)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, item->ri_buf[0].i_addr, item->ri_buf[0].i_len); return -EFSCORRUPTED; } - attr_value = item->ri_buf[2].i_addr; + attr_value = item->ri_buf[i].i_addr; + i++; + } + + /* + * Make sure we got the correct number of buffers for the operation + * that we just loaded. + */ + if (i != item->ri_total) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + + switch (op) { + case XFS_ATTRI_OP_FLAGS_REMOVE: + /* Regular remove operations operate only on names. */ + if (attr_value != NULL || attri_formatp->alfi_value_len != 0) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + fallthrough; + case XFS_ATTRI_OP_FLAGS_SET: + case XFS_ATTRI_OP_FLAGS_REPLACE: + /* + * Regular xattr set/remove/replace operations require a name + * and do not take a newname. Values are optional for set and + * replace. + */ + if (attr_name == NULL || attri_formatp->alfi_name_len == 0) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + break; } /* -- cgit v1.2.3-59-g8ed1b From 0aeeeb796980f74bf87ef175335ee1a9a1229767 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:31 -0700 Subject: xfs: always set args->value in xfs_attri_item_recover Always set args->value to the recovered value buffer. This reduces the amount of code in the switch statement, and hence the amount of thinking that I have to do. We validated the recovered buffers, supposedly. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index ebd6e98d9c66..8a13e2840692 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -572,6 +572,8 @@ xfs_attri_recover_work( args->name = nv->name.i_addr; args->namelen = nv->name.i_len; args->hashval = xfs_da_hashname(args->name, args->namelen); + args->value = nv->value.i_addr; + args->valuelen = nv->value.i_len; args->attr_filter = attrp->alfi_attr_filter & XFS_ATTRI_FILTER_MASK; args->op_flags = XFS_DA_OP_RECOVERY | XFS_DA_OP_OKNOENT | XFS_DA_OP_LOGGED; @@ -580,8 +582,6 @@ xfs_attri_recover_work( switch (xfs_attr_intent_op(attr)) { case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: - args->value = nv->value.i_addr; - args->valuelen = nv->value.i_len; args->total = xfs_attr_calc_size(args, &local); if (xfs_inode_hasattr(args->dp)) attr->xattri_dela_state = xfs_attr_init_replace_state(args); -- cgit v1.2.3-59-g8ed1b From c07f018bc094c5f30cb827ec9f11a23ace3435ec Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:32 -0700 Subject: xfs: use local variables for name and value length in _attri_commit_pass2 We're about to start using tagged unions in the xattr log format, so create a bunch of local variables in the recovery function so we only have to decode the log item fields once. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 8a13e2840692..59723e5f483e 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -738,9 +738,11 @@ xlog_recover_attri_commit_pass2( struct xfs_attri_log_item *attrip; struct xfs_attri_log_format *attri_formatp; struct xfs_attri_log_nameval *nv; - const void *attr_value = NULL; const void *attr_name; + const void *attr_value = NULL; size_t len; + unsigned int name_len = 0; + unsigned int value_len = 0; unsigned int op, i = 0; /* Validate xfs_attri_log_format before the large memory allocation */ @@ -769,6 +771,8 @@ xlog_recover_attri_commit_pass2( attri_formatp, len); return -EFSCORRUPTED; } + name_len = attri_formatp->alfi_name_len; + value_len = attri_formatp->alfi_value_len; break; case XFS_ATTRI_OP_FLAGS_REMOVE: /* Log item, attr name */ @@ -777,6 +781,7 @@ xlog_recover_attri_commit_pass2( attri_formatp, len); return -EFSCORRUPTED; } + name_len = attri_formatp->alfi_name_len; break; default: XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, @@ -786,15 +791,14 @@ xlog_recover_attri_commit_pass2( i++; /* Validate the attr name */ - if (item->ri_buf[i].i_len != - xlog_calc_iovec_len(attri_formatp->alfi_name_len)) { + if (item->ri_buf[i].i_len != xlog_calc_iovec_len(name_len)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, attri_formatp, len); return -EFSCORRUPTED; } attr_name = item->ri_buf[i].i_addr; - if (!xfs_attr_namecheck(attr_name, attri_formatp->alfi_name_len)) { + if (!xfs_attr_namecheck(attr_name, name_len)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, attri_formatp, len); return -EFSCORRUPTED; @@ -802,8 +806,8 @@ xlog_recover_attri_commit_pass2( i++; /* Validate the attr value, if present */ - if (attri_formatp->alfi_value_len != 0) { - if (item->ri_buf[i].i_len != xlog_calc_iovec_len(attri_formatp->alfi_value_len)) { + if (value_len != 0) { + if (item->ri_buf[i].i_len != xlog_calc_iovec_len(value_len)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, item->ri_buf[0].i_addr, item->ri_buf[0].i_len); @@ -827,7 +831,7 @@ xlog_recover_attri_commit_pass2( switch (op) { case XFS_ATTRI_OP_FLAGS_REMOVE: /* Regular remove operations operate only on names. */ - if (attr_value != NULL || attri_formatp->alfi_value_len != 0) { + if (attr_value != NULL || value_len != 0) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, attri_formatp, len); return -EFSCORRUPTED; @@ -840,7 +844,7 @@ xlog_recover_attri_commit_pass2( * and do not take a newname. Values are optional for set and * replace. */ - if (attr_name == NULL || attri_formatp->alfi_name_len == 0) { + if (attr_name == NULL || name_len == 0) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, attri_formatp, len); return -EFSCORRUPTED; @@ -853,9 +857,8 @@ xlog_recover_attri_commit_pass2( * name/value buffer to the recovered incore log item and drop our * reference. */ - nv = xfs_attri_log_nameval_alloc(attr_name, - attri_formatp->alfi_name_len, attr_value, - attri_formatp->alfi_value_len); + nv = xfs_attri_log_nameval_alloc(attr_name, name_len, + attr_value, value_len); attrip = xfs_attri_init(mp, nv); memcpy(&attrip->attri_format, attri_formatp, len); -- cgit v1.2.3-59-g8ed1b From 50855427c25426afbd98e9b4b00cb4a383614d88 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:33 -0700 Subject: xfs: refactor name/length checks in xfs_attri_validate Move the name and length checks into the attr op switch statement so that we can perform more specific checks of the value length. Over the next few patches we're going to add new attr op flags with different validation requirements. While we're at it, remove the incorrect comment. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 59723e5f483e..c8f92166b9ad 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -466,6 +466,12 @@ xfs_attri_item_match( return ATTRI_ITEM(lip)->attri_format.alfi_id == intent_id; } +static inline bool +xfs_attri_validate_namelen(unsigned int namelen) +{ + return namelen > 0 && namelen <= XATTR_NAME_MAX; +} + /* Is this recovered ATTRI format ok? */ static inline bool xfs_attri_validate( @@ -486,23 +492,24 @@ xfs_attri_validate( if (attrp->alfi_attr_filter & ~XFS_ATTRI_FILTER_MASK) return false; - /* alfi_op_flags should be either a set or remove */ switch (op) { case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: + if (attrp->alfi_value_len > XATTR_SIZE_MAX) + return false; + if (!xfs_attri_validate_namelen(attrp->alfi_name_len)) + return false; + break; case XFS_ATTRI_OP_FLAGS_REMOVE: + if (attrp->alfi_value_len != 0) + return false; + if (!xfs_attri_validate_namelen(attrp->alfi_name_len)) + return false; break; default: return false; } - if (attrp->alfi_value_len > XATTR_SIZE_MAX) - return false; - - if ((attrp->alfi_name_len > XATTR_NAME_MAX) || - (attrp->alfi_name_len == 0)) - return false; - return xfs_verify_ino(mp, attrp->alfi_ino); } -- cgit v1.2.3-59-g8ed1b From ffdcc3b8eb4d5ab263d04b9c4b2c6072c7b3c1e9 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:33 -0700 Subject: xfs: refactor name/value iovec validation in xlog_recover_attri_commit_pass2 Hoist the code that checks the attr name and value iovecs into separate helpers so that we can add more callsites for the new parent pointer attr intent items. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 64 ++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 46 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index c8f92166b9ad..39536303a7b6 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -734,6 +734,46 @@ const struct xfs_defer_op_type xfs_attr_defer_type = { .relog_intent = xfs_attr_relog_intent, }; +static inline void * +xfs_attri_validate_name_iovec( + struct xfs_mount *mp, + struct xfs_attri_log_format *attri_formatp, + const struct xfs_log_iovec *iovec, + unsigned int name_len) +{ + if (iovec->i_len != xlog_calc_iovec_len(name_len)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, sizeof(*attri_formatp)); + return NULL; + } + + if (!xfs_attr_namecheck(iovec->i_addr, name_len)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, sizeof(*attri_formatp)); + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + iovec->i_addr, iovec->i_len); + return NULL; + } + + return iovec->i_addr; +} + +static inline void * +xfs_attri_validate_value_iovec( + struct xfs_mount *mp, + struct xfs_attri_log_format *attri_formatp, + const struct xfs_log_iovec *iovec, + unsigned int value_len) +{ + if (iovec->i_len != xlog_calc_iovec_len(value_len)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, sizeof(*attri_formatp)); + return NULL; + } + + return iovec->i_addr; +} + STATIC int xlog_recover_attri_commit_pass2( struct xlog *log, @@ -798,30 +838,18 @@ xlog_recover_attri_commit_pass2( i++; /* Validate the attr name */ - if (item->ri_buf[i].i_len != xlog_calc_iovec_len(name_len)) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, - attri_formatp, len); - return -EFSCORRUPTED; - } - - attr_name = item->ri_buf[i].i_addr; - if (!xfs_attr_namecheck(attr_name, name_len)) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, - attri_formatp, len); + attr_name = xfs_attri_validate_name_iovec(mp, attri_formatp, + &item->ri_buf[i], name_len); + if (!attr_name) return -EFSCORRUPTED; - } i++; /* Validate the attr value, if present */ if (value_len != 0) { - if (item->ri_buf[i].i_len != xlog_calc_iovec_len(value_len)) { - XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, - item->ri_buf[0].i_addr, - item->ri_buf[0].i_len); + attr_value = xfs_attri_validate_value_iovec(mp, attri_formatp, + &item->ri_buf[i], value_len); + if (!attr_value) return -EFSCORRUPTED; - } - - attr_value = item->ri_buf[i].i_addr; i++; } -- cgit v1.2.3-59-g8ed1b From ea0b3e814741fb64e7785b564ea619578058e0b0 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:34 -0700 Subject: xfs: enforce one namespace per attribute Create a standardized helper function to enforce one namespace bit per extended attribute, and refactor all the open-coded hweight logic. This function is not a static inline to avoid porting hassles in userspace. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 11 +++++++++++ fs/xfs/libxfs/xfs_attr.h | 4 +++- fs/xfs/libxfs/xfs_attr_leaf.c | 7 ++++++- fs/xfs/scrub/attr.c | 12 +++++------- fs/xfs/scrub/attr_repair.c | 4 +--- fs/xfs/xfs_attr_item.c | 10 ++++++++-- fs/xfs/xfs_attr_list.c | 11 +++++++---- 7 files changed, 41 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index ba59dab6c56d..629fb25d149c 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -1532,12 +1532,23 @@ out_release: return error; } +/* Enforce that there is at most one namespace bit per attr. */ +inline bool xfs_attr_check_namespace(unsigned int attr_flags) +{ + return hweight32(attr_flags & XFS_ATTR_NSP_ONDISK_MASK) < 2; +} + /* Returns true if the attribute entry name is valid. */ bool xfs_attr_namecheck( + unsigned int attr_flags, const void *name, size_t length) { + /* Only one namespace bit allowed. */ + if (!xfs_attr_check_namespace(attr_flags)) + return false; + /* * MAXNAMELEN includes the trailing null, but (name/length) leave it * out, so use >= for the length check. diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index 79b457adb7bd..cd106b0a424f 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -560,7 +560,9 @@ enum xfs_attr_update { int xfs_attr_set(struct xfs_da_args *args, enum xfs_attr_update op); int xfs_attr_set_iter(struct xfs_attr_intent *attr); int xfs_attr_remove_iter(struct xfs_attr_intent *attr); -bool xfs_attr_namecheck(const void *name, size_t length); +bool xfs_attr_check_namespace(unsigned int attr_flags); +bool xfs_attr_namecheck(unsigned int attr_flags, const void *name, + size_t length); int xfs_attr_calc_size(struct xfs_da_args *args, int *local); void xfs_init_attr_trans(struct xfs_da_args *args, struct xfs_trans_res *tres, unsigned int *total); diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index 17ec5ff5a4e3..3b024ab892e6 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -950,6 +950,11 @@ xfs_attr_shortform_to_leaf( nargs.hashval = xfs_da_hashname(sfe->nameval, sfe->namelen); nargs.attr_filter = sfe->flags & XFS_ATTR_NSP_ONDISK_MASK; + if (!xfs_attr_check_namespace(sfe->flags)) { + xfs_da_mark_sick(args); + error = -EFSCORRUPTED; + goto out; + } error = xfs_attr3_leaf_lookup_int(bp, &nargs); /* set a->index */ ASSERT(error == -ENOATTR); error = xfs_attr3_leaf_add(bp, &nargs); @@ -1063,7 +1068,7 @@ xfs_attr_shortform_verify( * one namespace flag per xattr, so we can just count the * bits (i.e. hweight) here. */ - if (hweight8(sfep->flags & XFS_ATTR_NSP_ONDISK_MASK) > 1) + if (!xfs_attr_check_namespace(sfep->flags)) return __this_address; sfep = next_sfep; diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index fd22d652a63a..7789bd2f0950 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -203,14 +203,8 @@ xchk_xattr_actor( return 0; } - /* Only one namespace bit allowed. */ - if (hweight32(attr_flags & XFS_ATTR_NSP_ONDISK_MASK) > 1) { - xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, args.blkno); - return -ECANCELED; - } - /* Does this name make sense? */ - if (!xfs_attr_namecheck(name, namelen)) { + if (!xfs_attr_namecheck(attr_flags, name, namelen)) { xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, args.blkno); return -ECANCELED; } @@ -519,6 +513,10 @@ xchk_xattr_rec( xchk_da_set_corrupt(ds, level); return 0; } + if (!xfs_attr_check_namespace(ent->flags)) { + xchk_da_set_corrupt(ds, level); + return 0; + } if (ent->flags & XFS_ATTR_LOCAL) { lentry = (struct xfs_attr_leaf_name_local *) diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index 3066d662ea13..8b89c112c492 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -123,12 +123,10 @@ xrep_xattr_want_salvage( return false; if (namelen > XATTR_NAME_MAX || namelen <= 0) return false; - if (!xfs_attr_namecheck(name, namelen)) + if (!xfs_attr_namecheck(attr_flags, name, namelen)) return false; if (valuelen > XATTR_SIZE_MAX || valuelen < 0) return false; - if (hweight32(attr_flags & XFS_ATTR_NSP_ONDISK_MASK) > 1) - return false; return true; } diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 39536303a7b6..a65ac7479768 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -492,6 +492,10 @@ xfs_attri_validate( if (attrp->alfi_attr_filter & ~XFS_ATTRI_FILTER_MASK) return false; + if (!xfs_attr_check_namespace(attrp->alfi_attr_filter & + XFS_ATTR_NSP_ONDISK_MASK)) + return false; + switch (op) { case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: @@ -633,7 +637,8 @@ xfs_attr_recover_work( */ attrp = &attrip->attri_format; if (!xfs_attri_validate(mp, attrp) || - !xfs_attr_namecheck(nv->name.i_addr, nv->name.i_len)) + !xfs_attr_namecheck(attrp->alfi_attr_filter, nv->name.i_addr, + nv->name.i_len)) return -EFSCORRUPTED; attr = xfs_attri_recover_work(mp, dfp, attrp, &ip, nv); @@ -747,7 +752,8 @@ xfs_attri_validate_name_iovec( return NULL; } - if (!xfs_attr_namecheck(iovec->i_addr, name_len)) { + if (!xfs_attr_namecheck(attri_formatp->alfi_attr_filter, iovec->i_addr, + name_len)) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, attri_formatp, sizeof(*attri_formatp)); XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index 97c8f3dcfb89..903ed46c6887 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -82,7 +82,8 @@ xfs_attr_shortform_list( (dp->i_af.if_bytes + sf->count * 16) < context->bufsize)) { for (i = 0, sfe = xfs_attr_sf_firstentry(sf); i < sf->count; i++) { if (XFS_IS_CORRUPT(context->dp->i_mount, - !xfs_attr_namecheck(sfe->nameval, + !xfs_attr_namecheck(sfe->flags, + sfe->nameval, sfe->namelen))) { xfs_dirattr_mark_sick(context->dp, XFS_ATTR_FORK); return -EFSCORRUPTED; @@ -122,7 +123,8 @@ xfs_attr_shortform_list( for (i = 0, sfe = xfs_attr_sf_firstentry(sf); i < sf->count; i++) { if (unlikely( ((char *)sfe < (char *)sf) || - ((char *)sfe >= ((char *)sf + dp->i_af.if_bytes)))) { + ((char *)sfe >= ((char *)sf + dp->i_af.if_bytes)) || + !xfs_attr_check_namespace(sfe->flags))) { XFS_CORRUPTION_ERROR("xfs_attr_shortform_list", XFS_ERRLEVEL_LOW, context->dp->i_mount, sfe, @@ -177,7 +179,7 @@ xfs_attr_shortform_list( cursor->offset = 0; } if (XFS_IS_CORRUPT(context->dp->i_mount, - !xfs_attr_namecheck(sbp->name, + !xfs_attr_namecheck(sbp->flags, sbp->name, sbp->namelen))) { xfs_dirattr_mark_sick(context->dp, XFS_ATTR_FORK); error = -EFSCORRUPTED; @@ -502,7 +504,8 @@ xfs_attr3_leaf_list_int( } if (XFS_IS_CORRUPT(context->dp->i_mount, - !xfs_attr_namecheck(name, namelen))) { + !xfs_attr_namecheck(entry->flags, name, + namelen))) { xfs_dirattr_mark_sick(context->dp, XFS_ATTR_FORK); return -EFSCORRUPTED; } -- cgit v1.2.3-59-g8ed1b From 63211876ced33fbb730f515e8d830de53533fc82 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:35 -0700 Subject: xfs: rearrange xfs_attr_match parameters Rearrange the parameters to this function so that they match the order of attr listent: attr_flags -> name -> namelen -> value -> valuelen. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_leaf.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index 3b024ab892e6..bb00183d1349 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -510,9 +510,9 @@ xfs_attr3_leaf_read( static bool xfs_attr_match( struct xfs_da_args *args, - uint8_t namelen, - unsigned char *name, - int flags) + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen) { if (args->namelen != namelen) @@ -522,12 +522,12 @@ xfs_attr_match( /* Recovery ignores the INCOMPLETE flag. */ if ((args->op_flags & XFS_DA_OP_RECOVERY) && - args->attr_filter == (flags & XFS_ATTR_NSP_ONDISK_MASK)) + args->attr_filter == (attr_flags & XFS_ATTR_NSP_ONDISK_MASK)) return true; /* All remaining matches need to be filtered by INCOMPLETE state. */ if (args->attr_filter != - (flags & (XFS_ATTR_NSP_ONDISK_MASK | XFS_ATTR_INCOMPLETE))) + (attr_flags & (XFS_ATTR_NSP_ONDISK_MASK | XFS_ATTR_INCOMPLETE))) return false; return true; } @@ -746,8 +746,8 @@ xfs_attr_sf_findname( for (sfe = xfs_attr_sf_firstentry(sf); sfe < xfs_attr_sf_endptr(sf); sfe = xfs_attr_sf_nextentry(sfe)) { - if (xfs_attr_match(args, sfe->namelen, sfe->nameval, - sfe->flags)) + if (xfs_attr_match(args, sfe->flags, sfe->nameval, + sfe->namelen)) return sfe; } @@ -2443,15 +2443,16 @@ xfs_attr3_leaf_lookup_int( */ if (entry->flags & XFS_ATTR_LOCAL) { name_loc = xfs_attr3_leaf_name_local(leaf, probe); - if (!xfs_attr_match(args, name_loc->namelen, - name_loc->nameval, entry->flags)) + if (!xfs_attr_match(args, entry->flags, + name_loc->nameval, + name_loc->namelen)) continue; args->index = probe; return -EEXIST; } else { name_rmt = xfs_attr3_leaf_name_remote(leaf, probe); - if (!xfs_attr_match(args, name_rmt->namelen, - name_rmt->name, entry->flags)) + if (!xfs_attr_match(args, entry->flags, name_rmt->name, + name_rmt->namelen)) continue; args->index = probe; args->rmtvaluelen = be32_to_cpu(name_rmt->valuelen); -- cgit v1.2.3-59-g8ed1b From f49af061f49c004fb6df7f791f39f9ed370f767b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 22 Apr 2024 09:47:36 -0700 Subject: xfs: check the flags earlier in xfs_attr_match Checking the flags match is much cheaper than a memcmp, so do it early on in xfs_attr_match, and also add a little helper to calculate the match mask right under the comment explaining the logic for it. Signed-off-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_attr_leaf.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index bb00183d1349..c47fad39744e 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -507,6 +507,13 @@ xfs_attr3_leaf_read( * INCOMPLETE flag will not be set in attr->attr_filter, but rather * XFS_DA_OP_RECOVERY will be set in args->op_flags. */ +static inline unsigned int xfs_attr_match_mask(const struct xfs_da_args *args) +{ + if (args->op_flags & XFS_DA_OP_RECOVERY) + return XFS_ATTR_NSP_ONDISK_MASK; + return XFS_ATTR_NSP_ONDISK_MASK | XFS_ATTR_INCOMPLETE; +} + static bool xfs_attr_match( struct xfs_da_args *args, @@ -514,21 +521,15 @@ xfs_attr_match( const unsigned char *name, unsigned int namelen) { + unsigned int mask = xfs_attr_match_mask(args); if (args->namelen != namelen) return false; + if ((args->attr_filter & mask) != (attr_flags & mask)) + return false; if (memcmp(args->name, name, namelen) != 0) return false; - /* Recovery ignores the INCOMPLETE flag. */ - if ((args->op_flags & XFS_DA_OP_RECOVERY) && - args->attr_filter == (attr_flags & XFS_ATTR_NSP_ONDISK_MASK)) - return true; - - /* All remaining matches need to be filtered by INCOMPLETE state. */ - if (args->attr_filter != - (attr_flags & (XFS_ATTR_NSP_ONDISK_MASK | XFS_ATTR_INCOMPLETE))) - return false; return true; } -- cgit v1.2.3-59-g8ed1b From 9713dc88773d066413ae23aa474b13241507a89e Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:37 -0700 Subject: xfs: move xfs_attr_defer_add to xfs_attr_item.c Move the code that adds the incore xfs_attr_item deferred work data to a transaction live with the ATTRI log item code. This means that the upper level extended attribute code no longer has to know about the inner workings of the ATTRI log items. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 37 +++---------------------------------- fs/xfs/xfs_attr_item.c | 30 ++++++++++++++++++++++++++++++ fs/xfs/xfs_attr_item.h | 8 ++++++++ 3 files changed, 41 insertions(+), 34 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 629fb25d149c..50eab63ff3be 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -901,37 +901,6 @@ xfs_attr_lookup( return error; } -static void -xfs_attr_defer_add( - struct xfs_da_args *args, - unsigned int op_flags) -{ - - struct xfs_attr_intent *new; - - new = kmem_cache_zalloc(xfs_attr_intent_cache, - GFP_KERNEL | __GFP_NOFAIL); - new->xattri_op_flags = op_flags; - new->xattri_da_args = args; - - switch (op_flags) { - case XFS_ATTRI_OP_FLAGS_SET: - new->xattri_dela_state = xfs_attr_init_add_state(args); - break; - case XFS_ATTRI_OP_FLAGS_REPLACE: - new->xattri_dela_state = xfs_attr_init_replace_state(args); - break; - case XFS_ATTRI_OP_FLAGS_REMOVE: - new->xattri_dela_state = xfs_attr_init_remove_state(args); - break; - default: - ASSERT(0); - } - - xfs_defer_add(args->trans, &new->xattri_list, &xfs_attr_defer_type); - trace_xfs_attr_defer_add(new->xattri_dela_state, args->dp); -} - int xfs_attr_set( struct xfs_da_args *args, @@ -1021,14 +990,14 @@ xfs_attr_set( case -EEXIST: if (op == XFS_ATTRUPDATE_REMOVE) { /* if no value, we are performing a remove operation */ - xfs_attr_defer_add(args, XFS_ATTRI_OP_FLAGS_REMOVE); + xfs_attr_defer_add(args, XFS_ATTR_DEFER_REMOVE); break; } /* Pure create fails if the attr already exists */ if (op == XFS_ATTRUPDATE_CREATE) goto out_trans_cancel; - xfs_attr_defer_add(args, XFS_ATTRI_OP_FLAGS_REPLACE); + xfs_attr_defer_add(args, XFS_ATTR_DEFER_REPLACE); break; case -ENOATTR: /* Can't remove what isn't there. */ @@ -1038,7 +1007,7 @@ xfs_attr_set( /* Pure replace fails if no existing attr to replace. */ if (op == XFS_ATTRUPDATE_REPLACE) goto out_trans_cancel; - xfs_attr_defer_add(args, XFS_ATTRI_OP_FLAGS_SET); + xfs_attr_defer_add(args, XFS_ATTR_DEFER_SET); break; default: goto out_trans_cancel; diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index a65ac7479768..a7d6c9af47e8 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -727,6 +727,36 @@ xfs_attr_create_done( return &attrdp->attrd_item; } +void +xfs_attr_defer_add( + struct xfs_da_args *args, + enum xfs_attr_defer_op op) +{ + struct xfs_attr_intent *new; + + new = kmem_cache_zalloc(xfs_attr_intent_cache, + GFP_NOFS | __GFP_NOFAIL); + new->xattri_da_args = args; + + switch (op) { + case XFS_ATTR_DEFER_SET: + new->xattri_op_flags = XFS_ATTRI_OP_FLAGS_SET; + new->xattri_dela_state = xfs_attr_init_add_state(args); + break; + case XFS_ATTR_DEFER_REPLACE: + new->xattri_op_flags = XFS_ATTRI_OP_FLAGS_REPLACE; + new->xattri_dela_state = xfs_attr_init_replace_state(args); + break; + case XFS_ATTR_DEFER_REMOVE: + new->xattri_op_flags = XFS_ATTRI_OP_FLAGS_REMOVE; + new->xattri_dela_state = xfs_attr_init_remove_state(args); + break; + } + + xfs_defer_add(args->trans, &new->xattri_list, &xfs_attr_defer_type); + trace_xfs_attr_defer_add(new->xattri_dela_state, args->dp); +} + const struct xfs_defer_op_type xfs_attr_defer_type = { .name = "attr", .max_items = 1, diff --git a/fs/xfs/xfs_attr_item.h b/fs/xfs/xfs_attr_item.h index 3280a7930287..c32b669b0e16 100644 --- a/fs/xfs/xfs_attr_item.h +++ b/fs/xfs/xfs_attr_item.h @@ -51,4 +51,12 @@ struct xfs_attrd_log_item { extern struct kmem_cache *xfs_attri_cache; extern struct kmem_cache *xfs_attrd_cache; +enum xfs_attr_defer_op { + XFS_ATTR_DEFER_SET, + XFS_ATTR_DEFER_REMOVE, + XFS_ATTR_DEFER_REPLACE, +}; + +void xfs_attr_defer_add(struct xfs_da_args *args, enum xfs_attr_defer_op op); + #endif /* __XFS_ATTR_ITEM_H__ */ -- cgit v1.2.3-59-g8ed1b From a64e0134754bf88021e937aa34f1fbb5b524e585 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:38 -0700 Subject: xfs: create a separate hashname function for extended attributes Create a separate function to compute name hashvalues for extended attributes. When we get to parent pointers we'll be altering the rules so that metadump obfuscation doesn't turn heinous. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 28 ++++++++++++++++++++++++++-- fs/xfs/libxfs/xfs_attr.h | 14 ++++++++++++++ fs/xfs/libxfs/xfs_attr_leaf.c | 3 +-- fs/xfs/scrub/attr.c | 11 ++++++++--- fs/xfs/xfs_attr_item.c | 2 +- fs/xfs/xfs_attr_list.c | 5 ++++- 6 files changed, 54 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 50eab63ff3be..8262c263be9d 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -280,7 +280,7 @@ xfs_attr_get( args->owner = args->dp->i_ino; args->geo = args->dp->i_mount->m_attr_geo; args->whichfork = XFS_ATTR_FORK; - args->hashval = xfs_da_hashname(args->name, args->namelen); + xfs_attr_sethash(args); /* Entirely possible to look up a name which doesn't exist */ args->op_flags = XFS_DA_OP_OKNOENT; @@ -415,6 +415,30 @@ out: return error; } +/* Compute the hash value for a user/root/secure extended attribute */ +xfs_dahash_t +xfs_attr_hashname( + const uint8_t *name, + int namelen) +{ + return xfs_da_hashname(name, namelen); +} + +/* Compute the hash value for any extended attribute from any namespace. */ +xfs_dahash_t +xfs_attr_hashval( + struct xfs_mount *mp, + unsigned int attr_flags, + const uint8_t *name, + int namelen, + const void *value, + int valuelen) +{ + ASSERT(xfs_attr_check_namespace(attr_flags)); + + return xfs_attr_hashname(name, namelen); +} + /* * Handle the state change on completion of a multi-state attr operation. * @@ -925,7 +949,7 @@ xfs_attr_set( args->owner = args->dp->i_ino; args->geo = mp->m_attr_geo; args->whichfork = XFS_ATTR_FORK; - args->hashval = xfs_da_hashname(args->name, args->namelen); + xfs_attr_sethash(args); /* * We have no control over the attribute names that userspace passes us diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index cd106b0a424f..c63b1d610e53 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -628,6 +628,20 @@ xfs_attr_init_replace_state(struct xfs_da_args *args) return xfs_attr_init_add_state(args); } +xfs_dahash_t xfs_attr_hashname(const uint8_t *name, int namelen); + +xfs_dahash_t xfs_attr_hashval(struct xfs_mount *mp, unsigned int attr_flags, + const uint8_t *name, int namelen, const void *value, + int valuelen); + +/* Set the hash value for any extended attribute from any namespace. */ +static inline void xfs_attr_sethash(struct xfs_da_args *args) +{ + args->hashval = xfs_attr_hashval(args->dp->i_mount, args->attr_filter, + args->name, args->namelen, + args->value, args->valuelen); +} + extern struct kmem_cache *xfs_attr_intent_cache; int __init xfs_attr_intent_init_cache(void); void xfs_attr_intent_destroy_cache(void); diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index c47fad39744e..e54a8372a30a 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -948,14 +948,13 @@ xfs_attr_shortform_to_leaf( nargs.namelen = sfe->namelen; nargs.value = &sfe->nameval[nargs.namelen]; nargs.valuelen = sfe->valuelen; - nargs.hashval = xfs_da_hashname(sfe->nameval, - sfe->namelen); nargs.attr_filter = sfe->flags & XFS_ATTR_NSP_ONDISK_MASK; if (!xfs_attr_check_namespace(sfe->flags)) { xfs_da_mark_sick(args); error = -EFSCORRUPTED; goto out; } + xfs_attr_sethash(&nargs); error = xfs_attr3_leaf_lookup_int(bp, &nargs); /* set a->index */ ASSERT(error == -ENOATTR); error = xfs_attr3_leaf_add(bp, &nargs); diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 7789bd2f0950..22d7ef4df169 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -179,7 +179,6 @@ xchk_xattr_actor( .dp = ip, .name = name, .namelen = namelen, - .hashval = xfs_da_hashname(name, namelen), .trans = sc->tp, .valuelen = valuelen, .owner = ip->i_ino, @@ -230,6 +229,7 @@ xchk_xattr_actor( args.value = ab->value; + xfs_attr_sethash(&args); error = xfs_attr_get_ilocked(&args); /* ENODATA means the hash lookup failed and the attr is bad */ if (error == -ENODATA) @@ -525,7 +525,10 @@ xchk_xattr_rec( xchk_da_set_corrupt(ds, level); goto out; } - calc_hash = xfs_da_hashname(lentry->nameval, lentry->namelen); + calc_hash = xfs_attr_hashval(mp, ent->flags, lentry->nameval, + lentry->namelen, + lentry->nameval + lentry->namelen, + be16_to_cpu(lentry->valuelen)); } else { rentry = (struct xfs_attr_leaf_name_remote *) (((char *)bp->b_addr) + nameidx); @@ -533,7 +536,9 @@ xchk_xattr_rec( xchk_da_set_corrupt(ds, level); goto out; } - calc_hash = xfs_da_hashname(rentry->name, rentry->namelen); + calc_hash = xfs_attr_hashval(mp, ent->flags, rentry->name, + rentry->namelen, NULL, + be32_to_cpu(rentry->valuelen)); } if (calc_hash != hash) xchk_da_set_corrupt(ds, level); diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index a7d6c9af47e8..4a57bcff49eb 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -582,13 +582,13 @@ xfs_attri_recover_work( args->whichfork = XFS_ATTR_FORK; args->name = nv->name.i_addr; args->namelen = nv->name.i_len; - args->hashval = xfs_da_hashname(args->name, args->namelen); args->value = nv->value.i_addr; args->valuelen = nv->value.i_len; args->attr_filter = attrp->alfi_attr_filter & XFS_ATTRI_FILTER_MASK; args->op_flags = XFS_DA_OP_RECOVERY | XFS_DA_OP_OKNOENT | XFS_DA_OP_LOGGED; args->owner = args->dp->i_ino; + xfs_attr_sethash(args); switch (xfs_attr_intent_op(attr)) { case XFS_ATTRI_OP_FLAGS_SET: diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index 903ed46c6887..9bc4b5322539 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -135,12 +135,15 @@ xfs_attr_shortform_list( } sbp->entno = i; - sbp->hash = xfs_da_hashname(sfe->nameval, sfe->namelen); sbp->name = sfe->nameval; sbp->namelen = sfe->namelen; /* These are bytes, and both on-disk, don't endian-flip */ sbp->valuelen = sfe->valuelen; sbp->flags = sfe->flags; + sbp->hash = xfs_attr_hashval(dp->i_mount, sfe->flags, + sfe->nameval, sfe->namelen, + sfe->nameval + sfe->namelen, + sfe->valuelen); sfe = xfs_attr_sf_nextentry(sfe); sbp++; nsbuf++; -- cgit v1.2.3-59-g8ed1b From 98493ff878859eb0adefbc57a49ad47a92dfd252 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:38 -0700 Subject: xfs: add parent pointer support to attribute code Add the new parent attribute type. XFS_ATTR_PARENT is used only for parent pointer entries; it uses reserved blocks like XFS_ATTR_ROOT. Signed-off-by: Mark Tinguely Signed-off-by: Dave Chinner Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_da_format.h | 9 +++++++-- fs/xfs/libxfs/xfs_log_format.h | 1 + fs/xfs/xfs_trace.h | 3 ++- 3 files changed, 10 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h index ecd0616f5776..0c80f7ab9475 100644 --- a/fs/xfs/libxfs/xfs_da_format.h +++ b/fs/xfs/libxfs/xfs_da_format.h @@ -714,13 +714,17 @@ struct xfs_attr3_leafblock { #define XFS_ATTR_LOCAL_BIT 0 /* attr is stored locally */ #define XFS_ATTR_ROOT_BIT 1 /* limit access to trusted attrs */ #define XFS_ATTR_SECURE_BIT 2 /* limit access to secure attrs */ +#define XFS_ATTR_PARENT_BIT 3 /* parent pointer attrs */ #define XFS_ATTR_INCOMPLETE_BIT 7 /* attr in middle of create/delete */ #define XFS_ATTR_LOCAL (1u << XFS_ATTR_LOCAL_BIT) #define XFS_ATTR_ROOT (1u << XFS_ATTR_ROOT_BIT) #define XFS_ATTR_SECURE (1u << XFS_ATTR_SECURE_BIT) +#define XFS_ATTR_PARENT (1u << XFS_ATTR_PARENT_BIT) #define XFS_ATTR_INCOMPLETE (1u << XFS_ATTR_INCOMPLETE_BIT) -#define XFS_ATTR_NSP_ONDISK_MASK (XFS_ATTR_ROOT | XFS_ATTR_SECURE) +#define XFS_ATTR_NSP_ONDISK_MASK (XFS_ATTR_ROOT | \ + XFS_ATTR_SECURE | \ + XFS_ATTR_PARENT) #define XFS_ATTR_ONDISK_MASK (XFS_ATTR_NSP_ONDISK_MASK | \ XFS_ATTR_LOCAL | \ @@ -729,7 +733,8 @@ struct xfs_attr3_leafblock { #define XFS_ATTR_NAMESPACE_STR \ { XFS_ATTR_LOCAL, "local" }, \ { XFS_ATTR_ROOT, "root" }, \ - { XFS_ATTR_SECURE, "secure" } + { XFS_ATTR_SECURE, "secure" }, \ + { XFS_ATTR_PARENT, "parent" } /* * Alignment for namelist and valuelist entries (since they are mixed diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index accba2acd623..020aebd10143 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -1034,6 +1034,7 @@ struct xfs_icreate_log { */ #define XFS_ATTRI_FILTER_MASK (XFS_ATTR_ROOT | \ XFS_ATTR_SECURE | \ + XFS_ATTR_PARENT | \ XFS_ATTR_INCOMPLETE) /* diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 57f225ba7e8a..5621db48e763 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -91,7 +91,8 @@ struct xfs_exchrange; #define XFS_ATTR_FILTER_FLAGS \ { XFS_ATTR_ROOT, "ROOT" }, \ { XFS_ATTR_SECURE, "SECURE" }, \ - { XFS_ATTR_INCOMPLETE, "INCOMPLETE" } + { XFS_ATTR_INCOMPLETE, "INCOMPLETE" }, \ + { XFS_ATTR_PARENT, "PARENT" } DECLARE_EVENT_CLASS(xfs_attr_list_class, TP_PROTO(struct xfs_attr_list_context *ctx), -- cgit v1.2.3-59-g8ed1b From 8337d58ab2868f231a29824cd86d2e309bd36fa9 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:39 -0700 Subject: xfs: define parent pointer ondisk extended attribute format We need to define the parent pointer attribute format before we start adding support for it into all the code that needs to use it. The EA format we will use encodes the following information: name={dirent name} value={parent inumber, parent inode generation} hash=xfs_dir2_hashname(dirent name) ^ (parent_inumber) The inode/gen gives all the information we need to reliably identify the parent without requiring child->parent lock ordering, and allows userspace to do pathname component level reconstruction without the kernel ever needing to verify the parent itself as part of ioctl calls. By using the name-value lookup mode in the extended attribute code to match parent pointers using both the xattr name and value, we can identify the exact parent pointer EA we need to modify/remove in rename/unlink operations without searching the entire EA space. By storing the dirent name, we have enough information to be able to validate and reconstruct damaged directory trees. Earlier iterations of this patchset encoded the directory offset in the parent pointer key, but this format required repair to keep that in sync across directory rebuilds, which is unnecessary complexity. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_da_format.h | 13 +++++++++++++ fs/xfs/libxfs/xfs_ondisk.h | 1 + 2 files changed, 14 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h index 0c80f7ab9475..1395ad1937c5 100644 --- a/fs/xfs/libxfs/xfs_da_format.h +++ b/fs/xfs/libxfs/xfs_da_format.h @@ -890,4 +890,17 @@ static inline unsigned int xfs_dir2_dirblock_bytes(struct xfs_sb *sbp) xfs_failaddr_t xfs_da3_blkinfo_verify(struct xfs_buf *bp, struct xfs_da3_blkinfo *hdr3); +/* + * Parent pointer attribute format definition + * + * The xattr name contains the dirent name. + * The xattr value encodes the parent inode number and generation to ease + * opening parents by handle. + * The xattr hashval is xfs_dir2_namehash() ^ p_ino + */ +struct xfs_parent_rec { + __be64 p_ino; + __be32 p_gen; +} __packed; + #endif /* __XFS_DA_FORMAT_H__ */ diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 81885a6a028e..25952ef584ee 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -119,6 +119,7 @@ xfs_check_ondisk_structs(void) XFS_CHECK_OFFSET(xfs_dir2_sf_entry_t, offset, 1); XFS_CHECK_OFFSET(xfs_dir2_sf_entry_t, name, 3); XFS_CHECK_STRUCT_SIZE(xfs_dir2_sf_hdr_t, 10); + XFS_CHECK_STRUCT_SIZE(struct xfs_parent_rec, 12); /* log structures */ XFS_CHECK_STRUCT_SIZE(struct xfs_buf_log_format, 88); -- cgit v1.2.3-59-g8ed1b From f041455eb5773eda3291903ad6d1f33d4798e9a2 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:40 -0700 Subject: xfs: allow xattr matching on name and value for parent pointers If a file is hardlinked with the same name but from multiple parents, the parent pointers will all have the same dirent name (== attr name) but with different parent_ino/parent_gen values. To disambiguate, we need to be able to match on both the attr name and the attr value. This is in contrast to regular xattrs, which are matchtg edit d only on name. Therefore, plumb in the ability to match shortform and local attrs on name and value in the XFS_ATTR_PARENT namespace. Parent pointer attr values are never large enough to be stored in a remote attr, so we need can reject these cases as corruption. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_leaf.c | 52 ++++++++++++++++++++++++++++++++++++++----- 1 file changed, 46 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index e54a8372a30a..1a374c6885d7 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -514,12 +514,37 @@ static inline unsigned int xfs_attr_match_mask(const struct xfs_da_args *args) return XFS_ATTR_NSP_ONDISK_MASK | XFS_ATTR_INCOMPLETE; } +static inline bool +xfs_attr_parent_match( + const struct xfs_da_args *args, + const void *value, + unsigned int valuelen) +{ + ASSERT(args->value != NULL); + + /* Parent pointers do not use remote values */ + if (!value) + return false; + + /* + * The only value we support is a parent rec. However, we'll accept + * any valuelen so that offline repair can delete ATTR_PARENT values + * that are not parent pointers. + */ + if (valuelen != args->valuelen) + return false; + + return memcmp(args->value, value, valuelen) == 0; +} + static bool xfs_attr_match( struct xfs_da_args *args, unsigned int attr_flags, const unsigned char *name, - unsigned int namelen) + unsigned int namelen, + const void *value, + unsigned int valuelen) { unsigned int mask = xfs_attr_match_mask(args); @@ -530,6 +555,9 @@ xfs_attr_match( if (memcmp(args->name, name, namelen) != 0) return false; + if (attr_flags & XFS_ATTR_PARENT) + return xfs_attr_parent_match(args, value, valuelen); + return true; } @@ -539,6 +567,13 @@ xfs_attr_copy_value( unsigned char *value, int valuelen) { + /* + * Parent pointer lookups require the caller to specify the name and + * value, so don't copy anything. + */ + if (args->attr_filter & XFS_ATTR_PARENT) + return 0; + /* * No copy if all we have to do is get the length */ @@ -748,7 +783,8 @@ xfs_attr_sf_findname( sfe < xfs_attr_sf_endptr(sf); sfe = xfs_attr_sf_nextentry(sfe)) { if (xfs_attr_match(args, sfe->flags, sfe->nameval, - sfe->namelen)) + sfe->namelen, &sfe->nameval[sfe->namelen], + sfe->valuelen)) return sfe; } @@ -2444,18 +2480,22 @@ xfs_attr3_leaf_lookup_int( if (entry->flags & XFS_ATTR_LOCAL) { name_loc = xfs_attr3_leaf_name_local(leaf, probe); if (!xfs_attr_match(args, entry->flags, - name_loc->nameval, - name_loc->namelen)) + name_loc->nameval, name_loc->namelen, + &name_loc->nameval[name_loc->namelen], + be16_to_cpu(name_loc->valuelen))) continue; args->index = probe; return -EEXIST; } else { + unsigned int valuelen; + name_rmt = xfs_attr3_leaf_name_remote(leaf, probe); + valuelen = be32_to_cpu(name_rmt->valuelen); if (!xfs_attr_match(args, entry->flags, name_rmt->name, - name_rmt->namelen)) + name_rmt->namelen, NULL, valuelen)) continue; args->index = probe; - args->rmtvaluelen = be32_to_cpu(name_rmt->valuelen); + args->rmtvaluelen = valuelen; args->rmtblkno = be32_to_cpu(name_rmt->valueblk); args->rmtblkcnt = xfs_attr3_rmt_blocks( args->dp->i_mount, -- cgit v1.2.3-59-g8ed1b From a918f5f2cd2c9d2bf94f485c5cebbf47fb0627df Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:41 -0700 Subject: xfs: refactor xfs_is_using_logged_xattrs checks in attr item recovery Move this feature check down to the per-op checks so that we can ensure that we never see parent pointer attr items on non-pptr filesystems, and that logged xattrs are turned on for non-pptr attr items. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_attr_item.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 4a57bcff49eb..413e3d3959a5 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -480,9 +480,6 @@ xfs_attri_validate( { unsigned int op = xfs_attr_log_item_op(attrp); - if (!xfs_is_using_logged_xattrs(mp)) - return false; - if (attrp->__pad != 0) return false; @@ -499,12 +496,16 @@ xfs_attri_validate( switch (op) { case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: + if (!xfs_is_using_logged_xattrs(mp)) + return false; if (attrp->alfi_value_len > XATTR_SIZE_MAX) return false; if (!xfs_attri_validate_namelen(attrp->alfi_name_len)) return false; break; case XFS_ATTRI_OP_FLAGS_REMOVE: + if (!xfs_is_using_logged_xattrs(mp)) + return false; if (attrp->alfi_value_len != 0) return false; if (!xfs_attri_validate_namelen(attrp->alfi_name_len)) -- cgit v1.2.3-59-g8ed1b From 5773f7f82be5aa98e4883566072d33342814cebe Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:42 -0700 Subject: xfs: create attr log item opcodes and formats for parent pointers Make the necessary alterations to the extended attribute log intent item ondisk format so that we can log parent pointer operations. This requires the creation of new opcodes specific to parent pointers, and a new four-argument replace operation to handle renames. At this point this part of the patchset has changed so much from what Allison original wrote that I no longer think her SoB applies. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 19 +++ fs/xfs/libxfs/xfs_attr.h | 4 +- fs/xfs/libxfs/xfs_da_btree.h | 4 + fs/xfs/libxfs/xfs_log_format.h | 22 +++- fs/xfs/xfs_attr_item.c | 259 +++++++++++++++++++++++++++++++++++++---- fs/xfs/xfs_attr_item.h | 2 + 6 files changed, 284 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 8262c263be9d..78c87c405e33 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -439,6 +439,23 @@ xfs_attr_hashval( return xfs_attr_hashname(name, namelen); } +/* + * PPTR_REPLACE operations require the caller to set the old and new names and + * values explicitly. Update the canonical fields to the new name and value + * here now that the removal phase has finished. + */ +static void +xfs_attr_update_pptr_replace_args( + struct xfs_da_args *args) +{ + ASSERT(args->new_namelen > 0); + args->name = args->new_name; + args->namelen = args->new_namelen; + args->value = args->new_value; + args->valuelen = args->new_valuelen; + xfs_attr_sethash(args); +} + /* * Handle the state change on completion of a multi-state attr operation. * @@ -459,6 +476,8 @@ xfs_attr_complete_op( if (!(args->op_flags & XFS_DA_OP_REPLACE)) replace_state = XFS_DAS_DONE; + else if (xfs_attr_intent_op(attr) == XFS_ATTRI_OP_FLAGS_PPTR_REPLACE) + xfs_attr_update_pptr_replace_args(args); args->op_flags &= ~XFS_DA_OP_REPLACE; args->attr_filter &= ~XFS_ATTR_INCOMPLETE; diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index c63b1d610e53..d0ed7ea58ab0 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -510,8 +510,8 @@ struct xfs_attr_intent { struct xfs_da_args *xattri_da_args; /* - * Shared buffer containing the attr name and value so that the logging - * code can share large memory buffers between log items. + * Shared buffer containing the attr name, new name, and value so that + * the logging code can share large memory buffers between log items. */ struct xfs_attri_log_nameval *xattri_nameval; diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 17cef594b5bb..354d5d65043e 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -55,7 +55,9 @@ enum xfs_dacmp { typedef struct xfs_da_args { struct xfs_da_geometry *geo; /* da block geometry */ const uint8_t *name; /* string (maybe not NULL terminated) */ + const uint8_t *new_name; /* new attr name */ void *value; /* set of bytes (maybe contain NULLs) */ + void *new_value; /* new xattr value (may contain NULLs) */ struct xfs_inode *dp; /* directory inode to manipulate */ struct xfs_trans *trans; /* current trans (changes over time) */ @@ -63,10 +65,12 @@ typedef struct xfs_da_args { xfs_ino_t owner; /* inode that owns the dir/attr data */ int valuelen; /* length of value */ + int new_valuelen; /* length of new_value */ uint8_t filetype; /* filetype of inode for directories */ uint8_t op_flags; /* operation flags */ uint8_t attr_filter; /* XFS_ATTR_{ROOT,SECURE,INCOMPLETE} */ short namelen; /* length of string (maybe no NULL) */ + short new_namelen; /* length of new attr name */ xfs_dahash_t hashval; /* hash value of name */ xfs_extlen_t total; /* total blocks needed, for 1st bmap */ int whichfork; /* data or attribute fork */ diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index 020aebd10143..632dd9732455 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -115,11 +115,13 @@ struct xfs_unmount_log_format { #define XLOG_REG_TYPE_BUD_FORMAT 26 #define XLOG_REG_TYPE_ATTRI_FORMAT 27 #define XLOG_REG_TYPE_ATTRD_FORMAT 28 -#define XLOG_REG_TYPE_ATTR_NAME 29 +#define XLOG_REG_TYPE_ATTR_NAME 29 #define XLOG_REG_TYPE_ATTR_VALUE 30 #define XLOG_REG_TYPE_XMI_FORMAT 31 #define XLOG_REG_TYPE_XMD_FORMAT 32 -#define XLOG_REG_TYPE_MAX 32 +#define XLOG_REG_TYPE_ATTR_NEWNAME 33 +#define XLOG_REG_TYPE_ATTR_NEWVALUE 34 +#define XLOG_REG_TYPE_MAX 34 /* * Flags to log operation header @@ -1026,6 +1028,9 @@ struct xfs_icreate_log { #define XFS_ATTRI_OP_FLAGS_SET 1 /* Set the attribute */ #define XFS_ATTRI_OP_FLAGS_REMOVE 2 /* Remove the attribute */ #define XFS_ATTRI_OP_FLAGS_REPLACE 3 /* Replace the attribute */ +#define XFS_ATTRI_OP_FLAGS_PPTR_SET 4 /* Set parent pointer */ +#define XFS_ATTRI_OP_FLAGS_PPTR_REMOVE 5 /* Remove parent pointer */ +#define XFS_ATTRI_OP_FLAGS_PPTR_REPLACE 6 /* Replace parent pointer */ #define XFS_ATTRI_OP_FLAGS_TYPE_MASK 0xFF /* Flags type mask */ /* @@ -1048,7 +1053,18 @@ struct xfs_attri_log_format { uint64_t alfi_id; /* attri identifier */ uint64_t alfi_ino; /* the inode for this attr operation */ uint32_t alfi_op_flags; /* marks the op as a set or remove */ - uint32_t alfi_name_len; /* attr name length */ + union { + uint32_t alfi_name_len; /* attr name length */ + struct { + /* + * For PPTR_REPLACE, these are the lengths of the old + * and new attr names. The new and old values must + * have the same length. + */ + uint16_t alfi_old_name_len; + uint16_t alfi_new_name_len; + }; + }; uint32_t alfi_value_len; /* attr value length */ uint32_t alfi_attr_filter;/* attr filter flags */ }; diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 413e3d3959a5..be5064f5a531 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -73,8 +73,12 @@ static inline struct xfs_attri_log_nameval * xfs_attri_log_nameval_alloc( const void *name, unsigned int name_len, + const void *new_name, + unsigned int new_name_len, const void *value, - unsigned int value_len) + unsigned int value_len, + const void *new_value, + unsigned int new_value_len) { struct xfs_attri_log_nameval *nv; @@ -83,15 +87,26 @@ xfs_attri_log_nameval_alloc( * this. But kvmalloc() utterly sucks, so we use our own version. */ nv = xlog_kvmalloc(sizeof(struct xfs_attri_log_nameval) + - name_len + value_len); + name_len + new_name_len + value_len + + new_value_len); nv->name.i_addr = nv + 1; nv->name.i_len = name_len; nv->name.i_type = XLOG_REG_TYPE_ATTR_NAME; memcpy(nv->name.i_addr, name, name_len); + if (new_name_len) { + nv->new_name.i_addr = nv->name.i_addr + name_len; + nv->new_name.i_len = new_name_len; + memcpy(nv->new_name.i_addr, new_name, new_name_len); + } else { + nv->new_name.i_addr = NULL; + nv->new_name.i_len = 0; + } + nv->new_name.i_type = XLOG_REG_TYPE_ATTR_NEWNAME; + if (value_len) { - nv->value.i_addr = nv->name.i_addr + name_len; + nv->value.i_addr = nv->name.i_addr + name_len + new_name_len; nv->value.i_len = value_len; memcpy(nv->value.i_addr, value, value_len); } else { @@ -100,6 +115,17 @@ xfs_attri_log_nameval_alloc( } nv->value.i_type = XLOG_REG_TYPE_ATTR_VALUE; + if (new_value_len) { + nv->new_value.i_addr = nv->name.i_addr + name_len + + new_name_len + value_len; + nv->new_value.i_len = new_value_len; + memcpy(nv->new_value.i_addr, new_value, new_value_len); + } else { + nv->new_value.i_addr = NULL; + nv->new_value.i_len = 0; + } + nv->new_value.i_type = XLOG_REG_TYPE_ATTR_NEWVALUE; + refcount_set(&nv->refcount, 1); return nv; } @@ -145,11 +171,20 @@ xfs_attri_item_size( *nbytes += sizeof(struct xfs_attri_log_format) + xlog_calc_iovec_len(nv->name.i_len); - if (!nv->value.i_len) - return; + if (nv->new_name.i_len) { + *nvecs += 1; + *nbytes += xlog_calc_iovec_len(nv->new_name.i_len); + } - *nvecs += 1; - *nbytes += xlog_calc_iovec_len(nv->value.i_len); + if (nv->value.i_len) { + *nvecs += 1; + *nbytes += xlog_calc_iovec_len(nv->value.i_len); + } + + if (nv->new_value.i_len) { + *nvecs += 1; + *nbytes += xlog_calc_iovec_len(nv->new_value.i_len); + } } /* @@ -179,15 +214,28 @@ xfs_attri_item_format( ASSERT(nv->name.i_len > 0); attrip->attri_format.alfi_size++; + if (nv->new_name.i_len > 0) + attrip->attri_format.alfi_size++; + if (nv->value.i_len > 0) attrip->attri_format.alfi_size++; + if (nv->new_value.i_len > 0) + attrip->attri_format.alfi_size++; + xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_ATTRI_FORMAT, &attrip->attri_format, sizeof(struct xfs_attri_log_format)); xlog_copy_from_iovec(lv, &vecp, &nv->name); + + if (nv->new_name.i_len > 0) + xlog_copy_from_iovec(lv, &vecp, &nv->new_name); + if (nv->value.i_len > 0) xlog_copy_from_iovec(lv, &vecp, &nv->value); + + if (nv->new_value.i_len > 0) + xlog_copy_from_iovec(lv, &vecp, &nv->new_value); } /* @@ -322,6 +370,8 @@ xfs_attr_log_item( const struct xfs_attr_intent *attr) { struct xfs_attri_log_format *attrp; + struct xfs_attri_log_nameval *nv = attr->xattri_nameval; + struct xfs_da_args *args = attr->xattri_da_args; /* * At this point the xfs_attr_intent has been constructed, and we've @@ -329,13 +379,25 @@ xfs_attr_log_item( * structure with fields from this xfs_attr_intent */ attrp = &attrip->attri_format; - attrp->alfi_ino = attr->xattri_da_args->dp->i_ino; + attrp->alfi_ino = args->dp->i_ino; ASSERT(!(attr->xattri_op_flags & ~XFS_ATTRI_OP_FLAGS_TYPE_MASK)); attrp->alfi_op_flags = attr->xattri_op_flags; - attrp->alfi_value_len = attr->xattri_nameval->value.i_len; - attrp->alfi_name_len = attr->xattri_nameval->name.i_len; - ASSERT(!(attr->xattri_da_args->attr_filter & ~XFS_ATTRI_FILTER_MASK)); - attrp->alfi_attr_filter = attr->xattri_da_args->attr_filter; + attrp->alfi_value_len = nv->value.i_len; + + switch (xfs_attr_log_item_op(attrp)) { + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: + ASSERT(nv->value.i_len == nv->new_value.i_len); + + attrp->alfi_old_name_len = nv->name.i_len; + attrp->alfi_new_name_len = nv->new_name.i_len; + break; + default: + attrp->alfi_name_len = nv->name.i_len; + break; + } + + ASSERT(!(args->attr_filter & ~XFS_ATTRI_FILTER_MASK)); + attrp->alfi_attr_filter = args->attr_filter; } /* Get an ATTRI. */ @@ -374,8 +436,11 @@ xfs_attr_create_intent( * Transfer our reference to the name/value buffer to the * deferred work state structure. */ - attr->xattri_nameval = xfs_attri_log_nameval_alloc(args->name, - args->namelen, args->value, args->valuelen); + attr->xattri_nameval = xfs_attri_log_nameval_alloc( + args->name, args->namelen, + args->new_name, args->new_namelen, + args->value, args->valuelen, + args->new_value, args->new_valuelen); } attrip = xfs_attri_init(mp, attr->xattri_nameval); @@ -494,6 +559,17 @@ xfs_attri_validate( return false; switch (op) { + case XFS_ATTRI_OP_FLAGS_PPTR_SET: + case XFS_ATTRI_OP_FLAGS_PPTR_REMOVE: + if (!xfs_has_parent(mp)) + return false; + if (attrp->alfi_value_len != sizeof(struct xfs_parent_rec)) + return false; + if (!xfs_attri_validate_namelen(attrp->alfi_name_len)) + return false; + if (!(attrp->alfi_attr_filter & XFS_ATTR_PARENT)) + return false; + break; case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: if (!xfs_is_using_logged_xattrs(mp)) @@ -511,6 +587,18 @@ xfs_attri_validate( if (!xfs_attri_validate_namelen(attrp->alfi_name_len)) return false; break; + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: + if (!xfs_has_parent(mp)) + return false; + if (!xfs_attri_validate_namelen(attrp->alfi_old_name_len)) + return false; + if (!xfs_attri_validate_namelen(attrp->alfi_new_name_len)) + return false; + if (attrp->alfi_value_len != sizeof(struct xfs_parent_rec)) + return false; + if (!(attrp->alfi_attr_filter & XFS_ATTR_PARENT)) + return false; + break; default: return false; } @@ -583,8 +671,12 @@ xfs_attri_recover_work( args->whichfork = XFS_ATTR_FORK; args->name = nv->name.i_addr; args->namelen = nv->name.i_len; + args->new_name = nv->new_name.i_addr; + args->new_namelen = nv->new_name.i_len; args->value = nv->value.i_addr; args->valuelen = nv->value.i_len; + args->new_value = nv->new_value.i_addr; + args->new_valuelen = nv->new_value.i_len; args->attr_filter = attrp->alfi_attr_filter & XFS_ATTRI_FILTER_MASK; args->op_flags = XFS_DA_OP_RECOVERY | XFS_DA_OP_OKNOENT | XFS_DA_OP_LOGGED; @@ -592,6 +684,8 @@ xfs_attri_recover_work( xfs_attr_sethash(args); switch (xfs_attr_intent_op(attr)) { + case XFS_ATTRI_OP_FLAGS_PPTR_SET: + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: args->total = xfs_attr_calc_size(args, &local); @@ -600,6 +694,7 @@ xfs_attri_recover_work( else attr->xattri_dela_state = xfs_attr_init_add_state(args); break; + case XFS_ATTRI_OP_FLAGS_PPTR_REMOVE: case XFS_ATTRI_OP_FLAGS_REMOVE: attr->xattri_dela_state = xfs_attr_init_remove_state(args); break; @@ -700,7 +795,17 @@ xfs_attr_relog_intent( new_attrp->alfi_ino = old_attrp->alfi_ino; new_attrp->alfi_op_flags = old_attrp->alfi_op_flags; new_attrp->alfi_value_len = old_attrp->alfi_value_len; - new_attrp->alfi_name_len = old_attrp->alfi_name_len; + + switch (xfs_attr_log_item_op(old_attrp)) { + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: + new_attrp->alfi_new_name_len = old_attrp->alfi_new_name_len; + new_attrp->alfi_old_name_len = old_attrp->alfi_old_name_len; + break; + default: + new_attrp->alfi_name_len = old_attrp->alfi_name_len; + break; + } + new_attrp->alfi_attr_filter = old_attrp->alfi_attr_filter; return &new_attrip->attri_item; @@ -734,22 +839,61 @@ xfs_attr_defer_add( enum xfs_attr_defer_op op) { struct xfs_attr_intent *new; + unsigned int log_op = 0; + bool is_pptr = args->attr_filter & XFS_ATTR_PARENT; + + if (is_pptr) { + ASSERT(xfs_has_parent(args->dp->i_mount)); + ASSERT((args->attr_filter & ~XFS_ATTR_PARENT) == 0); + ASSERT(args->op_flags & XFS_DA_OP_LOGGED); + ASSERT(args->valuelen == sizeof(struct xfs_parent_rec)); + } new = kmem_cache_zalloc(xfs_attr_intent_cache, GFP_NOFS | __GFP_NOFAIL); new->xattri_da_args = args; + /* Compute log operation from the higher level op and namespace. */ switch (op) { case XFS_ATTR_DEFER_SET: - new->xattri_op_flags = XFS_ATTRI_OP_FLAGS_SET; - new->xattri_dela_state = xfs_attr_init_add_state(args); + if (is_pptr) + log_op = XFS_ATTRI_OP_FLAGS_PPTR_SET; + else + log_op = XFS_ATTRI_OP_FLAGS_SET; break; case XFS_ATTR_DEFER_REPLACE: - new->xattri_op_flags = XFS_ATTRI_OP_FLAGS_REPLACE; - new->xattri_dela_state = xfs_attr_init_replace_state(args); + if (is_pptr) + log_op = XFS_ATTRI_OP_FLAGS_PPTR_REPLACE; + else + log_op = XFS_ATTRI_OP_FLAGS_REPLACE; break; case XFS_ATTR_DEFER_REMOVE: - new->xattri_op_flags = XFS_ATTRI_OP_FLAGS_REMOVE; + if (is_pptr) + log_op = XFS_ATTRI_OP_FLAGS_PPTR_REMOVE; + else + log_op = XFS_ATTRI_OP_FLAGS_REMOVE; + break; + default: + ASSERT(0); + break; + } + new->xattri_op_flags = log_op; + + /* Set up initial attr operation state. */ + switch (log_op) { + case XFS_ATTRI_OP_FLAGS_PPTR_SET: + case XFS_ATTRI_OP_FLAGS_SET: + new->xattri_dela_state = xfs_attr_init_add_state(args); + break; + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: + ASSERT(args->new_valuelen == args->valuelen); + new->xattri_dela_state = xfs_attr_init_replace_state(args); + break; + case XFS_ATTRI_OP_FLAGS_REPLACE: + new->xattri_dela_state = xfs_attr_init_replace_state(args); + break; + case XFS_ATTRI_OP_FLAGS_PPTR_REMOVE: + case XFS_ATTRI_OP_FLAGS_REMOVE: new->xattri_dela_state = xfs_attr_init_remove_state(args); break; } @@ -824,9 +968,13 @@ xlog_recover_attri_commit_pass2( struct xfs_attri_log_nameval *nv; const void *attr_name; const void *attr_value = NULL; + const void *attr_new_name = NULL; + const void *attr_new_value = NULL; size_t len; unsigned int name_len = 0; unsigned int value_len = 0; + unsigned int new_name_len = 0; + unsigned int new_value_len = 0; unsigned int op, i = 0; /* Validate xfs_attri_log_format before the large memory allocation */ @@ -847,6 +995,17 @@ xlog_recover_attri_commit_pass2( /* Check the number of log iovecs makes sense for the op code. */ op = xfs_attr_log_item_op(attri_formatp); switch (op) { + case XFS_ATTRI_OP_FLAGS_PPTR_REMOVE: + case XFS_ATTRI_OP_FLAGS_PPTR_SET: + /* Log item, attr name, attr value */ + if (item->ri_total != 3) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + name_len = attri_formatp->alfi_name_len; + value_len = attri_formatp->alfi_value_len; + break; case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: /* Log item, attr name, attr value */ @@ -867,6 +1026,20 @@ xlog_recover_attri_commit_pass2( } name_len = attri_formatp->alfi_name_len; break; + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: + /* + * Log item, attr name, new attr name, attr value, new attr + * value + */ + if (item->ri_total != 5) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + name_len = attri_formatp->alfi_old_name_len; + new_name_len = attri_formatp->alfi_new_name_len; + new_value_len = value_len = attri_formatp->alfi_value_len; + break; default: XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, attri_formatp, len); @@ -881,6 +1054,16 @@ xlog_recover_attri_commit_pass2( return -EFSCORRUPTED; i++; + /* Validate the new attr name */ + if (new_name_len > 0) { + attr_new_name = xfs_attri_validate_name_iovec(mp, + attri_formatp, &item->ri_buf[i], + new_name_len); + if (!attr_new_name) + return -EFSCORRUPTED; + i++; + } + /* Validate the attr value, if present */ if (value_len != 0) { attr_value = xfs_attri_validate_value_iovec(mp, attri_formatp, @@ -890,6 +1073,16 @@ xlog_recover_attri_commit_pass2( i++; } + /* Validate the new attr value, if present */ + if (new_value_len != 0) { + attr_new_value = xfs_attri_validate_value_iovec(mp, + attri_formatp, &item->ri_buf[i], + new_value_len); + if (!attr_new_value) + return -EFSCORRUPTED; + i++; + } + /* * Make sure we got the correct number of buffers for the operation * that we just loaded. @@ -909,12 +1102,17 @@ xlog_recover_attri_commit_pass2( return -EFSCORRUPTED; } fallthrough; + case XFS_ATTRI_OP_FLAGS_PPTR_REMOVE: + case XFS_ATTRI_OP_FLAGS_PPTR_SET: case XFS_ATTRI_OP_FLAGS_SET: case XFS_ATTRI_OP_FLAGS_REPLACE: /* * Regular xattr set/remove/replace operations require a name * and do not take a newname. Values are optional for set and * replace. + * + * Name-value set/remove operations must have a name, do not + * take a newname, and can take a value. */ if (attr_name == NULL || name_len == 0) { XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, @@ -922,6 +1120,23 @@ xlog_recover_attri_commit_pass2( return -EFSCORRUPTED; } break; + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: + /* + * Name-value replace operations require the caller to + * specify the old and new names and values explicitly. + * Values are optional. + */ + if (attr_name == NULL || name_len == 0) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + if (attr_new_name == NULL || new_name_len == 0) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, len); + return -EFSCORRUPTED; + } + break; } /* @@ -930,7 +1145,9 @@ xlog_recover_attri_commit_pass2( * reference. */ nv = xfs_attri_log_nameval_alloc(attr_name, name_len, - attr_value, value_len); + attr_new_name, new_name_len, + attr_value, value_len, + attr_new_value, new_value_len); attrip = xfs_attri_init(mp, nv); memcpy(&attrip->attri_format, attri_formatp, len); diff --git a/fs/xfs/xfs_attr_item.h b/fs/xfs/xfs_attr_item.h index c32b669b0e16..e74128cbb722 100644 --- a/fs/xfs/xfs_attr_item.h +++ b/fs/xfs/xfs_attr_item.h @@ -13,7 +13,9 @@ struct kmem_zone; struct xfs_attri_log_nameval { struct xfs_log_iovec name; + struct xfs_log_iovec new_name; /* PPTR_REPLACE only */ struct xfs_log_iovec value; + struct xfs_log_iovec new_value; /* PPTR_REPLACE only */ refcount_t refcount; /* name and value follow the end of this struct */ -- cgit v1.2.3-59-g8ed1b From ae673f534a30976ce5e709c4535a59c12b786ef3 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:43 -0700 Subject: xfs: record inode generation in xattr update log intent items For parent pointer updates, record the i_generation of the file that is being updated so that we don't accidentally jump generations. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_log_format.h | 2 +- fs/xfs/xfs_attr_item.c | 33 +++++++++++++++++++++++++++------ 2 files changed, 28 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index 632dd9732455..3e6682ed656b 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -1049,7 +1049,7 @@ struct xfs_icreate_log { struct xfs_attri_log_format { uint16_t alfi_type; /* attri log item type */ uint16_t alfi_size; /* size of this item */ - uint32_t __pad; /* pad to 64 bit aligned */ + uint32_t alfi_igen; /* generation of alfi_ino for pptr ops */ uint64_t alfi_id; /* attri identifier */ uint64_t alfi_ino; /* the inode for this attr operation */ uint32_t alfi_op_flags; /* marks the op as a set or remove */ diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index be5064f5a531..2898eeb16366 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -388,9 +388,14 @@ xfs_attr_log_item( case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: ASSERT(nv->value.i_len == nv->new_value.i_len); + attrp->alfi_igen = VFS_I(args->dp)->i_generation; attrp->alfi_old_name_len = nv->name.i_len; attrp->alfi_new_name_len = nv->new_name.i_len; break; + case XFS_ATTRI_OP_FLAGS_PPTR_REMOVE: + case XFS_ATTRI_OP_FLAGS_PPTR_SET: + attrp->alfi_igen = VFS_I(args->dp)->i_generation; + fallthrough; default: attrp->alfi_name_len = nv->name.i_len; break; @@ -545,9 +550,6 @@ xfs_attri_validate( { unsigned int op = xfs_attr_log_item_op(attrp); - if (attrp->__pad != 0) - return false; - if (attrp->alfi_op_flags & ~XFS_ATTRI_OP_FLAGS_TYPE_MASK) return false; @@ -639,9 +641,27 @@ xfs_attri_recover_work( int local; int error; - error = xlog_recover_iget(mp, attrp->alfi_ino, &ip); - if (error) - return ERR_PTR(error); + /* + * Parent pointer attr items record the generation but regular logged + * xattrs do not; select the right iget function. + */ + switch (xfs_attr_log_item_op(attrp)) { + case XFS_ATTRI_OP_FLAGS_PPTR_SET: + case XFS_ATTRI_OP_FLAGS_PPTR_REPLACE: + case XFS_ATTRI_OP_FLAGS_PPTR_REMOVE: + error = xlog_recover_iget_handle(mp, attrp->alfi_ino, + attrp->alfi_igen, &ip); + break; + default: + error = xlog_recover_iget(mp, attrp->alfi_ino, &ip); + break; + } + if (error) { + xfs_irele(ip); + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, attrp, + sizeof(*attrp)); + return ERR_PTR(-EFSCORRUPTED); + } if (xfs_inode_has_attr_fork(ip)) { error = xfs_attri_iread_extents(ip); @@ -793,6 +813,7 @@ xfs_attr_relog_intent( new_attrp = &new_attrip->attri_format; new_attrp->alfi_ino = old_attrp->alfi_ino; + new_attrp->alfi_igen = old_attrp->alfi_igen; new_attrp->alfi_op_flags = old_attrp->alfi_op_flags; new_attrp->alfi_value_len = old_attrp->alfi_value_len; -- cgit v1.2.3-59-g8ed1b From 297da63379c6cba504a33aa7c526f36b148d4610 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:43 -0700 Subject: xfs: Expose init_xattrs in xfs_create_tmpfile Tmp files are used as part of rename operations and will need attr forks initialized for parent pointers. Expose the init_xattrs parameter to the calling function to initialize the fork. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_inode.c | 5 +++-- fs/xfs/xfs_inode.h | 2 +- fs/xfs/xfs_iops.c | 2 +- 3 files changed, 5 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 3c843223b4ed..060e4e767b51 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1185,6 +1185,7 @@ xfs_create_tmpfile( struct mnt_idmap *idmap, struct xfs_inode *dp, umode_t mode, + bool init_xattrs, struct xfs_inode **ipp) { struct xfs_mount *mp = dp->i_mount; @@ -1225,7 +1226,7 @@ xfs_create_tmpfile( error = xfs_dialloc(&tp, dp->i_ino, mode, &ino); if (!error) error = xfs_init_new_inode(idmap, tp, dp, ino, mode, - 0, 0, prid, false, &ip); + 0, 0, prid, init_xattrs, &ip); if (error) goto out_trans_cancel; @@ -3037,7 +3038,7 @@ xfs_rename_alloc_whiteout( int error; error = xfs_create_tmpfile(idmap, dp, S_IFCHR | WHITEOUT_MODE, - &tmpfile); + false, &tmpfile); if (error) return error; diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index a6da1ab8ab13..04a91e312993 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -522,7 +522,7 @@ int xfs_create(struct mnt_idmap *idmap, umode_t mode, dev_t rdev, bool need_xattr, struct xfs_inode **ipp); int xfs_create_tmpfile(struct mnt_idmap *idmap, - struct xfs_inode *dp, umode_t mode, + struct xfs_inode *dp, umode_t mode, bool init_xattrs, struct xfs_inode **ipp); int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode *ip); diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 659fd10c0cda..d32322f9ecde 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -201,7 +201,7 @@ xfs_generic_create( xfs_create_need_xattr(dir, default_acl, acl), &ip); } else { - error = xfs_create_tmpfile(idmap, XFS_I(dir), mode, &ip); + error = xfs_create_tmpfile(idmap, XFS_I(dir), mode, false, &ip); } if (unlikely(error)) goto out_free_acl; -- cgit v1.2.3-59-g8ed1b From a08d6729637428b6ef8c6a5a94d8c6db7b805a44 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:44 -0700 Subject: xfs: add parent pointer validator functions The attr name of a parent pointer is a string, and the attr value of a parent pointer is (more or less) a file handle. So we need to modify attr_namecheck to verify the parent pointer name, and add a xfs_parent_valuecheck function to sanitize the handle. At the same time, we need to validate attr values during log recovery if the xattr is really a parent pointer. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong [djwong: move functions to xfs_parent.c, adjust for new disk format] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/libxfs/xfs_attr.c | 5 +++ fs/xfs/libxfs/xfs_parent.c | 92 ++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_parent.h | 15 ++++++++ fs/xfs/xfs_attr_item.c | 10 +++++ 5 files changed, 123 insertions(+) create mode 100644 fs/xfs/libxfs/xfs_parent.c create mode 100644 fs/xfs/libxfs/xfs_parent.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 4e1eb3b6dbc4..4956ea9a307b 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -42,6 +42,7 @@ xfs-y += $(addprefix libxfs/, \ xfs_inode_buf.o \ xfs_log_rlimit.o \ xfs_ag_resv.o \ + xfs_parent.o \ xfs_rmap.o \ xfs_rmap_btree.o \ xfs_refcount.o \ diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 78c87c405e33..93524efa6e56 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -26,6 +26,7 @@ #include "xfs_trace.h" #include "xfs_attr_item.h" #include "xfs_xattr.h" +#include "xfs_parent.h" struct kmem_cache *xfs_attr_intent_cache; @@ -1568,6 +1569,10 @@ xfs_attr_namecheck( if (length >= MAXNAMELEN) return false; + /* Parent pointers have their own validation. */ + if (attr_flags & XFS_ATTR_PARENT) + return xfs_parent_namecheck(attr_flags, name, length); + /* There shouldn't be any nulls here */ return !memchr(name, 0, length); } diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c new file mode 100644 index 000000000000..5961fa8c8561 --- /dev/null +++ b/fs/xfs/libxfs/xfs_parent.c @@ -0,0 +1,92 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2022-2024 Oracle. + * All rights reserved. + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_da_format.h" +#include "xfs_log_format.h" +#include "xfs_shared.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_error.h" +#include "xfs_trace.h" +#include "xfs_trans.h" +#include "xfs_da_btree.h" +#include "xfs_attr.h" +#include "xfs_dir2.h" +#include "xfs_dir2_priv.h" +#include "xfs_attr_sf.h" +#include "xfs_bmap.h" +#include "xfs_defer.h" +#include "xfs_log.h" +#include "xfs_xattr.h" +#include "xfs_parent.h" +#include "xfs_trans_space.h" + +/* + * Parent pointer attribute handling. + * + * Because the attribute name is a filename component, it will never be longer + * than 255 bytes and must not contain nulls or slashes. These are roughly the + * same constraints that apply to attribute names. + * + * The attribute value must always be a struct xfs_parent_rec. This means the + * attribute will never be in remote format because 12 bytes is nowhere near + * xfs_attr_leaf_entsize_local_max() (~75% of block size). + * + * Creating a new parent attribute will always create a new attribute - there + * should never, ever be an existing attribute in the tree for a new inode. + * ENOSPC behavior is problematic - creating the inode without the parent + * pointer is effectively a corruption, so we allow parent attribute creation + * to dip into the reserve block pool to avoid unexpected ENOSPC errors from + * occurring. + */ + +/* Return true if parent pointer attr name is valid. */ +bool +xfs_parent_namecheck( + unsigned int attr_flags, + const void *name, + size_t length) +{ + /* + * Parent pointers always use logged operations, so there should never + * be incomplete xattrs. + */ + if (attr_flags & XFS_ATTR_INCOMPLETE) + return false; + + return xfs_dir2_namecheck(name, length); +} + +/* Return true if parent pointer attr value is valid. */ +bool +xfs_parent_valuecheck( + struct xfs_mount *mp, + const void *value, + size_t valuelen) +{ + const struct xfs_parent_rec *rec = value; + + if (!xfs_has_parent(mp)) + return false; + + /* The xattr value must be a parent record. */ + if (valuelen != sizeof(struct xfs_parent_rec)) + return false; + + /* The parent record must be local. */ + if (value == NULL) + return false; + + /* The parent inumber must be valid. */ + if (!xfs_verify_dir_ino(mp, be64_to_cpu(rec->p_ino))) + return false; + + return true; +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h new file mode 100644 index 000000000000..ef8aff860780 --- /dev/null +++ b/fs/xfs/libxfs/xfs_parent.h @@ -0,0 +1,15 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2022-2024 Oracle. + * All Rights Reserved. + */ +#ifndef __XFS_PARENT_H__ +#define __XFS_PARENT_H__ + +/* Metadata validators */ +bool xfs_parent_namecheck(unsigned int attr_flags, const void *name, + size_t length); +bool xfs_parent_valuecheck(struct xfs_mount *mp, const void *value, + size_t valuelen); + +#endif /* __XFS_PARENT_H__ */ diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 2898eeb16366..2b10ac4c5fce 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -27,6 +27,7 @@ #include "xfs_error.h" #include "xfs_log_priv.h" #include "xfs_log_recover.h" +#include "xfs_parent.h" struct kmem_cache *xfs_attri_cache; struct kmem_cache *xfs_attrd_cache; @@ -973,6 +974,15 @@ xfs_attri_validate_value_iovec( return NULL; } + if ((attri_formatp->alfi_attr_filter & XFS_ATTR_PARENT) && + !xfs_parent_valuecheck(mp, iovec->i_addr, value_len)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + attri_formatp, sizeof(*attri_formatp)); + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, + iovec->i_addr, iovec->i_len); + return NULL; + } + return iovec->i_addr; } -- cgit v1.2.3-59-g8ed1b From 7dba4a5fe1c5cdf0859830380c52f29295cbf345 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:45 -0700 Subject: xfs: extend transaction reservations for parent attributes We need to add, remove or modify parent pointer attributes during create/link/unlink/rename operations atomically with the dirents in the parent directories being modified. This means they need to be modified in the same transaction as the parent directories, and so we need to add the required space for the attribute modifications to the transaction reservations. Signed-off-by: Dave Chinner Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong [djwong: fix indenting errors, adjust for new log format] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_trans_resv.c | 326 ++++++++++++++++++++++++++++++++++------- 1 file changed, 274 insertions(+), 52 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c index 6cd45e8c118d..6dbe6e7251e7 100644 --- a/fs/xfs/libxfs/xfs_trans_resv.c +++ b/fs/xfs/libxfs/xfs_trans_resv.c @@ -20,6 +20,9 @@ #include "xfs_qm.h" #include "xfs_trans_space.h" #include "xfs_rtbitmap.h" +#include "xfs_attr_item.h" +#include "xfs_log.h" +#include "xfs_da_format.h" #define _ALLOC true #define _FREE false @@ -422,29 +425,110 @@ xfs_calc_itruncate_reservation_minlogsize( return xfs_calc_itruncate_reservation(mp, true); } +static inline unsigned int xfs_calc_pptr_link_overhead(void) +{ + return sizeof(struct xfs_attri_log_format) + + xlog_calc_iovec_len(sizeof(struct xfs_parent_rec)) + + xlog_calc_iovec_len(MAXNAMELEN - 1); +} +static inline unsigned int xfs_calc_pptr_unlink_overhead(void) +{ + return sizeof(struct xfs_attri_log_format) + + xlog_calc_iovec_len(sizeof(struct xfs_parent_rec)) + + xlog_calc_iovec_len(MAXNAMELEN - 1); +} +static inline unsigned int xfs_calc_pptr_replace_overhead(void) +{ + return sizeof(struct xfs_attri_log_format) + + xlog_calc_iovec_len(sizeof(struct xfs_parent_rec)) + + xlog_calc_iovec_len(MAXNAMELEN - 1) + + xlog_calc_iovec_len(sizeof(struct xfs_parent_rec)) + + xlog_calc_iovec_len(MAXNAMELEN - 1); +} + /* * In renaming a files we can modify: * the five inodes involved: 5 * inode size * the two directory btrees: 2 * (max depth + v2) * dir block size * the two directory bmap btrees: 2 * max depth * block size * And the bmap_finish transaction can free dir and bmap blocks (two sets - * of bmap blocks) giving: + * of bmap blocks) giving (t2): * the agf for the ags in which the blocks live: 3 * sector size * the agfl for the ags in which the blocks live: 3 * sector size * the superblock for the free block count: sector size * the allocation btrees: 3 exts * 2 trees * (2 * max depth - 1) * block size + * If parent pointers are enabled (t3), then each transaction in the chain + * must be capable of setting or removing the extended attribute + * containing the parent information. It must also be able to handle + * the three xattr intent items that track the progress of the parent + * pointer update. */ STATIC uint xfs_calc_rename_reservation( struct xfs_mount *mp) { - return XFS_DQUOT_LOGRES(mp) + - max((xfs_calc_inode_res(mp, 5) + - xfs_calc_buf_res(2 * XFS_DIROP_LOG_COUNT(mp), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(7, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(xfs_allocfree_block_count(mp, 3), - XFS_FSB_TO_B(mp, 1)))); + unsigned int overhead = XFS_DQUOT_LOGRES(mp); + struct xfs_trans_resv *resp = M_RES(mp); + unsigned int t1, t2, t3 = 0; + + t1 = xfs_calc_inode_res(mp, 5) + + xfs_calc_buf_res(2 * XFS_DIROP_LOG_COUNT(mp), + XFS_FSB_TO_B(mp, 1)); + + t2 = xfs_calc_buf_res(7, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(xfs_allocfree_block_count(mp, 3), + XFS_FSB_TO_B(mp, 1)); + + if (xfs_has_parent(mp)) { + unsigned int rename_overhead, exchange_overhead; + + t3 = max(resp->tr_attrsetm.tr_logres, + resp->tr_attrrm.tr_logres); + + /* + * For a standard rename, the three xattr intent log items + * are (1) replacing the pptr for the source file; (2) + * removing the pptr on the dest file; and (3) adding a + * pptr for the whiteout file in the src dir. + * + * For an RENAME_EXCHANGE, there are two xattr intent + * items to replace the pptr for both src and dest + * files. Link counts don't change and there is no + * whiteout. + * + * In the worst case we can end up relogging all log + * intent items to allow the log tail to move ahead, so + * they become overhead added to each transaction in a + * processing chain. + */ + rename_overhead = xfs_calc_pptr_replace_overhead() + + xfs_calc_pptr_unlink_overhead() + + xfs_calc_pptr_link_overhead(); + exchange_overhead = 2 * xfs_calc_pptr_replace_overhead(); + + overhead += max(rename_overhead, exchange_overhead); + } + + return overhead + max3(t1, t2, t3); +} + +static inline unsigned int +xfs_rename_log_count( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + /* One for the rename, one more for freeing blocks */ + unsigned int ret = XFS_RENAME_LOG_COUNT; + + /* + * Pre-reserve enough log reservation to handle the transaction + * rolling needed to remove or add one parent pointer. + */ + if (xfs_has_parent(mp)) + ret += max(resp->tr_attrsetm.tr_logcount, + resp->tr_attrrm.tr_logcount); + + return ret; } /* @@ -461,6 +545,23 @@ xfs_calc_iunlink_remove_reservation( 2 * M_IGEO(mp)->inode_cluster_size; } +static inline unsigned int +xfs_link_log_count( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + unsigned int ret = XFS_LINK_LOG_COUNT; + + /* + * Pre-reserve enough log reservation to handle the transaction + * rolling needed to add one parent pointer. + */ + if (xfs_has_parent(mp)) + ret += resp->tr_attrsetm.tr_logcount; + + return ret; +} + /* * For creating a link to an inode: * the parent directory inode: inode size @@ -477,14 +578,23 @@ STATIC uint xfs_calc_link_reservation( struct xfs_mount *mp) { - return XFS_DQUOT_LOGRES(mp) + - xfs_calc_iunlink_remove_reservation(mp) + - max((xfs_calc_inode_res(mp, 2) + - xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(xfs_allocfree_block_count(mp, 1), - XFS_FSB_TO_B(mp, 1)))); + unsigned int overhead = XFS_DQUOT_LOGRES(mp); + struct xfs_trans_resv *resp = M_RES(mp); + unsigned int t1, t2, t3 = 0; + + overhead += xfs_calc_iunlink_remove_reservation(mp); + t1 = xfs_calc_inode_res(mp, 2) + + xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1)); + t2 = xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(xfs_allocfree_block_count(mp, 1), + XFS_FSB_TO_B(mp, 1)); + + if (xfs_has_parent(mp)) { + t3 = resp->tr_attrsetm.tr_logres; + overhead += xfs_calc_pptr_link_overhead(); + } + + return overhead + max3(t1, t2, t3); } /* @@ -499,6 +609,23 @@ xfs_calc_iunlink_add_reservation(xfs_mount_t *mp) M_IGEO(mp)->inode_cluster_size; } +static inline unsigned int +xfs_remove_log_count( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + unsigned int ret = XFS_REMOVE_LOG_COUNT; + + /* + * Pre-reserve enough log reservation to handle the transaction + * rolling needed to add one parent pointer. + */ + if (xfs_has_parent(mp)) + ret += resp->tr_attrrm.tr_logcount; + + return ret; +} + /* * For removing a directory entry we can modify: * the parent directory inode: inode size @@ -515,14 +642,24 @@ STATIC uint xfs_calc_remove_reservation( struct xfs_mount *mp) { - return XFS_DQUOT_LOGRES(mp) + - xfs_calc_iunlink_add_reservation(mp) + - max((xfs_calc_inode_res(mp, 2) + - xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), - XFS_FSB_TO_B(mp, 1))), - (xfs_calc_buf_res(4, mp->m_sb.sb_sectsize) + - xfs_calc_buf_res(xfs_allocfree_block_count(mp, 2), - XFS_FSB_TO_B(mp, 1)))); + unsigned int overhead = XFS_DQUOT_LOGRES(mp); + struct xfs_trans_resv *resp = M_RES(mp); + unsigned int t1, t2, t3 = 0; + + overhead += xfs_calc_iunlink_add_reservation(mp); + + t1 = xfs_calc_inode_res(mp, 2) + + xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1)); + t2 = xfs_calc_buf_res(4, mp->m_sb.sb_sectsize) + + xfs_calc_buf_res(xfs_allocfree_block_count(mp, 2), + XFS_FSB_TO_B(mp, 1)); + + if (xfs_has_parent(mp)) { + t3 = resp->tr_attrrm.tr_logres; + overhead += xfs_calc_pptr_unlink_overhead(); + } + + return overhead + max3(t1, t2, t3); } /* @@ -571,12 +708,40 @@ xfs_calc_icreate_resv_alloc( xfs_calc_finobt_res(mp); } +static inline unsigned int +xfs_icreate_log_count( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + unsigned int ret = XFS_CREATE_LOG_COUNT; + + /* + * Pre-reserve enough log reservation to handle the transaction + * rolling needed to add one parent pointer. + */ + if (xfs_has_parent(mp)) + ret += resp->tr_attrsetm.tr_logcount; + + return ret; +} + STATIC uint -xfs_calc_icreate_reservation(xfs_mount_t *mp) +xfs_calc_icreate_reservation( + struct xfs_mount *mp) { - return XFS_DQUOT_LOGRES(mp) + - max(xfs_calc_icreate_resv_alloc(mp), - xfs_calc_create_resv_modify(mp)); + struct xfs_trans_resv *resp = M_RES(mp); + unsigned int overhead = XFS_DQUOT_LOGRES(mp); + unsigned int t1, t2, t3 = 0; + + t1 = xfs_calc_icreate_resv_alloc(mp); + t2 = xfs_calc_create_resv_modify(mp); + + if (xfs_has_parent(mp)) { + t3 = resp->tr_attrsetm.tr_logres; + overhead += xfs_calc_pptr_link_overhead(); + } + + return overhead + max3(t1, t2, t3); } STATIC uint @@ -589,6 +754,23 @@ xfs_calc_create_tmpfile_reservation( return res + xfs_calc_iunlink_add_reservation(mp); } +static inline unsigned int +xfs_mkdir_log_count( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + unsigned int ret = XFS_MKDIR_LOG_COUNT; + + /* + * Pre-reserve enough log reservation to handle the transaction + * rolling needed to add one parent pointer. + */ + if (xfs_has_parent(mp)) + ret += resp->tr_attrsetm.tr_logcount; + + return ret; +} + /* * Making a new directory is the same as creating a new file. */ @@ -599,6 +781,22 @@ xfs_calc_mkdir_reservation( return xfs_calc_icreate_reservation(mp); } +static inline unsigned int +xfs_symlink_log_count( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + unsigned int ret = XFS_SYMLINK_LOG_COUNT; + + /* + * Pre-reserve enough log reservation to handle the transaction + * rolling needed to add one parent pointer. + */ + if (xfs_has_parent(mp)) + ret += resp->tr_attrsetm.tr_logcount; + + return ret; +} /* * Making a new symplink is the same as creating a new file, but @@ -911,54 +1109,76 @@ xfs_calc_sb_reservation( return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); } -void -xfs_trans_resv_calc( +/* + * Namespace reservations. + * + * These get tricky when parent pointers are enabled as we have attribute + * modifications occurring from within these transactions. Rather than confuse + * each of these reservation calculations with the conditional attribute + * reservations, add them here in a clear and concise manner. This requires that + * the attribute reservations have already been calculated. + * + * Note that we only include the static attribute reservation here; the runtime + * reservation will have to be modified by the size of the attributes being + * added/removed/modified. See the comments on the attribute reservation + * calculations for more details. + */ +STATIC void +xfs_calc_namespace_reservations( struct xfs_mount *mp, struct xfs_trans_resv *resp) { - int logcount_adj = 0; - - /* - * The following transactions are logged in physical format and - * require a permanent reservation on space. - */ - resp->tr_write.tr_logres = xfs_calc_write_reservation(mp, false); - resp->tr_write.tr_logcount = XFS_WRITE_LOG_COUNT; - resp->tr_write.tr_logflags |= XFS_TRANS_PERM_LOG_RES; - - resp->tr_itruncate.tr_logres = xfs_calc_itruncate_reservation(mp, false); - resp->tr_itruncate.tr_logcount = XFS_ITRUNCATE_LOG_COUNT; - resp->tr_itruncate.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + ASSERT(resp->tr_attrsetm.tr_logres > 0); resp->tr_rename.tr_logres = xfs_calc_rename_reservation(mp); - resp->tr_rename.tr_logcount = XFS_RENAME_LOG_COUNT; + resp->tr_rename.tr_logcount = xfs_rename_log_count(mp, resp); resp->tr_rename.tr_logflags |= XFS_TRANS_PERM_LOG_RES; resp->tr_link.tr_logres = xfs_calc_link_reservation(mp); - resp->tr_link.tr_logcount = XFS_LINK_LOG_COUNT; + resp->tr_link.tr_logcount = xfs_link_log_count(mp, resp); resp->tr_link.tr_logflags |= XFS_TRANS_PERM_LOG_RES; resp->tr_remove.tr_logres = xfs_calc_remove_reservation(mp); - resp->tr_remove.tr_logcount = XFS_REMOVE_LOG_COUNT; + resp->tr_remove.tr_logcount = xfs_remove_log_count(mp, resp); resp->tr_remove.tr_logflags |= XFS_TRANS_PERM_LOG_RES; resp->tr_symlink.tr_logres = xfs_calc_symlink_reservation(mp); - resp->tr_symlink.tr_logcount = XFS_SYMLINK_LOG_COUNT; + resp->tr_symlink.tr_logcount = xfs_symlink_log_count(mp, resp); resp->tr_symlink.tr_logflags |= XFS_TRANS_PERM_LOG_RES; resp->tr_create.tr_logres = xfs_calc_icreate_reservation(mp); - resp->tr_create.tr_logcount = XFS_CREATE_LOG_COUNT; + resp->tr_create.tr_logcount = xfs_icreate_log_count(mp, resp); resp->tr_create.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + resp->tr_mkdir.tr_logres = xfs_calc_mkdir_reservation(mp); + resp->tr_mkdir.tr_logcount = xfs_mkdir_log_count(mp, resp); + resp->tr_mkdir.tr_logflags |= XFS_TRANS_PERM_LOG_RES; +} + +void +xfs_trans_resv_calc( + struct xfs_mount *mp, + struct xfs_trans_resv *resp) +{ + int logcount_adj = 0; + + /* + * The following transactions are logged in physical format and + * require a permanent reservation on space. + */ + resp->tr_write.tr_logres = xfs_calc_write_reservation(mp, false); + resp->tr_write.tr_logcount = XFS_WRITE_LOG_COUNT; + resp->tr_write.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + + resp->tr_itruncate.tr_logres = xfs_calc_itruncate_reservation(mp, false); + resp->tr_itruncate.tr_logcount = XFS_ITRUNCATE_LOG_COUNT; + resp->tr_itruncate.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + resp->tr_create_tmpfile.tr_logres = xfs_calc_create_tmpfile_reservation(mp); resp->tr_create_tmpfile.tr_logcount = XFS_CREATE_TMPFILE_LOG_COUNT; resp->tr_create_tmpfile.tr_logflags |= XFS_TRANS_PERM_LOG_RES; - resp->tr_mkdir.tr_logres = xfs_calc_mkdir_reservation(mp); - resp->tr_mkdir.tr_logcount = XFS_MKDIR_LOG_COUNT; - resp->tr_mkdir.tr_logflags |= XFS_TRANS_PERM_LOG_RES; - resp->tr_ifree.tr_logres = xfs_calc_ifree_reservation(mp); resp->tr_ifree.tr_logcount = XFS_INACTIVE_LOG_COUNT; resp->tr_ifree.tr_logflags |= XFS_TRANS_PERM_LOG_RES; @@ -988,6 +1208,8 @@ xfs_trans_resv_calc( resp->tr_qm_dqalloc.tr_logcount = XFS_WRITE_LOG_COUNT; resp->tr_qm_dqalloc.tr_logflags |= XFS_TRANS_PERM_LOG_RES; + xfs_calc_namespace_reservations(mp, resp); + /* * The following transactions are logged in logical format with * a default log count. -- cgit v1.2.3-59-g8ed1b From fb102fe7fe02e70f8a49cc7f74bc0769cdab2912 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:46 -0700 Subject: xfs: create a hashname function for parent pointers Although directory entry and parent pointer recordsets look very similar (name -> ino), there's one major difference between them: a file can be hardlinked from multiple parent directories with the same filename. This is common in shared container environments where a base directory tree might be hardlink-copied multiple times. IOWs the same 'ls' program might be hardlinked to multiple /srv/*/bin/ls paths. We don't want parent pointer operations to bog down on hash collisions between the same dirent name, so create a special hash function that mixes in the parent directory inode number. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 3 +++ fs/xfs/libxfs/xfs_parent.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_parent.h | 5 +++++ fs/xfs/scrub/attr.c | 4 ++++ 4 files changed, 59 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 93524efa6e56..8c283e5c2470 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -437,6 +437,9 @@ xfs_attr_hashval( { ASSERT(xfs_attr_check_namespace(attr_flags)); + if (attr_flags & XFS_ATTR_PARENT) + return xfs_parent_hashattr(mp, name, namelen, value, valuelen); + return xfs_attr_hashname(name, namelen); } diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c index 5961fa8c8561..d564baf2549c 100644 --- a/fs/xfs/libxfs/xfs_parent.c +++ b/fs/xfs/libxfs/xfs_parent.c @@ -90,3 +90,50 @@ xfs_parent_valuecheck( return true; } + +/* Compute the attribute name hash for a parent pointer. */ +xfs_dahash_t +xfs_parent_hashval( + struct xfs_mount *mp, + const uint8_t *name, + int namelen, + xfs_ino_t parent_ino) +{ + struct xfs_name xname = { + .name = name, + .len = namelen, + }; + + /* + * Use the same dirent name hash as would be used on the directory, but + * mix in the parent inode number to avoid collisions on hardlinked + * files with identical names but different parents. + */ + return xfs_dir2_hashname(mp, &xname) ^ + upper_32_bits(parent_ino) ^ lower_32_bits(parent_ino); +} + +/* Compute the attribute name hash from the xattr components. */ +xfs_dahash_t +xfs_parent_hashattr( + struct xfs_mount *mp, + const uint8_t *name, + int namelen, + const void *value, + int valuelen) +{ + const struct xfs_parent_rec *rec = value; + + /* Requires a local attr value in xfs_parent_rec format */ + if (valuelen != sizeof(struct xfs_parent_rec)) { + ASSERT(valuelen == sizeof(struct xfs_parent_rec)); + return 0; + } + + if (!value) { + ASSERT(value != NULL); + return 0; + } + + return xfs_parent_hashval(mp, name, namelen, be64_to_cpu(rec->p_ino)); +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h index ef8aff860780..6a4028871b72 100644 --- a/fs/xfs/libxfs/xfs_parent.h +++ b/fs/xfs/libxfs/xfs_parent.h @@ -12,4 +12,9 @@ bool xfs_parent_namecheck(unsigned int attr_flags, const void *name, bool xfs_parent_valuecheck(struct xfs_mount *mp, const void *value, size_t valuelen); +xfs_dahash_t xfs_parent_hashval(struct xfs_mount *mp, const uint8_t *name, + int namelen, xfs_ino_t parent_ino); +xfs_dahash_t xfs_parent_hashattr(struct xfs_mount *mp, const uint8_t *name, + int namelen, const void *value, int valuelen); + #endif /* __XFS_PARENT_H__ */ diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 22d7ef4df169..c07d050b39b2 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -536,6 +536,10 @@ xchk_xattr_rec( xchk_da_set_corrupt(ds, level); goto out; } + if (ent->flags & XFS_ATTR_PARENT) { + xchk_da_set_corrupt(ds, level); + goto out; + } calc_hash = xfs_attr_hashval(mp, ent->flags, rentry->name, rentry->namelen, NULL, be32_to_cpu(rentry->valuelen)); -- cgit v1.2.3-59-g8ed1b From b7c62d90c12c6cc86f10b8a62cefe0029374b6ff Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:47 -0700 Subject: xfs: parent pointer attribute creation Add parent pointer attribute during xfs_create, and subroutines to initialize attributes. Note that the xfs_attr_intent object contains a pointer to the caller's xfs_da_args object, so the latter must persist until transaction commit. Signed-off-by: Dave Chinner Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong [djwong: shorten names, adjust to new format, set init_xattrs for parent pointers] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/libxfs/xfs_parent.c | 68 +++++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_parent.h | 65 +++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_trans_space.c | 52 +++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_trans_space.h | 9 +++--- fs/xfs/scrub/tempfile.c | 2 +- fs/xfs/xfs_inode.c | 32 +++++++++++++++---- fs/xfs/xfs_iops.c | 15 ++++++++- fs/xfs/xfs_super.c | 10 ++++++ 9 files changed, 242 insertions(+), 12 deletions(-) create mode 100644 fs/xfs/libxfs/xfs_trans_space.c (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 4956ea9a307b..0c1a0b67af93 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -51,6 +51,7 @@ xfs-y += $(addprefix libxfs/, \ xfs_symlink_remote.o \ xfs_trans_inode.o \ xfs_trans_resv.o \ + xfs_trans_space.o \ xfs_types.o \ ) # xfs_rtbitmap is shared with libxfs diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c index d564baf2549c..65616cfc1a2b 100644 --- a/fs/xfs/libxfs/xfs_parent.c +++ b/fs/xfs/libxfs/xfs_parent.c @@ -27,6 +27,10 @@ #include "xfs_xattr.h" #include "xfs_parent.h" #include "xfs_trans_space.h" +#include "xfs_attr_item.h" +#include "xfs_health.h" + +struct kmem_cache *xfs_parent_args_cache; /* * Parent pointer attribute handling. @@ -137,3 +141,67 @@ xfs_parent_hashattr( return xfs_parent_hashval(mp, name, namelen, be64_to_cpu(rec->p_ino)); } + +/* + * Initialize the parent pointer arguments structure. Caller must have zeroed + * the contents of @args. @tp is only required for updates. + */ +static void +xfs_parent_da_args_init( + struct xfs_da_args *args, + struct xfs_trans *tp, + struct xfs_parent_rec *rec, + struct xfs_inode *child, + xfs_ino_t owner, + const struct xfs_name *parent_name) +{ + args->geo = child->i_mount->m_attr_geo; + args->whichfork = XFS_ATTR_FORK; + args->attr_filter = XFS_ATTR_PARENT; + args->op_flags = XFS_DA_OP_LOGGED | XFS_DA_OP_OKNOENT; + args->trans = tp; + args->dp = child; + args->owner = owner; + args->name = parent_name->name; + args->namelen = parent_name->len; + args->value = rec; + args->valuelen = sizeof(struct xfs_parent_rec); + xfs_attr_sethash(args); +} + +/* Make sure the incore state is ready for a parent pointer query/update. */ +static inline int +xfs_parent_iread_extents( + struct xfs_trans *tp, + struct xfs_inode *child) +{ + /* Parent pointers require that the attr fork must exist. */ + if (XFS_IS_CORRUPT(child->i_mount, !xfs_inode_has_attr_fork(child))) { + xfs_inode_mark_sick(child, XFS_SICK_INO_PARENT); + return -EFSCORRUPTED; + } + + return xfs_iread_extents(tp, child, XFS_ATTR_FORK); +} + +/* Add a parent pointer to reflect a dirent addition. */ +int +xfs_parent_addname( + struct xfs_trans *tp, + struct xfs_parent_args *ppargs, + struct xfs_inode *dp, + const struct xfs_name *parent_name, + struct xfs_inode *child) +{ + int error; + + error = xfs_parent_iread_extents(tp, child); + if (error) + return error; + + xfs_inode_to_parent_rec(&ppargs->rec, dp); + xfs_parent_da_args_init(&ppargs->args, tp, &ppargs->rec, child, + child->i_ino, parent_name); + xfs_attr_defer_add(&ppargs->args, XFS_ATTR_DEFER_SET); + return 0; +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h index 6a4028871b72..6de24e3ef318 100644 --- a/fs/xfs/libxfs/xfs_parent.h +++ b/fs/xfs/libxfs/xfs_parent.h @@ -17,4 +17,69 @@ xfs_dahash_t xfs_parent_hashval(struct xfs_mount *mp, const uint8_t *name, xfs_dahash_t xfs_parent_hashattr(struct xfs_mount *mp, const uint8_t *name, int namelen, const void *value, int valuelen); +/* Initializes a xfs_parent_rec to be stored as an attribute name. */ +static inline void +xfs_parent_rec_init( + struct xfs_parent_rec *rec, + xfs_ino_t ino, + uint32_t gen) +{ + rec->p_ino = cpu_to_be64(ino); + rec->p_gen = cpu_to_be32(gen); +} + +/* Initializes a xfs_parent_rec to be stored as an attribute name. */ +static inline void +xfs_inode_to_parent_rec( + struct xfs_parent_rec *rec, + const struct xfs_inode *dp) +{ + xfs_parent_rec_init(rec, dp->i_ino, VFS_IC(dp)->i_generation); +} + +extern struct kmem_cache *xfs_parent_args_cache; + +/* + * Parent pointer information needed to pass around the deferred xattr update + * machinery. + */ +struct xfs_parent_args { + struct xfs_parent_rec rec; + struct xfs_da_args args; +}; + +/* + * Start a parent pointer update by allocating the context object we need to + * perform a parent pointer update. + */ +static inline int +xfs_parent_start( + struct xfs_mount *mp, + struct xfs_parent_args **ppargsp) +{ + if (!xfs_has_parent(mp)) { + *ppargsp = NULL; + return 0; + } + + *ppargsp = kmem_cache_zalloc(xfs_parent_args_cache, GFP_KERNEL); + if (!*ppargsp) + return -ENOMEM; + return 0; +} + +/* Finish a parent pointer update by freeing the context object. */ +static inline void +xfs_parent_finish( + struct xfs_mount *mp, + struct xfs_parent_args *ppargs) +{ + if (ppargs) + kmem_cache_free(xfs_parent_args_cache, ppargs); +} + +int xfs_parent_addname(struct xfs_trans *tp, struct xfs_parent_args *ppargs, + struct xfs_inode *dp, const struct xfs_name *parent_name, + struct xfs_inode *child); + #endif /* __XFS_PARENT_H__ */ diff --git a/fs/xfs/libxfs/xfs_trans_space.c b/fs/xfs/libxfs/xfs_trans_space.c new file mode 100644 index 000000000000..90532c3fa205 --- /dev/null +++ b/fs/xfs/libxfs/xfs_trans_space.c @@ -0,0 +1,52 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_da_format.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_trans_space.h" + +/* Calculate the disk space required to add a parent pointer. */ +unsigned int +xfs_parent_calc_space_res( + struct xfs_mount *mp, + unsigned int namelen) +{ + /* + * Parent pointers are always the first attr in an attr tree, and never + * larger than a block + */ + return XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK) + + XFS_NEXTENTADD_SPACE_RES(mp, namelen, XFS_ATTR_FORK); +} + +unsigned int +xfs_create_space_res( + struct xfs_mount *mp, + unsigned int namelen) +{ + unsigned int ret; + + ret = XFS_IALLOC_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp, namelen); + if (xfs_has_parent(mp)) + ret += xfs_parent_calc_space_res(mp, namelen); + + return ret; +} + +unsigned int +xfs_mkdir_space_res( + struct xfs_mount *mp, + unsigned int namelen) +{ + return xfs_create_space_res(mp, namelen); +} diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h index 9640fc232c14..6cda87153b38 100644 --- a/fs/xfs/libxfs/xfs_trans_space.h +++ b/fs/xfs/libxfs/xfs_trans_space.h @@ -80,8 +80,6 @@ /* This macro is not used - see inline code in xfs_attr_set */ #define XFS_ATTRSET_SPACE_RES(mp, v) \ (XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK) + XFS_B_TO_FSB(mp, v)) -#define XFS_CREATE_SPACE_RES(mp,nl) \ - (XFS_IALLOC_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl)) #define XFS_DIOSTRAT_SPACE_RES(mp, v) \ (XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK) + (v)) #define XFS_GROWFS_SPACE_RES(mp) \ @@ -90,8 +88,6 @@ ((b) + XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK)) #define XFS_LINK_SPACE_RES(mp,nl) \ XFS_DIRENTER_SPACE_RES(mp,nl) -#define XFS_MKDIR_SPACE_RES(mp,nl) \ - (XFS_IALLOC_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl)) #define XFS_QM_DQALLOC_SPACE_RES(mp) \ (XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK) + \ XFS_DQUOT_CLUSTER_SIZE_FSB) @@ -106,5 +102,10 @@ #define XFS_IFREE_SPACE_RES(mp) \ (xfs_has_finobt(mp) ? M_IGEO(mp)->inobt_maxlevels : 0) +unsigned int xfs_parent_calc_space_res(struct xfs_mount *mp, + unsigned int namelen); + +unsigned int xfs_create_space_res(struct xfs_mount *mp, unsigned int namelen); +unsigned int xfs_mkdir_space_res(struct xfs_mount *mp, unsigned int namelen); #endif /* __XFS_TRANS_SPACE_H__ */ diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index 6f39504a216e..ddbcccb3dba1 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -71,7 +71,7 @@ xrep_tempfile_create( return error; if (is_dir) { - resblks = XFS_MKDIR_SPACE_RES(mp, 0); + resblks = xfs_mkdir_space_res(mp, 0); tres = &M_RES(mp)->tr_mkdir; } else { resblks = XFS_IALLOC_SPACE_RES(mp); diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 060e4e767b51..0dd4111a6773 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -40,6 +40,8 @@ #include "xfs_log_priv.h" #include "xfs_health.h" #include "xfs_pnfs.h" +#include "xfs_parent.h" +#include "xfs_xattr.h" struct kmem_cache *xfs_inode_cache; @@ -1017,7 +1019,7 @@ xfs_dir_hook_setup( int xfs_create( struct mnt_idmap *idmap, - xfs_inode_t *dp, + struct xfs_inode *dp, struct xfs_name *name, umode_t mode, dev_t rdev, @@ -1029,7 +1031,7 @@ xfs_create( struct xfs_inode *ip = NULL; struct xfs_trans *tp = NULL; int error; - bool unlock_dp_on_error = false; + bool unlock_dp_on_error = false; prid_t prid; struct xfs_dquot *udqp = NULL; struct xfs_dquot *gdqp = NULL; @@ -1037,6 +1039,7 @@ xfs_create( struct xfs_trans_res *tres; uint resblks; xfs_ino_t ino; + struct xfs_parent_args *ppargs; trace_xfs_create(dp, name); @@ -1058,13 +1061,17 @@ xfs_create( return error; if (is_dir) { - resblks = XFS_MKDIR_SPACE_RES(mp, name->len); + resblks = xfs_mkdir_space_res(mp, name->len); tres = &M_RES(mp)->tr_mkdir; } else { - resblks = XFS_CREATE_SPACE_RES(mp, name->len); + resblks = xfs_create_space_res(mp, name->len); tres = &M_RES(mp)->tr_create; } + error = xfs_parent_start(mp, &ppargs); + if (error) + goto out_release_dquots; + /* * Initially assume that the file does not exist and * reserve the resources for that case. If that is not @@ -1080,7 +1087,7 @@ xfs_create( resblks, &tp); } if (error) - goto out_release_dquots; + goto out_parent; xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); unlock_dp_on_error = true; @@ -1123,6 +1130,16 @@ xfs_create( xfs_bumplink(tp, dp); } + /* + * If we have parent pointers, we need to add the attribute containing + * the parent information now. + */ + if (ppargs) { + error = xfs_parent_addname(tp, ppargs, dp, name, ip); + if (error) + goto out_trans_cancel; + } + /* * Create ip with a reference from dp, and add '.' and '..' references * if it's a directory. @@ -1155,6 +1172,7 @@ xfs_create( *ipp = ip; xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_iunlock(dp, XFS_ILOCK_EXCL); + xfs_parent_finish(mp, ppargs); return 0; out_trans_cancel: @@ -1170,6 +1188,8 @@ xfs_create( xfs_finish_inode_setup(ip); xfs_irele(ip); } + out_parent: + xfs_parent_finish(mp, ppargs); out_release_dquots: xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); @@ -3038,7 +3058,7 @@ xfs_rename_alloc_whiteout( int error; error = xfs_create_tmpfile(idmap, dp, S_IFCHR | WHITEOUT_MODE, - false, &tmpfile); + xfs_has_parent(dp->i_mount), &tmpfile); if (error) return error; diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index d32322f9ecde..ff222827e550 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -157,6 +157,8 @@ xfs_create_need_xattr( if (dir->i_sb->s_security) return true; #endif + if (xfs_has_parent(XFS_I(dir)->i_mount)) + return true; return false; } @@ -201,7 +203,18 @@ xfs_generic_create( xfs_create_need_xattr(dir, default_acl, acl), &ip); } else { - error = xfs_create_tmpfile(idmap, XFS_I(dir), mode, false, &ip); + bool init_xattrs = false; + + /* + * If this temporary file will be linkable, set up the file + * with an attr fork to receive a parent pointer. + */ + if (!(tmpfile->f_flags & O_EXCL) && + xfs_has_parent(XFS_I(dir)->i_mount)) + init_xattrs = true; + + error = xfs_create_tmpfile(idmap, XFS_I(dir), mode, + init_xattrs, &ip); } if (unlikely(error)) goto out_free_acl; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index e525a6c477ff..c303d7ff9597 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -44,6 +44,7 @@ #include "xfs_dahash_test.h" #include "xfs_rtbitmap.h" #include "xfs_exchmaps_item.h" +#include "xfs_parent.h" #include "scrub/stats.h" #include "scrub/rcbag_btree.h" @@ -2211,8 +2212,16 @@ xfs_init_caches(void) if (!xfs_xmi_cache) goto out_destroy_xmd_cache; + xfs_parent_args_cache = kmem_cache_create("xfs_parent_args", + sizeof(struct xfs_parent_args), + 0, 0, NULL); + if (!xfs_parent_args_cache) + goto out_destroy_xmi_cache; + return 0; + out_destroy_xmi_cache: + kmem_cache_destroy(xfs_xmi_cache); out_destroy_xmd_cache: kmem_cache_destroy(xfs_xmd_cache); out_destroy_iul_cache: @@ -2273,6 +2282,7 @@ xfs_destroy_caches(void) * destroy caches. */ rcu_barrier(); + kmem_cache_destroy(xfs_parent_args_cache); kmem_cache_destroy(xfs_xmd_cache); kmem_cache_destroy(xfs_xmi_cache); kmem_cache_destroy(xfs_iunlink_cache); -- cgit v1.2.3-59-g8ed1b From f1097be220fa938de5114db57a1ddb5de2bf6046 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:48 -0700 Subject: xfs: add parent attributes to link This patch modifies xfs_link to add a parent pointer to the inode. Signed-off-by: Dave Chinner Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong [djwong: minor rebase fixes] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_trans_space.c | 14 ++++++++++++++ fs/xfs/libxfs/xfs_trans_space.h | 3 +-- fs/xfs/scrub/dir_repair.c | 2 +- fs/xfs/scrub/orphanage.c | 2 +- fs/xfs/xfs_inode.c | 43 +++++++++++++++++++++++++++++++++++------ 5 files changed, 54 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_trans_space.c b/fs/xfs/libxfs/xfs_trans_space.c index 90532c3fa205..cf775750120e 100644 --- a/fs/xfs/libxfs/xfs_trans_space.c +++ b/fs/xfs/libxfs/xfs_trans_space.c @@ -50,3 +50,17 @@ xfs_mkdir_space_res( { return xfs_create_space_res(mp, namelen); } + +unsigned int +xfs_link_space_res( + struct xfs_mount *mp, + unsigned int namelen) +{ + unsigned int ret; + + ret = XFS_DIRENTER_SPACE_RES(mp, namelen); + if (xfs_has_parent(mp)) + ret += xfs_parent_calc_space_res(mp, namelen); + + return ret; +} diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h index 6cda87153b38..5539634009fb 100644 --- a/fs/xfs/libxfs/xfs_trans_space.h +++ b/fs/xfs/libxfs/xfs_trans_space.h @@ -86,8 +86,6 @@ (2 * (mp)->m_alloc_maxlevels) #define XFS_GROWFSRT_SPACE_RES(mp,b) \ ((b) + XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK)) -#define XFS_LINK_SPACE_RES(mp,nl) \ - XFS_DIRENTER_SPACE_RES(mp,nl) #define XFS_QM_DQALLOC_SPACE_RES(mp) \ (XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK) + \ XFS_DQUOT_CLUSTER_SIZE_FSB) @@ -107,5 +105,6 @@ unsigned int xfs_parent_calc_space_res(struct xfs_mount *mp, unsigned int xfs_create_space_res(struct xfs_mount *mp, unsigned int namelen); unsigned int xfs_mkdir_space_res(struct xfs_mount *mp, unsigned int namelen); +unsigned int xfs_link_space_res(struct xfs_mount *mp, unsigned int namelen); #endif /* __XFS_TRANS_SPACE_H__ */ diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 38957da26b94..575397aef1f7 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -704,7 +704,7 @@ xrep_dir_replay_update( uint resblks; int error; - resblks = XFS_LINK_SPACE_RES(mp, xname->len); + resblks = xfs_link_space_res(mp, xname->len); error = xchk_trans_alloc(rd->sc, resblks); if (error) return error; diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index 885b7d478a0a..5e2c3546f2e9 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -326,7 +326,7 @@ xrep_adoption_trans_alloc( /* Compute the worst case space reservation that we need. */ adopt->sc = sc; - adopt->orphanage_blkres = XFS_LINK_SPACE_RES(mp, MAXNAMELEN); + adopt->orphanage_blkres = xfs_link_space_res(mp, MAXNAMELEN); if (S_ISDIR(VFS_I(sc->ip)->i_mode)) child_blkres = XFS_RENAME_SPACE_RES(mp, xfs_name_dotdot.len); adopt->child_blkres = child_blkres; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 0dd4111a6773..61a390f5e2ae 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1299,14 +1299,15 @@ xfs_create_tmpfile( int xfs_link( - xfs_inode_t *tdp, - xfs_inode_t *sip, + struct xfs_inode *tdp, + struct xfs_inode *sip, struct xfs_name *target_name) { - xfs_mount_t *mp = tdp->i_mount; - xfs_trans_t *tp; + struct xfs_mount *mp = tdp->i_mount; + struct xfs_trans *tp; int error, nospace_error = 0; int resblks; + struct xfs_parent_args *ppargs; trace_xfs_link(tdp, target_name); @@ -1325,11 +1326,25 @@ xfs_link( if (error) goto std_return; - resblks = XFS_LINK_SPACE_RES(mp, target_name->len); + error = xfs_parent_start(mp, &ppargs); + if (error) + goto std_return; + + resblks = xfs_link_space_res(mp, target_name->len); error = xfs_trans_alloc_dir(tdp, &M_RES(mp)->tr_link, sip, &resblks, &tp, &nospace_error); if (error) - goto std_return; + goto out_parent; + + /* + * We don't allow reservationless or quotaless hardlinking when parent + * pointers are enabled because we can't back out if the xattrs must + * grow. + */ + if (ppargs && nospace_error) { + error = nospace_error; + goto error_return; + } /* * If we are using project inheritance, we only allow hard link @@ -1380,6 +1395,19 @@ xfs_link( xfs_trans_log_inode(tp, tdp, XFS_ILOG_CORE); xfs_bumplink(tp, sip); + + /* + * If we have parent pointers, we now need to add the parent record to + * the attribute fork of the inode. If this is the initial parent + * attribute, we need to create it correctly, otherwise we can just add + * the parent to the inode. + */ + if (ppargs) { + error = xfs_parent_addname(tp, ppargs, tdp, target_name, sip); + if (error) + goto error_return; + } + xfs_dir_update_hook(tdp, sip, 1, target_name); /* @@ -1393,12 +1421,15 @@ xfs_link( error = xfs_trans_commit(tp); xfs_iunlock(tdp, XFS_ILOCK_EXCL); xfs_iunlock(sip, XFS_ILOCK_EXCL); + xfs_parent_finish(mp, ppargs); return error; error_return: xfs_trans_cancel(tp); xfs_iunlock(tdp, XFS_ILOCK_EXCL); xfs_iunlock(sip, XFS_ILOCK_EXCL); + out_parent: + xfs_parent_finish(mp, ppargs); std_return: if (error == -ENOSPC && nospace_error) error = nospace_error; -- cgit v1.2.3-59-g8ed1b From 5d31a85dcc1fa4c5d4a925c6da67751653a700ba Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:49 -0700 Subject: xfs: add parent attributes to symlink This patch modifies xfs_symlink to add a parent pointer to the inode. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong [djwong: minor rebase fixups] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_trans_space.c | 17 +++++++++++++++++ fs/xfs/libxfs/xfs_trans_space.h | 4 ++-- fs/xfs/scrub/symlink_repair.c | 2 +- fs/xfs/xfs_symlink.c | 30 +++++++++++++++++++++++++----- 4 files changed, 45 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_trans_space.c b/fs/xfs/libxfs/xfs_trans_space.c index cf775750120e..c8adda82debe 100644 --- a/fs/xfs/libxfs/xfs_trans_space.c +++ b/fs/xfs/libxfs/xfs_trans_space.c @@ -64,3 +64,20 @@ xfs_link_space_res( return ret; } + +unsigned int +xfs_symlink_space_res( + struct xfs_mount *mp, + unsigned int namelen, + unsigned int fsblocks) +{ + unsigned int ret; + + ret = XFS_IALLOC_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp, namelen) + + fsblocks; + + if (xfs_has_parent(mp)) + ret += xfs_parent_calc_space_res(mp, namelen); + + return ret; +} diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h index 5539634009fb..354ad1d6e18d 100644 --- a/fs/xfs/libxfs/xfs_trans_space.h +++ b/fs/xfs/libxfs/xfs_trans_space.h @@ -95,8 +95,6 @@ XFS_DIRREMOVE_SPACE_RES(mp) #define XFS_RENAME_SPACE_RES(mp,nl) \ (XFS_DIRREMOVE_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl)) -#define XFS_SYMLINK_SPACE_RES(mp,nl,b) \ - (XFS_IALLOC_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl) + (b)) #define XFS_IFREE_SPACE_RES(mp) \ (xfs_has_finobt(mp) ? M_IGEO(mp)->inobt_maxlevels : 0) @@ -106,5 +104,7 @@ unsigned int xfs_parent_calc_space_res(struct xfs_mount *mp, unsigned int xfs_create_space_res(struct xfs_mount *mp, unsigned int namelen); unsigned int xfs_mkdir_space_res(struct xfs_mount *mp, unsigned int namelen); unsigned int xfs_link_space_res(struct xfs_mount *mp, unsigned int namelen); +unsigned int xfs_symlink_space_res(struct xfs_mount *mp, unsigned int namelen, + unsigned int fsblocks); #endif /* __XFS_TRANS_SPACE_H__ */ diff --git a/fs/xfs/scrub/symlink_repair.c b/fs/xfs/scrub/symlink_repair.c index 178304959535..c8b5a5b878ac 100644 --- a/fs/xfs/scrub/symlink_repair.c +++ b/fs/xfs/scrub/symlink_repair.c @@ -421,7 +421,7 @@ xrep_symlink_rebuild( * unlikely. */ fs_blocks = xfs_symlink_blocks(sc->mp, target_len); - resblks = XFS_SYMLINK_SPACE_RES(sc->mp, target_len, fs_blocks); + resblks = xfs_symlink_space_res(sc->mp, target_len, fs_blocks); error = xfs_trans_reserve_quota_nblks(sc->tp, sc->tempip, resblks, 0, true); if (error) diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 85ef56fdd7df..17aee806ec2e 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -25,6 +25,8 @@ #include "xfs_error.h" #include "xfs_health.h" #include "xfs_symlink_remote.h" +#include "xfs_parent.h" +#include "xfs_defer.h" int xfs_readlink( @@ -100,6 +102,7 @@ xfs_symlink( struct xfs_dquot *pdqp = NULL; uint resblks; xfs_ino_t ino; + struct xfs_parent_args *ppargs; *ipp = NULL; @@ -130,18 +133,24 @@ xfs_symlink( /* * The symlink will fit into the inode data fork? - * There can't be any attributes so we get the whole variable part. + * If there are no parent pointers, then there wont't be any attributes. + * So we get the whole variable part, and do not need to reserve extra + * blocks. Otherwise, we need to reserve the blocks. */ - if (pathlen <= XFS_LITINO(mp)) + if (pathlen <= XFS_LITINO(mp) && !xfs_has_parent(mp)) fs_blocks = 0; else fs_blocks = xfs_symlink_blocks(mp, pathlen); - resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks); + resblks = xfs_symlink_space_res(mp, link_name->len, fs_blocks); + + error = xfs_parent_start(mp, &ppargs); + if (error) + goto out_release_dquots; error = xfs_trans_alloc_icreate(mp, &M_RES(mp)->tr_symlink, udqp, gdqp, pdqp, resblks, &tp); if (error) - goto out_release_dquots; + goto out_parent; xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); unlock_dp_on_error = true; @@ -161,7 +170,7 @@ xfs_symlink( if (!error) error = xfs_init_new_inode(idmap, tp, dp, ino, S_IFLNK | (mode & ~S_IFMT), 1, 0, prid, - false, &ip); + xfs_has_parent(mp), &ip); if (error) goto out_trans_cancel; @@ -195,6 +204,14 @@ xfs_symlink( goto out_trans_cancel; xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); + + /* Add parent pointer for the new symlink. */ + if (ppargs) { + error = xfs_parent_addname(tp, ppargs, dp, link_name, ip); + if (error) + goto out_trans_cancel; + } + xfs_dir_update_hook(dp, ip, 1, link_name); /* @@ -216,6 +233,7 @@ xfs_symlink( *ipp = ip; xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_iunlock(dp, XFS_ILOCK_EXCL); + xfs_parent_finish(mp, ppargs); return 0; out_trans_cancel: @@ -231,6 +249,8 @@ out_release_inode: xfs_finish_inode_setup(ip); xfs_irele(ip); } +out_parent: + xfs_parent_finish(mp, ppargs); out_release_dquots: xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); -- cgit v1.2.3-59-g8ed1b From d2d18330f63cd70b50eddac76de7c59a36f2faa7 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:49 -0700 Subject: xfs: remove parent pointers in unlink This patch removes the parent pointer attribute during unlink Signed-off-by: Dave Chinner Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong [djwong: adjust to new ondisk format, minor rebase fixes] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_parent.c | 22 ++++++++++++++++++++++ fs/xfs/libxfs/xfs_parent.h | 3 +++ fs/xfs/libxfs/xfs_trans_space.c | 13 +++++++++++++ fs/xfs/libxfs/xfs_trans_space.h | 3 +-- fs/xfs/xfs_inode.c | 27 +++++++++++++++++++++------ 5 files changed, 60 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c index 65616cfc1a2b..6142e68f2338 100644 --- a/fs/xfs/libxfs/xfs_parent.c +++ b/fs/xfs/libxfs/xfs_parent.c @@ -205,3 +205,25 @@ xfs_parent_addname( xfs_attr_defer_add(&ppargs->args, XFS_ATTR_DEFER_SET); return 0; } + +/* Remove a parent pointer to reflect a dirent removal. */ +int +xfs_parent_removename( + struct xfs_trans *tp, + struct xfs_parent_args *ppargs, + struct xfs_inode *dp, + const struct xfs_name *parent_name, + struct xfs_inode *child) +{ + int error; + + error = xfs_parent_iread_extents(tp, child); + if (error) + return error; + + xfs_inode_to_parent_rec(&ppargs->rec, dp); + xfs_parent_da_args_init(&ppargs->args, tp, &ppargs->rec, child, + child->i_ino, parent_name); + xfs_attr_defer_add(&ppargs->args, XFS_ATTR_DEFER_REMOVE); + return 0; +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h index 6de24e3ef318..4a7fd48c226a 100644 --- a/fs/xfs/libxfs/xfs_parent.h +++ b/fs/xfs/libxfs/xfs_parent.h @@ -81,5 +81,8 @@ xfs_parent_finish( int xfs_parent_addname(struct xfs_trans *tp, struct xfs_parent_args *ppargs, struct xfs_inode *dp, const struct xfs_name *parent_name, struct xfs_inode *child); +int xfs_parent_removename(struct xfs_trans *tp, struct xfs_parent_args *ppargs, + struct xfs_inode *dp, const struct xfs_name *parent_name, + struct xfs_inode *child); #endif /* __XFS_PARENT_H__ */ diff --git a/fs/xfs/libxfs/xfs_trans_space.c b/fs/xfs/libxfs/xfs_trans_space.c index c8adda82debe..df729e4f1a4c 100644 --- a/fs/xfs/libxfs/xfs_trans_space.c +++ b/fs/xfs/libxfs/xfs_trans_space.c @@ -81,3 +81,16 @@ xfs_symlink_space_res( return ret; } + +unsigned int +xfs_remove_space_res( + struct xfs_mount *mp, + unsigned int namelen) +{ + unsigned int ret = XFS_DIRREMOVE_SPACE_RES(mp); + + if (xfs_has_parent(mp)) + ret += xfs_parent_calc_space_res(mp, namelen); + + return ret; +} diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h index 354ad1d6e18d..a4490813c56f 100644 --- a/fs/xfs/libxfs/xfs_trans_space.h +++ b/fs/xfs/libxfs/xfs_trans_space.h @@ -91,8 +91,6 @@ XFS_DQUOT_CLUSTER_SIZE_FSB) #define XFS_QM_QINOCREATE_SPACE_RES(mp) \ XFS_IALLOC_SPACE_RES(mp) -#define XFS_REMOVE_SPACE_RES(mp) \ - XFS_DIRREMOVE_SPACE_RES(mp) #define XFS_RENAME_SPACE_RES(mp,nl) \ (XFS_DIRREMOVE_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl)) #define XFS_IFREE_SPACE_RES(mp) \ @@ -106,5 +104,6 @@ unsigned int xfs_mkdir_space_res(struct xfs_mount *mp, unsigned int namelen); unsigned int xfs_link_space_res(struct xfs_mount *mp, unsigned int namelen); unsigned int xfs_symlink_space_res(struct xfs_mount *mp, unsigned int namelen, unsigned int fsblocks); +unsigned int xfs_remove_space_res(struct xfs_mount *mp, unsigned int namelen); #endif /* __XFS_TRANS_SPACE_H__ */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 61a390f5e2ae..c4a1c2dd5261 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2721,16 +2721,17 @@ xfs_iunpin_wait( */ int xfs_remove( - xfs_inode_t *dp, + struct xfs_inode *dp, struct xfs_name *name, - xfs_inode_t *ip) + struct xfs_inode *ip) { - xfs_mount_t *mp = dp->i_mount; - xfs_trans_t *tp = NULL; + struct xfs_mount *mp = dp->i_mount; + struct xfs_trans *tp = NULL; int is_dir = S_ISDIR(VFS_I(ip)->i_mode); int dontcare; int error = 0; uint resblks; + struct xfs_parent_args *ppargs; trace_xfs_remove(dp, name); @@ -2747,6 +2748,10 @@ xfs_remove( if (error) goto std_return; + error = xfs_parent_start(mp, &ppargs); + if (error) + goto std_return; + /* * We try to get the real space reservation first, allowing for * directory btree deletion(s) implying possible bmap insert(s). If we @@ -2758,12 +2763,12 @@ xfs_remove( * the directory code can handle a reservationless update and we don't * want to prevent a user from trying to free space by deleting things. */ - resblks = XFS_REMOVE_SPACE_RES(mp); + resblks = xfs_remove_space_res(mp, name->len); error = xfs_trans_alloc_dir(dp, &M_RES(mp)->tr_remove, ip, &resblks, &tp, &dontcare); if (error) { ASSERT(error != -ENOSPC); - goto std_return; + goto out_parent; } /* @@ -2823,6 +2828,13 @@ xfs_remove( goto out_trans_cancel; } + /* Remove parent pointer. */ + if (ppargs) { + error = xfs_parent_removename(tp, ppargs, dp, name, ip); + if (error) + goto out_trans_cancel; + } + /* * Drop the link from dp to ip, and if ip was a directory, remove the * '.' and '..' references since we freed the directory. @@ -2846,6 +2858,7 @@ xfs_remove( xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_iunlock(dp, XFS_ILOCK_EXCL); + xfs_parent_finish(mp, ppargs); return 0; out_trans_cancel: @@ -2853,6 +2866,8 @@ xfs_remove( out_unlock: xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_iunlock(dp, XFS_ILOCK_EXCL); + out_parent: + xfs_parent_finish(mp, ppargs); std_return: return error; } -- cgit v1.2.3-59-g8ed1b From 5a8338c88284df4e9e697225aa65f2709333a659 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:50 -0700 Subject: xfs: Add parent pointers to rename This patch removes the old parent pointer attribute during the rename operation, and re-adds the updated parent pointer. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong [djwong: adjust to new ondisk format] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_parent.c | 30 ++++++++++++++++ fs/xfs/libxfs/xfs_parent.h | 6 ++++ fs/xfs/libxfs/xfs_trans_space.c | 25 +++++++++++++ fs/xfs/libxfs/xfs_trans_space.h | 6 ++-- fs/xfs/scrub/orphanage.c | 3 +- fs/xfs/scrub/parent_repair.c | 3 +- fs/xfs/xfs_inode.c | 80 +++++++++++++++++++++++++++++++++++++---- 7 files changed, 142 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c index 6142e68f2338..fdf643bfde4d 100644 --- a/fs/xfs/libxfs/xfs_parent.c +++ b/fs/xfs/libxfs/xfs_parent.c @@ -227,3 +227,33 @@ xfs_parent_removename( xfs_attr_defer_add(&ppargs->args, XFS_ATTR_DEFER_REMOVE); return 0; } + +/* Replace one parent pointer with another to reflect a rename. */ +int +xfs_parent_replacename( + struct xfs_trans *tp, + struct xfs_parent_args *ppargs, + struct xfs_inode *old_dp, + const struct xfs_name *old_name, + struct xfs_inode *new_dp, + const struct xfs_name *new_name, + struct xfs_inode *child) +{ + int error; + + error = xfs_parent_iread_extents(tp, child); + if (error) + return error; + + xfs_inode_to_parent_rec(&ppargs->rec, old_dp); + xfs_parent_da_args_init(&ppargs->args, tp, &ppargs->rec, child, + child->i_ino, old_name); + + xfs_inode_to_parent_rec(&ppargs->new_rec, new_dp); + ppargs->args.new_name = new_name->name; + ppargs->args.new_namelen = new_name->len; + ppargs->args.new_value = &ppargs->new_rec; + ppargs->args.new_valuelen = sizeof(struct xfs_parent_rec); + xfs_attr_defer_add(&ppargs->args, XFS_ATTR_DEFER_REPLACE); + return 0; +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h index 4a7fd48c226a..768633b31367 100644 --- a/fs/xfs/libxfs/xfs_parent.h +++ b/fs/xfs/libxfs/xfs_parent.h @@ -45,6 +45,7 @@ extern struct kmem_cache *xfs_parent_args_cache; */ struct xfs_parent_args { struct xfs_parent_rec rec; + struct xfs_parent_rec new_rec; struct xfs_da_args args; }; @@ -84,5 +85,10 @@ int xfs_parent_addname(struct xfs_trans *tp, struct xfs_parent_args *ppargs, int xfs_parent_removename(struct xfs_trans *tp, struct xfs_parent_args *ppargs, struct xfs_inode *dp, const struct xfs_name *parent_name, struct xfs_inode *child); +int xfs_parent_replacename(struct xfs_trans *tp, + struct xfs_parent_args *ppargs, + struct xfs_inode *old_dp, const struct xfs_name *old_name, + struct xfs_inode *new_dp, const struct xfs_name *new_name, + struct xfs_inode *child); #endif /* __XFS_PARENT_H__ */ diff --git a/fs/xfs/libxfs/xfs_trans_space.c b/fs/xfs/libxfs/xfs_trans_space.c index df729e4f1a4c..b9dc3752f702 100644 --- a/fs/xfs/libxfs/xfs_trans_space.c +++ b/fs/xfs/libxfs/xfs_trans_space.c @@ -94,3 +94,28 @@ xfs_remove_space_res( return ret; } + +unsigned int +xfs_rename_space_res( + struct xfs_mount *mp, + unsigned int src_namelen, + bool target_exists, + unsigned int target_namelen, + bool has_whiteout) +{ + unsigned int ret; + + ret = XFS_DIRREMOVE_SPACE_RES(mp) + + XFS_DIRENTER_SPACE_RES(mp, target_namelen); + + if (xfs_has_parent(mp)) { + if (has_whiteout) + ret += xfs_parent_calc_space_res(mp, src_namelen); + ret += 2 * xfs_parent_calc_space_res(mp, target_namelen); + } + + if (target_exists) + ret += xfs_parent_calc_space_res(mp, target_namelen); + + return ret; +} diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h index a4490813c56f..1155ff2d37e2 100644 --- a/fs/xfs/libxfs/xfs_trans_space.h +++ b/fs/xfs/libxfs/xfs_trans_space.h @@ -91,8 +91,6 @@ XFS_DQUOT_CLUSTER_SIZE_FSB) #define XFS_QM_QINOCREATE_SPACE_RES(mp) \ XFS_IALLOC_SPACE_RES(mp) -#define XFS_RENAME_SPACE_RES(mp,nl) \ - (XFS_DIRREMOVE_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl)) #define XFS_IFREE_SPACE_RES(mp) \ (xfs_has_finobt(mp) ? M_IGEO(mp)->inobt_maxlevels : 0) @@ -106,4 +104,8 @@ unsigned int xfs_symlink_space_res(struct xfs_mount *mp, unsigned int namelen, unsigned int fsblocks); unsigned int xfs_remove_space_res(struct xfs_mount *mp, unsigned int namelen); +unsigned int xfs_rename_space_res(struct xfs_mount *mp, + unsigned int src_namelen, bool target_exists, + unsigned int target_namelen, bool has_whiteout); + #endif /* __XFS_TRANS_SPACE_H__ */ diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index 5e2c3546f2e9..94bcc2799188 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -328,7 +328,8 @@ xrep_adoption_trans_alloc( adopt->sc = sc; adopt->orphanage_blkres = xfs_link_space_res(mp, MAXNAMELEN); if (S_ISDIR(VFS_I(sc->ip)->i_mode)) - child_blkres = XFS_RENAME_SPACE_RES(mp, xfs_name_dotdot.len); + child_blkres = xfs_rename_space_res(mp, 0, false, + xfs_name_dotdot.len, false); adopt->child_blkres = child_blkres; /* diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index ebb5791bf839..63590e1b3506 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -171,7 +171,8 @@ xrep_parent_reset_dotdot( * Reserve more space just in case we have to expand the dir. We're * allowed to exceed quota to repair inconsistent metadata. */ - spaceres = XFS_RENAME_SPACE_RES(sc->mp, xfs_name_dotdot.len); + spaceres = xfs_rename_space_res(sc->mp, 0, false, xfs_name_dotdot.len, + false); error = xfs_trans_reserve_more_inode(sc->tp, sc->ip, spaceres, 0, true); if (error) diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index c4a1c2dd5261..59488c17e1c1 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -3148,6 +3148,9 @@ xfs_rename( struct xfs_trans *tp; struct xfs_inode *wip = NULL; /* whiteout inode */ struct xfs_inode *inodes[__XFS_SORT_INODES]; + struct xfs_parent_args *src_ppargs = NULL; + struct xfs_parent_args *tgt_ppargs = NULL; + struct xfs_parent_args *wip_ppargs = NULL; int i; int num_inodes = __XFS_SORT_INODES; bool new_parent = (src_dp != target_dp); @@ -3179,9 +3182,26 @@ xfs_rename( xfs_sort_for_rename(src_dp, target_dp, src_ip, target_ip, wip, inodes, &num_inodes); + error = xfs_parent_start(mp, &src_ppargs); + if (error) + goto out_release_wip; + + if (wip) { + error = xfs_parent_start(mp, &wip_ppargs); + if (error) + goto out_src_ppargs; + } + + if (target_ip) { + error = xfs_parent_start(mp, &tgt_ppargs); + if (error) + goto out_wip_ppargs; + } + retry: nospace_error = 0; - spaceres = XFS_RENAME_SPACE_RES(mp, target_name->len); + spaceres = xfs_rename_space_res(mp, src_name->len, target_ip != NULL, + target_name->len, wip != NULL); error = xfs_trans_alloc(mp, &M_RES(mp)->tr_rename, spaceres, 0, 0, &tp); if (error == -ENOSPC) { nospace_error = error; @@ -3190,7 +3210,17 @@ retry: &tp); } if (error) - goto out_release_wip; + goto out_tgt_ppargs; + + /* + * We don't allow reservationless renaming when parent pointers are + * enabled because we can't back out if the xattrs must grow. + */ + if (src_ppargs && nospace_error) { + error = nospace_error; + xfs_trans_cancel(tp); + goto out_tgt_ppargs; + } /* * Attach the dquots to the inodes @@ -3198,7 +3228,7 @@ retry: error = xfs_qm_vop_rename_dqattach(inodes); if (error) { xfs_trans_cancel(tp); - goto out_release_wip; + goto out_tgt_ppargs; } /* @@ -3267,6 +3297,15 @@ retry: goto out_trans_cancel; } + /* + * We don't allow quotaless renaming when parent pointers are enabled + * because we can't back out if the xattrs must grow. + */ + if (src_ppargs && nospace_error) { + error = nospace_error; + goto out_trans_cancel; + } + /* * Check for expected errors before we dirty the transaction * so we can return an error without a transaction abort. @@ -3459,6 +3498,28 @@ retry: if (error) goto out_trans_cancel; + /* Schedule parent pointer updates. */ + if (wip_ppargs) { + error = xfs_parent_addname(tp, wip_ppargs, src_dp, src_name, + wip); + if (error) + goto out_trans_cancel; + } + + if (src_ppargs) { + error = xfs_parent_replacename(tp, src_ppargs, src_dp, + src_name, target_dp, target_name, src_ip); + if (error) + goto out_trans_cancel; + } + + if (tgt_ppargs) { + error = xfs_parent_removename(tp, tgt_ppargs, target_dp, + target_name, target_ip); + if (error) + goto out_trans_cancel; + } + xfs_trans_ichgtime(tp, src_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); xfs_trans_log_inode(tp, src_dp, XFS_ILOG_CORE); if (new_parent) @@ -3480,14 +3541,19 @@ retry: xfs_dir_update_hook(src_dp, wip, 1, src_name); error = xfs_finish_rename(tp); - xfs_iunlock_rename(inodes, num_inodes); - if (wip) - xfs_irele(wip); - return error; + nospace_error = 0; + goto out_unlock; out_trans_cancel: xfs_trans_cancel(tp); +out_unlock: xfs_iunlock_rename(inodes, num_inodes); +out_tgt_ppargs: + xfs_parent_finish(mp, tgt_ppargs); +out_wip_ppargs: + xfs_parent_finish(mp, wip_ppargs); +out_src_ppargs: + xfs_parent_finish(mp, src_ppargs); out_release_wip: if (wip) xfs_irele(wip); -- cgit v1.2.3-59-g8ed1b From 1c12949e50e191933c08758ae53e31b852e730d6 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:51 -0700 Subject: xfs: Add parent pointers to xfs_cross_rename Cross renames are handled separately from standard renames, and need different handling to update the parent attributes correctly. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/xfs_inode.c | 33 +++++++++++++++++++++++++-------- 1 file changed, 25 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 59488c17e1c1..ebe2ce9bd9ee 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2972,15 +2972,17 @@ xfs_cross_rename( struct xfs_inode *dp1, struct xfs_name *name1, struct xfs_inode *ip1, + struct xfs_parent_args *ip1_ppargs, struct xfs_inode *dp2, struct xfs_name *name2, struct xfs_inode *ip2, + struct xfs_parent_args *ip2_ppargs, int spaceres) { - int error = 0; - int ip1_flags = 0; - int ip2_flags = 0; - int dp2_flags = 0; + int error = 0; + int ip1_flags = 0; + int ip2_flags = 0; + int dp2_flags = 0; /* Swap inode number for dirent in first parent */ error = xfs_dir_replace(tp, dp1, name1, ip2->i_ino, spaceres); @@ -3049,6 +3051,21 @@ xfs_cross_rename( } } + /* Schedule parent pointer replacements */ + if (ip1_ppargs) { + error = xfs_parent_replacename(tp, ip1_ppargs, dp1, name1, dp2, + name2, ip1); + if (error) + goto out_trans_abort; + } + + if (ip2_ppargs) { + error = xfs_parent_replacename(tp, ip2_ppargs, dp2, name2, dp1, + name1, ip2); + if (error) + goto out_trans_abort; + } + if (ip1_flags) { xfs_trans_ichgtime(tp, ip1, ip1_flags); xfs_trans_log_inode(tp, ip1, XFS_ILOG_CORE); @@ -3265,10 +3282,10 @@ retry: /* RENAME_EXCHANGE is unique from here on. */ if (flags & RENAME_EXCHANGE) { error = xfs_cross_rename(tp, src_dp, src_name, src_ip, - target_dp, target_name, target_ip, - spaceres); - xfs_iunlock_rename(inodes, num_inodes); - return error; + src_ppargs, target_dp, target_name, target_ip, + tgt_ppargs, spaceres); + nospace_error = 0; + goto out_unlock; } /* -- cgit v1.2.3-59-g8ed1b From daf9f884906bcfcffe26967aee9ece893fba019b Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:52 -0700 Subject: xfs: don't return XFS_ATTR_PARENT attributes via listxattr Parent pointers are internal filesystem metadata. They're not intended to be directly visible to userspace, so filter them out of xfs_xattr_put_listent so that they don't appear in listxattr. Signed-off-by: Allison Henderson Inspired-by: Andrey Albershteyn Reviewed-by: Darrick J. Wong [djwong: change this to XFS_ATTR_PRIVATE_NSP_MASK per fsverity patchset] Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_da_format.h | 3 +++ fs/xfs/xfs_xattr.c | 4 ++++ 2 files changed, 7 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h index 1395ad1937c5..ebde6eb1da65 100644 --- a/fs/xfs/libxfs/xfs_da_format.h +++ b/fs/xfs/libxfs/xfs_da_format.h @@ -726,6 +726,9 @@ struct xfs_attr3_leafblock { XFS_ATTR_SECURE | \ XFS_ATTR_PARENT) +/* Private attr namespaces not exposed to userspace */ +#define XFS_ATTR_PRIVATE_NSP_MASK (XFS_ATTR_PARENT) + #define XFS_ATTR_ONDISK_MASK (XFS_ATTR_NSP_ONDISK_MASK | \ XFS_ATTR_LOCAL | \ XFS_ATTR_INCOMPLETE) diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index ba56a9e73144..1e82d11d980f 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -229,6 +229,10 @@ xfs_xattr_put_listent( ASSERT(context->count >= 0); + /* Don't expose private xattr namespaces. */ + if (flags & XFS_ATTR_PRIVATE_NSP_MASK) + return; + if (flags & XFS_ATTR_ROOT) { #ifdef CONFIG_XFS_POSIX_ACL if (namelen == SGI_ACL_FILE_SIZE && -- cgit v1.2.3-59-g8ed1b From 8f4b980ee67fe53a77b70b1fdd8e15f2fe37180c Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:53 -0700 Subject: xfs: pass the attr value to put_listent when possible Pass the attr value to put_listent when we have local xattrs or shortform xattrs. This will enable the GETPARENTS ioctl to use xfs_attr_list as its backend. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.h | 5 +++-- fs/xfs/libxfs/xfs_attr_sf.h | 1 + fs/xfs/xfs_attr_list.c | 8 +++++++- fs/xfs/xfs_ioctl.c | 1 + fs/xfs/xfs_xattr.c | 1 + 5 files changed, 13 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index d0ed7ea58ab0..d12583dd7eec 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -47,8 +47,9 @@ struct xfs_attrlist_cursor_kern { /* void; state communicated via *context */ -typedef void (*put_listent_func_t)(struct xfs_attr_list_context *, int, - unsigned char *, int, int); +typedef void (*put_listent_func_t)(struct xfs_attr_list_context *context, + int flags, unsigned char *name, int namelen, void *value, + int valuelen); struct xfs_attr_list_context { struct xfs_trans *tp; diff --git a/fs/xfs/libxfs/xfs_attr_sf.h b/fs/xfs/libxfs/xfs_attr_sf.h index bc4422223024..73bdc0e55682 100644 --- a/fs/xfs/libxfs/xfs_attr_sf.h +++ b/fs/xfs/libxfs/xfs_attr_sf.h @@ -16,6 +16,7 @@ typedef struct xfs_attr_sf_sort { uint8_t flags; /* flags bits (see xfs_attr_leaf.h) */ xfs_dahash_t hash; /* this entry's hash value */ unsigned char *name; /* name value, pointer into buffer */ + void *value; } xfs_attr_sf_sort_t; #define XFS_ATTR_SF_ENTSIZE_MAX /* max space for name&value */ \ diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c index 9bc4b5322539..5c947e5ce8b8 100644 --- a/fs/xfs/xfs_attr_list.c +++ b/fs/xfs/xfs_attr_list.c @@ -92,6 +92,7 @@ xfs_attr_shortform_list( sfe->flags, sfe->nameval, (int)sfe->namelen, + &sfe->nameval[sfe->namelen], (int)sfe->valuelen); /* * Either search callback finished early or @@ -138,6 +139,7 @@ xfs_attr_shortform_list( sbp->name = sfe->nameval; sbp->namelen = sfe->namelen; /* These are bytes, and both on-disk, don't endian-flip */ + sbp->value = &sfe->nameval[sfe->namelen], sbp->valuelen = sfe->valuelen; sbp->flags = sfe->flags; sbp->hash = xfs_attr_hashval(dp->i_mount, sfe->flags, @@ -192,6 +194,7 @@ xfs_attr_shortform_list( sbp->flags, sbp->name, sbp->namelen, + sbp->value, sbp->valuelen); if (context->seen_enough) break; @@ -479,6 +482,7 @@ xfs_attr3_leaf_list_int( */ for (; i < ichdr.count; entry++, i++) { char *name; + void *value; int namelen, valuelen; if (be32_to_cpu(entry->hashval) != cursor->hashval) { @@ -496,6 +500,7 @@ xfs_attr3_leaf_list_int( name_loc = xfs_attr3_leaf_name_local(leaf, i); name = name_loc->nameval; namelen = name_loc->namelen; + value = &name_loc->nameval[name_loc->namelen]; valuelen = be16_to_cpu(name_loc->valuelen); } else { xfs_attr_leaf_name_remote_t *name_rmt; @@ -503,6 +508,7 @@ xfs_attr3_leaf_list_int( name_rmt = xfs_attr3_leaf_name_remote(leaf, i); name = name_rmt->name; namelen = name_rmt->namelen; + value = NULL; valuelen = be32_to_cpu(name_rmt->valuelen); } @@ -513,7 +519,7 @@ xfs_attr3_leaf_list_int( return -EFSCORRUPTED; } context->put_listent(context, entry->flags, - name, namelen, valuelen); + name, namelen, value, valuelen); if (context->seen_enough) break; cursor->offset++; diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index e30f9f40f086..7a2a5cf06a5c 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -310,6 +310,7 @@ xfs_ioc_attr_put_listent( int flags, unsigned char *name, int namelen, + void *value, int valuelen) { struct xfs_attrlist *alist = context->buffer; diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 1e82d11d980f..b43f7081b0f4 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -222,6 +222,7 @@ xfs_xattr_put_listent( int flags, unsigned char *name, int namelen, + void *value, int valuelen) { char *prefix; -- cgit v1.2.3-59-g8ed1b From af69d852dfe62b925d0df401eafad40698c889c6 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:54 -0700 Subject: xfs: move handle ioctl code to xfs_handle.c Move the handle managemnet code (and the attrmulti code that uses it) to xfs_handle.c. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/xfs_handle.c | 618 +++++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_handle.h | 28 +++ fs/xfs/xfs_ioctl.c | 592 +----------------------------------------------- fs/xfs/xfs_ioctl.h | 28 --- fs/xfs/xfs_ioctl32.c | 1 + 6 files changed, 649 insertions(+), 619 deletions(-) create mode 100644 fs/xfs/xfs_handle.c create mode 100644 fs/xfs/xfs_handle.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 0c1a0b67af93..c969b11ce0f4 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -78,6 +78,7 @@ xfs-y += xfs_aops.o \ xfs_fsmap.o \ xfs_fsops.o \ xfs_globals.o \ + xfs_handle.o \ xfs_health.o \ xfs_icache.o \ xfs_ioctl.o \ diff --git a/fs/xfs/xfs_handle.c b/fs/xfs/xfs_handle.c new file mode 100644 index 000000000000..13c2479a3053 --- /dev/null +++ b/fs/xfs/xfs_handle.c @@ -0,0 +1,618 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All rights reserved. + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_format.h" +#include "xfs_log_format.h" +#include "xfs_shared.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_error.h" +#include "xfs_trace.h" +#include "xfs_trans.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" +#include "xfs_attr.h" +#include "xfs_ioctl.h" +#include "xfs_parent.h" +#include "xfs_da_btree.h" +#include "xfs_handle.h" +#include "xfs_health.h" +#include "xfs_icache.h" +#include "xfs_export.h" +#include "xfs_xattr.h" +#include "xfs_acl.h" + +#include + +/* + * xfs_find_handle maps from userspace xfs_fsop_handlereq structure to + * a file or fs handle. + * + * XFS_IOC_PATH_TO_FSHANDLE + * returns fs handle for a mount point or path within that mount point + * XFS_IOC_FD_TO_HANDLE + * returns full handle for a FD opened in user space + * XFS_IOC_PATH_TO_HANDLE + * returns full handle for a path + */ +int +xfs_find_handle( + unsigned int cmd, + xfs_fsop_handlereq_t *hreq) +{ + int hsize; + xfs_handle_t handle; + struct inode *inode; + struct fd f = {NULL}; + struct path path; + int error; + struct xfs_inode *ip; + + if (cmd == XFS_IOC_FD_TO_HANDLE) { + f = fdget(hreq->fd); + if (!f.file) + return -EBADF; + inode = file_inode(f.file); + } else { + error = user_path_at(AT_FDCWD, hreq->path, 0, &path); + if (error) + return error; + inode = d_inode(path.dentry); + } + ip = XFS_I(inode); + + /* + * We can only generate handles for inodes residing on a XFS filesystem, + * and only for regular files, directories or symbolic links. + */ + error = -EINVAL; + if (inode->i_sb->s_magic != XFS_SB_MAGIC) + goto out_put; + + error = -EBADF; + if (!S_ISREG(inode->i_mode) && + !S_ISDIR(inode->i_mode) && + !S_ISLNK(inode->i_mode)) + goto out_put; + + + memcpy(&handle.ha_fsid, ip->i_mount->m_fixedfsid, sizeof(xfs_fsid_t)); + + if (cmd == XFS_IOC_PATH_TO_FSHANDLE) { + /* + * This handle only contains an fsid, zero the rest. + */ + memset(&handle.ha_fid, 0, sizeof(handle.ha_fid)); + hsize = sizeof(xfs_fsid_t); + } else { + handle.ha_fid.fid_len = sizeof(xfs_fid_t) - + sizeof(handle.ha_fid.fid_len); + handle.ha_fid.fid_pad = 0; + handle.ha_fid.fid_gen = inode->i_generation; + handle.ha_fid.fid_ino = ip->i_ino; + hsize = sizeof(xfs_handle_t); + } + + error = -EFAULT; + if (copy_to_user(hreq->ohandle, &handle, hsize) || + copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) + goto out_put; + + error = 0; + + out_put: + if (cmd == XFS_IOC_FD_TO_HANDLE) + fdput(f); + else + path_put(&path); + return error; +} + +/* + * No need to do permission checks on the various pathname components + * as the handle operations are privileged. + */ +STATIC int +xfs_handle_acceptable( + void *context, + struct dentry *dentry) +{ + return 1; +} + +/* + * Convert userspace handle data into a dentry. + */ +struct dentry * +xfs_handle_to_dentry( + struct file *parfilp, + void __user *uhandle, + u32 hlen) +{ + xfs_handle_t handle; + struct xfs_fid64 fid; + + /* + * Only allow handle opens under a directory. + */ + if (!S_ISDIR(file_inode(parfilp)->i_mode)) + return ERR_PTR(-ENOTDIR); + + if (hlen != sizeof(xfs_handle_t)) + return ERR_PTR(-EINVAL); + if (copy_from_user(&handle, uhandle, hlen)) + return ERR_PTR(-EFAULT); + if (handle.ha_fid.fid_len != + sizeof(handle.ha_fid) - sizeof(handle.ha_fid.fid_len)) + return ERR_PTR(-EINVAL); + + memset(&fid, 0, sizeof(struct fid)); + fid.ino = handle.ha_fid.fid_ino; + fid.gen = handle.ha_fid.fid_gen; + + return exportfs_decode_fh(parfilp->f_path.mnt, (struct fid *)&fid, 3, + FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG, + xfs_handle_acceptable, NULL); +} + +STATIC struct dentry * +xfs_handlereq_to_dentry( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq) +{ + return xfs_handle_to_dentry(parfilp, hreq->ihandle, hreq->ihandlen); +} + +int +xfs_open_by_handle( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq) +{ + const struct cred *cred = current_cred(); + int error; + int fd; + int permflag; + struct file *filp; + struct inode *inode; + struct dentry *dentry; + fmode_t fmode; + struct path path; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + dentry = xfs_handlereq_to_dentry(parfilp, hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + inode = d_inode(dentry); + + /* Restrict xfs_open_by_handle to directories & regular files. */ + if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) { + error = -EPERM; + goto out_dput; + } + +#if BITS_PER_LONG != 32 + hreq->oflags |= O_LARGEFILE; +#endif + + permflag = hreq->oflags; + fmode = OPEN_FMODE(permflag); + if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) && + (fmode & FMODE_WRITE) && IS_APPEND(inode)) { + error = -EPERM; + goto out_dput; + } + + if ((fmode & FMODE_WRITE) && IS_IMMUTABLE(inode)) { + error = -EPERM; + goto out_dput; + } + + /* Can't write directories. */ + if (S_ISDIR(inode->i_mode) && (fmode & FMODE_WRITE)) { + error = -EISDIR; + goto out_dput; + } + + fd = get_unused_fd_flags(0); + if (fd < 0) { + error = fd; + goto out_dput; + } + + path.mnt = parfilp->f_path.mnt; + path.dentry = dentry; + filp = dentry_open(&path, hreq->oflags, cred); + dput(dentry); + if (IS_ERR(filp)) { + put_unused_fd(fd); + return PTR_ERR(filp); + } + + if (S_ISREG(inode->i_mode)) { + filp->f_flags |= O_NOATIME; + filp->f_mode |= FMODE_NOCMTIME; + } + + fd_install(fd, filp); + return fd; + + out_dput: + dput(dentry); + return error; +} + +int +xfs_readlink_by_handle( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq) +{ + struct dentry *dentry; + __u32 olen; + int error; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + dentry = xfs_handlereq_to_dentry(parfilp, hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + /* Restrict this handle operation to symlinks only. */ + if (!d_is_symlink(dentry)) { + error = -EINVAL; + goto out_dput; + } + + if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) { + error = -EFAULT; + goto out_dput; + } + + error = vfs_readlink(dentry, hreq->ohandle, olen); + + out_dput: + dput(dentry); + return error; +} + +/* + * Format an attribute and copy it out to the user's buffer. + * Take care to check values and protect against them changing later, + * we may be reading them directly out of a user buffer. + */ +static void +xfs_ioc_attr_put_listent( + struct xfs_attr_list_context *context, + int flags, + unsigned char *name, + int namelen, + void *value, + int valuelen) +{ + struct xfs_attrlist *alist = context->buffer; + struct xfs_attrlist_ent *aep; + int arraytop; + + ASSERT(!context->seen_enough); + ASSERT(context->count >= 0); + ASSERT(context->count < (ATTR_MAX_VALUELEN/8)); + ASSERT(context->firstu >= sizeof(*alist)); + ASSERT(context->firstu <= context->bufsize); + + /* + * Only list entries in the right namespace. + */ + if (context->attr_filter != (flags & XFS_ATTR_NSP_ONDISK_MASK)) + return; + + arraytop = sizeof(*alist) + + context->count * sizeof(alist->al_offset[0]); + + /* decrement by the actual bytes used by the attr */ + context->firstu -= round_up(offsetof(struct xfs_attrlist_ent, a_name) + + namelen + 1, sizeof(uint32_t)); + if (context->firstu < arraytop) { + trace_xfs_attr_list_full(context); + alist->al_more = 1; + context->seen_enough = 1; + return; + } + + aep = context->buffer + context->firstu; + aep->a_valuelen = valuelen; + memcpy(aep->a_name, name, namelen); + aep->a_name[namelen] = 0; + alist->al_offset[context->count++] = context->firstu; + alist->al_count = context->count; + trace_xfs_attr_list_add(context); +} + +static unsigned int +xfs_attr_filter( + u32 ioc_flags) +{ + if (ioc_flags & XFS_IOC_ATTR_ROOT) + return XFS_ATTR_ROOT; + if (ioc_flags & XFS_IOC_ATTR_SECURE) + return XFS_ATTR_SECURE; + return 0; +} + +static inline enum xfs_attr_update +xfs_xattr_flags( + u32 ioc_flags, + void *value) +{ + if (!value) + return XFS_ATTRUPDATE_REMOVE; + if (ioc_flags & XFS_IOC_ATTR_CREATE) + return XFS_ATTRUPDATE_CREATE; + if (ioc_flags & XFS_IOC_ATTR_REPLACE) + return XFS_ATTRUPDATE_REPLACE; + return XFS_ATTRUPDATE_UPSERT; +} + +int +xfs_ioc_attr_list( + struct xfs_inode *dp, + void __user *ubuf, + size_t bufsize, + int flags, + struct xfs_attrlist_cursor __user *ucursor) +{ + struct xfs_attr_list_context context = { }; + struct xfs_attrlist *alist; + void *buffer; + int error; + + if (bufsize < sizeof(struct xfs_attrlist) || + bufsize > XFS_XATTR_LIST_MAX) + return -EINVAL; + + /* + * Reject flags, only allow namespaces. + */ + if (flags & ~(XFS_IOC_ATTR_ROOT | XFS_IOC_ATTR_SECURE)) + return -EINVAL; + if (flags == (XFS_IOC_ATTR_ROOT | XFS_IOC_ATTR_SECURE)) + return -EINVAL; + + /* + * Validate the cursor. + */ + if (copy_from_user(&context.cursor, ucursor, sizeof(context.cursor))) + return -EFAULT; + if (context.cursor.pad1 || context.cursor.pad2) + return -EINVAL; + if (!context.cursor.initted && + (context.cursor.hashval || context.cursor.blkno || + context.cursor.offset)) + return -EINVAL; + + buffer = kvzalloc(bufsize, GFP_KERNEL); + if (!buffer) + return -ENOMEM; + + /* + * Initialize the output buffer. + */ + context.dp = dp; + context.resynch = 1; + context.attr_filter = xfs_attr_filter(flags); + context.buffer = buffer; + context.bufsize = round_down(bufsize, sizeof(uint32_t)); + context.firstu = context.bufsize; + context.put_listent = xfs_ioc_attr_put_listent; + + alist = context.buffer; + alist->al_count = 0; + alist->al_more = 0; + alist->al_offset[0] = context.bufsize; + + error = xfs_attr_list(&context); + if (error) + goto out_free; + + if (copy_to_user(ubuf, buffer, bufsize) || + copy_to_user(ucursor, &context.cursor, sizeof(context.cursor))) + error = -EFAULT; +out_free: + kvfree(buffer); + return error; +} + +int +xfs_attrlist_by_handle( + struct file *parfilp, + struct xfs_fsop_attrlist_handlereq __user *p) +{ + struct xfs_fsop_attrlist_handlereq al_hreq; + struct dentry *dentry; + int error = -ENOMEM; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (copy_from_user(&al_hreq, p, sizeof(al_hreq))) + return -EFAULT; + + dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + error = xfs_ioc_attr_list(XFS_I(d_inode(dentry)), al_hreq.buffer, + al_hreq.buflen, al_hreq.flags, &p->pos); + dput(dentry); + return error; +} + +static int +xfs_attrmulti_attr_get( + struct inode *inode, + unsigned char *name, + unsigned char __user *ubuf, + uint32_t *len, + uint32_t flags) +{ + struct xfs_da_args args = { + .dp = XFS_I(inode), + .attr_filter = xfs_attr_filter(flags), + .name = name, + .namelen = strlen(name), + .valuelen = *len, + }; + int error; + + if (*len > XFS_XATTR_SIZE_MAX) + return -EINVAL; + + error = xfs_attr_get(&args); + if (error) + goto out_kfree; + + *len = args.valuelen; + if (copy_to_user(ubuf, args.value, args.valuelen)) + error = -EFAULT; + +out_kfree: + kvfree(args.value); + return error; +} + +static int +xfs_attrmulti_attr_set( + struct inode *inode, + unsigned char *name, + const unsigned char __user *ubuf, + uint32_t len, + uint32_t flags) +{ + struct xfs_da_args args = { + .dp = XFS_I(inode), + .attr_filter = xfs_attr_filter(flags), + .name = name, + .namelen = strlen(name), + }; + int error; + + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + return -EPERM; + + if (ubuf) { + if (len > XFS_XATTR_SIZE_MAX) + return -EINVAL; + args.value = memdup_user(ubuf, len); + if (IS_ERR(args.value)) + return PTR_ERR(args.value); + args.valuelen = len; + } + + error = xfs_attr_change(&args, xfs_xattr_flags(flags, args.value)); + if (!error && (flags & XFS_IOC_ATTR_ROOT)) + xfs_forget_acl(inode, name); + kfree(args.value); + return error; +} + +int +xfs_ioc_attrmulti_one( + struct file *parfilp, + struct inode *inode, + uint32_t opcode, + void __user *uname, + void __user *value, + uint32_t *len, + uint32_t flags) +{ + unsigned char *name; + int error; + + if ((flags & XFS_IOC_ATTR_ROOT) && (flags & XFS_IOC_ATTR_SECURE)) + return -EINVAL; + + name = strndup_user(uname, MAXNAMELEN); + if (IS_ERR(name)) + return PTR_ERR(name); + + switch (opcode) { + case ATTR_OP_GET: + error = xfs_attrmulti_attr_get(inode, name, value, len, flags); + break; + case ATTR_OP_REMOVE: + value = NULL; + *len = 0; + fallthrough; + case ATTR_OP_SET: + error = mnt_want_write_file(parfilp); + if (error) + break; + error = xfs_attrmulti_attr_set(inode, name, value, *len, flags); + mnt_drop_write_file(parfilp); + break; + default: + error = -EINVAL; + break; + } + + kfree(name); + return error; +} + +int +xfs_attrmulti_by_handle( + struct file *parfilp, + void __user *arg) +{ + int error; + xfs_attr_multiop_t *ops; + xfs_fsop_attrmulti_handlereq_t am_hreq; + struct dentry *dentry; + unsigned int i, size; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t))) + return -EFAULT; + + /* overflow check */ + if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t)) + return -E2BIG; + + dentry = xfs_handlereq_to_dentry(parfilp, &am_hreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + error = -E2BIG; + size = am_hreq.opcount * sizeof(xfs_attr_multiop_t); + if (!size || size > 16 * PAGE_SIZE) + goto out_dput; + + ops = memdup_user(am_hreq.ops, size); + if (IS_ERR(ops)) { + error = PTR_ERR(ops); + goto out_dput; + } + + error = 0; + for (i = 0; i < am_hreq.opcount; i++) { + ops[i].am_error = xfs_ioc_attrmulti_one(parfilp, + d_inode(dentry), ops[i].am_opcode, + ops[i].am_attrname, ops[i].am_attrvalue, + &ops[i].am_length, ops[i].am_flags); + } + + if (copy_to_user(am_hreq.ops, ops, size)) + error = -EFAULT; + + kfree(ops); + out_dput: + dput(dentry); + return error; +} diff --git a/fs/xfs/xfs_handle.h b/fs/xfs/xfs_handle.h new file mode 100644 index 000000000000..e39eaf4689da --- /dev/null +++ b/fs/xfs/xfs_handle.h @@ -0,0 +1,28 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All rights reserved. + */ +#ifndef __XFS_HANDLE_H__ +#define __XFS_HANDLE_H__ + +int xfs_attrlist_by_handle(struct file *parfilp, + struct xfs_fsop_attrlist_handlereq __user *p); +int xfs_attrmulti_by_handle(struct file *parfilp, void __user *arg); + +int xfs_find_handle(unsigned int cmd, struct xfs_fsop_handlereq *hreq); +int xfs_open_by_handle(struct file *parfilp, struct xfs_fsop_handlereq *hreq); +int xfs_readlink_by_handle(struct file *parfilp, + struct xfs_fsop_handlereq *hreq); + +int xfs_ioc_attrmulti_one(struct file *parfilp, struct inode *inode, + uint32_t opcode, void __user *uname, void __user *value, + uint32_t *len, uint32_t flags); +int xfs_ioc_attr_list(struct xfs_inode *dp, void __user *ubuf, + size_t bufsize, int flags, + struct xfs_attrlist_cursor __user *ucursor); + +struct dentry *xfs_handle_to_dentry(struct file *parfilp, void __user *uhandle, + u32 hlen); + +#endif /* __XFS_HANDLE_H__ */ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 7a2a5cf06a5c..ed05fcd6261d 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -23,11 +23,9 @@ #include "xfs_fsops.h" #include "xfs_discard.h" #include "xfs_quota.h" -#include "xfs_export.h" #include "xfs_trace.h" #include "xfs_icache.h" #include "xfs_trans.h" -#include "xfs_acl.h" #include "xfs_btree.h" #include #include "xfs_fsmap.h" @@ -37,602 +35,14 @@ #include "xfs_health.h" #include "xfs_reflink.h" #include "xfs_ioctl.h" -#include "xfs_xattr.h" #include "xfs_rtbitmap.h" #include "xfs_file.h" #include "xfs_exchrange.h" +#include "xfs_handle.h" #include -#include #include -/* - * xfs_find_handle maps from userspace xfs_fsop_handlereq structure to - * a file or fs handle. - * - * XFS_IOC_PATH_TO_FSHANDLE - * returns fs handle for a mount point or path within that mount point - * XFS_IOC_FD_TO_HANDLE - * returns full handle for a FD opened in user space - * XFS_IOC_PATH_TO_HANDLE - * returns full handle for a path - */ -int -xfs_find_handle( - unsigned int cmd, - xfs_fsop_handlereq_t *hreq) -{ - int hsize; - xfs_handle_t handle; - struct inode *inode; - struct fd f = {NULL}; - struct path path; - int error; - struct xfs_inode *ip; - - if (cmd == XFS_IOC_FD_TO_HANDLE) { - f = fdget(hreq->fd); - if (!f.file) - return -EBADF; - inode = file_inode(f.file); - } else { - error = user_path_at(AT_FDCWD, hreq->path, 0, &path); - if (error) - return error; - inode = d_inode(path.dentry); - } - ip = XFS_I(inode); - - /* - * We can only generate handles for inodes residing on a XFS filesystem, - * and only for regular files, directories or symbolic links. - */ - error = -EINVAL; - if (inode->i_sb->s_magic != XFS_SB_MAGIC) - goto out_put; - - error = -EBADF; - if (!S_ISREG(inode->i_mode) && - !S_ISDIR(inode->i_mode) && - !S_ISLNK(inode->i_mode)) - goto out_put; - - - memcpy(&handle.ha_fsid, ip->i_mount->m_fixedfsid, sizeof(xfs_fsid_t)); - - if (cmd == XFS_IOC_PATH_TO_FSHANDLE) { - /* - * This handle only contains an fsid, zero the rest. - */ - memset(&handle.ha_fid, 0, sizeof(handle.ha_fid)); - hsize = sizeof(xfs_fsid_t); - } else { - handle.ha_fid.fid_len = sizeof(xfs_fid_t) - - sizeof(handle.ha_fid.fid_len); - handle.ha_fid.fid_pad = 0; - handle.ha_fid.fid_gen = inode->i_generation; - handle.ha_fid.fid_ino = ip->i_ino; - hsize = sizeof(xfs_handle_t); - } - - error = -EFAULT; - if (copy_to_user(hreq->ohandle, &handle, hsize) || - copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) - goto out_put; - - error = 0; - - out_put: - if (cmd == XFS_IOC_FD_TO_HANDLE) - fdput(f); - else - path_put(&path); - return error; -} - -/* - * No need to do permission checks on the various pathname components - * as the handle operations are privileged. - */ -STATIC int -xfs_handle_acceptable( - void *context, - struct dentry *dentry) -{ - return 1; -} - -/* - * Convert userspace handle data into a dentry. - */ -struct dentry * -xfs_handle_to_dentry( - struct file *parfilp, - void __user *uhandle, - u32 hlen) -{ - xfs_handle_t handle; - struct xfs_fid64 fid; - - /* - * Only allow handle opens under a directory. - */ - if (!S_ISDIR(file_inode(parfilp)->i_mode)) - return ERR_PTR(-ENOTDIR); - - if (hlen != sizeof(xfs_handle_t)) - return ERR_PTR(-EINVAL); - if (copy_from_user(&handle, uhandle, hlen)) - return ERR_PTR(-EFAULT); - if (handle.ha_fid.fid_len != - sizeof(handle.ha_fid) - sizeof(handle.ha_fid.fid_len)) - return ERR_PTR(-EINVAL); - - memset(&fid, 0, sizeof(struct fid)); - fid.ino = handle.ha_fid.fid_ino; - fid.gen = handle.ha_fid.fid_gen; - - return exportfs_decode_fh(parfilp->f_path.mnt, (struct fid *)&fid, 3, - FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG, - xfs_handle_acceptable, NULL); -} - -STATIC struct dentry * -xfs_handlereq_to_dentry( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - return xfs_handle_to_dentry(parfilp, hreq->ihandle, hreq->ihandlen); -} - -int -xfs_open_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - const struct cred *cred = current_cred(); - int error; - int fd; - int permflag; - struct file *filp; - struct inode *inode; - struct dentry *dentry; - fmode_t fmode; - struct path path; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - dentry = xfs_handlereq_to_dentry(parfilp, hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - inode = d_inode(dentry); - - /* Restrict xfs_open_by_handle to directories & regular files. */ - if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) { - error = -EPERM; - goto out_dput; - } - -#if BITS_PER_LONG != 32 - hreq->oflags |= O_LARGEFILE; -#endif - - permflag = hreq->oflags; - fmode = OPEN_FMODE(permflag); - if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) && - (fmode & FMODE_WRITE) && IS_APPEND(inode)) { - error = -EPERM; - goto out_dput; - } - - if ((fmode & FMODE_WRITE) && IS_IMMUTABLE(inode)) { - error = -EPERM; - goto out_dput; - } - - /* Can't write directories. */ - if (S_ISDIR(inode->i_mode) && (fmode & FMODE_WRITE)) { - error = -EISDIR; - goto out_dput; - } - - fd = get_unused_fd_flags(0); - if (fd < 0) { - error = fd; - goto out_dput; - } - - path.mnt = parfilp->f_path.mnt; - path.dentry = dentry; - filp = dentry_open(&path, hreq->oflags, cred); - dput(dentry); - if (IS_ERR(filp)) { - put_unused_fd(fd); - return PTR_ERR(filp); - } - - if (S_ISREG(inode->i_mode)) { - filp->f_flags |= O_NOATIME; - filp->f_mode |= FMODE_NOCMTIME; - } - - fd_install(fd, filp); - return fd; - - out_dput: - dput(dentry); - return error; -} - -int -xfs_readlink_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - struct dentry *dentry; - __u32 olen; - int error; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - dentry = xfs_handlereq_to_dentry(parfilp, hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - /* Restrict this handle operation to symlinks only. */ - if (!d_is_symlink(dentry)) { - error = -EINVAL; - goto out_dput; - } - - if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) { - error = -EFAULT; - goto out_dput; - } - - error = vfs_readlink(dentry, hreq->ohandle, olen); - - out_dput: - dput(dentry); - return error; -} - -/* - * Format an attribute and copy it out to the user's buffer. - * Take care to check values and protect against them changing later, - * we may be reading them directly out of a user buffer. - */ -static void -xfs_ioc_attr_put_listent( - struct xfs_attr_list_context *context, - int flags, - unsigned char *name, - int namelen, - void *value, - int valuelen) -{ - struct xfs_attrlist *alist = context->buffer; - struct xfs_attrlist_ent *aep; - int arraytop; - - ASSERT(!context->seen_enough); - ASSERT(context->count >= 0); - ASSERT(context->count < (ATTR_MAX_VALUELEN/8)); - ASSERT(context->firstu >= sizeof(*alist)); - ASSERT(context->firstu <= context->bufsize); - - /* - * Only list entries in the right namespace. - */ - if (context->attr_filter != (flags & XFS_ATTR_NSP_ONDISK_MASK)) - return; - - arraytop = sizeof(*alist) + - context->count * sizeof(alist->al_offset[0]); - - /* decrement by the actual bytes used by the attr */ - context->firstu -= round_up(offsetof(struct xfs_attrlist_ent, a_name) + - namelen + 1, sizeof(uint32_t)); - if (context->firstu < arraytop) { - trace_xfs_attr_list_full(context); - alist->al_more = 1; - context->seen_enough = 1; - return; - } - - aep = context->buffer + context->firstu; - aep->a_valuelen = valuelen; - memcpy(aep->a_name, name, namelen); - aep->a_name[namelen] = 0; - alist->al_offset[context->count++] = context->firstu; - alist->al_count = context->count; - trace_xfs_attr_list_add(context); -} - -static unsigned int -xfs_attr_filter( - u32 ioc_flags) -{ - if (ioc_flags & XFS_IOC_ATTR_ROOT) - return XFS_ATTR_ROOT; - if (ioc_flags & XFS_IOC_ATTR_SECURE) - return XFS_ATTR_SECURE; - return 0; -} - -static inline enum xfs_attr_update -xfs_xattr_flags( - u32 ioc_flags, - void *value) -{ - if (!value) - return XFS_ATTRUPDATE_REMOVE; - if (ioc_flags & XFS_IOC_ATTR_CREATE) - return XFS_ATTRUPDATE_CREATE; - if (ioc_flags & XFS_IOC_ATTR_REPLACE) - return XFS_ATTRUPDATE_REPLACE; - return XFS_ATTRUPDATE_UPSERT; -} - -int -xfs_ioc_attr_list( - struct xfs_inode *dp, - void __user *ubuf, - size_t bufsize, - int flags, - struct xfs_attrlist_cursor __user *ucursor) -{ - struct xfs_attr_list_context context = { }; - struct xfs_attrlist *alist; - void *buffer; - int error; - - if (bufsize < sizeof(struct xfs_attrlist) || - bufsize > XFS_XATTR_LIST_MAX) - return -EINVAL; - - /* - * Reject flags, only allow namespaces. - */ - if (flags & ~(XFS_IOC_ATTR_ROOT | XFS_IOC_ATTR_SECURE)) - return -EINVAL; - if (flags == (XFS_IOC_ATTR_ROOT | XFS_IOC_ATTR_SECURE)) - return -EINVAL; - - /* - * Validate the cursor. - */ - if (copy_from_user(&context.cursor, ucursor, sizeof(context.cursor))) - return -EFAULT; - if (context.cursor.pad1 || context.cursor.pad2) - return -EINVAL; - if (!context.cursor.initted && - (context.cursor.hashval || context.cursor.blkno || - context.cursor.offset)) - return -EINVAL; - - buffer = kvzalloc(bufsize, GFP_KERNEL); - if (!buffer) - return -ENOMEM; - - /* - * Initialize the output buffer. - */ - context.dp = dp; - context.resynch = 1; - context.attr_filter = xfs_attr_filter(flags); - context.buffer = buffer; - context.bufsize = round_down(bufsize, sizeof(uint32_t)); - context.firstu = context.bufsize; - context.put_listent = xfs_ioc_attr_put_listent; - - alist = context.buffer; - alist->al_count = 0; - alist->al_more = 0; - alist->al_offset[0] = context.bufsize; - - error = xfs_attr_list(&context); - if (error) - goto out_free; - - if (copy_to_user(ubuf, buffer, bufsize) || - copy_to_user(ucursor, &context.cursor, sizeof(context.cursor))) - error = -EFAULT; -out_free: - kvfree(buffer); - return error; -} - -STATIC int -xfs_attrlist_by_handle( - struct file *parfilp, - struct xfs_fsop_attrlist_handlereq __user *p) -{ - struct xfs_fsop_attrlist_handlereq al_hreq; - struct dentry *dentry; - int error = -ENOMEM; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - if (copy_from_user(&al_hreq, p, sizeof(al_hreq))) - return -EFAULT; - - dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - error = xfs_ioc_attr_list(XFS_I(d_inode(dentry)), al_hreq.buffer, - al_hreq.buflen, al_hreq.flags, &p->pos); - dput(dentry); - return error; -} - -static int -xfs_attrmulti_attr_get( - struct inode *inode, - unsigned char *name, - unsigned char __user *ubuf, - uint32_t *len, - uint32_t flags) -{ - struct xfs_da_args args = { - .dp = XFS_I(inode), - .attr_filter = xfs_attr_filter(flags), - .name = name, - .namelen = strlen(name), - .valuelen = *len, - }; - int error; - - if (*len > XFS_XATTR_SIZE_MAX) - return -EINVAL; - - error = xfs_attr_get(&args); - if (error) - goto out_kfree; - - *len = args.valuelen; - if (copy_to_user(ubuf, args.value, args.valuelen)) - error = -EFAULT; - -out_kfree: - kvfree(args.value); - return error; -} - -static int -xfs_attrmulti_attr_set( - struct inode *inode, - unsigned char *name, - const unsigned char __user *ubuf, - uint32_t len, - uint32_t flags) -{ - struct xfs_da_args args = { - .dp = XFS_I(inode), - .attr_filter = xfs_attr_filter(flags), - .name = name, - .namelen = strlen(name), - }; - int error; - - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - return -EPERM; - - if (ubuf) { - if (len > XFS_XATTR_SIZE_MAX) - return -EINVAL; - args.value = memdup_user(ubuf, len); - if (IS_ERR(args.value)) - return PTR_ERR(args.value); - args.valuelen = len; - } - - error = xfs_attr_change(&args, xfs_xattr_flags(flags, args.value)); - if (!error && (flags & XFS_IOC_ATTR_ROOT)) - xfs_forget_acl(inode, name); - kfree(args.value); - return error; -} - -int -xfs_ioc_attrmulti_one( - struct file *parfilp, - struct inode *inode, - uint32_t opcode, - void __user *uname, - void __user *value, - uint32_t *len, - uint32_t flags) -{ - unsigned char *name; - int error; - - if ((flags & XFS_IOC_ATTR_ROOT) && (flags & XFS_IOC_ATTR_SECURE)) - return -EINVAL; - - name = strndup_user(uname, MAXNAMELEN); - if (IS_ERR(name)) - return PTR_ERR(name); - - switch (opcode) { - case ATTR_OP_GET: - error = xfs_attrmulti_attr_get(inode, name, value, len, flags); - break; - case ATTR_OP_REMOVE: - value = NULL; - *len = 0; - fallthrough; - case ATTR_OP_SET: - error = mnt_want_write_file(parfilp); - if (error) - break; - error = xfs_attrmulti_attr_set(inode, name, value, *len, flags); - mnt_drop_write_file(parfilp); - break; - default: - error = -EINVAL; - break; - } - - kfree(name); - return error; -} - -STATIC int -xfs_attrmulti_by_handle( - struct file *parfilp, - void __user *arg) -{ - int error; - xfs_attr_multiop_t *ops; - xfs_fsop_attrmulti_handlereq_t am_hreq; - struct dentry *dentry; - unsigned int i, size; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t))) - return -EFAULT; - - /* overflow check */ - if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t)) - return -E2BIG; - - dentry = xfs_handlereq_to_dentry(parfilp, &am_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - error = -E2BIG; - size = am_hreq.opcount * sizeof(xfs_attr_multiop_t); - if (!size || size > 16 * PAGE_SIZE) - goto out_dput; - - ops = memdup_user(am_hreq.ops, size); - if (IS_ERR(ops)) { - error = PTR_ERR(ops); - goto out_dput; - } - - error = 0; - for (i = 0; i < am_hreq.opcount; i++) { - ops[i].am_error = xfs_ioc_attrmulti_one(parfilp, - d_inode(dentry), ops[i].am_opcode, - ops[i].am_attrname, ops[i].am_attrvalue, - &ops[i].am_length, ops[i].am_flags); - } - - if (copy_to_user(am_hreq.ops, ops, size)) - error = -EFAULT; - - kfree(ops); - out_dput: - dput(dentry); - return error; -} - /* Return 0 on success or positive error */ int xfs_fsbulkstat_one_fmt( diff --git a/fs/xfs/xfs_ioctl.h b/fs/xfs/xfs_ioctl.h index 38be600b5e1e..12124946f347 100644 --- a/fs/xfs/xfs_ioctl.h +++ b/fs/xfs/xfs_ioctl.h @@ -14,34 +14,6 @@ int xfs_ioc_swapext( xfs_swapext_t *sxp); -extern int -xfs_find_handle( - unsigned int cmd, - xfs_fsop_handlereq_t *hreq); - -extern int -xfs_open_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq); - -extern int -xfs_readlink_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq); - -int xfs_ioc_attrmulti_one(struct file *parfilp, struct inode *inode, - uint32_t opcode, void __user *uname, void __user *value, - uint32_t *len, uint32_t flags); -int xfs_ioc_attr_list(struct xfs_inode *dp, void __user *ubuf, - size_t bufsize, int flags, - struct xfs_attrlist_cursor __user *ucursor); - -extern struct dentry * -xfs_handle_to_dentry( - struct file *parfilp, - void __user *uhandle, - u32 hlen); - extern int xfs_fileattr_get( struct dentry *dentry, diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c index ee35eea1ecce..b64785dc4354 100644 --- a/fs/xfs/xfs_ioctl32.c +++ b/fs/xfs/xfs_ioctl32.c @@ -24,6 +24,7 @@ #include "xfs_ioctl32.h" #include "xfs_trace.h" #include "xfs_sb.h" +#include "xfs_handle.h" #define _NATIVE_IOC(cmd, type) \ _IOC(_IOC_DIR(cmd), _IOC_TYPE(cmd), _IOC_NR(cmd), sizeof(type)) -- cgit v1.2.3-59-g8ed1b From b8c9d4253da43c02b287831f7e576568f24fbe58 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:55 -0700 Subject: xfs: split out handle management helpers a bit Split out the functions that generate file/fs handles and map them back into dentries in preparation for the GETPARENTS ioctl next. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_fs.h | 4 ++- fs/xfs/xfs_handle.c | 98 ++++++++++++++++++++++++++++++++++---------------- 2 files changed, 70 insertions(+), 32 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index 53526fca7386..97384ab95de4 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -633,7 +633,9 @@ typedef struct xfs_fsop_attrmulti_handlereq { /* * per machine unique filesystem identifier types. */ -typedef struct { __u32 val[2]; } xfs_fsid_t; /* file system id type */ +typedef struct xfs_fsid { + __u32 val[2]; /* file system id type */ +} xfs_fsid_t; typedef struct xfs_fid { __u16 fid_len; /* length of remainder */ diff --git a/fs/xfs/xfs_handle.c b/fs/xfs/xfs_handle.c index 13c2479a3053..b9f4d9860682 100644 --- a/fs/xfs/xfs_handle.c +++ b/fs/xfs/xfs_handle.c @@ -30,6 +30,42 @@ #include +static inline size_t +xfs_filehandle_fid_len(void) +{ + struct xfs_handle *handle = NULL; + + return sizeof(struct xfs_fid) - sizeof(handle->ha_fid.fid_len); +} + +static inline size_t +xfs_filehandle_init( + struct xfs_mount *mp, + xfs_ino_t ino, + uint32_t gen, + struct xfs_handle *handle) +{ + memcpy(&handle->ha_fsid, mp->m_fixedfsid, sizeof(struct xfs_fsid)); + + handle->ha_fid.fid_len = xfs_filehandle_fid_len(); + handle->ha_fid.fid_pad = 0; + handle->ha_fid.fid_gen = gen; + handle->ha_fid.fid_ino = ino; + + return sizeof(struct xfs_handle); +} + +static inline size_t +xfs_fshandle_init( + struct xfs_mount *mp, + struct xfs_handle *handle) +{ + memcpy(&handle->ha_fsid, mp->m_fixedfsid, sizeof(struct xfs_fsid)); + memset(&handle->ha_fid, 0, sizeof(handle->ha_fid)); + + return sizeof(struct xfs_fsid); +} + /* * xfs_find_handle maps from userspace xfs_fsop_handlereq structure to * a file or fs handle. @@ -84,20 +120,11 @@ xfs_find_handle( memcpy(&handle.ha_fsid, ip->i_mount->m_fixedfsid, sizeof(xfs_fsid_t)); - if (cmd == XFS_IOC_PATH_TO_FSHANDLE) { - /* - * This handle only contains an fsid, zero the rest. - */ - memset(&handle.ha_fid, 0, sizeof(handle.ha_fid)); - hsize = sizeof(xfs_fsid_t); - } else { - handle.ha_fid.fid_len = sizeof(xfs_fid_t) - - sizeof(handle.ha_fid.fid_len); - handle.ha_fid.fid_pad = 0; - handle.ha_fid.fid_gen = inode->i_generation; - handle.ha_fid.fid_ino = ip->i_ino; - hsize = sizeof(xfs_handle_t); - } + if (cmd == XFS_IOC_PATH_TO_FSHANDLE) + hsize = xfs_fshandle_init(ip->i_mount, &handle); + else + hsize = xfs_filehandle_init(ip->i_mount, ip->i_ino, + inode->i_generation, &handle); error = -EFAULT; if (copy_to_user(hreq->ohandle, &handle, hsize) || @@ -126,6 +153,31 @@ xfs_handle_acceptable( return 1; } +/* Convert handle already copied to kernel space into a dentry. */ +static struct dentry * +xfs_khandle_to_dentry( + struct file *file, + struct xfs_handle *handle) +{ + struct xfs_fid64 fid = { + .ino = handle->ha_fid.fid_ino, + .gen = handle->ha_fid.fid_gen, + }; + + /* + * Only allow handle opens under a directory. + */ + if (!S_ISDIR(file_inode(file)->i_mode)) + return ERR_PTR(-ENOTDIR); + + if (handle->ha_fid.fid_len != xfs_filehandle_fid_len()) + return ERR_PTR(-EINVAL); + + return exportfs_decode_fh(file->f_path.mnt, (struct fid *)&fid, 3, + FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG, + xfs_handle_acceptable, NULL); +} + /* * Convert userspace handle data into a dentry. */ @@ -136,29 +188,13 @@ xfs_handle_to_dentry( u32 hlen) { xfs_handle_t handle; - struct xfs_fid64 fid; - - /* - * Only allow handle opens under a directory. - */ - if (!S_ISDIR(file_inode(parfilp)->i_mode)) - return ERR_PTR(-ENOTDIR); if (hlen != sizeof(xfs_handle_t)) return ERR_PTR(-EINVAL); if (copy_from_user(&handle, uhandle, hlen)) return ERR_PTR(-EFAULT); - if (handle.ha_fid.fid_len != - sizeof(handle.ha_fid) - sizeof(handle.ha_fid.fid_len)) - return ERR_PTR(-EINVAL); - - memset(&fid, 0, sizeof(struct fid)); - fid.ino = handle.ha_fid.fid_ino; - fid.gen = handle.ha_fid.fid_gen; - return exportfs_decode_fh(parfilp->f_path.mnt, (struct fid *)&fid, 3, - FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG, - xfs_handle_acceptable, NULL); + return xfs_khandle_to_dentry(parfilp, &handle); } STATIC struct dentry * -- cgit v1.2.3-59-g8ed1b From 233f4e12bbb2c5fb1588b857336a26e8bb6942af Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:55 -0700 Subject: xfs: add parent pointer ioctls This patch adds a pair of new file ioctls to retrieve the parent pointer of a given inode. They both return the same results, but one operates on the file descriptor passed to ioctl() whereas the other allows the caller to specify a file handle for which the caller wants results. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_fs.h | 74 +++++++++++ fs/xfs/libxfs/xfs_ondisk.h | 5 + fs/xfs/libxfs/xfs_parent.c | 34 ++++++ fs/xfs/libxfs/xfs_parent.h | 5 + fs/xfs/xfs_export.c | 2 +- fs/xfs/xfs_export.h | 2 + fs/xfs/xfs_handle.c | 298 +++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_handle.h | 5 + fs/xfs/xfs_ioctl.c | 6 +- fs/xfs/xfs_trace.c | 1 + fs/xfs/xfs_trace.h | 92 ++++++++++++++ 11 files changed, 522 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index 97384ab95de4..ea654df0505f 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -816,6 +816,78 @@ struct xfs_exchange_range { XFS_EXCHANGE_RANGE_DRY_RUN | \ XFS_EXCHANGE_RANGE_FILE1_WRITTEN) +/* Iterating parent pointers of files. */ + +/* target was the root directory */ +#define XFS_GETPARENTS_OFLAG_ROOT (1U << 0) + +/* Cursor is done iterating pptrs */ +#define XFS_GETPARENTS_OFLAG_DONE (1U << 1) + +#define XFS_GETPARENTS_OFLAGS_ALL (XFS_GETPARENTS_OFLAG_ROOT | \ + XFS_GETPARENTS_OFLAG_DONE) + +#define XFS_GETPARENTS_IFLAGS_ALL (0) + +struct xfs_getparents_rec { + struct xfs_handle gpr_parent; /* Handle to parent */ + __u32 gpr_reclen; /* Length of entire record */ + __u32 gpr_reserved; /* zero */ + char gpr_name[]; /* Null-terminated filename */ +}; + +/* Iterate through this file's directory parent pointers */ +struct xfs_getparents { + /* + * Structure to track progress in iterating the parent pointers. + * Must be initialized to zeroes before the first ioctl call, and + * not touched by callers after that. + */ + struct xfs_attrlist_cursor gp_cursor; + + /* Input flags: XFS_GETPARENTS_IFLAG* */ + __u16 gp_iflags; + + /* Output flags: XFS_GETPARENTS_OFLAG* */ + __u16 gp_oflags; + + /* Size of the gp_buffer in bytes */ + __u32 gp_bufsize; + + /* Must be set to zero */ + __u64 gp_reserved; + + /* Pointer to a buffer in which to place xfs_getparents_rec */ + __u64 gp_buffer; +}; + +static inline struct xfs_getparents_rec * +xfs_getparents_first_rec(struct xfs_getparents *gp) +{ + return (struct xfs_getparents_rec *)(uintptr_t)gp->gp_buffer; +} + +static inline struct xfs_getparents_rec * +xfs_getparents_next_rec(struct xfs_getparents *gp, + struct xfs_getparents_rec *gpr) +{ + void *next = ((void *)gpr + gpr->gpr_reclen); + void *end = (void *)(uintptr_t)(gp->gp_buffer + gp->gp_bufsize); + + if (next >= end) + return NULL; + + return next; +} + +/* Iterate through this file handle's directory parent pointers. */ +struct xfs_getparents_by_handle { + /* Handle to file whose parents we want. */ + struct xfs_handle gph_handle; + + struct xfs_getparents gph_request; +}; + /* * ioctl commands that are used by Linux filesystems */ @@ -851,6 +923,8 @@ struct xfs_exchange_range { /* XFS_IOC_GETFSMAP ------ hoisted 59 */ #define XFS_IOC_SCRUB_METADATA _IOWR('X', 60, struct xfs_scrub_metadata) #define XFS_IOC_AG_GEOMETRY _IOWR('X', 61, struct xfs_ag_geometry) +#define XFS_IOC_GETPARENTS _IOWR('X', 62, struct xfs_getparents) +#define XFS_IOC_GETPARENTS_BY_HANDLE _IOWR('X', 63, struct xfs_getparents_by_handle) /* * ioctl commands that replace IRIX syssgi()'s diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 25952ef584ee..e8cdd77d03fa 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -156,6 +156,11 @@ xfs_check_ondisk_structs(void) XFS_CHECK_OFFSET(struct xfs_efi_log_format_32, efi_extents, 16); XFS_CHECK_OFFSET(struct xfs_efi_log_format_64, efi_extents, 16); + /* parent pointer ioctls */ + XFS_CHECK_STRUCT_SIZE(struct xfs_getparents_rec, 32); + XFS_CHECK_STRUCT_SIZE(struct xfs_getparents, 40); + XFS_CHECK_STRUCT_SIZE(struct xfs_getparents_by_handle, 64); + /* * The v5 superblock format extended several v4 header structures with * additional data. While new fields are only accessible on v5 diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c index fdf643bfde4d..504de1ef3387 100644 --- a/fs/xfs/libxfs/xfs_parent.c +++ b/fs/xfs/libxfs/xfs_parent.c @@ -257,3 +257,37 @@ xfs_parent_replacename( xfs_attr_defer_add(&ppargs->args, XFS_ATTR_DEFER_REPLACE); return 0; } + +/* + * Extract parent pointer information from any parent pointer xattr into + * @parent_ino/gen. The last two parameters can be NULL pointers. + * + * Returns 0 if this is not a parent pointer xattr at all; or -EFSCORRUPTED for + * garbage. + */ +int +xfs_parent_from_attr( + struct xfs_mount *mp, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + xfs_ino_t *parent_ino, + uint32_t *parent_gen) +{ + const struct xfs_parent_rec *rec = value; + + ASSERT(attr_flags & XFS_ATTR_PARENT); + + if (!xfs_parent_namecheck(attr_flags, name, namelen)) + return -EFSCORRUPTED; + if (!xfs_parent_valuecheck(mp, value, valuelen)) + return -EFSCORRUPTED; + + if (parent_ino) + *parent_ino = be64_to_cpu(rec->p_ino); + if (parent_gen) + *parent_gen = be32_to_cpu(rec->p_gen); + return 0; +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h index 768633b31367..d7ab09e738ad 100644 --- a/fs/xfs/libxfs/xfs_parent.h +++ b/fs/xfs/libxfs/xfs_parent.h @@ -91,4 +91,9 @@ int xfs_parent_replacename(struct xfs_trans *tp, struct xfs_inode *new_dp, const struct xfs_name *new_name, struct xfs_inode *child); +int xfs_parent_from_attr(struct xfs_mount *mp, unsigned int attr_flags, + const unsigned char *name, unsigned int namelen, + const void *value, unsigned int valuelen, + xfs_ino_t *parent_ino, uint32_t *parent_gen); + #endif /* __XFS_PARENT_H__ */ diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c index 4b03221351c0..201489d3de08 100644 --- a/fs/xfs/xfs_export.c +++ b/fs/xfs/xfs_export.c @@ -102,7 +102,7 @@ xfs_fs_encode_fh( return fileid_type; } -STATIC struct inode * +struct inode * xfs_nfs_get_inode( struct super_block *sb, u64 ino, diff --git a/fs/xfs/xfs_export.h b/fs/xfs/xfs_export.h index 64471a3ddb04..3cd85e8901a5 100644 --- a/fs/xfs/xfs_export.h +++ b/fs/xfs/xfs_export.h @@ -57,4 +57,6 @@ struct xfs_fid64 { /* This flag goes on the wire. Don't play with it. */ #define XFS_FILEID_TYPE_64FLAG 0x80 /* NFS fileid has 64bit inodes */ +struct inode *xfs_nfs_get_inode(struct super_block *sb, u64 ino, u32 gen); + #endif /* __XFS_EXPORT_H__ */ diff --git a/fs/xfs/xfs_handle.c b/fs/xfs/xfs_handle.c index b9f4d9860682..c8785ed59543 100644 --- a/fs/xfs/xfs_handle.c +++ b/fs/xfs/xfs_handle.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2022-2024 Oracle. * All rights reserved. */ #include "xfs.h" @@ -178,6 +179,30 @@ xfs_khandle_to_dentry( xfs_handle_acceptable, NULL); } +/* Convert handle already copied to kernel space into an xfs_inode. */ +static struct xfs_inode * +xfs_khandle_to_inode( + struct file *file, + struct xfs_handle *handle) +{ + struct xfs_inode *ip = XFS_I(file_inode(file)); + struct xfs_mount *mp = ip->i_mount; + struct inode *inode; + + if (!S_ISDIR(VFS_I(ip)->i_mode)) + return ERR_PTR(-ENOTDIR); + + if (handle->ha_fid.fid_len != xfs_filehandle_fid_len()) + return ERR_PTR(-EINVAL); + + inode = xfs_nfs_get_inode(mp->m_super, handle->ha_fid.fid_ino, + handle->ha_fid.fid_gen); + if (IS_ERR(inode)) + return ERR_CAST(inode); + + return XFS_I(inode); +} + /* * Convert userspace handle data into a dentry. */ @@ -652,3 +677,276 @@ xfs_attrmulti_by_handle( dput(dentry); return error; } + +struct xfs_getparents_ctx { + struct xfs_attr_list_context context; + struct xfs_getparents_by_handle gph; + + /* File to target */ + struct xfs_inode *ip; + + /* Internal buffer where we format records */ + void *krecords; + + /* Last record filled out */ + struct xfs_getparents_rec *lastrec; + + unsigned int count; +}; + +static inline unsigned int +xfs_getparents_rec_sizeof( + unsigned int namelen) +{ + return round_up(sizeof(struct xfs_getparents_rec) + namelen + 1, + sizeof(uint64_t)); +} + +static void +xfs_getparents_put_listent( + struct xfs_attr_list_context *context, + int flags, + unsigned char *name, + int namelen, + void *value, + int valuelen) +{ + struct xfs_getparents_ctx *gpx = + container_of(context, struct xfs_getparents_ctx, context); + struct xfs_inode *ip = context->dp; + struct xfs_mount *mp = ip->i_mount; + struct xfs_getparents *gp = &gpx->gph.gph_request; + struct xfs_getparents_rec *gpr = gpx->krecords + context->firstu; + unsigned short reclen = + xfs_getparents_rec_sizeof(namelen); + xfs_ino_t ino; + uint32_t gen; + int error; + + if (!(flags & XFS_ATTR_PARENT)) + return; + + error = xfs_parent_from_attr(mp, flags, name, namelen, value, valuelen, + &ino, &gen); + if (error) { + xfs_inode_mark_sick(ip, XFS_SICK_INO_PARENT); + context->seen_enough = -EFSCORRUPTED; + return; + } + + /* + * We found a parent pointer, but we've filled up the buffer. Signal + * to the caller that we did /not/ reach the end of the parent pointer + * recordset. + */ + if (context->firstu > context->bufsize - reclen) { + context->seen_enough = 1; + return; + } + + /* Format the parent pointer directly into the caller buffer. */ + gpr->gpr_reclen = reclen; + xfs_filehandle_init(mp, ino, gen, &gpr->gpr_parent); + memcpy(gpr->gpr_name, name, namelen); + gpr->gpr_name[namelen] = 0; + + trace_xfs_getparents_put_listent(ip, gp, context, gpr); + + context->firstu += reclen; + gpx->count++; + gpx->lastrec = gpr; +} + +/* Expand the last record to fill the rest of the caller's buffer. */ +static inline void +xfs_getparents_expand_lastrec( + struct xfs_getparents_ctx *gpx) +{ + struct xfs_getparents *gp = &gpx->gph.gph_request; + struct xfs_getparents_rec *gpr = gpx->lastrec; + + if (!gpx->lastrec) + gpr = gpx->krecords; + + gpr->gpr_reclen = gp->gp_bufsize - ((void *)gpr - gpx->krecords); + + trace_xfs_getparents_expand_lastrec(gpx->ip, gp, &gpx->context, gpr); +} + +static inline void __user *u64_to_uptr(u64 val) +{ + return (void __user *)(uintptr_t)val; +} + +/* Retrieve the parent pointers for a given inode. */ +STATIC int +xfs_getparents( + struct xfs_getparents_ctx *gpx) +{ + struct xfs_getparents *gp = &gpx->gph.gph_request; + struct xfs_inode *ip = gpx->ip; + struct xfs_mount *mp = ip->i_mount; + size_t bufsize; + int error; + + /* Check size of buffer requested by user */ + if (gp->gp_bufsize > XFS_XATTR_LIST_MAX) + return -ENOMEM; + if (gp->gp_bufsize < xfs_getparents_rec_sizeof(1)) + return -EINVAL; + + if (gp->gp_iflags & ~XFS_GETPARENTS_IFLAGS_ALL) + return -EINVAL; + if (gp->gp_reserved) + return -EINVAL; + + bufsize = round_down(gp->gp_bufsize, sizeof(uint64_t)); + gpx->krecords = kvzalloc(bufsize, GFP_KERNEL); + if (!gpx->krecords) { + bufsize = min(bufsize, PAGE_SIZE); + gpx->krecords = kvzalloc(bufsize, GFP_KERNEL); + if (!gpx->krecords) + return -ENOMEM; + } + + gpx->context.dp = ip; + gpx->context.resynch = 1; + gpx->context.put_listent = xfs_getparents_put_listent; + gpx->context.bufsize = bufsize; + /* firstu is used to track the bytes filled in the buffer */ + gpx->context.firstu = 0; + + /* Copy the cursor provided by caller */ + memcpy(&gpx->context.cursor, &gp->gp_cursor, + sizeof(struct xfs_attrlist_cursor)); + gpx->count = 0; + gp->gp_oflags = 0; + + trace_xfs_getparents_begin(ip, gp, &gpx->context.cursor); + + error = xfs_attr_list(&gpx->context); + if (error) + goto out_free_buf; + if (gpx->context.seen_enough < 0) { + error = gpx->context.seen_enough; + goto out_free_buf; + } + xfs_getparents_expand_lastrec(gpx); + + /* Update the caller with the current cursor position */ + memcpy(&gp->gp_cursor, &gpx->context.cursor, + sizeof(struct xfs_attrlist_cursor)); + + /* Is this the root directory? */ + if (ip->i_ino == mp->m_sb.sb_rootino) + gp->gp_oflags |= XFS_GETPARENTS_OFLAG_ROOT; + + if (gpx->context.seen_enough == 0) { + /* + * If we did not run out of buffer space, then we reached the + * end of the pptr recordset, so set the DONE flag. + */ + gp->gp_oflags |= XFS_GETPARENTS_OFLAG_DONE; + } else if (gpx->count == 0) { + /* + * If we ran out of buffer space before copying any parent + * pointers at all, the caller's buffer was too short. Tell + * userspace that, erm, the message is too long. + */ + error = -EMSGSIZE; + goto out_free_buf; + } + + trace_xfs_getparents_end(ip, gp, &gpx->context.cursor); + + ASSERT(gpx->context.firstu <= gpx->gph.gph_request.gp_bufsize); + + /* Copy the records to userspace. */ + if (copy_to_user(u64_to_uptr(gpx->gph.gph_request.gp_buffer), + gpx->krecords, gpx->context.firstu)) + error = -EFAULT; + +out_free_buf: + kvfree(gpx->krecords); + gpx->krecords = NULL; + return error; +} + +/* Retrieve the parents of this file and pass them back to userspace. */ +int +xfs_ioc_getparents( + struct file *file, + struct xfs_getparents __user *ureq) +{ + struct xfs_getparents_ctx gpx = { + .ip = XFS_I(file_inode(file)), + }; + struct xfs_getparents *kreq = &gpx.gph.gph_request; + struct xfs_mount *mp = gpx.ip->i_mount; + int error; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (!xfs_has_parent(mp)) + return -EOPNOTSUPP; + if (copy_from_user(kreq, ureq, sizeof(*kreq))) + return -EFAULT; + + error = xfs_getparents(&gpx); + if (error) + return error; + + if (copy_to_user(ureq, kreq, sizeof(*kreq))) + return -EFAULT; + + return 0; +} + +/* Retrieve the parents of this file handle and pass them back to userspace. */ +int +xfs_ioc_getparents_by_handle( + struct file *file, + struct xfs_getparents_by_handle __user *ureq) +{ + struct xfs_getparents_ctx gpx = { }; + struct xfs_inode *ip = XFS_I(file_inode(file)); + struct xfs_mount *mp = ip->i_mount; + struct xfs_getparents_by_handle *kreq = &gpx.gph; + struct xfs_handle *handle = &kreq->gph_handle; + int error; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (!xfs_has_parent(mp)) + return -EOPNOTSUPP; + if (copy_from_user(kreq, ureq, sizeof(*kreq))) + return -EFAULT; + + /* + * We don't use exportfs_decode_fh because it does too much work here. + * If the handle refers to a directory, the exportfs code will walk + * upwards through the directory tree to connect the dentries to the + * root directory dentry. For GETPARENTS we don't care about that + * because we're not actually going to open a file descriptor; we only + * want to open an inode and read its parent pointers. + * + * Note that xfs_scrub uses GETPARENTS to log that it will try to fix a + * corrupted file's metadata. For this usecase we would really rather + * userspace single-step the path reconstruction to avoid loops or + * other strange things if the directory tree is corrupt. + */ + gpx.ip = xfs_khandle_to_inode(file, handle); + if (IS_ERR(gpx.ip)) + return PTR_ERR(gpx.ip); + + error = xfs_getparents(&gpx); + if (error) + goto out_rele; + + if (copy_to_user(ureq, kreq, sizeof(*kreq))) + error = -EFAULT; + +out_rele: + xfs_irele(gpx.ip); + return error; +} diff --git a/fs/xfs/xfs_handle.h b/fs/xfs/xfs_handle.h index e39eaf4689da..6799a86d8565 100644 --- a/fs/xfs/xfs_handle.h +++ b/fs/xfs/xfs_handle.h @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2022-2024 Oracle. * All rights reserved. */ #ifndef __XFS_HANDLE_H__ @@ -25,4 +26,8 @@ int xfs_ioc_attr_list(struct xfs_inode *dp, void __user *ubuf, struct dentry *xfs_handle_to_dentry(struct file *parfilp, void __user *uhandle, u32 hlen); +int xfs_ioc_getparents(struct file *file, struct xfs_getparents __user *arg); +int xfs_ioc_getparents_by_handle(struct file *file, + struct xfs_getparents_by_handle __user *arg); + #endif /* __XFS_HANDLE_H__ */ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index ed05fcd6261d..0e97070abe80 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -35,6 +35,7 @@ #include "xfs_health.h" #include "xfs_reflink.h" #include "xfs_ioctl.h" +#include "xfs_xattr.h" #include "xfs_rtbitmap.h" #include "xfs_file.h" #include "xfs_exchrange.h" @@ -1424,7 +1425,10 @@ xfs_file_ioctl( case XFS_IOC_FSGETXATTRA: return xfs_ioc_fsgetxattra(ip, arg); - + case XFS_IOC_GETPARENTS: + return xfs_ioc_getparents(filp, arg); + case XFS_IOC_GETPARENTS_BY_HANDLE: + return xfs_ioc_getparents_by_handle(filp, arg); case XFS_IOC_GETBMAP: case XFS_IOC_GETBMAPA: case XFS_IOC_GETBMAPX: diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c index cf92a3bd56c7..9c7fbaae2717 100644 --- a/fs/xfs/xfs_trace.c +++ b/fs/xfs/xfs_trace.c @@ -41,6 +41,7 @@ #include "xfs_bmap.h" #include "xfs_exchmaps.h" #include "xfs_exchrange.h" +#include "xfs_parent.h" /* * We include this last to have the helpers above available for the trace diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 5621db48e763..05cb59bd0b80 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -87,6 +87,9 @@ struct xfs_bmap_intent; struct xfs_exchmaps_intent; struct xfs_exchmaps_req; struct xfs_exchrange; +struct xfs_getparents; +struct xfs_parent_irec; +struct xfs_attrlist_cursor_kern; #define XFS_ATTR_FILTER_FLAGS \ { XFS_ATTR_ROOT, "ROOT" }, \ @@ -5096,6 +5099,95 @@ TRACE_EVENT(xfs_exchmaps_delta_nextents, __entry->d_nexts1, __entry->d_nexts2) ); +DECLARE_EVENT_CLASS(xfs_getparents_rec_class, + TP_PROTO(struct xfs_inode *ip, const struct xfs_getparents *ppi, + const struct xfs_attr_list_context *context, + const struct xfs_getparents_rec *pptr), + TP_ARGS(ip, ppi, context, pptr), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(unsigned int, firstu) + __field(unsigned short, reclen) + __field(unsigned int, bufsize) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, parent_gen) + __string(name, pptr->gpr_name) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->firstu = context->firstu; + __entry->reclen = pptr->gpr_reclen; + __entry->bufsize = ppi->gp_bufsize; + __entry->parent_ino = pptr->gpr_parent.ha_fid.fid_ino; + __entry->parent_gen = pptr->gpr_parent.ha_fid.fid_gen; + __assign_str(name, pptr->gpr_name); + ), + TP_printk("dev %d:%d ino 0x%llx firstu %u reclen %u bufsize %u parent_ino 0x%llx parent_gen 0x%x name '%s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->firstu, + __entry->reclen, + __entry->bufsize, + __entry->parent_ino, + __entry->parent_gen, + __get_str(name)) +) +#define DEFINE_XFS_GETPARENTS_REC_EVENT(name) \ +DEFINE_EVENT(xfs_getparents_rec_class, name, \ + TP_PROTO(struct xfs_inode *ip, const struct xfs_getparents *ppi, \ + const struct xfs_attr_list_context *context, \ + const struct xfs_getparents_rec *pptr), \ + TP_ARGS(ip, ppi, context, pptr)) +DEFINE_XFS_GETPARENTS_REC_EVENT(xfs_getparents_put_listent); +DEFINE_XFS_GETPARENTS_REC_EVENT(xfs_getparents_expand_lastrec); + +DECLARE_EVENT_CLASS(xfs_getparents_class, + TP_PROTO(struct xfs_inode *ip, const struct xfs_getparents *ppi, + const struct xfs_attrlist_cursor_kern *cur), + TP_ARGS(ip, ppi, cur), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(unsigned short, iflags) + __field(unsigned short, oflags) + __field(unsigned int, bufsize) + __field(unsigned int, hashval) + __field(unsigned int, blkno) + __field(unsigned int, offset) + __field(int, initted) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->iflags = ppi->gp_iflags; + __entry->oflags = ppi->gp_oflags; + __entry->bufsize = ppi->gp_bufsize; + __entry->hashval = cur->hashval; + __entry->blkno = cur->blkno; + __entry->offset = cur->offset; + __entry->initted = cur->initted; + ), + TP_printk("dev %d:%d ino 0x%llx iflags 0x%x oflags 0x%x bufsize %u cur_init? %d hashval 0x%x blkno %u offset %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->iflags, + __entry->oflags, + __entry->bufsize, + __entry->initted, + __entry->hashval, + __entry->blkno, + __entry->offset) +) +#define DEFINE_XFS_GETPARENTS_EVENT(name) \ +DEFINE_EVENT(xfs_getparents_class, name, \ + TP_PROTO(struct xfs_inode *ip, const struct xfs_getparents *ppi, \ + const struct xfs_attrlist_cursor_kern *cur), \ + TP_ARGS(ip, ppi, cur)) +DEFINE_XFS_GETPARENTS_EVENT(xfs_getparents_begin); +DEFINE_XFS_GETPARENTS_EVENT(xfs_getparents_end); + #endif /* _TRACE_XFS_H */ #undef TRACE_INCLUDE_PATH -- cgit v1.2.3-59-g8ed1b From 7dafb449b7922c1eec6fee3ed85b679d51f0f431 Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:56 -0700 Subject: xfs: don't remove the attr fork when parent pointers are enabled When an inode is removed, it may also cause the attribute fork to be removed if it is the last attribute. This transaction gets flushed to the log, but if the system goes down before we could inactivate the symlink, the log recovery tries to inactivate this inode (since it is on the unlinked list) but the verifier trips over the remote value and leaks it. Hence we ended up with a file in this odd state on a "clean" mount. The "obvious" fix is to prohibit erasure of the attr fork to avoid tripping over the verifiers when pptrs are enabled. Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_leaf.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index 1a374c6885d7..b9e98950eb3d 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -891,7 +891,8 @@ xfs_attr_sf_removename( */ if (totsize == sizeof(struct xfs_attr_sf_hdr) && xfs_has_attr2(mp) && (dp->i_df.if_format != XFS_DINODE_FMT_BTREE) && - !(args->op_flags & (XFS_DA_OP_ADDNAME | XFS_DA_OP_REPLACE))) { + !(args->op_flags & (XFS_DA_OP_ADDNAME | XFS_DA_OP_REPLACE)) && + !xfs_has_parent(mp)) { xfs_attr_fork_remove(dp, args->trans); } else { xfs_idata_realloc(dp, -size, XFS_ATTR_FORK); @@ -900,7 +901,8 @@ xfs_attr_sf_removename( ASSERT(totsize > sizeof(struct xfs_attr_sf_hdr) || (args->op_flags & XFS_DA_OP_ADDNAME) || !xfs_has_attr2(mp) || - dp->i_df.if_format == XFS_DINODE_FMT_BTREE); + dp->i_df.if_format == XFS_DINODE_FMT_BTREE || + xfs_has_parent(mp)); xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_ADATA); } -- cgit v1.2.3-59-g8ed1b From 5f98ec1cb5c264e4815e21d632ee0b3d6e700e3d Mon Sep 17 00:00:00 2001 From: Allison Henderson Date: Mon, 22 Apr 2024 09:47:57 -0700 Subject: xfs: add a incompat feature bit for parent pointers Create an incompat feature bit and a fs geometry flag so that we can enable the feature in the ondisk superblock and advertise its existence to userspace. Signed-off-by: Mark Tinguely Signed-off-by: Dave Chinner Signed-off-by: Allison Henderson Reviewed-by: Darrick J. Wong Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong --- fs/xfs/libxfs/xfs_format.h | 1 + fs/xfs/libxfs/xfs_fs.h | 1 + fs/xfs/libxfs/xfs_sb.c | 4 ++++ fs/xfs/xfs_super.c | 4 ++++ 4 files changed, 10 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index f1818c54af6f..b457e457e1f7 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -374,6 +374,7 @@ xfs_sb_has_ro_compat_feature( #define XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR (1 << 4) /* needs xfs_repair */ #define XFS_SB_FEAT_INCOMPAT_NREXT64 (1 << 5) /* large extent counters */ #define XFS_SB_FEAT_INCOMPAT_EXCHRANGE (1 << 6) /* exchangerange supported */ +#define XFS_SB_FEAT_INCOMPAT_PARENT (1 << 7) /* parent pointers */ #define XFS_SB_FEAT_INCOMPAT_ALL \ (XFS_SB_FEAT_INCOMPAT_FTYPE | \ XFS_SB_FEAT_INCOMPAT_SPINODES | \ diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index ea654df0505f..dd13bfa500f2 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -240,6 +240,7 @@ typedef struct xfs_fsop_resblks { #define XFS_FSOP_GEOM_FLAGS_INOBTCNT (1 << 22) /* inobt btree counter */ #define XFS_FSOP_GEOM_FLAGS_NREXT64 (1 << 23) /* large extent counters */ #define XFS_FSOP_GEOM_FLAGS_EXCHANGE_RANGE (1 << 24) /* exchange range */ +#define XFS_FSOP_GEOM_FLAGS_PARENT (1 << 25) /* linux parent pointers */ /* * Minimum and maximum sizes need for growth checks. diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index c350e259b685..09e4bf949bf8 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -178,6 +178,8 @@ xfs_sb_version_to_features( features |= XFS_FEAT_NREXT64; if (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_EXCHRANGE) features |= XFS_FEAT_EXCHANGE_RANGE; + if (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_PARENT) + features |= XFS_FEAT_PARENT; return features; } @@ -1254,6 +1256,8 @@ xfs_fs_geometry( geo->flags |= XFS_FSOP_GEOM_FLAGS_BIGTIME; if (xfs_has_inobtcounts(mp)) geo->flags |= XFS_FSOP_GEOM_FLAGS_INOBTCNT; + if (xfs_has_parent(mp)) + geo->flags |= XFS_FSOP_GEOM_FLAGS_PARENT; if (xfs_has_sector(mp)) { geo->flags |= XFS_FSOP_GEOM_FLAGS_SECTOR; geo->logsectsize = sbp->sb_logsectsize; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index c303d7ff9597..27e9f749c4c7 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1746,6 +1746,10 @@ xfs_fs_fill_super( xfs_warn(mp, "EXPERIMENTAL exchange-range feature enabled. Use at your own risk!"); + if (xfs_has_parent(mp)) + xfs_warn(mp, + "EXPERIMENTAL parent pointer feature enabled. Use at your own risk!"); + error = xfs_mountfs(mp); if (error) goto out_filestream_unmount; -- cgit v1.2.3-59-g8ed1b From 7ea816ca4043c2bc6052f696b6aebe2c22980a03 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:58 -0700 Subject: xfs: fix unit conversion error in xfs_log_calc_max_attrsetm_res Dave and I were discussing some recent test regressions as a result of me turning on nrext64=1 on realtime filesystems, when we noticed that the minimum log size of a 32M filesystem jumped from 954 blocks to 4287 blocks. Digging through xfs_log_calc_max_attrsetm_res, Dave noticed that @size contains the maximum estimated amount of space needed for a local format xattr, in bytes, but we feed this quantity to XFS_NEXTENTADD_SPACE_RES, which requires units of blocks. This has resulted in an overestimation of the minimum log size over the years. We should nominally correct this, but there's a backwards compatibility problem -- if we enable it now, the minimum log size will decrease. If a corrected mkfs formats a filesystem with this new smaller log size, a user will encounter mount failures on an uncorrected kernel due to the larger minimum log size computations there. Therefore, turn this on for parent pointers because it wasn't merged at all upstream when this issue was discovered. Signed-off-by: Darrick J. Wong Reviewed-by: Allison Henderson Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_log_rlimit.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_log_rlimit.c b/fs/xfs/libxfs/xfs_log_rlimit.c index 9975b93a7412..3518d5e21df0 100644 --- a/fs/xfs/libxfs/xfs_log_rlimit.c +++ b/fs/xfs/libxfs/xfs_log_rlimit.c @@ -16,6 +16,29 @@ #include "xfs_bmap_btree.h" #include "xfs_trace.h" +/* + * Shortly after enabling the large extents count feature in 2023, longstanding + * bugs were found in the code that computes the minimum log size. Luckily, + * the bugs resulted in over-estimates of that size, so there's no impact to + * existing users. However, we don't want to reduce the minimum log size + * because that can create the situation where a newer mkfs writes a new + * filesystem that an older kernel won't mount. + * + * Therefore, we only may correct the computation starting with filesystem + * features that didn't exist in 2023. In other words, only turn this on if + * the filesystem has parent pointers. + * + * This function can be called before the XFS_HAS_* flags have been set up, + * (e.g. mkfs) so we must check the ondisk superblock. + */ +static inline bool +xfs_want_minlogsize_fixes( + struct xfs_sb *sb) +{ + return xfs_sb_is_v5(sb) && + xfs_sb_has_incompat_feature(sb, XFS_SB_FEAT_INCOMPAT_PARENT); +} + /* * Calculate the maximum length in bytes that would be required for a local * attribute value as large attributes out of line are not logged. @@ -31,6 +54,15 @@ xfs_log_calc_max_attrsetm_res( MAXNAMELEN - 1; nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK); nblks += XFS_B_TO_FSB(mp, size); + + /* + * If the feature set is new enough, correct a unit conversion error in + * the xattr transaction reservation code that resulted in oversized + * minimum log size computations. + */ + if (xfs_want_minlogsize_fixes(&mp->m_sb)) + size = XFS_B_TO_FSB(mp, size); + nblks += XFS_NEXTENTADD_SPACE_RES(mp, size, XFS_ATTR_FORK); return M_RES(mp)->tr_attrsetm.tr_logres + -- cgit v1.2.3-59-g8ed1b From 6ed858c7c678218aa8df9d9e75d5e9955c105415 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:47:59 -0700 Subject: xfs: drop compatibility minimum log size computations for reflink Let's also drop the oversized minimum log computations for reflink and rmap that were the result of bugs introduced many years ago. Signed-off-by: Darrick J. Wong Reviewed-by: Allison Henderson Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_log_rlimit.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_log_rlimit.c b/fs/xfs/libxfs/xfs_log_rlimit.c index 3518d5e21df0..d3bd6a86c8fe 100644 --- a/fs/xfs/libxfs/xfs_log_rlimit.c +++ b/fs/xfs/libxfs/xfs_log_rlimit.c @@ -24,6 +24,11 @@ * because that can create the situation where a newer mkfs writes a new * filesystem that an older kernel won't mount. * + * Several years prior, we also discovered that the transaction reservations + * for rmap and reflink operations were unnecessarily large. That was fixed, + * but the minimum log size computation was left alone to avoid the + * compatibility problems noted above. Fix that too. + * * Therefore, we only may correct the computation starting with filesystem * features that didn't exist in 2023. In other words, only turn this on if * the filesystem has parent pointers. @@ -80,6 +85,15 @@ xfs_log_calc_trans_resv_for_minlogblocks( { unsigned int rmap_maxlevels = mp->m_rmap_maxlevels; + /* + * If the feature set is new enough, drop the oversized minimum log + * size computation introduced by the original reflink code. + */ + if (xfs_want_minlogsize_fixes(&mp->m_sb)) { + xfs_trans_resv_calc(mp, resv); + return; + } + /* * In the early days of rmap+reflink, we always set the rmap maxlevels * to 9 even if the AG was small enough that it would never grow to -- cgit v1.2.3-59-g8ed1b From 67ac7091e35bd34b75c0ec77331b53ca052e0cb3 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:00 -0700 Subject: xfs: enable parent pointers Add parent pointers to the list of supported features. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_format.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index b457e457e1f7..61f51becff4f 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -382,7 +382,8 @@ xfs_sb_has_ro_compat_feature( XFS_SB_FEAT_INCOMPAT_BIGTIME | \ XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR | \ XFS_SB_FEAT_INCOMPAT_NREXT64 | \ - XFS_SB_FEAT_INCOMPAT_EXCHRANGE) + XFS_SB_FEAT_INCOMPAT_EXCHRANGE | \ + XFS_SB_FEAT_INCOMPAT_PARENT) #define XFS_SB_FEAT_INCOMPAT_UNKNOWN ~XFS_SB_FEAT_INCOMPAT_ALL static inline bool -- cgit v1.2.3-59-g8ed1b From 2a009397eb5ae178670cbd7101e9635cf6412b35 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:00 -0700 Subject: xfs: revert commit 44af6c7e59b12 In my haste to fix what I thought was a performance problem in the attr scrub code, I neglected to notice that the xfs_attr_get_ilocked also had the effect of checking that attributes can actually be looked up through the attr dabtree. Fix this. Fixes: 44af6c7e59b12 ("xfs: don't load local xattr values during scrub") Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index c07d050b39b2..393ed36709b3 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -208,14 +208,6 @@ xchk_xattr_actor( return -ECANCELED; } - /* - * Local and shortform xattr values are stored in the attr leaf block, - * so we don't need to retrieve the value from a remote block to detect - * corruption problems. - */ - if (value) - return 0; - /* * Try to allocate enough memory to extract the attr value. If that * doesn't work, return -EDEADLOCK as a signal to try again with a @@ -229,6 +221,11 @@ xchk_xattr_actor( args.value = ab->value; + /* + * Get the attr value to ensure that lookup can find this attribute + * through the dabtree indexing and that remote value retrieval also + * works correctly. + */ xfs_attr_sethash(&args); error = xfs_attr_get_ilocked(&args); /* ENODATA means the hash lookup failed and the attr is bad */ -- cgit v1.2.3-59-g8ed1b From 61b3f0df5c235806d372aaf696ce9aee7746d18f Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:01 -0700 Subject: xfs: check dirents have parent pointers If the fs has parent pointers, we need to check that each child dirent points to a file that has a parent pointer pointing back at us. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_parent.c | 22 +++++++++ fs/xfs/libxfs/xfs_parent.h | 5 ++ fs/xfs/scrub/dir.c | 112 ++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 138 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c index 504de1ef3387..1e53df5f8332 100644 --- a/fs/xfs/libxfs/xfs_parent.c +++ b/fs/xfs/libxfs/xfs_parent.c @@ -291,3 +291,25 @@ xfs_parent_from_attr( *parent_gen = be32_to_cpu(rec->p_gen); return 0; } + +/* + * Look up a parent pointer record (@parent_name -> @pptr) of @ip. + * + * Caller must hold at least ILOCK_SHARED. The scratchpad need not be + * initialized. + * + * Returns 0 if the pointer is found, -ENOATTR if there is no match, or a + * negative errno. + */ +int +xfs_parent_lookup( + struct xfs_trans *tp, + struct xfs_inode *ip, + const struct xfs_name *parent_name, + struct xfs_parent_rec *pptr, + struct xfs_da_args *scratch) +{ + memset(scratch, 0, sizeof(struct xfs_da_args)); + xfs_parent_da_args_init(scratch, tp, pptr, ip, ip->i_ino, parent_name); + return xfs_attr_get_ilocked(scratch); +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h index d7ab09e738ad..97788582321a 100644 --- a/fs/xfs/libxfs/xfs_parent.h +++ b/fs/xfs/libxfs/xfs_parent.h @@ -96,4 +96,9 @@ int xfs_parent_from_attr(struct xfs_mount *mp, unsigned int attr_flags, const void *value, unsigned int valuelen, xfs_ino_t *parent_ino, uint32_t *parent_gen); +/* Repair functions */ +int xfs_parent_lookup(struct xfs_trans *tp, struct xfs_inode *ip, + const struct xfs_name *name, struct xfs_parent_rec *pptr, + struct xfs_da_args *scratch); + #endif /* __XFS_PARENT_H__ */ diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 3fe6ffcf9c06..e11d73eb8935 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -16,6 +16,8 @@ #include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_health.h" +#include "xfs_attr.h" +#include "xfs_parent.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/dabtree.h" @@ -41,6 +43,14 @@ xchk_setup_directory( /* Directories */ +struct xchk_dir { + struct xfs_scrub *sc; + + /* information for parent pointer validation. */ + struct xfs_parent_rec pptr_rec; + struct xfs_da_args pptr_args; +}; + /* Scrub a directory entry. */ /* Check that an inode's mode matches a given XFS_DIR3_FT_* type. */ @@ -63,6 +73,90 @@ xchk_dir_check_ftype( xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, offset); } +/* + * Try to lock a child file for checking parent pointers. Returns the inode + * flags for the locks we now hold, or zero if we failed. + */ +STATIC unsigned int +xchk_dir_lock_child( + struct xfs_scrub *sc, + struct xfs_inode *ip) +{ + if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) + return 0; + + if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) { + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + return 0; + } + + if (!xfs_inode_has_attr_fork(ip) || !xfs_need_iread_extents(&ip->i_af)) + return XFS_IOLOCK_SHARED | XFS_ILOCK_SHARED; + + xfs_iunlock(ip, XFS_ILOCK_SHARED); + + if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) { + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + return 0; + } + + return XFS_IOLOCK_SHARED | XFS_ILOCK_EXCL; +} + +/* Check the backwards link (parent pointer) associated with this dirent. */ +STATIC int +xchk_dir_parent_pointer( + struct xchk_dir *sd, + const struct xfs_name *name, + struct xfs_inode *ip) +{ + struct xfs_scrub *sc = sd->sc; + int error; + + xfs_inode_to_parent_rec(&sd->pptr_rec, sc->ip); + error = xfs_parent_lookup(sc->tp, ip, name, &sd->pptr_rec, + &sd->pptr_args); + if (error == -ENOATTR) + xchk_fblock_xref_set_corrupt(sc, XFS_DATA_FORK, 0); + + return 0; +} + +/* Look for a parent pointer matching this dirent, if the child isn't busy. */ +STATIC int +xchk_dir_check_pptr_fast( + struct xchk_dir *sd, + xfs_dir2_dataptr_t dapos, + const struct xfs_name *name, + struct xfs_inode *ip) +{ + struct xfs_scrub *sc = sd->sc; + unsigned int lockmode; + int error; + + /* dot and dotdot entries do not have parent pointers */ + if (xfs_dir2_samename(name, &xfs_name_dot) || + xfs_dir2_samename(name, &xfs_name_dotdot)) + return 0; + + /* No self-referential non-dot or dotdot dirents. */ + if (ip == sc->ip) { + xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); + return -ECANCELED; + } + + /* Try to lock the inode. */ + lockmode = xchk_dir_lock_child(sc, ip); + if (!lockmode) { + xchk_set_incomplete(sc); + return -ECANCELED; + } + + error = xchk_dir_parent_pointer(sd, name, ip); + xfs_iunlock(ip, lockmode); + return error; +} + /* * Scrub a single directory entry. * @@ -80,6 +174,7 @@ xchk_dir_actor( { struct xfs_mount *mp = dp->i_mount; struct xfs_inode *ip; + struct xchk_dir *sd = priv; xfs_ino_t lookup_ino; xfs_dablk_t offset; int error = 0; @@ -146,6 +241,14 @@ xchk_dir_actor( goto out; xchk_dir_check_ftype(sc, offset, ip, name->type); + + if (xfs_has_parent(mp)) { + error = xchk_dir_check_pptr_fast(sd, dapos, name, ip); + if (error) + goto out_rele; + } + +out_rele: xchk_irele(sc, ip); out: if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) @@ -767,6 +870,7 @@ int xchk_directory( struct xfs_scrub *sc) { + struct xchk_dir *sd; int error; if (!S_ISDIR(VFS_I(sc->ip)->i_mode)) @@ -799,8 +903,14 @@ xchk_directory( if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) return 0; + sd = kvzalloc(sizeof(struct xchk_dir), XCHK_GFP_FLAGS); + if (!sd) + return -ENOMEM; + sd->sc = sc; + /* Look up every name in this directory by hash. */ - error = xchk_dir_walk(sc, sc->ip, xchk_dir_actor, NULL); + error = xchk_dir_walk(sc, sc->ip, xchk_dir_actor, sd); + kvfree(sd); if (error && error != -ECANCELED) return error; -- cgit v1.2.3-59-g8ed1b From b961c8bf1fc3d0232209a49f47d1cd7a55b7a861 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:02 -0700 Subject: xfs: deferred scrub of dirents If the trylock-based parent pointer check fails, retain those dirents and check them at the end. This may involve dropping the locks on the file being scanned, so yay. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/dir.c | 234 ++++++++++++++++++++++++++++++++++++++++++++++++- fs/xfs/scrub/readdir.c | 78 +++++++++++++++++ fs/xfs/scrub/readdir.h | 3 + fs/xfs/scrub/trace.h | 34 +++++++ 4 files changed, 346 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index e11d73eb8935..62474d0557c4 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -24,6 +24,10 @@ #include "scrub/readdir.h" #include "scrub/health.h" #include "scrub/repair.h" +#include "scrub/trace.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" /* Set us up to scrub directories. */ int @@ -43,12 +47,37 @@ xchk_setup_directory( /* Directories */ +/* Deferred directory entry that we saved for later. */ +struct xchk_dirent { + /* Cookie for retrieval of the dirent name. */ + xfblob_cookie name_cookie; + + /* Child inode number. */ + xfs_ino_t ino; + + /* Length of the pptr name. */ + uint8_t namelen; +}; + struct xchk_dir { struct xfs_scrub *sc; /* information for parent pointer validation. */ struct xfs_parent_rec pptr_rec; struct xfs_da_args pptr_args; + + /* Fixed-size array of xchk_dirent structures. */ + struct xfarray *dir_entries; + + /* Blobs containing dirent names. */ + struct xfblob *dir_names; + + /* If we've cycled the ILOCK, we must revalidate deferred dirents. */ + bool need_revalidate; + + /* Name buffer for dirent revalidation. */ + struct xfs_name xname; + uint8_t namebuf[MAXNAMELEN]; }; /* Scrub a directory entry. */ @@ -148,8 +177,26 @@ xchk_dir_check_pptr_fast( /* Try to lock the inode. */ lockmode = xchk_dir_lock_child(sc, ip); if (!lockmode) { - xchk_set_incomplete(sc); - return -ECANCELED; + struct xchk_dirent save_de = { + .namelen = name->len, + .ino = ip->i_ino, + }; + + /* Couldn't lock the inode, so save the dirent for later. */ + trace_xchk_dir_defer(sc->ip, name, ip->i_ino); + + error = xfblob_storename(sd->dir_names, &save_de.name_cookie, + name); + if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, + &error)) + return error; + + error = xfarray_append(sd->dir_entries, &save_de); + if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, + &error)) + return error; + + return 0; } error = xchk_dir_parent_pointer(sd, name, ip); @@ -865,6 +912,142 @@ out: return error; } +/* + * Revalidate a dirent that we collected in the past but couldn't check because + * of lock contention. Returns 0 if the dirent is still valid, -ENOENT if it + * has gone away on us, or a negative errno. + */ +STATIC int +xchk_dir_revalidate_dirent( + struct xchk_dir *sd, + const struct xfs_name *xname, + xfs_ino_t ino) +{ + struct xfs_scrub *sc = sd->sc; + xfs_ino_t child_ino; + int error; + + /* + * Look up the directory entry. If we get -ENOENT, the directory entry + * went away and there's nothing to revalidate. Return any other + * error. + */ + error = xchk_dir_lookup(sc, sc->ip, xname, &child_ino); + if (error) + return error; + + /* The inode number changed, nothing to revalidate. */ + if (ino != child_ino) + return -ENOENT; + + return 0; +} + +/* + * Check a directory entry's parent pointers the slow way, which means we cycle + * locks a bunch and put up with revalidation until we get it done. + */ +STATIC int +xchk_dir_slow_dirent( + struct xchk_dir *sd, + struct xchk_dirent *dirent, + const struct xfs_name *xname) +{ + struct xfs_scrub *sc = sd->sc; + struct xfs_inode *ip; + unsigned int lockmode; + int error; + + /* Check that the deferred dirent still exists. */ + if (sd->need_revalidate) { + error = xchk_dir_revalidate_dirent(sd, xname, dirent->ino); + if (error == -ENOENT) + return 0; + if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, + &error)) + return error; + } + + error = xchk_iget(sc, dirent->ino, &ip); + if (error == -EINVAL || error == -ENOENT) { + xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); + return 0; + } + if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error)) + return error; + + /* + * If we can grab both IOLOCK and ILOCK of the alleged child, we can + * proceed with the validation. + */ + lockmode = xchk_dir_lock_child(sc, ip); + if (lockmode) { + trace_xchk_dir_slowpath(sc->ip, xname, ip->i_ino); + goto check_pptr; + } + + /* + * We couldn't lock the child file. Drop all the locks and try to + * get them again, one at a time. + */ + xchk_iunlock(sc, sc->ilock_flags); + sd->need_revalidate = true; + + trace_xchk_dir_ultraslowpath(sc->ip, xname, ip->i_ino); + + error = xchk_dir_trylock_for_pptrs(sc, ip, &lockmode); + if (error) + goto out_rele; + + /* Revalidate, since we just cycled the locks. */ + error = xchk_dir_revalidate_dirent(sd, xname, dirent->ino); + if (error == -ENOENT) { + error = 0; + goto out_unlock; + } + if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error)) + goto out_unlock; + +check_pptr: + error = xchk_dir_parent_pointer(sd, xname, ip); +out_unlock: + xfs_iunlock(ip, lockmode); +out_rele: + xchk_irele(sc, ip); + return error; +} + +/* Check all the dirents that we deferred the first time around. */ +STATIC int +xchk_dir_finish_slow_dirents( + struct xchk_dir *sd) +{ + xfarray_idx_t array_cur; + int error; + + foreach_xfarray_idx(sd->dir_entries, array_cur) { + struct xchk_dirent dirent; + + if (sd->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + return 0; + + error = xfarray_load(sd->dir_entries, array_cur, &dirent); + if (error) + return error; + + error = xfblob_loadname(sd->dir_names, dirent.name_cookie, + &sd->xname, dirent.namelen); + if (error) + return error; + + error = xchk_dir_slow_dirent(sd, &dirent, &sd->xname); + if (error) + return error; + } + + return 0; +} + /* Scrub a whole directory. */ int xchk_directory( @@ -907,11 +1090,56 @@ xchk_directory( if (!sd) return -ENOMEM; sd->sc = sc; + sd->xname.name = sd->namebuf; + + if (xfs_has_parent(sc->mp)) { + char *descr; + + /* + * Set up some staging memory for dirents that we can't check + * due to locking contention. + */ + descr = xchk_xfile_ino_descr(sc, "slow directory entries"); + error = xfarray_create(descr, 0, sizeof(struct xchk_dirent), + &sd->dir_entries); + kfree(descr); + if (error) + goto out_sd; + + descr = xchk_xfile_ino_descr(sc, "slow directory entry names"); + error = xfblob_create(descr, &sd->dir_names); + kfree(descr); + if (error) + goto out_entries; + } /* Look up every name in this directory by hash. */ error = xchk_dir_walk(sc, sc->ip, xchk_dir_actor, sd); + if (error == -ECANCELED) + error = 0; + if (error) + goto out_names; + + if (xfs_has_parent(sc->mp)) { + error = xchk_dir_finish_slow_dirents(sd); + if (error == -ETIMEDOUT) { + /* Couldn't grab a lock, scrub was marked incomplete */ + error = 0; + goto out_names; + } + if (error) + goto out_names; + } + +out_names: + if (sd->dir_names) + xfblob_destroy(sd->dir_names); +out_entries: + if (sd->dir_entries) + xfarray_destroy(sd->dir_entries); +out_sd: kvfree(sd); - if (error && error != -ECANCELED) + if (error) return error; /* If the dir is clean, it is clearly not zapped. */ diff --git a/fs/xfs/scrub/readdir.c b/fs/xfs/scrub/readdir.c index 028690761c62..28a94c78b0b1 100644 --- a/fs/xfs/scrub/readdir.c +++ b/fs/xfs/scrub/readdir.c @@ -18,6 +18,7 @@ #include "xfs_trans.h" #include "xfs_error.h" #include "scrub/scrub.h" +#include "scrub/common.h" #include "scrub/readdir.h" /* Call a function for every entry in a shortform directory. */ @@ -380,3 +381,80 @@ out_check_rval: *ino = args.inumber; return error; } + +/* + * Try to grab the IOLOCK and ILOCK of sc->ip and ip, returning @ip's lock + * state. The caller may have a transaction, so we must use trylock for both + * IOLOCKs. + */ +static inline unsigned int +xchk_dir_trylock_both( + struct xfs_scrub *sc, + struct xfs_inode *ip) +{ + if (!xchk_ilock_nowait(sc, XFS_IOLOCK_EXCL)) + return 0; + + if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) + goto parent_iolock; + + xchk_ilock(sc, XFS_ILOCK_EXCL); + if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) + goto parent_ilock; + + return XFS_IOLOCK_SHARED | XFS_ILOCK_EXCL; + +parent_ilock: + xchk_iunlock(sc, XFS_ILOCK_EXCL); + xfs_iunlock(ip, XFS_IOLOCK_SHARED); +parent_iolock: + xchk_iunlock(sc, XFS_IOLOCK_EXCL); + return 0; +} + +/* + * Try for a limited time to grab the IOLOCK and ILOCK of both the scrub target + * (@sc->ip) and the inode at the other end (@ip) of a directory or parent + * pointer link so that we can check that link. + * + * We do not know ahead of time that the directory tree is /not/ corrupt, so we + * cannot use the "lock two inode" functions because we do not know that there + * is not a racing thread trying to take the locks in opposite order. First + * take IOLOCK_EXCL of the scrub target, and then try to take IOLOCK_SHARED + * of @ip to synchronize with the VFS. Next, take ILOCK_EXCL of the scrub + * target and @ip to synchronize with XFS. + * + * If the trylocks succeed, *lockmode will be set to the locks held for @ip; + * @sc->ilock_flags will be set for the locks held for @sc->ip; and zero will + * be returned. If not, returns -EDEADLOCK to try again; or -ETIMEDOUT if + * XCHK_TRY_HARDER was set. Returns -EINTR if the process has been killed. + */ +int +xchk_dir_trylock_for_pptrs( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int *lockmode) +{ + unsigned int nr; + int error = 0; + + ASSERT(sc->ilock_flags == 0); + + for (nr = 0; nr < HZ; nr++) { + *lockmode = xchk_dir_trylock_both(sc, ip); + if (*lockmode) + return 0; + + if (xchk_should_terminate(sc, &error)) + return error; + + delay(1); + } + + if (sc->flags & XCHK_TRY_HARDER) { + xchk_set_incomplete(sc); + return -ETIMEDOUT; + } + + return -EDEADLOCK; +} diff --git a/fs/xfs/scrub/readdir.h b/fs/xfs/scrub/readdir.h index 55787f4df123..da501877a64d 100644 --- a/fs/xfs/scrub/readdir.h +++ b/fs/xfs/scrub/readdir.h @@ -16,4 +16,7 @@ int xchk_dir_walk(struct xfs_scrub *sc, struct xfs_inode *dp, int xchk_dir_lookup(struct xfs_scrub *sc, struct xfs_inode *dp, const struct xfs_name *name, xfs_ino_t *ino); +int xchk_dir_trylock_for_pptrs(struct xfs_scrub *sc, struct xfs_inode *ip, + unsigned int *lockmode); + #endif /* __XFS_SCRUB_READDIR_H__ */ diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 814db1d1747a..4db762480b8d 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -1511,6 +1511,40 @@ DEFINE_EVENT(xchk_nlinks_diff_class, name, \ TP_ARGS(mp, ip, live)) DEFINE_SCRUB_NLINKS_DIFF_EVENT(xchk_nlinks_compare_inode); +DECLARE_EVENT_CLASS(xchk_pptr_class, + TP_PROTO(struct xfs_inode *ip, const struct xfs_name *name, + xfs_ino_t far_ino), + TP_ARGS(ip, name, far_ino), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + __field(xfs_ino_t, far_ino) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->namelen = name->len; + memcpy(__get_str(name), name, name->len); + __entry->far_ino = far_ino; + ), + TP_printk("dev %d:%d ino 0x%llx name '%.*s' far_ino 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->namelen, + __get_str(name), + __entry->far_ino) +) +#define DEFINE_XCHK_PPTR_EVENT(name) \ +DEFINE_EVENT(xchk_pptr_class, name, \ + TP_PROTO(struct xfs_inode *ip, const struct xfs_name *name, \ + xfs_ino_t far_ino), \ + TP_ARGS(ip, name, far_ino)) +DEFINE_XCHK_PPTR_EVENT(xchk_dir_defer); +DEFINE_XCHK_PPTR_EVENT(xchk_dir_slowpath); +DEFINE_XCHK_PPTR_EVENT(xchk_dir_ultraslowpath); + /* repair tracepoints */ #if IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) -- cgit v1.2.3-59-g8ed1b From 0d29a20fbdba89501bd2ac003faeee666c9a5008 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:03 -0700 Subject: xfs: scrub parent pointers Actually check parent pointers now. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/parent.c | 371 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 371 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index acb6282c3d14..6ebbb7104126 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -15,11 +15,15 @@ #include "xfs_icache.h" #include "xfs_dir2.h" #include "xfs_dir2_priv.h" +#include "xfs_attr.h" +#include "xfs_parent.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/readdir.h" #include "scrub/tempfile.h" #include "scrub/repair.h" +#include "scrub/listxattr.h" +#include "scrub/trace.h" /* Set us up to scrub parents. */ int @@ -197,6 +201,370 @@ out_rele: return error; } +/* + * Checking of Parent Pointers + * =========================== + * + * On filesystems with directory parent pointers, we check the referential + * integrity by visiting each parent pointer of a child file and checking that + * the directory referenced by the pointer actually has a dirent pointing + * forward to the child file. + */ + +struct xchk_pptrs { + struct xfs_scrub *sc; + + /* How many parent pointers did we find at the end? */ + unsigned long long pptrs_found; + + /* Parent of this directory. */ + xfs_ino_t parent_ino; +}; + +/* Does this parent pointer match the dotdot entry? */ +STATIC int +xchk_parent_scan_dotdot( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xchk_pptrs *pp = priv; + xfs_ino_t parent_ino; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, &parent_ino, NULL); + if (error) + return error; + + if (pp->parent_ino == parent_ino) + return -ECANCELED; + + return 0; +} + +/* Look up the dotdot entry so that we can check it as we walk the pptrs. */ +STATIC int +xchk_parent_pptr_and_dotdot( + struct xchk_pptrs *pp) +{ + struct xfs_scrub *sc = pp->sc; + int error; + + /* Look up '..' */ + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &pp->parent_ino); + if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, 0, &error)) + return error; + if (!xfs_verify_dir_ino(sc->mp, pp->parent_ino)) { + xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); + return 0; + } + + /* Is this the root dir? Then '..' must point to itself. */ + if (sc->ip == sc->mp->m_rootip) { + if (sc->ip->i_ino != pp->parent_ino) + xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); + return 0; + } + + /* + * If this is now an unlinked directory, the dotdot value is + * meaningless as long as it points to a valid inode. + */ + if (VFS_I(sc->ip)->i_nlink == 0) + return 0; + + if (pp->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + return 0; + + /* Otherwise, walk the pptrs again, and check. */ + error = xchk_xattr_walk(sc, sc->ip, xchk_parent_scan_dotdot, pp); + if (error == -ECANCELED) { + /* Found a parent pointer that matches dotdot. */ + return 0; + } + if (!error || error == -EFSCORRUPTED) { + /* Found a broken parent pointer or no match. */ + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); + return 0; + } + return error; +} + +/* + * Try to lock a parent directory for checking dirents. Returns the inode + * flags for the locks we now hold, or zero if we failed. + */ +STATIC unsigned int +xchk_parent_lock_dir( + struct xfs_scrub *sc, + struct xfs_inode *dp) +{ + if (!xfs_ilock_nowait(dp, XFS_IOLOCK_SHARED)) + return 0; + + if (!xfs_ilock_nowait(dp, XFS_ILOCK_SHARED)) { + xfs_iunlock(dp, XFS_IOLOCK_SHARED); + return 0; + } + + if (!xfs_need_iread_extents(&dp->i_df)) + return XFS_IOLOCK_SHARED | XFS_ILOCK_SHARED; + + xfs_iunlock(dp, XFS_ILOCK_SHARED); + + if (!xfs_ilock_nowait(dp, XFS_ILOCK_EXCL)) { + xfs_iunlock(dp, XFS_IOLOCK_SHARED); + return 0; + } + + return XFS_IOLOCK_SHARED | XFS_ILOCK_EXCL; +} + +/* Check the forward link (dirent) associated with this parent pointer. */ +STATIC int +xchk_parent_dirent( + struct xchk_pptrs *pp, + const struct xfs_name *xname, + struct xfs_inode *dp) +{ + struct xfs_scrub *sc = pp->sc; + xfs_ino_t child_ino; + int error; + + /* + * Use the name attached to this parent pointer to look up the + * directory entry in the alleged parent. + */ + error = xchk_dir_lookup(sc, dp, xname, &child_ino); + if (error == -ENOENT) { + xchk_fblock_xref_set_corrupt(sc, XFS_ATTR_FORK, 0); + return 0; + } + if (!xchk_fblock_xref_process_error(sc, XFS_ATTR_FORK, 0, &error)) + return error; + + /* Does the inode number match? */ + if (child_ino != sc->ip->i_ino) { + xchk_fblock_xref_set_corrupt(sc, XFS_ATTR_FORK, 0); + return 0; + } + + return 0; +} + +/* Try to grab a parent directory. */ +STATIC int +xchk_parent_iget( + struct xchk_pptrs *pp, + const struct xfs_parent_rec *pptr, + struct xfs_inode **dpp) +{ + struct xfs_scrub *sc = pp->sc; + struct xfs_inode *ip; + xfs_ino_t parent_ino = be64_to_cpu(pptr->p_ino); + int error; + + /* Validate inode number. */ + error = xfs_dir_ino_validate(sc->mp, parent_ino); + if (error) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); + return -ECANCELED; + } + + error = xchk_iget(sc, parent_ino, &ip); + if (error == -EINVAL || error == -ENOENT) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); + return -ECANCELED; + } + if (!xchk_fblock_xref_process_error(sc, XFS_ATTR_FORK, 0, &error)) + return error; + + /* The parent must be a directory. */ + if (!S_ISDIR(VFS_I(ip)->i_mode)) { + xchk_fblock_xref_set_corrupt(sc, XFS_ATTR_FORK, 0); + goto out_rele; + } + + /* Validate generation number. */ + if (VFS_I(ip)->i_generation != be32_to_cpu(pptr->p_gen)) { + xchk_fblock_xref_set_corrupt(sc, XFS_ATTR_FORK, 0); + goto out_rele; + } + + *dpp = ip; + return 0; +out_rele: + xchk_irele(sc, ip); + return 0; +} + +/* + * Walk an xattr of a file. If this xattr is a parent pointer, follow it up + * to a parent directory and check that the parent has a dirent pointing back + * to us. + */ +STATIC int +xchk_parent_scan_attr( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xfs_name xname = { + .name = name, + .len = namelen, + }; + struct xchk_pptrs *pp = priv; + struct xfs_inode *dp = NULL; + const struct xfs_parent_rec *pptr_rec = value; + xfs_ino_t parent_ino; + unsigned int lockmode; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, &parent_ino, NULL); + if (error) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); + return error; + } + + /* No self-referential parent pointers. */ + if (parent_ino == sc->ip->i_ino) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); + return -ECANCELED; + } + + pp->pptrs_found++; + + error = xchk_parent_iget(pp, pptr_rec, &dp); + if (error) + return error; + if (!dp) + return 0; + + /* Try to lock the inode. */ + lockmode = xchk_parent_lock_dir(sc, dp); + if (!lockmode) { + xchk_set_incomplete(sc); + error = -ECANCELED; + goto out_rele; + } + + error = xchk_parent_dirent(pp, &xname, dp); + if (error) + goto out_unlock; + +out_unlock: + xfs_iunlock(dp, lockmode); +out_rele: + xchk_irele(sc, dp); + return error; +} + +/* + * Compare the number of parent pointers to the link count. For + * non-directories these should be the same. For unlinked directories the + * count should be zero; for linked directories, it should be nonzero. + */ +STATIC int +xchk_parent_count_pptrs( + struct xchk_pptrs *pp) +{ + struct xfs_scrub *sc = pp->sc; + + if (S_ISDIR(VFS_I(sc->ip)->i_mode)) { + if (sc->ip == sc->mp->m_rootip) + pp->pptrs_found++; + + if (VFS_I(sc->ip)->i_nlink == 0 && pp->pptrs_found > 0) + xchk_ino_set_corrupt(sc, sc->ip->i_ino); + else if (VFS_I(sc->ip)->i_nlink > 0 && + pp->pptrs_found == 0) + xchk_ino_set_corrupt(sc, sc->ip->i_ino); + } else { + if (VFS_I(sc->ip)->i_nlink != pp->pptrs_found) + xchk_ino_set_corrupt(sc, sc->ip->i_ino); + } + + return 0; +} + +/* Check parent pointers of a file. */ +STATIC int +xchk_parent_pptr( + struct xfs_scrub *sc) +{ + struct xchk_pptrs *pp; + int error; + + pp = kvzalloc(sizeof(struct xchk_pptrs), XCHK_GFP_FLAGS); + if (!pp) + return -ENOMEM; + pp->sc = sc; + + error = xchk_xattr_walk(sc, sc->ip, xchk_parent_scan_attr, pp); + if (error == -ECANCELED) { + error = 0; + goto out_pp; + } + if (error) + goto out_pp; + + if (pp->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out_pp; + + /* + * For subdirectories, make sure the dotdot entry references the same + * inode as the parent pointers. + * + * If we're scanning a /consistent/ directory, there should only be + * one parent pointer, and it should point to the same directory as + * the dotdot entry. + * + * However, a corrupt directory tree might feature a subdirectory with + * multiple parents. The directory loop scanner is responsible for + * correcting that kind of problem, so for now we only validate that + * the dotdot entry matches /one/ of the parents. + */ + if (S_ISDIR(VFS_I(sc->ip)->i_mode)) { + error = xchk_parent_pptr_and_dotdot(pp); + if (error) + goto out_pp; + } + + if (pp->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out_pp; + + /* + * Complain if the number of parent pointers doesn't match the link + * count. This could be a sign of missing parent pointers (or an + * incorrect link count). + */ + error = xchk_parent_count_pptrs(pp); + if (error) + goto out_pp; + +out_pp: + kvfree(pp); + return error; +} + /* Scrub a parent pointer. */ int xchk_parent( @@ -206,6 +574,9 @@ xchk_parent( xfs_ino_t parent_ino; int error = 0; + if (xfs_has_parent(mp)) + return xchk_parent_pptr(sc); + /* * If we're a directory, check that the '..' link points up to * a directory that has one entry pointing to us. -- cgit v1.2.3-59-g8ed1b From 8ad345306d1eba337a3b88f9acbe762f9e770f76 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:04 -0700 Subject: xfs: deferred scrub of parent pointers If the trylock-based dirent check fails, retain those parent pointers and check them at the end. This may involve dropping the locks on the file being scanned, so yay. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 2 +- fs/xfs/scrub/parent.c | 267 ++++++++++++++++++++++++++++++++++++++++++++++++-- fs/xfs/scrub/trace.h | 3 + 3 files changed, 264 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index c969b11ce0f4..af99a455ce4d 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -177,6 +177,7 @@ xfs-y += $(addprefix scrub/, \ scrub.o \ symlink.o \ xfarray.o \ + xfblob.o \ xfile.o \ ) @@ -218,7 +219,6 @@ xfs-y += $(addprefix scrub/, \ rmap_repair.o \ symlink_repair.o \ tempfile.o \ - xfblob.o \ ) xfs-$(CONFIG_XFS_RT) += $(addprefix scrub/, \ diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index 6ebbb7104126..f14e6f643fb6 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -23,6 +23,9 @@ #include "scrub/tempfile.h" #include "scrub/repair.h" #include "scrub/listxattr.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" #include "scrub/trace.h" /* Set us up to scrub parents. */ @@ -211,6 +214,18 @@ out_rele: * forward to the child file. */ +/* Deferred parent pointer entry that we saved for later. */ +struct xchk_pptr { + /* Cookie for retrieval of the pptr name. */ + xfblob_cookie name_cookie; + + /* Parent pointer record. */ + struct xfs_parent_rec pptr_rec; + + /* Length of the pptr name. */ + uint8_t namelen; +}; + struct xchk_pptrs { struct xfs_scrub *sc; @@ -219,6 +234,22 @@ struct xchk_pptrs { /* Parent of this directory. */ xfs_ino_t parent_ino; + + /* Fixed-size array of xchk_pptr structures. */ + struct xfarray *pptr_entries; + + /* Blobs containing parent pointer names. */ + struct xfblob *pptr_names; + + /* Scratch buffer for scanning pptr xattrs */ + struct xfs_da_args pptr_args; + + /* If we've cycled the ILOCK, we must revalidate all deferred pptrs. */ + bool need_revalidate; + + /* Name buffer */ + struct xfs_name xname; + char namebuf[MAXNAMELEN]; }; /* Does this parent pointer match the dotdot entry? */ @@ -461,8 +492,25 @@ xchk_parent_scan_attr( /* Try to lock the inode. */ lockmode = xchk_parent_lock_dir(sc, dp); if (!lockmode) { - xchk_set_incomplete(sc); - error = -ECANCELED; + struct xchk_pptr save_pp = { + .pptr_rec = *pptr_rec, /* struct copy */ + .namelen = namelen, + }; + + /* Couldn't lock the inode, so save the pptr for later. */ + trace_xchk_parent_defer(sc->ip, &xname, dp->i_ino); + + error = xfblob_storename(pp->pptr_names, &save_pp.name_cookie, + &xname); + if (!xchk_fblock_xref_process_error(sc, XFS_ATTR_FORK, 0, + &error)) + goto out_rele; + + error = xfarray_append(pp->pptr_entries, &save_pp); + if (!xchk_fblock_xref_process_error(sc, XFS_ATTR_FORK, 0, + &error)) + goto out_rele; + goto out_rele; } @@ -477,6 +525,162 @@ out_rele: return error; } +/* + * Revalidate a parent pointer that we collected in the past but couldn't check + * because of lock contention. Returns 0 if the parent pointer is still valid, + * -ENOENT if it has gone away on us, or a negative errno. + */ +STATIC int +xchk_parent_revalidate_pptr( + struct xchk_pptrs *pp, + const struct xfs_name *xname, + struct xfs_parent_rec *pptr) +{ + struct xfs_scrub *sc = pp->sc; + int error; + + error = xfs_parent_lookup(sc->tp, sc->ip, xname, pptr, &pp->pptr_args); + if (error == -ENOATTR) { + /* Parent pointer went away, nothing to revalidate. */ + return -ENOENT; + } + + return error; +} + +/* + * Check a parent pointer the slow way, which means we cycle locks a bunch + * and put up with revalidation until we get it done. + */ +STATIC int +xchk_parent_slow_pptr( + struct xchk_pptrs *pp, + const struct xfs_name *xname, + struct xfs_parent_rec *pptr) +{ + struct xfs_scrub *sc = pp->sc; + struct xfs_inode *dp = NULL; + unsigned int lockmode; + int error; + + /* Check that the deferred parent pointer still exists. */ + if (pp->need_revalidate) { + error = xchk_parent_revalidate_pptr(pp, xname, pptr); + if (error == -ENOENT) + return 0; + if (!xchk_fblock_xref_process_error(sc, XFS_ATTR_FORK, 0, + &error)) + return error; + } + + error = xchk_parent_iget(pp, pptr, &dp); + if (error) + return error; + if (!dp) + return 0; + + /* + * If we can grab both IOLOCK and ILOCK of the alleged parent, we + * can proceed with the validation. + */ + lockmode = xchk_parent_lock_dir(sc, dp); + if (lockmode) { + trace_xchk_parent_slowpath(sc->ip, xname, dp->i_ino); + goto check_dirent; + } + + /* + * We couldn't lock the parent dir. Drop all the locks and try to + * get them again, one at a time. + */ + xchk_iunlock(sc, sc->ilock_flags); + pp->need_revalidate = true; + + trace_xchk_parent_ultraslowpath(sc->ip, xname, dp->i_ino); + + error = xchk_dir_trylock_for_pptrs(sc, dp, &lockmode); + if (error) + goto out_rele; + + /* Revalidate the parent pointer now that we cycled locks. */ + error = xchk_parent_revalidate_pptr(pp, xname, pptr); + if (error == -ENOENT) { + error = 0; + goto out_unlock; + } + if (!xchk_fblock_xref_process_error(sc, XFS_ATTR_FORK, 0, &error)) + goto out_unlock; + +check_dirent: + error = xchk_parent_dirent(pp, xname, dp); +out_unlock: + xfs_iunlock(dp, lockmode); +out_rele: + xchk_irele(sc, dp); + return error; +} + +/* Check all the parent pointers that we deferred the first time around. */ +STATIC int +xchk_parent_finish_slow_pptrs( + struct xchk_pptrs *pp) +{ + xfarray_idx_t array_cur; + int error; + + foreach_xfarray_idx(pp->pptr_entries, array_cur) { + struct xchk_pptr pptr; + + if (pp->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + return 0; + + error = xfarray_load(pp->pptr_entries, array_cur, &pptr); + if (error) + return error; + + error = xfblob_loadname(pp->pptr_names, pptr.name_cookie, + &pp->xname, pptr.namelen); + if (error) + return error; + + error = xchk_parent_slow_pptr(pp, &pp->xname, &pptr.pptr_rec); + if (error) + return error; + } + + /* Empty out both xfiles now that we've checked everything. */ + xfarray_truncate(pp->pptr_entries); + xfblob_truncate(pp->pptr_names); + return 0; +} + +/* Count the number of parent pointers. */ +STATIC int +xchk_parent_count_pptr( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xchk_pptrs *pp = priv; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, NULL, NULL); + if (error) + return error; + + pp->pptrs_found++; + return 0; +} + /* * Compare the number of parent pointers to the link count. For * non-directories these should be the same. For unlinked directories the @@ -487,6 +691,23 @@ xchk_parent_count_pptrs( struct xchk_pptrs *pp) { struct xfs_scrub *sc = pp->sc; + int error; + + /* + * If we cycled the ILOCK while cross-checking parent pointers with + * dirents, then we need to recalculate the number of parent pointers. + */ + if (pp->need_revalidate) { + pp->pptrs_found = 0; + error = xchk_xattr_walk(sc, sc->ip, xchk_parent_count_pptr, pp); + if (error == -EFSCORRUPTED) { + /* Found a bad parent pointer */ + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); + return 0; + } + if (error) + return error; + } if (S_ISDIR(VFS_I(sc->ip)->i_mode)) { if (sc->ip == sc->mp->m_rootip) @@ -511,23 +732,51 @@ xchk_parent_pptr( struct xfs_scrub *sc) { struct xchk_pptrs *pp; + char *descr; int error; pp = kvzalloc(sizeof(struct xchk_pptrs), XCHK_GFP_FLAGS); if (!pp) return -ENOMEM; pp->sc = sc; + pp->xname.name = pp->namebuf; + + /* + * Set up some staging memory for parent pointers that we can't check + * due to locking contention. + */ + descr = xchk_xfile_ino_descr(sc, "slow parent pointer entries"); + error = xfarray_create(descr, 0, sizeof(struct xchk_pptr), + &pp->pptr_entries); + kfree(descr); + if (error) + goto out_pp; + + descr = xchk_xfile_ino_descr(sc, "slow parent pointer names"); + error = xfblob_create(descr, &pp->pptr_names); + kfree(descr); + if (error) + goto out_entries; error = xchk_xattr_walk(sc, sc->ip, xchk_parent_scan_attr, pp); if (error == -ECANCELED) { error = 0; - goto out_pp; + goto out_names; } if (error) - goto out_pp; + goto out_names; + + error = xchk_parent_finish_slow_pptrs(pp); + if (error == -ETIMEDOUT) { + /* Couldn't grab a lock, scrub was marked incomplete */ + error = 0; + goto out_names; + } + if (error) + goto out_names; if (pp->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) - goto out_pp; + goto out_names; /* * For subdirectories, make sure the dotdot entry references the same @@ -545,7 +794,7 @@ xchk_parent_pptr( if (S_ISDIR(VFS_I(sc->ip)->i_mode)) { error = xchk_parent_pptr_and_dotdot(pp); if (error) - goto out_pp; + goto out_names; } if (pp->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) @@ -558,8 +807,12 @@ xchk_parent_pptr( */ error = xchk_parent_count_pptrs(pp); if (error) - goto out_pp; + goto out_names; +out_names: + xfblob_destroy(pp->pptr_names); +out_entries: + xfarray_destroy(pp->pptr_entries); out_pp: kvfree(pp); return error; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 4db762480b8d..97a106519b53 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -1544,6 +1544,9 @@ DEFINE_EVENT(xchk_pptr_class, name, \ DEFINE_XCHK_PPTR_EVENT(xchk_dir_defer); DEFINE_XCHK_PPTR_EVENT(xchk_dir_slowpath); DEFINE_XCHK_PPTR_EVENT(xchk_dir_ultraslowpath); +DEFINE_XCHK_PPTR_EVENT(xchk_parent_defer); +DEFINE_XCHK_PPTR_EVENT(xchk_parent_slowpath); +DEFINE_XCHK_PPTR_EVENT(xchk_parent_ultraslowpath); /* repair tracepoints */ #if IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) -- cgit v1.2.3-59-g8ed1b From 77ede5f44b0d86c2ec812442846f512884009766 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:05 -0700 Subject: xfs: walk directory parent pointers to determine backref count If the filesystem has parent pointers enabled, walk the parent pointers of subdirectories to determine the true backref count. In theory each subdir should have a single parent reachable via dotdot, but in the case of (corrupt) subdirs with multiple parents, we need to keep the link counts high enough that the directory loop detector will be able to correct the multiple parents problems. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/common.h | 1 + fs/xfs/scrub/nlinks.c | 85 +++++++++++++++++++++++++++++++++++++++++++- fs/xfs/scrub/nlinks_repair.c | 2 ++ fs/xfs/scrub/parent.c | 61 +++++++++++++++++++++++++++++++ fs/xfs/scrub/trace.c | 1 + fs/xfs/scrub/trace.h | 28 +++++++++++++++ 6 files changed, 177 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h index 89f7bbec887e..e00466f40482 100644 --- a/fs/xfs/scrub/common.h +++ b/fs/xfs/scrub/common.h @@ -212,6 +212,7 @@ static inline bool xchk_skip_xref(struct xfs_scrub_metadata *sm) } bool xchk_dir_looks_zapped(struct xfs_inode *dp); +bool xchk_pptr_looks_zapped(struct xfs_inode *ip); #ifdef CONFIG_XFS_ONLINE_REPAIR /* Decide if a repair is required. */ diff --git a/fs/xfs/scrub/nlinks.c b/fs/xfs/scrub/nlinks.c index fcb9c473f372..d27b32e6f33d 100644 --- a/fs/xfs/scrub/nlinks.c +++ b/fs/xfs/scrub/nlinks.c @@ -18,6 +18,7 @@ #include "xfs_dir2.h" #include "xfs_dir2_priv.h" #include "xfs_ag.h" +#include "xfs_parent.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/repair.h" @@ -29,6 +30,7 @@ #include "scrub/trace.h" #include "scrub/readdir.h" #include "scrub/tempfile.h" +#include "scrub/listxattr.h" /* * Live Inode Link Count Checking @@ -272,12 +274,17 @@ xchk_nlinks_collect_dirent( * number of parents of the root directory. * * Otherwise, increment the number of backrefs pointing back to ino. + * + * If the filesystem has parent pointers, we walk the pptrs to + * determine the backref count. */ if (dotdot) { if (dp == sc->mp->m_rootip) error = xchk_nlinks_update_incore(xnc, ino, 1, 0, 0); - else + else if (!xfs_has_parent(sc->mp)) error = xchk_nlinks_update_incore(xnc, ino, 0, 1, 0); + else + error = 0; if (error) goto out_unlock; } @@ -314,6 +321,61 @@ out_incomplete: return error; } +/* Bump the backref count for the inode referenced by this parent pointer. */ +STATIC int +xchk_nlinks_collect_pptr( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xfs_name xname = { + .name = name, + .len = namelen, + }; + struct xchk_nlink_ctrs *xnc = priv; + const struct xfs_parent_rec *pptr_rec = value; + xfs_ino_t parent_ino; + int error; + + /* Update the shadow link counts if we haven't already failed. */ + + if (xchk_iscan_aborted(&xnc->collect_iscan)) { + error = -ECANCELED; + goto out_incomplete; + } + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, &parent_ino, NULL); + if (error) + return error; + + trace_xchk_nlinks_collect_pptr(sc->mp, ip, &xname, pptr_rec); + + mutex_lock(&xnc->lock); + + error = xchk_nlinks_update_incore(xnc, parent_ino, 0, 1, 0); + if (error) + goto out_unlock; + + mutex_unlock(&xnc->lock); + return 0; + +out_unlock: + mutex_unlock(&xnc->lock); + xchk_iscan_abort(&xnc->collect_iscan); +out_incomplete: + xchk_set_incomplete(sc); + return error; +} + /* Walk a directory to bump the observed link counts of the children. */ STATIC int xchk_nlinks_collect_dir( @@ -360,6 +422,27 @@ xchk_nlinks_collect_dir( if (error) goto out_abort; + /* Walk the parent pointers to get real backref counts. */ + if (xfs_has_parent(sc->mp)) { + /* + * If the extended attributes look as though they has been + * zapped by the inode record repair code, we cannot scan for + * parent pointers. + */ + if (xchk_pptr_looks_zapped(dp)) { + error = -EBUSY; + goto out_unlock; + } + + error = xchk_xattr_walk(sc, dp, xchk_nlinks_collect_pptr, xnc); + if (error == -ECANCELED) { + error = 0; + goto out_unlock; + } + if (error) + goto out_abort; + } + xchk_iscan_mark_visited(&xnc->collect_iscan, dp); goto out_unlock; diff --git a/fs/xfs/scrub/nlinks_repair.c b/fs/xfs/scrub/nlinks_repair.c index 83f8637bb08f..78d0f650fe89 100644 --- a/fs/xfs/scrub/nlinks_repair.c +++ b/fs/xfs/scrub/nlinks_repair.c @@ -18,6 +18,8 @@ #include "xfs_ialloc.h" #include "xfs_sb.h" #include "xfs_ag.h" +#include "xfs_dir2.h" +#include "xfs_parent.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/repair.h" diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index f14e6f643fb6..068691434be1 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -873,3 +873,64 @@ xchk_parent( return error; } + +/* + * Decide if this file's extended attributes (and therefore its parent + * pointers) have been zapped to satisfy the inode and ifork verifiers. + * Checking and repairing should be postponed until the extended attribute + * structure is fixed. + */ +bool +xchk_pptr_looks_zapped( + struct xfs_inode *ip) +{ + struct xfs_mount *mp = ip->i_mount; + struct inode *inode = VFS_I(ip); + + ASSERT(xfs_has_parent(mp)); + + /* + * Temporary files that cannot be linked into the directory tree do not + * have attr forks because they cannot ever have parents. + */ + if (inode->i_nlink == 0 && !(inode->i_state & I_LINKABLE)) + return false; + + /* + * Directory tree roots do not have parents, so the expected outcome + * of a parent pointer scan is always the empty set. It's safe to scan + * them even if the attr fork was zapped. + */ + if (ip == mp->m_rootip) + return false; + + /* + * Metadata inodes are all rooted in the superblock and do not have + * any parents. Hence the attr fork will not be initialized, but + * there are no parent pointers that might have been zapped. + */ + if (xfs_is_metadata_inode(ip)) + return false; + + /* + * Linked and linkable non-rootdir files should always have an + * attribute fork because that is where parent pointers are + * stored. If the fork is absent, something is amiss. + */ + if (!xfs_inode_has_attr_fork(ip)) + return true; + + /* Repair zapped this file's attr fork a short time ago */ + if (xfs_ifork_zapped(ip, XFS_ATTR_FORK)) + return true; + + /* + * If the dinode repair found a bad attr fork, it will reset the fork + * to extents format with zero records and wait for the bmapbta + * scrubber to reconstruct the block mappings. The extended attribute + * structure always contain some content when parent pointers are + * enabled, so this is a clear sign of a zapped attr fork. + */ + return ip->i_af.if_format == XFS_DINODE_FMT_EXTENTS && + ip->i_af.if_nextents == 0; +} diff --git a/fs/xfs/scrub/trace.c b/fs/xfs/scrub/trace.c index b2ce7b22cad3..4a8cc2c98d99 100644 --- a/fs/xfs/scrub/trace.c +++ b/fs/xfs/scrub/trace.c @@ -19,6 +19,7 @@ #include "xfs_da_format.h" #include "xfs_dir2.h" #include "xfs_rmap.h" +#include "xfs_parent.h" #include "scrub/scrub.h" #include "scrub/xfile.h" #include "scrub/xfarray.h" diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 97a106519b53..3e726610b9e3 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -26,6 +26,7 @@ struct xchk_iscan; struct xchk_nlink; struct xchk_fscounters; struct xfs_rmap_update_params; +struct xfs_parent_rec; /* * ftrace's __print_symbolic requires that all enum values be wrapped in the @@ -1363,6 +1364,33 @@ TRACE_EVENT(xchk_nlinks_collect_dirent, __get_str(name)) ); +TRACE_EVENT(xchk_nlinks_collect_pptr, + TP_PROTO(struct xfs_mount *mp, struct xfs_inode *dp, + const struct xfs_name *name, + const struct xfs_parent_rec *pptr), + TP_ARGS(mp, dp, name, pptr), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, dir) + __field(xfs_ino_t, ino) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->dir = dp->i_ino; + __entry->ino = be64_to_cpu(pptr->p_ino); + __entry->namelen = name->len; + memcpy(__get_str(name), name->name, name->len); + ), + TP_printk("dev %d:%d dir 0x%llx -> ino 0x%llx name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->dir, + __entry->ino, + __entry->namelen, + __get_str(name)) +); + TRACE_EVENT(xchk_nlinks_collect_metafile, TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino), TP_ARGS(mp, ino), -- cgit v1.2.3-59-g8ed1b From 59a2af9086f0d60fc8de7346da67db7d764c7221 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:05 -0700 Subject: xfs: check parent pointer xattrs when scrubbing Check parent pointer xattrs as part of scrubbing xattrs. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 393ed36709b3..b550f3e34ffc 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -17,6 +17,7 @@ #include "xfs_attr.h" #include "xfs_attr_leaf.h" #include "xfs_attr_sf.h" +#include "xfs_parent.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/dabtree.h" @@ -208,6 +209,13 @@ xchk_xattr_actor( return -ECANCELED; } + /* Check parent pointer record. */ + if ((attr_flags & XFS_ATTR_PARENT) && + !xfs_parent_valuecheck(sc->mp, value, valuelen)) { + xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, args.blkno); + return -ECANCELED; + } + /* * Try to allocate enough memory to extract the attr value. If that * doesn't work, return -EDEADLOCK as a signal to try again with a @@ -219,6 +227,14 @@ xchk_xattr_actor( if (error) return error; + /* + * Parent pointers are matched on attr name and value, so we must + * supply the xfs_parent_rec here when confirming that the dabtree + * indexing works correctly. + */ + if (attr_flags & XFS_ATTR_PARENT) + memcpy(ab->value, value, valuelen); + args.value = ab->value; /* -- cgit v1.2.3-59-g8ed1b From e7420e75ef04787bc51688fc9bbca7da4d164a1e Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:06 -0700 Subject: xfs: remove some boilerplate from xfs_attr_set In preparation for online/offline repair wanting to use xfs_attr_set, move some of the boilerplate out of this function into the callers. Repair can initialize the da_args completely, and the userspace flag handling/twisting goes away once we move it to xfs_attr_change. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 33 ++++++++++++--------------------- fs/xfs/scrub/attr_repair.c | 4 ++++ fs/xfs/xfs_xattr.c | 24 ++++++++++++++++++++++-- 3 files changed, 38 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 8c283e5c2470..df8418671c37 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -948,6 +948,16 @@ xfs_attr_lookup( return error; } +/* + * Make a change to the xattr structure. + * + * The caller must have initialized @args, attached dquots, and must not hold + * any ILOCKs. + * + * Returns -EEXIST for XFS_ATTRUPDATE_CREATE if the name already exists. + * Returns -ENOATTR for XFS_ATTRUPDATE_REMOVE if the name does not exist. + * Returns 0 on success, or a negative errno if something else went wrong. + */ int xfs_attr_set( struct xfs_da_args *args, @@ -961,27 +971,7 @@ xfs_attr_set( int rmt_blks = 0; unsigned int total; - if (xfs_is_shutdown(dp->i_mount)) - return -EIO; - - error = xfs_qm_dqattach(dp); - if (error) - return error; - - if (!args->owner) - args->owner = args->dp->i_ino; - args->geo = mp->m_attr_geo; - args->whichfork = XFS_ATTR_FORK; - xfs_attr_sethash(args); - - /* - * We have no control over the attribute names that userspace passes us - * to remove, so we have to allow the name lookup prior to attribute - * removal to fail as well. Preserve the logged flag, since we need - * to pass that through to the logging code. - */ - args->op_flags = XFS_DA_OP_OKNOENT | - (args->op_flags & XFS_DA_OP_LOGGED); + ASSERT(!args->trans); switch (op) { case XFS_ATTRUPDATE_UPSERT: @@ -1076,6 +1066,7 @@ xfs_attr_set( error = xfs_trans_commit(args->trans); out_unlock: xfs_iunlock(dp, XFS_ILOCK_EXCL); + args->trans = NULL; return error; out_trans_cancel: diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index 8b89c112c492..67c0ec0d1dbb 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -558,6 +558,9 @@ xrep_xattr_insert_rec( .namelen = key->namelen, .valuelen = key->valuelen, .owner = rx->sc->ip->i_ino, + .geo = rx->sc->mp->m_attr_geo, + .whichfork = XFS_ATTR_FORK, + .op_flags = XFS_DA_OP_OKNOENT, }; struct xchk_xattr_buf *ab = rx->sc->buf; int error; @@ -602,6 +605,7 @@ xrep_xattr_insert_rec( * xfs_attr_set creates and commits its own transaction. If the attr * already exists, we'll just drop it during the rebuild. */ + xfs_attr_sethash(&args); error = xfs_attr_set(&args, XFS_ATTRUPDATE_CREATE); if (error == -EEXIST) error = 0; diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index b43f7081b0f4..bbdbe9026658 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -17,6 +17,7 @@ #include "xfs_acl.h" #include "xfs_log.h" #include "xfs_xattr.h" +#include "xfs_quota.h" #include @@ -70,7 +71,9 @@ xfs_attr_want_log_assist( /* * Set or remove an xattr, having grabbed the appropriate logging resources - * prior to calling libxfs. + * prior to calling libxfs. Callers of this function are only required to + * initialize the inode, attr_filter, name, namelen, value, and valuelen fields + * of @args. */ int xfs_attr_change( @@ -80,7 +83,19 @@ xfs_attr_change( struct xfs_mount *mp = args->dp->i_mount; int error; - ASSERT(!(args->op_flags & XFS_DA_OP_LOGGED)); + if (xfs_is_shutdown(mp)) + return -EIO; + + error = xfs_qm_dqattach(args->dp); + if (error) + return error; + + /* + * We have no control over the attribute names that userspace passes us + * to remove, so we have to allow the name lookup prior to attribute + * removal to fail as well. + */ + args->op_flags = XFS_DA_OP_OKNOENT; if (xfs_attr_want_log_assist(mp)) { error = xfs_attr_grab_log_assist(mp); @@ -90,6 +105,11 @@ xfs_attr_change( args->op_flags |= XFS_DA_OP_LOGGED; } + args->owner = args->dp->i_ino; + args->geo = mp->m_attr_geo; + args->whichfork = XFS_ATTR_FORK; + xfs_attr_sethash(args); + return xfs_attr_set(args, op); } -- cgit v1.2.3-59-g8ed1b From bf61c36a45d4c215994699a7a06a00c58d22e8a2 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:07 -0700 Subject: xfs: make the reserved block permission flag explicit in xfs_attr_set Make the use of reserved blocks an explicit parameter to xfs_attr_set. Userspace setting XFS_ATTR_ROOT attrs should continue to be able to use it, but for online repairs we can back out and therefore do not care. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 6 +++--- fs/xfs/libxfs/xfs_attr.h | 2 +- fs/xfs/scrub/attr_repair.c | 2 +- fs/xfs/xfs_xattr.c | 2 +- 4 files changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index df8418671c37..c98145596f02 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -952,7 +952,7 @@ xfs_attr_lookup( * Make a change to the xattr structure. * * The caller must have initialized @args, attached dquots, and must not hold - * any ILOCKs. + * any ILOCKs. Reserved data blocks may be used if @rsvd is set. * * Returns -EEXIST for XFS_ATTRUPDATE_CREATE if the name already exists. * Returns -ENOATTR for XFS_ATTRUPDATE_REMOVE if the name does not exist. @@ -961,12 +961,12 @@ xfs_attr_lookup( int xfs_attr_set( struct xfs_da_args *args, - enum xfs_attr_update op) + enum xfs_attr_update op, + bool rsvd) { struct xfs_inode *dp = args->dp; struct xfs_mount *mp = dp->i_mount; struct xfs_trans_res tres; - bool rsvd = (args->attr_filter & XFS_ATTR_ROOT); int error, local; int rmt_blks = 0; unsigned int total; diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index d12583dd7eec..43dee4cbaab2 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -558,7 +558,7 @@ enum xfs_attr_update { XFS_ATTRUPDATE_REPLACE, /* set value, fail if attr does not exist */ }; -int xfs_attr_set(struct xfs_da_args *args, enum xfs_attr_update op); +int xfs_attr_set(struct xfs_da_args *args, enum xfs_attr_update op, bool rsvd); int xfs_attr_set_iter(struct xfs_attr_intent *attr); int xfs_attr_remove_iter(struct xfs_attr_intent *attr); bool xfs_attr_check_namespace(unsigned int attr_flags); diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index 67c0ec0d1dbb..cbcc446d5119 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -606,7 +606,7 @@ xrep_xattr_insert_rec( * already exists, we'll just drop it during the rebuild. */ xfs_attr_sethash(&args); - error = xfs_attr_set(&args, XFS_ATTRUPDATE_CREATE); + error = xfs_attr_set(&args, XFS_ATTRUPDATE_CREATE, false); if (error == -EEXIST) error = 0; diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index bbdbe9026658..ab3d22f662f2 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -110,7 +110,7 @@ xfs_attr_change( args->whichfork = XFS_ATTR_FORK; xfs_attr_sethash(args); - return xfs_attr_set(args, op); + return xfs_attr_set(args, op, args->attr_filter & XFS_ATTR_ROOT); } -- cgit v1.2.3-59-g8ed1b From 086e934fe9c741f25a269ae74cc891eaf3f5c4e2 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:08 -0700 Subject: xfs: salvage parent pointers when rebuilding xattr structures When we're salvaging extended attributes, make sure we validate the ones that claim to be parent pointers before adding them to the salvage pile. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr_repair.c | 36 +++++++++++++++++++++++++++--------- fs/xfs/scrub/trace.h | 38 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 65 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index cbcc446d5119..48443e88e298 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -28,6 +28,7 @@ #include "xfs_exchmaps.h" #include "xfs_exchrange.h" #include "xfs_acl.h" +#include "xfs_parent.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" @@ -127,6 +128,9 @@ xrep_xattr_want_salvage( return false; if (valuelen > XATTR_SIZE_MAX || valuelen < 0) return false; + if (attr_flags & XFS_ATTR_PARENT) + return xfs_parent_valuecheck(rx->sc->mp, value, valuelen); + return true; } @@ -154,14 +158,21 @@ xrep_xattr_salvage_key( * Truncate the name to the first character that would trip namecheck. * If we no longer have a name after that, ignore this attribute. */ - while (i < namelen && name[i] != 0) - i++; - if (i == 0) - return 0; - key.namelen = i; + if (flags & XFS_ATTR_PARENT) { + key.namelen = namelen; + + trace_xrep_xattr_salvage_pptr(rx->sc->ip, flags, name, + key.namelen, value, valuelen); + } else { + while (i < namelen && name[i] != 0) + i++; + if (i == 0) + return 0; + key.namelen = i; - trace_xrep_xattr_salvage_rec(rx->sc->ip, flags, name, key.namelen, - valuelen); + trace_xrep_xattr_salvage_rec(rx->sc->ip, flags, name, + key.namelen, valuelen); + } error = xfblob_store(rx->xattr_blobs, &key.name_cookie, name, key.namelen); @@ -598,8 +609,15 @@ xrep_xattr_insert_rec( ab->name[key->namelen] = 0; - trace_xrep_xattr_insert_rec(rx->sc->tempip, key->flags, ab->name, - key->namelen, key->valuelen); + if (key->flags & XFS_ATTR_PARENT) { + trace_xrep_xattr_insert_pptr(rx->sc->tempip, key->flags, + ab->name, key->namelen, ab->value, + key->valuelen); + args.op_flags |= XFS_DA_OP_LOGGED; + } else { + trace_xrep_xattr_insert_rec(rx->sc->tempip, key->flags, + ab->name, key->namelen, key->valuelen); + } /* * xfs_attr_set creates and commits its own transaction. If the attr diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 3e726610b9e3..4b968df3d840 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2540,6 +2540,44 @@ DEFINE_EVENT(xrep_xattr_salvage_class, name, \ DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_xattr_salvage_rec); DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_xattr_insert_rec); +DECLARE_EVENT_CLASS(xrep_pptr_salvage_class, + TP_PROTO(struct xfs_inode *ip, unsigned int flags, const void *name, + unsigned int namelen, const void *value, unsigned int valuelen), + TP_ARGS(ip, flags, name, namelen, value, valuelen), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, parent_gen) + __field(unsigned int, namelen) + __dynamic_array(char, name, namelen) + ), + TP_fast_assign( + const struct xfs_parent_rec *rec = value; + + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->parent_ino = be64_to_cpu(rec->p_ino); + __entry->parent_gen = be32_to_cpu(rec->p_gen); + __entry->namelen = namelen; + memcpy(__get_str(name), name, namelen); + ), + TP_printk("dev %d:%d ino 0x%llx parent_ino 0x%llx parent_gen 0x%x name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->parent_ino, + __entry->parent_gen, + __entry->namelen, + __get_str(name)) +) +#define DEFINE_XREP_PPTR_SALVAGE_EVENT(name) \ +DEFINE_EVENT(xrep_pptr_salvage_class, name, \ + TP_PROTO(struct xfs_inode *ip, unsigned int flags, const void *name, \ + unsigned int namelen, const void *value, unsigned int valuelen), \ + TP_ARGS(ip, flags, name, namelen, value, valuelen)) +DEFINE_XREP_PPTR_SALVAGE_EVENT(xrep_xattr_salvage_pptr); +DEFINE_XREP_PPTR_SALVAGE_EVENT(xrep_xattr_insert_pptr); + TRACE_EVENT(xrep_xattr_class, TP_PROTO(struct xfs_inode *ip, struct xfs_inode *arg_ip), TP_ARGS(ip, arg_ip), -- cgit v1.2.3-59-g8ed1b From 5769aa41ee34d4d1cc2b35376107b8e9694698f0 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:09 -0700 Subject: xfs: add raw parent pointer apis to support repair Add a couple of utility functions to set or remove parent pointers from a file. These functions will be used by repair code, hence they skip the xattr logging that regular parent pointer updates use. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_dir2.c | 2 +- fs/xfs/libxfs/xfs_dir2.h | 2 +- fs/xfs/libxfs/xfs_parent.c | 64 ++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_parent.h | 6 +++++ 4 files changed, 72 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index 9da99fa20c75..7634344dc515 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -434,7 +434,7 @@ int xfs_dir_removename( struct xfs_trans *tp, struct xfs_inode *dp, - struct xfs_name *name, + const struct xfs_name *name, xfs_ino_t ino, xfs_extlen_t total) /* bmap's total block count */ { diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index eb3a5c35025b..b580a78bcf4f 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -58,7 +58,7 @@ extern int xfs_dir_lookup(struct xfs_trans *tp, struct xfs_inode *dp, const struct xfs_name *name, xfs_ino_t *inum, struct xfs_name *ci_name); extern int xfs_dir_removename(struct xfs_trans *tp, struct xfs_inode *dp, - struct xfs_name *name, xfs_ino_t ino, + const struct xfs_name *name, xfs_ino_t ino, xfs_extlen_t tot); extern int xfs_dir_replace(struct xfs_trans *tp, struct xfs_inode *dp, const struct xfs_name *name, xfs_ino_t inum, diff --git a/fs/xfs/libxfs/xfs_parent.c b/fs/xfs/libxfs/xfs_parent.c index 1e53df5f8332..69366c44a701 100644 --- a/fs/xfs/libxfs/xfs_parent.c +++ b/fs/xfs/libxfs/xfs_parent.c @@ -313,3 +313,67 @@ xfs_parent_lookup( xfs_parent_da_args_init(scratch, tp, pptr, ip, ip->i_ino, parent_name); return xfs_attr_get_ilocked(scratch); } + +/* Sanity-check a parent pointer before we try to perform repairs. */ +static inline bool +xfs_parent_sanity_check( + struct xfs_mount *mp, + const struct xfs_name *parent_name, + const struct xfs_parent_rec *pptr) +{ + if (!xfs_parent_namecheck(XFS_ATTR_PARENT, parent_name->name, + parent_name->len)) + return false; + + if (!xfs_parent_valuecheck(mp, pptr, sizeof(*pptr))) + return false; + + return true; +} + + +/* + * Attach the parent pointer (@parent_name -> @pptr) to @ip immediately. + * Caller must not have a transaction or hold the ILOCK. This is for + * specialized repair functions only. The scratchpad need not be initialized. + */ +int +xfs_parent_set( + struct xfs_inode *ip, + xfs_ino_t owner, + const struct xfs_name *parent_name, + struct xfs_parent_rec *pptr, + struct xfs_da_args *scratch) +{ + if (!xfs_parent_sanity_check(ip->i_mount, parent_name, pptr)) { + ASSERT(0); + return -EFSCORRUPTED; + } + + memset(scratch, 0, sizeof(struct xfs_da_args)); + xfs_parent_da_args_init(scratch, NULL, pptr, ip, owner, parent_name); + return xfs_attr_set(scratch, XFS_ATTRUPDATE_CREATE, false); +} + +/* + * Remove the parent pointer (@parent_name -> @pptr) from @ip immediately. + * Caller must not have a transaction or hold the ILOCK. This is for + * specialized repair functions only. The scratchpad need not be initialized. + */ +int +xfs_parent_unset( + struct xfs_inode *ip, + xfs_ino_t owner, + const struct xfs_name *parent_name, + struct xfs_parent_rec *pptr, + struct xfs_da_args *scratch) +{ + if (!xfs_parent_sanity_check(ip->i_mount, parent_name, pptr)) { + ASSERT(0); + return -EFSCORRUPTED; + } + + memset(scratch, 0, sizeof(struct xfs_da_args)); + xfs_parent_da_args_init(scratch, NULL, pptr, ip, owner, parent_name); + return xfs_attr_set(scratch, XFS_ATTRUPDATE_REMOVE, false); +} diff --git a/fs/xfs/libxfs/xfs_parent.h b/fs/xfs/libxfs/xfs_parent.h index 97788582321a..b8036527cdc7 100644 --- a/fs/xfs/libxfs/xfs_parent.h +++ b/fs/xfs/libxfs/xfs_parent.h @@ -100,5 +100,11 @@ int xfs_parent_from_attr(struct xfs_mount *mp, unsigned int attr_flags, int xfs_parent_lookup(struct xfs_trans *tp, struct xfs_inode *ip, const struct xfs_name *name, struct xfs_parent_rec *pptr, struct xfs_da_args *scratch); +int xfs_parent_set(struct xfs_inode *ip, xfs_ino_t owner, + const struct xfs_name *name, struct xfs_parent_rec *pptr, + struct xfs_da_args *scratch); +int xfs_parent_unset(struct xfs_inode *ip, xfs_ino_t owner, + const struct xfs_name *name, struct xfs_parent_rec *pptr, + struct xfs_da_args *scratch); #endif /* __XFS_PARENT_H__ */ -- cgit v1.2.3-59-g8ed1b From 7b72ab2c6a468305449db8f204bf1e406fd3e147 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 23 Apr 2024 08:52:31 -0400 Subject: dlm: return -ENOMEM if ls_recover_buf fails This patch fixes to return -ENOMEM in case of an allocation failure that was forgotten to change in commit 6c648035cbe7 ("dlm: switch to use rhashtable for rsbs"). Reported-by: kernel test robot Reported-by: Dan Carpenter Closes: https://lore.kernel.org/r/202404200536.jGi6052v-lkp@intel.com/ Fixes: 6c648035cbe7 ("dlm: switch to use rhashtable for rsbs") Signed-off-by: Alexander Aring Signed-off-by: David Teigland --- fs/dlm/lockspace.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index ed23787271b1..475ab4370dda 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -482,8 +482,10 @@ static int new_lockspace(const char *name, const char *cluster, * might send less. */ ls->ls_recover_buf = kmalloc(DLM_MAX_SOCKET_BUFSIZE, GFP_NOFS); - if (!ls->ls_recover_buf) + if (!ls->ls_recover_buf) { + error = -ENOMEM; goto out_lkbidr; + } ls->ls_slot = 0; ls->ls_num_slots = 0; -- cgit v1.2.3-59-g8ed1b From 72a5425adc7ade7f7808e85d949dde26c05ec127 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:50:04 +0100 Subject: ext2: Remove call to folio_set_error() Nobody checks this flag on ext2 folios, stop setting it. Cc: Jan Kara Cc: linux-ext4@vger.kernel.org Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jan Kara Message-Id: <20240420025029.2166544-10-willy@infradead.org> --- fs/ext2/dir.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index 4fb155b5a958..087457061c6e 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -175,7 +175,6 @@ Eend: (unsigned long) le32_to_cpu(p->inode)); } fail: - folio_set_error(folio); return false; } -- cgit v1.2.3-59-g8ed1b From 91ba8cb1b09393b1c2222082e087f066fe13fd9b Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:50:08 +0100 Subject: isofs: Remove calls to set/clear the error flag Nobody checks the error flag on isofs folios, so stop setting and clearing it. Cc: Jan Kara Signed-off-by: Matthew Wilcox (Oracle) Acked-by: Jan Kara Signed-off-by: Jan Kara Message-Id: <20240420025029.2166544-14-willy@infradead.org> --- fs/isofs/compress.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'fs') diff --git a/fs/isofs/compress.c b/fs/isofs/compress.c index c4da3f634b92..34d5baa5d88a 100644 --- a/fs/isofs/compress.c +++ b/fs/isofs/compress.c @@ -346,8 +346,6 @@ static int zisofs_read_folio(struct file *file, struct folio *folio) for (i = 0; i < pcount; i++, index++) { if (i != full_page) pages[i] = grab_cache_page_nowait(mapping, index); - if (pages[i]) - ClearPageError(pages[i]); } err = zisofs_fill_pages(inode, full_page, pcount, pages); @@ -356,8 +354,6 @@ static int zisofs_read_folio(struct file *file, struct folio *folio) for (i = 0; i < pcount; i++) { if (pages[i]) { flush_dcache_page(pages[i]); - if (i == full_page && err) - SetPageError(pages[i]); unlock_page(pages[i]); if (i != full_page) put_page(pages[i]); -- cgit v1.2.3-59-g8ed1b From 76fc23b695f4717bb5e7b616eaad7d6213fdea9f Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:10 -0700 Subject: xfs: repair directories by scanning directory parent pointers For filesystems with parent pointers, scan the entire filesystem looking for parent pointers that target the directory we're rebuilding instead of trying to salvage whatever we can from the directory data blocks. This will be more robust than salvaging, but there's more code to come. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/dir_repair.c | 347 +++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 341 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 575397aef1f7..4e6b0e88b996 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -28,6 +28,7 @@ #include "xfs_exchmaps.h" #include "xfs_exchrange.h" #include "xfs_ag.h" +#include "xfs_parent.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" @@ -43,6 +44,7 @@ #include "scrub/reap.h" #include "scrub/findparent.h" #include "scrub/orphanage.h" +#include "scrub/listxattr.h" /* * Directory Repair @@ -57,6 +59,15 @@ * being repaired and the temporary directory, and will later become important * for parent pointer scanning. * + * If parent pointers are enabled on this filesystem, we instead reconstruct + * the directory by visiting each parent pointer of each file in the filesystem + * and translating the relevant parent pointer records into dirents. In this + * case, it is advantageous to stash all directory entries created from parent + * pointers for a single child file before replaying them into the temporary + * directory. To save memory, the live filesystem scan reuses the findparent + * fields. Directory repair chooses either parent pointer scanning or + * directory entry salvaging, but not both. + * * Directory entries added to the temporary directory do not elevate the link * counts of the inodes found. When salvaging completes, the remaining stashed * entries are replayed to the temporary directory. An atomic mapping exchange @@ -112,7 +123,15 @@ struct xrep_dir { /* * Information used to scan the filesystem to find the inumber of the - * dotdot entry for this directory. + * dotdot entry for this directory. For directory salvaging when + * parent pointers are not enabled, we use the findparent_* functions + * on this object and access only the parent_ino field directly. + * + * When parent pointers are enabled, however, the pptr scanner uses the + * iscan, hooks, lock, and parent_ino fields of this object directly. + * @pscan.lock coordinates access to dir_entries, dir_names, + * parent_ino, subdirs, dirents, and args. This reduces the memory + * requirements of this structure. */ struct xrep_parent_scan_info pscan; @@ -763,28 +782,35 @@ xrep_dir_replay_updates( int error; /* Add all the salvaged dirents to the temporary directory. */ + mutex_lock(&rd->pscan.lock); foreach_xfarray_idx(rd->dir_entries, array_cur) { struct xrep_dirent dirent; error = xfarray_load(rd->dir_entries, array_cur, &dirent); if (error) - return error; + goto out_unlock; error = xfblob_loadname(rd->dir_names, dirent.name_cookie, &rd->xname, dirent.namelen); if (error) - return error; + goto out_unlock; rd->xname.type = dirent.ftype; + mutex_unlock(&rd->pscan.lock); error = xrep_dir_replay_update(rd, &rd->xname, &dirent); if (error) return error; + mutex_lock(&rd->pscan.lock); } /* Empty out both arrays now that we've added the entries. */ xfarray_truncate(rd->dir_entries); xfblob_truncate(rd->dir_names); + mutex_unlock(&rd->pscan.lock); return 0; +out_unlock: + mutex_unlock(&rd->pscan.lock); + return error; } /* @@ -995,6 +1021,269 @@ xrep_dir_salvage_entries( } +/* + * Examine a parent pointer of a file. If it leads us back to the directory + * that we're rebuilding, create an incore dirent from the parent pointer and + * stash it. + */ +STATIC int +xrep_dir_scan_pptr( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xfs_name xname = { + .name = name, + .len = namelen, + .type = xfs_mode_to_ftype(VFS_I(ip)->i_mode), + }; + xfs_ino_t parent_ino; + uint32_t parent_gen; + struct xrep_dir *rd = priv; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + /* + * Ignore parent pointers that point back to a different dir, list the + * wrong generation number, or are invalid. + */ + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, &parent_ino, &parent_gen); + if (error) + return error; + + if (parent_ino != sc->ip->i_ino || + parent_gen != VFS_I(sc->ip)->i_generation) + return 0; + + mutex_lock(&rd->pscan.lock); + error = xrep_dir_stash_createname(rd, &xname, ip->i_ino); + mutex_unlock(&rd->pscan.lock); + return error; +} + +/* + * If this child dirent points to the directory being repaired, remember that + * fact so that we can reset the dotdot entry if necessary. + */ +STATIC int +xrep_dir_scan_dirent( + struct xfs_scrub *sc, + struct xfs_inode *dp, + xfs_dir2_dataptr_t dapos, + const struct xfs_name *name, + xfs_ino_t ino, + void *priv) +{ + struct xrep_dir *rd = priv; + + /* Dirent doesn't point to this directory. */ + if (ino != rd->sc->ip->i_ino) + return 0; + + /* Ignore garbage inum. */ + if (!xfs_verify_dir_ino(rd->sc->mp, ino)) + return 0; + + /* No weird looking names. */ + if (name->len >= MAXNAMELEN || name->len <= 0) + return 0; + + /* Don't pick up dot or dotdot entries; we only want child dirents. */ + if (xfs_dir2_samename(name, &xfs_name_dotdot) || + xfs_dir2_samename(name, &xfs_name_dot)) + return 0; + + trace_xrep_dir_stash_createname(sc->tempip, &xfs_name_dotdot, + dp->i_ino); + + xrep_findparent_scan_found(&rd->pscan, dp->i_ino); + return 0; +} + +/* + * Decide if we want to look for child dirents or parent pointers in this file. + * Skip the dir being repaired and any files being used to stage repairs. + */ +static inline bool +xrep_dir_want_scan( + struct xrep_dir *rd, + const struct xfs_inode *ip) +{ + return ip != rd->sc->ip && !xrep_is_tempfile(ip); +} + +/* + * Take ILOCK on a file that we want to scan. + * + * Select ILOCK_EXCL if the file is a directory with an unloaded data bmbt or + * has an unloaded attr bmbt. Otherwise, take ILOCK_SHARED. + */ +static inline unsigned int +xrep_dir_scan_ilock( + struct xrep_dir *rd, + struct xfs_inode *ip) +{ + uint lock_mode = XFS_ILOCK_SHARED; + + /* Need to take the shared ILOCK to advance the iscan cursor. */ + if (!xrep_dir_want_scan(rd, ip)) + goto lock; + + if (S_ISDIR(VFS_I(ip)->i_mode) && xfs_need_iread_extents(&ip->i_df)) { + lock_mode = XFS_ILOCK_EXCL; + goto lock; + } + + if (xfs_inode_has_attr_fork(ip) && xfs_need_iread_extents(&ip->i_af)) + lock_mode = XFS_ILOCK_EXCL; + +lock: + xfs_ilock(ip, lock_mode); + return lock_mode; +} + +/* + * Scan this file for relevant child dirents or parent pointers that point to + * the directory we're rebuilding. + */ +STATIC int +xrep_dir_scan_file( + struct xrep_dir *rd, + struct xfs_inode *ip) +{ + unsigned int lock_mode; + int error = 0; + + lock_mode = xrep_dir_scan_ilock(rd, ip); + + if (!xrep_dir_want_scan(rd, ip)) + goto scan_done; + + /* + * If the extended attributes look as though they has been zapped by + * the inode record repair code, we cannot scan for parent pointers. + */ + if (xchk_pptr_looks_zapped(ip)) { + error = -EBUSY; + goto scan_done; + } + + error = xchk_xattr_walk(rd->sc, ip, xrep_dir_scan_pptr, rd); + if (error) + goto scan_done; + + if (S_ISDIR(VFS_I(ip)->i_mode)) { + /* + * If the directory looks as though it has been zapped by the + * inode record repair code, we cannot scan for child dirents. + */ + if (xchk_dir_looks_zapped(ip)) { + error = -EBUSY; + goto scan_done; + } + + error = xchk_dir_walk(rd->sc, ip, xrep_dir_scan_dirent, rd); + if (error) + goto scan_done; + } + +scan_done: + xchk_iscan_mark_visited(&rd->pscan.iscan, ip); + xfs_iunlock(ip, lock_mode); + return error; +} + +/* + * Scan all files in the filesystem for parent pointers that we can turn into + * replacement dirents, and a dirent that we can use to set the dotdot pointer. + */ +STATIC int +xrep_dir_scan_dirtree( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + struct xfs_inode *ip; + int error; + + /* Roots of directory trees are their own parents. */ + if (sc->ip == sc->mp->m_rootip) + xrep_findparent_scan_found(&rd->pscan, sc->ip->i_ino); + + /* + * Filesystem scans are time consuming. Drop the directory ILOCK and + * all other resources for the duration of the scan and hope for the + * best. The live update hooks will keep our scan information up to + * date even though we've dropped the locks. + */ + xchk_trans_cancel(sc); + if (sc->ilock_flags & (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)) + xchk_iunlock(sc, sc->ilock_flags & (XFS_ILOCK_SHARED | + XFS_ILOCK_EXCL)); + error = xchk_trans_alloc_empty(sc); + if (error) + return error; + + while ((error = xchk_iscan_iter(&rd->pscan.iscan, &ip)) == 1) { + bool flush; + + error = xrep_dir_scan_file(rd, ip); + xchk_irele(sc, ip); + if (error) + break; + + /* Flush stashed dirent updates to constrain memory usage. */ + mutex_lock(&rd->pscan.lock); + flush = xrep_dir_want_flush_stashed(rd); + mutex_unlock(&rd->pscan.lock); + if (flush) { + xchk_trans_cancel(sc); + + error = xrep_tempfile_iolock_polled(sc); + if (error) + break; + + error = xrep_dir_replay_updates(rd); + xrep_tempfile_iounlock(sc); + if (error) + break; + + error = xchk_trans_alloc_empty(sc); + if (error) + break; + } + + if (xchk_should_terminate(sc, &error)) + break; + } + xchk_iscan_iter_finish(&rd->pscan.iscan); + if (error) { + /* + * If we couldn't grab an inode that was busy with a state + * change, change the error code so that we exit to userspace + * as quickly as possible. + */ + if (error == -EBUSY) + return -ECANCELED; + return error; + } + + /* + * Cancel the empty transaction so that we can (later) use the atomic + * file mapping exchange functions to lock files and commit the new + * directory. + */ + xchk_trans_cancel(rd->sc); + return 0; +} + /* * Free all the directory blocks and reset the data fork. The caller must * join the inode to the transaction. This function returns with the inode @@ -1194,6 +1483,45 @@ reset_nlink: return 0; } +/* + * Finish replaying stashed dirent updates, allocate a transaction for + * exchanging data fork mappings, and take the ILOCKs of both directories + * before we commit the new directory structure. + */ +STATIC int +xrep_dir_finalize_tempdir( + struct xrep_dir *rd) +{ + struct xfs_scrub *sc = rd->sc; + int error; + + if (!xfs_has_parent(sc->mp)) + return xrep_tempexch_trans_alloc(sc, XFS_DATA_FORK, &rd->tx); + + /* + * Repair relies on the ILOCK to quiesce all possible dirent updates. + * Replay all queued dirent updates into the tempdir before exchanging + * the contents, even if that means dropping the ILOCKs and the + * transaction. + */ + do { + error = xrep_dir_replay_updates(rd); + if (error) + return error; + + error = xrep_tempexch_trans_alloc(sc, XFS_DATA_FORK, &rd->tx); + if (error) + return error; + + if (xfarray_length(rd->dir_entries) == 0) + break; + + xchk_trans_cancel(sc); + xrep_tempfile_iunlock_both(sc); + } while (!xchk_should_terminate(sc, &error)); + return error; +} + /* Exchange the temporary directory's data fork with the one being repaired. */ STATIC int xrep_dir_swap( @@ -1296,8 +1624,12 @@ xrep_dir_rebuild_tree( if (error) return error; - /* Allocate transaction and ILOCK the scrub file and the temp file. */ - error = xrep_tempexch_trans_alloc(sc, XFS_DATA_FORK, &rd->tx); + /* + * Allocate transaction, lock inodes, and make sure that we've replayed + * all the stashed dirent updates to the tempdir. After this point, + * we're ready to exchange data fork mappings. + */ + error = xrep_dir_finalize_tempdir(rd); if (error) return error; @@ -1482,7 +1814,10 @@ xrep_directory( if (error) return error; - error = xrep_dir_salvage_entries(rd); + if (xfs_has_parent(sc->mp)) + error = xrep_dir_scan_dirtree(rd); + else + error = xrep_dir_salvage_entries(rd); if (error) goto out_teardown; -- cgit v1.2.3-59-g8ed1b From 8559b21a64d983315bdf1bd9f8dfdf732c56d057 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:10 -0700 Subject: xfs: implement live updates for directory repairs While we're scanning the filesystem for parent pointers that we can turn into dirents, we cannot hold the IOLOCK or ILOCK of the directory being repaired. Therefore, we need to set up a dirent hook so that we can keep the temporary directory up to date with the rest of the filesystem. Hence we add the ability to *remove* entries from the temporary dir. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/dir_repair.c | 218 ++++++++++++++++++++++++++++++++++++++++++---- fs/xfs/scrub/findparent.c | 10 ++- fs/xfs/scrub/findparent.h | 10 ++- fs/xfs/scrub/trace.h | 2 + 4 files changed, 218 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 4e6b0e88b996..60e31da4451e 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -85,6 +85,12 @@ * other threads. */ +/* Create a dirent in the tempdir. */ +#define XREP_DIRENT_ADD (1) + +/* Remove a dirent from the tempdir. */ +#define XREP_DIRENT_REMOVE (2) + /* Directory entry to be restored in the new directory. */ struct xrep_dirent { /* Cookie for retrieval of the dirent name. */ @@ -98,6 +104,9 @@ struct xrep_dirent { /* File type of the dirent. */ uint8_t ftype; + + /* XREP_DIRENT_{ADD,REMOVE} */ + uint8_t action; }; /* @@ -339,6 +348,7 @@ xrep_dir_stash_createname( xfs_ino_t ino) { struct xrep_dirent dirent = { + .action = XREP_DIRENT_ADD, .ino = ino, .namelen = name->len, .ftype = name->type, @@ -354,6 +364,33 @@ xrep_dir_stash_createname( return xfarray_append(rd->dir_entries, &dirent); } +/* + * Remember that we want to remove a dirent from the tempdir. These stashed + * actions will be replayed later. + */ +STATIC int +xrep_dir_stash_removename( + struct xrep_dir *rd, + const struct xfs_name *name, + xfs_ino_t ino) +{ + struct xrep_dirent dirent = { + .action = XREP_DIRENT_REMOVE, + .ino = ino, + .namelen = name->len, + .ftype = name->type, + }; + int error; + + trace_xrep_dir_stash_removename(rd->sc->tempip, name, ino); + + error = xfblob_storename(rd->dir_names, &dirent.name_cookie, name); + if (error) + return error; + + return xfarray_append(rd->dir_entries, &dirent); +} + /* Allocate an in-core record to hold entries while we rebuild the dir data. */ STATIC int xrep_dir_salvage_entry( @@ -705,6 +742,43 @@ xrep_dir_replay_createname( return xfs_dir2_node_addname(&rd->args); } +/* Replay a stashed removename onto the temporary directory. */ +STATIC int +xrep_dir_replay_removename( + struct xrep_dir *rd, + const struct xfs_name *name, + xfs_extlen_t total) +{ + struct xfs_inode *dp = rd->args.dp; + bool is_block, is_leaf; + int error; + + ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); + + xrep_dir_init_args(rd, dp, name); + rd->args.op_flags = 0; + rd->args.total = total; + + trace_xrep_dir_replay_removename(dp, name, 0); + + if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return xfs_dir2_sf_removename(&rd->args); + + error = xfs_dir2_isblock(&rd->args, &is_block); + if (error) + return error; + if (is_block) + return xfs_dir2_block_removename(&rd->args); + + error = xfs_dir2_isleaf(&rd->args, &is_leaf); + if (error) + return error; + if (is_leaf) + return xfs_dir2_leaf_removename(&rd->args); + + return xfs_dir2_node_removename(&rd->args); +} + /* * Add this stashed incore directory entry to the temporary directory. * The caller must hold the tempdir's IOLOCK, must not hold any ILOCKs, and @@ -732,26 +806,64 @@ xrep_dir_replay_update( xrep_tempfile_ilock(rd->sc); xfs_trans_ijoin(rd->sc->tp, rd->sc->tempip, 0); - /* - * Create a replacement dirent in the temporary directory. Note that - * _createname doesn't check for existing entries. There shouldn't be - * any in the temporary dir, but we'll verify this in debug mode. - */ + switch (dirent->action) { + case XREP_DIRENT_ADD: + /* + * Create a replacement dirent in the temporary directory. + * Note that _createname doesn't check for existing entries. + * There shouldn't be any in the temporary dir, but we'll + * verify this in debug mode. + */ #ifdef DEBUG - error = xchk_dir_lookup(rd->sc, rd->sc->tempip, xname, &ino); - if (error != -ENOENT) { - ASSERT(error != -ENOENT); - goto out_cancel; - } + error = xchk_dir_lookup(rd->sc, rd->sc->tempip, xname, &ino); + if (error != -ENOENT) { + ASSERT(error != -ENOENT); + goto out_cancel; + } #endif - error = xrep_dir_replay_createname(rd, xname, dirent->ino, resblks); - if (error) - goto out_cancel; + error = xrep_dir_replay_createname(rd, xname, dirent->ino, + resblks); + if (error) + goto out_cancel; + + if (xname->type == XFS_DIR3_FT_DIR) + rd->subdirs++; + rd->dirents++; + break; + case XREP_DIRENT_REMOVE: + /* + * Remove a dirent from the temporary directory. Note that + * _removename doesn't check the inode target of the exist + * entry. There should be a perfect match in the temporary + * dir, but we'll verify this in debug mode. + */ +#ifdef DEBUG + error = xchk_dir_lookup(rd->sc, rd->sc->tempip, xname, &ino); + if (error) { + ASSERT(error != 0); + goto out_cancel; + } + if (ino != dirent->ino) { + ASSERT(ino == dirent->ino); + error = -EIO; + goto out_cancel; + } +#endif - if (xname->type == XFS_DIR3_FT_DIR) - rd->subdirs++; - rd->dirents++; + error = xrep_dir_replay_removename(rd, xname, resblks); + if (error) + goto out_cancel; + + if (xname->type == XFS_DIR3_FT_DIR) + rd->subdirs--; + rd->dirents--; + break; + default: + ASSERT(0); + error = -EIO; + goto out_cancel; + } /* Commit and unlock. */ error = xrep_trans_commit(rd->sc); @@ -1284,6 +1396,71 @@ xrep_dir_scan_dirtree( return 0; } +/* + * Capture dirent updates being made by other threads which are relevant to the + * directory being repaired. + */ +STATIC int +xrep_dir_live_update( + struct notifier_block *nb, + unsigned long action, + void *data) +{ + struct xfs_dir_update_params *p = data; + struct xrep_dir *rd; + struct xfs_scrub *sc; + int error = 0; + + rd = container_of(nb, struct xrep_dir, pscan.dhook.dirent_hook.nb); + sc = rd->sc; + + /* + * This thread updated a child dirent in the directory that we're + * rebuilding. Stash the update for replay against the temporary + * directory. + */ + if (p->dp->i_ino == sc->ip->i_ino && + xchk_iscan_want_live_update(&rd->pscan.iscan, p->ip->i_ino)) { + mutex_lock(&rd->pscan.lock); + if (p->delta > 0) + error = xrep_dir_stash_createname(rd, p->name, + p->ip->i_ino); + else + error = xrep_dir_stash_removename(rd, p->name, + p->ip->i_ino); + mutex_unlock(&rd->pscan.lock); + if (error) + goto out_abort; + } + + /* + * This thread updated another directory's child dirent that points to + * the directory that we're rebuilding, so remember the new dotdot + * target. + */ + if (p->ip->i_ino == sc->ip->i_ino && + xchk_iscan_want_live_update(&rd->pscan.iscan, p->dp->i_ino)) { + if (p->delta > 0) { + trace_xrep_dir_stash_createname(sc->tempip, + &xfs_name_dotdot, + p->dp->i_ino); + + xrep_findparent_scan_found(&rd->pscan, p->dp->i_ino); + } else { + trace_xrep_dir_stash_removename(sc->tempip, + &xfs_name_dotdot, + rd->pscan.parent_ino); + + xrep_findparent_scan_found(&rd->pscan, NULLFSINO); + } + } + + return NOTIFY_DONE; +out_abort: + xchk_iscan_abort(&rd->pscan.iscan); + return NOTIFY_DONE; +} + /* * Free all the directory blocks and reset the data fork. The caller must * join the inode to the transaction. This function returns with the inode @@ -1633,6 +1810,9 @@ xrep_dir_rebuild_tree( if (error) return error; + if (xchk_iscan_aborted(&rd->pscan.iscan)) + return -ECANCELED; + /* * Exchange the tempdir's data fork with the file being repaired. This * recreates the transaction and re-takes the ILOCK in the scrub @@ -1688,7 +1868,11 @@ xrep_dir_setup_scan( if (error) goto out_xfarray; - error = xrep_findparent_scan_start(sc, &rd->pscan); + if (xfs_has_parent(sc->mp)) + error = __xrep_findparent_scan_start(sc, &rd->pscan, + xrep_dir_live_update); + else + error = xrep_findparent_scan_start(sc, &rd->pscan); if (error) goto out_xfblob; diff --git a/fs/xfs/scrub/findparent.c b/fs/xfs/scrub/findparent.c index 712dd73e4789..c78422ad757b 100644 --- a/fs/xfs/scrub/findparent.c +++ b/fs/xfs/scrub/findparent.c @@ -238,9 +238,10 @@ xrep_findparent_live_update( * will be called when there is a dotdot update for the inode being repaired. */ int -xrep_findparent_scan_start( +__xrep_findparent_scan_start( struct xfs_scrub *sc, - struct xrep_parent_scan_info *pscan) + struct xrep_parent_scan_info *pscan, + notifier_fn_t custom_fn) { int error; @@ -262,7 +263,10 @@ xrep_findparent_scan_start( * ILOCK, which means that any in-progress inode updates will finish * before we can scan the inode. */ - xfs_dir_hook_setup(&pscan->dhook, xrep_findparent_live_update); + if (custom_fn) + xfs_dir_hook_setup(&pscan->dhook, custom_fn); + else + xfs_dir_hook_setup(&pscan->dhook, xrep_findparent_live_update); error = xfs_dir_hook_add(sc->mp, &pscan->dhook); if (error) goto out_iscan; diff --git a/fs/xfs/scrub/findparent.h b/fs/xfs/scrub/findparent.h index 501f99d3164e..d998c7a88152 100644 --- a/fs/xfs/scrub/findparent.h +++ b/fs/xfs/scrub/findparent.h @@ -24,8 +24,14 @@ struct xrep_parent_scan_info { bool lookup_parent; }; -int xrep_findparent_scan_start(struct xfs_scrub *sc, - struct xrep_parent_scan_info *pscan); +int __xrep_findparent_scan_start(struct xfs_scrub *sc, + struct xrep_parent_scan_info *pscan, + notifier_fn_t custom_fn); +static inline int xrep_findparent_scan_start(struct xfs_scrub *sc, + struct xrep_parent_scan_info *pscan) +{ + return __xrep_findparent_scan_start(sc, pscan, NULL); +} int xrep_findparent_scan(struct xrep_parent_scan_info *pscan); void xrep_findparent_scan_teardown(struct xrep_parent_scan_info *pscan); diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 4b968df3d840..64db413b1888 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2692,6 +2692,8 @@ DEFINE_XREP_DIRENT_EVENT(xrep_dir_salvage_entry); DEFINE_XREP_DIRENT_EVENT(xrep_dir_stash_createname); DEFINE_XREP_DIRENT_EVENT(xrep_dir_replay_createname); DEFINE_XREP_DIRENT_EVENT(xrep_adoption_reparent); +DEFINE_XREP_DIRENT_EVENT(xrep_dir_stash_removename); +DEFINE_XREP_DIRENT_EVENT(xrep_dir_replay_removename); DECLARE_EVENT_CLASS(xrep_adoption_class, TP_PROTO(struct xfs_inode *dp, struct xfs_inode *ip, bool moved), -- cgit v1.2.3-59-g8ed1b From e5d7ce0364d8ee6821fd93814885c3bef775b9c3 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:11 -0700 Subject: xfs: replay unlocked parent pointer updates that accrue during xattr repair There are a few places where the extended attribute repair code drops the ILOCK to apply stashed xattrs to the temporary file. Although setxattr and removexattr are still locked out because we retain our hold on the IOLOCK, this doesn't prevent renames from updating parent pointers, because the VFS doesn't take i_rwsem on children that are being moved. Therefore, set up a dirent hook to capture parent pointer updates for this file, and replay(?) the updates. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr_repair.c | 438 ++++++++++++++++++++++++++++++++++++++++++++- fs/xfs/scrub/trace.h | 73 ++++++++ 2 files changed, 509 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index 48443e88e298..87f6c9cb54eb 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -96,6 +96,52 @@ struct xrep_xattr { /* Number of attributes that we are salvaging. */ unsigned long long attrs_found; + + /* Can we flush stashed attrs to the tempfile? */ + bool can_flush; + + /* Did the live update fail, and hence the repair is now out of date? */ + bool live_update_aborted; + + /* Lock protecting parent pointer updates */ + struct mutex lock; + + /* Fixed-size array of xrep_xattr_pptr structures. */ + struct xfarray *pptr_recs; + + /* Blobs containing parent pointer names. */ + struct xfblob *pptr_names; + + /* Hook to capture parent pointer updates. */ + struct xfs_dir_hook dhook; + + /* Scratch buffer for capturing parent pointers. */ + struct xfs_da_args pptr_args; + + /* Name buffer */ + struct xfs_name xname; + char namebuf[MAXNAMELEN]; +}; + +/* Create a parent pointer in the tempfile. */ +#define XREP_XATTR_PPTR_ADD (1) + +/* Remove a parent pointer from the tempfile. */ +#define XREP_XATTR_PPTR_REMOVE (2) + +/* A stashed parent pointer update. */ +struct xrep_xattr_pptr { + /* Cookie for retrieval of the pptr name. */ + xfblob_cookie name_cookie; + + /* Parent pointer record. */ + struct xfs_parent_rec pptr_rec; + + /* Length of the pptr name. */ + uint8_t namelen; + + /* XREP_XATTR_PPTR_{ADD,REMOVE} */ + uint8_t action; }; /* Set up to recreate the extended attributes. */ @@ -103,6 +149,9 @@ int xrep_setup_xattr( struct xfs_scrub *sc) { + if (xfs_has_parent(sc->mp)) + xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); + return xrep_tempfile_create(sc, S_IFREG); } @@ -711,11 +760,122 @@ xrep_xattr_want_flush_stashed( { unsigned long long bytes; + if (!rx->can_flush) + return false; + bytes = xfarray_bytes(rx->xattr_records) + xfblob_bytes(rx->xattr_blobs); return bytes > XREP_XATTR_MAX_STASH_BYTES; } +/* + * Did we observe rename changing parent pointer xattrs while we were flushing + * salvaged attrs? + */ +static inline bool +xrep_xattr_saw_pptr_conflict( + struct xrep_xattr *rx) +{ + bool ret; + + ASSERT(rx->can_flush); + + if (!xfs_has_parent(rx->sc->mp)) + return false; + + xfs_assert_ilocked(rx->sc->ip, XFS_ILOCK_EXCL); + + mutex_lock(&rx->lock); + ret = xfarray_bytes(rx->pptr_recs) > 0; + mutex_unlock(&rx->lock); + + return ret; +} + +/* + * Reset the entire repair state back to initial conditions, now that we've + * detected a parent pointer update to the attr structure while we were + * flushing salvaged attrs. See the locking notes in dir_repair.c for more + * information on why this is all necessary. + */ +STATIC int +xrep_xattr_full_reset( + struct xrep_xattr *rx) +{ + struct xfs_scrub *sc = rx->sc; + struct xfs_attr_sf_hdr *hdr; + struct xfs_ifork *ifp = &sc->tempip->i_af; + int error; + + trace_xrep_xattr_full_reset(sc->ip, sc->tempip); + + /* The temporary file's data fork had better not be in btree format. */ + if (sc->tempip->i_df.if_format == XFS_DINODE_FMT_BTREE) { + ASSERT(0); + return -EIO; + } + + /* + * We begin in transaction context with sc->ip ILOCKed but not joined + * to the transaction. To reset to the initial state, we must hold + * sc->ip's ILOCK to prevent rename from updating parent pointer + * information and the tempfile's ILOCK to clear its contents. + */ + xchk_iunlock(rx->sc, XFS_ILOCK_EXCL); + xrep_tempfile_ilock_both(sc); + xfs_trans_ijoin(sc->tp, sc->ip, 0); + xfs_trans_ijoin(sc->tp, sc->tempip, 0); + + /* + * Free all the blocks of the attr fork of the temp file, and reset + * it back to local format. + */ + if (xfs_ifork_has_extents(&sc->tempip->i_af)) { + error = xrep_reap_ifork(sc, sc->tempip, XFS_ATTR_FORK); + if (error) + return error; + + ASSERT(ifp->if_bytes == 0); + ifp->if_format = XFS_DINODE_FMT_LOCAL; + xfs_idata_realloc(sc->tempip, sizeof(*hdr), XFS_ATTR_FORK); + } + + /* Reinitialize the attr fork to an empty shortform structure. */ + hdr = ifp->if_data; + memset(hdr, 0, sizeof(*hdr)); + hdr->totsize = cpu_to_be16(sizeof(*hdr)); + xfs_trans_log_inode(sc->tp, sc->tempip, XFS_ILOG_CORE | XFS_ILOG_ADATA); + + /* + * Roll this transaction to commit our reset ondisk. The tempfile + * should no longer be joined to the transaction, so we drop its ILOCK. + * This should leave us in transaction context with sc->ip ILOCKed but + * not joined to the transaction. + */ + error = xrep_roll_trans(sc); + if (error) + return error; + xrep_tempfile_iunlock(sc); + + /* + * Erase any accumulated parent pointer updates now that we've erased + * the tempfile's attr fork. We're resetting the entire repair state + * back to where we were initially, except now we won't flush salvaged + * xattrs until the very end. + */ + mutex_lock(&rx->lock); + xfarray_truncate(rx->pptr_recs); + xfblob_truncate(rx->pptr_names); + mutex_unlock(&rx->lock); + + rx->can_flush = false; + rx->attrs_found = 0; + + ASSERT(xfarray_bytes(rx->xattr_records) == 0); + ASSERT(xfblob_bytes(rx->xattr_blobs) == 0); + return 0; +} + /* Extract as many attribute keys and values as we can. */ STATIC int xrep_xattr_recover( @@ -730,6 +890,7 @@ xrep_xattr_recover( int nmap; int error; +restart: /* * Iterate each xattr leaf block in the attr fork to scan them for any * attributes that we might salvage. @@ -768,6 +929,14 @@ xrep_xattr_recover( error = xrep_xattr_flush_stashed(rx); if (error) return error; + + if (xrep_xattr_saw_pptr_conflict(rx)) { + error = xrep_xattr_full_reset(rx); + if (error) + return error; + + goto restart; + } } } } @@ -927,6 +1096,180 @@ xrep_xattr_salvage_attributes( return xrep_xattr_flush_stashed(rx); } +/* + * Add this stashed incore parent pointer to the temporary file. The caller + * must hold the tempdir's IOLOCK, must not hold any ILOCKs, and must not be in + * transaction context. + */ +STATIC int +xrep_xattr_replay_pptr_update( + struct xrep_xattr *rx, + const struct xfs_name *xname, + struct xrep_xattr_pptr *pptr) +{ + struct xfs_scrub *sc = rx->sc; + int error; + + switch (pptr->action) { + case XREP_XATTR_PPTR_ADD: + /* Create parent pointer. */ + trace_xrep_xattr_replay_parentadd(sc->tempip, xname, + &pptr->pptr_rec); + + error = xfs_parent_set(sc->tempip, sc->ip->i_ino, xname, + &pptr->pptr_rec, &rx->pptr_args); + ASSERT(error != -EEXIST); + return error; + case XREP_XATTR_PPTR_REMOVE: + /* Remove parent pointer. */ + trace_xrep_xattr_replay_parentremove(sc->tempip, xname, + &pptr->pptr_rec); + + error = xfs_parent_unset(sc->tempip, sc->ip->i_ino, xname, + &pptr->pptr_rec, &rx->pptr_args); + ASSERT(error != -ENOATTR); + return error; + } + + ASSERT(0); + return -EIO; +} + +/* + * Flush stashed parent pointer updates that have been recorded by the scanner. + * This is done to reduce the memory requirements of the xattr rebuild, since + * files can have a lot of hardlinks and the fs can be busy. + * + * Caller must not hold transactions or ILOCKs. Caller must hold the tempfile + * IOLOCK. + */ +STATIC int +xrep_xattr_replay_pptr_updates( + struct xrep_xattr *rx) +{ + xfarray_idx_t array_cur; + int error; + + mutex_lock(&rx->lock); + foreach_xfarray_idx(rx->pptr_recs, array_cur) { + struct xrep_xattr_pptr pptr; + + error = xfarray_load(rx->pptr_recs, array_cur, &pptr); + if (error) + goto out_unlock; + + error = xfblob_loadname(rx->pptr_names, pptr.name_cookie, + &rx->xname, pptr.namelen); + if (error) + goto out_unlock; + mutex_unlock(&rx->lock); + + error = xrep_xattr_replay_pptr_update(rx, &rx->xname, &pptr); + if (error) + return error; + + mutex_lock(&rx->lock); + } + + /* Empty out both arrays now that we've added the entries. */ + xfarray_truncate(rx->pptr_recs); + xfblob_truncate(rx->pptr_names); + mutex_unlock(&rx->lock); + return 0; +out_unlock: + mutex_unlock(&rx->lock); + return error; +} + +/* + * Remember that we want to create a parent pointer in the tempfile. These + * stashed actions will be replayed later. + */ +STATIC int +xrep_xattr_stash_parentadd( + struct xrep_xattr *rx, + const struct xfs_name *name, + const struct xfs_inode *dp) +{ + struct xrep_xattr_pptr pptr = { + .action = XREP_XATTR_PPTR_ADD, + .namelen = name->len, + }; + int error; + + trace_xrep_xattr_stash_parentadd(rx->sc->tempip, dp, name); + + xfs_inode_to_parent_rec(&pptr.pptr_rec, dp); + error = xfblob_storename(rx->pptr_names, &pptr.name_cookie, name); + if (error) + return error; + + return xfarray_append(rx->pptr_recs, &pptr); +} + +/* + * Remember that we want to remove a parent pointer from the tempfile. These + * stashed actions will be replayed later. + */ +STATIC int +xrep_xattr_stash_parentremove( + struct xrep_xattr *rx, + const struct xfs_name *name, + const struct xfs_inode *dp) +{ + struct xrep_xattr_pptr pptr = { + .action = XREP_XATTR_PPTR_REMOVE, + .namelen = name->len, + }; + int error; + + trace_xrep_xattr_stash_parentremove(rx->sc->tempip, dp, name); + + xfs_inode_to_parent_rec(&pptr.pptr_rec, dp); + error = xfblob_storename(rx->pptr_names, &pptr.name_cookie, name); + if (error) + return error; + + return xfarray_append(rx->pptr_recs, &pptr); +} + +/* + * Capture dirent updates being made by other threads. We will have to replay + * the parent pointer updates before exchanging attr forks. + */ +STATIC int +xrep_xattr_live_dirent_update( + struct notifier_block *nb, + unsigned long action, + void *data) +{ + struct xfs_dir_update_params *p = data; + struct xrep_xattr *rx; + struct xfs_scrub *sc; + int error; + + rx = container_of(nb, struct xrep_xattr, dhook.dirent_hook.nb); + sc = rx->sc; + + /* + * This thread updated a dirent that points to the file that we're + * repairing, so stash the update for replay against the temporary + * file. + */ + if (p->ip->i_ino != sc->ip->i_ino) + return NOTIFY_DONE; + + mutex_lock(&rx->lock); + if (p->delta > 0) + error = xrep_xattr_stash_parentadd(rx, p->name, p->dp); + else + error = xrep_xattr_stash_parentremove(rx, p->name, p->dp); + if (error) + rx->live_update_aborted = true; + mutex_unlock(&rx->lock); + return NOTIFY_DONE; +} + /* * Prepare both inodes' attribute forks for an exchange. Promote the tempfile * from short format to leaf format, and if the file being repaired has a short @@ -1030,6 +1373,45 @@ xrep_xattr_swap( return xrep_tempexch_contents(sc, tx); } +/* + * Finish replaying stashed parent pointer updates, allocate a transaction for + * exchanging extent mappings, and take the ILOCKs of both files before we + * commit the new extended attribute structure. + */ +STATIC int +xrep_xattr_finalize_tempfile( + struct xrep_xattr *rx) +{ + struct xfs_scrub *sc = rx->sc; + int error; + + if (!xfs_has_parent(sc->mp)) + return xrep_tempexch_trans_alloc(sc, XFS_ATTR_FORK, &rx->tx); + + /* + * Repair relies on the ILOCK to quiesce all possible xattr updates. + * Replay all queued parent pointer updates into the tempfile before + * exchanging the contents, even if that means dropping the ILOCKs and + * the transaction. + */ + do { + error = xrep_xattr_replay_pptr_updates(rx); + if (error) + return error; + + error = xrep_tempexch_trans_alloc(sc, XFS_ATTR_FORK, &rx->tx); + if (error) + return error; + + if (xfarray_length(rx->pptr_recs) == 0) + break; + + xchk_trans_cancel(sc); + xrep_tempfile_iunlock_both(sc); + } while (!xchk_should_terminate(sc, &error)); + return error; +} + /* * Exchange the new extended attribute data (which we created in the tempfile) * with the file being repaired. @@ -1082,8 +1464,12 @@ xrep_xattr_rebuild_tree( if (error) return error; - /* Allocate exchange transaction and lock both inodes. */ - error = xrep_tempexch_trans_alloc(rx->sc, XFS_ATTR_FORK, &rx->tx); + /* + * Allocate transaction, lock inodes, and make sure that we've replayed + * all the stashed parent pointer updates to the temp file. After this + * point, we're ready to exchange attr fork mappings. + */ + error = xrep_xattr_finalize_tempfile(rx); if (error) return error; @@ -1124,8 +1510,15 @@ STATIC void xrep_xattr_teardown( struct xrep_xattr *rx) { + if (xfs_has_parent(rx->sc->mp)) + xfs_dir_hook_del(rx->sc->mp, &rx->dhook); + if (rx->pptr_names) + xfblob_destroy(rx->pptr_names); + if (rx->pptr_recs) + xfarray_destroy(rx->pptr_recs); xfblob_destroy(rx->xattr_blobs); xfarray_destroy(rx->xattr_records); + mutex_destroy(&rx->lock); kfree(rx); } @@ -1144,6 +1537,10 @@ xrep_xattr_setup_scan( if (!rx) return -ENOMEM; rx->sc = sc; + rx->can_flush = true; + rx->xname.name = rx->namebuf; + + mutex_init(&rx->lock); /* * Allocate enough memory to handle loading local attr values from the @@ -1171,11 +1568,43 @@ xrep_xattr_setup_scan( if (error) goto out_keys; + if (xfs_has_parent(sc->mp)) { + ASSERT(sc->flags & XCHK_FSGATES_DIRENTS); + + descr = xchk_xfile_ino_descr(sc, + "xattr retained parent pointer entries"); + error = xfarray_create(descr, 0, + sizeof(struct xrep_xattr_pptr), + &rx->pptr_recs); + kfree(descr); + if (error) + goto out_values; + + descr = xchk_xfile_ino_descr(sc, + "xattr retained parent pointer names"); + error = xfblob_create(descr, &rx->pptr_names); + kfree(descr); + if (error) + goto out_pprecs; + + xfs_dir_hook_setup(&rx->dhook, xrep_xattr_live_dirent_update); + error = xfs_dir_hook_add(sc->mp, &rx->dhook); + if (error) + goto out_ppnames; + } + *rxp = rx; return 0; +out_ppnames: + xfblob_destroy(rx->pptr_names); +out_pprecs: + xfarray_destroy(rx->pptr_recs); +out_values: + xfblob_destroy(rx->xattr_blobs); out_keys: xfarray_destroy(rx->xattr_records); out_rx: + mutex_destroy(&rx->lock); kfree(rx); return error; } @@ -1212,6 +1641,11 @@ xrep_xattr( if (error) goto out_scan; + if (rx->live_update_aborted) { + error = -EIO; + goto out_scan; + } + /* Last chance to abort before we start committing fixes. */ if (xchk_should_terminate(sc, &error)) goto out_scan; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 64db413b1888..68532f686eeb 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2602,6 +2602,43 @@ DEFINE_EVENT(xrep_xattr_class, name, \ TP_ARGS(ip, arg_ip)) DEFINE_XREP_XATTR_EVENT(xrep_xattr_rebuild_tree); DEFINE_XREP_XATTR_EVENT(xrep_xattr_reset_fork); +DEFINE_XREP_XATTR_EVENT(xrep_xattr_full_reset); + +DECLARE_EVENT_CLASS(xrep_xattr_pptr_scan_class, + TP_PROTO(struct xfs_inode *ip, const struct xfs_inode *dp, + const struct xfs_name *name), + TP_ARGS(ip, dp, name), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, parent_gen) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->parent_ino = dp->i_ino; + __entry->parent_gen = VFS_IC(dp)->i_generation; + __entry->namelen = name->len; + memcpy(__get_str(name), name->name, name->len); + ), + TP_printk("dev %d:%d ino 0x%llx parent_ino 0x%llx parent_gen 0x%x name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->parent_ino, + __entry->parent_gen, + __entry->namelen, + __get_str(name)) +) +#define DEFINE_XREP_XATTR_PPTR_SCAN_EVENT(name) \ +DEFINE_EVENT(xrep_xattr_pptr_scan_class, name, \ + TP_PROTO(struct xfs_inode *ip, const struct xfs_inode *dp, \ + const struct xfs_name *name), \ + TP_ARGS(ip, dp, name)) +DEFINE_XREP_XATTR_PPTR_SCAN_EVENT(xrep_xattr_stash_parentadd); +DEFINE_XREP_XATTR_PPTR_SCAN_EVENT(xrep_xattr_stash_parentremove); TRACE_EVENT(xrep_dir_recover_dirblock, TP_PROTO(struct xfs_inode *dp, xfs_dablk_t dabno, uint32_t magic, @@ -2748,6 +2785,42 @@ DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_dir_salvaged_parent); DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_findparent_dirent); DEFINE_XREP_PARENT_SALVAGE_EVENT(xrep_findparent_from_dcache); +DECLARE_EVENT_CLASS(xrep_pptr_class, + TP_PROTO(struct xfs_inode *ip, const struct xfs_name *name, + const struct xfs_parent_rec *pptr), + TP_ARGS(ip, name, pptr), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, parent_gen) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->parent_ino = be64_to_cpu(pptr->p_ino); + __entry->parent_gen = be32_to_cpu(pptr->p_gen); + __entry->namelen = name->len; + memcpy(__get_str(name), name->name, name->len); + ), + TP_printk("dev %d:%d ino 0x%llx parent_ino 0x%llx parent_gen 0x%x name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->parent_ino, + __entry->parent_gen, + __entry->namelen, + __get_str(name)) +) +#define DEFINE_XREP_PPTR_EVENT(name) \ +DEFINE_EVENT(xrep_pptr_class, name, \ + TP_PROTO(struct xfs_inode *ip, const struct xfs_name *name, \ + const struct xfs_parent_rec *pptr), \ + TP_ARGS(ip, name, pptr)) +DEFINE_XREP_PPTR_EVENT(xrep_xattr_replay_parentadd); +DEFINE_XREP_PPTR_EVENT(xrep_xattr_replay_parentremove); + TRACE_EVENT(xrep_nlinks_set_record, TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, const struct xchk_nlink *obs), -- cgit v1.2.3-59-g8ed1b From b334f7fab57a07fbece40648b4a22ab3f173bd48 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:12 -0700 Subject: xfs: repair directory parent pointers by scanning for dirents If parent pointers are enabled on the filesystem, we can repair the entire dataset by walking the directories of the filesystem looking for dirents that we can turn into parent pointers. Once we have a full incore dataset, we'll figure out what to do with it, but that's for a subsequent patch. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/parent_repair.c | 414 ++++++++++++++++++++++++++++++++++++++++++- fs/xfs/scrub/trace.h | 36 ++++ 2 files changed, 447 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index 63590e1b3506..b4084a9f0e9c 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -24,6 +24,7 @@ #include "xfs_trans_space.h" #include "xfs_health.h" #include "xfs_exchmaps.h" +#include "xfs_parent.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" @@ -34,6 +35,9 @@ #include "scrub/readdir.h" #include "scrub/tempfile.h" #include "scrub/orphanage.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" /* * Repairing The Directory Parent Pointer @@ -49,14 +53,61 @@ * See the section on locking issues in dir_repair.c for more information about * conflicts with the VFS. The findparent code wll keep our incore parent * inode up to date. + * + * If parent pointers are enabled, we instead reconstruct the parent pointer + * information by visiting every directory entry of every directory in the + * system and translating the relevant dirents into parent pointers. In this + * case, it is advantageous to stash all parent pointers created from dirents + * from a single parent file before replaying them into the temporary file. To + * save memory, the live filesystem scan reuses the findparent object. Parent + * pointer repair chooses either directory scanning or findparent, but not + * both. + * + * When salvaging completes, the remaining stashed entries are replayed to the + * temporary file. All non-parent pointer extended attributes are copied to + * the temporary file's extended attributes. An atomic extent swap is used to + * commit the new directory blocks to the directory being repaired. This will + * disrupt attrmulti cursors. + */ + +/* A stashed parent pointer update. */ +struct xrep_pptr { + /* Cookie for retrieval of the pptr name. */ + xfblob_cookie name_cookie; + + /* Parent pointer record. */ + struct xfs_parent_rec pptr_rec; + + /* Length of the pptr name. */ + uint8_t namelen; +}; + +/* + * Stash up to 8 pages of recovered parent pointers in pptr_recs and + * pptr_names before we write them to the temp file. */ +#define XREP_PARENT_MAX_STASH_BYTES (PAGE_SIZE * 8) struct xrep_parent { struct xfs_scrub *sc; + /* Fixed-size array of xrep_pptr structures. */ + struct xfarray *pptr_recs; + + /* Blobs containing parent pointer names. */ + struct xfblob *pptr_names; + /* * Information used to scan the filesystem to find the inumber of the - * dotdot entry for this directory. + * dotdot entry for this directory. On filesystems without parent + * pointers, we use the findparent_* functions on this object and + * access only the parent_ino field directly. + * + * When parent pointers are enabled, the directory entry scanner uses + * the iscan, hooks, and lock fields of this object directly. + * @pscan.lock coordinates access to pptr_recs, pptr_names, pptr, and + * pptr_scratch. This reduces the memory requirements of this + * structure. */ struct xrep_parent_scan_info pscan; @@ -66,6 +117,9 @@ struct xrep_parent { /* Directory entry name, plus the trailing null. */ struct xfs_name xname; unsigned char namebuf[MAXNAMELEN]; + + /* Scratch buffer for scanning pptr xattrs */ + struct xfs_da_args pptr_args; }; /* Tear down all the incore stuff we created. */ @@ -74,6 +128,12 @@ xrep_parent_teardown( struct xrep_parent *rp) { xrep_findparent_scan_teardown(&rp->pscan); + if (rp->pptr_names) + xfblob_destroy(rp->pptr_names); + rp->pptr_names = NULL; + if (rp->pptr_recs) + xfarray_destroy(rp->pptr_recs); + rp->pptr_recs = NULL; } /* Set up for a parent repair. */ @@ -82,6 +142,7 @@ xrep_setup_parent( struct xfs_scrub *sc) { struct xrep_parent *rp; + int error; xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); @@ -92,6 +153,10 @@ xrep_setup_parent( rp->xname.name = rp->namebuf; sc->buf = rp; + error = xrep_tempfile_create(sc, S_IFREG); + if (error) + return error; + return xrep_orphanage_try_create(sc); } @@ -147,6 +212,307 @@ out_relock: return error; } +/* + * Add this stashed incore parent pointer to the temporary file. + * The caller must hold the tempdir's IOLOCK, must not hold any ILOCKs, and + * must not be in transaction context. + */ +STATIC int +xrep_parent_replay_update( + struct xrep_parent *rp, + const struct xfs_name *xname, + struct xrep_pptr *pptr) +{ + struct xfs_scrub *sc = rp->sc; + + /* Create parent pointer. */ + trace_xrep_parent_replay_parentadd(sc->tempip, xname, &pptr->pptr_rec); + + return xfs_parent_set(sc->tempip, sc->ip->i_ino, xname, + &pptr->pptr_rec, &rp->pptr_args); +} + +/* + * Flush stashed parent pointer updates that have been recorded by the scanner. + * This is done to reduce the memory requirements of the parent pointer + * rebuild, since files can have a lot of hardlinks and the fs can be busy. + * + * Caller must not hold transactions or ILOCKs. Caller must hold the tempfile + * IOLOCK. + */ +STATIC int +xrep_parent_replay_updates( + struct xrep_parent *rp) +{ + xfarray_idx_t array_cur; + int error; + + mutex_lock(&rp->pscan.lock); + foreach_xfarray_idx(rp->pptr_recs, array_cur) { + struct xrep_pptr pptr; + + error = xfarray_load(rp->pptr_recs, array_cur, &pptr); + if (error) + goto out_unlock; + + error = xfblob_loadname(rp->pptr_names, pptr.name_cookie, + &rp->xname, pptr.namelen); + if (error) + goto out_unlock; + rp->xname.len = pptr.namelen; + mutex_unlock(&rp->pscan.lock); + + error = xrep_parent_replay_update(rp, &rp->xname, &pptr); + if (error) + return error; + + mutex_lock(&rp->pscan.lock); + } + + /* Empty out both arrays now that we've added the entries. */ + xfarray_truncate(rp->pptr_recs); + xfblob_truncate(rp->pptr_names); + mutex_unlock(&rp->pscan.lock); + return 0; +out_unlock: + mutex_unlock(&rp->pscan.lock); + return error; +} + +/* + * Remember that we want to create a parent pointer in the tempfile. These + * stashed actions will be replayed later. + */ +STATIC int +xrep_parent_stash_parentadd( + struct xrep_parent *rp, + const struct xfs_name *name, + const struct xfs_inode *dp) +{ + struct xrep_pptr pptr = { + .namelen = name->len, + }; + int error; + + trace_xrep_parent_stash_parentadd(rp->sc->tempip, dp, name); + + xfs_inode_to_parent_rec(&pptr.pptr_rec, dp); + error = xfblob_storename(rp->pptr_names, &pptr.name_cookie, name); + if (error) + return error; + + return xfarray_append(rp->pptr_recs, &pptr); +} + +/* + * Examine an entry of a directory. If this dirent leads us back to the file + * whose parent pointers we're rebuilding, add a pptr to the temporary + * directory. + */ +STATIC int +xrep_parent_scan_dirent( + struct xfs_scrub *sc, + struct xfs_inode *dp, + xfs_dir2_dataptr_t dapos, + const struct xfs_name *name, + xfs_ino_t ino, + void *priv) +{ + struct xrep_parent *rp = priv; + int error; + + /* Dirent doesn't point to this directory. */ + if (ino != rp->sc->ip->i_ino) + return 0; + + /* No weird looking names. */ + if (name->len == 0 || !xfs_dir2_namecheck(name->name, name->len)) + return -EFSCORRUPTED; + + /* No mismatching ftypes. */ + if (name->type != xfs_mode_to_ftype(VFS_I(sc->ip)->i_mode)) + return -EFSCORRUPTED; + + /* Don't pick up dot or dotdot entries; we only want child dirents. */ + if (xfs_dir2_samename(name, &xfs_name_dotdot) || + xfs_dir2_samename(name, &xfs_name_dot)) + return 0; + + /* + * Transform this dirent into a parent pointer and queue it for later + * addition to the temporary file. + */ + mutex_lock(&rp->pscan.lock); + error = xrep_parent_stash_parentadd(rp, name, dp); + mutex_unlock(&rp->pscan.lock); + return error; +} + +/* + * Decide if we want to look for dirents in this directory. Skip the file + * being repaired and any files being used to stage repairs. + */ +static inline bool +xrep_parent_want_scan( + struct xrep_parent *rp, + const struct xfs_inode *ip) +{ + return ip != rp->sc->ip && !xrep_is_tempfile(ip); +} + +/* + * Take ILOCK on a file that we want to scan. + * + * Select ILOCK_EXCL if the file is a directory with an unloaded data bmbt. + * Otherwise, take ILOCK_SHARED. + */ +static inline unsigned int +xrep_parent_scan_ilock( + struct xrep_parent *rp, + struct xfs_inode *ip) +{ + uint lock_mode = XFS_ILOCK_SHARED; + + /* Still need to take the shared ILOCK to advance the iscan cursor. */ + if (!xrep_parent_want_scan(rp, ip)) + goto lock; + + if (S_ISDIR(VFS_I(ip)->i_mode) && xfs_need_iread_extents(&ip->i_df)) { + lock_mode = XFS_ILOCK_EXCL; + goto lock; + } + +lock: + xfs_ilock(ip, lock_mode); + return lock_mode; +} + +/* + * Scan this file for relevant child dirents that point to the file whose + * parent pointers we're rebuilding. + */ +STATIC int +xrep_parent_scan_file( + struct xrep_parent *rp, + struct xfs_inode *ip) +{ + unsigned int lock_mode; + int error = 0; + + lock_mode = xrep_parent_scan_ilock(rp, ip); + + if (!xrep_parent_want_scan(rp, ip)) + goto scan_done; + + if (S_ISDIR(VFS_I(ip)->i_mode)) { + /* + * If the directory looks as though it has been zapped by the + * inode record repair code, we cannot scan for child dirents. + */ + if (xchk_dir_looks_zapped(ip)) { + error = -EBUSY; + goto scan_done; + } + + error = xchk_dir_walk(rp->sc, ip, xrep_parent_scan_dirent, rp); + if (error) + goto scan_done; + } + +scan_done: + xchk_iscan_mark_visited(&rp->pscan.iscan, ip); + xfs_iunlock(ip, lock_mode); + return error; +} + +/* Decide if we've stashed too much pptr data in memory. */ +static inline bool +xrep_parent_want_flush_stashed( + struct xrep_parent *rp) +{ + unsigned long long bytes; + + bytes = xfarray_bytes(rp->pptr_recs) + xfblob_bytes(rp->pptr_names); + return bytes > XREP_PARENT_MAX_STASH_BYTES; +} + +/* + * Scan all directories in the filesystem to look for dirents that we can turn + * into parent pointers. + */ +STATIC int +xrep_parent_scan_dirtree( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + struct xfs_inode *ip; + int error; + + /* + * Filesystem scans are time consuming. Drop the file ILOCK and all + * other resources for the duration of the scan and hope for the best. + * The live update hooks will keep our scan information up to date. + */ + xchk_trans_cancel(sc); + if (sc->ilock_flags & (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)) + xchk_iunlock(sc, sc->ilock_flags & (XFS_ILOCK_SHARED | + XFS_ILOCK_EXCL)); + error = xchk_trans_alloc_empty(sc); + if (error) + return error; + + while ((error = xchk_iscan_iter(&rp->pscan.iscan, &ip)) == 1) { + bool flush; + + error = xrep_parent_scan_file(rp, ip); + xchk_irele(sc, ip); + if (error) + break; + + /* Flush stashed pptr updates to constrain memory usage. */ + mutex_lock(&rp->pscan.lock); + flush = xrep_parent_want_flush_stashed(rp); + mutex_unlock(&rp->pscan.lock); + if (flush) { + xchk_trans_cancel(sc); + + error = xrep_tempfile_iolock_polled(sc); + if (error) + break; + + error = xrep_parent_replay_updates(rp); + xrep_tempfile_iounlock(sc); + if (error) + break; + + error = xchk_trans_alloc_empty(sc); + if (error) + break; + } + + if (xchk_should_terminate(sc, &error)) + break; + } + xchk_iscan_iter_finish(&rp->pscan.iscan); + if (error) { + /* + * If we couldn't grab an inode that was busy with a state + * change, change the error code so that we exit to userspace + * as quickly as possible. + */ + if (error == -EBUSY) + return -ECANCELED; + return error; + } + + /* + * Cancel the empty transaction so that we can (later) use the atomic + * extent swap helpers to lock files and commit the new directory. + */ + xchk_trans_cancel(rp->sc); + return 0; +} + /* Reset a directory's dotdot entry, if needed. */ STATIC int xrep_parent_reset_dotdot( @@ -298,8 +664,39 @@ xrep_parent_setup_scan( struct xrep_parent *rp) { struct xfs_scrub *sc = rp->sc; + char *descr; + int error; - return xrep_findparent_scan_start(sc, &rp->pscan); + if (!xfs_has_parent(sc->mp)) + return xrep_findparent_scan_start(sc, &rp->pscan); + + /* Set up some staging memory for logging parent pointer updates. */ + descr = xchk_xfile_ino_descr(sc, "parent pointer entries"); + error = xfarray_create(descr, 0, sizeof(struct xrep_pptr), + &rp->pptr_recs); + kfree(descr); + if (error) + return error; + + descr = xchk_xfile_ino_descr(sc, "parent pointer names"); + error = xfblob_create(descr, &rp->pptr_names); + kfree(descr); + if (error) + goto out_recs; + + error = xrep_findparent_scan_start(sc, &rp->pscan); + if (error) + goto out_names; + + return 0; + +out_names: + xfblob_destroy(rp->pptr_names); + rp->pptr_names = NULL; +out_recs: + xfarray_destroy(rp->pptr_recs); + rp->pptr_recs = NULL; + return error; } int @@ -309,11 +706,22 @@ xrep_parent( struct xrep_parent *rp = sc->buf; int error; + /* + * When the parent pointers feature is enabled, repairs are committed + * by atomically committing a new xattr structure and reaping the old + * attr fork. Reaping requires rmap to be enabled. + */ + if (xfs_has_parent(sc->mp) && !xfs_has_rmapbt(sc->mp)) + return -EOPNOTSUPP; + error = xrep_parent_setup_scan(rp); if (error) return error; - error = xrep_parent_find_dotdot(rp); + if (xfs_has_parent(sc->mp)) + error = xrep_parent_scan_dirtree(rp); + else + error = xrep_parent_find_dotdot(rp); if (error) goto out_teardown; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 68532f686eeb..10c2a8d10058 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2820,6 +2820,42 @@ DEFINE_EVENT(xrep_pptr_class, name, \ TP_ARGS(ip, name, pptr)) DEFINE_XREP_PPTR_EVENT(xrep_xattr_replay_parentadd); DEFINE_XREP_PPTR_EVENT(xrep_xattr_replay_parentremove); +DEFINE_XREP_PPTR_EVENT(xrep_parent_replay_parentadd); + +DECLARE_EVENT_CLASS(xrep_pptr_scan_class, + TP_PROTO(struct xfs_inode *ip, const struct xfs_inode *dp, + const struct xfs_name *name), + TP_ARGS(ip, dp, name), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, parent_gen) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->parent_ino = dp->i_ino; + __entry->parent_gen = VFS_IC(dp)->i_generation; + __entry->namelen = name->len; + memcpy(__get_str(name), name->name, name->len); + ), + TP_printk("dev %d:%d ino 0x%llx parent_ino 0x%llx parent_gen 0x%x name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->parent_ino, + __entry->parent_gen, + __entry->namelen, + __get_str(name)) +) +#define DEFINE_XREP_PPTR_SCAN_EVENT(name) \ +DEFINE_EVENT(xrep_pptr_scan_class, name, \ + TP_PROTO(struct xfs_inode *ip, const struct xfs_inode *dp, \ + const struct xfs_name *name), \ + TP_ARGS(ip, dp, name)) +DEFINE_XREP_PPTR_SCAN_EVENT(xrep_parent_stash_parentadd); TRACE_EVENT(xrep_nlinks_set_record, TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, -- cgit v1.2.3-59-g8ed1b From 65a1fb7a11291f361d36e6ebf3bb5e60e9ca8d13 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:13 -0700 Subject: xfs: implement live updates for parent pointer repairs While we're scanning the filesystem for dirents that we can turn into parent pointers, we cannot hold the IOLOCK or ILOCK of the file being repaired. Therefore, we need to set up a dirent hook so that we can keep the temporary file's parent pionters up to date with the rest of the filesystem. Hence we add the ability to *remove* pptrs from the temporary file. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/parent_repair.c | 103 ++++++++++++++++++++++++++++++++++++++++--- fs/xfs/scrub/trace.h | 2 + 2 files changed, 100 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index b4084a9f0e9c..311bc7990d7c 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -70,6 +70,12 @@ * disrupt attrmulti cursors. */ +/* Create a parent pointer in the tempfile. */ +#define XREP_PPTR_ADD (1) + +/* Remove a parent pointer from the tempfile. */ +#define XREP_PPTR_REMOVE (2) + /* A stashed parent pointer update. */ struct xrep_pptr { /* Cookie for retrieval of the pptr name. */ @@ -80,6 +86,9 @@ struct xrep_pptr { /* Length of the pptr name. */ uint8_t namelen; + + /* XREP_PPTR_{ADD,REMOVE} */ + uint8_t action; }; /* @@ -225,11 +234,25 @@ xrep_parent_replay_update( { struct xfs_scrub *sc = rp->sc; - /* Create parent pointer. */ - trace_xrep_parent_replay_parentadd(sc->tempip, xname, &pptr->pptr_rec); + switch (pptr->action) { + case XREP_PPTR_ADD: + /* Create parent pointer. */ + trace_xrep_parent_replay_parentadd(sc->tempip, xname, + &pptr->pptr_rec); + + return xfs_parent_set(sc->tempip, sc->ip->i_ino, xname, + &pptr->pptr_rec, &rp->pptr_args); + case XREP_PPTR_REMOVE: + /* Remove parent pointer. */ + trace_xrep_parent_replay_parentremove(sc->tempip, xname, + &pptr->pptr_rec); + + return xfs_parent_unset(sc->tempip, sc->ip->i_ino, xname, + &pptr->pptr_rec, &rp->pptr_args); + } - return xfs_parent_set(sc->tempip, sc->ip->i_ino, xname, - &pptr->pptr_rec, &rp->pptr_args); + ASSERT(0); + return -EIO; } /* @@ -290,6 +313,7 @@ xrep_parent_stash_parentadd( const struct xfs_inode *dp) { struct xrep_pptr pptr = { + .action = XREP_PPTR_ADD, .namelen = name->len, }; int error; @@ -304,6 +328,32 @@ xrep_parent_stash_parentadd( return xfarray_append(rp->pptr_recs, &pptr); } +/* + * Remember that we want to remove a parent pointer from the tempfile. These + * stashed actions will be replayed later. + */ +STATIC int +xrep_parent_stash_parentremove( + struct xrep_parent *rp, + const struct xfs_name *name, + const struct xfs_inode *dp) +{ + struct xrep_pptr pptr = { + .action = XREP_PPTR_REMOVE, + .namelen = name->len, + }; + int error; + + trace_xrep_parent_stash_parentremove(rp->sc->tempip, dp, name); + + xfs_inode_to_parent_rec(&pptr.pptr_rec, dp); + error = xfblob_storename(rp->pptr_names, &pptr.name_cookie, name); + if (error) + return error; + + return xfarray_append(rp->pptr_recs, &pptr); +} + /* * Examine an entry of a directory. If this dirent leads us back to the file * whose parent pointers we're rebuilding, add a pptr to the temporary @@ -513,6 +563,48 @@ xrep_parent_scan_dirtree( return 0; } +/* + * Capture dirent updates being made by other threads which are relevant to the + * file being repaired. + */ +STATIC int +xrep_parent_live_update( + struct notifier_block *nb, + unsigned long action, + void *data) +{ + struct xfs_dir_update_params *p = data; + struct xrep_parent *rp; + struct xfs_scrub *sc; + int error; + + rp = container_of(nb, struct xrep_parent, pscan.dhook.dirent_hook.nb); + sc = rp->sc; + + /* + * This thread updated a dirent that points to the file that we're + * repairing, so stash the update for replay against the temporary + * file. + */ + if (p->ip->i_ino == sc->ip->i_ino && + xchk_iscan_want_live_update(&rp->pscan.iscan, p->dp->i_ino)) { + mutex_lock(&rp->pscan.lock); + if (p->delta > 0) + error = xrep_parent_stash_parentadd(rp, p->name, p->dp); + else + error = xrep_parent_stash_parentremove(rp, p->name, + p->dp); + mutex_unlock(&rp->pscan.lock); + if (error) + goto out_abort; + } + + return NOTIFY_DONE; +out_abort: + xchk_iscan_abort(&rp->pscan.iscan); + return NOTIFY_DONE; +} + /* Reset a directory's dotdot entry, if needed. */ STATIC int xrep_parent_reset_dotdot( @@ -684,7 +776,8 @@ xrep_parent_setup_scan( if (error) goto out_recs; - error = xrep_findparent_scan_start(sc, &rp->pscan); + error = __xrep_findparent_scan_start(sc, &rp->pscan, + xrep_parent_live_update); if (error) goto out_names; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 10c2a8d10058..3e0cd482379c 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2821,6 +2821,7 @@ DEFINE_EVENT(xrep_pptr_class, name, \ DEFINE_XREP_PPTR_EVENT(xrep_xattr_replay_parentadd); DEFINE_XREP_PPTR_EVENT(xrep_xattr_replay_parentremove); DEFINE_XREP_PPTR_EVENT(xrep_parent_replay_parentadd); +DEFINE_XREP_PPTR_EVENT(xrep_parent_replay_parentremove); DECLARE_EVENT_CLASS(xrep_pptr_scan_class, TP_PROTO(struct xfs_inode *ip, const struct xfs_inode *dp, @@ -2856,6 +2857,7 @@ DEFINE_EVENT(xrep_pptr_scan_class, name, \ const struct xfs_name *name), \ TP_ARGS(ip, dp, name)) DEFINE_XREP_PPTR_SCAN_EVENT(xrep_parent_stash_parentadd); +DEFINE_XREP_PPTR_SCAN_EVENT(xrep_parent_stash_parentremove); TRACE_EVENT(xrep_nlinks_set_record, TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, -- cgit v1.2.3-59-g8ed1b From 13db7007892694c891fc37feccbd2ac8f227af78 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:14 -0700 Subject: xfs: remove pointless unlocked assertion Remove this assertion about the inode not having an attr fork from xfs_bmap_add_attrfork because the function handles that case just fine. Weirder still, the function actually /requires/ the caller not to hold the ILOCK, which means that its accesses are not stabilized. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_bmap.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 8a1446e025e0..1f528cf2d906 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -1041,8 +1041,6 @@ xfs_bmap_add_attrfork( int logflags; /* logging flags */ int error; /* error return value */ - ASSERT(xfs_inode_has_attr_fork(ip) == 0); - mp = ip->i_mount; ASSERT(!XFS_NOT_DQATTACHED(mp, ip)); -- cgit v1.2.3-59-g8ed1b From 55edcd1f86474f973fccf5c5ccc8bc7908893142 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:15 -0700 Subject: xfs: split xfs_bmap_add_attrfork into two pieces Split this function into two pieces -- one to make the actual changes to the inode core to add the attr fork, and another one to deal with getting the transaction and locking the inodes. The next couple of patches will need this to be split into two. One patch implements committing new parent pointer recordsets to damaged files. If one file has an attr fork and the other does not, we have to create the missing attr fork before the atomic swap transaction, and can use the behavior encoded in the current xfs_bmap_add_attrfork. The second patch adapts /lost+found adoptions to handle parent pointers correctly. The adoption process will add a parent pointer to a child that is being moved to /lost+found, but this requires that the attr fork already exists. We don't know if we're actually going to commit the adoption until we've already reserved a transaction and taken the ILOCKs, which means that we must have a way to bypass the start of the current xfs_bmap_add_attrfork. Therefore, create xfs_attr_add_fork as the helper that creates a transaction and takes locks; and make xfs_bmap_add_attrfork the function that updates the inode core and allocates the incore attr fork. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 39 ++++++++++++++++++++++++++++++++++++++- fs/xfs/libxfs/xfs_bmap.c | 36 ++++++++++-------------------------- fs/xfs/libxfs/xfs_bmap.h | 3 ++- 3 files changed, 50 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index c98145596f02..ab6ec2f15d76 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -948,6 +948,43 @@ xfs_attr_lookup( return error; } +STATIC int +xfs_attr_add_fork( + struct xfs_inode *ip, /* incore inode pointer */ + int size, /* space new attribute needs */ + int rsvd) /* xact may use reserved blks */ +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; /* transaction pointer */ + unsigned int blks; /* space reservation */ + int error; /* error return value */ + + ASSERT(!XFS_NOT_DQATTACHED(mp, ip)); + + blks = XFS_ADDAFORK_SPACE_RES(mp); + + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_addafork, blks, 0, + rsvd, &tp); + if (error) + return error; + + if (xfs_inode_has_attr_fork(ip)) + goto trans_cancel; + + error = xfs_bmap_add_attrfork(tp, ip, size, rsvd); + if (error) + goto trans_cancel; + + error = xfs_trans_commit(tp); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + return error; + +trans_cancel: + xfs_trans_cancel(tp); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + return error; +} + /* * Make a change to the xattr structure. * @@ -989,7 +1026,7 @@ xfs_attr_set( xfs_attr_sf_entsize_byname(args->namelen, args->valuelen); - error = xfs_bmap_add_attrfork(dp, sf_size, rsvd); + error = xfs_attr_add_fork(dp, sf_size, rsvd); if (error) return error; } diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 1f528cf2d906..6053f5e5c71e 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -1025,38 +1025,29 @@ xfs_bmap_set_attrforkoff( } /* - * Convert inode from non-attributed to attributed. - * Must not be in a transaction, ip must not be locked. + * Convert inode from non-attributed to attributed. Caller must hold the + * ILOCK_EXCL and the file cannot have an attr fork. */ int /* error code */ xfs_bmap_add_attrfork( - xfs_inode_t *ip, /* incore inode pointer */ + struct xfs_trans *tp, + struct xfs_inode *ip, /* incore inode pointer */ int size, /* space new attribute needs */ int rsvd) /* xact may use reserved blks */ { - xfs_mount_t *mp; /* mount structure */ - xfs_trans_t *tp; /* transaction pointer */ - int blks; /* space reservation */ + struct xfs_mount *mp = tp->t_mountp; int version = 1; /* superblock attr version */ int logflags; /* logging flags */ int error; /* error return value */ - mp = ip->i_mount; + xfs_assert_ilocked(ip, XFS_ILOCK_EXCL); ASSERT(!XFS_NOT_DQATTACHED(mp, ip)); - - blks = XFS_ADDAFORK_SPACE_RES(mp); - - error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_addafork, blks, 0, - rsvd, &tp); - if (error) - return error; - if (xfs_inode_has_attr_fork(ip)) - goto trans_cancel; + ASSERT(!xfs_inode_has_attr_fork(ip)); xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); error = xfs_bmap_set_attrforkoff(ip, size, &version); if (error) - goto trans_cancel; + return error; xfs_ifork_init_attr(ip, XFS_DINODE_FMT_EXTENTS, 0); logflags = 0; @@ -1077,7 +1068,7 @@ xfs_bmap_add_attrfork( if (logflags) xfs_trans_log_inode(tp, ip, logflags); if (error) - goto trans_cancel; + return error; if (!xfs_has_attr(mp) || (!xfs_has_attr2(mp) && version == 2)) { bool log_sb = false; @@ -1096,14 +1087,7 @@ xfs_bmap_add_attrfork( xfs_log_sb(tp); } - error = xfs_trans_commit(tp); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - return error; - -trans_cancel: - xfs_trans_cancel(tp); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - return error; + return 0; } /* diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index 32fb2a455c29..e98849eb9bba 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -176,7 +176,8 @@ int xfs_bmap_longest_free_extent(struct xfs_perag *pag, void xfs_trim_extent(struct xfs_bmbt_irec *irec, xfs_fileoff_t bno, xfs_filblks_t len); unsigned int xfs_bmap_compute_attr_offset(struct xfs_mount *mp); -int xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd); +int xfs_bmap_add_attrfork(struct xfs_trans *tp, struct xfs_inode *ip, + int size, int rsvd); void xfs_bmap_local_to_extents_empty(struct xfs_trans *tp, struct xfs_inode *ip, int whichfork); int xfs_bmap_local_to_extents(struct xfs_trans *tp, struct xfs_inode *ip, -- cgit v1.2.3-59-g8ed1b From 6efbbdeb140603351e1413aee79e789bf2279a2b Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:15 -0700 Subject: xfs: add a per-leaf block callback to xchk_xattr_walk Add a second callback function to xchk_xattr_walk so that we can do something in between attr leaf blocks. This will be used by the next patch to see if we should flush cached parent pointer updates to constrain memory usage. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr.c | 2 +- fs/xfs/scrub/dir_repair.c | 2 +- fs/xfs/scrub/listxattr.c | 10 +++++++++- fs/xfs/scrub/listxattr.h | 4 +++- fs/xfs/scrub/nlinks.c | 3 ++- fs/xfs/scrub/parent.c | 7 ++++--- 6 files changed, 20 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index b550f3e34ffc..708334f9b2bd 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -675,7 +675,7 @@ xchk_xattr( * iteration, which doesn't really follow the usual buffer * locking order. */ - error = xchk_xattr_walk(sc, sc->ip, xchk_xattr_actor, NULL); + error = xchk_xattr_walk(sc, sc->ip, xchk_xattr_actor, NULL, NULL); if (!xchk_fblock_process_error(sc, XFS_ATTR_FORK, 0, &error)) return error; diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 60e31da4451e..e968150fe0f0 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -1288,7 +1288,7 @@ xrep_dir_scan_file( goto scan_done; } - error = xchk_xattr_walk(rd->sc, ip, xrep_dir_scan_pptr, rd); + error = xchk_xattr_walk(rd->sc, ip, xrep_dir_scan_pptr, NULL, rd); if (error) goto scan_done; diff --git a/fs/xfs/scrub/listxattr.c b/fs/xfs/scrub/listxattr.c index cbe5911ecbbc..256ff7700c94 100644 --- a/fs/xfs/scrub/listxattr.c +++ b/fs/xfs/scrub/listxattr.c @@ -221,6 +221,7 @@ xchk_xattr_walk_node( struct xfs_scrub *sc, struct xfs_inode *ip, xchk_xattr_fn attr_fn, + xchk_xattrleaf_fn leaf_fn, void *priv) { struct xfs_attr3_icleaf_hdr leafhdr; @@ -252,6 +253,12 @@ xchk_xattr_walk_node( xfs_trans_brelse(sc->tp, leaf_bp); + if (leaf_fn) { + error = leaf_fn(sc, priv); + if (error) + goto out_bitmap; + } + /* Make sure we haven't seen this new leaf already. */ len = 1; if (xdab_bitmap_test(&seen_dablks, leafhdr.forw, &len)) { @@ -288,6 +295,7 @@ xchk_xattr_walk( struct xfs_scrub *sc, struct xfs_inode *ip, xchk_xattr_fn attr_fn, + xchk_xattrleaf_fn leaf_fn, void *priv) { int error; @@ -308,5 +316,5 @@ xchk_xattr_walk( if (xfs_attr_is_leaf(ip)) return xchk_xattr_walk_leaf(sc, ip, attr_fn, priv); - return xchk_xattr_walk_node(sc, ip, attr_fn, priv); + return xchk_xattr_walk_node(sc, ip, attr_fn, leaf_fn, priv); } diff --git a/fs/xfs/scrub/listxattr.h b/fs/xfs/scrub/listxattr.h index 48fe89d05946..703cfb7b14cf 100644 --- a/fs/xfs/scrub/listxattr.h +++ b/fs/xfs/scrub/listxattr.h @@ -11,7 +11,9 @@ typedef int (*xchk_xattr_fn)(struct xfs_scrub *sc, struct xfs_inode *ip, unsigned int namelen, const void *value, unsigned int valuelen, void *priv); +typedef int (*xchk_xattrleaf_fn)(struct xfs_scrub *sc, void *priv); + int xchk_xattr_walk(struct xfs_scrub *sc, struct xfs_inode *ip, - xchk_xattr_fn attr_fn, void *priv); + xchk_xattr_fn attr_fn, xchk_xattrleaf_fn leaf_fn, void *priv); #endif /* __XFS_SCRUB_LISTXATTR_H__ */ diff --git a/fs/xfs/scrub/nlinks.c b/fs/xfs/scrub/nlinks.c index d27b32e6f33d..80aee30886c4 100644 --- a/fs/xfs/scrub/nlinks.c +++ b/fs/xfs/scrub/nlinks.c @@ -434,7 +434,8 @@ xchk_nlinks_collect_dir( goto out_unlock; } - error = xchk_xattr_walk(sc, dp, xchk_nlinks_collect_pptr, xnc); + error = xchk_xattr_walk(sc, dp, xchk_nlinks_collect_pptr, NULL, + xnc); if (error == -ECANCELED) { error = 0; goto out_unlock; diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index 068691434be1..733c410a2279 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -317,7 +317,7 @@ xchk_parent_pptr_and_dotdot( return 0; /* Otherwise, walk the pptrs again, and check. */ - error = xchk_xattr_walk(sc, sc->ip, xchk_parent_scan_dotdot, pp); + error = xchk_xattr_walk(sc, sc->ip, xchk_parent_scan_dotdot, NULL, pp); if (error == -ECANCELED) { /* Found a parent pointer that matches dotdot. */ return 0; @@ -699,7 +699,8 @@ xchk_parent_count_pptrs( */ if (pp->need_revalidate) { pp->pptrs_found = 0; - error = xchk_xattr_walk(sc, sc->ip, xchk_parent_count_pptr, pp); + error = xchk_xattr_walk(sc, sc->ip, xchk_parent_count_pptr, + NULL, pp); if (error == -EFSCORRUPTED) { /* Found a bad parent pointer */ xchk_fblock_set_corrupt(sc, XFS_ATTR_FORK, 0); @@ -758,7 +759,7 @@ xchk_parent_pptr( if (error) goto out_entries; - error = xchk_xattr_walk(sc, sc->ip, xchk_parent_scan_attr, pp); + error = xchk_xattr_walk(sc, sc->ip, xchk_parent_scan_attr, NULL, pp); if (error == -ECANCELED) { error = 0; goto out_names; -- cgit v1.2.3-59-g8ed1b From a26dc21309af68623b82b4e366cbbeb5a85ce65b Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:16 -0700 Subject: xfs: actually rebuild the parent pointer xattrs Once we've assembled all the parent pointers for a file, we need to commit the new dataset atomically to that file. Parent pointer records are embedded in the xattr structure, which means that we must write a new extended attribute structure, again, atomically. Therefore, we must copy the non-parent-pointer attributes from the file being repaired into the temporary file's extended attributes and then call the atomic extent swap mechanism to exchange the blocks. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 2 +- fs/xfs/libxfs/xfs_attr.h | 1 + fs/xfs/scrub/attr_repair.c | 4 +- fs/xfs/scrub/attr_repair.h | 4 + fs/xfs/scrub/findparent.c | 2 + fs/xfs/scrub/parent_repair.c | 709 +++++++++++++++++++++++++++++++++++++++++-- fs/xfs/scrub/trace.h | 2 + 7 files changed, 701 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index ab6ec2f15d76..1c2a27fce08a 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -948,7 +948,7 @@ xfs_attr_lookup( return error; } -STATIC int +int xfs_attr_add_fork( struct xfs_inode *ip, /* incore inode pointer */ int size, /* space new attribute needs */ diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index 43dee4cbaab2..088cb7b30168 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -648,5 +648,6 @@ int __init xfs_attr_intent_init_cache(void); void xfs_attr_intent_destroy_cache(void); int xfs_attr_sf_totsize(struct xfs_inode *dp); +int xfs_attr_add_fork(struct xfs_inode *ip, int size, int rsvd); #endif /* __XFS_ATTR_H__ */ diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index 87f6c9cb54eb..e059813b92b7 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -1030,7 +1030,7 @@ xrep_xattr_reset_fork( * fork. The caller must ILOCK the tempfile and join it to the transaction. * This function returns with the inode joined to a clean scrub transaction. */ -STATIC int +int xrep_xattr_reset_tempfile_fork( struct xfs_scrub *sc) { @@ -1336,7 +1336,7 @@ xrep_xattr_swap_prep( } /* Exchange the temporary file's attribute fork with the one being repaired. */ -STATIC int +int xrep_xattr_swap( struct xfs_scrub *sc, struct xrep_tempexch *tx) diff --git a/fs/xfs/scrub/attr_repair.h b/fs/xfs/scrub/attr_repair.h index 0a9ffa7cfa90..979729bd4a5f 100644 --- a/fs/xfs/scrub/attr_repair.h +++ b/fs/xfs/scrub/attr_repair.h @@ -6,6 +6,10 @@ #ifndef __XFS_SCRUB_ATTR_REPAIR_H__ #define __XFS_SCRUB_ATTR_REPAIR_H__ +struct xrep_tempexch; + +int xrep_xattr_swap(struct xfs_scrub *sc, struct xrep_tempexch *tx); int xrep_xattr_reset_fork(struct xfs_scrub *sc); +int xrep_xattr_reset_tempfile_fork(struct xfs_scrub *sc); #endif /* __XFS_SCRUB_ATTR_REPAIR_H__ */ diff --git a/fs/xfs/scrub/findparent.c b/fs/xfs/scrub/findparent.c index c78422ad757b..01766041ba2c 100644 --- a/fs/xfs/scrub/findparent.c +++ b/fs/xfs/scrub/findparent.c @@ -24,6 +24,7 @@ #include "xfs_trans_space.h" #include "xfs_health.h" #include "xfs_exchmaps.h" +#include "xfs_parent.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" @@ -33,6 +34,7 @@ #include "scrub/findparent.h" #include "scrub/readdir.h" #include "scrub/tempfile.h" +#include "scrub/listxattr.h" /* * Finding the Parent of a Directory diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index 311bc7990d7c..28e9746c0663 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -25,6 +25,8 @@ #include "xfs_health.h" #include "xfs_exchmaps.h" #include "xfs_parent.h" +#include "xfs_attr.h" +#include "xfs_bmap.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" @@ -34,10 +36,13 @@ #include "scrub/findparent.h" #include "scrub/readdir.h" #include "scrub/tempfile.h" +#include "scrub/tempexch.h" #include "scrub/orphanage.h" #include "scrub/xfile.h" #include "scrub/xfarray.h" #include "scrub/xfblob.h" +#include "scrub/attr_repair.h" +#include "scrub/listxattr.h" /* * Repairing The Directory Parent Pointer @@ -65,9 +70,9 @@ * * When salvaging completes, the remaining stashed entries are replayed to the * temporary file. All non-parent pointer extended attributes are copied to - * the temporary file's extended attributes. An atomic extent swap is used to - * commit the new directory blocks to the directory being repaired. This will - * disrupt attrmulti cursors. + * the temporary file's extended attributes. An atomic file mapping exchange + * is used to commit the new xattr blocks to the file being repaired. This + * will disrupt attrmulti cursors. */ /* Create a parent pointer in the tempfile. */ @@ -106,6 +111,23 @@ struct xrep_parent { /* Blobs containing parent pointer names. */ struct xfblob *pptr_names; + /* xattr keys */ + struct xfarray *xattr_records; + + /* xattr values */ + struct xfblob *xattr_blobs; + + /* Scratch buffers for saving extended attributes */ + unsigned char *xattr_name; + void *xattr_value; + unsigned int xattr_value_sz; + + /* + * Information used to exchange the attr fork mappings, if the fs + * supports parent pointers. + */ + struct xrep_tempexch tx; + /* * Information used to scan the filesystem to find the inumber of the * dotdot entry for this directory. On filesystems without parent @@ -117,6 +139,8 @@ struct xrep_parent { * @pscan.lock coordinates access to pptr_recs, pptr_names, pptr, and * pptr_scratch. This reduces the memory requirements of this * structure. + * + * The lock also controls access to xattr_records and xattr_blobs(?) */ struct xrep_parent_scan_info pscan; @@ -129,14 +153,48 @@ struct xrep_parent { /* Scratch buffer for scanning pptr xattrs */ struct xfs_da_args pptr_args; + + /* Have we seen any live updates of parent pointers recently? */ + bool saw_pptr_updates; }; +struct xrep_parent_xattr { + /* Cookie for retrieval of the xattr name. */ + xfblob_cookie name_cookie; + + /* Cookie for retrieval of the xattr value. */ + xfblob_cookie value_cookie; + + /* XFS_ATTR_* flags */ + int flags; + + /* Length of the value and name. */ + uint32_t valuelen; + uint16_t namelen; +}; + +/* + * Stash up to 8 pages of attrs in xattr_records/xattr_blobs before we write + * them to the temp file. + */ +#define XREP_PARENT_XATTR_MAX_STASH_BYTES (PAGE_SIZE * 8) + /* Tear down all the incore stuff we created. */ static void xrep_parent_teardown( struct xrep_parent *rp) { xrep_findparent_scan_teardown(&rp->pscan); + kvfree(rp->xattr_name); + rp->xattr_name = NULL; + kvfree(rp->xattr_value); + rp->xattr_value = NULL; + if (rp->xattr_blobs) + xfblob_destroy(rp->xattr_blobs); + rp->xattr_blobs = NULL; + if (rp->xattr_records) + xfarray_destroy(rp->xattr_records); + rp->xattr_records = NULL; if (rp->pptr_names) xfblob_destroy(rp->pptr_names); rp->pptr_names = NULL; @@ -556,10 +614,11 @@ xrep_parent_scan_dirtree( } /* - * Cancel the empty transaction so that we can (later) use the atomic - * extent swap helpers to lock files and commit the new directory. + * Retake sc->ip's ILOCK now that we're done flushing stashed parent + * pointers. We end this function with an empty transaction and the + * ILOCK. */ - xchk_trans_cancel(rp->sc); + xchk_ilock(rp->sc, XFS_ILOCK_EXCL); return 0; } @@ -594,6 +653,8 @@ xrep_parent_live_update( else error = xrep_parent_stash_parentremove(rp, p->name, p->dp); + if (!error) + rp->saw_pptr_updates = true; mutex_unlock(&rp->pscan.lock); if (error) goto out_abort; @@ -648,6 +709,55 @@ xrep_parent_reset_dotdot( return xfs_trans_roll(&sc->tp); } +/* Pass back the parent inumber if this a parent pointer */ +STATIC int +xrep_parent_lookup_pptr( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + xfs_ino_t *inop = priv; + xfs_ino_t parent_ino; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, &parent_ino, NULL); + if (error) + return error; + + *inop = parent_ino; + return -ECANCELED; +} + +/* + * Find the first parent of the scrub target by walking parent pointers for + * the purpose of deciding if we're going to move it to the orphanage. + * We don't care if the attr fork is zapped. + */ +STATIC int +xrep_parent_lookup_pptrs( + struct xfs_scrub *sc, + xfs_ino_t *inop) +{ + int error; + + *inop = NULLFSINO; + + error = xchk_xattr_walk(sc, sc->ip, xrep_parent_lookup_pptr, NULL, + inop); + if (error && error != -ECANCELED) + return error; + return 0; +} + /* * Move the current file to the orphanage. * @@ -664,14 +774,26 @@ xrep_parent_move_to_orphanage( xfs_ino_t orig_parent, new_parent; int error; - /* - * We are about to drop the ILOCK on sc->ip to lock the orphanage and - * prepare for the adoption. Therefore, look up the old dotdot entry - * for sc->ip so that we can compare it after we re-lock sc->ip. - */ - error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &orig_parent); - if (error) - return error; + if (S_ISDIR(VFS_I(sc->ip)->i_mode)) { + /* + * We are about to drop the ILOCK on sc->ip to lock the + * orphanage and prepare for the adoption. Therefore, look up + * the old dotdot entry for sc->ip so that we can compare it + * after we re-lock sc->ip. + */ + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, + &orig_parent); + if (error) + return error; + } else { + /* + * We haven't dropped the ILOCK since we committed the new + * xattr structure (and hence the new parent pointer records), + * which means that the file cannot have been moved in the + * directory tree, and there are no parents. + */ + orig_parent = NULLFSINO; + } /* * Drop the ILOCK on the scrub target and commit the transaction. @@ -704,9 +826,14 @@ xrep_parent_move_to_orphanage( * Now that we've reacquired the ILOCK on sc->ip, look up the dotdot * entry again. If the parent changed or the child was unlinked while * the child directory was unlocked, we don't need to move the child to - * the orphanage after all. + * the orphanage after all. For a non-directory, we have to scan for + * the first parent pointer to see if one has been added. */ - error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &new_parent); + if (S_ISDIR(VFS_I(sc->ip)->i_mode)) + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, + &new_parent); + else + error = xrep_parent_lookup_pptrs(sc, &new_parent); if (error) return error; @@ -733,6 +860,488 @@ xrep_parent_move_to_orphanage( return 0; } +/* Ensure that the xattr value buffer is large enough. */ +STATIC int +xrep_parent_alloc_xattr_value( + struct xrep_parent *rp, + size_t bufsize) +{ + void *new_val; + + if (rp->xattr_value_sz >= bufsize) + return 0; + + if (rp->xattr_value) { + kvfree(rp->xattr_value); + rp->xattr_value = NULL; + rp->xattr_value_sz = 0; + } + + new_val = kvmalloc(bufsize, XCHK_GFP_FLAGS); + if (!new_val) + return -ENOMEM; + + rp->xattr_value = new_val; + rp->xattr_value_sz = bufsize; + return 0; +} + +/* Retrieve the (remote) value of a non-pptr xattr. */ +STATIC int +xrep_parent_fetch_xattr_remote( + struct xrep_parent *rp, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + unsigned int valuelen) +{ + struct xfs_scrub *sc = rp->sc; + struct xfs_da_args args = { + .attr_filter = attr_flags & XFS_ATTR_NSP_ONDISK_MASK, + .geo = sc->mp->m_attr_geo, + .whichfork = XFS_ATTR_FORK, + .dp = ip, + .name = name, + .namelen = namelen, + .trans = sc->tp, + .valuelen = valuelen, + .owner = ip->i_ino, + }; + int error; + + /* + * If we need a larger value buffer, try to allocate one. If that + * fails, return with -EDEADLOCK to try harder. + */ + error = xrep_parent_alloc_xattr_value(rp, valuelen); + if (error == -ENOMEM) + return -EDEADLOCK; + if (error) + return error; + + args.value = rp->xattr_value; + xfs_attr_sethash(&args); + return xfs_attr_get_ilocked(&args); +} + +/* Stash non-pptr attributes for later replay into the temporary file. */ +STATIC int +xrep_parent_stash_xattr( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xrep_parent_xattr key = { + .valuelen = valuelen, + .namelen = namelen, + .flags = attr_flags & XFS_ATTR_NSP_ONDISK_MASK, + }; + struct xrep_parent *rp = priv; + int error; + + if (attr_flags & (XFS_ATTR_INCOMPLETE | XFS_ATTR_PARENT)) + return 0; + + if (!value) { + error = xrep_parent_fetch_xattr_remote(rp, ip, attr_flags, + name, namelen, valuelen); + if (error) + return error; + + value = rp->xattr_value; + } + + trace_xrep_parent_stash_xattr(rp->sc->tempip, key.flags, (void *)name, + key.namelen, key.valuelen); + + error = xfblob_store(rp->xattr_blobs, &key.name_cookie, name, + key.namelen); + if (error) + return error; + + error = xfblob_store(rp->xattr_blobs, &key.value_cookie, value, + key.valuelen); + if (error) + return error; + + return xfarray_append(rp->xattr_records, &key); +} + +/* Insert one xattr key/value. */ +STATIC int +xrep_parent_insert_xattr( + struct xrep_parent *rp, + const struct xrep_parent_xattr *key) +{ + struct xfs_da_args args = { + .dp = rp->sc->tempip, + .attr_filter = key->flags, + .namelen = key->namelen, + .valuelen = key->valuelen, + .owner = rp->sc->ip->i_ino, + .geo = rp->sc->mp->m_attr_geo, + .whichfork = XFS_ATTR_FORK, + .op_flags = XFS_DA_OP_OKNOENT, + }; + int error; + + ASSERT(!(key->flags & XFS_ATTR_PARENT)); + + /* + * Grab pointers to the scrub buffer so that we can use them to insert + * attrs into the temp file. + */ + args.name = rp->xattr_name; + args.value = rp->xattr_value; + + /* + * The attribute name is stored near the end of the in-core buffer, + * though we reserve one more byte to ensure null termination. + */ + rp->xattr_name[XATTR_NAME_MAX] = 0; + + error = xfblob_load(rp->xattr_blobs, key->name_cookie, rp->xattr_name, + key->namelen); + if (error) + return error; + + error = xfblob_free(rp->xattr_blobs, key->name_cookie); + if (error) + return error; + + error = xfblob_load(rp->xattr_blobs, key->value_cookie, args.value, + key->valuelen); + if (error) + return error; + + error = xfblob_free(rp->xattr_blobs, key->value_cookie); + if (error) + return error; + + rp->xattr_name[key->namelen] = 0; + + trace_xrep_parent_insert_xattr(rp->sc->tempip, key->flags, + rp->xattr_name, key->namelen, key->valuelen); + + xfs_attr_sethash(&args); + return xfs_attr_set(&args, XFS_ATTRUPDATE_UPSERT, false); +} + +/* + * Periodically flush salvaged attributes to the temporary file. This is done + * to reduce the memory requirements of the xattr rebuild because files can + * contain millions of attributes. + */ +STATIC int +xrep_parent_flush_xattrs( + struct xrep_parent *rp) +{ + xfarray_idx_t array_cur; + int error; + + /* + * Entering this function, the scrub context has a reference to the + * inode being repaired, the temporary file, and the empty scrub + * transaction that we created for the xattr scan. We hold ILOCK_EXCL + * on the inode being repaired. + * + * To constrain kernel memory use, we occasionally flush salvaged + * xattrs from the xfarray and xfblob structures into the temporary + * file in preparation for exchanging the xattr structures at the end. + * Updating the temporary file requires a transaction, so we commit the + * scrub transaction and drop the ILOCK so that xfs_attr_set can + * allocate whatever transaction it wants. + * + * We still hold IOLOCK_EXCL on the inode being repaired, which + * prevents anyone from adding xattrs (or parent pointers) while we're + * flushing. + */ + xchk_trans_cancel(rp->sc); + xchk_iunlock(rp->sc, XFS_ILOCK_EXCL); + + /* + * Take the IOLOCK of the temporary file while we modify xattrs. This + * isn't strictly required because the temporary file is never revealed + * to userspace, but we follow the same locking rules. We still hold + * sc->ip's IOLOCK. + */ + error = xrep_tempfile_iolock_polled(rp->sc); + if (error) + return error; + + /* Add all the salvaged attrs to the temporary file. */ + foreach_xfarray_idx(rp->xattr_records, array_cur) { + struct xrep_parent_xattr key; + + error = xfarray_load(rp->xattr_records, array_cur, &key); + if (error) + return error; + + error = xrep_parent_insert_xattr(rp, &key); + if (error) + return error; + } + + /* Empty out both arrays now that we've added the entries. */ + xfarray_truncate(rp->xattr_records); + xfblob_truncate(rp->xattr_blobs); + + xrep_tempfile_iounlock(rp->sc); + + /* Recreate the empty transaction and relock the inode. */ + error = xchk_trans_alloc_empty(rp->sc); + if (error) + return error; + xchk_ilock(rp->sc, XFS_ILOCK_EXCL); + return 0; +} + +/* Decide if we've stashed too much xattr data in memory. */ +static inline bool +xrep_parent_want_flush_xattrs( + struct xrep_parent *rp) +{ + unsigned long long bytes; + + bytes = xfarray_bytes(rp->xattr_records) + + xfblob_bytes(rp->xattr_blobs); + return bytes > XREP_PARENT_XATTR_MAX_STASH_BYTES; +} + +/* Flush staged attributes to the temporary file if we're over the limit. */ +STATIC int +xrep_parent_try_flush_xattrs( + struct xfs_scrub *sc, + void *priv) +{ + struct xrep_parent *rp = priv; + int error; + + if (!xrep_parent_want_flush_xattrs(rp)) + return 0; + + error = xrep_parent_flush_xattrs(rp); + if (error) + return error; + + /* + * If there were any parent pointer updates to the xattr structure + * while we dropped the ILOCK, the xattr structure is now stale. + * Signal to the attr copy process that we need to start over, but + * this time without opportunistic attr flushing. + * + * This is unlikely to happen, so we're ok with restarting the copy. + */ + mutex_lock(&rp->pscan.lock); + if (rp->saw_pptr_updates) + error = -ESTALE; + mutex_unlock(&rp->pscan.lock); + return error; +} + +/* Copy all the non-pptr extended attributes into the temporary file. */ +STATIC int +xrep_parent_copy_xattrs( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + int error; + + /* + * Clear the pptr updates flag. We hold sc->ip ILOCKed, so there + * can't be any parent pointer updates in progress. + */ + mutex_lock(&rp->pscan.lock); + rp->saw_pptr_updates = false; + mutex_unlock(&rp->pscan.lock); + + /* Copy xattrs, stopping periodically to flush the incore buffers. */ + error = xchk_xattr_walk(sc, sc->ip, xrep_parent_stash_xattr, + xrep_parent_try_flush_xattrs, rp); + if (error && error != -ESTALE) + return error; + + if (error == -ESTALE) { + /* + * The xattr copy collided with a parent pointer update. + * Restart the copy, but this time hold the ILOCK all the way + * to the end to lock out any directory parent pointer updates. + */ + error = xchk_xattr_walk(sc, sc->ip, xrep_parent_stash_xattr, + NULL, rp); + if (error) + return error; + } + + /* Flush any remaining stashed xattrs to the temporary file. */ + if (xfarray_bytes(rp->xattr_records) == 0) + return 0; + + return xrep_parent_flush_xattrs(rp); +} + +/* + * Ensure that @sc->ip and @sc->tempip both have attribute forks before we head + * into the attr fork exchange transaction. All files on a filesystem with + * parent pointers must have an attr fork because the parent pointer code does + * not itself add attribute forks. + * + * Note: Unlinkable unlinked files don't need one, but the overhead of having + * an unnecessary attr fork is not justified by the additional code complexity + * that would be needed to track that state correctly. + */ +STATIC int +xrep_parent_ensure_attr_fork( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + int error; + + error = xfs_attr_add_fork(sc->tempip, + sizeof(struct xfs_attr_sf_hdr), 1); + if (error) + return error; + return xfs_attr_add_fork(sc->ip, sizeof(struct xfs_attr_sf_hdr), 1); +} + +/* + * Finish replaying stashed parent pointer updates, allocate a transaction for + * exchanging extent mappings, and take the ILOCKs of both files before we + * commit the new attribute structure. + */ +STATIC int +xrep_parent_finalize_tempfile( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + int error; + + /* + * Repair relies on the ILOCK to quiesce all possible xattr updates. + * Replay all queued parent pointer updates into the tempfile before + * exchanging the contents, even if that means dropping the ILOCKs and + * the transaction. + */ + do { + error = xrep_parent_replay_updates(rp); + if (error) + return error; + + error = xrep_parent_ensure_attr_fork(rp); + if (error) + return error; + + error = xrep_tempexch_trans_alloc(sc, XFS_ATTR_FORK, &rp->tx); + if (error) + return error; + + if (xfarray_length(rp->pptr_recs) == 0) + break; + + xchk_trans_cancel(sc); + xrep_tempfile_iunlock_both(sc); + } while (!xchk_should_terminate(sc, &error)); + return error; +} + +/* + * Replay all the stashed parent pointers into the temporary file, copy all + * the non-pptr xattrs from the file being repaired into the temporary file, + * and exchange the attr fork contents atomically. + */ +STATIC int +xrep_parent_rebuild_pptrs( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + xfs_ino_t parent_ino = NULLFSINO; + int error; + + /* + * Copy non-ppttr xattrs from the file being repaired into the + * temporary file's xattr structure. We hold sc->ip's IOLOCK, which + * prevents setxattr/removexattr calls from occurring, but renames + * update the parent pointers without holding IOLOCK. If we detect + * stale attr structures, we restart the scan but only flush at the + * end. + */ + error = xrep_parent_copy_xattrs(rp); + if (error) + return error; + + /* + * Cancel the empty transaction that we used to walk and copy attrs, + * and drop the ILOCK so that we can take the IOLOCK on the temporary + * file. We still hold sc->ip's IOLOCK. + */ + xchk_trans_cancel(sc); + xchk_iunlock(sc, XFS_ILOCK_EXCL); + + error = xrep_tempfile_iolock_polled(sc); + if (error) + return error; + + /* + * Allocate transaction, lock inodes, and make sure that we've replayed + * all the stashed pptr updates to the tempdir. After this point, + * we're ready to exchange the attr fork mappings. + */ + error = xrep_parent_finalize_tempfile(rp); + if (error) + return error; + + /* Last chance to abort before we start committing pptr fixes. */ + if (xchk_should_terminate(sc, &error)) + return error; + + if (xchk_iscan_aborted(&rp->pscan.iscan)) + return -ECANCELED; + + /* + * Exchange the attr fork contents and junk the old attr fork contents, + * which are now in the tempfile. + */ + error = xrep_xattr_swap(sc, &rp->tx); + if (error) + return error; + error = xrep_xattr_reset_tempfile_fork(sc); + if (error) + return error; + + /* + * Roll to get a transaction without any inodes joined to it. Then we + * can drop the tempfile's ILOCK and IOLOCK before doing more work on + * the scrub target file. + */ + error = xfs_trans_roll(&sc->tp); + if (error) + return error; + xrep_tempfile_iunlock(sc); + xrep_tempfile_iounlock(sc); + + /* + * We've committed the new parent pointers. Find at least one parent + * so that we can decide if we're moving this file to the orphanage. + * For this purpose, root directories are their own parents. + */ + if (sc->ip == sc->mp->m_rootip) { + xrep_findparent_scan_found(&rp->pscan, sc->ip->i_ino); + } else { + error = xrep_parent_lookup_pptrs(sc, &parent_ino); + if (error) + return error; + if (parent_ino != NULLFSINO) + xrep_findparent_scan_found(&rp->pscan, parent_ino); + } + return 0; +} + /* * Commit the new parent pointer structure (currently only the dotdot entry) to * the file that we're repairing. @@ -741,13 +1350,24 @@ STATIC int xrep_parent_rebuild_tree( struct xrep_parent *rp) { + int error; + + if (xfs_has_parent(rp->sc->mp)) { + error = xrep_parent_rebuild_pptrs(rp); + if (error) + return error; + } + if (rp->pscan.parent_ino == NULLFSINO) { if (xrep_orphanage_can_adopt(rp->sc)) return xrep_parent_move_to_orphanage(rp); return -EFSCORRUPTED; } - return xrep_parent_reset_dotdot(rp); + if (S_ISDIR(VFS_I(rp->sc->ip)->i_mode)) + return xrep_parent_reset_dotdot(rp); + + return 0; } /* Set up the filesystem scan so we can look for parents. */ @@ -757,18 +1377,39 @@ xrep_parent_setup_scan( { struct xfs_scrub *sc = rp->sc; char *descr; + struct xfs_da_geometry *geo = sc->mp->m_attr_geo; + int max_len; int error; if (!xfs_has_parent(sc->mp)) return xrep_findparent_scan_start(sc, &rp->pscan); + /* Buffers for copying non-pptr attrs to the tempfile */ + rp->xattr_name = kvmalloc(XATTR_NAME_MAX + 1, XCHK_GFP_FLAGS); + if (!rp->xattr_name) + return -ENOMEM; + + /* + * Allocate enough memory to handle loading local attr values from the + * xfblob data while flushing stashed attrs to the temporary file. + * We only realloc the buffer when salvaging remote attr values, so + * TRY_HARDER means we allocate the maximal attr value size. + */ + if (sc->flags & XCHK_TRY_HARDER) + max_len = XATTR_SIZE_MAX; + else + max_len = xfs_attr_leaf_entsize_local_max(geo->blksize); + error = xrep_parent_alloc_xattr_value(rp, max_len); + if (error) + goto out_xattr_name; + /* Set up some staging memory for logging parent pointer updates. */ descr = xchk_xfile_ino_descr(sc, "parent pointer entries"); error = xfarray_create(descr, 0, sizeof(struct xrep_pptr), &rp->pptr_recs); kfree(descr); if (error) - return error; + goto out_xattr_value; descr = xchk_xfile_ino_descr(sc, "parent pointer names"); error = xfblob_create(descr, &rp->pptr_names); @@ -776,19 +1417,47 @@ xrep_parent_setup_scan( if (error) goto out_recs; + /* Set up some storage for copying attrs before the mapping exchange */ + descr = xchk_xfile_ino_descr(sc, + "parent pointer retained xattr entries"); + error = xfarray_create(descr, 0, sizeof(struct xrep_parent_xattr), + &rp->xattr_records); + kfree(descr); + if (error) + goto out_names; + + descr = xchk_xfile_ino_descr(sc, + "parent pointer retained xattr values"); + error = xfblob_create(descr, &rp->xattr_blobs); + kfree(descr); + if (error) + goto out_attr_keys; + error = __xrep_findparent_scan_start(sc, &rp->pscan, xrep_parent_live_update); if (error) - goto out_names; + goto out_attr_values; return 0; +out_attr_values: + xfblob_destroy(rp->xattr_blobs); + rp->xattr_blobs = NULL; +out_attr_keys: + xfarray_destroy(rp->xattr_records); + rp->xattr_records = NULL; out_names: xfblob_destroy(rp->pptr_names); rp->pptr_names = NULL; out_recs: xfarray_destroy(rp->pptr_recs); rp->pptr_recs = NULL; +out_xattr_value: + kvfree(rp->xattr_value); + rp->xattr_value = NULL; +out_xattr_name: + kvfree(rp->xattr_name); + rp->xattr_name = NULL; return error; } @@ -818,7 +1487,7 @@ xrep_parent( if (error) goto out_teardown; - /* Last chance to abort before we start committing fixes. */ + /* Last chance to abort before we start committing dotdot fixes. */ if (xchk_should_terminate(sc, &error)) goto out_teardown; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 3e0cd482379c..ecfaa4b88910 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -2539,6 +2539,8 @@ DEFINE_EVENT(xrep_xattr_salvage_class, name, \ TP_ARGS(ip, flags, name, namelen, valuelen)) DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_xattr_salvage_rec); DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_xattr_insert_rec); +DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_parent_stash_xattr); +DEFINE_XREP_XATTR_SALVAGE_EVENT(xrep_parent_insert_xattr); DECLARE_EVENT_CLASS(xrep_pptr_salvage_class, TP_PROTO(struct xfs_inode *ip, unsigned int flags, const void *name, -- cgit v1.2.3-59-g8ed1b From 7be3d20bbeda6602d6c5f67ec2b8f189a07ea0e3 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:17 -0700 Subject: xfs: adapt the orphanage code to handle parent pointers Adapt the orphanage's adoption code to update the child file's parent pointers as part of the reparenting process. Also ensure that the child has an attr fork to receive the parent pointer update, since the runtime code assumes one exists. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/orphanage.c | 38 ++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/orphanage.h | 3 +++ fs/xfs/scrub/scrub.c | 2 ++ 3 files changed, 43 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index 94bcc2799188..b2f905924d0d 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -19,6 +19,8 @@ #include "xfs_icache.h" #include "xfs_bmap.h" #include "xfs_bmap_btree.h" +#include "xfs_parent.h" +#include "xfs_attr_sf.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/repair.h" @@ -330,6 +332,8 @@ xrep_adoption_trans_alloc( if (S_ISDIR(VFS_I(sc->ip)->i_mode)) child_blkres = xfs_rename_space_res(mp, 0, false, xfs_name_dotdot.len, false); + if (xfs_has_parent(mp)) + child_blkres += XFS_ADDAFORK_SPACE_RES(mp); adopt->child_blkres = child_blkres; /* @@ -503,6 +507,19 @@ xrep_adoption_zap_dcache( dput(d_orphanage); } +/* + * If we have to add an attr fork ahead of a parent pointer update, how much + * space should we ask for? + */ +static inline int +xrep_adoption_attr_sizeof( + const struct xrep_adoption *adopt) +{ + return sizeof(struct xfs_attr_sf_hdr) + + xfs_attr_sf_entsize_byname(sizeof(struct xfs_parent_rec), + adopt->xname->len); +} + /* * Move the current file to the orphanage under the computed name. * @@ -524,6 +541,19 @@ xrep_adoption_move( if (error) return error; + /* + * If this filesystem has parent pointers, ensure that the file being + * moved to the orphanage has an attribute fork. This is required + * because the parent pointer code does not itself add attr forks. + */ + if (!xfs_inode_has_attr_fork(sc->ip) && xfs_has_parent(sc->mp)) { + int sf_size = xrep_adoption_attr_sizeof(adopt); + + error = xfs_bmap_add_attrfork(sc->tp, sc->ip, sf_size, true); + if (error) + return error; + } + /* Create the new name in the orphanage. */ error = xfs_dir_createname(sc->tp, sc->orphanage, adopt->xname, sc->ip->i_ino, adopt->orphanage_blkres); @@ -548,6 +578,14 @@ xrep_adoption_move( return error; } + /* Add a parent pointer from the file back to the lost+found. */ + if (xfs_has_parent(sc->mp)) { + error = xfs_parent_addname(sc->tp, &adopt->ppargs, + sc->orphanage, adopt->xname, sc->ip); + if (error) + return error; + } + /* * Notify dirent hooks that we moved the file to /lost+found, and * finish all the deferred work so that we know the adoption is fully diff --git a/fs/xfs/scrub/orphanage.h b/fs/xfs/scrub/orphanage.h index 319179ab788d..beb6b686784e 100644 --- a/fs/xfs/scrub/orphanage.h +++ b/fs/xfs/scrub/orphanage.h @@ -54,6 +54,9 @@ struct xrep_adoption { /* Name used for the adoption. */ struct xfs_name *xname; + /* Parent pointer context tracking */ + struct xfs_parent_args ppargs; + /* Block reservations for orphanage and child (if directory). */ unsigned int orphanage_blkres; unsigned int child_blkres; diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index ebb06838c31b..7b1f1abdc7a9 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -19,6 +19,8 @@ #include "xfs_rmap.h" #include "xfs_exchrange.h" #include "xfs_exchmaps.h" +#include "xfs_dir2.h" +#include "xfs_parent.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" -- cgit v1.2.3-59-g8ed1b From 3f50ddbf4b470f8566f99a83597da50180d937c0 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:18 -0700 Subject: xfs: repair link count of nondirectories after rebuilding parent pointers Since the parent pointer scrubber does not exhaustively search the filesystem for missing parent pointers, it doesn't have a good way to determine that there are pointers missing from an otherwise uncorrupt xattr structure. Instead, for nondirectories it employs a heuristic of comparing the file link count to the number of parent pointers found. However, we don't want this heuristic flagging a false corruption after a repair has actually scanned the entire filesystem to rebuild the parent pointers. Therefore, reset the file link count in this one case because we actually know the correct link count. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/parent_repair.c | 107 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 107 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index 28e9746c0663..ee88ce5a12b8 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -27,6 +27,7 @@ #include "xfs_parent.h" #include "xfs_attr.h" #include "xfs_bmap.h" +#include "xfs_ag.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" @@ -156,6 +157,9 @@ struct xrep_parent { /* Have we seen any live updates of parent pointers recently? */ bool saw_pptr_updates; + + /* Number of parents we found after all other repairs */ + unsigned long long parents; }; struct xrep_parent_xattr { @@ -1370,6 +1374,102 @@ xrep_parent_rebuild_tree( return 0; } +/* Count the number of parent pointers. */ +STATIC int +xrep_parent_count_pptr( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xrep_parent *rp = priv; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, NULL, NULL); + if (error) + return error; + + rp->parents++; + return 0; +} + +/* + * After all parent pointer rebuilding and adoption activity completes, reset + * the link count of this nondirectory, having scanned the fs to rebuild all + * parent pointers. + */ +STATIC int +xrep_parent_set_nondir_nlink( + struct xrep_parent *rp) +{ + struct xfs_scrub *sc = rp->sc; + struct xfs_inode *ip = sc->ip; + struct xfs_perag *pag; + bool joined = false; + int error; + + /* Count parent pointers so we can reset the file link count. */ + rp->parents = 0; + error = xchk_xattr_walk(sc, ip, xrep_parent_count_pptr, NULL, rp); + if (error) + return error; + + if (rp->parents > 0 && xfs_inode_on_unlinked_list(ip)) { + xfs_trans_ijoin(sc->tp, sc->ip, 0); + joined = true; + + /* + * The file is on the unlinked list but we found parents. + * Remove the file from the unlinked list. + */ + pag = xfs_perag_get(sc->mp, XFS_INO_TO_AGNO(sc->mp, ip->i_ino)); + if (!pag) { + ASSERT(0); + return -EFSCORRUPTED; + } + + error = xfs_iunlink_remove(sc->tp, pag, ip); + xfs_perag_put(pag); + if (error) + return error; + } else if (rp->parents == 0 && !xfs_inode_on_unlinked_list(ip)) { + xfs_trans_ijoin(sc->tp, sc->ip, 0); + joined = true; + + /* + * The file is not on the unlinked list but we found no + * parents. Add the file to the unlinked list. + */ + error = xfs_iunlink(sc->tp, ip); + if (error) + return error; + } + + /* Set the correct link count. */ + if (VFS_I(ip)->i_nlink != rp->parents) { + if (!joined) { + xfs_trans_ijoin(sc->tp, sc->ip, 0); + joined = true; + } + + set_nlink(VFS_I(ip), min_t(unsigned long long, rp->parents, + XFS_NLINK_PINNED)); + } + + /* Log the inode to keep it moving forward if we dirtied anything. */ + if (joined) + xfs_trans_log_inode(sc->tp, ip, XFS_ILOG_CORE); + return 0; +} + /* Set up the filesystem scan so we can look for parents. */ STATIC int xrep_parent_setup_scan( @@ -1494,6 +1594,13 @@ xrep_parent( error = xrep_parent_rebuild_tree(rp); if (error) goto out_teardown; + if (xfs_has_parent(sc->mp) && !S_ISDIR(VFS_I(sc->ip)->i_mode)) { + error = xrep_parent_set_nondir_nlink(rp); + if (error) + goto out_teardown; + } + + error = xrep_defer_finish(sc); out_teardown: xrep_parent_teardown(rp); -- cgit v1.2.3-59-g8ed1b From 327ed702d84034879572812f580cb769848af7ae Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:19 -0700 Subject: xfs: inode repair should ensure there's an attr fork to store parent pointers The runtime parent pointer update code expects that any file being moved around the directory tree already has an attr fork. However, if we had to rebuild an inode core record, there's a chance that we zeroed forkoff as part of the inode to pass the iget verifiers. Therefore, if we performed any repairs on an inode core, ensure that the inode has a nonzero forkoff before unlocking the inode. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/inode_repair.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c index e3b74ea50fde..daf9f1ee7c2c 100644 --- a/fs/xfs/scrub/inode_repair.c +++ b/fs/xfs/scrub/inode_repair.c @@ -1736,6 +1736,44 @@ xrep_inode_extsize( } } +/* Ensure this file has an attr fork if it needs to hold a parent pointer. */ +STATIC int +xrep_inode_pptr( + struct xfs_scrub *sc) +{ + struct xfs_mount *mp = sc->mp; + struct xfs_inode *ip = sc->ip; + struct inode *inode = VFS_I(ip); + + if (!xfs_has_parent(mp)) + return 0; + + /* + * Unlinked inodes that cannot be added to the directory tree will not + * have a parent pointer. + */ + if (inode->i_nlink == 0 && !(inode->i_state & I_LINKABLE)) + return 0; + + /* The root directory doesn't have a parent pointer. */ + if (ip == mp->m_rootip) + return 0; + + /* + * Metadata inodes are rooted in the superblock and do not have any + * parents. + */ + if (xfs_is_metadata_inode(ip)) + return 0; + + /* Inode already has an attr fork; no further work possible here. */ + if (xfs_inode_has_attr_fork(ip)) + return 0; + + return xfs_bmap_add_attrfork(sc->tp, ip, + sizeof(struct xfs_attr_sf_hdr), true); +} + /* Fix any irregularities in an inode that the verifiers don't catch. */ STATIC int xrep_inode_problems( @@ -1744,6 +1782,9 @@ xrep_inode_problems( int error; error = xrep_inode_blockcounts(sc); + if (error) + return error; + error = xrep_inode_pptr(sc); if (error) return error; xrep_inode_timestamps(sc->ip); -- cgit v1.2.3-59-g8ed1b From 928b721a11789a9363d6d7c32a1f3166a79f3b5f Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:20 -0700 Subject: xfs: teach online scrub to find directory tree structure problems Create a new scrubber that detects corruptions within the directory tree structure itself. It can detect directories with multiple parents; loops within the directory tree; and directory loops not accessible from the root. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/libxfs/xfs_fs.h | 3 +- fs/xfs/scrub/common.h | 1 + fs/xfs/scrub/dirtree.c | 795 ++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/dirtree.h | 129 ++++++++ fs/xfs/scrub/ino_bitmap.h | 37 +++ fs/xfs/scrub/scrub.c | 7 + fs/xfs/scrub/scrub.h | 1 + fs/xfs/scrub/stats.c | 1 + fs/xfs/scrub/trace.c | 4 + fs/xfs/scrub/trace.h | 190 ++++++++++- fs/xfs/scrub/xfarray.h | 1 + 12 files changed, 1168 insertions(+), 2 deletions(-) create mode 100644 fs/xfs/scrub/dirtree.c create mode 100644 fs/xfs/scrub/dirtree.h create mode 100644 fs/xfs/scrub/ino_bitmap.h (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index af99a455ce4d..8ec0dd257a98 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -163,6 +163,7 @@ xfs-y += $(addprefix scrub/, \ common.o \ dabtree.o \ dir.o \ + dirtree.o \ fscounters.o \ health.o \ ialloc.o \ diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index dd13bfa500f2..85f2a7e20ac0 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -719,9 +719,10 @@ struct xfs_scrub_metadata { #define XFS_SCRUB_TYPE_QUOTACHECK 25 /* quota counters */ #define XFS_SCRUB_TYPE_NLINKS 26 /* inode link counts */ #define XFS_SCRUB_TYPE_HEALTHY 27 /* everything checked out ok */ +#define XFS_SCRUB_TYPE_DIRTREE 28 /* directory tree structure */ /* Number of scrub subcommands. */ -#define XFS_SCRUB_TYPE_NR 28 +#define XFS_SCRUB_TYPE_NR 29 /* i: Repair this metadata. */ #define XFS_SCRUB_IFLAG_REPAIR (1u << 0) diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h index e00466f40482..39465e39dc5f 100644 --- a/fs/xfs/scrub/common.h +++ b/fs/xfs/scrub/common.h @@ -92,6 +92,7 @@ int xchk_setup_directory(struct xfs_scrub *sc); int xchk_setup_xattr(struct xfs_scrub *sc); int xchk_setup_symlink(struct xfs_scrub *sc); int xchk_setup_parent(struct xfs_scrub *sc); +int xchk_setup_dirtree(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xchk_setup_rtbitmap(struct xfs_scrub *sc); int xchk_setup_rtsummary(struct xfs_scrub *sc); diff --git a/fs/xfs/scrub/dirtree.c b/fs/xfs/scrub/dirtree.c new file mode 100644 index 000000000000..807c8a8ca7d4 --- /dev/null +++ b/fs/xfs/scrub/dirtree.c @@ -0,0 +1,795 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2023-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_inode.h" +#include "xfs_icache.h" +#include "xfs_dir2.h" +#include "xfs_dir2_priv.h" +#include "xfs_attr.h" +#include "xfs_parent.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/bitmap.h" +#include "scrub/ino_bitmap.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" +#include "scrub/listxattr.h" +#include "scrub/trace.h" +#include "scrub/dirtree.h" + +/* + * Directory Tree Structure Validation + * =================================== + * + * Validating the tree qualities of the directory tree structure can be + * difficult. If the tree is frozen, running a depth (or breadth) first search + * and marking a bitmap suffices to determine if there is a cycle. XORing the + * mark bitmap with the inode bitmap afterwards tells us if there are + * disconnected cycles. If the tree is not frozen, directory updates can move + * subtrees across the scanner wavefront, which complicates the design greatly. + * + * Directory parent pointers change that by enabling an incremental approach to + * validation of the tree structure. Instead of using one thread to scan the + * entire filesystem, we instead can have multiple threads walking individual + * subdirectories upwards to the root. In a perfect world, the IOLOCK would + * suffice to stabilize two directories in a parent -> child relationship. + * Unfortunately, the VFS does not take the IOLOCK when moving a child + * subdirectory, so we instead synchronize on ILOCK and use dirent update hooks + * to detect a race. If a race occurs in a path, we restart the scan. + * + * If the walk terminates without reaching the root, we know the path is + * disconnected and ought to be attached to the lost and found. If on the walk + * we find the same subdir that we're scanning, we know this is a cycle and + * should delete an incoming edge. If we find multiple paths to the root, we + * know to delete an incoming edge. + * + * There are two big hitches with this approach: first, all file link counts + * must be correct to prevent other writers from doing the wrong thing with the + * directory tree structure. Second, because we're walking upwards in a tree + * of arbitrary depth, we cannot hold all the ILOCKs. Instead, we will use a + * directory update hook to invalidate the scan results if one of the paths + * we've scanned has changed. + */ + +/* Clean up the dirtree checking resources. */ +STATIC void +xchk_dirtree_buf_cleanup( + void *buf) +{ + struct xchk_dirtree *dl = buf; + struct xchk_dirpath *path, *n; + + xchk_dirtree_for_each_path_safe(dl, path, n) { + list_del_init(&path->list); + xino_bitmap_destroy(&path->seen_inodes); + kfree(path); + } + + xfblob_destroy(dl->path_names); + xfarray_destroy(dl->path_steps); + mutex_destroy(&dl->lock); +} + +/* Set us up to look for directory loops. */ +int +xchk_setup_dirtree( + struct xfs_scrub *sc) +{ + struct xchk_dirtree *dl; + char *descr; + int error; + + dl = kvzalloc(sizeof(struct xchk_dirtree), XCHK_GFP_FLAGS); + if (!dl) + return -ENOMEM; + dl->sc = sc; + dl->xname.name = dl->namebuf; + INIT_LIST_HEAD(&dl->path_list); + dl->root_ino = NULLFSINO; + + mutex_init(&dl->lock); + + descr = xchk_xfile_ino_descr(sc, "dirtree path steps"); + error = xfarray_create(descr, 0, sizeof(struct xchk_dirpath_step), + &dl->path_steps); + kfree(descr); + if (error) + goto out_dl; + + descr = xchk_xfile_ino_descr(sc, "dirtree path names"); + error = xfblob_create(descr, &dl->path_names); + kfree(descr); + if (error) + goto out_steps; + + error = xchk_setup_inode_contents(sc, 0); + if (error) + goto out_names; + + sc->buf = dl; + sc->buf_cleanup = xchk_dirtree_buf_cleanup; + return 0; + +out_names: + xfblob_destroy(dl->path_names); +out_steps: + xfarray_destroy(dl->path_steps); +out_dl: + mutex_destroy(&dl->lock); + kvfree(dl); + return error; +} + +/* + * Add the parent pointer described by @dl->pptr to the given path as a new + * step. Returns -ELNRNG if the path is too deep. + */ +STATIC int +xchk_dirpath_append( + struct xchk_dirtree *dl, + struct xfs_inode *ip, + struct xchk_dirpath *path, + const struct xfs_name *name, + const struct xfs_parent_rec *pptr) +{ + struct xchk_dirpath_step step = { + .pptr_rec = *pptr, /* struct copy */ + .name_len = name->len, + }; + int error; + + /* + * If this path is more than 2 billion steps long, this directory tree + * is too far gone to fix. + */ + if (path->nr_steps >= XFS_MAXLINK) + return -ELNRNG; + + error = xfblob_storename(dl->path_names, &step.name_cookie, name); + if (error) + return error; + + error = xino_bitmap_set(&path->seen_inodes, ip->i_ino); + if (error) + return error; + + error = xfarray_append(dl->path_steps, &step); + if (error) + return error; + + path->nr_steps++; + return 0; +} + +/* + * Create an xchk_path for each parent pointer of the directory that we're + * scanning. For each path created, we will eventually try to walk towards the + * root with the goal of deleting all parents except for one that leads to the + * root. + * + * Returns -EFSCORRUPTED to signal that the inode being scanned has a corrupt + * parent pointer and hence there's no point in continuing; or -ENOSR if there + * are too many parent pointers for this directory. + */ +STATIC int +xchk_dirtree_create_path( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xfs_name xname = { + .name = name, + .len = namelen, + }; + struct xchk_dirtree *dl = priv; + struct xchk_dirpath *path; + const struct xfs_parent_rec *rec = value; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, NULL, NULL); + if (error) + return error; + + /* + * If there are more than 2 billion actual parent pointers for this + * subdirectory, this fs is too far gone to fix. + */ + if (dl->nr_paths >= XFS_MAXLINK) + return -ENOSR; + + trace_xchk_dirtree_create_path(sc, ip, dl->nr_paths, &xname, rec); + + /* + * Create a new xchk_path structure to remember this parent pointer + * and record the first name step. + */ + path = kmalloc(sizeof(struct xchk_dirpath), XCHK_GFP_FLAGS); + if (!path) + return -ENOMEM; + + INIT_LIST_HEAD(&path->list); + xino_bitmap_init(&path->seen_inodes); + path->nr_steps = 0; + path->outcome = XCHK_DIRPATH_SCANNING; + + error = xchk_dirpath_append(dl, sc->ip, path, &xname, rec); + if (error) + goto out_path; + + path->first_step = xfarray_length(dl->path_steps) - 1; + path->second_step = XFARRAY_NULLIDX; + path->path_nr = dl->nr_paths; + + list_add_tail(&path->list, &dl->path_list); + dl->nr_paths++; + return 0; +out_path: + kfree(path); + return error; +} + +/* + * Validate that the first step of this path still has a corresponding + * parent pointer in @sc->ip. We probably dropped @sc->ip's ILOCK while + * walking towards the roots, which is why this is necessary. + * + * This function has a side effect of loading the first parent pointer of this + * path into the parent pointer scratch pad. This prepares us to walk up the + * directory tree towards the root. Returns -ESTALE if the scan data is now + * out of date. + */ +STATIC int +xchk_dirpath_revalidate( + struct xchk_dirtree *dl, + struct xchk_dirpath *path) +{ + struct xfs_scrub *sc = dl->sc; + int error; + + /* + * Look up the parent pointer that corresponds to the start of this + * path. If the parent pointer has disappeared on us, dump all the + * scan results and try again. + */ + error = xfs_parent_lookup(sc->tp, sc->ip, &dl->xname, &dl->pptr_rec, + &dl->pptr_args); + if (error == -ENOATTR) { + trace_xchk_dirpath_disappeared(dl->sc, sc->ip, path->path_nr, + path->first_step, &dl->xname, &dl->pptr_rec); + dl->stale = true; + return -ESTALE; + } + + return error; +} + +/* + * Walk the parent pointers of a directory at the end of a path and record + * the parent that we find in @dl->xname/pptr_rec. + */ +STATIC int +xchk_dirpath_find_next_step( + struct xfs_scrub *sc, + struct xfs_inode *ip, + unsigned int attr_flags, + const unsigned char *name, + unsigned int namelen, + const void *value, + unsigned int valuelen, + void *priv) +{ + struct xchk_dirtree *dl = priv; + const struct xfs_parent_rec *rec = value; + int error; + + if (!(attr_flags & XFS_ATTR_PARENT)) + return 0; + + error = xfs_parent_from_attr(sc->mp, attr_flags, name, namelen, value, + valuelen, NULL, NULL); + if (error) + return error; + + /* + * If we've already set @dl->pptr_rec, then this directory has multiple + * parents. Signal this back to the caller via -EMLINK. + */ + if (dl->parents_found > 0) + return -EMLINK; + + dl->parents_found++; + memcpy(dl->namebuf, name, namelen); + dl->xname.len = namelen; + dl->pptr_rec = *rec; /* struct copy */ + return 0; +} + +/* Set and log the outcome of a path walk. */ +static inline void +xchk_dirpath_set_outcome( + struct xchk_dirtree *dl, + struct xchk_dirpath *path, + enum xchk_dirpath_outcome outcome) +{ + trace_xchk_dirpath_set_outcome(dl->sc, path->path_nr, path->nr_steps, + outcome); + + path->outcome = outcome; +} + +/* + * Scan the directory at the end of this path for its parent directory link. + * If we find one, extend the path. Returns -ESTALE if the scan data out of + * date. Returns -EFSCORRUPTED if the parent pointer is bad; or -ELNRNG if + * the path got too deep. + */ +STATIC int +xchk_dirpath_step_up( + struct xchk_dirtree *dl, + struct xchk_dirpath *path) +{ + struct xfs_scrub *sc = dl->sc; + struct xfs_inode *dp; + xfs_ino_t parent_ino = be64_to_cpu(dl->pptr_rec.p_ino); + unsigned int lock_mode; + int error; + + /* Grab and lock the parent directory. */ + error = xchk_iget(sc, parent_ino, &dp); + if (error) + return error; + + lock_mode = xfs_ilock_attr_map_shared(dp); + mutex_lock(&dl->lock); + + if (dl->stale) { + error = -ESTALE; + goto out_scanlock; + } + + /* We've reached the root directory; the path is ok. */ + if (parent_ino == dl->root_ino) { + xchk_dirpath_set_outcome(dl, path, XCHK_DIRPATH_OK); + error = 0; + goto out_scanlock; + } + + /* + * The inode being scanned is its own distant ancestor! Get rid of + * this path. + */ + if (parent_ino == sc->ip->i_ino) { + xchk_dirpath_set_outcome(dl, path, XCHK_DIRPATH_DELETE); + error = 0; + goto out_scanlock; + } + + /* + * We've seen this inode before during the path walk. There's a loop + * above us in the directory tree. This probably means that we cannot + * continue, but let's keep walking paths to get a full picture. + */ + if (xino_bitmap_test(&path->seen_inodes, parent_ino)) { + xchk_dirpath_set_outcome(dl, path, XCHK_DIRPATH_LOOP); + error = 0; + goto out_scanlock; + } + + /* The handle encoded in the parent pointer must match. */ + if (VFS_I(dp)->i_generation != be32_to_cpu(dl->pptr_rec.p_gen)) { + trace_xchk_dirpath_badgen(dl->sc, dp, path->path_nr, + path->nr_steps, &dl->xname, &dl->pptr_rec); + error = -EFSCORRUPTED; + goto out_scanlock; + } + + /* Parent pointer must point up to a directory. */ + if (!S_ISDIR(VFS_I(dp)->i_mode)) { + trace_xchk_dirpath_nondir_parent(dl->sc, dp, path->path_nr, + path->nr_steps, &dl->xname, &dl->pptr_rec); + error = -EFSCORRUPTED; + goto out_scanlock; + } + + /* Parent cannot be an unlinked directory. */ + if (VFS_I(dp)->i_nlink == 0) { + trace_xchk_dirpath_unlinked_parent(dl->sc, dp, path->path_nr, + path->nr_steps, &dl->xname, &dl->pptr_rec); + error = -EFSCORRUPTED; + goto out_scanlock; + } + + /* + * If the extended attributes look as though they has been zapped by + * the inode record repair code, we cannot scan for parent pointers. + */ + if (xchk_pptr_looks_zapped(dp)) { + error = -EBUSY; + xchk_set_incomplete(sc); + goto out_scanlock; + } + + /* + * Walk the parent pointers of @dp to find the parent of this directory + * to find the next step in our walk. If we find that @dp has exactly + * one parent, the parent pointer information will be stored in + * @dl->pptr_rec. This prepares us for the next step of the walk. + */ + mutex_unlock(&dl->lock); + dl->parents_found = 0; + error = xchk_xattr_walk(sc, dp, xchk_dirpath_find_next_step, NULL, dl); + mutex_lock(&dl->lock); + if (error == -EFSCORRUPTED || error == -EMLINK || + (!error && dl->parents_found == 0)) { + /* + * Further up the directory tree from @sc->ip, we found a + * corrupt parent pointer, multiple parent pointers while + * finding this directory's parent, or zero parents despite + * having a nonzero link count. Keep looking for other paths. + */ + xchk_dirpath_set_outcome(dl, path, XCHK_DIRPATH_CORRUPT); + error = 0; + goto out_scanlock; + } + if (error) + goto out_scanlock; + + if (dl->stale) { + error = -ESTALE; + goto out_scanlock; + } + + trace_xchk_dirpath_found_next_step(sc, dp, path->path_nr, + path->nr_steps, &dl->xname, &dl->pptr_rec); + + /* Append to the path steps */ + error = xchk_dirpath_append(dl, dp, path, &dl->xname, &dl->pptr_rec); + if (error) + goto out_scanlock; + + if (path->second_step == XFARRAY_NULLIDX) + path->second_step = xfarray_length(dl->path_steps) - 1; + +out_scanlock: + mutex_unlock(&dl->lock); + xfs_iunlock(dp, lock_mode); + xchk_irele(sc, dp); + return error; +} + +/* + * Walk the directory tree upwards towards what is hopefully the root + * directory, recording path steps as we go. The current path components are + * stored in dl->pptr_rec and dl->xname. + * + * Returns -ESTALE if the scan data are out of date. Returns -EFSCORRUPTED + * only if the direct parent pointer of @sc->ip associated with this path is + * corrupt. + */ +STATIC int +xchk_dirpath_walk_upwards( + struct xchk_dirtree *dl, + struct xchk_dirpath *path) +{ + struct xfs_scrub *sc = dl->sc; + int error; + + ASSERT(sc->ilock_flags & XFS_ILOCK_EXCL); + + /* Reload the start of this path and make sure it's still there. */ + error = xchk_dirpath_revalidate(dl, path); + if (error) + return error; + + trace_xchk_dirpath_walk_upwards(sc, sc->ip, path->path_nr, &dl->xname, + &dl->pptr_rec); + + /* + * The inode being scanned is its own direct ancestor! + * Get rid of this path. + */ + if (be64_to_cpu(dl->pptr_rec.p_ino) == sc->ip->i_ino) { + xchk_dirpath_set_outcome(dl, path, XCHK_DIRPATH_DELETE); + return 0; + } + + /* + * Drop ILOCK_EXCL on the inode being scanned. We still hold + * IOLOCK_EXCL on it, so it cannot move around or be renamed. + * + * Beyond this point we're walking up the directory tree, which means + * that we can acquire and drop the ILOCK on an alias of sc->ip. The + * ILOCK state is no longer tracked in the scrub context. Hence we + * must drop @sc->ip's ILOCK during the walk. + */ + mutex_unlock(&dl->lock); + xchk_iunlock(sc, XFS_ILOCK_EXCL); + + /* + * Take the first step in the walk towards the root by checking the + * start of this path, which is a direct parent pointer of @sc->ip. + * If we see any kind of error here (including corruptions), the parent + * pointer of @sc->ip is corrupt. Stop the whole scan. + */ + error = xchk_dirpath_step_up(dl, path); + if (error) { + xchk_ilock(sc, XFS_ILOCK_EXCL); + mutex_lock(&dl->lock); + return error; + } + + /* + * Take steps upward from the second step in this path towards the + * root. If we hit corruption errors here, there's a problem + * *somewhere* in the path, but we don't need to stop scanning. + */ + while (!error && path->outcome == XCHK_DIRPATH_SCANNING) + error = xchk_dirpath_step_up(dl, path); + + /* Retake the locks we had, mark paths, etc. */ + xchk_ilock(sc, XFS_ILOCK_EXCL); + mutex_lock(&dl->lock); + if (error == -EFSCORRUPTED) { + xchk_dirpath_set_outcome(dl, path, XCHK_DIRPATH_CORRUPT); + error = 0; + } + if (!error && dl->stale) + return -ESTALE; + return error; +} + +/* Delete all the collected path information. */ +STATIC void +xchk_dirtree_reset( + void *buf) +{ + struct xchk_dirtree *dl = buf; + struct xchk_dirpath *path, *n; + + ASSERT(dl->sc->ilock_flags & XFS_ILOCK_EXCL); + + xchk_dirtree_for_each_path_safe(dl, path, n) { + list_del_init(&path->list); + xino_bitmap_destroy(&path->seen_inodes); + kfree(path); + } + dl->nr_paths = 0; + + xfarray_truncate(dl->path_steps); + xfblob_truncate(dl->path_names); + + dl->stale = false; +} + +/* + * Load the name/pptr from the first step in this path into @dl->pptr_rec and + * @dl->xname. + */ +STATIC int +xchk_dirtree_load_path( + struct xchk_dirtree *dl, + struct xchk_dirpath *path) +{ + struct xchk_dirpath_step step; + int error; + + error = xfarray_load(dl->path_steps, path->first_step, &step); + if (error) + return error; + + error = xfblob_loadname(dl->path_names, step.name_cookie, &dl->xname, + step.name_len); + if (error) + return error; + + dl->pptr_rec = step.pptr_rec; /* struct copy */ + return 0; +} + +/* + * For each parent pointer of this subdir, trace a path upwards towards the + * root directory and record what we find. Returns 0 for success; + * -EFSCORRUPTED if walking the parent pointers of @sc->ip failed, -ELNRNG if a + * path was too deep; -ENOSR if there were too many parent pointers; or + * a negative errno. + */ +STATIC int +xchk_dirtree_find_paths_to_root( + struct xchk_dirtree *dl) +{ + struct xfs_scrub *sc = dl->sc; + struct xchk_dirpath *path; + int error = 0; + + do { + if (xchk_should_terminate(sc, &error)) + return error; + + xchk_dirtree_reset(dl); + + /* + * If the extended attributes look as though they has been + * zapped by the inode record repair code, we cannot scan for + * parent pointers. + */ + if (xchk_pptr_looks_zapped(sc->ip)) { + xchk_set_incomplete(sc); + return -EBUSY; + } + + /* + * Create path walk contexts for each parent of the directory + * that is being scanned. Directories are supposed to have + * only one parent, but this is how we detect multiple parents. + */ + error = xchk_xattr_walk(sc, sc->ip, xchk_dirtree_create_path, + NULL, dl); + if (error) + return error; + + xchk_dirtree_for_each_path(dl, path) { + /* Load path components into dl->pptr/xname */ + error = xchk_dirtree_load_path(dl, path); + if (error) + return error; + + /* + * Try to walk up each path to the root. This enables + * us to find directory loops in ancestors, and the + * like. + */ + error = xchk_dirpath_walk_upwards(dl, path); + if (error == -EFSCORRUPTED) { + /* + * A parent pointer of @sc->ip is bad, don't + * bother continuing. + */ + break; + } + if (error == -ESTALE) { + /* This had better be an invalidation. */ + ASSERT(dl->stale); + break; + } + if (error) + return error; + } + } while (dl->stale); + + return error; +} + +/* + * Figure out what to do with the paths we tried to find. Do not call this + * if the scan results are stale. + */ +STATIC void +xchk_dirtree_evaluate( + struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc) +{ + struct xchk_dirpath *path; + + ASSERT(!dl->stale); + + /* Scan the paths we have to decide what to do. */ + memset(oc, 0, sizeof(struct xchk_dirtree_outcomes)); + xchk_dirtree_for_each_path(dl, path) { + trace_xchk_dirpath_evaluate_path(dl->sc, path->path_nr, + path->nr_steps, path->outcome); + + switch (path->outcome) { + case XCHK_DIRPATH_SCANNING: + /* shouldn't get here */ + ASSERT(0); + break; + case XCHK_DIRPATH_DELETE: + /* This one is already going away. */ + oc->bad++; + break; + case XCHK_DIRPATH_CORRUPT: + case XCHK_DIRPATH_LOOP: + /* Couldn't find the end of this path. */ + oc->suspect++; + break; + case XCHK_DIRPATH_STALE: + /* shouldn't get here either */ + ASSERT(0); + break; + case XCHK_DIRPATH_OK: + /* This path got all the way to the root. */ + oc->good++; + break; + } + } + + trace_xchk_dirtree_evaluate(dl, oc); +} + +/* Look for directory loops. */ +int +xchk_dirtree( + struct xfs_scrub *sc) +{ + struct xchk_dirtree_outcomes oc; + struct xchk_dirtree *dl = sc->buf; + int error; + + /* + * Nondirectories do not point downwards to other files, so they cannot + * cause a cycle in the directory tree. + */ + if (!S_ISDIR(VFS_I(sc->ip)->i_mode)) + return -ENOENT; + + ASSERT(xfs_has_parent(sc->mp)); + + /* Find the root of the directory tree. */ + dl->root_ino = sc->mp->m_rootip->i_ino; + + trace_xchk_dirtree_start(sc->ip, sc->sm, 0); + + mutex_lock(&dl->lock); + + /* Trace each parent pointer's path to the root. */ + error = xchk_dirtree_find_paths_to_root(dl); + if (error == -EFSCORRUPTED || error == -ELNRNG || error == -ENOSR) { + /* + * Don't bother walking the paths if the xattr structure or the + * parent pointers are corrupt; this scan cannot be completed + * without full information. + */ + xchk_ino_xref_set_corrupt(sc, sc->ip->i_ino); + error = 0; + goto out_scanlock; + } + if (error == -EBUSY) { + /* + * We couldn't scan some directory's parent pointers because + * the attr fork looked like it had been zapped. The + * scan was marked incomplete, so no further error code + * is necessary. + */ + error = 0; + goto out_scanlock; + } + if (error) + goto out_scanlock; + + /* Assess what we found in our path evaluation. */ + xchk_dirtree_evaluate(dl, &oc); + if (xchk_dirtree_parentless(dl)) { + if (oc.good || oc.bad || oc.suspect) + xchk_ino_set_corrupt(sc, sc->ip->i_ino); + } else { + if (oc.bad || oc.good + oc.suspect != 1) + xchk_ino_set_corrupt(sc, sc->ip->i_ino); + if (oc.suspect) + xchk_ino_xref_set_corrupt(sc, sc->ip->i_ino); + } + +out_scanlock: + mutex_unlock(&dl->lock); + trace_xchk_dirtree_done(sc->ip, sc->sm, error); + return error; +} diff --git a/fs/xfs/scrub/dirtree.h b/fs/xfs/scrub/dirtree.h new file mode 100644 index 000000000000..50fefd64ae50 --- /dev/null +++ b/fs/xfs/scrub/dirtree.h @@ -0,0 +1,129 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2023-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_DIRTREE_H__ +#define __XFS_SCRUB_DIRTREE_H__ + +/* + * Each of these represents one parent pointer path step in a chain going + * up towards the directory tree root. These are stored inside an xfarray. + */ +struct xchk_dirpath_step { + /* Directory entry name associated with this parent link. */ + xfblob_cookie name_cookie; + unsigned int name_len; + + /* Handle of the parent directory. */ + struct xfs_parent_rec pptr_rec; +}; + +enum xchk_dirpath_outcome { + XCHK_DIRPATH_SCANNING = 0, /* still being put together */ + XCHK_DIRPATH_DELETE, /* delete this path */ + XCHK_DIRPATH_CORRUPT, /* corruption detected in path */ + XCHK_DIRPATH_LOOP, /* cycle detected further up */ + XCHK_DIRPATH_STALE, /* path is stale */ + XCHK_DIRPATH_OK, /* path reaches the root */ +}; + +/* + * Each of these represents one parent pointer path out of the directory being + * scanned. These exist in-core, and hopefully there aren't more than a + * handful of them. + */ +struct xchk_dirpath { + struct list_head list; + + /* Index of the first step in this path. */ + xfarray_idx_t first_step; + + /* Index of the second step in this path. */ + xfarray_idx_t second_step; + + /* Inodes seen while walking this path. */ + struct xino_bitmap seen_inodes; + + /* Number of steps in this path. */ + unsigned int nr_steps; + + /* Which path is this? */ + unsigned int path_nr; + + /* What did we conclude from following this path? */ + enum xchk_dirpath_outcome outcome; +}; + +struct xchk_dirtree_outcomes { + /* Number of XCHK_DIRPATH_DELETE */ + unsigned int bad; + + /* Number of XCHK_DIRPATH_CORRUPT or XCHK_DIRPATH_LOOP */ + unsigned int suspect; + + /* Number of XCHK_DIRPATH_OK */ + unsigned int good; +}; + +struct xchk_dirtree { + struct xfs_scrub *sc; + + /* Root inode that we're looking for. */ + xfs_ino_t root_ino; + + /* Scratch buffer for scanning pptr xattrs */ + struct xfs_parent_rec pptr_rec; + struct xfs_da_args pptr_args; + + /* Name buffer */ + struct xfs_name xname; + char namebuf[MAXNAMELEN]; + + /* lock for everything below here */ + struct mutex lock; + + /* + * All path steps observed during this scan. Each of the path + * steps for a particular pathwalk are recorded in sequential + * order in the xfarray. A pathwalk ends either with a step + * pointing to the root directory (success) or pointing to NULLFSINO + * (loop detected, empty dir detected, etc). + */ + struct xfarray *path_steps; + + /* All names observed during this scan. */ + struct xfblob *path_names; + + /* All paths being tracked by this scanner. */ + struct list_head path_list; + + /* Number of paths in path_list. */ + unsigned int nr_paths; + + /* Number of parents found by a pptr scan. */ + unsigned int parents_found; + + /* Have the path data been invalidated by a concurrent update? */ + bool stale:1; +}; + +#define xchk_dirtree_for_each_path_safe(dl, path, n) \ + list_for_each_entry_safe((path), (n), &(dl)->path_list, list) + +#define xchk_dirtree_for_each_path(dl, path) \ + list_for_each_entry((path), &(dl)->path_list, list) + +static inline bool +xchk_dirtree_parentless(const struct xchk_dirtree *dl) +{ + struct xfs_scrub *sc = dl->sc; + + if (sc->ip == sc->mp->m_rootip) + return true; + if (VFS_I(sc->ip)->i_nlink == 0) + return true; + return false; +} + +#endif /* __XFS_SCRUB_DIRTREE_H__ */ diff --git a/fs/xfs/scrub/ino_bitmap.h b/fs/xfs/scrub/ino_bitmap.h new file mode 100644 index 000000000000..1300833679ab --- /dev/null +++ b/fs/xfs/scrub/ino_bitmap.h @@ -0,0 +1,37 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2023-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#ifndef __XFS_SCRUB_INO_BITMAP_H__ +#define __XFS_SCRUB_INO_BITMAP_H__ + +/* Bitmaps, but for type-checked for xfs_ino_t */ + +struct xino_bitmap { + struct xbitmap64 inobitmap; +}; + +static inline void xino_bitmap_init(struct xino_bitmap *bitmap) +{ + xbitmap64_init(&bitmap->inobitmap); +} + +static inline void xino_bitmap_destroy(struct xino_bitmap *bitmap) +{ + xbitmap64_destroy(&bitmap->inobitmap); +} + +static inline int xino_bitmap_set(struct xino_bitmap *bitmap, xfs_ino_t ino) +{ + return xbitmap64_set(&bitmap->inobitmap, ino, 1); +} + +static inline int xino_bitmap_test(struct xino_bitmap *bitmap, xfs_ino_t ino) +{ + uint64_t len = 1; + + return xbitmap64_test(&bitmap->inobitmap, ino, &len); +} + +#endif /* __XFS_SCRUB_INO_BITMAP_H__ */ diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 7b1f1abdc7a9..8f1431db7739 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -436,6 +436,13 @@ static const struct xchk_meta_ops meta_scrub_ops[] = { .scrub = xchk_health_record, .repair = xrep_notsupported, }, + [XFS_SCRUB_TYPE_DIRTREE] = { /* directory tree structure */ + .type = ST_INODE, + .setup = xchk_setup_dirtree, + .scrub = xchk_dirtree, + .has = xfs_has_parent, + .repair = xrep_notsupported, + }, }; static int diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index 54a4242bc79c..391027047146 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -185,6 +185,7 @@ int xchk_directory(struct xfs_scrub *sc); int xchk_xattr(struct xfs_scrub *sc); int xchk_symlink(struct xfs_scrub *sc); int xchk_parent(struct xfs_scrub *sc); +int xchk_dirtree(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xchk_rtbitmap(struct xfs_scrub *sc); int xchk_rtsummary(struct xfs_scrub *sc); diff --git a/fs/xfs/scrub/stats.c b/fs/xfs/scrub/stats.c index 42cafbed94ac..7996c2335476 100644 --- a/fs/xfs/scrub/stats.c +++ b/fs/xfs/scrub/stats.c @@ -79,6 +79,7 @@ static const char *name_map[XFS_SCRUB_TYPE_NR] = { [XFS_SCRUB_TYPE_FSCOUNTERS] = "fscounters", [XFS_SCRUB_TYPE_QUOTACHECK] = "quotacheck", [XFS_SCRUB_TYPE_NLINKS] = "nlinks", + [XFS_SCRUB_TYPE_DIRTREE] = "dirtree", }; /* Format the scrub stats into a text buffer, similar to pcp style. */ diff --git a/fs/xfs/scrub/trace.c b/fs/xfs/scrub/trace.c index 4a8cc2c98d99..4470ad0533b8 100644 --- a/fs/xfs/scrub/trace.c +++ b/fs/xfs/scrub/trace.c @@ -28,6 +28,10 @@ #include "scrub/orphanage.h" #include "scrub/nlinks.h" #include "scrub/fscounters.h" +#include "scrub/bitmap.h" +#include "scrub/ino_bitmap.h" +#include "scrub/xfblob.h" +#include "scrub/dirtree.h" /* Figure out which block the btree cursor was pointing to. */ static inline xfs_fsblock_t diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index ecfaa4b88910..c474bcd7d54b 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -27,6 +27,9 @@ struct xchk_nlink; struct xchk_fscounters; struct xfs_rmap_update_params; struct xfs_parent_rec; +enum xchk_dirpath_outcome; +struct xchk_dirtree; +struct xchk_dirtree_outcomes; /* * ftrace's __print_symbolic requires that all enum values be wrapped in the @@ -65,6 +68,7 @@ TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_FSCOUNTERS); TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_QUOTACHECK); TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_NLINKS); TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_HEALTHY); +TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_DIRTREE); #define XFS_SCRUB_TYPE_STRINGS \ { XFS_SCRUB_TYPE_PROBE, "probe" }, \ @@ -94,7 +98,8 @@ TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_HEALTHY); { XFS_SCRUB_TYPE_FSCOUNTERS, "fscounters" }, \ { XFS_SCRUB_TYPE_QUOTACHECK, "quotacheck" }, \ { XFS_SCRUB_TYPE_NLINKS, "nlinks" }, \ - { XFS_SCRUB_TYPE_HEALTHY, "healthy" } + { XFS_SCRUB_TYPE_HEALTHY, "healthy" }, \ + { XFS_SCRUB_TYPE_DIRTREE, "dirtree" } #define XFS_SCRUB_FLAG_STRINGS \ { XFS_SCRUB_IFLAG_REPAIR, "repair" }, \ @@ -171,6 +176,8 @@ DEFINE_EVENT(xchk_class, name, \ DEFINE_SCRUB_EVENT(xchk_start); DEFINE_SCRUB_EVENT(xchk_done); DEFINE_SCRUB_EVENT(xchk_deadlock_retry); +DEFINE_SCRUB_EVENT(xchk_dirtree_start); +DEFINE_SCRUB_EVENT(xchk_dirtree_done); DEFINE_SCRUB_EVENT(xrep_attempt); DEFINE_SCRUB_EVENT(xrep_done); @@ -1576,6 +1583,187 @@ DEFINE_XCHK_PPTR_EVENT(xchk_parent_defer); DEFINE_XCHK_PPTR_EVENT(xchk_parent_slowpath); DEFINE_XCHK_PPTR_EVENT(xchk_parent_ultraslowpath); +DECLARE_EVENT_CLASS(xchk_dirtree_class, + TP_PROTO(struct xfs_scrub *sc, struct xfs_inode *ip, + unsigned int path_nr, const struct xfs_name *name, + const struct xfs_parent_rec *pptr), + TP_ARGS(sc, ip, path_nr, name, pptr), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned int, path_nr) + __field(xfs_ino_t, child_ino) + __field(unsigned int, child_gen) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, parent_gen) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->path_nr = path_nr; + __entry->child_ino = ip->i_ino; + __entry->child_gen = VFS_I(ip)->i_generation; + __entry->parent_ino = be64_to_cpu(pptr->p_ino); + __entry->parent_gen = be32_to_cpu(pptr->p_gen); + __entry->namelen = name->len; + memcpy(__get_str(name), name->name, name->len); + ), + TP_printk("dev %d:%d path %u child_ino 0x%llx child_gen 0x%x parent_ino 0x%llx parent_gen 0x%x name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->path_nr, + __entry->child_ino, + __entry->child_gen, + __entry->parent_ino, + __entry->parent_gen, + __entry->namelen, + __get_str(name)) +); +#define DEFINE_XCHK_DIRTREE_EVENT(name) \ +DEFINE_EVENT(xchk_dirtree_class, name, \ + TP_PROTO(struct xfs_scrub *sc, struct xfs_inode *ip, \ + unsigned int path_nr, const struct xfs_name *name, \ + const struct xfs_parent_rec *pptr), \ + TP_ARGS(sc, ip, path_nr, name, pptr)) +DEFINE_XCHK_DIRTREE_EVENT(xchk_dirtree_create_path); +DEFINE_XCHK_DIRTREE_EVENT(xchk_dirpath_walk_upwards); + +DECLARE_EVENT_CLASS(xchk_dirpath_class, + TP_PROTO(struct xfs_scrub *sc, struct xfs_inode *ip, + unsigned int path_nr, unsigned int step_nr, + const struct xfs_name *name, + const struct xfs_parent_rec *pptr), + TP_ARGS(sc, ip, path_nr, step_nr, name, pptr), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned int, path_nr) + __field(unsigned int, step_nr) + __field(xfs_ino_t, child_ino) + __field(unsigned int, child_gen) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, parent_gen) + __field(unsigned int, namelen) + __dynamic_array(char, name, name->len) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->path_nr = path_nr; + __entry->step_nr = step_nr; + __entry->child_ino = ip->i_ino; + __entry->child_gen = VFS_I(ip)->i_generation; + __entry->parent_ino = be64_to_cpu(pptr->p_ino); + __entry->parent_gen = be32_to_cpu(pptr->p_gen); + __entry->namelen = name->len; + memcpy(__get_str(name), name->name, name->len); + ), + TP_printk("dev %d:%d path %u step %u child_ino 0x%llx child_gen 0x%x parent_ino 0x%llx parent_gen 0x%x name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->path_nr, + __entry->step_nr, + __entry->child_ino, + __entry->child_gen, + __entry->parent_ino, + __entry->parent_gen, + __entry->namelen, + __get_str(name)) +); +#define DEFINE_XCHK_DIRPATH_EVENT(name) \ +DEFINE_EVENT(xchk_dirpath_class, name, \ + TP_PROTO(struct xfs_scrub *sc, struct xfs_inode *ip, \ + unsigned int path_nr, unsigned int step_nr, \ + const struct xfs_name *name, \ + const struct xfs_parent_rec *pptr), \ + TP_ARGS(sc, ip, path_nr, step_nr, name, pptr)) +DEFINE_XCHK_DIRPATH_EVENT(xchk_dirpath_disappeared); +DEFINE_XCHK_DIRPATH_EVENT(xchk_dirpath_badgen); +DEFINE_XCHK_DIRPATH_EVENT(xchk_dirpath_nondir_parent); +DEFINE_XCHK_DIRPATH_EVENT(xchk_dirpath_unlinked_parent); +DEFINE_XCHK_DIRPATH_EVENT(xchk_dirpath_found_next_step); + +TRACE_DEFINE_ENUM(XCHK_DIRPATH_SCANNING); +TRACE_DEFINE_ENUM(XCHK_DIRPATH_DELETE); +TRACE_DEFINE_ENUM(XCHK_DIRPATH_CORRUPT); +TRACE_DEFINE_ENUM(XCHK_DIRPATH_LOOP); +TRACE_DEFINE_ENUM(XCHK_DIRPATH_STALE); +TRACE_DEFINE_ENUM(XCHK_DIRPATH_OK); + +#define XCHK_DIRPATH_OUTCOME_STRINGS \ + { XCHK_DIRPATH_SCANNING, "scanning" }, \ + { XCHK_DIRPATH_DELETE, "delete" }, \ + { XCHK_DIRPATH_CORRUPT, "corrupt" }, \ + { XCHK_DIRPATH_LOOP, "loop" }, \ + { XCHK_DIRPATH_STALE, "stale" }, \ + { XCHK_DIRPATH_OK, "ok" } + +DECLARE_EVENT_CLASS(xchk_dirpath_outcome_class, + TP_PROTO(struct xfs_scrub *sc, unsigned long long path_nr, + unsigned int nr_steps, \ + unsigned int outcome), + TP_ARGS(sc, path_nr, nr_steps, outcome), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned long long, path_nr) + __field(unsigned int, nr_steps) + __field(unsigned int, outcome) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->path_nr = path_nr; + __entry->nr_steps = nr_steps; + __entry->outcome = outcome; + ), + TP_printk("dev %d:%d path %llu steps %u outcome %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->path_nr, + __entry->nr_steps, + __print_symbolic(__entry->outcome, XCHK_DIRPATH_OUTCOME_STRINGS)) +); +#define DEFINE_XCHK_DIRPATH_OUTCOME_EVENT(name) \ +DEFINE_EVENT(xchk_dirpath_outcome_class, name, \ + TP_PROTO(struct xfs_scrub *sc, unsigned long long path_nr, \ + unsigned int nr_steps, \ + unsigned int outcome), \ + TP_ARGS(sc, path_nr, nr_steps, outcome)) +DEFINE_XCHK_DIRPATH_OUTCOME_EVENT(xchk_dirpath_set_outcome); +DEFINE_XCHK_DIRPATH_OUTCOME_EVENT(xchk_dirpath_evaluate_path); + +DECLARE_EVENT_CLASS(xchk_dirtree_evaluate_class, + TP_PROTO(const struct xchk_dirtree *dl, + const struct xchk_dirtree_outcomes *oc), + TP_ARGS(dl, oc), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_ino_t, rootino) + __field(unsigned int, nr_paths) + __field(unsigned int, bad) + __field(unsigned int, suspect) + __field(unsigned int, good) + ), + TP_fast_assign( + __entry->dev = dl->sc->mp->m_super->s_dev; + __entry->ino = dl->sc->ip->i_ino; + __entry->rootino = dl->root_ino; + __entry->nr_paths = dl->nr_paths; + __entry->bad = oc->bad; + __entry->suspect = oc->suspect; + __entry->good = oc->good; + ), + TP_printk("dev %d:%d ino 0x%llx rootino 0x%llx nr_paths %u bad %u suspect %u good %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->rootino, + __entry->nr_paths, + __entry->bad, + __entry->suspect, + __entry->good) +); +#define DEFINE_XCHK_DIRTREE_EVALUATE_EVENT(name) \ +DEFINE_EVENT(xchk_dirtree_evaluate_class, name, \ + TP_PROTO(const struct xchk_dirtree *dl, \ + const struct xchk_dirtree_outcomes *oc), \ + TP_ARGS(dl, oc)) +DEFINE_XCHK_DIRTREE_EVALUATE_EVENT(xchk_dirtree_evaluate); + /* repair tracepoints */ #if IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) diff --git a/fs/xfs/scrub/xfarray.h b/fs/xfs/scrub/xfarray.h index 3b10a58e9f14..8f54c8fc888f 100644 --- a/fs/xfs/scrub/xfarray.h +++ b/fs/xfs/scrub/xfarray.h @@ -8,6 +8,7 @@ /* xfile array index type, along with cursor initialization */ typedef uint64_t xfarray_idx_t; +#define XFARRAY_NULLIDX ((__force xfarray_idx_t)-1ULL) #define XFARRAY_CURSOR_INIT ((__force xfarray_idx_t)0) /* Iterate each index of an xfile array. */ -- cgit v1.2.3-59-g8ed1b From d54c5ac80f8f180eb064ca255d61a879aef6f06d Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:21 -0700 Subject: xfs: invalidate dirloop scrub path data when concurrent updates happen Add a dirent update hook so that we can detect directory tree updates that affect any of the paths found by this scrubber and force it to rescan. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/dirtree.c | 160 ++++++++++++++++++++++++++++++++++++++++++++++++- fs/xfs/scrub/dirtree.h | 20 +++++++ fs/xfs/scrub/trace.h | 65 ++++++++++++++++++++ 3 files changed, 244 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/dirtree.c b/fs/xfs/scrub/dirtree.c index 807c8a8ca7d4..ecc56eb5ed27 100644 --- a/fs/xfs/scrub/dirtree.c +++ b/fs/xfs/scrub/dirtree.c @@ -70,6 +70,9 @@ xchk_dirtree_buf_cleanup( struct xchk_dirtree *dl = buf; struct xchk_dirpath *path, *n; + if (dl->scan_ino != NULLFSINO) + xfs_dir_hook_del(dl->sc->mp, &dl->dhook); + xchk_dirtree_for_each_path_safe(dl, path, n) { list_del_init(&path->list); xino_bitmap_destroy(&path->seen_inodes); @@ -90,13 +93,17 @@ xchk_setup_dirtree( char *descr; int error; + xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); + dl = kvzalloc(sizeof(struct xchk_dirtree), XCHK_GFP_FLAGS); if (!dl) return -ENOMEM; dl->sc = sc; dl->xname.name = dl->namebuf; + dl->hook_xname.name = dl->hook_namebuf; INIT_LIST_HEAD(&dl->path_list); dl->root_ino = NULLFSINO; + dl->scan_ino = NULLFSINO; mutex_init(&dl->lock); @@ -558,6 +565,133 @@ xchk_dirpath_walk_upwards( return error; } +/* + * Decide if this path step has been touched by this live update. Returns + * 1 for yes, 0 for no, or a negative errno. + */ +STATIC int +xchk_dirpath_step_is_stale( + struct xchk_dirtree *dl, + struct xchk_dirpath *path, + unsigned int step_nr, + xfarray_idx_t step_idx, + struct xfs_dir_update_params *p, + xfs_ino_t *cursor) +{ + struct xchk_dirpath_step step; + xfs_ino_t child_ino = *cursor; + int error; + + error = xfarray_load(dl->path_steps, step_idx, &step); + if (error) + return error; + *cursor = be64_to_cpu(step.pptr_rec.p_ino); + + /* + * If the parent and child being updated are not the ones mentioned in + * this path step, the scan data is still ok. + */ + if (p->ip->i_ino != child_ino || p->dp->i_ino != *cursor) + return 0; + + /* + * If the dirent name lengths or byte sequences are different, the scan + * data is still ok. + */ + if (p->name->len != step.name_len) + return 0; + + error = xfblob_loadname(dl->path_names, step.name_cookie, + &dl->hook_xname, step.name_len); + if (error) + return error; + + if (memcmp(dl->hook_xname.name, p->name->name, p->name->len) != 0) + return 0; + + /* Exact match, scan data is out of date. */ + trace_xchk_dirpath_changed(dl->sc, path->path_nr, step_nr, p->dp, + p->ip, p->name); + return 1; +} + +/* + * Decide if this path has been touched by this live update. Returns 1 for + * yes, 0 for no, or a negative errno. + */ +STATIC int +xchk_dirpath_is_stale( + struct xchk_dirtree *dl, + struct xchk_dirpath *path, + struct xfs_dir_update_params *p) +{ + xfs_ino_t cursor = dl->scan_ino; + xfarray_idx_t idx = path->first_step; + unsigned int i; + int ret; + + /* + * The child being updated has not been seen by this path at all; this + * path cannot be stale. + */ + if (!xino_bitmap_test(&path->seen_inodes, p->ip->i_ino)) + return 0; + + ret = xchk_dirpath_step_is_stale(dl, path, 0, idx, p, &cursor); + if (ret != 0) + return ret; + + for (i = 1, idx = path->second_step; i < path->nr_steps; i++, idx++) { + ret = xchk_dirpath_step_is_stale(dl, path, i, idx, p, &cursor); + if (ret != 0) + return ret; + } + + return 0; +} + +/* + * Decide if a directory update from the regular filesystem touches any of the + * paths we've scanned, and invalidate the scan data if true. + */ +STATIC int +xchk_dirtree_live_update( + struct notifier_block *nb, + unsigned long action, + void *data) +{ + struct xfs_dir_update_params *p = data; + struct xchk_dirtree *dl; + struct xchk_dirpath *path; + int ret; + + dl = container_of(nb, struct xchk_dirtree, dhook.dirent_hook.nb); + + trace_xchk_dirtree_live_update(dl->sc, p->dp, action, p->ip, p->delta, + p->name); + + mutex_lock(&dl->lock); + + if (dl->stale || dl->aborted) + goto out_unlock; + + xchk_dirtree_for_each_path(dl, path) { + ret = xchk_dirpath_is_stale(dl, path, p); + if (ret < 0) { + dl->aborted = true; + break; + } + if (ret == 1) { + dl->stale = true; + break; + } + } + +out_unlock: + mutex_unlock(&dl->lock); + return NOTIFY_DONE; +} + /* Delete all the collected path information. */ STATIC void xchk_dirtree_reset( @@ -673,6 +807,8 @@ xchk_dirtree_find_paths_to_root( } if (error) return error; + if (dl->aborted) + return 0; } } while (dl->stale); @@ -744,11 +880,28 @@ xchk_dirtree( ASSERT(xfs_has_parent(sc->mp)); - /* Find the root of the directory tree. */ + /* + * Find the root of the directory tree. Remember which directory to + * scan, because the hook doesn't detach until after sc->ip gets + * released during teardown. + */ dl->root_ino = sc->mp->m_rootip->i_ino; + dl->scan_ino = sc->ip->i_ino; trace_xchk_dirtree_start(sc->ip, sc->sm, 0); + /* + * Hook into the directory entry code so that we can capture updates to + * paths that we have already scanned. The scanner thread takes each + * directory's ILOCK, which means that any in-progress directory update + * will finish before we can scan the directory. + */ + ASSERT(sc->flags & XCHK_FSGATES_DIRENTS); + xfs_dir_hook_setup(&dl->dhook, xchk_dirtree_live_update); + error = xfs_dir_hook_add(sc->mp, &dl->dhook); + if (error) + goto out; + mutex_lock(&dl->lock); /* Trace each parent pointer's path to the root. */ @@ -775,6 +928,10 @@ xchk_dirtree( } if (error) goto out_scanlock; + if (dl->aborted) { + xchk_set_incomplete(sc); + goto out_scanlock; + } /* Assess what we found in our path evaluation. */ xchk_dirtree_evaluate(dl, &oc); @@ -790,6 +947,7 @@ xchk_dirtree( out_scanlock: mutex_unlock(&dl->lock); +out: trace_xchk_dirtree_done(sc->ip, sc->sm, error); return error; } diff --git a/fs/xfs/scrub/dirtree.h b/fs/xfs/scrub/dirtree.h index 50fefd64ae50..2ddbcf43c291 100644 --- a/fs/xfs/scrub/dirtree.h +++ b/fs/xfs/scrub/dirtree.h @@ -72,6 +72,13 @@ struct xchk_dirtree { /* Root inode that we're looking for. */ xfs_ino_t root_ino; + /* + * This is the inode that we're scanning. The live update hook can + * continue to be called after xchk_teardown drops sc->ip but before + * it calls buf_cleanup, so we keep a copy. + */ + xfs_ino_t scan_ino; + /* Scratch buffer for scanning pptr xattrs */ struct xfs_parent_rec pptr_rec; struct xfs_da_args pptr_args; @@ -80,9 +87,19 @@ struct xchk_dirtree { struct xfs_name xname; char namebuf[MAXNAMELEN]; + /* + * Hook into directory updates so that we can receive live updates + * from other writer threads. + */ + struct xfs_dir_hook dhook; + /* lock for everything below here */ struct mutex lock; + /* buffer for the live update functions to use for dirent names */ + struct xfs_name hook_xname; + unsigned char hook_namebuf[MAXNAMELEN]; + /* * All path steps observed during this scan. Each of the path * steps for a particular pathwalk are recorded in sequential @@ -106,6 +123,9 @@ struct xchk_dirtree { /* Have the path data been invalidated by a concurrent update? */ bool stale:1; + + /* Has the scan been aborted? */ + bool aborted:1; }; #define xchk_dirtree_for_each_path_safe(dl, path, n) \ diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index c474bcd7d54b..509b6f4fd0cd 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -1764,6 +1764,71 @@ DEFINE_EVENT(xchk_dirtree_evaluate_class, name, \ TP_ARGS(dl, oc)) DEFINE_XCHK_DIRTREE_EVALUATE_EVENT(xchk_dirtree_evaluate); +TRACE_EVENT(xchk_dirpath_changed, + TP_PROTO(struct xfs_scrub *sc, unsigned int path_nr, + unsigned int step_nr, const struct xfs_inode *dp, + const struct xfs_inode *ip, const struct xfs_name *xname), + TP_ARGS(sc, path_nr, step_nr, dp, ip, xname), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned int, path_nr) + __field(unsigned int, step_nr) + __field(xfs_ino_t, child_ino) + __field(xfs_ino_t, parent_ino) + __field(unsigned int, namelen) + __dynamic_array(char, name, xname->len) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->path_nr = path_nr; + __entry->step_nr = step_nr; + __entry->child_ino = ip->i_ino; + __entry->parent_ino = dp->i_ino; + __entry->namelen = xname->len; + memcpy(__get_str(name), xname->name, xname->len); + ), + TP_printk("dev %d:%d path %u step %u child_ino 0x%llx parent_ino 0x%llx name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->path_nr, + __entry->step_nr, + __entry->child_ino, + __entry->parent_ino, + __entry->namelen, + __get_str(name)) +); + +TRACE_EVENT(xchk_dirtree_live_update, + TP_PROTO(struct xfs_scrub *sc, const struct xfs_inode *dp, + int action, const struct xfs_inode *ip, int delta, + const struct xfs_name *xname), + TP_ARGS(sc, dp, action, ip, delta, xname), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, parent_ino) + __field(int, action) + __field(xfs_ino_t, child_ino) + __field(int, delta) + __field(unsigned int, namelen) + __dynamic_array(char, name, xname->len) + ), + TP_fast_assign( + __entry->dev = sc->mp->m_super->s_dev; + __entry->parent_ino = dp->i_ino; + __entry->action = action; + __entry->child_ino = ip->i_ino; + __entry->delta = delta; + __entry->namelen = xname->len; + memcpy(__get_str(name), xname->name, xname->len); + ), + TP_printk("dev %d:%d parent_ino 0x%llx child_ino 0x%llx nlink_delta %d name '%.*s'", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->parent_ino, + __entry->child_ino, + __entry->delta, + __entry->namelen, + __get_str(name)) +); + /* repair tracepoints */ #if IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) -- cgit v1.2.3-59-g8ed1b From 37056912d5721324ac28787a4f903798f7361099 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:21 -0700 Subject: xfs: report directory tree corruption in the health information Report directories that are the source of corruption in the directory tree. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_fs.h | 1 + fs/xfs/libxfs/xfs_health.h | 4 +++- fs/xfs/scrub/health.c | 1 + fs/xfs/xfs_health.c | 1 + 4 files changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index 85f2a7e20ac0..7ae1912cdbfc 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -411,6 +411,7 @@ struct xfs_bulkstat { #define XFS_BS_SICK_XATTR (1 << 5) /* extended attributes */ #define XFS_BS_SICK_SYMLINK (1 << 6) /* symbolic link remote target */ #define XFS_BS_SICK_PARENT (1 << 7) /* parent pointers */ +#define XFS_BS_SICK_DIRTREE (1 << 8) /* directory tree structure */ /* * Project quota id helpers (previously projid was 16bit only diff --git a/fs/xfs/libxfs/xfs_health.h b/fs/xfs/libxfs/xfs_health.h index 3c64b5f9bd68..b0edb4288e59 100644 --- a/fs/xfs/libxfs/xfs_health.h +++ b/fs/xfs/libxfs/xfs_health.h @@ -95,6 +95,7 @@ struct xfs_da_args; /* Don't propagate sick status to ag health summary during inactivation */ #define XFS_SICK_INO_FORGET (1 << 12) +#define XFS_SICK_INO_DIRTREE (1 << 13) /* directory tree structure */ /* Primary evidence of health problems in a given group. */ #define XFS_SICK_FS_PRIMARY (XFS_SICK_FS_COUNTERS | \ @@ -125,7 +126,8 @@ struct xfs_da_args; XFS_SICK_INO_DIR | \ XFS_SICK_INO_XATTR | \ XFS_SICK_INO_SYMLINK | \ - XFS_SICK_INO_PARENT) + XFS_SICK_INO_PARENT | \ + XFS_SICK_INO_DIRTREE) #define XFS_SICK_INO_ZAPPED (XFS_SICK_INO_BMBTD_ZAPPED | \ XFS_SICK_INO_BMBTA_ZAPPED | \ diff --git a/fs/xfs/scrub/health.c b/fs/xfs/scrub/health.c index 9020a6bef7f1..b712a8bd34f5 100644 --- a/fs/xfs/scrub/health.c +++ b/fs/xfs/scrub/health.c @@ -108,6 +108,7 @@ static const struct xchk_health_map type_to_health_flag[XFS_SCRUB_TYPE_NR] = { [XFS_SCRUB_TYPE_FSCOUNTERS] = { XHG_FS, XFS_SICK_FS_COUNTERS }, [XFS_SCRUB_TYPE_QUOTACHECK] = { XHG_FS, XFS_SICK_FS_QUOTACHECK }, [XFS_SCRUB_TYPE_NLINKS] = { XHG_FS, XFS_SICK_FS_NLINKS }, + [XFS_SCRUB_TYPE_DIRTREE] = { XHG_INO, XFS_SICK_INO_DIRTREE }, }; /* Return the health status mask for this scrub type. */ diff --git a/fs/xfs/xfs_health.c b/fs/xfs/xfs_health.c index b39f959146bc..10f116d093a2 100644 --- a/fs/xfs/xfs_health.c +++ b/fs/xfs/xfs_health.c @@ -470,6 +470,7 @@ static const struct ioctl_sick_map ino_map[] = { { XFS_SICK_INO_BMBTA_ZAPPED, XFS_BS_SICK_BMBTA }, { XFS_SICK_INO_DIR_ZAPPED, XFS_BS_SICK_DIR }, { XFS_SICK_INO_SYMLINK_ZAPPED, XFS_BS_SICK_SYMLINK }, + { XFS_SICK_INO_DIRTREE, XFS_BS_SICK_DIRTREE }, { 0, 0 }, }; -- cgit v1.2.3-59-g8ed1b From 3f31406aef493b3f19020909d29974e28253f91c Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:22 -0700 Subject: xfs: fix corruptions in the directory tree Repair corruptions in the directory tree itself. Cycles are broken by removing an incoming parent->child link. Multiply-owned directories are fixed by pruning the extra parent -> child links Disconnected subtrees are reconnected to the lost and found. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/Makefile | 1 + fs/xfs/scrub/dirtree.c | 38 +- fs/xfs/scrub/dirtree.h | 29 ++ fs/xfs/scrub/dirtree_repair.c | 821 ++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/orphanage.c | 6 + fs/xfs/scrub/orphanage.h | 8 + fs/xfs/scrub/repair.h | 4 + fs/xfs/scrub/scrub.c | 2 +- fs/xfs/scrub/trace.h | 23 +- fs/xfs/xfs_inode.c | 2 +- fs/xfs/xfs_inode.h | 1 + 11 files changed, 927 insertions(+), 8 deletions(-) create mode 100644 fs/xfs/scrub/dirtree_repair.c (limited to 'fs') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 8ec0dd257a98..d1ce1213797b 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -204,6 +204,7 @@ xfs-y += $(addprefix scrub/, \ bmap_repair.o \ cow_repair.o \ dir_repair.o \ + dirtree_repair.o \ findparent.o \ fscounters_repair.o \ ialloc_repair.o \ diff --git a/fs/xfs/scrub/dirtree.c b/fs/xfs/scrub/dirtree.c index ecc56eb5ed27..bde58fb561ea 100644 --- a/fs/xfs/scrub/dirtree.c +++ b/fs/xfs/scrub/dirtree.c @@ -26,6 +26,8 @@ #include "scrub/xfblob.h" #include "scrub/listxattr.h" #include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/orphanage.h" #include "scrub/dirtree.h" /* @@ -95,6 +97,12 @@ xchk_setup_dirtree( xchk_fsgates_enable(sc, XCHK_FSGATES_DIRENTS); + if (xchk_could_repair(sc)) { + error = xrep_setup_dirtree(sc); + if (error) + return error; + } + dl = kvzalloc(sizeof(struct xchk_dirtree), XCHK_GFP_FLAGS); if (!dl) return -ENOMEM; @@ -104,6 +112,7 @@ xchk_setup_dirtree( INIT_LIST_HEAD(&dl->path_list); dl->root_ino = NULLFSINO; dl->scan_ino = NULLFSINO; + dl->parent_ino = NULLFSINO; mutex_init(&dl->lock); @@ -142,7 +151,7 @@ out_dl: * Add the parent pointer described by @dl->pptr to the given path as a new * step. Returns -ELNRNG if the path is too deep. */ -STATIC int +int xchk_dirpath_append( struct xchk_dirtree *dl, struct xfs_inode *ip, @@ -609,6 +618,22 @@ xchk_dirpath_step_is_stale( if (memcmp(dl->hook_xname.name, p->name->name, p->name->len) != 0) return 0; + /* + * If the update comes from the repair code itself, walk the state + * machine forward. + */ + if (p->ip->i_ino == dl->scan_ino && + path->outcome == XREP_DIRPATH_ADOPTING) { + xchk_dirpath_set_outcome(dl, path, XREP_DIRPATH_ADOPTED); + return 0; + } + + if (p->ip->i_ino == dl->scan_ino && + path->outcome == XREP_DIRPATH_DELETING) { + xchk_dirpath_set_outcome(dl, path, XREP_DIRPATH_DELETED); + return 0; + } + /* Exact match, scan data is out of date. */ trace_xchk_dirpath_changed(dl->sc, path->path_nr, step_nr, p->dp, p->ip, p->name); @@ -747,7 +772,7 @@ xchk_dirtree_load_path( * path was too deep; -ENOSR if there were too many parent pointers; or * a negative errno. */ -STATIC int +int xchk_dirtree_find_paths_to_root( struct xchk_dirtree *dl) { @@ -819,7 +844,7 @@ xchk_dirtree_find_paths_to_root( * Figure out what to do with the paths we tried to find. Do not call this * if the scan results are stale. */ -STATIC void +void xchk_dirtree_evaluate( struct xchk_dirtree *dl, struct xchk_dirtree_outcomes *oc) @@ -856,6 +881,13 @@ xchk_dirtree_evaluate( /* This path got all the way to the root. */ oc->good++; break; + case XREP_DIRPATH_DELETING: + case XREP_DIRPATH_DELETED: + case XREP_DIRPATH_ADOPTING: + case XREP_DIRPATH_ADOPTED: + /* These should not be in progress! */ + ASSERT(0); + break; } } diff --git a/fs/xfs/scrub/dirtree.h b/fs/xfs/scrub/dirtree.h index 2ddbcf43c291..1e1686365c61 100644 --- a/fs/xfs/scrub/dirtree.h +++ b/fs/xfs/scrub/dirtree.h @@ -26,6 +26,11 @@ enum xchk_dirpath_outcome { XCHK_DIRPATH_LOOP, /* cycle detected further up */ XCHK_DIRPATH_STALE, /* path is stale */ XCHK_DIRPATH_OK, /* path reaches the root */ + + XREP_DIRPATH_DELETING, /* path is being deleted */ + XREP_DIRPATH_DELETED, /* path has been deleted */ + XREP_DIRPATH_ADOPTING, /* path is being adopted */ + XREP_DIRPATH_ADOPTED, /* path has been adopted */ }; /* @@ -64,6 +69,9 @@ struct xchk_dirtree_outcomes { /* Number of XCHK_DIRPATH_OK */ unsigned int good; + + /* Directory needs to be added to lost+found */ + bool needs_adoption; }; struct xchk_dirtree { @@ -79,6 +87,14 @@ struct xchk_dirtree { */ xfs_ino_t scan_ino; + /* + * If we start deleting redundant paths to this subdirectory, this is + * the inode number of the surviving parent and the dotdot entry will + * be set to this value. If the value is NULLFSINO, then use @root_ino + * as a stand-in until the orphanage can adopt the subdirectory. + */ + xfs_ino_t parent_ino; + /* Scratch buffer for scanning pptr xattrs */ struct xfs_parent_rec pptr_rec; struct xfs_da_args pptr_args; @@ -87,12 +103,18 @@ struct xchk_dirtree { struct xfs_name xname; char namebuf[MAXNAMELEN]; + /* Information for reparenting this directory. */ + struct xrep_adoption adoption; + /* * Hook into directory updates so that we can receive live updates * from other writer threads. */ struct xfs_dir_hook dhook; + /* Parent pointer update arguments. */ + struct xfs_parent_args ppargs; + /* lock for everything below here */ struct mutex lock; @@ -146,4 +168,11 @@ xchk_dirtree_parentless(const struct xchk_dirtree *dl) return false; } +int xchk_dirtree_find_paths_to_root(struct xchk_dirtree *dl); +int xchk_dirpath_append(struct xchk_dirtree *dl, struct xfs_inode *ip, + struct xchk_dirpath *path, const struct xfs_name *name, + const struct xfs_parent_rec *pptr); +void xchk_dirtree_evaluate(struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc); + #endif /* __XFS_SCRUB_DIRTREE_H__ */ diff --git a/fs/xfs/scrub/dirtree_repair.c b/fs/xfs/scrub/dirtree_repair.c new file mode 100644 index 000000000000..5c04e70ba951 --- /dev/null +++ b/fs/xfs/scrub/dirtree_repair.c @@ -0,0 +1,821 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (c) 2023-2024 Oracle. All Rights Reserved. + * Author: Darrick J. Wong + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_trans_space.h" +#include "xfs_mount.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_inode.h" +#include "xfs_icache.h" +#include "xfs_dir2.h" +#include "xfs_dir2_priv.h" +#include "xfs_attr.h" +#include "xfs_parent.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/bitmap.h" +#include "scrub/ino_bitmap.h" +#include "scrub/xfile.h" +#include "scrub/xfarray.h" +#include "scrub/xfblob.h" +#include "scrub/listxattr.h" +#include "scrub/trace.h" +#include "scrub/repair.h" +#include "scrub/orphanage.h" +#include "scrub/dirtree.h" +#include "scrub/readdir.h" + +/* + * Directory Tree Structure Repairs + * ================================ + * + * If we decide that the directory being scanned is participating in a + * directory loop, the only change we can make is to remove directory entries + * pointing down to @sc->ip. If that leaves it with no parents, the directory + * should be adopted by the orphanage. + */ + +/* Set up to repair directory loops. */ +int +xrep_setup_dirtree( + struct xfs_scrub *sc) +{ + return xrep_orphanage_try_create(sc); +} + +/* Change the outcome of this path. */ +static inline void +xrep_dirpath_set_outcome( + struct xchk_dirtree *dl, + struct xchk_dirpath *path, + enum xchk_dirpath_outcome outcome) +{ + trace_xrep_dirpath_set_outcome(dl->sc, path->path_nr, path->nr_steps, + outcome); + + path->outcome = outcome; +} + +/* Delete all paths. */ +STATIC void +xrep_dirtree_delete_all_paths( + struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc) +{ + struct xchk_dirpath *path; + + xchk_dirtree_for_each_path(dl, path) { + switch (path->outcome) { + case XCHK_DIRPATH_CORRUPT: + case XCHK_DIRPATH_LOOP: + oc->suspect--; + oc->bad++; + xrep_dirpath_set_outcome(dl, path, XCHK_DIRPATH_DELETE); + break; + case XCHK_DIRPATH_OK: + oc->good--; + oc->bad++; + xrep_dirpath_set_outcome(dl, path, XCHK_DIRPATH_DELETE); + break; + default: + break; + } + } + + ASSERT(oc->suspect == 0); + ASSERT(oc->good == 0); +} + +/* Since this is the surviving path, set the dotdot entry to this value. */ +STATIC void +xrep_dirpath_retain_parent( + struct xchk_dirtree *dl, + struct xchk_dirpath *path) +{ + struct xchk_dirpath_step step; + int error; + + error = xfarray_load(dl->path_steps, path->first_step, &step); + if (error) + return; + + dl->parent_ino = be64_to_cpu(step.pptr_rec.p_ino); +} + +/* Find the one surviving path so we know how to set dotdot. */ +STATIC void +xrep_dirtree_find_surviving_path( + struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc) +{ + struct xchk_dirpath *path; + bool foundit = false; + + xchk_dirtree_for_each_path(dl, path) { + switch (path->outcome) { + case XCHK_DIRPATH_CORRUPT: + case XCHK_DIRPATH_LOOP: + case XCHK_DIRPATH_OK: + if (!foundit) { + xrep_dirpath_retain_parent(dl, path); + foundit = true; + continue; + } + ASSERT(foundit == false); + break; + default: + break; + } + } + + ASSERT(oc->suspect + oc->good == 1); +} + +/* Delete all paths except for the one good one. */ +STATIC void +xrep_dirtree_keep_one_good_path( + struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc) +{ + struct xchk_dirpath *path; + bool foundit = false; + + xchk_dirtree_for_each_path(dl, path) { + switch (path->outcome) { + case XCHK_DIRPATH_CORRUPT: + case XCHK_DIRPATH_LOOP: + oc->suspect--; + oc->bad++; + xrep_dirpath_set_outcome(dl, path, XCHK_DIRPATH_DELETE); + break; + case XCHK_DIRPATH_OK: + if (!foundit) { + xrep_dirpath_retain_parent(dl, path); + foundit = true; + continue; + } + oc->good--; + oc->bad++; + xrep_dirpath_set_outcome(dl, path, XCHK_DIRPATH_DELETE); + break; + default: + break; + } + } + + ASSERT(oc->suspect == 0); + ASSERT(oc->good < 2); +} + +/* Delete all paths except for one suspect one. */ +STATIC void +xrep_dirtree_keep_one_suspect_path( + struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc) +{ + struct xchk_dirpath *path; + bool foundit = false; + + xchk_dirtree_for_each_path(dl, path) { + switch (path->outcome) { + case XCHK_DIRPATH_CORRUPT: + case XCHK_DIRPATH_LOOP: + if (!foundit) { + xrep_dirpath_retain_parent(dl, path); + foundit = true; + continue; + } + oc->suspect--; + oc->bad++; + xrep_dirpath_set_outcome(dl, path, XCHK_DIRPATH_DELETE); + break; + case XCHK_DIRPATH_OK: + ASSERT(0); + break; + default: + break; + } + } + + ASSERT(oc->suspect == 1); + ASSERT(oc->good == 0); +} + +/* + * Figure out what to do with the paths we tried to find. Returns -EDEADLOCK + * if the scan results have become stale. + */ +STATIC void +xrep_dirtree_decide_fate( + struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc) +{ + xchk_dirtree_evaluate(dl, oc); + + /* Parentless directories should not have any paths at all. */ + if (xchk_dirtree_parentless(dl)) { + xrep_dirtree_delete_all_paths(dl, oc); + return; + } + + /* One path is exactly the number of paths we want. */ + if (oc->good + oc->suspect == 1) { + xrep_dirtree_find_surviving_path(dl, oc); + return; + } + + /* Zero paths means we should reattach the subdir to the orphanage. */ + if (oc->good + oc->suspect == 0) { + if (dl->sc->orphanage) + oc->needs_adoption = true; + return; + } + + /* + * Otherwise, this subdirectory has too many parents. If there's at + * least one good path, keep it and delete the others. + */ + if (oc->good > 0) { + xrep_dirtree_keep_one_good_path(dl, oc); + return; + } + + /* + * There are no good paths and there are too many suspect paths. + * Keep the first suspect path and delete the rest. + */ + xrep_dirtree_keep_one_suspect_path(dl, oc); +} + +/* + * Load the first step of this path into @step and @dl->xname/pptr + * for later repair work. + */ +STATIC int +xrep_dirtree_prep_path( + struct xchk_dirtree *dl, + struct xchk_dirpath *path, + struct xchk_dirpath_step *step) +{ + int error; + + error = xfarray_load(dl->path_steps, path->first_step, step); + if (error) + return error; + + error = xfblob_loadname(dl->path_names, step->name_cookie, &dl->xname, + step->name_len); + if (error) + return error; + + dl->pptr_rec = step->pptr_rec; /* struct copy */ + return 0; +} + +/* Delete the VFS dentry for a removed child. */ +STATIC int +xrep_dirtree_purge_dentry( + struct xchk_dirtree *dl, + struct xfs_inode *dp, + const struct xfs_name *name) +{ + struct qstr qname = QSTR_INIT(name->name, name->len); + struct dentry *parent_dentry, *child_dentry; + int error = 0; + + /* + * Find the dentry for the parent directory. If there isn't one, we're + * done. Caller already holds i_rwsem for parent and child. + */ + parent_dentry = d_find_alias(VFS_I(dp)); + if (!parent_dentry) + return 0; + + /* The VFS thinks the parent is a directory, right? */ + if (!d_is_dir(parent_dentry)) { + ASSERT(d_is_dir(parent_dentry)); + error = -EFSCORRUPTED; + goto out_dput_parent; + } + + /* + * Try to find the dirent pointing to the child. If there isn't one, + * we're done. + */ + qname.hash = full_name_hash(parent_dentry, name->name, name->len); + child_dentry = d_lookup(parent_dentry, &qname); + if (!child_dentry) { + error = 0; + goto out_dput_parent; + } + + trace_xrep_dirtree_delete_child(dp->i_mount, child_dentry); + + /* Child is not a directory? We're screwed. */ + if (!d_is_dir(child_dentry)) { + ASSERT(d_is_dir(child_dentry)); + error = -EFSCORRUPTED; + goto out_dput_child; + } + + /* Replace the child dentry with a negative one. */ + d_delete(child_dentry); + +out_dput_child: + dput(child_dentry); +out_dput_parent: + dput(parent_dentry); + return error; +} + +/* + * Prepare to delete a link by taking the IOLOCK of the parent and the child + * (scrub target). Caller must hold IOLOCK_EXCL on @sc->ip. Returns 0 if we + * took both locks, or a negative errno if we couldn't lock the parent in time. + */ +static inline int +xrep_dirtree_unlink_iolock( + struct xfs_scrub *sc, + struct xfs_inode *dp) +{ + int error; + + ASSERT(sc->ilock_flags & XFS_IOLOCK_EXCL); + + if (xfs_ilock_nowait(dp, XFS_IOLOCK_EXCL)) + return 0; + + xchk_iunlock(sc, XFS_IOLOCK_EXCL); + do { + xfs_ilock(dp, XFS_IOLOCK_EXCL); + if (xchk_ilock_nowait(sc, XFS_IOLOCK_EXCL)) + break; + xfs_iunlock(dp, XFS_IOLOCK_EXCL); + + if (xchk_should_terminate(sc, &error)) { + xchk_ilock(sc, XFS_IOLOCK_EXCL); + return error; + } + + delay(1); + } while (1); + + return 0; +} + +/* + * Remove a link from the directory tree and update the dcache. Returns + * -ESTALE if the scan data are now out of date. + */ +STATIC int +xrep_dirtree_unlink( + struct xchk_dirtree *dl, + struct xfs_inode *dp, + struct xchk_dirpath *path, + struct xchk_dirpath_step *step) +{ + struct xfs_scrub *sc = dl->sc; + struct xfs_mount *mp = sc->mp; + xfs_ino_t dotdot_ino; + xfs_ino_t parent_ino = dl->parent_ino; + unsigned int resblks; + int dontcare; + int error; + + /* Take IOLOCK_EXCL of the parent and child. */ + error = xrep_dirtree_unlink_iolock(sc, dp); + if (error) + return error; + + /* + * Create the transaction that we need to sever the path. Ignore + * EDQUOT and ENOSPC being returned via nospace_error because the + * directory code can handle a reservationless update. + */ + resblks = xfs_remove_space_res(mp, step->name_len); + error = xfs_trans_alloc_dir(dp, &M_RES(mp)->tr_remove, sc->ip, + &resblks, &sc->tp, &dontcare); + if (error) + goto out_iolock; + + /* + * Cancel if someone invalidate the paths while we were trying to get + * the ILOCK. + */ + mutex_lock(&dl->lock); + if (dl->stale) { + mutex_unlock(&dl->lock); + error = -ESTALE; + goto out_trans_cancel; + } + xrep_dirpath_set_outcome(dl, path, XREP_DIRPATH_DELETING); + mutex_unlock(&dl->lock); + + trace_xrep_dirtree_delete_path(dl->sc, sc->ip, path->path_nr, + &dl->xname, &dl->pptr_rec); + + /* + * Decide if we need to reset the dotdot entry. Rules: + * + * - If there's a surviving parent, we want dotdot to point there. + * - If we don't have any surviving parents, then point dotdot at the + * root dir. + * - If dotdot is already set to the value we want, pass in NULLFSINO + * for no change necessary. + * + * Do this /before/ we dirty anything, in case the dotdot lookup + * fails. + */ + error = xchk_dir_lookup(sc, sc->ip, &xfs_name_dotdot, &dotdot_ino); + if (error) + goto out_trans_cancel; + if (parent_ino == NULLFSINO) + parent_ino = dl->root_ino; + if (dotdot_ino == parent_ino) + parent_ino = NULLFSINO; + + /* Drop the link from sc->ip's dotdot entry. */ + error = xfs_droplink(sc->tp, dp); + if (error) + goto out_trans_cancel; + + /* Reset the dotdot entry to a surviving parent. */ + if (parent_ino != NULLFSINO) { + error = xfs_dir_replace(sc->tp, sc->ip, &xfs_name_dotdot, + parent_ino, 0); + if (error) + goto out_trans_cancel; + } + + /* Drop the link from dp to sc->ip. */ + error = xfs_droplink(sc->tp, sc->ip); + if (error) + goto out_trans_cancel; + + error = xfs_dir_removename(sc->tp, dp, &dl->xname, sc->ip->i_ino, + resblks); + if (error) { + ASSERT(error != -ENOENT); + goto out_trans_cancel; + } + + if (xfs_has_parent(sc->mp)) { + error = xfs_parent_removename(sc->tp, &dl->ppargs, dp, + &dl->xname, sc->ip); + if (error) + goto out_trans_cancel; + } + + /* + * Notify dirent hooks that we removed the bad link, invalidate the + * dcache, and commit the repair. + */ + xfs_dir_update_hook(dp, sc->ip, -1, &dl->xname); + error = xrep_dirtree_purge_dentry(dl, dp, &dl->xname); + if (error) + goto out_trans_cancel; + + error = xrep_trans_commit(sc); + goto out_ilock; + +out_trans_cancel: + xchk_trans_cancel(sc); +out_ilock: + xfs_iunlock(sc->ip, XFS_ILOCK_EXCL); + xfs_iunlock(dp, XFS_ILOCK_EXCL); +out_iolock: + xfs_iunlock(dp, XFS_IOLOCK_EXCL); + return error; +} + +/* + * Delete a directory entry that points to this directory. Returns -ESTALE + * if the scan data are now out of date. + */ +STATIC int +xrep_dirtree_delete_path( + struct xchk_dirtree *dl, + struct xchk_dirpath *path) +{ + struct xchk_dirpath_step step; + struct xfs_scrub *sc = dl->sc; + struct xfs_inode *dp; + int error; + + /* + * Load the parent pointer and directory inode for this path, then + * drop the scan lock, the ILOCK, and the transaction so that + * _delete_path can reserve the proper transaction. This sets up + * @dl->xname for the deletion. + */ + error = xrep_dirtree_prep_path(dl, path, &step); + if (error) + return error; + + error = xchk_iget(sc, be64_to_cpu(step.pptr_rec.p_ino), &dp); + if (error) + return error; + + mutex_unlock(&dl->lock); + xchk_trans_cancel(sc); + xchk_iunlock(sc, XFS_ILOCK_EXCL); + + /* Delete the directory link and release the parent. */ + error = xrep_dirtree_unlink(dl, dp, path, &step); + xchk_irele(sc, dp); + + /* + * Retake all the resources we had at the beginning even if the repair + * failed or the scan data are now stale. This keeps things simple for + * the caller. + */ + xchk_trans_alloc_empty(sc); + xchk_ilock(sc, XFS_ILOCK_EXCL); + mutex_lock(&dl->lock); + + if (!error && dl->stale) + error = -ESTALE; + return error; +} + +/* Add a new path to represent our in-progress adoption. */ +STATIC int +xrep_dirtree_create_adoption_path( + struct xchk_dirtree *dl) +{ + struct xfs_scrub *sc = dl->sc; + struct xchk_dirpath *path; + int error; + + /* + * We should have capped the number of paths at XFS_MAXLINK-1 in the + * scanner. + */ + if (dl->nr_paths > XFS_MAXLINK) { + ASSERT(dl->nr_paths <= XFS_MAXLINK); + return -EFSCORRUPTED; + } + + /* + * Create a new xchk_path structure to remember this parent pointer + * and record the first name step. + */ + path = kmalloc(sizeof(struct xchk_dirpath), XCHK_GFP_FLAGS); + if (!path) + return -ENOMEM; + + INIT_LIST_HEAD(&path->list); + xino_bitmap_init(&path->seen_inodes); + path->nr_steps = 0; + path->outcome = XREP_DIRPATH_ADOPTING; + + /* + * Record the new link that we just created in the orphanage. Because + * adoption is the last repair that we perform, we don't bother filling + * in the path all the way back to the root. + */ + xfs_inode_to_parent_rec(&dl->pptr_rec, sc->orphanage); + + error = xino_bitmap_set(&path->seen_inodes, sc->orphanage->i_ino); + if (error) + goto out_path; + + trace_xrep_dirtree_create_adoption(sc, sc->ip, dl->nr_paths, + &dl->xname, &dl->pptr_rec); + + error = xchk_dirpath_append(dl, sc->ip, path, &dl->xname, + &dl->pptr_rec); + if (error) + goto out_path; + + path->first_step = xfarray_length(dl->path_steps) - 1; + path->second_step = XFARRAY_NULLIDX; + path->path_nr = dl->nr_paths; + + list_add_tail(&path->list, &dl->path_list); + dl->nr_paths++; + return 0; + +out_path: + kfree(path); + return error; +} + +/* + * Prepare to move a file to the orphanage by taking the IOLOCK of the + * orphanage and the child (scrub target). Caller must hold IOLOCK_EXCL on + * @sc->ip. Returns 0 if we took both locks, or a negative errno if we + * couldn't lock the orphanage in time. + */ +static inline int +xrep_dirtree_adopt_iolock( + struct xfs_scrub *sc) +{ + int error; + + ASSERT(sc->ilock_flags & XFS_IOLOCK_EXCL); + + if (xrep_orphanage_ilock_nowait(sc, XFS_IOLOCK_EXCL)) + return 0; + + xchk_iunlock(sc, XFS_IOLOCK_EXCL); + do { + xrep_orphanage_ilock(sc, XFS_IOLOCK_EXCL); + if (xchk_ilock_nowait(sc, XFS_IOLOCK_EXCL)) + break; + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + + if (xchk_should_terminate(sc, &error)) { + xchk_ilock(sc, XFS_IOLOCK_EXCL); + return error; + } + + delay(1); + } while (1); + + return 0; +} + +/* + * Reattach this orphaned directory to the orphanage. Do not call this with + * any resources held. Returns -ESTALE if the scan data have become out of + * date. + */ +STATIC int +xrep_dirtree_adopt( + struct xchk_dirtree *dl) +{ + struct xfs_scrub *sc = dl->sc; + int error; + + /* Take the IOLOCK of the orphanage and the scrub target. */ + error = xrep_dirtree_adopt_iolock(sc); + if (error) + return error; + + /* + * Set up for an adoption. The directory tree fixer runs after the + * link counts have been corrected. Therefore, we must bump the + * child's link count since there will be no further opportunity to fix + * errors. + */ + error = xrep_adoption_trans_alloc(sc, &dl->adoption); + if (error) + goto out_iolock; + dl->adoption.bump_child_nlink = true; + + /* Figure out what name we're going to use here. */ + error = xrep_adoption_compute_name(&dl->adoption, &dl->xname); + if (error) + goto out_trans; + + /* + * Now that we have a proposed name for the orphanage entry, create + * a faux path so that the live update hook will see it. + */ + mutex_lock(&dl->lock); + if (dl->stale) { + mutex_unlock(&dl->lock); + error = -ESTALE; + goto out_trans; + } + error = xrep_dirtree_create_adoption_path(dl); + mutex_unlock(&dl->lock); + if (error) + goto out_trans; + + /* Reparent the directory. */ + error = xrep_adoption_move(&dl->adoption); + if (error) + goto out_trans; + + /* + * Commit the name and release all inode locks except for the scrub + * target's IOLOCK. + */ + error = xrep_trans_commit(sc); + goto out_ilock; + +out_trans: + xchk_trans_cancel(sc); +out_ilock: + xchk_iunlock(sc, XFS_ILOCK_EXCL); + xrep_orphanage_iunlock(sc, XFS_ILOCK_EXCL); +out_iolock: + xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + return error; +} + +/* + * This newly orphaned directory needs to be adopted by the orphanage. + * Make this happen. + */ +STATIC int +xrep_dirtree_move_to_orphanage( + struct xchk_dirtree *dl) +{ + struct xfs_scrub *sc = dl->sc; + int error; + + /* + * Start by dropping all the resources that we hold so that we can grab + * all the resources that we need for the adoption. + */ + mutex_unlock(&dl->lock); + xchk_trans_cancel(sc); + xchk_iunlock(sc, XFS_ILOCK_EXCL); + + /* Perform the adoption. */ + error = xrep_dirtree_adopt(dl); + + /* + * Retake all the resources we had at the beginning even if the repair + * failed or the scan data are now stale. This keeps things simple for + * the caller. + */ + xchk_trans_alloc_empty(sc); + xchk_ilock(sc, XFS_ILOCK_EXCL); + mutex_lock(&dl->lock); + + if (!error && dl->stale) + error = -ESTALE; + return error; +} + +/* + * Try to fix all the problems. Returns -ESTALE if the scan data have become + * out of date. + */ +STATIC int +xrep_dirtree_fix_problems( + struct xchk_dirtree *dl, + struct xchk_dirtree_outcomes *oc) +{ + struct xchk_dirpath *path; + int error; + + /* Delete all the paths we don't want. */ + xchk_dirtree_for_each_path(dl, path) { + if (path->outcome != XCHK_DIRPATH_DELETE) + continue; + + error = xrep_dirtree_delete_path(dl, path); + if (error) + return error; + } + + /* Reparent this directory to the orphanage. */ + if (oc->needs_adoption) { + if (xrep_orphanage_can_adopt(dl->sc)) + return xrep_dirtree_move_to_orphanage(dl); + return -EFSCORRUPTED; + } + + return 0; +} + +/* Fix directory loops involving this directory. */ +int +xrep_dirtree( + struct xfs_scrub *sc) +{ + struct xchk_dirtree *dl = sc->buf; + struct xchk_dirtree_outcomes oc; + int error; + + /* + * Prepare to fix the directory tree by retaking the scan lock. The + * order of resource acquisition is still IOLOCK -> transaction -> + * ILOCK -> scan lock. + */ + mutex_lock(&dl->lock); + do { + /* + * Decide what we're going to do, then do it. An -ESTALE + * return here means the scan results are invalid and we have + * to walk again. + */ + if (!dl->stale) { + xrep_dirtree_decide_fate(dl, &oc); + + trace_xrep_dirtree_decided_fate(dl, &oc); + + error = xrep_dirtree_fix_problems(dl, &oc); + if (!error || error != -ESTALE) + break; + } + error = xchk_dirtree_find_paths_to_root(dl); + if (error == -ELNRNG || error == -ENOSR) + error = -EFSCORRUPTED; + } while (!error); + mutex_unlock(&dl->lock); + + return error; +} diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index b2f905924d0d..b1c6c60ee1da 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -570,6 +570,12 @@ xrep_adoption_move( xfs_bumplink(sc->tp, sc->orphanage); xfs_trans_log_inode(sc->tp, sc->orphanage, XFS_ILOG_CORE); + /* Bump the link count of the child. */ + if (adopt->bump_child_nlink) { + xfs_bumplink(sc->tp, sc->ip); + xfs_trans_log_inode(sc->tp, sc->ip, XFS_ILOG_CORE); + } + /* Replace the dotdot entry if the child is a subdirectory. */ if (isdir) { error = xfs_dir_replace(sc->tp, sc->ip, &xfs_name_dotdot, diff --git a/fs/xfs/scrub/orphanage.h b/fs/xfs/scrub/orphanage.h index beb6b686784e..7c7a2e7d81db 100644 --- a/fs/xfs/scrub/orphanage.h +++ b/fs/xfs/scrub/orphanage.h @@ -60,6 +60,14 @@ struct xrep_adoption { /* Block reservations for orphanage and child (if directory). */ unsigned int orphanage_blkres; unsigned int child_blkres; + + /* + * Does the caller want us to bump the child link count? This is not + * needed when reattaching files that have become disconnected but have + * nlink > 1. It is necessary when changing the directory tree + * structure. + */ + bool bump_child_nlink:1; }; bool xrep_orphanage_can_adopt(struct xfs_scrub *sc); diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h index 622eb486a16f..0e0dc2bf985c 100644 --- a/fs/xfs/scrub/repair.h +++ b/fs/xfs/scrub/repair.h @@ -95,6 +95,7 @@ int xrep_setup_directory(struct xfs_scrub *sc); int xrep_setup_parent(struct xfs_scrub *sc); int xrep_setup_nlinks(struct xfs_scrub *sc); int xrep_setup_symlink(struct xfs_scrub *sc, unsigned int *resblks); +int xrep_setup_dirtree(struct xfs_scrub *sc); /* Repair setup functions */ int xrep_setup_ag_allocbt(struct xfs_scrub *sc); @@ -132,6 +133,7 @@ int xrep_xattr(struct xfs_scrub *sc); int xrep_directory(struct xfs_scrub *sc); int xrep_parent(struct xfs_scrub *sc); int xrep_symlink(struct xfs_scrub *sc); +int xrep_dirtree(struct xfs_scrub *sc); #ifdef CONFIG_XFS_RT int xrep_rtbitmap(struct xfs_scrub *sc); @@ -205,6 +207,7 @@ xrep_setup_nothing( #define xrep_setup_directory xrep_setup_nothing #define xrep_setup_parent xrep_setup_nothing #define xrep_setup_nlinks xrep_setup_nothing +#define xrep_setup_dirtree xrep_setup_nothing #define xrep_setup_inode(sc, imap) ((void)0) @@ -239,6 +242,7 @@ static inline int xrep_setup_symlink(struct xfs_scrub *sc, unsigned int *x) #define xrep_directory xrep_notsupported #define xrep_parent xrep_notsupported #define xrep_symlink xrep_notsupported +#define xrep_dirtree xrep_notsupported #endif /* CONFIG_XFS_ONLINE_REPAIR */ diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 8f1431db7739..e813b66b603a 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -441,7 +441,7 @@ static const struct xchk_meta_ops meta_scrub_ops[] = { .setup = xchk_setup_dirtree, .scrub = xchk_dirtree, .has = xfs_has_parent, - .repair = xrep_notsupported, + .repair = xrep_dirtree, }, }; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 509b6f4fd0cd..b3756722bee1 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -1685,6 +1685,10 @@ TRACE_DEFINE_ENUM(XCHK_DIRPATH_CORRUPT); TRACE_DEFINE_ENUM(XCHK_DIRPATH_LOOP); TRACE_DEFINE_ENUM(XCHK_DIRPATH_STALE); TRACE_DEFINE_ENUM(XCHK_DIRPATH_OK); +TRACE_DEFINE_ENUM(XREP_DIRPATH_DELETING); +TRACE_DEFINE_ENUM(XREP_DIRPATH_DELETED); +TRACE_DEFINE_ENUM(XREP_DIRPATH_ADOPTING); +TRACE_DEFINE_ENUM(XREP_DIRPATH_ADOPTED); #define XCHK_DIRPATH_OUTCOME_STRINGS \ { XCHK_DIRPATH_SCANNING, "scanning" }, \ @@ -1692,7 +1696,11 @@ TRACE_DEFINE_ENUM(XCHK_DIRPATH_OK); { XCHK_DIRPATH_CORRUPT, "corrupt" }, \ { XCHK_DIRPATH_LOOP, "loop" }, \ { XCHK_DIRPATH_STALE, "stale" }, \ - { XCHK_DIRPATH_OK, "ok" } + { XCHK_DIRPATH_OK, "ok" }, \ + { XREP_DIRPATH_DELETING, "deleting" }, \ + { XREP_DIRPATH_DELETED, "deleted" }, \ + { XREP_DIRPATH_ADOPTING, "adopting" }, \ + { XREP_DIRPATH_ADOPTED, "adopted" } DECLARE_EVENT_CLASS(xchk_dirpath_outcome_class, TP_PROTO(struct xfs_scrub *sc, unsigned long long path_nr, @@ -1738,6 +1746,7 @@ DECLARE_EVENT_CLASS(xchk_dirtree_evaluate_class, __field(unsigned int, bad) __field(unsigned int, suspect) __field(unsigned int, good) + __field(bool, needs_adoption) ), TP_fast_assign( __entry->dev = dl->sc->mp->m_super->s_dev; @@ -1747,15 +1756,17 @@ DECLARE_EVENT_CLASS(xchk_dirtree_evaluate_class, __entry->bad = oc->bad; __entry->suspect = oc->suspect; __entry->good = oc->good; + __entry->needs_adoption = oc->needs_adoption ? 1 : 0; ), - TP_printk("dev %d:%d ino 0x%llx rootino 0x%llx nr_paths %u bad %u suspect %u good %u", + TP_printk("dev %d:%d ino 0x%llx rootino 0x%llx nr_paths %u bad %u suspect %u good %u adopt? %d", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->rootino, __entry->nr_paths, __entry->bad, __entry->suspect, - __entry->good) + __entry->good, + __entry->needs_adoption) ); #define DEFINE_XCHK_DIRTREE_EVALUATE_EVENT(name) \ DEFINE_EVENT(xchk_dirtree_evaluate_class, name, \ @@ -3181,6 +3192,7 @@ DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_child); DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_alias); DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_dentry); DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_invalidate_child); +DEFINE_REPAIR_DENTRY_EVENT(xrep_dirtree_delete_child); TRACE_EVENT(xrep_symlink_salvage_target, TP_PROTO(struct xfs_inode *ip, char *target, unsigned int targetlen), @@ -3483,6 +3495,11 @@ TRACE_EVENT(xrep_iunlink_commit_bucket, __entry->agino) ); +DEFINE_XCHK_DIRPATH_OUTCOME_EVENT(xrep_dirpath_set_outcome); +DEFINE_XCHK_DIRTREE_EVENT(xrep_dirtree_delete_path); +DEFINE_XCHK_DIRTREE_EVENT(xrep_dirtree_create_adoption); +DEFINE_XCHK_DIRTREE_EVALUATE_EVENT(xrep_dirtree_decided_fate); + #endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ #endif /* _TRACE_XFS_SCRUB_TRACE_H */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index ebe2ce9bd9ee..58fb7a5062e1 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -893,7 +893,7 @@ xfs_init_new_inode( * link count to go to zero, move the inode to AGI unlinked list so that it can * be freed when the last active reference goes away via xfs_inactive(). */ -static int /* error */ +int xfs_droplink( struct xfs_trans *tp, struct xfs_inode *ip) diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 04a91e312993..9fd4d29a5713 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -626,6 +626,7 @@ void xfs_end_io(struct work_struct *work); int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2); void xfs_iunlock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2); void xfs_iunlock2_remapping(struct xfs_inode *ip1, struct xfs_inode *ip2); +int xfs_droplink(struct xfs_trans *tp, struct xfs_inode *ip); void xfs_bumplink(struct xfs_trans *tp, struct xfs_inode *ip); void xfs_lock_inodes(struct xfs_inode **ips, int inodes, uint lock_mode); void xfs_sort_inodes(struct xfs_inode **i_tab, unsigned int num_inodes); -- cgit v1.2.3-59-g8ed1b From 271557de7cbfdecb08e89ae1ca74647ceb57224f Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:23 -0700 Subject: xfs: reduce the rate of cond_resched calls inside scrub We really don't want to call cond_resched every single time we go through a loop in scrub -- there may be billions of records, and probing into the scheduler itself has overhead. Reduce this overhead by only calling cond_resched 10x per second; and add a counter so that we only check jiffies once every 1000 records or so. Surprisingly, this reduces scrub-only fstests runtime by about 2%. I used the bmapinflate xfs_db command to produce a billion-extent file and this stupid gadget reduced the scrub runtime by about 4%. From a stupid microbenchmark of calling these things 1 billion times, I estimate that cond_resched costs about 5.5ns per call; jiffes costs about 0.3ns per read; and fatal_signal_pending costs about 0.4ns per call. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/common.h | 25 -------------------- fs/xfs/scrub/scrub.c | 1 + fs/xfs/scrub/scrub.h | 64 ++++++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/xfarray.c | 10 ++++---- fs/xfs/scrub/xfarray.h | 3 +++ fs/xfs/scrub/xfile.c | 2 +- 6 files changed, 74 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h index 39465e39dc5f..3d5f1f6b4b7b 100644 --- a/fs/xfs/scrub/common.h +++ b/fs/xfs/scrub/common.h @@ -6,31 +6,6 @@ #ifndef __XFS_SCRUB_COMMON_H__ #define __XFS_SCRUB_COMMON_H__ -/* - * We /could/ terminate a scrub/repair operation early. If we're not - * in a good place to continue (fatal signal, etc.) then bail out. - * Note that we're careful not to make any judgements about *error. - */ -static inline bool -xchk_should_terminate( - struct xfs_scrub *sc, - int *error) -{ - /* - * If preemption is disabled, we need to yield to the scheduler every - * few seconds so that we don't run afoul of the soft lockup watchdog - * or RCU stall detector. - */ - cond_resched(); - - if (fatal_signal_pending(current)) { - if (*error == 0) - *error = -EINTR; - return true; - } - return false; -} - int xchk_trans_alloc(struct xfs_scrub *sc, uint resblks); int xchk_trans_alloc_empty(struct xfs_scrub *sc); void xchk_trans_cancel(struct xfs_scrub *sc); diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index e813b66b603a..4a81f828f9f1 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -620,6 +620,7 @@ xfs_scrub_metadata( sc->sm = sm; sc->ops = &meta_scrub_ops[sm->sm_type]; sc->sick_mask = xchk_health_mask_for_scrub_type(sm->sm_type); + sc->relax = INIT_XCHK_RELAX; retry_op: /* * When repairs are allowed, prevent freezing or readonly remount while diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index 391027047146..4e7e3edb6350 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -8,6 +8,49 @@ struct xfs_scrub; +struct xchk_relax { + unsigned long next_resched; + unsigned int resched_nr; + bool interruptible; +}; + +/* Yield to the scheduler at most 10x per second. */ +#define XCHK_RELAX_NEXT (jiffies + (HZ / 10)) + +#define INIT_XCHK_RELAX \ + (struct xchk_relax){ \ + .next_resched = XCHK_RELAX_NEXT, \ + .resched_nr = 0, \ + .interruptible = true, \ + } + +/* + * Relax during a scrub operation and exit if there's a fatal signal pending. + * + * If preemption is disabled, we need to yield to the scheduler every now and + * then so that we don't run afoul of the soft lockup watchdog or RCU stall + * detector. cond_resched calls are somewhat expensive (~5ns) so we want to + * ratelimit this to 10x per second. Amortize the cost of the other checks by + * only doing it once every 100 calls. + */ +static inline int xchk_maybe_relax(struct xchk_relax *widget) +{ + /* Amortize the cost of scheduling and checking signals. */ + if (likely(++widget->resched_nr < 100)) + return 0; + widget->resched_nr = 0; + + if (unlikely(widget->next_resched <= jiffies)) { + cond_resched(); + widget->next_resched = XCHK_RELAX_NEXT; + } + + if (widget->interruptible && fatal_signal_pending(current)) + return -EINTR; + + return 0; +} + /* * Standard flags for allocating memory within scrub. NOFS context is * configured by the process allocation scope. Scrub and repair must be able @@ -123,6 +166,9 @@ struct xfs_scrub { */ unsigned int sick_mask; + /* next time we want to cond_resched() */ + struct xchk_relax relax; + /* State tracking for single-AG operations. */ struct xchk_ag sa; }; @@ -167,6 +213,24 @@ struct xfs_scrub_subord *xchk_scrub_create_subord(struct xfs_scrub *sc, unsigned int subtype); void xchk_scrub_free_subord(struct xfs_scrub_subord *sub); +/* + * We /could/ terminate a scrub/repair operation early. If we're not + * in a good place to continue (fatal signal, etc.) then bail out. + * Note that we're careful not to make any judgements about *error. + */ +static inline bool +xchk_should_terminate( + struct xfs_scrub *sc, + int *error) +{ + if (xchk_maybe_relax(&sc->relax)) { + if (*error == 0) + *error = -EINTR; + return true; + } + return false; +} + /* Metadata scrubbers */ int xchk_tester(struct xfs_scrub *sc); int xchk_superblock(struct xfs_scrub *sc); diff --git a/fs/xfs/scrub/xfarray.c b/fs/xfs/scrub/xfarray.c index b65cd3fc5ac9..9185ae7088d4 100644 --- a/fs/xfs/scrub/xfarray.c +++ b/fs/xfs/scrub/xfarray.c @@ -7,9 +7,9 @@ #include "xfs_fs.h" #include "xfs_shared.h" #include "xfs_format.h" +#include "scrub/scrub.h" #include "scrub/xfile.h" #include "scrub/xfarray.h" -#include "scrub/scrub.h" #include "scrub/trace.h" /* @@ -486,6 +486,9 @@ xfarray_sortinfo_alloc( xfarray_sortinfo_lo(si)[0] = 0; xfarray_sortinfo_hi(si)[0] = array->nr - 1; + si->relax = INIT_XCHK_RELAX; + if (flags & XFARRAY_SORT_KILLABLE) + si->relax.interruptible = false; trace_xfarray_sort(si, nr_bytes); *infop = si; @@ -503,10 +506,7 @@ xfarray_sort_terminated( * few seconds so that we don't run afoul of the soft lockup watchdog * or RCU stall detector. */ - cond_resched(); - - if ((si->flags & XFARRAY_SORT_KILLABLE) && - fatal_signal_pending(current)) { + if (xchk_maybe_relax(&si->relax)) { if (*error == 0) *error = -EINTR; return true; diff --git a/fs/xfs/scrub/xfarray.h b/fs/xfs/scrub/xfarray.h index 8f54c8fc888f..5eeeeed13ae2 100644 --- a/fs/xfs/scrub/xfarray.h +++ b/fs/xfs/scrub/xfarray.h @@ -127,6 +127,9 @@ struct xfarray_sortinfo { /* XFARRAY_SORT_* flags; see below. */ unsigned int flags; + /* next time we want to cond_resched() */ + struct xchk_relax relax; + /* Cache a folio here for faster scanning for pivots */ struct folio *folio; diff --git a/fs/xfs/scrub/xfile.c b/fs/xfs/scrub/xfile.c index 4e254a0ba003..d848222f802b 100644 --- a/fs/xfs/scrub/xfile.c +++ b/fs/xfs/scrub/xfile.c @@ -10,9 +10,9 @@ #include "xfs_log_format.h" #include "xfs_trans_resv.h" #include "xfs_mount.h" +#include "scrub/scrub.h" #include "scrub/xfile.h" #include "scrub/xfarray.h" -#include "scrub/scrub.h" #include "scrub/trace.h" #include -- cgit v1.2.3-59-g8ed1b From be7cf174e908b1f350dd3ae4fbdf335f22af3273 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:24 -0700 Subject: xfs: move xfs_ioc_scrub_metadata to scrub.c Move the scrub ioctl handler to scrub.c to keep the code together and to reduce unnecessary code when CONFIG_XFS_ONLINE_SCRUB=n. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/scrub.c | 27 ++++++++++++++++++++++++++- fs/xfs/scrub/xfs_scrub.h | 4 ++-- fs/xfs/xfs_ioctl.c | 24 ------------------------ 3 files changed, 28 insertions(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 4a81f828f9f1..1456cc11c406 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -578,7 +578,7 @@ xchk_scrub_create_subord( } /* Dispatch metadata scrubbing. */ -int +STATIC int xfs_scrub_metadata( struct file *file, struct xfs_scrub_metadata *sm) @@ -724,3 +724,28 @@ try_harder: run.retries++; goto retry_op; } + +/* Scrub one aspect of one piece of metadata. */ +int +xfs_ioc_scrub_metadata( + struct file *file, + void __user *arg) +{ + struct xfs_scrub_metadata scrub; + int error; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (copy_from_user(&scrub, arg, sizeof(scrub))) + return -EFAULT; + + error = xfs_scrub_metadata(file, &scrub); + if (error) + return error; + + if (copy_to_user(arg, &scrub, sizeof(scrub))) + return -EFAULT; + + return 0; +} diff --git a/fs/xfs/scrub/xfs_scrub.h b/fs/xfs/scrub/xfs_scrub.h index a39befa743ce..02c930f175d0 100644 --- a/fs/xfs/scrub/xfs_scrub.h +++ b/fs/xfs/scrub/xfs_scrub.h @@ -7,9 +7,9 @@ #define __XFS_SCRUB_H__ #ifndef CONFIG_XFS_ONLINE_SCRUB -# define xfs_scrub_metadata(file, sm) (-ENOTTY) +# define xfs_ioc_scrub_metadata(f, a) (-ENOTTY) #else -int xfs_scrub_metadata(struct file *file, struct xfs_scrub_metadata *sm); +int xfs_ioc_scrub_metadata(struct file *file, void __user *arg); #endif /* CONFIG_XFS_ONLINE_SCRUB */ #endif /* __XFS_SCRUB_H__ */ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 0e97070abe80..857b19428984 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -1055,30 +1055,6 @@ out_free: return error; } -STATIC int -xfs_ioc_scrub_metadata( - struct file *file, - void __user *arg) -{ - struct xfs_scrub_metadata scrub; - int error; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (copy_from_user(&scrub, arg, sizeof(scrub))) - return -EFAULT; - - error = xfs_scrub_metadata(file, &scrub); - if (error) - return error; - - if (copy_to_user(arg, &scrub, sizeof(scrub))) - return -EFAULT; - - return 0; -} - int xfs_ioc_swapext( xfs_swapext_t *sxp) -- cgit v1.2.3-59-g8ed1b From c77b37584c2d1054452853e47e42c7350b8fe687 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:25 -0700 Subject: xfs: introduce vectored scrub mode Introduce a variant on XFS_SCRUB_METADATA that allows for a vectored mode. The caller specifies the principal metadata object that they want to scrub (allocation group, inode, etc.) once, followed by an array of scrub types they want called on that object. The kernel runs the scrub operations and writes the output flags and errno code to the corresponding array element. A new pseudo scrub type BARRIER is introduced to force the kernel to return to userspace if any corruptions have been found when scrubbing the previous scrub types in the array. This enables userspace to schedule, for example, the sequence: 1. data fork 2. barrier 3. directory If the data fork scrub is clean, then the kernel will perform the directory scrub. If not, the barrier in 2 will exit back to userspace. The alternative would have been an interface where userspace passes a pointer to an empty buffer, and the kernel formats that with xfs_scrub_vecs that tell userspace what it scrubbed and what the outcome was. With that the kernel would have to communicate that the buffer needed to have been at least X size, even though for our cases XFS_SCRUB_TYPE_NR + 2 would always be enough. Compared to that, this design keeps all the dependency policy and ordering logic in userspace where it already resides instead of duplicating it in the kernel. The downside of that is that it needs the barrier logic. When running fstests in "rebuild all metadata after each test" mode, I observed a 10% reduction in runtime due to fewer transitions across the system call boundary. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_fs.h | 33 +++++++++++ fs/xfs/scrub/scrub.c | 149 +++++++++++++++++++++++++++++++++++++++++++++++ fs/xfs/scrub/trace.h | 79 ++++++++++++++++++++++++- fs/xfs/scrub/xfs_scrub.h | 2 + fs/xfs/xfs_ioctl.c | 2 + 5 files changed, 264 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index 7ae1912cdbfc..97996cb79aaa 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -725,6 +725,15 @@ struct xfs_scrub_metadata { /* Number of scrub subcommands. */ #define XFS_SCRUB_TYPE_NR 29 +/* + * This special type code only applies to the vectored scrub implementation. + * + * If any of the previous scrub vectors recorded runtime errors or have + * sv_flags bits set that match the OFLAG bits in the barrier vector's + * sv_flags, set the barrier's sv_ret to -ECANCELED and return to userspace. + */ +#define XFS_SCRUB_TYPE_BARRIER (0xFFFFFFFF) + /* i: Repair this metadata. */ #define XFS_SCRUB_IFLAG_REPAIR (1u << 0) @@ -769,6 +778,29 @@ struct xfs_scrub_metadata { XFS_SCRUB_OFLAG_NO_REPAIR_NEEDED) #define XFS_SCRUB_FLAGS_ALL (XFS_SCRUB_FLAGS_IN | XFS_SCRUB_FLAGS_OUT) +/* Vectored scrub calls to reduce the number of kernel transitions. */ + +struct xfs_scrub_vec { + __u32 sv_type; /* XFS_SCRUB_TYPE_* */ + __u32 sv_flags; /* XFS_SCRUB_FLAGS_* */ + __s32 sv_ret; /* 0 or a negative error code */ + __u32 sv_reserved; /* must be zero */ +}; + +/* Vectored metadata scrub control structure. */ +struct xfs_scrub_vec_head { + __u64 svh_ino; /* inode number. */ + __u32 svh_gen; /* inode generation. */ + __u32 svh_agno; /* ag number. */ + __u32 svh_flags; /* XFS_SCRUB_VEC_FLAGS_* */ + __u16 svh_rest_us; /* wait this much time between vector items */ + __u16 svh_nr; /* number of svh_vectors */ + __u64 svh_reserved; /* must be zero */ + __u64 svh_vectors; /* pointer to buffer of xfs_scrub_vec */ +}; + +#define XFS_SCRUB_VEC_FLAGS_ALL (0) + /* * ioctl limits */ @@ -928,6 +960,7 @@ struct xfs_getparents_by_handle { #define XFS_IOC_AG_GEOMETRY _IOWR('X', 61, struct xfs_ag_geometry) #define XFS_IOC_GETPARENTS _IOWR('X', 62, struct xfs_getparents) #define XFS_IOC_GETPARENTS_BY_HANDLE _IOWR('X', 63, struct xfs_getparents_by_handle) +#define XFS_IOC_SCRUBV_METADATA _IOWR('X', 64, struct xfs_scrub_vec_head) /* * ioctl commands that replace IRIX syssgi()'s diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 1456cc11c406..78b00ab85c9c 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -21,6 +21,7 @@ #include "xfs_exchmaps.h" #include "xfs_dir2.h" #include "xfs_parent.h" +#include "xfs_icache.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" @@ -749,3 +750,151 @@ xfs_ioc_scrub_metadata( return 0; } + +/* Decide if there have been any scrub failures up to this point. */ +static inline int +xfs_scrubv_check_barrier( + struct xfs_mount *mp, + const struct xfs_scrub_vec *vectors, + const struct xfs_scrub_vec *stop_vec) +{ + const struct xfs_scrub_vec *v; + __u32 failmask; + + failmask = stop_vec->sv_flags & XFS_SCRUB_FLAGS_OUT; + + for (v = vectors; v < stop_vec; v++) { + if (v->sv_type == XFS_SCRUB_TYPE_BARRIER) + continue; + + /* + * Runtime errors count as a previous failure, except the ones + * used to ask userspace to retry. + */ + switch (v->sv_ret) { + case -EBUSY: + case -ENOENT: + case -EUSERS: + case 0: + break; + default: + return -ECANCELED; + } + + /* + * If any of the out-flags on the scrub vector match the mask + * that was set on the barrier vector, that's a previous fail. + */ + if (v->sv_flags & failmask) + return -ECANCELED; + } + + return 0; +} + +/* Vectored scrub implementation to reduce ioctl calls. */ +int +xfs_ioc_scrubv_metadata( + struct file *file, + void __user *arg) +{ + struct xfs_scrub_vec_head head; + struct xfs_scrub_vec_head __user *uhead = arg; + struct xfs_scrub_vec *vectors; + struct xfs_scrub_vec __user *uvectors; + struct xfs_inode *ip_in = XFS_I(file_inode(file)); + struct xfs_mount *mp = ip_in->i_mount; + struct xfs_scrub_vec *v; + size_t vec_bytes; + unsigned int i; + int error = 0; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (copy_from_user(&head, uhead, sizeof(head))) + return -EFAULT; + + if (head.svh_reserved) + return -EINVAL; + if (head.svh_flags & ~XFS_SCRUB_VEC_FLAGS_ALL) + return -EINVAL; + if (head.svh_nr == 0) + return 0; + + vec_bytes = array_size(head.svh_nr, sizeof(struct xfs_scrub_vec)); + if (vec_bytes > PAGE_SIZE) + return -ENOMEM; + + uvectors = (void __user *)(uintptr_t)head.svh_vectors; + vectors = memdup_user(uvectors, vec_bytes); + if (IS_ERR(vectors)) + return PTR_ERR(vectors); + + trace_xchk_scrubv_start(ip_in, &head); + + for (i = 0, v = vectors; i < head.svh_nr; i++, v++) { + if (v->sv_reserved) { + error = -EINVAL; + goto out_free; + } + + if (v->sv_type == XFS_SCRUB_TYPE_BARRIER && + (v->sv_flags & ~XFS_SCRUB_FLAGS_OUT)) { + error = -EINVAL; + goto out_free; + } + + trace_xchk_scrubv_item(mp, &head, i, v); + } + + /* Run all the scrubbers. */ + for (i = 0, v = vectors; i < head.svh_nr; i++, v++) { + struct xfs_scrub_metadata sm = { + .sm_type = v->sv_type, + .sm_flags = v->sv_flags, + .sm_ino = head.svh_ino, + .sm_gen = head.svh_gen, + .sm_agno = head.svh_agno, + }; + + if (v->sv_type == XFS_SCRUB_TYPE_BARRIER) { + v->sv_ret = xfs_scrubv_check_barrier(mp, vectors, v); + if (v->sv_ret) { + trace_xchk_scrubv_barrier_fail(mp, &head, i, v); + break; + } + + continue; + } + + v->sv_ret = xfs_scrub_metadata(file, &sm); + v->sv_flags = sm.sm_flags; + + trace_xchk_scrubv_outcome(mp, &head, i, v); + + if (head.svh_rest_us) { + ktime_t expires; + + expires = ktime_add_ns(ktime_get(), + head.svh_rest_us * 1000); + set_current_state(TASK_KILLABLE); + schedule_hrtimeout(&expires, HRTIMER_MODE_ABS); + } + + if (fatal_signal_pending(current)) { + error = -EINTR; + goto out_free; + } + } + + if (copy_to_user(uvectors, vectors, vec_bytes) || + copy_to_user(uhead, &head, sizeof(head))) { + error = -EFAULT; + goto out_free; + } + +out_free: + kfree(vectors); + return error; +} diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index b3756722bee1..8ce74bd8530a 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -69,6 +69,7 @@ TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_QUOTACHECK); TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_NLINKS); TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_HEALTHY); TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_DIRTREE); +TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_BARRIER); #define XFS_SCRUB_TYPE_STRINGS \ { XFS_SCRUB_TYPE_PROBE, "probe" }, \ @@ -99,7 +100,8 @@ TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_DIRTREE); { XFS_SCRUB_TYPE_QUOTACHECK, "quotacheck" }, \ { XFS_SCRUB_TYPE_NLINKS, "nlinks" }, \ { XFS_SCRUB_TYPE_HEALTHY, "healthy" }, \ - { XFS_SCRUB_TYPE_DIRTREE, "dirtree" } + { XFS_SCRUB_TYPE_DIRTREE, "dirtree" }, \ + { XFS_SCRUB_TYPE_BARRIER, "barrier" } #define XFS_SCRUB_FLAG_STRINGS \ { XFS_SCRUB_IFLAG_REPAIR, "repair" }, \ @@ -208,6 +210,81 @@ DEFINE_EVENT(xchk_fsgate_class, name, \ DEFINE_SCRUB_FSHOOK_EVENT(xchk_fsgates_enable); DEFINE_SCRUB_FSHOOK_EVENT(xchk_fsgates_disable); +DECLARE_EVENT_CLASS(xchk_vector_head_class, + TP_PROTO(struct xfs_inode *ip, struct xfs_scrub_vec_head *vhead), + TP_ARGS(ip, vhead), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_agnumber_t, agno) + __field(xfs_ino_t, inum) + __field(unsigned int, gen) + __field(unsigned int, flags) + __field(unsigned short, rest_us) + __field(unsigned short, nr_vecs) + ), + TP_fast_assign( + __entry->dev = ip->i_mount->m_super->s_dev; + __entry->ino = ip->i_ino; + __entry->agno = vhead->svh_agno; + __entry->inum = vhead->svh_ino; + __entry->gen = vhead->svh_gen; + __entry->flags = vhead->svh_flags; + __entry->rest_us = vhead->svh_rest_us; + __entry->nr_vecs = vhead->svh_nr; + ), + TP_printk("dev %d:%d ino 0x%llx agno 0x%x inum 0x%llx gen 0x%x flags 0x%x rest_us %u nr_vecs %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->agno, + __entry->inum, + __entry->gen, + __entry->flags, + __entry->rest_us, + __entry->nr_vecs) +) +#define DEFINE_SCRUBV_HEAD_EVENT(name) \ +DEFINE_EVENT(xchk_vector_head_class, name, \ + TP_PROTO(struct xfs_inode *ip, struct xfs_scrub_vec_head *vhead), \ + TP_ARGS(ip, vhead)) + +DEFINE_SCRUBV_HEAD_EVENT(xchk_scrubv_start); + +DECLARE_EVENT_CLASS(xchk_vector_class, + TP_PROTO(struct xfs_mount *mp, struct xfs_scrub_vec_head *vhead, + unsigned int vec_nr, struct xfs_scrub_vec *v), + TP_ARGS(mp, vhead, vec_nr, v), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned int, vec_nr) + __field(unsigned int, vec_type) + __field(unsigned int, vec_flags) + __field(int, vec_ret) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->vec_nr = vec_nr; + __entry->vec_type = v->sv_type; + __entry->vec_flags = v->sv_flags; + __entry->vec_ret = v->sv_ret; + ), + TP_printk("dev %d:%d vec[%u] type %s flags %s ret %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->vec_nr, + __print_symbolic(__entry->vec_type, XFS_SCRUB_TYPE_STRINGS), + __print_flags(__entry->vec_flags, "|", XFS_SCRUB_FLAG_STRINGS), + __entry->vec_ret) +) +#define DEFINE_SCRUBV_EVENT(name) \ +DEFINE_EVENT(xchk_vector_class, name, \ + TP_PROTO(struct xfs_mount *mp, struct xfs_scrub_vec_head *vhead, \ + unsigned int vec_nr, struct xfs_scrub_vec *v), \ + TP_ARGS(mp, vhead, vec_nr, v)) + +DEFINE_SCRUBV_EVENT(xchk_scrubv_barrier_fail); +DEFINE_SCRUBV_EVENT(xchk_scrubv_item); +DEFINE_SCRUBV_EVENT(xchk_scrubv_outcome); + TRACE_EVENT(xchk_op_error, TP_PROTO(struct xfs_scrub *sc, xfs_agnumber_t agno, xfs_agblock_t bno, int error, void *ret_ip), diff --git a/fs/xfs/scrub/xfs_scrub.h b/fs/xfs/scrub/xfs_scrub.h index 02c930f175d0..f17173b83e6f 100644 --- a/fs/xfs/scrub/xfs_scrub.h +++ b/fs/xfs/scrub/xfs_scrub.h @@ -8,8 +8,10 @@ #ifndef CONFIG_XFS_ONLINE_SCRUB # define xfs_ioc_scrub_metadata(f, a) (-ENOTTY) +# define xfs_ioc_scrubv_metadata(f, a) (-ENOTTY) #else int xfs_ioc_scrub_metadata(struct file *file, void __user *arg); +int xfs_ioc_scrubv_metadata(struct file *file, void __user *arg); #endif /* CONFIG_XFS_ONLINE_SCRUB */ #endif /* __XFS_SCRUB_H__ */ diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 857b19428984..f0117188f302 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -1413,6 +1413,8 @@ xfs_file_ioctl( case FS_IOC_GETFSMAP: return xfs_ioc_getfsmap(ip, arg); + case XFS_IOC_SCRUBV_METADATA: + return xfs_ioc_scrubv_metadata(filp, arg); case XFS_IOC_SCRUB_METADATA: return xfs_ioc_scrub_metadata(filp, arg); -- cgit v1.2.3-59-g8ed1b From b27ce0da60a523fc32e3795f96b2de5490642235 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:26 -0700 Subject: xfs: use dontcache for grabbing inodes during scrub Back when I wrote commit a03297a0ca9f2, I had thought that we'd be doing users a favor by only marking inodes dontcache at the end of a scrub operation, and only if there's only one reference to that inode. This was more or less true back when I_DONTCACHE was an XFS iflag and the only thing it did was change the outcome of xfs_fs_drop_inode to 1. Note: If there are dentries pointing to the inode when scrub finishes, the inode will have positive i_count and stay around in cache until dentry reclaim. But now we have d_mark_dontcache, which cause the inode *and* the dentries attached to it all to be marked I_DONTCACHE, which means that we drop the dentries ASAP, which drops the inode ASAP. This is bad if scrub found problems with the inode, because now they can be scheduled for inactivation, which can cause inodegc to trip on it and shut down the filesystem. Even if the inode isn't bad, this is still suboptimal because phases 3-7 each initiate inode scans. Dropping the inode immediately during phase 3 is silly because phase 5 will reload it and drop it immediately, etc. It's fine to mark the inodes dontcache, but if there have been accesses to the file that set up dentries, we should keep them. I validated this by setting up ftrace to capture xfs_iget_recycle* tracepoints and ran xfs/285 for 30 seconds. With current djwong-wtf I saw ~30,000 recycle events. I then dropped the d_mark_dontcache calls and set XFS_IGET_DONTCACHE, and the recycle events dropped to ~5,000 per 30 seconds. Therefore, grab the inode with XFS_IGET_DONTCACHE, which only has the effect of setting I_DONTCACHE for cache misses. Remove the d_mark_dontcache call that can happen in xchk_irele. Fixes: a03297a0ca9f2 ("xfs: manage inode DONTCACHE status at irele time") Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/common.c | 12 +++--------- fs/xfs/scrub/iscan.c | 13 +++++++++++-- fs/xfs/scrub/scrub.h | 7 +++++++ 3 files changed, 21 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index a7d3a2279662..1ad8ec63a7f4 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -783,7 +783,7 @@ xchk_iget( { ASSERT(sc->tp != NULL); - return xfs_iget(sc->mp, sc->tp, inum, XFS_IGET_UNTRUSTED, 0, ipp); + return xfs_iget(sc->mp, sc->tp, inum, XCHK_IGET_FLAGS, 0, ipp); } /* @@ -834,8 +834,8 @@ again: if (error) return error; - error = xfs_iget(mp, tp, inum, - XFS_IGET_NORETRY | XFS_IGET_UNTRUSTED, 0, ipp); + error = xfs_iget(mp, tp, inum, XFS_IGET_NORETRY | XCHK_IGET_FLAGS, 0, + ipp); if (error == -EAGAIN) { /* * The inode may be in core but temporarily unavailable and may @@ -1062,12 +1062,6 @@ xchk_irele( spin_lock(&VFS_I(ip)->i_lock); VFS_I(ip)->i_state &= ~I_DONTCACHE; spin_unlock(&VFS_I(ip)->i_lock); - } else if (atomic_read(&VFS_I(ip)->i_count) == 1) { - /* - * If this is the last reference to the inode and the caller - * permits it, set DONTCACHE to avoid thrashing. - */ - d_mark_dontcache(VFS_I(ip)); } xfs_irele(ip); diff --git a/fs/xfs/scrub/iscan.c b/fs/xfs/scrub/iscan.c index c380207702e2..cf9d983667ce 100644 --- a/fs/xfs/scrub/iscan.c +++ b/fs/xfs/scrub/iscan.c @@ -407,6 +407,15 @@ xchk_iscan_iget_retry( return -EAGAIN; } +/* + * For an inode scan, we hold the AGI and want to try to grab a batch of + * inodes. Holding the AGI prevents inodegc from clearing freed inodes, + * so we must use noretry here. For every inode after the first one in the + * batch, we don't want to wait, so we use retry there too. Finally, use + * dontcache to avoid polluting the cache. + */ +#define ISCAN_IGET_FLAGS (XFS_IGET_NORETRY | XFS_IGET_DONTCACHE) + /* * Grab an inode as part of an inode scan. While scanning this inode, the * caller must ensure that no other threads can modify the inode until a call @@ -434,7 +443,7 @@ xchk_iscan_iget( ASSERT(iscan->__inodes[0] == NULL); /* Fill the first slot in the inode array. */ - error = xfs_iget(sc->mp, sc->tp, ino, XFS_IGET_NORETRY, 0, + error = xfs_iget(sc->mp, sc->tp, ino, ISCAN_IGET_FLAGS, 0, &iscan->__inodes[idx]); trace_xchk_iscan_iget(iscan, error); @@ -507,7 +516,7 @@ xchk_iscan_iget( ASSERT(iscan->__inodes[idx] == NULL); - error = xfs_iget(sc->mp, sc->tp, ino, XFS_IGET_NORETRY, 0, + error = xfs_iget(sc->mp, sc->tp, ino, ISCAN_IGET_FLAGS, 0, &iscan->__inodes[idx]); if (error) break; diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index 4e7e3edb6350..1da10182f7f4 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -60,6 +60,13 @@ static inline int xchk_maybe_relax(struct xchk_relax *widget) #define XCHK_GFP_FLAGS ((__force gfp_t)(GFP_KERNEL | __GFP_NOWARN | \ __GFP_RETRY_MAYFAIL)) +/* + * For opening files by handle for fsck operations, we don't trust the inumber + * or the allocation state; therefore, perform an untrusted lookup. We don't + * want these inodes to pollute the cache, so mark them for immediate removal. + */ +#define XCHK_IGET_FLAGS (XFS_IGET_UNTRUSTED | XFS_IGET_DONTCACHE) + /* Type info and names for the scrub types. */ enum xchk_type { ST_NONE = 1, /* disabled */ -- cgit v1.2.3-59-g8ed1b From 4ad350ac58627bfe81f71f43f6738e36b4eb75c6 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:26 -0700 Subject: xfs: only iget the file once when doing vectored scrub-by-handle If a program wants us to perform a scrub on a file handle and the fd passed to ioctl() is not the file referenced in the handle, iget the file once and pass it into the scrub code. This amortizes the untrusted iget lookup over /all/ the scrubbers mentioned in the scrubv call. When running fstests in "rebuild all metadata after each test" mode, I observed a 10% reduction in runtime on account of avoiding repeated inobt lookups. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/scrub.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+) (limited to 'fs') diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 78b00ab85c9c..43af5ce1f99f 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -792,6 +792,37 @@ xfs_scrubv_check_barrier( return 0; } +/* + * If the caller provided us with a nonzero inode number that isn't the ioctl + * file, try to grab a reference to it to eliminate all further untrusted inode + * lookups. If we can't get the inode, let each scrub function try again. + */ +STATIC struct xfs_inode * +xchk_scrubv_open_by_handle( + struct xfs_mount *mp, + const struct xfs_scrub_vec_head *head) +{ + struct xfs_trans *tp; + struct xfs_inode *ip; + int error; + + error = xfs_trans_alloc_empty(mp, &tp); + if (error) + return NULL; + + error = xfs_iget(mp, tp, head->svh_ino, XCHK_IGET_FLAGS, 0, &ip); + xfs_trans_cancel(tp); + if (error) + return NULL; + + if (VFS_I(ip)->i_generation != head->svh_gen) { + xfs_irele(ip); + return NULL; + } + + return ip; +} + /* Vectored scrub implementation to reduce ioctl calls. */ int xfs_ioc_scrubv_metadata( @@ -804,6 +835,7 @@ xfs_ioc_scrubv_metadata( struct xfs_scrub_vec __user *uvectors; struct xfs_inode *ip_in = XFS_I(file_inode(file)); struct xfs_mount *mp = ip_in->i_mount; + struct xfs_inode *handle_ip = NULL; struct xfs_scrub_vec *v; size_t vec_bytes; unsigned int i; @@ -848,6 +880,17 @@ xfs_ioc_scrubv_metadata( trace_xchk_scrubv_item(mp, &head, i, v); } + /* + * If the caller wants us to do a scrub-by-handle and the file used to + * call the ioctl is not the same file, load the incore inode and pin + * it across all the scrubv actions to avoid repeated UNTRUSTED + * lookups. The reference is not passed to deeper layers of scrub + * because each scrubber gets to decide its own strategy and return + * values for getting an inode. + */ + if (head.svh_ino && head.svh_ino != ip_in->i_ino) + handle_ip = xchk_scrubv_open_by_handle(mp, &head); + /* Run all the scrubbers. */ for (i = 0, v = vectors; i < head.svh_nr; i++, v++) { struct xfs_scrub_metadata sm = { @@ -895,6 +938,8 @@ xfs_ioc_scrubv_metadata( } out_free: + if (handle_ip) + xfs_irele(handle_ip); kfree(vectors); return error; } -- cgit v1.2.3-59-g8ed1b From 66917537522312a6e462787f0f347d6998cf7038 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:27 -0700 Subject: xfs: drop the scrub file's iolock when transaction allocation fails If the transaction allocation in the !orphanage_available case of xrep_nlinks_repair_inode fails, we need to drop the IOLOCK of the file being scrubbed before exiting. Found by fuzzing u3.sfdir3.list[1].name = zeroes in xfs/1546. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/nlinks_repair.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/nlinks_repair.c b/fs/xfs/scrub/nlinks_repair.c index 78d0f650fe89..b3e707f47b7b 100644 --- a/fs/xfs/scrub/nlinks_repair.c +++ b/fs/xfs/scrub/nlinks_repair.c @@ -138,8 +138,10 @@ xrep_nlinks_repair_inode( error = xfs_trans_alloc(mp, &M_RES(mp)->tr_link, 0, 0, 0, &sc->tp); - if (error) + if (error) { + xchk_iunlock(sc, XFS_IOLOCK_EXCL); return error; + } xchk_ilock(sc, XFS_ILOCK_EXCL); xfs_trans_ijoin(sc->tp, ip, 0); -- cgit v1.2.3-59-g8ed1b From b44bfc06958f49ccb611632a4fb7c7df4fdcbc06 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:28 -0700 Subject: xfs: fix iunlock calls in xrep_adoption_trans_alloc If the transaction allocation in xrep_adoption_trans_alloc fails, we should drop only the locks that we took. In this case this is ILOCK_EXCL of both the orphanage and the file being repaired. Dropping any IOLOCK here is incorrect. Found by fuzzing u3.sfdir3.list[1].name = zeroes in xfs/1546. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/orphanage.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index b1c6c60ee1da..2b142e6de8f3 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -382,7 +382,7 @@ xrep_adoption_trans_alloc( out_cancel: xchk_trans_cancel(sc); xrep_orphanage_iunlock(sc, XFS_ILOCK_EXCL); - xrep_orphanage_iunlock(sc, XFS_IOLOCK_EXCL); + xchk_iunlock(sc, XFS_ILOCK_EXCL); return error; } -- cgit v1.2.3-59-g8ed1b From 6d335233fe6952189c949d65ab16d92afb0b8fb4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:29 -0700 Subject: xfs: exchange-range for repairs is no longer dynamic The atomic file exchange-range functionality is now a permanent filesystem feature instead of a dynamic log-incompat feature. It cannot be turned on at runtime, so we no longer need the XCHK_FSGATES flags and whatnot that supported it. Remove the flag and the enable function, and move the xfs_has_exchange_range checks to the start of the repair functions. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/attr_repair.c | 3 +++ fs/xfs/scrub/dir_repair.c | 3 +++ fs/xfs/scrub/parent_repair.c | 10 +++++++--- fs/xfs/scrub/rtsummary_repair.c | 10 ++++------ fs/xfs/scrub/scrub.c | 8 +++----- fs/xfs/scrub/scrub.h | 7 ------- fs/xfs/scrub/symlink_repair.c | 3 +++ fs/xfs/scrub/tempexch.h | 1 - fs/xfs/scrub/tempfile.c | 24 ++---------------------- fs/xfs/scrub/trace.h | 1 - 10 files changed, 25 insertions(+), 45 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/attr_repair.c b/fs/xfs/scrub/attr_repair.c index e059813b92b7..c7eb94069caf 100644 --- a/fs/xfs/scrub/attr_repair.c +++ b/fs/xfs/scrub/attr_repair.c @@ -1630,6 +1630,9 @@ xrep_xattr( /* The rmapbt is required to reap the old attr fork. */ if (!xfs_has_rmapbt(sc->mp)) return -EOPNOTSUPP; + /* We require atomic file exchange range to rebuild anything. */ + if (!xfs_has_exchange_range(sc->mp)) + return -EOPNOTSUPP; error = xrep_xattr_setup_scan(sc, &rx); if (error) diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index e968150fe0f0..6ad40f8aafb8 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -1993,6 +1993,9 @@ xrep_directory( /* The rmapbt is required to reap the old data fork. */ if (!xfs_has_rmapbt(sc->mp)) return -EOPNOTSUPP; + /* We require atomic file exchange range to rebuild anything. */ + if (!xfs_has_exchange_range(sc->mp)) + return -EOPNOTSUPP; error = xrep_dir_setup_scan(rd); if (error) diff --git a/fs/xfs/scrub/parent_repair.c b/fs/xfs/scrub/parent_repair.c index ee88ce5a12b8..7b42b7f65a0b 100644 --- a/fs/xfs/scrub/parent_repair.c +++ b/fs/xfs/scrub/parent_repair.c @@ -1571,10 +1571,14 @@ xrep_parent( /* * When the parent pointers feature is enabled, repairs are committed * by atomically committing a new xattr structure and reaping the old - * attr fork. Reaping requires rmap to be enabled. + * attr fork. Reaping requires rmap and exchange-range to be enabled. */ - if (xfs_has_parent(sc->mp) && !xfs_has_rmapbt(sc->mp)) - return -EOPNOTSUPP; + if (xfs_has_parent(sc->mp)) { + if (!xfs_has_rmapbt(sc->mp)) + return -EOPNOTSUPP; + if (!xfs_has_exchange_range(sc->mp)) + return -EOPNOTSUPP; + } error = xrep_parent_setup_scan(rp); if (error) diff --git a/fs/xfs/scrub/rtsummary_repair.c b/fs/xfs/scrub/rtsummary_repair.c index c8bb6c4f15d0..d9e971c4c79f 100644 --- a/fs/xfs/scrub/rtsummary_repair.c +++ b/fs/xfs/scrub/rtsummary_repair.c @@ -62,12 +62,7 @@ xrep_setup_rtsummary( return -EOPNOTSUPP; rts->resblks += blocks; - - /* - * Grab support for atomic file content exchanges before we allocate - * any transactions or grab ILOCKs. - */ - return xrep_tempexch_enable(sc); + return 0; } static int @@ -111,6 +106,9 @@ xrep_rtsummary( /* We require the rmapbt to rebuild anything. */ if (!xfs_has_rmapbt(mp)) return -EOPNOTSUPP; + /* We require atomic file exchange range to rebuild anything. */ + if (!xfs_has_exchange_range(mp)) + return -EOPNOTSUPP; /* Walk away if we disagree on the size of the rt bitmap. */ if (rts->rbmblocks != mp->m_sb.sb_rbmblocks) diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 43af5ce1f99f..c013f0ba4f36 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -154,15 +154,14 @@ xchk_probe( /* Scrub setup and teardown */ -#define FSGATES_MASK (XCHK_FSGATES_ALL | XREP_FSGATES_ALL) static inline void xchk_fsgates_disable( struct xfs_scrub *sc) { - if (!(sc->flags & FSGATES_MASK)) + if (!(sc->flags & XCHK_FSGATES_ALL)) return; - trace_xchk_fsgates_disable(sc, sc->flags & FSGATES_MASK); + trace_xchk_fsgates_disable(sc, sc->flags & XCHK_FSGATES_ALL); if (sc->flags & XCHK_FSGATES_DRAIN) xfs_drain_wait_disable(); @@ -176,9 +175,8 @@ xchk_fsgates_disable( if (sc->flags & XCHK_FSGATES_RMAP) xfs_rmap_hook_disable(); - sc->flags &= ~FSGATES_MASK; + sc->flags &= ~XCHK_FSGATES_ALL; } -#undef FSGATES_MASK /* Free the resources associated with a scrub subtype. */ void diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index 1da10182f7f4..1bc33f010d0e 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -188,7 +188,6 @@ struct xfs_scrub { #define XCHK_FSGATES_QUOTA (1U << 4) /* quota live update enabled */ #define XCHK_FSGATES_DIRENTS (1U << 5) /* directory live update enabled */ #define XCHK_FSGATES_RMAP (1U << 6) /* rmapbt live update enabled */ -#define XREP_FSGATES_EXCHANGE_RANGE (1U << 29) /* uses file content exchange */ #define XREP_RESET_PERAG_RESV (1U << 30) /* must reset AG space reservation */ #define XREP_ALREADY_FIXED (1U << 31) /* checking our repair work */ @@ -203,12 +202,6 @@ struct xfs_scrub { XCHK_FSGATES_DIRENTS | \ XCHK_FSGATES_RMAP) -/* - * The sole XREP_FSGATES* flag reflects a log intent item that is protected - * by a log-incompat feature flag. No code patching in use here. - */ -#define XREP_FSGATES_ALL (XREP_FSGATES_EXCHANGE_RANGE) - struct xfs_scrub_subord { struct xfs_scrub sc; struct xfs_scrub *parent_sc; diff --git a/fs/xfs/scrub/symlink_repair.c b/fs/xfs/scrub/symlink_repair.c index c8b5a5b878ac..d015a86ef460 100644 --- a/fs/xfs/scrub/symlink_repair.c +++ b/fs/xfs/scrub/symlink_repair.c @@ -490,6 +490,9 @@ xrep_symlink( /* The rmapbt is required to reap the old data fork. */ if (!xfs_has_rmapbt(sc->mp)) return -EOPNOTSUPP; + /* We require atomic file exchange range to rebuild anything. */ + if (!xfs_has_exchange_range(sc->mp)) + return -EOPNOTSUPP; ASSERT(sc->ilock_flags & XFS_ILOCK_EXCL); diff --git a/fs/xfs/scrub/tempexch.h b/fs/xfs/scrub/tempexch.h index c1dd4adec4f1..995ba187c5aa 100644 --- a/fs/xfs/scrub/tempexch.h +++ b/fs/xfs/scrub/tempexch.h @@ -11,7 +11,6 @@ struct xrep_tempexch { struct xfs_exchmaps_req req; }; -int xrep_tempexch_enable(struct xfs_scrub *sc); int xrep_tempexch_trans_reserve(struct xfs_scrub *sc, int whichfork, struct xrep_tempexch *ti); int xrep_tempexch_trans_alloc(struct xfs_scrub *sc, int whichfork, diff --git a/fs/xfs/scrub/tempfile.c b/fs/xfs/scrub/tempfile.c index ddbcccb3dba1..b747b625c5ee 100644 --- a/fs/xfs/scrub/tempfile.c +++ b/fs/xfs/scrub/tempfile.c @@ -486,23 +486,6 @@ xrep_tempfile_roll_trans( return 0; } -/* Enable file content exchanges. */ -int -xrep_tempexch_enable( - struct xfs_scrub *sc) -{ - if (sc->flags & XREP_FSGATES_EXCHANGE_RANGE) - return 0; - - if (!xfs_has_exchange_range(sc->mp)) - return -EOPNOTSUPP; - - trace_xchk_fsgates_enable(sc, XREP_FSGATES_EXCHANGE_RANGE); - - sc->flags |= XREP_FSGATES_EXCHANGE_RANGE; - return 0; -} - /* * Fill out the mapping exchange request in preparation for atomically * committing the contents of a metadata file that we've rebuilt in the temp @@ -745,6 +728,7 @@ xrep_tempexch_trans_alloc( int error; ASSERT(sc->tp == NULL); + ASSERT(xfs_has_exchange_range(sc->mp)); error = xrep_tempexch_prep_request(sc, whichfork, tx); if (error) @@ -757,10 +741,6 @@ xrep_tempexch_trans_alloc( if (xfs_has_lazysbcount(sc->mp)) flags |= XFS_TRANS_RES_FDBLKS; - error = xrep_tempexch_enable(sc); - if (error) - return error; - error = xfs_trans_alloc(sc->mp, &M_RES(sc->mp)->tr_itruncate, tx->req.resblks, 0, flags, &sc->tp); if (error) @@ -785,7 +765,7 @@ xrep_tempexch_contents( { int error; - ASSERT(sc->flags & XREP_FSGATES_EXCHANGE_RANGE); + ASSERT(xfs_has_exchange_range(sc->mp)); xfs_exchange_mappings(sc->tp, &tx->req); error = xfs_defer_finish(&sc->tp); diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 8ce74bd8530a..4b945007ca6c 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -122,7 +122,6 @@ TRACE_DEFINE_ENUM(XFS_SCRUB_TYPE_BARRIER); { XCHK_FSGATES_QUOTA, "fsgates_quota" }, \ { XCHK_FSGATES_DIRENTS, "fsgates_dirents" }, \ { XCHK_FSGATES_RMAP, "fsgates_rmap" }, \ - { XREP_FSGATES_EXCHANGE_RANGE, "fsgates_exchrange" }, \ { XREP_RESET_PERAG_RESV, "reset_perag_resv" }, \ { XREP_ALREADY_FIXED, "already_fixed" } -- cgit v1.2.3-59-g8ed1b From 5e1c7d0b29f7e05b01e448d2579a469cf3a0d350 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Mon, 22 Apr 2024 09:48:30 -0700 Subject: xfs: invalidate dentries for a file before moving it to the orphanage Invalidate the cached dentries that point to the file that we're moving to lost+found before we actually move it. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig --- fs/xfs/scrub/orphanage.c | 47 ++++++++++++++++++++--------------------------- fs/xfs/scrub/trace.h | 2 -- 2 files changed, 20 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index 2b142e6de8f3..7148d8362db8 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -434,16 +434,17 @@ xrep_adoption_check_dcache( { struct qstr qname = QSTR_INIT(adopt->xname->name, adopt->xname->len); + struct xfs_scrub *sc = adopt->sc; struct dentry *d_orphanage, *d_child; int error = 0; - d_orphanage = d_find_alias(VFS_I(adopt->sc->orphanage)); + d_orphanage = d_find_alias(VFS_I(sc->orphanage)); if (!d_orphanage) return 0; d_child = d_hash_and_lookup(d_orphanage, &qname); if (d_child) { - trace_xrep_adoption_check_child(adopt->sc->mp, d_child); + trace_xrep_adoption_check_child(sc->mp, d_child); if (d_is_positive(d_child)) { ASSERT(d_is_negative(d_child)); @@ -454,33 +455,15 @@ xrep_adoption_check_dcache( } dput(d_orphanage); - if (error) - return error; - - /* - * Do we need to update d_parent of the dentry for the file being - * repaired? There shouldn't be a hashed dentry with a parent since - * the file had nonzero nlink but wasn't connected to any parent dir. - */ - d_child = d_find_alias(VFS_I(adopt->sc->ip)); - if (!d_child) - return 0; - - trace_xrep_adoption_check_alias(adopt->sc->mp, d_child); - - if (d_child->d_parent && !d_unhashed(d_child)) { - ASSERT(d_child->d_parent == NULL || d_unhashed(d_child)); - error = -EFSCORRUPTED; - } - - dput(d_child); return error; } /* - * Remove all negative dentries from the dcache. There should not be any - * positive entries, since we've maintained our lock on the orphanage - * directory. + * Invalidate all dentries for the name that was added to the orphanage + * directory, and all dentries pointing to the child inode that was moved. + * + * There should not be any positive entries for the name, since we've + * maintained our lock on the orphanage directory. */ static void xrep_adoption_zap_dcache( @@ -488,15 +471,17 @@ xrep_adoption_zap_dcache( { struct qstr qname = QSTR_INIT(adopt->xname->name, adopt->xname->len); + struct xfs_scrub *sc = adopt->sc; struct dentry *d_orphanage, *d_child; - d_orphanage = d_find_alias(VFS_I(adopt->sc->orphanage)); + /* Invalidate all dentries for the adoption name */ + d_orphanage = d_find_alias(VFS_I(sc->orphanage)); if (!d_orphanage) return; d_child = d_hash_and_lookup(d_orphanage, &qname); while (d_child != NULL) { - trace_xrep_adoption_invalidate_child(adopt->sc->mp, d_child); + trace_xrep_adoption_invalidate_child(sc->mp, d_child); ASSERT(d_is_negative(d_child)); d_invalidate(d_child); @@ -505,6 +490,14 @@ xrep_adoption_zap_dcache( } dput(d_orphanage); + + /* Invalidate all the dentries pointing down to this file. */ + while ((d_child = d_find_alias(VFS_I(sc->ip))) != NULL) { + trace_xrep_adoption_invalidate_child(sc->mp, d_child); + + d_invalidate(d_child); + dput(d_child); + } } /* diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 4b945007ca6c..e27daa51cab6 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -3265,8 +3265,6 @@ DEFINE_EVENT(xrep_dentry_class, name, \ TP_PROTO(struct xfs_mount *mp, const struct dentry *dentry), \ TP_ARGS(mp, dentry)) DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_child); -DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_alias); -DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_check_dentry); DEFINE_REPAIR_DENTRY_EVENT(xrep_adoption_invalidate_child); DEFINE_REPAIR_DENTRY_EVENT(xrep_dirtree_delete_child); -- cgit v1.2.3-59-g8ed1b From 9c8ecb9308d8013ff9ac9d36fdd8ae746033b93c Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 23 Apr 2024 17:48:43 -0400 Subject: Revert "NFSD: Reschedule CB operations when backchannel rpc_clnt is shut down" The reverted commit attempted to enable NFSD to retransmit pending callback operations if an NFS client disconnects, but unintentionally introduces a hazardous behavior regression if the client becomes permanently unreachable while callback operations are still pending. A disconnect can occur due to network partition or if the NFS server needs to force the NFS client to retransmit (for example, if a GSS window under-run occurs). Reverting the commit will make NFSD behave the same as it did in v6.8 and before. Pending callback operations are permanently lost if the client connection is terminated before the client receives them. For some callback operations, this loss is not harmful. However, for CB_RECALL, the loss means a delegation might be revoked unnecessarily. For CB_OFFLOAD, pending COPY operations will never complete unless the NFS client subsequently sends an OFFLOAD_STATUS operation, which the Linux NFS client does not currently implement. These issues still need to be addressed somehow. Reported-by: Dai Ngo Link: https://bugzilla.kernel.org/show_bug.cgi?id=218735 Signed-off-by: Chuck Lever --- fs/nfsd/nfs4callback.c | 20 ++------------------ 1 file changed, 2 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index e440f72b9d4e..d153af81f406 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -986,14 +986,6 @@ static bool nfsd4_queue_cb(struct nfsd4_callback *cb) return queue_delayed_work(callback_wq, &cb->cb_work, 0); } -static void nfsd4_queue_cb_delayed(struct nfsd4_callback *cb, - unsigned long msecs) -{ - trace_nfsd_cb_queue(cb->cb_clp, cb); - queue_delayed_work(callback_wq, &cb->cb_work, - msecs_to_jiffies(msecs)); -} - static void nfsd41_cb_inflight_begin(struct nfs4_client *clp) { atomic_inc(&clp->cl_cb_inflight); @@ -1502,16 +1494,8 @@ nfsd4_run_cb_work(struct work_struct *work) clnt = clp->cl_cb_client; if (!clnt) { - if (test_bit(NFSD4_CLIENT_CB_KILL, &clp->cl_flags)) - nfsd41_destroy_cb(cb); - else { - /* - * XXX: Ideally, we could wait for the client to - * reconnect, but I haven't figured out how - * to do that yet. - */ - nfsd4_queue_cb_delayed(cb, 25); - } + /* Callback channel broken, or client killed; give up: */ + nfsd41_destroy_cb(cb); return; } -- cgit v1.2.3-59-g8ed1b From 8ddb7142c8ab37371c6fd167a8aded97922c6268 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 23 Apr 2024 17:52:24 -0400 Subject: Revert "NFSD: Convert the callback workqueue to use delayed_work" This commit was a pre-requisite for commit c1ccfcf1a9bf ("NFSD: Reschedule CB operations when backchannel rpc_clnt is shut down"), which has already been reverted. Signed-off-by: Chuck Lever --- fs/nfsd/nfs4callback.c | 6 +++--- fs/nfsd/state.h | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index d153af81f406..6aff46701aa1 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -983,7 +983,7 @@ static struct workqueue_struct *callback_wq; static bool nfsd4_queue_cb(struct nfsd4_callback *cb) { trace_nfsd_cb_queue(cb->cb_clp, cb); - return queue_delayed_work(callback_wq, &cb->cb_work, 0); + return queue_work(callback_wq, &cb->cb_work); } static void nfsd41_cb_inflight_begin(struct nfs4_client *clp) @@ -1482,7 +1482,7 @@ static void nfsd4_run_cb_work(struct work_struct *work) { struct nfsd4_callback *cb = - container_of(work, struct nfsd4_callback, cb_work.work); + container_of(work, struct nfsd4_callback, cb_work); struct nfs4_client *clp = cb->cb_clp; struct rpc_clnt *clnt; int flags; @@ -1528,7 +1528,7 @@ void nfsd4_init_cb(struct nfsd4_callback *cb, struct nfs4_client *clp, cb->cb_msg.rpc_argp = cb; cb->cb_msg.rpc_resp = cb; cb->cb_ops = ops; - INIT_DELAYED_WORK(&cb->cb_work, nfsd4_run_cb_work); + INIT_WORK(&cb->cb_work, nfsd4_run_cb_work); cb->cb_status = 0; cb->cb_need_restart = false; cb->cb_holds_slot = false; diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index 01c6f3445646..2ed0fcf879fd 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -68,7 +68,7 @@ struct nfsd4_callback { struct nfs4_client *cb_clp; struct rpc_message cb_msg; const struct nfsd4_callback_ops *cb_ops; - struct delayed_work cb_work; + struct work_struct cb_work; int cb_seq_status; int cb_status; bool cb_need_restart; -- cgit v1.2.3-59-g8ed1b From 9a1f1d04f63c59550a5364858b46eeffdf03e8d6 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 23 Apr 2024 20:41:22 -0600 Subject: smb: client: Fix struct_group() usage in __packed structs Use struct_group_attr() in __packed structs, instead of struct_group(). Below you can see the pahole output before/after changes: pahole -C smb2_file_network_open_info fs/smb/client/smb2ops.o struct smb2_file_network_open_info { union { struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le64 AllocationSize; /* 32 8 */ __le64 EndOfFile; /* 40 8 */ __le32 Attributes; /* 48 4 */ }; /* 0 56 */ struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le64 AllocationSize; /* 32 8 */ __le64 EndOfFile; /* 40 8 */ __le32 Attributes; /* 48 4 */ } network_open_info; /* 0 56 */ }; /* 0 56 */ __le32 Reserved; /* 56 4 */ /* size: 60, cachelines: 1, members: 2 */ /* last cacheline: 60 bytes */ } __attribute__((__packed__)); pahole -C smb2_file_network_open_info fs/smb/client/smb2ops.o struct smb2_file_network_open_info { union { struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le64 AllocationSize; /* 32 8 */ __le64 EndOfFile; /* 40 8 */ __le32 Attributes; /* 48 4 */ } __attribute__((__packed__)); /* 0 52 */ struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le64 AllocationSize; /* 32 8 */ __le64 EndOfFile; /* 40 8 */ __le32 Attributes; /* 48 4 */ } __attribute__((__packed__)) network_open_info; /* 0 52 */ }; /* 0 52 */ __le32 Reserved; /* 52 4 */ /* size: 56, cachelines: 1, members: 2 */ /* last cacheline: 56 bytes */ }; pahole -C smb_com_open_rsp fs/smb/client/cifssmb.o struct smb_com_open_rsp { ... union { struct { __le64 CreationTime; /* 48 8 */ __le64 LastAccessTime; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ __le64 LastWriteTime; /* 64 8 */ __le64 ChangeTime; /* 72 8 */ __le32 FileAttributes; /* 80 4 */ }; /* 48 40 */ struct { __le64 CreationTime; /* 48 8 */ __le64 LastAccessTime; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ __le64 LastWriteTime; /* 64 8 */ __le64 ChangeTime; /* 72 8 */ __le32 FileAttributes; /* 80 4 */ } common_attributes; /* 48 40 */ }; /* 48 40 */ ... /* size: 111, cachelines: 2, members: 14 */ /* last cacheline: 47 bytes */ } __attribute__((__packed__)); pahole -C smb_com_open_rsp fs/smb/client/cifssmb.o struct smb_com_open_rsp { ... union { struct { __le64 CreationTime; /* 48 8 */ __le64 LastAccessTime; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ __le64 LastWriteTime; /* 64 8 */ __le64 ChangeTime; /* 72 8 */ __le32 FileAttributes; /* 80 4 */ } __attribute__((__packed__)); /* 48 36 */ struct { __le64 CreationTime; /* 48 8 */ __le64 LastAccessTime; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ __le64 LastWriteTime; /* 64 8 */ __le64 ChangeTime; /* 72 8 */ __le32 FileAttributes; /* 80 4 */ } __attribute__((__packed__)) common_attributes; /* 48 36 */ }; /* 48 36 */ ... /* size: 107, cachelines: 2, members: 14 */ /* last cacheline: 43 bytes */ } __attribute__((__packed__)); pahole -C FILE_ALL_INFO fs/smb/client/cifssmb.o typedef struct { union { struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le32 Attributes; /* 32 4 */ }; /* 0 40 */ struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le32 Attributes; /* 32 4 */ } common_attributes; /* 0 40 */ }; /* 0 40 */ ... /* size: 113, cachelines: 2, members: 17 */ /* last cacheline: 49 bytes */ } __attribute__((__packed__)) FILE_ALL_INFO; pahole -C FILE_ALL_INFO fs/smb/client/cifssmb.o typedef struct { union { struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le32 Attributes; /* 32 4 */ } __attribute__((__packed__)); /* 0 36 */ struct { __le64 CreationTime; /* 0 8 */ __le64 LastAccessTime; /* 8 8 */ __le64 LastWriteTime; /* 16 8 */ __le64 ChangeTime; /* 24 8 */ __le32 Attributes; /* 32 4 */ } __attribute__((__packed__)) common_attributes; /* 0 36 */ }; /* 0 36 */ ... /* size: 109, cachelines: 2, members: 17 */ /* last cacheline: 45 bytes */ } __attribute__((__packed__)) FILE_ALL_INFO; Fixes: 0015eb6e1238 ("smb: client, common: fix fortify warnings") Cc: stable@vger.kernel.org Reviewed-by: Namjae Jeon Signed-off-by: Gustavo A. R. Silva Signed-off-by: Steve French --- fs/smb/client/cifspdu.h | 4 ++-- fs/smb/client/smb2pdu.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifspdu.h b/fs/smb/client/cifspdu.h index c0513fbb8a59..c46d418c1c0c 100644 --- a/fs/smb/client/cifspdu.h +++ b/fs/smb/client/cifspdu.h @@ -882,7 +882,7 @@ typedef struct smb_com_open_rsp { __u8 OplockLevel; __u16 Fid; __le32 CreateAction; - struct_group(common_attributes, + struct_group_attr(common_attributes, __packed, __le64 CreationTime; __le64 LastAccessTime; __le64 LastWriteTime; @@ -2266,7 +2266,7 @@ typedef struct { /* QueryFileInfo/QueryPathinfo (also for SetPath/SetFile) data buffer formats */ /******************************************************************************/ typedef struct { /* data block encoding of response to level 263 QPathInfo */ - struct_group(common_attributes, + struct_group_attr(common_attributes, __packed, __le64 CreationTime; __le64 LastAccessTime; __le64 LastWriteTime; diff --git a/fs/smb/client/smb2pdu.h b/fs/smb/client/smb2pdu.h index c72a3b2886b7..2fccf0d4f53d 100644 --- a/fs/smb/client/smb2pdu.h +++ b/fs/smb/client/smb2pdu.h @@ -320,7 +320,7 @@ struct smb2_file_reparse_point_info { } __packed; struct smb2_file_network_open_info { - struct_group(network_open_info, + struct_group_attr(network_open_info, __packed, __le64 CreationTime; __le64 LastAccessTime; __le64 LastWriteTime; -- cgit v1.2.3-59-g8ed1b From 08e012a62de877e77d7d44d5bddace63d760741b Mon Sep 17 00:00:00 2001 From: Jiapeng Chong Date: Wed, 24 Apr 2024 10:06:38 +0800 Subject: xfs: Remove unused function xrep_dir_self_parent The function are defined in the dir_repair.c file, but not called elsewhere, so delete the unused function. fs/xfs/scrub/dir_repair.c:186:1: warning: unused function 'xrep_dir_self_parent'. Reported-by: Abaci Robot Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=8867 Signed-off-by: Jiapeng Chong Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/scrub/dir_repair.c | 21 --------------------- 1 file changed, 21 deletions(-) (limited to 'fs') diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 6ad40f8aafb8..fa1a9954d48d 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -204,27 +204,6 @@ xrep_setup_directory( return 0; } -/* - * If we're the root of a directory tree, we are our own parent. If we're an - * unlinked directory, the parent /won't/ have a link to us. Set the parent - * directory to the root for both cases. Returns NULLFSINO if we don't know - * what to do. - */ -static inline xfs_ino_t -xrep_dir_self_parent( - struct xrep_dir *rd) -{ - struct xfs_scrub *sc = rd->sc; - - if (sc->ip->i_ino == sc->mp->m_sb.sb_rootino) - return sc->mp->m_sb.sb_rootino; - - if (VFS_I(sc->ip)->i_nlink == 0) - return sc->mp->m_sb.sb_rootino; - - return NULLFSINO; -} - /* * Look up the dotdot entry and confirm that it's really the parent. * Returns NULLFSINO if we don't know what to do. -- cgit v1.2.3-59-g8ed1b From 520713a93d550406dae14d49cdb8778d70cecdfd Mon Sep 17 00:00:00 2001 From: Thomas Weißschuh Date: Fri, 15 Mar 2024 19:11:30 +0100 Subject: sysctl: treewide: drop unused argument ctl_table_root::set_ownership(table) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove the 'table' argument from set_ownership as it is never used. This change is a step towards putting "struct ctl_table" into .rodata and eventually having sysctl core only use "const struct ctl_table". The patch was created with the following coccinelle script: @@ identifier func, head, table, uid, gid; @@ void func( struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { ... } No additional occurrences of 'set_ownership' were found after doing a tree-wide search. Reviewed-by: Joel Granados Signed-off-by: Thomas Weißschuh Signed-off-by: Joel Granados --- fs/proc/proc_sysctl.c | 2 +- include/linux/sysctl.h | 1 - ipc/ipc_sysctl.c | 3 +-- ipc/mq_sysctl.c | 3 +-- net/sysctl_net.c | 1 - 5 files changed, 3 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 37cde0efee57..ed3a41ed9705 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -480,7 +480,7 @@ static struct inode *proc_sys_make_inode(struct super_block *sb, } if (root->set_ownership) - root->set_ownership(head, table, &inode->i_uid, &inode->i_gid); + root->set_ownership(head, &inode->i_uid, &inode->i_gid); else { inode->i_uid = GLOBAL_ROOT_UID; inode->i_gid = GLOBAL_ROOT_GID; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index ee7d33b89e9e..60333a6b9370 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -205,7 +205,6 @@ struct ctl_table_root { struct ctl_table_set default_set; struct ctl_table_set *(*lookup)(struct ctl_table_root *root); void (*set_ownership)(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid); int (*permissions)(struct ctl_table_header *head, struct ctl_table *table); }; diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c index 45cb1dabce29..1a5085e5b178 100644 --- a/ipc/ipc_sysctl.c +++ b/ipc/ipc_sysctl.c @@ -192,7 +192,6 @@ static int set_is_seen(struct ctl_table_set *set) } static void ipc_set_ownership(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { struct ipc_namespace *ns = @@ -224,7 +223,7 @@ static int ipc_permissions(struct ctl_table_header *head, struct ctl_table *tabl kuid_t ns_root_uid; kgid_t ns_root_gid; - ipc_set_ownership(head, table, &ns_root_uid, &ns_root_gid); + ipc_set_ownership(head, &ns_root_uid, &ns_root_gid); if (uid_eq(current_euid(), ns_root_uid)) mode >>= 6; diff --git a/ipc/mq_sysctl.c b/ipc/mq_sysctl.c index 21fba3a6edaf..6bb1c5397c69 100644 --- a/ipc/mq_sysctl.c +++ b/ipc/mq_sysctl.c @@ -78,7 +78,6 @@ static int set_is_seen(struct ctl_table_set *set) } static void mq_set_ownership(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { struct ipc_namespace *ns = @@ -97,7 +96,7 @@ static int mq_permissions(struct ctl_table_header *head, struct ctl_table *table kuid_t ns_root_uid; kgid_t ns_root_gid; - mq_set_ownership(head, table, &ns_root_uid, &ns_root_gid); + mq_set_ownership(head, &ns_root_uid, &ns_root_gid); if (uid_eq(current_euid(), ns_root_uid)) mode >>= 6; diff --git a/net/sysctl_net.c b/net/sysctl_net.c index 051ed5f6fc93..a0a7a79991f9 100644 --- a/net/sysctl_net.c +++ b/net/sysctl_net.c @@ -54,7 +54,6 @@ static int net_ctl_permissions(struct ctl_table_header *head, } static void net_ctl_set_ownership(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { struct net *net = container_of(head->set, struct net, sysctls); -- cgit v1.2.3-59-g8ed1b From eb32d3adef461bc523c97321f19209f55512ee4a Mon Sep 17 00:00:00 2001 From: Thomas Weißschuh Date: Fri, 22 Mar 2024 18:05:56 +0100 Subject: sysctl: drop sysctl_is_perm_empty_ctl_table MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It is used only twice and those callers are simpler with sysctl_is_perm_empty_ctl_header(). So use this sibling function. This is part of an effort to constify definition of struct ctl_table. For this effort the mutable member 'type' is moved from struct ctl_table to struct ctl_table_header. Unifying the macros sysctl_is_perm_empty_ctl_* makes this easier. Signed-off-by: Thomas Weißschuh Signed-off-by: Joel Granados --- fs/proc/proc_sysctl.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index ed3a41ed9705..3f8abb6d2ad5 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -48,10 +48,8 @@ struct ctl_table_header *register_sysctl_mount_point(const char *path) } EXPORT_SYMBOL(register_sysctl_mount_point); -#define sysctl_is_perm_empty_ctl_table(tptr) \ - (tptr[0].type == SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY) #define sysctl_is_perm_empty_ctl_header(hptr) \ - (sysctl_is_perm_empty_ctl_table(hptr->ctl_table)) + (hptr->ctl_table[0].type == SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY) #define sysctl_set_perm_empty_ctl_header(hptr) \ (hptr->ctl_table[0].type = SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY) #define sysctl_clear_perm_empty_ctl_header(hptr) \ @@ -233,7 +231,7 @@ static int insert_header(struct ctl_dir *dir, struct ctl_table_header *header) /* Am I creating a permanently empty directory? */ if (header->ctl_table_size > 0 && - sysctl_is_perm_empty_ctl_table(header->ctl_table)) { + sysctl_is_perm_empty_ctl_header(header)) { if (!RB_EMPTY_ROOT(&dir->root)) return -EINVAL; sysctl_set_perm_empty_ctl_header(dir_h); @@ -1204,7 +1202,7 @@ static bool get_links(struct ctl_dir *dir, struct ctl_table *entry, *link; if (header->ctl_table_size == 0 || - sysctl_is_perm_empty_ctl_table(header->ctl_table)) + sysctl_is_perm_empty_ctl_header(header)) return true; /* Are there links available for every entry in table? */ -- cgit v1.2.3-59-g8ed1b From 4a7b29f6509470a7e63dd67e516268fbef226194 Mon Sep 17 00:00:00 2001 From: Thomas Weißschuh Date: Fri, 22 Mar 2024 18:05:57 +0100 Subject: sysctl: move sysctl type to ctl_table_header MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move the SYSCTL_TABLE_TYPE_{DEFAULT,PERMANENTLY_EMPTY} enums from ctl_table to ctl_table_header. Removing the mutable member is necessary to constify static instances of struct ctl_table. Move the initialization of the sysctl_mount_point type into init_header() where all the other header fields are also initialized. As a side-effect the memory usage of the sysctl core is reduced. Each ctl_table_header instance can manage multiple ctl_table instances and is only allocated when the table is actually registered. This saves 8 bytes of memory per ctl_table on 64bit, 4 due to the enum field itself and 4 due to padding. Signed-off-by: Thomas Weißschuh Signed-off-by: Joel Granados --- fs/proc/proc_sysctl.c | 10 ++++++---- include/linux/sysctl.h | 22 +++++++++++----------- 2 files changed, 17 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 3f8abb6d2ad5..a847b0bc63c4 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -31,7 +31,7 @@ static const struct inode_operations proc_sys_dir_operations; /* Support for permanently empty directories */ static struct ctl_table sysctl_mount_point[] = { - {.type = SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY } + { } }; /** @@ -49,11 +49,11 @@ struct ctl_table_header *register_sysctl_mount_point(const char *path) EXPORT_SYMBOL(register_sysctl_mount_point); #define sysctl_is_perm_empty_ctl_header(hptr) \ - (hptr->ctl_table[0].type == SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY) + (hptr->type == SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY) #define sysctl_set_perm_empty_ctl_header(hptr) \ - (hptr->ctl_table[0].type = SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY) + (hptr->type = SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY) #define sysctl_clear_perm_empty_ctl_header(hptr) \ - (hptr->ctl_table[0].type = SYSCTL_TABLE_TYPE_DEFAULT) + (hptr->type = SYSCTL_TABLE_TYPE_DEFAULT) void proc_sys_poll_notify(struct ctl_table_poll *poll) { @@ -208,6 +208,8 @@ static void init_header(struct ctl_table_header *head, node++; } } + if (table == sysctl_mount_point) + sysctl_set_perm_empty_ctl_header(head); } static void erase_header(struct ctl_table_header *head) diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index f9214de0490c..47bd28ffa88f 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -137,17 +137,6 @@ struct ctl_table { void *data; int maxlen; umode_t mode; - /** - * enum type - Enumeration to differentiate between ctl target types - * @SYSCTL_TABLE_TYPE_DEFAULT: ctl target with no special considerations - * @SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY: Used to identify a permanently - * empty directory target to serve - * as mount point. - */ - enum { - SYSCTL_TABLE_TYPE_DEFAULT, - SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY - } type; proc_handler *proc_handler; /* Callback for text formatting */ struct ctl_table_poll *poll; void *extra1; @@ -188,6 +177,17 @@ struct ctl_table_header { struct ctl_dir *parent; struct ctl_node *node; struct hlist_head inodes; /* head for proc_inode->sysctl_inodes */ + /** + * enum type - Enumeration to differentiate between ctl target types + * @SYSCTL_TABLE_TYPE_DEFAULT: ctl target with no special considerations + * @SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY: Used to identify a permanently + * empty directory target to serve + * as mount point. + */ + enum { + SYSCTL_TABLE_TYPE_DEFAULT, + SYSCTL_TABLE_TYPE_PERMANENTLY_EMPTY, + } type; }; struct ctl_dir { -- cgit v1.2.3-59-g8ed1b From a35dd3a786f57903151b18275b1eed105084cf72 Mon Sep 17 00:00:00 2001 From: Thomas Weißschuh Date: Fri, 22 Mar 2024 18:05:58 +0100 Subject: sysctl: drop now unnecessary out-of-bounds check MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove the now unneeded check for ctl_table_size; it is safe to do so as sysctl_set_perm_empty_ctl_header() does not access the ctl_table member anymore. This also makes the element of sysctl_mount_point unnecessary, so drop it at the same time. Signed-off-by: Thomas Weißschuh Signed-off-by: Joel Granados --- fs/proc/proc_sysctl.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index a847b0bc63c4..b1c2c0b82116 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -30,9 +30,7 @@ static const struct file_operations proc_sys_dir_file_operations; static const struct inode_operations proc_sys_dir_operations; /* Support for permanently empty directories */ -static struct ctl_table sysctl_mount_point[] = { - { } -}; +static struct ctl_table sysctl_mount_point[] = { }; /** * register_sysctl_mount_point() - registers a sysctl mount point @@ -232,8 +230,7 @@ static int insert_header(struct ctl_dir *dir, struct ctl_table_header *header) return -EROFS; /* Am I creating a permanently empty directory? */ - if (header->ctl_table_size > 0 && - sysctl_is_perm_empty_ctl_header(header)) { + if (sysctl_is_perm_empty_ctl_header(header)) { if (!RB_EMPTY_ROOT(&dir->root)) return -EINVAL; sysctl_set_perm_empty_ctl_header(dir_h); -- cgit v1.2.3-59-g8ed1b From 19e048641bc6e29a4c3ba1427481f86305f3b960 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 23 Apr 2024 14:46:06 +0200 Subject: xfs: fix overly long line in the file_operations Re-wrap the newly added fop_flags fields to not go over 80 characters. Signed-off-by: Christoph Hellwig Link: https://lore.kernel.org/r/20240423124608.537794-2-hch@lst.de Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/xfs/xfs_file.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 147439ad3581..c6b6809c1ffe 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1496,8 +1496,8 @@ const struct file_operations xfs_file_operations = { .fallocate = xfs_file_fallocate, .fadvise = xfs_file_fadvise, .remap_file_range = xfs_file_remap_range, - .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | FOP_BUFFER_WASYNC | - FOP_DIO_PARALLEL_WRITE, + .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | + FOP_BUFFER_WASYNC | FOP_DIO_PARALLEL_WRITE, }; const struct file_operations xfs_dir_file_operations = { @@ -1510,6 +1510,6 @@ const struct file_operations xfs_dir_file_operations = { .compat_ioctl = xfs_file_compat_ioctl, #endif .fsync = xfs_dir_fsync, - .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | FOP_BUFFER_WASYNC | - FOP_DIO_PARALLEL_WRITE, + .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | + FOP_BUFFER_WASYNC | FOP_DIO_PARALLEL_WRITE, }; -- cgit v1.2.3-59-g8ed1b From f50805713a6e8bd58bb69386c60f5c922b882016 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 23 Apr 2024 14:46:07 +0200 Subject: xfs: drop fop_flags for directories Directories have non of the capabilities, so drop the flags. Note that the current state is harmless as no one actually checks for the flags either. Signed-off-by: Christoph Hellwig Link: https://lore.kernel.org/r/20240423124608.537794-3-hch@lst.de Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/xfs/xfs_file.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index c6b6809c1ffe..4415d0f3bbc5 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1510,6 +1510,4 @@ const struct file_operations xfs_dir_file_operations = { .compat_ioctl = xfs_file_compat_ioctl, #endif .fsync = xfs_dir_fsync, - .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | - FOP_BUFFER_WASYNC | FOP_DIO_PARALLEL_WRITE, }; -- cgit v1.2.3-59-g8ed1b From 652efdeca5b142ee9c5197f45f64fc3d427d4b08 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 23 Apr 2024 14:46:08 +0200 Subject: xfs: don't call xfs_file_open from xfs_dir_open Directories do not support direct I/O and thus no non-blocking direct I/O either. Open code the shutdown check and call to generic_file_open instead. Signed-off-by: Christoph Hellwig Link: https://lore.kernel.org/r/20240423124608.537794-4-hch@lst.de Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/xfs/xfs_file.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 4415d0f3bbc5..2ce302b4885f 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1243,7 +1243,9 @@ xfs_dir_open( unsigned int mode; int error; - error = xfs_file_open(inode, file); + if (xfs_is_shutdown(ip->i_mount)) + return -EIO; + error = generic_file_open(inode, file); if (error) return error; -- cgit v1.2.3-59-g8ed1b From 7a1ad9d8120e3d510ee5de8924f14de089aa2e2d Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Tue, 9 Apr 2024 13:54:26 +0200 Subject: gfs2: Fix lru_count accounting Currently, gfs2_scan_glock_lru() decrements lru_count when a glock is moved onto the dispose list. When such a glock is then stolen from the dispose list while gfs2_dispose_glock_lru() doesn't hold the lru_lock, lru_count will be decremented again, so the counter will eventually go negative. This bug has existed in one form or another since at least commit 97cc1025b1a91 ("GFS2: Kill two daemons with one patch"). Fix this by only decrementing lru_count when we actually remove a glock and schedule for it to be unlocked and dropped. We also don't need to remove and then re-add glocks when we can just as well move them back onto the lru_list when necessary. In addition, return the number of glocks freed as we should, not the number of glocks moved onto the dispose list. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 27 +++++++++++++-------------- 1 file changed, 13 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 04e0a8ac61d7..8e6936b79460 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -2009,29 +2009,30 @@ static bool can_free_glock(struct gfs2_glock *gl) * private) */ -static void gfs2_dispose_glock_lru(struct list_head *list) +static unsigned long gfs2_dispose_glock_lru(struct list_head *list) __releases(&lru_lock) __acquires(&lru_lock) { struct gfs2_glock *gl; + unsigned long freed = 0; list_sort(NULL, list, glock_cmp); while(!list_empty(list)) { gl = list_first_entry(list, struct gfs2_glock, gl_lru); - list_del_init(&gl->gl_lru); - clear_bit(GLF_LRU, &gl->gl_flags); if (!spin_trylock(&gl->gl_lockref.lock)) { add_back_to_lru: - list_add(&gl->gl_lru, &lru_list); - set_bit(GLF_LRU, &gl->gl_flags); - atomic_inc(&lru_count); + list_move(&gl->gl_lru, &lru_list); continue; } if (!can_free_glock(gl)) { spin_unlock(&gl->gl_lockref.lock); goto add_back_to_lru; } + list_del_init(&gl->gl_lru); + atomic_dec(&lru_count); + clear_bit(GLF_LRU, &gl->gl_flags); + freed++; gl->gl_lockref.count++; if (demote_ok(gl)) handle_callback(gl, LM_ST_UNLOCKED, 0, false); @@ -2039,6 +2040,7 @@ add_back_to_lru: spin_unlock(&gl->gl_lockref.lock); cond_resched_lock(&lru_lock); } + return freed; } /** @@ -2050,24 +2052,21 @@ add_back_to_lru: * gfs2_dispose_glock_lru() above. */ -static long gfs2_scan_glock_lru(int nr) +static unsigned long gfs2_scan_glock_lru(unsigned long nr) { struct gfs2_glock *gl, *next; LIST_HEAD(dispose); - long freed = 0; + unsigned long freed = 0; spin_lock(&lru_lock); list_for_each_entry_safe(gl, next, &lru_list, gl_lru) { - if (nr-- <= 0) + if (!nr--) break; - if (can_free_glock(gl)) { + if (can_free_glock(gl)) list_move(&gl->gl_lru, &dispose); - atomic_dec(&lru_count); - freed++; - } } if (!list_empty(&dispose)) - gfs2_dispose_glock_lru(&dispose); + freed = gfs2_dispose_glock_lru(&dispose); spin_unlock(&lru_lock); return freed; -- cgit v1.2.3-59-g8ed1b From 59f60005797b4018d7b46620037e0c53d690795e Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 10 Apr 2024 04:24:56 +0200 Subject: gfs2: Remove ill-placed consistency check This consistency check was originally added by commit 9287c6452d2b1 ("gfs2: Fix occasional glock use-after-free"). It is ill-placed in gfs2_glock_free() because if it holds there, it must equally hold in __gfs2_glock_put() already. Either way, the check doesn't seem necessary anymore. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 8e6936b79460..bf7538274985 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -170,7 +170,6 @@ void gfs2_glock_free(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; - gfs2_glock_assert_withdraw(gl, atomic_read(&gl->gl_revokes) == 0); rhashtable_remove_fast(&gl_hash_table, &gl->gl_node, ht_parms); smp_mb(); wake_up_glock(gl); -- cgit v1.2.3-59-g8ed1b From d98779e687726d8f8860f1c54b5687eec5f63a73 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 10 Apr 2024 04:50:18 +0200 Subject: gfs2: Fix potential glock use-after-free on unmount When a DLM lockspace is released and there ares still locks in that lockspace, DLM will unlock those locks automatically. Commit fb6791d100d1b started exploiting this behavior to speed up filesystem unmount: gfs2 would simply free glocks it didn't want to unlock and then release the lockspace. This didn't take the bast callbacks for asynchronous lock contention notifications into account, which remain active until until a lock is unlocked or its lockspace is released. To prevent those callbacks from accessing deallocated objects, put the glocks that should not be unlocked on the sd_dead_glocks list, release the lockspace, and only then free those glocks. As an additional measure, ignore unexpected ast and bast callbacks if the receiving glock is dead. Fixes: fb6791d100d1b ("GFS2: skip dlm_unlock calls in unmount") Signed-off-by: Andreas Gruenbacher Cc: David Teigland --- fs/gfs2/glock.c | 35 ++++++++++++++++++++++++++++++++--- fs/gfs2/glock.h | 1 + fs/gfs2/incore.h | 1 + fs/gfs2/lock_dlm.c | 32 ++++++++++++++++++++++---------- fs/gfs2/ops_fstype.c | 1 + fs/gfs2/super.c | 3 --- 6 files changed, 57 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index bf7538274985..4cf8971ce8ee 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -166,18 +166,45 @@ static bool glock_blocked_by_withdraw(struct gfs2_glock *gl) return true; } -void gfs2_glock_free(struct gfs2_glock *gl) +static void __gfs2_glock_free(struct gfs2_glock *gl) { - struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; - rhashtable_remove_fast(&gl_hash_table, &gl->gl_node, ht_parms); smp_mb(); wake_up_glock(gl); call_rcu(&gl->gl_rcu, gfs2_glock_dealloc); +} + +void gfs2_glock_free(struct gfs2_glock *gl) { + struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; + + __gfs2_glock_free(gl); if (atomic_dec_and_test(&sdp->sd_glock_disposal)) wake_up(&sdp->sd_kill_wait); } +void gfs2_glock_free_later(struct gfs2_glock *gl) { + struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; + + spin_lock(&lru_lock); + list_add(&gl->gl_lru, &sdp->sd_dead_glocks); + spin_unlock(&lru_lock); + if (atomic_dec_and_test(&sdp->sd_glock_disposal)) + wake_up(&sdp->sd_kill_wait); +} + +static void gfs2_free_dead_glocks(struct gfs2_sbd *sdp) +{ + struct list_head *list = &sdp->sd_dead_glocks; + + while(!list_empty(list)) { + struct gfs2_glock *gl; + + gl = list_first_entry(list, struct gfs2_glock, gl_lru); + list_del_init(&gl->gl_lru); + __gfs2_glock_free(gl); + } +} + /** * gfs2_glock_hold() - increment reference count on glock * @gl: The glock to hold @@ -2233,6 +2260,8 @@ void gfs2_gl_hash_clear(struct gfs2_sbd *sdp) wait_event_timeout(sdp->sd_kill_wait, atomic_read(&sdp->sd_glock_disposal) == 0, HZ * 600); + gfs2_lm_unmount(sdp); + gfs2_free_dead_glocks(sdp); glock_hash_walk(dump_glock_func, sdp); } diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index 2c697674a86f..19aef6d53267 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -252,6 +252,7 @@ void gfs2_gl_dq_holders(struct gfs2_sbd *sdp); void gfs2_glock_thaw(struct gfs2_sbd *sdp); void gfs2_glock_add_to_lru(struct gfs2_glock *gl); void gfs2_glock_free(struct gfs2_glock *gl); +void gfs2_glock_free_later(struct gfs2_glock *gl); int __init gfs2_glock_init(void); void gfs2_glock_exit(void); diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 95a334d64da2..60abd7050c99 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -838,6 +838,7 @@ struct gfs2_sbd { /* For quiescing the filesystem */ struct gfs2_holder sd_freeze_gh; struct mutex sd_freeze_mutex; + struct list_head sd_dead_glocks; char sd_fsname[GFS2_FSNAME_LEN + 3 * sizeof(int) + 2]; char sd_table_name[GFS2_FSNAME_LEN]; diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c index d1ac5d0679ea..e028e55e67d9 100644 --- a/fs/gfs2/lock_dlm.c +++ b/fs/gfs2/lock_dlm.c @@ -121,6 +121,11 @@ static void gdlm_ast(void *arg) struct gfs2_glock *gl = arg; unsigned ret = gl->gl_state; + /* If the glock is dead, we only react to a dlm_unlock() reply. */ + if (__lockref_is_dead(&gl->gl_lockref) && + gl->gl_lksb.sb_status != -DLM_EUNLOCK) + return; + gfs2_update_reply_times(gl); BUG_ON(gl->gl_lksb.sb_flags & DLM_SBF_DEMOTED); @@ -171,6 +176,9 @@ static void gdlm_bast(void *arg, int mode) { struct gfs2_glock *gl = arg; + if (__lockref_is_dead(&gl->gl_lockref)) + return; + switch (mode) { case DLM_LOCK_EX: gfs2_glock_cb(gl, LM_ST_UNLOCKED); @@ -291,8 +299,12 @@ static void gdlm_put_lock(struct gfs2_glock *gl) struct lm_lockstruct *ls = &sdp->sd_lockstruct; int error; - if (gl->gl_lksb.sb_lkid == 0) - goto out_free; + BUG_ON(!__lockref_is_dead(&gl->gl_lockref)); + + if (gl->gl_lksb.sb_lkid == 0) { + gfs2_glock_free(gl); + return; + } clear_bit(GLF_BLOCKING, &gl->gl_flags); gfs2_glstats_inc(gl, GFS2_LKS_DCOUNT); @@ -300,13 +312,17 @@ static void gdlm_put_lock(struct gfs2_glock *gl) gfs2_update_request_times(gl); /* don't want to call dlm if we've unmounted the lock protocol */ - if (test_bit(DFL_UNMOUNT, &ls->ls_recover_flags)) - goto out_free; + if (test_bit(DFL_UNMOUNT, &ls->ls_recover_flags)) { + gfs2_glock_free(gl); + return; + } /* don't want to skip dlm_unlock writing the lvb when lock has one */ if (test_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags) && - !gl->gl_lksb.sb_lvbptr) - goto out_free; + !gl->gl_lksb.sb_lvbptr) { + gfs2_glock_free_later(gl); + return; + } again: error = dlm_unlock(ls->ls_dlm, gl->gl_lksb.sb_lkid, DLM_LKF_VALBLK, @@ -321,10 +337,6 @@ again: gl->gl_name.ln_type, (unsigned long long)gl->gl_name.ln_number, error); } - return; - -out_free: - gfs2_glock_free(gl); } static void gdlm_cancel(struct gfs2_glock *gl) diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 1281e60be639..db0df091a6a7 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -136,6 +136,7 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) atomic_set(&sdp->sd_log_in_flight, 0); init_waitqueue_head(&sdp->sd_log_flush_wait); mutex_init(&sdp->sd_freeze_mutex); + INIT_LIST_HEAD(&sdp->sd_dead_glocks); return sdp; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index d481db9510ac..32a37daaccbe 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -650,10 +650,7 @@ restart: gfs2_gl_hash_clear(sdp); truncate_inode_pages_final(&sdp->sd_aspace); gfs2_delete_debugfs_file(sdp); - /* Unmount the locking protocol */ - gfs2_lm_unmount(sdp); - /* At this point, we're through participating in the lockspace */ gfs2_sys_fs_del(sdp); free_sbd(sdp); } -- cgit v1.2.3-59-g8ed1b From a3730c5ec57b033ba6e437f7881a894d57b28a4a Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 10 Apr 2024 05:23:19 +0200 Subject: gfs2: Unlock fewer glocks on unmount At unmount time, we would generally like to explicitly unlock as few glocks as possible for efficiency. We are already skipping glocks that don't have a lock value block (LVB), but we can also skip glocks which are not held in DLM_LOCK_EX or DLM_LOCK_PW mode (of which gfs2 only uses DLM_LOCK_EX under the name LM_ST_EXCLUSIVE). Signed-off-by: Andreas Gruenbacher Cc: David Teigland --- fs/gfs2/lock_dlm.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c index e028e55e67d9..49059274a528 100644 --- a/fs/gfs2/lock_dlm.c +++ b/fs/gfs2/lock_dlm.c @@ -316,10 +316,16 @@ static void gdlm_put_lock(struct gfs2_glock *gl) gfs2_glock_free(gl); return; } - /* don't want to skip dlm_unlock writing the lvb when lock has one */ + + /* + * When the lockspace is released, all remaining glocks will be + * unlocked automatically. This is more efficient than unlocking them + * individually, but when the lock is held in DLM_LOCK_EX or + * DLM_LOCK_PW mode, the lock value block (LVB) will be lost. + */ if (test_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags) && - !gl->gl_lksb.sb_lvbptr) { + (!gl->gl_lksb.sb_lvbptr || gl->gl_state != LM_ST_EXCLUSIVE)) { gfs2_glock_free_later(gl); return; } -- cgit v1.2.3-59-g8ed1b From 1cd28e15864054f3c48baee9eecda1c0441c48ac Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 12 Apr 2024 19:16:58 +0200 Subject: gfs2: finish_xmote cleanup Currently, function finish_xmote() takes and releases the glock spinlock. However, all of its callers immediately take that spinlock again, so it makes more sense to take the spin lock before calling finish_xmote() already. With that, thaw_glock() is the only place that sets the GLF_HAVE_REPLY flag outside of the glock spinlock, but it also takes that spinlock immediately thereafter. Change that to set the bit when the spinlock is already held. This allows to switch from test_and_clear_bit() to test_bit() and clear_bit() in glock_work_func(). Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 4cf8971ce8ee..b1a2862d431d 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -625,7 +625,6 @@ static void finish_xmote(struct gfs2_glock *gl, unsigned int ret) struct gfs2_holder *gh; unsigned state = ret & LM_OUT_ST_MASK; - spin_lock(&gl->gl_lockref.lock); trace_gfs2_glock_state_change(gl, state); state_change(gl, state); gh = find_first_waiter(gl); @@ -673,7 +672,6 @@ retry: gl->gl_target, state); GLOCK_BUG_ON(gl, 1); } - spin_unlock(&gl->gl_lockref.lock); return; } @@ -696,7 +694,6 @@ retry: } out: clear_bit(GLF_LOCK, &gl->gl_flags); - spin_unlock(&gl->gl_lockref.lock); } static bool is_system_glock(struct gfs2_glock *gl) @@ -843,15 +840,19 @@ skip_inval: if (ret == -EINVAL && gl->gl_target == LM_ST_UNLOCKED && target == LM_ST_UNLOCKED && test_bit(DFL_UNMOUNT, &ls->ls_recover_flags)) { + spin_lock(&gl->gl_lockref.lock); finish_xmote(gl, target); - gfs2_glock_queue_work(gl, 0); + __gfs2_glock_queue_work(gl, 0); + spin_unlock(&gl->gl_lockref.lock); } else if (ret) { fs_err(sdp, "lm_lock ret %d\n", ret); GLOCK_BUG_ON(gl, !gfs2_withdrawing_or_withdrawn(sdp)); } } else { /* lock_nolock */ + spin_lock(&gl->gl_lockref.lock); finish_xmote(gl, target); - gfs2_glock_queue_work(gl, 0); + __gfs2_glock_queue_work(gl, 0); + spin_unlock(&gl->gl_lockref.lock); } out: spin_lock(&gl->gl_lockref.lock); @@ -1108,11 +1109,12 @@ static void glock_work_func(struct work_struct *work) struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_work.work); unsigned int drop_refs = 1; - if (test_and_clear_bit(GLF_REPLY_PENDING, &gl->gl_flags)) { + spin_lock(&gl->gl_lockref.lock); + if (test_bit(GLF_REPLY_PENDING, &gl->gl_flags)) { + clear_bit(GLF_REPLY_PENDING, &gl->gl_flags); finish_xmote(gl, gl->gl_reply); drop_refs++; } - spin_lock(&gl->gl_lockref.lock); if (test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) && gl->gl_state != LM_ST_UNLOCKED && gl->gl_demote_state != LM_ST_EXCLUSIVE) { @@ -2183,8 +2185,11 @@ static void thaw_glock(struct gfs2_glock *gl) return; if (!lockref_get_not_dead(&gl->gl_lockref)) return; + + spin_lock(&gl->gl_lockref.lock); set_bit(GLF_REPLY_PENDING, &gl->gl_flags); - gfs2_glock_queue_work(gl, 0); + __gfs2_glock_queue_work(gl, 0); + spin_unlock(&gl->gl_lockref.lock); } /** -- cgit v1.2.3-59-g8ed1b From 9947a06d29c0a30da88cdc6376ca5fd87083e130 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Mon, 15 Apr 2024 11:23:04 +0200 Subject: gfs2: do_xmote fixes Function do_xmote() is called with the glock spinlock held. Commit 86934198eefa added a 'goto skip_inval' statement at the beginning of the function to further below where the glock spinlock is expected not to be held anymore. Then it added code there that requires the glock spinlock to be held. This doesn't make sense; fix this up by dropping and retaking the spinlock where needed. In addition, when ->lm_lock() returned an error, do_xmote() didn't fail the locking operation, and simply left the glock hanging; fix that as well. (This is a much older error.) Fixes: 86934198eefa ("gfs2: Clear flags when withdraw prevents xmote") Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 44 +++++++++++++++++++++++++------------------- 1 file changed, 25 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index b1a2862d431d..fa374617d2f1 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -721,6 +721,7 @@ __acquires(&gl->gl_lockref.lock) { const struct gfs2_glock_operations *glops = gl->gl_ops; struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; + struct lm_lockstruct *ls = &sdp->sd_lockstruct; unsigned int lck_flags = (unsigned int)(gh ? gh->gh_flags : 0); int ret; @@ -749,6 +750,9 @@ __acquires(&gl->gl_lockref.lock) (gl->gl_state == LM_ST_EXCLUSIVE) || (lck_flags & (LM_FLAG_TRY|LM_FLAG_TRY_1CB))) clear_bit(GLF_BLOCKING, &gl->gl_flags); + if (!glops->go_inval && !glops->go_sync) + goto skip_inval; + spin_unlock(&gl->gl_lockref.lock); if (glops->go_sync) { ret = glops->go_sync(gl); @@ -761,6 +765,7 @@ __acquires(&gl->gl_lockref.lock) fs_err(sdp, "Error %d syncing glock \n", ret); gfs2_dump_glock(NULL, gl, true); } + spin_lock(&gl->gl_lockref.lock); goto skip_inval; } } @@ -781,9 +786,10 @@ __acquires(&gl->gl_lockref.lock) glops->go_inval(gl, target == LM_ST_DEFERRED ? 0 : DIO_METADATA); clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); } + spin_lock(&gl->gl_lockref.lock); skip_inval: - gfs2_glock_hold(gl); + gl->gl_lockref.count++; /* * Check for an error encountered since we called go_sync and go_inval. * If so, we can't withdraw from the glock code because the withdraw @@ -825,37 +831,37 @@ skip_inval: */ clear_bit(GLF_LOCK, &gl->gl_flags); clear_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags); - gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD); - goto out; + __gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD); + return; } else { clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); } } - if (sdp->sd_lockstruct.ls_ops->lm_lock) { - struct lm_lockstruct *ls = &sdp->sd_lockstruct; + if (ls->ls_ops->lm_lock) { + spin_unlock(&gl->gl_lockref.lock); + ret = ls->ls_ops->lm_lock(gl, target, lck_flags); + spin_lock(&gl->gl_lockref.lock); - /* lock_dlm */ - ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags); if (ret == -EINVAL && gl->gl_target == LM_ST_UNLOCKED && target == LM_ST_UNLOCKED && test_bit(DFL_UNMOUNT, &ls->ls_recover_flags)) { - spin_lock(&gl->gl_lockref.lock); - finish_xmote(gl, target); - __gfs2_glock_queue_work(gl, 0); - spin_unlock(&gl->gl_lockref.lock); + /* + * The lockspace has been released and the lock has + * been unlocked implicitly. + */ } else if (ret) { fs_err(sdp, "lm_lock ret %d\n", ret); - GLOCK_BUG_ON(gl, !gfs2_withdrawing_or_withdrawn(sdp)); + target = gl->gl_state | LM_OUT_ERROR; + } else { + /* The operation will be completed asynchronously. */ + return; } - } else { /* lock_nolock */ - spin_lock(&gl->gl_lockref.lock); - finish_xmote(gl, target); - __gfs2_glock_queue_work(gl, 0); - spin_unlock(&gl->gl_lockref.lock); } -out: - spin_lock(&gl->gl_lockref.lock); + + /* Complete the operation now. */ + finish_xmote(gl, target); + __gfs2_glock_queue_work(gl, 0); } /** -- cgit v1.2.3-59-g8ed1b From 1e86044402c45b70a9b31beeaefb5cc732a7470c Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 12 Apr 2024 21:58:15 +0200 Subject: gfs2: Remove and replace gfs2_glock_queue_work There are no more callers of gfs2_glock_queue_work() left, so remove that helper. With that, we can now rename __gfs2_glock_queue_work() back to gfs2_glock_queue_work() to get rid of some unnecessary clutter. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/glock.c | 35 +++++++++++++++-------------------- 1 file changed, 15 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index fa374617d2f1..9f11fc1e79eb 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -274,7 +274,7 @@ static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl) * Enqueue the glock on the work queue. Passes one glock reference on to the * work queue. */ -static void __gfs2_glock_queue_work(struct gfs2_glock *gl, unsigned long delay) { +static void gfs2_glock_queue_work(struct gfs2_glock *gl, unsigned long delay) { if (!queue_delayed_work(glock_workqueue, &gl->gl_work, delay)) { /* * We are holding the lockref spinlock, and the work was still @@ -287,12 +287,6 @@ static void __gfs2_glock_queue_work(struct gfs2_glock *gl, unsigned long delay) } } -static void gfs2_glock_queue_work(struct gfs2_glock *gl, unsigned long delay) { - spin_lock(&gl->gl_lockref.lock); - __gfs2_glock_queue_work(gl, delay); - spin_unlock(&gl->gl_lockref.lock); -} - static void __gfs2_glock_put(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; @@ -337,7 +331,8 @@ void gfs2_glock_put_async(struct gfs2_glock *gl) if (lockref_put_or_lock(&gl->gl_lockref)) return; - __gfs2_glock_queue_work(gl, 0); + GLOCK_BUG_ON(gl, gl->gl_lockref.count != 1); + gfs2_glock_queue_work(gl, 0); spin_unlock(&gl->gl_lockref.lock); } @@ -831,7 +826,7 @@ skip_inval: */ clear_bit(GLF_LOCK, &gl->gl_flags); clear_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags); - __gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD); + gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD); return; } else { clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); @@ -861,7 +856,7 @@ skip_inval: /* Complete the operation now. */ finish_xmote(gl, target); - __gfs2_glock_queue_work(gl, 0); + gfs2_glock_queue_work(gl, 0); } /** @@ -909,7 +904,7 @@ out_sched: clear_bit(GLF_LOCK, &gl->gl_flags); smp_mb__after_atomic(); gl->gl_lockref.count++; - __gfs2_glock_queue_work(gl, 0); + gfs2_glock_queue_work(gl, 0); return; out_unlock: @@ -1141,12 +1136,12 @@ static void glock_work_func(struct work_struct *work) drop_refs--; if (gl->gl_name.ln_type != LM_TYPE_INODE) delay = 0; - __gfs2_glock_queue_work(gl, delay); + gfs2_glock_queue_work(gl, delay); } /* * Drop the remaining glock references manually here. (Mind that - * __gfs2_glock_queue_work depends on the lockref spinlock begin held + * gfs2_glock_queue_work depends on the lockref spinlock begin held * here as well.) */ gl->gl_lockref.count -= drop_refs; @@ -1651,7 +1646,7 @@ unlock: test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))) { set_bit(GLF_REPLY_PENDING, &gl->gl_flags); gl->gl_lockref.count++; - __gfs2_glock_queue_work(gl, 0); + gfs2_glock_queue_work(gl, 0); } run_queue(gl, 1); spin_unlock(&gl->gl_lockref.lock); @@ -1717,7 +1712,7 @@ static void __gfs2_glock_dq(struct gfs2_holder *gh) !test_bit(GLF_DEMOTE, &gl->gl_flags) && gl->gl_name.ln_type == LM_TYPE_INODE) delay = gl->gl_hold_time; - __gfs2_glock_queue_work(gl, delay); + gfs2_glock_queue_work(gl, delay); } } @@ -1941,7 +1936,7 @@ void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state) delay = gl->gl_hold_time; } handle_callback(gl, state, delay, true); - __gfs2_glock_queue_work(gl, delay); + gfs2_glock_queue_work(gl, delay); spin_unlock(&gl->gl_lockref.lock); } @@ -2001,7 +1996,7 @@ void gfs2_glock_complete(struct gfs2_glock *gl, int ret) gl->gl_lockref.count++; set_bit(GLF_REPLY_PENDING, &gl->gl_flags); - __gfs2_glock_queue_work(gl, 0); + gfs2_glock_queue_work(gl, 0); spin_unlock(&gl->gl_lockref.lock); } @@ -2070,7 +2065,7 @@ add_back_to_lru: gl->gl_lockref.count++; if (demote_ok(gl)) handle_callback(gl, LM_ST_UNLOCKED, 0, false); - __gfs2_glock_queue_work(gl, 0); + gfs2_glock_queue_work(gl, 0); spin_unlock(&gl->gl_lockref.lock); cond_resched_lock(&lru_lock); } @@ -2194,7 +2189,7 @@ static void thaw_glock(struct gfs2_glock *gl) spin_lock(&gl->gl_lockref.lock); set_bit(GLF_REPLY_PENDING, &gl->gl_flags); - __gfs2_glock_queue_work(gl, 0); + gfs2_glock_queue_work(gl, 0); spin_unlock(&gl->gl_lockref.lock); } @@ -2213,7 +2208,7 @@ static void clear_glock(struct gfs2_glock *gl) gl->gl_lockref.count++; if (gl->gl_state != LM_ST_UNLOCKED) handle_callback(gl, LM_ST_UNLOCKED, 0, false); - __gfs2_glock_queue_work(gl, 0); + gfs2_glock_queue_work(gl, 0); } spin_unlock(&gl->gl_lockref.lock); } -- cgit v1.2.3-59-g8ed1b From 2a5eb9995528441447d33838727f6ec1caf08139 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 16 Feb 2024 22:25:44 -0800 Subject: binfmt_elf: Leave a gap between .bss and brk Currently the brk starts its randomization immediately after .bss, which means there is a chance that when the random offset is 0, linear overflows from .bss can reach into the brk area. Leave at least a single page gap between .bss and brk (when it has not already been explicitly relocated into the mmap range). Reported-by: Closes: https://lore.kernel.org/linux-hardening/CA+2EKTVLvc8hDZc+2Yhwmus=dzOUG5E4gV7ayCbu0MPJTZzWkw@mail.gmail.com/ Link: https://lore.kernel.org/r/20240217062545.1631668-2-keescook@chromium.org Signed-off-by: Kees Cook --- fs/binfmt_elf.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 5397b552fbeb..7862962f7a85 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1262,6 +1262,9 @@ out_free_interp: if (IS_ENABLED(CONFIG_ARCH_HAS_ELF_RANDOMIZE) && elf_ex->e_type == ET_DYN && !interpreter) { mm->brk = mm->start_brk = ELF_ET_DYN_BASE; + } else { + /* Otherwise leave a gap between .bss and brk. */ + mm->brk = mm->start_brk = mm->brk + PAGE_SIZE; } mm->brk = mm->start_brk = arch_randomize_brk(mm); -- cgit v1.2.3-59-g8ed1b From 10e29251be0e9f774910c1baaa89355859491769 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 22 Mar 2024 12:54:18 -0700 Subject: binfmt_elf_fdpic: fix /proc//auxv Althought FDPIC linux kernel provides /proc//auxv files they are empty because there's no code that initializes mm->saved_auxv in the FDPIC ELF loader. Synchronize FDPIC ELF aux vector setup with ELF. Replace entry-by-entry aux vector copying to userspace with initialization of mm->saved_auxv first and then copying it to userspace as a whole. Signed-off-by: Max Filippov Link: https://lore.kernel.org/r/20240322195418.2160164-1-jcmvbkbc@gmail.com Signed-off-by: Kees Cook --- fs/binfmt_elf_fdpic.c | 85 +++++++++++++++++++++++---------------------------- 1 file changed, 39 insertions(+), 46 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 3314249e8674..b799701454a9 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -505,8 +505,9 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, char *k_platform, *k_base_platform; char __user *u_platform, *u_base_platform, *p; int loop; - int nr; /* reset for each csp adjustment */ unsigned long flags = 0; + int ei_index; + elf_addr_t *elf_info; #ifdef CONFIG_MMU /* In some cases (e.g. Hyper-Threading), we want to avoid L1 evictions @@ -601,44 +602,24 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, csp -= sp & 15UL; sp -= sp & 15UL; - /* put the ELF interpreter info on the stack */ -#define NEW_AUX_ENT(id, val) \ - do { \ - struct { unsigned long _id, _val; } __user *ent, v; \ - \ - ent = (void __user *) csp; \ - v._id = (id); \ - v._val = (val); \ - if (copy_to_user(ent + nr, &v, sizeof(v))) \ - return -EFAULT; \ - nr++; \ + /* Create the ELF interpreter info */ + elf_info = (elf_addr_t *)mm->saved_auxv; + /* update AT_VECTOR_SIZE_BASE if the number of NEW_AUX_ENT() changes */ +#define NEW_AUX_ENT(id, val) \ + do { \ + *elf_info++ = id; \ + *elf_info++ = val; \ } while (0) - nr = 0; - csp -= 2 * sizeof(unsigned long); - NEW_AUX_ENT(AT_NULL, 0); - if (k_platform) { - nr = 0; - csp -= 2 * sizeof(unsigned long); - NEW_AUX_ENT(AT_PLATFORM, - (elf_addr_t) (unsigned long) u_platform); - } - - if (k_base_platform) { - nr = 0; - csp -= 2 * sizeof(unsigned long); - NEW_AUX_ENT(AT_BASE_PLATFORM, - (elf_addr_t) (unsigned long) u_base_platform); - } - - if (bprm->have_execfd) { - nr = 0; - csp -= 2 * sizeof(unsigned long); - NEW_AUX_ENT(AT_EXECFD, bprm->execfd); - } - - nr = 0; - csp -= DLINFO_ITEMS * 2 * sizeof(unsigned long); +#ifdef ARCH_DLINFO + /* + * ARCH_DLINFO must come first so PPC can do its special alignment of + * AUXV. + * update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT() in + * ARCH_DLINFO changes + */ + ARCH_DLINFO; +#endif NEW_AUX_ENT(AT_HWCAP, ELF_HWCAP); #ifdef ELF_HWCAP2 NEW_AUX_ENT(AT_HWCAP2, ELF_HWCAP2); @@ -659,17 +640,29 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, NEW_AUX_ENT(AT_EGID, (elf_addr_t) from_kgid_munged(cred->user_ns, cred->egid)); NEW_AUX_ENT(AT_SECURE, bprm->secureexec); NEW_AUX_ENT(AT_EXECFN, bprm->exec); + if (k_platform) + NEW_AUX_ENT(AT_PLATFORM, + (elf_addr_t)(unsigned long)u_platform); + if (k_base_platform) + NEW_AUX_ENT(AT_BASE_PLATFORM, + (elf_addr_t)(unsigned long)u_base_platform); + if (bprm->have_execfd) + NEW_AUX_ENT(AT_EXECFD, bprm->execfd); +#undef NEW_AUX_ENT + /* AT_NULL is zero; clear the rest too */ + memset(elf_info, 0, (char *)mm->saved_auxv + + sizeof(mm->saved_auxv) - (char *)elf_info); -#ifdef ARCH_DLINFO - nr = 0; - csp -= AT_VECTOR_SIZE_ARCH * 2 * sizeof(unsigned long); + /* And advance past the AT_NULL entry. */ + elf_info += 2; - /* ARCH_DLINFO must come last so platform specific code can enforce - * special alignment requirements on the AUXV if necessary (eg. PPC). - */ - ARCH_DLINFO; -#endif -#undef NEW_AUX_ENT + ei_index = elf_info - (elf_addr_t *)mm->saved_auxv; + csp -= ei_index * sizeof(elf_addr_t); + + /* Put the elf_info on the stack in the right place. */ + if (copy_to_user((void __user *)csp, mm->saved_auxv, + ei_index * sizeof(elf_addr_t))) + return -EFAULT; /* allocate room for argv[] and envv[] */ csp -= (bprm->envc + 1) * sizeof(elf_caddr_t); -- cgit v1.2.3-59-g8ed1b From 31ca7e77fd7efc3f98582bf682fcaadc31a7d505 Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Thu, 28 Mar 2024 21:16:52 +0000 Subject: reiserfs: replace deprecated strncpy with scnprintf strncpy() is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. Our goal here is to get @namebuf populated with @name's contents but surrounded with quotes. There is some careful handling done to ensure we properly truncate @name so that we have room for a literal quote as well as a NUL-term. All this careful handling can be done with scnprintf using the dynamic string width specifier %.*s which allows us to pass in the max size for a source string. Doing this, we can put literal quotes in our format specifier and ensure @name is truncated to fit inbetween these quotes (-3 is from 2 quotes + 1 NUL-byte). All in all, we get to remove a deprecated use of strncpy and clean up this code nicely! Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2] Link: https://github.com/KSPP/linux/issues/90 Cc: Signed-off-by: Justin Stitt Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/20240328-strncpy-fs-reiserfs-item_ops-c-v1-1-2dab6d22a996@google.com Signed-off-by: Kees Cook --- fs/reiserfs/item_ops.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/item_ops.c b/fs/reiserfs/item_ops.c index 3a5a752d96c7..5011c10287c6 100644 --- a/fs/reiserfs/item_ops.c +++ b/fs/reiserfs/item_ops.c @@ -389,16 +389,9 @@ static void direntry_print_item(struct item_head *ih, char *item) name = item + deh_location(deh); if (name[namelen - 1] == 0) namelen = strlen(name); - namebuf[0] = '"'; - if (namelen > sizeof(namebuf) - 3) { - strncpy(namebuf + 1, name, sizeof(namebuf) - 3); - namebuf[sizeof(namebuf) - 2] = '"'; - namebuf[sizeof(namebuf) - 1] = 0; - } else { - memcpy(namebuf + 1, name, namelen); - namebuf[namelen + 1] = '"'; - namebuf[namelen + 2] = 0; - } + + scnprintf(namebuf, sizeof(namebuf), "\"%.*s\"", + (int)sizeof(namebuf)-3, name); printk("%d: %-15s%-15d%-15d%-15lld%-15lld(%s)\n", i, namebuf, -- cgit v1.2.3-59-g8ed1b From 7dcbf17e3f917f691d2ae9ed4a34283bd0e74a95 Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Mon, 1 Apr 2024 18:10:48 +0000 Subject: hfsplus: refactor copy_name to not use strncpy strncpy() is deprecated with NUL-terminated destination strings [1]. The copy_name() method does a lot of manual buffer manipulation to eventually arrive with its desired string. If we don't know the namespace this attr has or belongs to we want to prepend "osx." to our final string. Following this, we're copying xattr_name and doing a bizarre manual NUL-byte assignment with a memset where n=1. Really, we can use some more obvious string APIs to acomplish this, improving readability and security. Following the same control flow as before: if we don't know the namespace let's use scnprintf() to form our prefix + xattr_name pairing (while NUL-terminating too!). Otherwise, use strscpy() to return the number of bytes copied into our buffer. Additionally, for non-empty strings, include the NUL-byte in the length -- matching the behavior of the previous implementation. Note that strscpy() _can_ return -E2BIG but this is already handled by all callsites: In both hfsplus_listxattr_finder_info() and hfsplus_listxattr(), ret is already type ssize_t so we can change the return type of copy_name() to match (understanding that scnprintf()'s return type is different yet fully representable by ssize_t). Furthermore, listxattr() in fs/xattr.c is well-equipped to handle a potential -E2BIG return result from vfs_listxattr(): | ssize_t error; ... | error = vfs_listxattr(d, klist, size); | if (error > 0) { | if (size && copy_to_user(list, klist, error)) | error = -EFAULT; | } else if (error == -ERANGE && size >= XATTR_LIST_MAX) { | /* The file system tried to returned a list bigger | than XATTR_LIST_MAX bytes. Not possible. */ | error = -E2BIG; | } ... the error can potentially already be -E2BIG, skipping this else-if and ending up at the same state as other errors. Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html Link: https://github.com/KSPP/linux/issues/90 Cc: Signed-off-by: Justin Stitt Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/20240401-strncpy-fs-hfsplus-xattr-c-v2-1-6e089999355e@google.com Signed-off-by: Kees Cook --- fs/hfsplus/xattr.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/hfsplus/xattr.c b/fs/hfsplus/xattr.c index 9c9ff6b8c6f7..5a400259ae74 100644 --- a/fs/hfsplus/xattr.c +++ b/fs/hfsplus/xattr.c @@ -400,21 +400,19 @@ static int name_len(const char *xattr_name, int xattr_name_len) return len; } -static int copy_name(char *buffer, const char *xattr_name, int name_len) +static ssize_t copy_name(char *buffer, const char *xattr_name, int name_len) { - int len = name_len; - int offset = 0; + ssize_t len; - if (!is_known_namespace(xattr_name)) { - memcpy(buffer, XATTR_MAC_OSX_PREFIX, XATTR_MAC_OSX_PREFIX_LEN); - offset += XATTR_MAC_OSX_PREFIX_LEN; - len += XATTR_MAC_OSX_PREFIX_LEN; - } - - strncpy(buffer + offset, xattr_name, name_len); - memset(buffer + offset + name_len, 0, 1); - len += 1; + if (!is_known_namespace(xattr_name)) + len = scnprintf(buffer, name_len + XATTR_MAC_OSX_PREFIX_LEN, + "%s%s", XATTR_MAC_OSX_PREFIX, xattr_name); + else + len = strscpy(buffer, xattr_name, name_len + 1); + /* include NUL-byte in length for non-empty name */ + if (len >= 0) + len++; return len; } -- cgit v1.2.3-59-g8ed1b From f700b71927017ca5697e8a24fd2ad5d3e06489d6 Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Thu, 21 Mar 2024 00:38:54 +0000 Subject: fs: ecryptfs: replace deprecated strncpy with strscpy strncpy() is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. A good alternative is strscpy() as it guarantees NUL-termination on the destination buffer. In crypto.c: We expect cipher_name to be NUL-terminated based on its use with the C-string format specifier %s and with other string apis like strlen(): | printk(KERN_ERR "Error attempting to initialize key TFM " | "cipher with name = [%s]; rc = [%d]\n", | tmp_tfm->cipher_name, rc); and | int cipher_name_len = strlen(cipher_name); In main.c: We can remove the manual NUL-byte assignments as well as the pointers to destinations (which I assume only existed to trim down on line length?) in favor of directly using the destination buffer which allows the compiler to get size information -- enabling the usage of the new 2-argument strscpy(). Note that this patch relies on the _new_ 2-argument versions of strscpy() and strscpy_pad() introduced in Commit e6584c3964f2f ("string: Allow 2-argument strscpy()"). Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2] Link: https://github.com/KSPP/linux/issues/90 Cc: Signed-off-by: Justin Stitt Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/20240321-strncpy-fs-ecryptfs-crypto-c-v1-1-d78b74c214ac@google.com Signed-off-by: Kees Cook --- fs/ecryptfs/crypto.c | 4 +--- fs/ecryptfs/main.c | 26 ++++++-------------------- 2 files changed, 7 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 2fe0f3af1a08..d39a1a69fecc 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1606,9 +1606,7 @@ ecryptfs_add_new_key_tfm(struct ecryptfs_key_tfm **key_tfm, char *cipher_name, goto out; } mutex_init(&tmp_tfm->key_tfm_mutex); - strncpy(tmp_tfm->cipher_name, cipher_name, - ECRYPTFS_MAX_CIPHER_NAME_SIZE); - tmp_tfm->cipher_name[ECRYPTFS_MAX_CIPHER_NAME_SIZE] = '\0'; + strscpy(tmp_tfm->cipher_name, cipher_name); tmp_tfm->key_size = key_size; rc = ecryptfs_process_key_cipher(&tmp_tfm->key_tfm, tmp_tfm->cipher_name, diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 2dc927ba067f..577c56302314 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -256,11 +256,8 @@ static int ecryptfs_parse_options(struct ecryptfs_sb_info *sbi, char *options, substring_t args[MAX_OPT_ARGS]; int token; char *sig_src; - char *cipher_name_dst; char *cipher_name_src; - char *fn_cipher_name_dst; char *fn_cipher_name_src; - char *fnek_dst; char *fnek_src; char *cipher_key_bytes_src; char *fn_cipher_key_bytes_src; @@ -293,12 +290,8 @@ static int ecryptfs_parse_options(struct ecryptfs_sb_info *sbi, char *options, case ecryptfs_opt_cipher: case ecryptfs_opt_ecryptfs_cipher: cipher_name_src = args[0].from; - cipher_name_dst = - mount_crypt_stat-> - global_default_cipher_name; - strncpy(cipher_name_dst, cipher_name_src, - ECRYPTFS_MAX_CIPHER_NAME_SIZE); - cipher_name_dst[ECRYPTFS_MAX_CIPHER_NAME_SIZE] = '\0'; + strscpy(mount_crypt_stat->global_default_cipher_name, + cipher_name_src); cipher_name_set = 1; break; case ecryptfs_opt_ecryptfs_key_bytes: @@ -326,11 +319,8 @@ static int ecryptfs_parse_options(struct ecryptfs_sb_info *sbi, char *options, break; case ecryptfs_opt_fnek_sig: fnek_src = args[0].from; - fnek_dst = - mount_crypt_stat->global_default_fnek_sig; - strncpy(fnek_dst, fnek_src, ECRYPTFS_SIG_SIZE_HEX); - mount_crypt_stat->global_default_fnek_sig[ - ECRYPTFS_SIG_SIZE_HEX] = '\0'; + strscpy(mount_crypt_stat->global_default_fnek_sig, + fnek_src); rc = ecryptfs_add_global_auth_tok( mount_crypt_stat, mount_crypt_stat->global_default_fnek_sig, @@ -348,12 +338,8 @@ static int ecryptfs_parse_options(struct ecryptfs_sb_info *sbi, char *options, break; case ecryptfs_opt_fn_cipher: fn_cipher_name_src = args[0].from; - fn_cipher_name_dst = - mount_crypt_stat->global_default_fn_cipher_name; - strncpy(fn_cipher_name_dst, fn_cipher_name_src, - ECRYPTFS_MAX_CIPHER_NAME_SIZE); - mount_crypt_stat->global_default_fn_cipher_name[ - ECRYPTFS_MAX_CIPHER_NAME_SIZE] = '\0'; + strscpy(mount_crypt_stat->global_default_fn_cipher_name, + fn_cipher_name_src); fn_cipher_name_set = 1; break; case ecryptfs_opt_fn_cipher_key_bytes: -- cgit v1.2.3-59-g8ed1b From fd1a745ce03e37945674c14833870a9af0882e2d Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Thu, 21 Mar 2024 14:24:42 +0000 Subject: mm: support page_mapcount() on page_has_type() pages Return 0 for pages which can't be mapped. This matches how page_mapped() works. It is more convenient for users to not have to filter out these pages. Link: https://lkml.kernel.org/r/20240321142448.1645400-5-willy@infradead.org Fixes: 9c5ccf2db04b ("mm: remove HUGETLB_PAGE_DTOR") Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Hildenbrand Acked-by: Vlastimil Babka Cc: Miaohe Lin Cc: Muchun Song Cc: Oscar Salvador Cc: Signed-off-by: Andrew Morton --- fs/proc/page.c | 7 ++----- include/linux/mm.h | 8 +++++--- include/linux/page-flags.h | 4 ++-- 3 files changed, 9 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/proc/page.c b/fs/proc/page.c index 195b077c0fac..9223856c934b 100644 --- a/fs/proc/page.c +++ b/fs/proc/page.c @@ -67,7 +67,7 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf, */ ppage = pfn_to_online_page(pfn); - if (!ppage || PageSlab(ppage) || page_has_type(ppage)) + if (!ppage) pcount = 0; else pcount = page_mapcount(ppage); @@ -124,11 +124,8 @@ u64 stable_page_flags(struct page *page) /* * pseudo flags for the well known (anonymous) memory mapped pages - * - * Note that page->_mapcount is overloaded in SLAB, so the - * simple test in page_mapped() is not enough. */ - if (!PageSlab(page) && page_mapped(page)) + if (page_mapped(page)) u |= 1 << KPF_MMAP; if (PageAnon(page)) u |= 1 << KPF_ANON; diff --git a/include/linux/mm.h b/include/linux/mm.h index 7b0ee64225de..b6bdaa18b9e9 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1223,14 +1223,16 @@ static inline void page_mapcount_reset(struct page *page) * a large folio, it includes the number of times this page is mapped * as part of that folio. * - * The result is undefined for pages which cannot be mapped into userspace. - * For example SLAB or special types of pages. See function page_has_type(). - * They use this field in struct page differently. + * Will report 0 for pages which cannot be mapped into userspace, eg + * slab, page tables and similar. */ static inline int page_mapcount(struct page *page) { int mapcount = atomic_read(&page->_mapcount) + 1; + /* Handle page_has_type() pages */ + if (mapcount < 0) + mapcount = 0; if (unlikely(PageCompound(page))) mapcount += folio_entire_mapcount(page_folio(page)); diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index dc1607f1415e..35a0087d0910 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -971,12 +971,12 @@ static inline bool is_page_hwpoison(struct page *page) * page_type may be used. Because it is initialised to -1, we invert the * sense of the bit, so __SetPageFoo *clears* the bit used for PageFoo, and * __ClearPageFoo *sets* the bit used for PageFoo. We reserve a few high and - * low bits so that an underflow or overflow of page_mapcount() won't be + * low bits so that an underflow or overflow of _mapcount won't be * mistaken for a page type value. */ #define PAGE_TYPE_BASE 0xf0000000 -/* Reserve 0x0000007f to catch underflows of page_mapcount */ +/* Reserve 0x0000007f to catch underflows of _mapcount */ #define PAGE_MAPCOUNT_RESERVE -128 #define PG_buddy 0x00000080 #define PG_offline 0x00000100 -- cgit v1.2.3-59-g8ed1b From 958b9f85f8d9d884045ed4b93b2082090e617f97 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 25 Apr 2024 00:15:46 -0400 Subject: erofs_buf: store address_space instead of inode ... seeing that ->i_mapping is the only thing we want from the inode. Signed-off-by: Al Viro --- fs/erofs/data.c | 7 +++---- fs/erofs/dir.c | 2 +- fs/erofs/internal.h | 2 +- fs/erofs/namei.c | 4 ++-- fs/erofs/xattr.c | 2 +- fs/erofs/zdata.c | 2 +- 6 files changed, 9 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/erofs/data.c b/fs/erofs/data.c index d3c446dda2ff..e1a170e45c70 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -32,7 +32,6 @@ void erofs_put_metabuf(struct erofs_buf *buf) void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset, enum erofs_kmap_type type) { - struct inode *inode = buf->inode; pgoff_t index = offset >> PAGE_SHIFT; struct page *page = buf->page; struct folio *folio; @@ -42,7 +41,7 @@ void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset, erofs_put_metabuf(buf); nofs_flag = memalloc_nofs_save(); - folio = read_cache_folio(inode->i_mapping, index, NULL, NULL); + folio = read_cache_folio(buf->mapping, index, NULL, NULL); memalloc_nofs_restore(nofs_flag); if (IS_ERR(folio)) return folio; @@ -67,9 +66,9 @@ void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset, void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb) { if (erofs_is_fscache_mode(sb)) - buf->inode = EROFS_SB(sb)->s_fscache->inode; + buf->mapping = EROFS_SB(sb)->s_fscache->inode->i_mapping; else - buf->inode = sb->s_bdev->bd_inode; + buf->mapping = sb->s_bdev->bd_inode->i_mapping; } void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb, diff --git a/fs/erofs/dir.c b/fs/erofs/dir.c index 9d38f39bb4f7..2193a6710c8f 100644 --- a/fs/erofs/dir.c +++ b/fs/erofs/dir.c @@ -58,7 +58,7 @@ static int erofs_readdir(struct file *f, struct dir_context *ctx) int err = 0; bool initial = true; - buf.inode = dir; + buf.mapping = dir->i_mapping; while (ctx->pos < dirsize) { struct erofs_dirent *de; unsigned int nameoff, maxsize; diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 9e30c67c135c..12a179818897 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -223,7 +223,7 @@ enum erofs_kmap_type { }; struct erofs_buf { - struct inode *inode; + struct address_space *mapping; struct page *page; void *base; enum erofs_kmap_type kmap_type; diff --git a/fs/erofs/namei.c b/fs/erofs/namei.c index 11afa48996a3..c94d0c1608a8 100644 --- a/fs/erofs/namei.c +++ b/fs/erofs/namei.c @@ -99,7 +99,7 @@ static void *erofs_find_target_block(struct erofs_buf *target, struct erofs_buf buf = __EROFS_BUF_INITIALIZER; struct erofs_dirent *de; - buf.inode = dir; + buf.mapping = dir->i_mapping; de = erofs_bread(&buf, erofs_pos(dir->i_sb, mid), EROFS_KMAP); if (!IS_ERR(de)) { const int nameoff = nameoff_from_disk(de->nameoff, bsz); @@ -171,7 +171,7 @@ int erofs_namei(struct inode *dir, const struct qstr *name, erofs_nid_t *nid, qn.name = name->name; qn.end = name->name + name->len; - buf.inode = dir; + buf.mapping = dir->i_mapping; ndirents = 0; de = erofs_find_target_block(&buf, dir, &qn, &ndirents); diff --git a/fs/erofs/xattr.c b/fs/erofs/xattr.c index ec233917830a..a90d7d649739 100644 --- a/fs/erofs/xattr.c +++ b/fs/erofs/xattr.c @@ -483,7 +483,7 @@ int erofs_xattr_prefixes_init(struct super_block *sb) return -ENOMEM; if (sbi->packed_inode) - buf.inode = sbi->packed_inode; + buf.mapping = sbi->packed_inode->i_mapping; else erofs_init_metabuf(&buf, sb); diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 9ffdae7fcd5b..283c9c3a611d 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -936,7 +936,7 @@ static int z_erofs_read_fragment(struct super_block *sb, struct page *page, if (!packed_inode) return -EFSCORRUPTED; - buf.inode = packed_inode; + buf.mapping = packed_inode->i_mapping; for (; cur < end; cur += cnt, pos += cnt) { cnt = min_t(unsigned int, end - cur, sb->s_blocksize - erofs_blkoff(sb, pos)); -- cgit v1.2.3-59-g8ed1b From 89c6c1d91ab2dc05c6f4ad504b03169aa32694cc Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Wed, 20 Mar 2024 19:05:44 +0800 Subject: iomap: drop the write failure handles when unsharing and zeroing Unsharing and zeroing can only happen within EOF, so there is never a need to perform posteof pagecache truncation if write begin fails, also partial write could never theoretically happened from iomap_write_end(), so remove both of them. Signed-off-by: Zhang Yi Link: https://lore.kernel.org/r/20240320110548.2200662-6-yi.zhang@huaweicloud.com Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index e09441f4fceb..2bd9e780ff34 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -824,7 +824,6 @@ static int iomap_write_begin(struct iomap_iter *iter, loff_t pos, out_unlock: __iomap_put_folio(iter, pos, 0, folio); - iomap_write_failed(iter->inode, pos, len); return status; } @@ -901,8 +900,6 @@ static size_t iomap_write_end(struct iomap_iter *iter, loff_t pos, size_t len, if (old_size < pos) pagecache_isize_extended(iter->inode, old_size, pos); - if (ret < len) - iomap_write_failed(iter->inode, pos + ret, len - ret); return ret; } @@ -950,8 +947,10 @@ retry: } status = iomap_write_begin(iter, pos, bytes, &folio); - if (unlikely(status)) + if (unlikely(status)) { + iomap_write_failed(iter->inode, pos, bytes); break; + } if (iter->iomap.flags & IOMAP_F_STALE) break; @@ -965,6 +964,9 @@ retry: copied = copy_folio_from_iter_atomic(folio, offset, bytes, i); status = iomap_write_end(iter, pos, bytes, copied, folio); + if (status < bytes) + iomap_write_failed(iter->inode, pos + status, + bytes - status); if (unlikely(copied != status)) iov_iter_revert(i, copied - status); -- cgit v1.2.3-59-g8ed1b From 943bc0882cebf482422640924062a7daac5a27ba Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Wed, 20 Mar 2024 19:05:45 +0800 Subject: iomap: don't increase i_size if it's not a write operation Increase i_size in iomap_zero_range() and iomap_unshare_iter() is not needed, the caller should handle it. Especially, when truncate partial block, we should not increase i_size beyond the new EOF here. It doesn't affect xfs and gfs2 now because they set the new file size after zero out, it doesn't matter that a transient increase in i_size, but it will affect ext4 because it set file size before truncate. So move the i_size updating logic to iomap_write_iter(). Signed-off-by: Zhang Yi Link: https://lore.kernel.org/r/20240320110548.2200662-7-yi.zhang@huaweicloud.com Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 50 +++++++++++++++++++++++++------------------------- 1 file changed, 25 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 2bd9e780ff34..620c830dfec4 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -875,32 +875,13 @@ static size_t iomap_write_end(struct iomap_iter *iter, loff_t pos, size_t len, size_t copied, struct folio *folio) { const struct iomap *srcmap = iomap_iter_srcmap(iter); - loff_t old_size = iter->inode->i_size; - size_t ret; - - if (srcmap->type == IOMAP_INLINE) { - ret = iomap_write_end_inline(iter, folio, pos, copied); - } else if (srcmap->flags & IOMAP_F_BUFFER_HEAD) { - ret = block_write_end(NULL, iter->inode->i_mapping, pos, len, - copied, &folio->page, NULL); - } else { - ret = __iomap_write_end(iter->inode, pos, len, copied, folio); - } - - /* - * Update the in-memory inode size after copying the data into the page - * cache. It's up to the file system to write the updated size to disk, - * preferably after I/O completion so that no stale data is exposed. - */ - if (pos + ret > old_size) { - i_size_write(iter->inode, pos + ret); - iter->iomap.flags |= IOMAP_F_SIZE_CHANGED; - } - __iomap_put_folio(iter, pos, ret, folio); - if (old_size < pos) - pagecache_isize_extended(iter->inode, old_size, pos); - return ret; + if (srcmap->type == IOMAP_INLINE) + return iomap_write_end_inline(iter, folio, pos, copied); + if (srcmap->flags & IOMAP_F_BUFFER_HEAD) + return block_write_end(NULL, iter->inode->i_mapping, pos, len, + copied, &folio->page, NULL); + return __iomap_write_end(iter->inode, pos, len, copied, folio); } static loff_t iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i) @@ -915,6 +896,7 @@ static loff_t iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i) do { struct folio *folio; + loff_t old_size; size_t offset; /* Offset into folio */ size_t bytes; /* Bytes to write to folio */ size_t copied; /* Bytes copied from user */ @@ -964,6 +946,22 @@ retry: copied = copy_folio_from_iter_atomic(folio, offset, bytes, i); status = iomap_write_end(iter, pos, bytes, copied, folio); + /* + * Update the in-memory inode size after copying the data into + * the page cache. It's up to the file system to write the + * updated size to disk, preferably after I/O completion so that + * no stale data is exposed. Only once that's done can we + * unlock and release the folio. + */ + old_size = iter->inode->i_size; + if (pos + status > old_size) { + i_size_write(iter->inode, pos + status); + iter->iomap.flags |= IOMAP_F_SIZE_CHANGED; + } + __iomap_put_folio(iter, pos, status, folio); + + if (old_size < pos) + pagecache_isize_extended(iter->inode, old_size, pos); if (status < bytes) iomap_write_failed(iter->inode, pos + status, bytes - status); @@ -1336,6 +1334,7 @@ static loff_t iomap_unshare_iter(struct iomap_iter *iter) bytes = folio_size(folio) - offset; bytes = iomap_write_end(iter, pos, bytes, bytes, folio); + __iomap_put_folio(iter, pos, bytes, folio); if (WARN_ON_ONCE(bytes == 0)) return -EIO; @@ -1400,6 +1399,7 @@ static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero) folio_mark_accessed(folio); bytes = iomap_write_end(iter, pos, bytes, bytes, folio); + __iomap_put_folio(iter, pos, bytes, folio); if (WARN_ON_ONCE(bytes == 0)) return -EIO; -- cgit v1.2.3-59-g8ed1b From 1a61d74932d460f47ffaf08927dbe9d43315841f Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Wed, 20 Mar 2024 19:05:46 +0800 Subject: iomap: use a new variable to handle the written bytes in iomap_write_iter() In iomap_write_iter(), the status variable used to receive the return value from iomap_write_end() is confusing, replace it with a new written variable to represent the written bytes in each cycle, no logic changes. Signed-off-by: Zhang Yi Link: https://lore.kernel.org/r/20240320110548.2200662-8-yi.zhang@huaweicloud.com Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 33 +++++++++++++++++---------------- 1 file changed, 17 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 620c830dfec4..1333a0d96f83 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -889,7 +889,7 @@ static loff_t iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i) loff_t length = iomap_length(iter); size_t chunk = PAGE_SIZE << MAX_PAGECACHE_ORDER; loff_t pos = iter->pos; - ssize_t written = 0; + ssize_t total_written = 0; long status = 0; struct address_space *mapping = iter->inode->i_mapping; unsigned int bdp_flags = (iter->flags & IOMAP_NOWAIT) ? BDP_ASYNC : 0; @@ -900,6 +900,7 @@ static loff_t iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i) size_t offset; /* Offset into folio */ size_t bytes; /* Bytes to write to folio */ size_t copied; /* Bytes copied from user */ + size_t written; /* Bytes have been written */ bytes = iov_iter_count(i); retry: @@ -944,7 +945,7 @@ retry: flush_dcache_folio(folio); copied = copy_folio_from_iter_atomic(folio, offset, bytes, i); - status = iomap_write_end(iter, pos, bytes, copied, folio); + written = iomap_write_end(iter, pos, bytes, copied, folio); /* * Update the in-memory inode size after copying the data into @@ -954,22 +955,22 @@ retry: * unlock and release the folio. */ old_size = iter->inode->i_size; - if (pos + status > old_size) { - i_size_write(iter->inode, pos + status); + if (pos + written > old_size) { + i_size_write(iter->inode, pos + written); iter->iomap.flags |= IOMAP_F_SIZE_CHANGED; } - __iomap_put_folio(iter, pos, status, folio); + __iomap_put_folio(iter, pos, written, folio); if (old_size < pos) pagecache_isize_extended(iter->inode, old_size, pos); - if (status < bytes) - iomap_write_failed(iter->inode, pos + status, - bytes - status); - if (unlikely(copied != status)) - iov_iter_revert(i, copied - status); + if (written < bytes) + iomap_write_failed(iter->inode, pos + written, + bytes - written); + if (unlikely(copied != written)) + iov_iter_revert(i, copied - written); cond_resched(); - if (unlikely(status == 0)) { + if (unlikely(written == 0)) { /* * A short copy made iomap_write_end() reject the * thing entirely. Might be memory poisoning @@ -983,17 +984,17 @@ retry: goto retry; } } else { - pos += status; - written += status; - length -= status; + pos += written; + total_written += written; + length -= written; } } while (iov_iter_count(i) && length); if (status == -EAGAIN) { - iov_iter_revert(i, written); + iov_iter_revert(i, total_written); return -EAGAIN; } - return written ? written : status; + return total_written ? total_written : status; } ssize_t -- cgit v1.2.3-59-g8ed1b From 815f4b633ba1c9c6a151f251616f23e2407fcf24 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Wed, 20 Mar 2024 19:05:47 +0800 Subject: iomap: make iomap_write_end() return a boolean For now, we can make sure iomap_write_end() always return 0 or copied bytes, so instead of return written bytes, convert to return a boolean to indicate the copied bytes have been written to the pagecache. Signed-off-by: Zhang Yi Link: https://lore.kernel.org/r/20240320110548.2200662-9-yi.zhang@huaweicloud.com Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 48 +++++++++++++++++++++++++++++++----------------- 1 file changed, 31 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 1333a0d96f83..73241fbaff6b 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -828,7 +828,7 @@ out_unlock: return status; } -static size_t __iomap_write_end(struct inode *inode, loff_t pos, size_t len, +static bool __iomap_write_end(struct inode *inode, loff_t pos, size_t len, size_t copied, struct folio *folio) { flush_dcache_folio(folio); @@ -845,14 +845,14 @@ static size_t __iomap_write_end(struct inode *inode, loff_t pos, size_t len, * redo the whole thing. */ if (unlikely(copied < len && !folio_test_uptodate(folio))) - return 0; + return false; iomap_set_range_uptodate(folio, offset_in_folio(folio, pos), len); iomap_set_range_dirty(folio, offset_in_folio(folio, pos), copied); filemap_dirty_folio(inode->i_mapping, folio); - return copied; + return true; } -static size_t iomap_write_end_inline(const struct iomap_iter *iter, +static void iomap_write_end_inline(const struct iomap_iter *iter, struct folio *folio, loff_t pos, size_t copied) { const struct iomap *iomap = &iter->iomap; @@ -867,20 +867,31 @@ static size_t iomap_write_end_inline(const struct iomap_iter *iter, kunmap_local(addr); mark_inode_dirty(iter->inode); - return copied; } -/* Returns the number of bytes copied. May be 0. Cannot be an errno. */ -static size_t iomap_write_end(struct iomap_iter *iter, loff_t pos, size_t len, +/* + * Returns true if all copied bytes have been written to the pagecache, + * otherwise return false. + */ +static bool iomap_write_end(struct iomap_iter *iter, loff_t pos, size_t len, size_t copied, struct folio *folio) { const struct iomap *srcmap = iomap_iter_srcmap(iter); - if (srcmap->type == IOMAP_INLINE) - return iomap_write_end_inline(iter, folio, pos, copied); - if (srcmap->flags & IOMAP_F_BUFFER_HEAD) - return block_write_end(NULL, iter->inode->i_mapping, pos, len, - copied, &folio->page, NULL); + if (srcmap->type == IOMAP_INLINE) { + iomap_write_end_inline(iter, folio, pos, copied); + return true; + } + + if (srcmap->flags & IOMAP_F_BUFFER_HEAD) { + size_t bh_written; + + bh_written = block_write_end(NULL, iter->inode->i_mapping, pos, + len, copied, &folio->page, NULL); + WARN_ON_ONCE(bh_written != copied && bh_written != 0); + return bh_written == copied; + } + return __iomap_write_end(iter->inode, pos, len, copied, folio); } @@ -945,7 +956,8 @@ retry: flush_dcache_folio(folio); copied = copy_folio_from_iter_atomic(folio, offset, bytes, i); - written = iomap_write_end(iter, pos, bytes, copied, folio); + written = iomap_write_end(iter, pos, bytes, copied, folio) ? + copied : 0; /* * Update the in-memory inode size after copying the data into @@ -1323,6 +1335,7 @@ static loff_t iomap_unshare_iter(struct iomap_iter *iter) int status; size_t offset; size_t bytes = min_t(u64, SIZE_MAX, length); + bool ret; status = iomap_write_begin(iter, pos, bytes, &folio); if (unlikely(status)) @@ -1334,9 +1347,9 @@ static loff_t iomap_unshare_iter(struct iomap_iter *iter) if (bytes > folio_size(folio) - offset) bytes = folio_size(folio) - offset; - bytes = iomap_write_end(iter, pos, bytes, bytes, folio); + ret = iomap_write_end(iter, pos, bytes, bytes, folio); __iomap_put_folio(iter, pos, bytes, folio); - if (WARN_ON_ONCE(bytes == 0)) + if (WARN_ON_ONCE(!ret)) return -EIO; cond_resched(); @@ -1385,6 +1398,7 @@ static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero) int status; size_t offset; size_t bytes = min_t(u64, SIZE_MAX, length); + bool ret; status = iomap_write_begin(iter, pos, bytes, &folio); if (status) @@ -1399,9 +1413,9 @@ static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero) folio_zero_range(folio, offset, bytes); folio_mark_accessed(folio); - bytes = iomap_write_end(iter, pos, bytes, bytes, folio); + ret = iomap_write_end(iter, pos, bytes, bytes, folio); __iomap_put_folio(iter, pos, bytes, folio); - if (WARN_ON_ONCE(bytes == 0)) + if (WARN_ON_ONCE(!ret)) return -EIO; pos += bytes; -- cgit v1.2.3-59-g8ed1b From e1f453d4336d5d7fbbd1910532201b4a07a20a5c Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Wed, 20 Mar 2024 19:05:48 +0800 Subject: iomap: do some small logical cleanup in buffered write Since iomap_write_end() can never return a partial write length, the comparison between written, copied and bytes becomes useless, just merge them with the unwritten branch. Signed-off-by: Zhang Yi Link: https://lore.kernel.org/r/20240320110548.2200662-10-yi.zhang@huaweicloud.com Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 73241fbaff6b..41c8f0c68ef5 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -975,11 +975,6 @@ retry: if (old_size < pos) pagecache_isize_extended(iter->inode, old_size, pos); - if (written < bytes) - iomap_write_failed(iter->inode, pos + written, - bytes - written); - if (unlikely(copied != written)) - iov_iter_revert(i, copied - written); cond_resched(); if (unlikely(written == 0)) { @@ -989,6 +984,9 @@ retry: * halfway through, might be a race with munmap, * might be severe memory pressure. */ + iomap_write_failed(iter->inode, pos, bytes); + iov_iter_revert(i, copied); + if (chunk > PAGE_SIZE) chunk /= 2; if (copied) { -- cgit v1.2.3-59-g8ed1b From f19257997d9c31864b4cf3fcef6e2d2b9ede148d Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Fri, 12 Jan 2024 14:48:46 +0800 Subject: exfat: zero the reserved fields of file and stream extension dentries From exFAT specification, the reserved fields should initialize to zero and should not use for any purpose. If create a new dentry set in the UNUSED dentries, all fields had been zeroed when allocating cluster to parent directory. But if create a new dentry set in the DELETED dentries, the reserved fields in file and stream extension dentries may be non-zero. Because only the valid bit of the type field of the dentry is cleared in exfat_remove_entries(), if the type of dentry is different from the original(For example, a dentry that was originally a file name dentry, then set to deleted dentry, and then set as a file dentry), the reserved fields is non-zero. So this commit initializes the dentry to 0 before createing file dentry and stream extension dentry. Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Aoyama Wataru Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- fs/exfat/dir.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/exfat/dir.c b/fs/exfat/dir.c index 077944d3c2c0..84572e11cc05 100644 --- a/fs/exfat/dir.c +++ b/fs/exfat/dir.c @@ -420,6 +420,7 @@ static void exfat_set_entry_type(struct exfat_dentry *ep, unsigned int type) static void exfat_init_stream_entry(struct exfat_dentry *ep, unsigned int start_clu, unsigned long long size) { + memset(ep, 0, sizeof(*ep)); exfat_set_entry_type(ep, TYPE_STREAM); if (size == 0) ep->dentry.stream.flags = ALLOC_FAT_CHAIN; @@ -457,6 +458,7 @@ void exfat_init_dir_entry(struct exfat_entry_set_cache *es, struct exfat_dentry *ep; ep = exfat_get_dentry_cached(es, ES_IDX_FILE); + memset(ep, 0, sizeof(*ep)); exfat_set_entry_type(ep, type); exfat_set_entry_time(sbi, ts, &ep->dentry.file.create_tz, -- cgit v1.2.3-59-g8ed1b From 9af503d91298c3f2945e73703f0e00995be08c30 Mon Sep 17 00:00:00 2001 From: Dominique Martinet Date: Fri, 19 Apr 2024 11:22:48 +0900 Subject: btrfs: add missing mutex_unlock in btrfs_relocate_sys_chunks() The previous patch that replaced BUG_ON by error handling forgot to unlock the mutex in the error path. Link: https://lore.kernel.org/all/Zh%2fHpAGFqa7YAFuM@duo.ucw.cz Reported-by: Pavel Machek Fixes: 7411055db5ce ("btrfs: handle chunk tree lookup error in btrfs_relocate_sys_chunks()") CC: stable@vger.kernel.org Reviewed-by: Pavel Machek Signed-off-by: Dominique Martinet Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index dedec3d9b111..c72c351fe7eb 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -3419,6 +3419,7 @@ again: * alignment and size). */ ret = -EUCLEAN; + mutex_unlock(&fs_info->reclaim_bgs_lock); goto error; } -- cgit v1.2.3-59-g8ed1b From 0f2b8098d72a93890e69aa24ec549ef4bc34f4db Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 19 Apr 2024 14:38:48 -0400 Subject: btrfs: take the cleaner_mutex earlier in qgroup disable One of my CI runs popped the following lockdep splat ====================================================== WARNING: possible circular locking dependency detected 6.9.0-rc4+ #1 Not tainted ------------------------------------------------------ btrfs/471533 is trying to acquire lock: ffff92ba46980850 (&fs_info->cleaner_mutex){+.+.}-{3:3}, at: btrfs_quota_disable+0x54/0x4c0 but task is already holding lock: ffff92ba46980bd0 (&fs_info->subvol_sem){++++}-{3:3}, at: btrfs_ioctl+0x1c8f/0x2600 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #2 (&fs_info->subvol_sem){++++}-{3:3}: down_read+0x42/0x170 btrfs_rename+0x607/0xb00 btrfs_rename2+0x2e/0x70 vfs_rename+0xaf8/0xfc0 do_renameat2+0x586/0x600 __x64_sys_rename+0x43/0x50 do_syscall_64+0x95/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e -> #1 (&sb->s_type->i_mutex_key#16){++++}-{3:3}: down_write+0x3f/0xc0 btrfs_inode_lock+0x40/0x70 prealloc_file_extent_cluster+0x1b0/0x370 relocate_file_extent_cluster+0xb2/0x720 relocate_data_extent+0x107/0x160 relocate_block_group+0x442/0x550 btrfs_relocate_block_group+0x2cb/0x4b0 btrfs_relocate_chunk+0x50/0x1b0 btrfs_balance+0x92f/0x13d0 btrfs_ioctl+0x1abf/0x2600 __x64_sys_ioctl+0x97/0xd0 do_syscall_64+0x95/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e -> #0 (&fs_info->cleaner_mutex){+.+.}-{3:3}: __lock_acquire+0x13e7/0x2180 lock_acquire+0xcb/0x2e0 __mutex_lock+0xbe/0xc00 btrfs_quota_disable+0x54/0x4c0 btrfs_ioctl+0x206b/0x2600 __x64_sys_ioctl+0x97/0xd0 do_syscall_64+0x95/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e other info that might help us debug this: Chain exists of: &fs_info->cleaner_mutex --> &sb->s_type->i_mutex_key#16 --> &fs_info->subvol_sem Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&fs_info->subvol_sem); lock(&sb->s_type->i_mutex_key#16); lock(&fs_info->subvol_sem); lock(&fs_info->cleaner_mutex); *** DEADLOCK *** 2 locks held by btrfs/471533: #0: ffff92ba4319e420 (sb_writers#14){.+.+}-{0:0}, at: btrfs_ioctl+0x3b5/0x2600 #1: ffff92ba46980bd0 (&fs_info->subvol_sem){++++}-{3:3}, at: btrfs_ioctl+0x1c8f/0x2600 stack backtrace: CPU: 1 PID: 471533 Comm: btrfs Kdump: loaded Not tainted 6.9.0-rc4+ #1 Call Trace: dump_stack_lvl+0x77/0xb0 check_noncircular+0x148/0x160 ? lock_acquire+0xcb/0x2e0 __lock_acquire+0x13e7/0x2180 lock_acquire+0xcb/0x2e0 ? btrfs_quota_disable+0x54/0x4c0 ? lock_is_held_type+0x9a/0x110 __mutex_lock+0xbe/0xc00 ? btrfs_quota_disable+0x54/0x4c0 ? srso_return_thunk+0x5/0x5f ? lock_acquire+0xcb/0x2e0 ? btrfs_quota_disable+0x54/0x4c0 ? btrfs_quota_disable+0x54/0x4c0 btrfs_quota_disable+0x54/0x4c0 btrfs_ioctl+0x206b/0x2600 ? srso_return_thunk+0x5/0x5f ? __do_sys_statfs+0x61/0x70 __x64_sys_ioctl+0x97/0xd0 do_syscall_64+0x95/0x180 ? srso_return_thunk+0x5/0x5f ? reacquire_held_locks+0xd1/0x1f0 ? do_user_addr_fault+0x307/0x8a0 ? srso_return_thunk+0x5/0x5f ? lock_acquire+0xcb/0x2e0 ? srso_return_thunk+0x5/0x5f ? srso_return_thunk+0x5/0x5f ? find_held_lock+0x2b/0x80 ? srso_return_thunk+0x5/0x5f ? lock_release+0xca/0x2a0 ? srso_return_thunk+0x5/0x5f ? do_user_addr_fault+0x35c/0x8a0 ? srso_return_thunk+0x5/0x5f ? trace_hardirqs_off+0x4b/0xc0 ? srso_return_thunk+0x5/0x5f ? lockdep_hardirqs_on_prepare+0xde/0x190 ? srso_return_thunk+0x5/0x5f This happens because when we call rename we already have the inode mutex held, and then we acquire the subvol_sem if we are a subvolume. This makes the dependency inode lock -> subvol sem When we're running data relocation we will preallocate space for the data relocation inode, and we always run the relocation under the ->cleaner_mutex. This now creates the dependency of cleaner_mutex -> inode lock (from the prealloc) -> subvol_sem Qgroup delete is doing this in the opposite order, it is acquiring the subvol_sem and then it is acquiring the cleaner_mutex, which results in this lockdep splat. This deadlock can't happen in reality, because we won't ever rename the data reloc inode, nor is the data reloc inode a subvolume. However this is fairly easy to fix, simply take the cleaner mutex in the case where we are disabling qgroups before we take the subvol_sem. This resolves the lockdep splat. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ioctl.c | 33 ++++++++++++++++++++++++++++++--- fs/btrfs/qgroup.c | 21 ++++++++------------- 2 files changed, 38 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 5a507237c4fa..12ede85018bf 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -3758,15 +3758,43 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg) goto drop_write; } - down_write(&fs_info->subvol_sem); - switch (sa->cmd) { case BTRFS_QUOTA_CTL_ENABLE: case BTRFS_QUOTA_CTL_ENABLE_SIMPLE_QUOTA: + down_write(&fs_info->subvol_sem); ret = btrfs_quota_enable(fs_info, sa); + up_write(&fs_info->subvol_sem); break; case BTRFS_QUOTA_CTL_DISABLE: + /* + * Lock the cleaner mutex to prevent races with concurrent + * relocation, because relocation may be building backrefs for + * blocks of the quota root while we are deleting the root. This + * is like dropping fs roots of deleted snapshots/subvolumes, we + * need the same protection. + * + * This also prevents races between concurrent tasks trying to + * disable quotas, because we will unlock and relock + * qgroup_ioctl_lock across BTRFS_FS_QUOTA_ENABLED changes. + * + * We take this here because we have the dependency of + * + * inode_lock -> subvol_sem + * + * because of rename. With relocation we can prealloc extents, + * so that makes the dependency chain + * + * cleaner_mutex -> inode_lock -> subvol_sem + * + * so we must take the cleaner_mutex here before we take the + * subvol_sem. The deadlock can't actually happen, but this + * quiets lockdep. + */ + mutex_lock(&fs_info->cleaner_mutex); + down_write(&fs_info->subvol_sem); ret = btrfs_quota_disable(fs_info); + up_write(&fs_info->subvol_sem); + mutex_unlock(&fs_info->cleaner_mutex); break; default: ret = -EINVAL; @@ -3774,7 +3802,6 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg) } kfree(sa); - up_write(&fs_info->subvol_sem); drop_write: mnt_drop_write_file(file); return ret; diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index cf8820ce7aa2..364acc9bbe73 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1342,16 +1342,10 @@ int btrfs_quota_disable(struct btrfs_fs_info *fs_info) lockdep_assert_held_write(&fs_info->subvol_sem); /* - * Lock the cleaner mutex to prevent races with concurrent relocation, - * because relocation may be building backrefs for blocks of the quota - * root while we are deleting the root. This is like dropping fs roots - * of deleted snapshots/subvolumes, we need the same protection. - * - * This also prevents races between concurrent tasks trying to disable - * quotas, because we will unlock and relock qgroup_ioctl_lock across - * BTRFS_FS_QUOTA_ENABLED changes. + * Relocation will mess with backrefs, so make sure we have the + * cleaner_mutex held to protect us from relocate. */ - mutex_lock(&fs_info->cleaner_mutex); + lockdep_assert_held(&fs_info->cleaner_mutex); mutex_lock(&fs_info->qgroup_ioctl_lock); if (!fs_info->quota_root) @@ -1373,9 +1367,13 @@ int btrfs_quota_disable(struct btrfs_fs_info *fs_info) clear_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags); btrfs_qgroup_wait_for_completion(fs_info, false); + /* + * We have nothing held here and no trans handle, just return the error + * if there is one. + */ ret = flush_reservations(fs_info); if (ret) - goto out_unlock_cleaner; + return ret; /* * 1 For the root item @@ -1439,9 +1437,6 @@ out: btrfs_end_transaction(trans); else if (trans) ret = btrfs_commit_transaction(trans); -out_unlock_cleaner: - mutex_unlock(&fs_info->cleaner_mutex); - return ret; } -- cgit v1.2.3-59-g8ed1b From 2174035a7f1148a52f5a3f371f04224168b5b00a Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Fri, 5 Apr 2024 01:15:42 +0000 Subject: f2fs: clear writeback when compression failed Let's stop issuing compressed writes and clear their writeback flags. Reviewed-by: Daeho Jeong Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 40 ++++++++++++++++++++++++++++++++++++++-- 1 file changed, 38 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index d67c471ab5df..b12d3a49bfda 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -1031,6 +1031,31 @@ static void set_cluster_writeback(struct compress_ctx *cc) } } +static void cancel_cluster_writeback(struct compress_ctx *cc, + struct compress_io_ctx *cic, int submitted) +{ + int i; + + /* Wait for submitted IOs. */ + if (submitted > 1) { + f2fs_submit_merged_write(F2FS_I_SB(cc->inode), DATA); + while (atomic_read(&cic->pending_pages) != + (cc->valid_nr_cpages - submitted + 1)) + f2fs_io_schedule_timeout(DEFAULT_IO_TIMEOUT); + } + + /* Cancel writeback and stay locked. */ + for (i = 0; i < cc->cluster_size; i++) { + if (i < submitted) { + inode_inc_dirty_pages(cc->inode); + lock_page(cc->rpages[i]); + } + clear_page_private_gcing(cc->rpages[i]); + if (folio_test_writeback(page_folio(cc->rpages[i]))) + end_page_writeback(cc->rpages[i]); + } +} + static void set_cluster_dirty(struct compress_ctx *cc) { int i; @@ -1232,7 +1257,6 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc, .page = NULL, .encrypted_page = NULL, .compressed_page = NULL, - .submitted = 0, .io_type = io_type, .io_wbc = wbc, .encrypted = fscrypt_inode_uses_fs_layer_crypto(cc->inode) ? @@ -1358,7 +1382,16 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc, fio.compressed_page = cc->cpages[i - 1]; cc->cpages[i - 1] = NULL; + fio.submitted = 0; f2fs_outplace_write_data(&dn, &fio); + if (unlikely(!fio.submitted)) { + cancel_cluster_writeback(cc, cic, i); + + /* To call fscrypt_finalize_bounce_page */ + i = cc->valid_nr_cpages; + *submitted = 0; + goto out_destroy_crypt; + } (*submitted)++; unlock_continue: inode_dec_dirty_pages(cc->inode); @@ -1392,8 +1425,11 @@ unlock_continue: out_destroy_crypt: page_array_free(cc->inode, cic->rpages, cc->cluster_size); - for (--i; i >= 0; i--) + for (--i; i >= 0; i--) { + if (!cc->cpages[i]) + continue; fscrypt_finalize_bounce_page(&cc->cpages[i]); + } out_put_cic: kmem_cache_free(cic_entry_slab, cic); out_put_dnode: -- cgit v1.2.3-59-g8ed1b From b864ddb57eb00c4ea1e6801c7b2f70f1db2a7f4b Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Tue, 9 Apr 2024 20:34:11 +0000 Subject: f2fs: fix false alarm on invalid block address f2fs_ra_meta_pages can try to read ahead on invalid block address which is not the corruption case. Cc: # v6.9+ Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=218770 Fixes: 31f85ccc84b8 ("f2fs: unify the error handling of f2fs_is_valid_blkaddr") Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/checkpoint.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 5d05a413f451..55d444bec5c0 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -179,22 +179,22 @@ static bool __f2fs_is_valid_blkaddr(struct f2fs_sb_info *sbi, break; case META_SIT: if (unlikely(blkaddr >= SIT_BLK_CNT(sbi))) - goto err; + goto check_only; break; case META_SSA: if (unlikely(blkaddr >= MAIN_BLKADDR(sbi) || blkaddr < SM_I(sbi)->ssa_blkaddr)) - goto err; + goto check_only; break; case META_CP: if (unlikely(blkaddr >= SIT_I(sbi)->sit_base_addr || blkaddr < __start_cp_addr(sbi))) - goto err; + goto check_only; break; case META_POR: if (unlikely(blkaddr >= MAX_BLKADDR(sbi) || blkaddr < MAIN_BLKADDR(sbi))) - goto err; + goto check_only; break; case DATA_GENERIC: case DATA_GENERIC_ENHANCE: @@ -228,6 +228,7 @@ static bool __f2fs_is_valid_blkaddr(struct f2fs_sb_info *sbi, return true; err: f2fs_handle_error(sbi, ERROR_INVALID_BLKADDR); +check_only: return false; } -- cgit v1.2.3-59-g8ed1b From 3763f9effcdccc38bde614c911f331203f204a01 Mon Sep 17 00:00:00 2001 From: Wu Bo Date: Tue, 23 Apr 2024 05:27:59 -0600 Subject: f2fs: use helper to print zone condition To make code clean, use blk_zone_cond_str() to print debug information. Signed-off-by: Wu Bo Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 17 +++-------------- 1 file changed, 3 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 2206199e8099..5d6dae4861f3 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -5025,17 +5025,6 @@ out: } #ifdef CONFIG_BLK_DEV_ZONED -static const char *f2fs_zone_status[BLK_ZONE_COND_OFFLINE + 1] = { - [BLK_ZONE_COND_NOT_WP] = "NOT_WP", - [BLK_ZONE_COND_EMPTY] = "EMPTY", - [BLK_ZONE_COND_IMP_OPEN] = "IMPLICIT_OPEN", - [BLK_ZONE_COND_EXP_OPEN] = "EXPLICIT_OPEN", - [BLK_ZONE_COND_CLOSED] = "CLOSED", - [BLK_ZONE_COND_READONLY] = "READONLY", - [BLK_ZONE_COND_FULL] = "FULL", - [BLK_ZONE_COND_OFFLINE] = "OFFLINE", -}; - static int check_zone_write_pointer(struct f2fs_sb_info *sbi, struct f2fs_dev_info *fdev, struct blk_zone *zone) @@ -5066,7 +5055,7 @@ static int check_zone_write_pointer(struct f2fs_sb_info *sbi, if (IS_CURSEC(sbi, GET_SEC_FROM_SEG(sbi, zone_segno))) { f2fs_notice(sbi, "Open zones: valid block[0x%x,0x%x] cond[%s]", zone_segno, valid_block_cnt, - f2fs_zone_status[zone->cond]); + blk_zone_cond_str(zone->cond)); return 0; } @@ -5077,7 +5066,7 @@ static int check_zone_write_pointer(struct f2fs_sb_info *sbi, if (!valid_block_cnt) { f2fs_notice(sbi, "Zone without valid block has non-zero write " "pointer. Reset the write pointer: cond[%s]", - f2fs_zone_status[zone->cond]); + blk_zone_cond_str(zone->cond)); ret = __f2fs_issue_discard_zone(sbi, fdev->bdev, zone_block, zone->len >> log_sectors_per_block); if (ret) @@ -5095,7 +5084,7 @@ static int check_zone_write_pointer(struct f2fs_sb_info *sbi, */ f2fs_notice(sbi, "Valid blocks are not aligned with write " "pointer: valid block[0x%x,0x%x] cond[%s]", - zone_segno, valid_block_cnt, f2fs_zone_status[zone->cond]); + zone_segno, valid_block_cnt, blk_zone_cond_str(zone->cond)); nofs_flags = memalloc_nofs_save(); ret = blkdev_zone_mgmt(fdev->bdev, REQ_OP_ZONE_FINISH, -- cgit v1.2.3-59-g8ed1b From 8094a600245e9b28eb36a13036f202ad67c1f887 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 25 Apr 2024 11:30:16 -0500 Subject: smb3: missing lock when picking channel Coverity spotted a place where we should have been holding the channel lock when accessing the ses channel index. Addresses-Coverity: 1582039 ("Data race condition (MISSING_LOCK)") Cc: stable@vger.kernel.org Reviewed-by: Shyam Prasad N Signed-off-by: Steve French --- fs/smb/client/transport.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index 994d70193432..e1a79e031b28 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -1057,9 +1057,11 @@ struct TCP_Server_Info *cifs_pick_channel(struct cifs_ses *ses) index = (uint)atomic_inc_return(&ses->chan_seq); index %= ses->chan_count; } + + server = ses->chans[index].server; spin_unlock(&ses->chan_lock); - return ses->chans[index].server; + return server; } int -- cgit v1.2.3-59-g8ed1b From 8861fd5180476f45f9e8853db154600469a0284f Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 25 Apr 2024 12:49:50 -0500 Subject: smb3: fix lock ordering potential deadlock in cifs_sync_mid_result Coverity spotted that the cifs_sync_mid_result function could deadlock "Thread deadlock (ORDER_REVERSAL) lock_order: Calling spin_lock acquires lock TCP_Server_Info.srv_lock while holding lock TCP_Server_Info.mid_lock" Addresses-Coverity: 1590401 ("Thread deadlock (ORDER_REVERSAL)") Cc: stable@vger.kernel.org Reviewed-by: Shyam Prasad N Signed-off-by: Steve French --- fs/smb/client/transport.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index e1a79e031b28..ddf1a3aafee5 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -909,12 +909,15 @@ cifs_sync_mid_result(struct mid_q_entry *mid, struct TCP_Server_Info *server) list_del_init(&mid->qhead); mid->mid_flags |= MID_DELETED; } + spin_unlock(&server->mid_lock); cifs_server_dbg(VFS, "%s: invalid mid state mid=%llu state=%d\n", __func__, mid->mid, mid->mid_state); rc = -EIO; + goto sync_mid_done; } spin_unlock(&server->mid_lock); +sync_mid_done: release_mid(mid); return rc; } -- cgit v1.2.3-59-g8ed1b From 18180a4550d08be4eb0387fe83f02f703f92d4e7 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Apr 2024 17:46:26 -0400 Subject: NFSD: Fix nfsd4_encode_fattr4() crasher Ensure that args.acl is initialized early. It is used in an unconditional call to kfree() on the way out of nfsd4_encode_fattr4(). Reported-by: Scott Mayhew Fixes: 83ab8678ad0c ("NFSD: Add struct nfsd4_fattr_args") Signed-off-by: Chuck Lever --- fs/nfsd/nfs4xdr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 1955481832e0..a644460f3a5e 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -3515,6 +3515,7 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, args.exp = exp; args.dentry = dentry; args.ignore_crossmnt = (ignore_crossmnt != 0); + args.acl = NULL; /* * Make a local copy of the attribute bitmap that can be modified. @@ -3573,7 +3574,6 @@ nfsd4_encode_fattr4(struct svc_rqst *rqstp, struct xdr_stream *xdr, } else args.fhp = fhp; - args.acl = NULL; if (attrmask[0] & FATTR4_WORD0_ACL) { err = nfsd4_get_nfs4_acl(rqstp, dentry, &args.acl); if (err == -EOPNOTSUPP) -- cgit v1.2.3-59-g8ed1b From 91cdcd8d624bfdf05e8db9d572759516f3c786b8 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Tue, 12 Mar 2024 11:34:11 -0400 Subject: mm: zswap: optimize zswap pool size tracking Profiling the munmap() of a zswapped memory region shows 60% of the total cycles currently going into updating the zswap_pool_total_size. There are three consumers of this counter: - store, to enforce the globally configured pool limit - meminfo & debugfs, to report the size to the user - shrink, to determine the batch size for each cycle Instead of aggregating everytime an entry enters or exits the zswap pool, aggregate the value from the zpools on-demand: - Stores aggregate the counter anyway upon success. Aggregating to check the limit instead is the same amount of work. - Meminfo & debugfs might benefit somewhat from a pre-aggregated counter, but aren't exactly hotpaths. - Shrinking can aggregate once for every cycle instead of doing it for every freed entry. As the shrinker might work on tens or hundreds of objects per scan cycle, this is a large reduction in aggregations. The paths that benefit dramatically are swapin, swapoff, and unmaps. There could be millions of pages being processed until somebody asks for the pool size again. This eliminates the pool size updates from those paths entirely. Top profile entries for a 24G range munmap(), before: 38.54% zswap-unmap [kernel.kallsyms] [k] zs_zpool_total_size 12.51% zswap-unmap [kernel.kallsyms] [k] zpool_get_total_size 9.10% zswap-unmap [kernel.kallsyms] [k] zswap_update_total_size 2.95% zswap-unmap [kernel.kallsyms] [k] obj_cgroup_uncharge_zswap 2.88% zswap-unmap [kernel.kallsyms] [k] __slab_free 2.86% zswap-unmap [kernel.kallsyms] [k] xas_store and after: 7.70% zswap-unmap [kernel.kallsyms] [k] __slab_free 7.16% zswap-unmap [kernel.kallsyms] [k] obj_cgroup_uncharge_zswap 6.74% zswap-unmap [kernel.kallsyms] [k] xas_store It was also briefly considered to move to a single atomic in zswap that is updated by the backends, since zswap only cares about the sum of all pools anyway. However, zram directly needs per-pool information out of zsmalloc. To keep the backend from having to update two atomics every time, I opted for the lazy aggregation instead for now. Link: https://lkml.kernel.org/r/20240312153901.3441-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Acked-by: Yosry Ahmed Reviewed-by: Chengming Zhou Reviewed-by: Nhat Pham Signed-off-by: Andrew Morton --- fs/proc/meminfo.c | 3 +- include/linux/zswap.h | 2 +- mm/zswap.c | 101 +++++++++++++++++++++++++------------------------- 3 files changed, 52 insertions(+), 54 deletions(-) (limited to 'fs') diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c index 45af9a989d40..245171d9164b 100644 --- a/fs/proc/meminfo.c +++ b/fs/proc/meminfo.c @@ -89,8 +89,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v) show_val_kb(m, "SwapTotal: ", i.totalswap); show_val_kb(m, "SwapFree: ", i.freeswap); #ifdef CONFIG_ZSWAP - seq_printf(m, "Zswap: %8lu kB\n", - (unsigned long)(zswap_pool_total_size >> 10)); + show_val_kb(m, "Zswap: ", zswap_total_pages()); seq_printf(m, "Zswapped: %8lu kB\n", (unsigned long)atomic_read(&zswap_stored_pages) << (PAGE_SHIFT - 10)); diff --git a/include/linux/zswap.h b/include/linux/zswap.h index 341aea490070..2a85b941db97 100644 --- a/include/linux/zswap.h +++ b/include/linux/zswap.h @@ -7,7 +7,6 @@ struct lruvec; -extern u64 zswap_pool_total_size; extern atomic_t zswap_stored_pages; #ifdef CONFIG_ZSWAP @@ -27,6 +26,7 @@ struct zswap_lruvec_state { atomic_long_t nr_zswap_protected; }; +unsigned long zswap_total_pages(void); bool zswap_store(struct folio *folio); bool zswap_load(struct folio *folio); void zswap_invalidate(swp_entry_t swp); diff --git a/mm/zswap.c b/mm/zswap.c index 2817e2791c13..a56b76852664 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -43,8 +43,6 @@ /********************************* * statistics **********************************/ -/* Total bytes used by the compressed storage */ -u64 zswap_pool_total_size; /* The number of compressed pages currently stored in zswap */ atomic_t zswap_stored_pages = ATOMIC_INIT(0); /* The number of same-value filled pages currently stored in zswap */ @@ -265,45 +263,6 @@ static inline struct zswap_tree *swap_zswap_tree(swp_entry_t swp) pr_debug("%s pool %s/%s\n", msg, (p)->tfm_name, \ zpool_get_type((p)->zpools[0])) -static bool zswap_is_full(void) -{ - return totalram_pages() * zswap_max_pool_percent / 100 < - DIV_ROUND_UP(zswap_pool_total_size, PAGE_SIZE); -} - -static bool zswap_can_accept(void) -{ - return totalram_pages() * zswap_accept_thr_percent / 100 * - zswap_max_pool_percent / 100 > - DIV_ROUND_UP(zswap_pool_total_size, PAGE_SIZE); -} - -static u64 get_zswap_pool_size(struct zswap_pool *pool) -{ - u64 pool_size = 0; - int i; - - for (i = 0; i < ZSWAP_NR_ZPOOLS; i++) - pool_size += zpool_get_total_size(pool->zpools[i]); - - return pool_size; -} - -static void zswap_update_total_size(void) -{ - struct zswap_pool *pool; - u64 total = 0; - - rcu_read_lock(); - - list_for_each_entry_rcu(pool, &zswap_pools, list) - total += get_zswap_pool_size(pool); - - rcu_read_unlock(); - - zswap_pool_total_size = total; -} - /********************************* * pool functions **********************************/ @@ -541,6 +500,33 @@ static struct zswap_pool *zswap_pool_find_get(char *type, char *compressor) return NULL; } +static unsigned long zswap_max_pages(void) +{ + return totalram_pages() * zswap_max_pool_percent / 100; +} + +static unsigned long zswap_accept_thr_pages(void) +{ + return zswap_max_pages() * zswap_accept_thr_percent / 100; +} + +unsigned long zswap_total_pages(void) +{ + struct zswap_pool *pool; + u64 total = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(pool, &zswap_pools, list) { + int i; + + for (i = 0; i < ZSWAP_NR_ZPOOLS; i++) + total += zpool_get_total_size(pool->zpools[i]); + } + rcu_read_unlock(); + + return total >> PAGE_SHIFT; +} + /********************************* * param callbacks **********************************/ @@ -913,7 +899,6 @@ static void zswap_entry_free(struct zswap_entry *entry) } zswap_entry_cache_free(entry); atomic_dec(&zswap_stored_pages); - zswap_update_total_size(); } /* @@ -1344,7 +1329,7 @@ static unsigned long zswap_shrinker_count(struct shrinker *shrinker, nr_backing = memcg_page_state(memcg, MEMCG_ZSWAP_B) >> PAGE_SHIFT; nr_stored = memcg_page_state(memcg, MEMCG_ZSWAPPED); } else { - nr_backing = zswap_pool_total_size >> PAGE_SHIFT; + nr_backing = zswap_total_pages(); nr_stored = atomic_read(&zswap_nr_stored); } @@ -1412,6 +1397,10 @@ static void shrink_worker(struct work_struct *w) { struct mem_cgroup *memcg; int ret, failures = 0; + unsigned long thr; + + /* Reclaim down to the accept threshold */ + thr = zswap_accept_thr_pages(); /* global reclaim will select cgroup in a round-robin fashion. */ do { @@ -1459,10 +1448,9 @@ static void shrink_worker(struct work_struct *w) break; if (ret && ++failures == MAX_RECLAIM_RETRIES) break; - resched: cond_resched(); - } while (!zswap_can_accept()); + } while (zswap_total_pages() > thr); } static int zswap_is_page_same_filled(void *ptr, unsigned long *value) @@ -1503,6 +1491,7 @@ bool zswap_store(struct folio *folio) struct zswap_entry *entry, *dupentry; struct obj_cgroup *objcg = NULL; struct mem_cgroup *memcg = NULL; + unsigned long max_pages, cur_pages; VM_WARN_ON_ONCE(!folio_test_locked(folio)); VM_WARN_ON_ONCE(!folio_test_swapcache(folio)); @@ -1514,6 +1503,7 @@ bool zswap_store(struct folio *folio) if (!zswap_enabled) goto check_old; + /* Check cgroup limits */ objcg = get_obj_cgroup_from_folio(folio); if (objcg && !obj_cgroup_may_zswap(objcg)) { memcg = get_mem_cgroup_from_objcg(objcg); @@ -1524,15 +1514,18 @@ bool zswap_store(struct folio *folio) mem_cgroup_put(memcg); } - /* reclaim space if needed */ - if (zswap_is_full()) { + /* Check global limits */ + cur_pages = zswap_total_pages(); + max_pages = zswap_max_pages(); + + if (cur_pages >= max_pages) { zswap_pool_limit_hit++; zswap_pool_reached_full = true; goto shrink; } if (zswap_pool_reached_full) { - if (!zswap_can_accept()) + if (cur_pages > zswap_accept_thr_pages()) goto shrink; else zswap_pool_reached_full = false; @@ -1608,7 +1601,6 @@ insert_entry: /* update stats */ atomic_inc(&zswap_stored_pages); - zswap_update_total_size(); count_vm_event(ZSWPOUT); return true; @@ -1752,6 +1744,13 @@ void zswap_swapoff(int type) static struct dentry *zswap_debugfs_root; +static int debugfs_get_total_size(void *data, u64 *val) +{ + *val = zswap_total_pages() * PAGE_SIZE; + return 0; +} +DEFINE_DEBUGFS_ATTRIBUTE(total_size_fops, debugfs_get_total_size, NULL, "%llu\n"); + static int zswap_debugfs_init(void) { if (!debugfs_initialized()) @@ -1773,8 +1772,8 @@ static int zswap_debugfs_init(void) zswap_debugfs_root, &zswap_reject_compress_poor); debugfs_create_u64("written_back_pages", 0444, zswap_debugfs_root, &zswap_written_back_pages); - debugfs_create_u64("pool_total_size", 0444, - zswap_debugfs_root, &zswap_pool_total_size); + debugfs_create_file("pool_total_size", 0444, + zswap_debugfs_root, NULL, &total_size_fops); debugfs_create_atomic_t("stored_pages", 0444, zswap_debugfs_root, &zswap_stored_pages); debugfs_create_atomic_t("same_filled_pages", 0444, -- cgit v1.2.3-59-g8ed1b From 2c321f3f70bc284510598f712b702ce8d60c4d14 Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Sun, 14 Apr 2024 19:07:31 -0700 Subject: mm: change inlined allocation helpers to account at the call site Main goal of memory allocation profiling patchset is to provide accounting that is cheap enough to run in production. To achieve that we inject counters using codetags at the allocation call sites to account every time allocation is made. This injection allows us to perform accounting efficiently because injected counters are immediately available as opposed to the alternative methods, such as using _RET_IP_, which would require counter lookup and appropriate locking that makes accounting much more expensive. This method requires all allocation functions to inject separate counters at their call sites so that their callers can be individually accounted. Counter injection is implemented by allocation hooks which should wrap all allocation functions. Inlined functions which perform allocations but do not use allocation hooks are directly charged for the allocations they perform. In most cases these functions are just specialized allocation wrappers used from multiple places to allocate objects of a specific type. It would be more useful to do the accounting at their call sites instead. Instrument these helpers to do accounting at the call site. Simple inlined allocation wrappers are converted directly into macros. More complex allocators or allocators with documentation are converted into _noprof versions and allocation hooks are added. This allows memory allocation profiling mechanism to charge allocations to the callers of these functions. Link: https://lkml.kernel.org/r/20240415020731.1152108-1-surenb@google.com Signed-off-by: Suren Baghdasaryan Acked-by: Jan Kara [jbd2] Cc: Anna Schumaker Cc: Arnd Bergmann Cc: Benjamin Tissoires Cc: Christoph Lameter Cc: David Rientjes Cc: David S. Miller Cc: Dennis Zhou Cc: Eric Dumazet Cc: Herbert Xu Cc: Jakub Kicinski Cc: Jakub Sitnicki Cc: Jiri Kosina Cc: Joerg Roedel Cc: Joonsoo Kim Cc: Kent Overstreet Cc: Matthew Wilcox (Oracle) Cc: Paolo Abeni Cc: Pekka Enberg Cc: Tejun Heo Cc: Theodore Ts'o Cc: Trond Myklebust Cc: Vlastimil Babka Cc: Will Deacon Signed-off-by: Andrew Morton --- drivers/iommu/amd/amd_iommu.h | 5 +++-- fs/nfs/iostat.h | 5 +---- include/acpi/platform/aclinuxex.h | 19 ++++++------------ include/asm-generic/pgalloc.h | 35 ++++++++++++++++++++------------- include/crypto/hash.h | 7 ++++--- include/crypto/internal/acompress.h | 5 +++-- include/crypto/skcipher.h | 7 ++++--- include/linux/bpf.h | 33 ++++++++----------------------- include/linux/bpfptr.h | 5 +++-- include/linux/dma-fence-chain.h | 6 ++---- include/linux/hid_bpf.h | 6 ++---- include/linux/jbd2.h | 12 ++++------- include/linux/mm.h | 5 +++-- include/linux/mm_types.h | 5 +++-- include/linux/percpu.h | 3 +++ include/linux/ptr_ring.h | 28 +++++++++++++++----------- include/linux/skb_array.h | 19 ++++++++++-------- include/linux/skbuff.h | 20 ++++++++----------- include/linux/skmsg.h | 8 +++----- include/linux/slab.h | 5 +++++ include/linux/sockptr.h | 10 ++++++---- include/net/netlabel.h | 16 +++++++++------ include/net/netlink.h | 5 +++-- include/net/request_sock.h | 5 +++-- include/net/tcx.h | 5 +++-- net/sunrpc/auth_gss/auth_gss_internal.h | 6 ++++-- 26 files changed, 142 insertions(+), 143 deletions(-) (limited to 'fs') diff --git a/drivers/iommu/amd/amd_iommu.h b/drivers/iommu/amd/amd_iommu.h index f482aab420f7..52575ba9a141 100644 --- a/drivers/iommu/amd/amd_iommu.h +++ b/drivers/iommu/amd/amd_iommu.h @@ -134,13 +134,14 @@ static inline int get_pci_sbdf_id(struct pci_dev *pdev) return PCI_SEG_DEVID_TO_SBDF(seg, devid); } -static inline void *alloc_pgtable_page(int nid, gfp_t gfp) +static inline void *alloc_pgtable_page_noprof(int nid, gfp_t gfp) { struct page *page; - page = alloc_pages_node(nid, gfp | __GFP_ZERO, 0); + page = alloc_pages_node_noprof(nid, gfp | __GFP_ZERO, 0); return page ? page_address(page) : NULL; } +#define alloc_pgtable_page(...) alloc_hooks(alloc_pgtable_page_noprof(__VA_ARGS__)) /* * This must be called after device probe completes. During probe diff --git a/fs/nfs/iostat.h b/fs/nfs/iostat.h index 5aa776b5a3e7..b17a9eb9b148 100644 --- a/fs/nfs/iostat.h +++ b/fs/nfs/iostat.h @@ -46,10 +46,7 @@ static inline void nfs_add_stats(const struct inode *inode, nfs_add_server_stats(NFS_SERVER(inode), stat, addend); } -static inline struct nfs_iostats __percpu *nfs_alloc_iostats(void) -{ - return alloc_percpu(struct nfs_iostats); -} +#define nfs_alloc_iostats() alloc_percpu(struct nfs_iostats) static inline void nfs_free_iostats(struct nfs_iostats __percpu *stats) { diff --git a/include/acpi/platform/aclinuxex.h b/include/acpi/platform/aclinuxex.h index 600d4e2641da..62cac266a1c8 100644 --- a/include/acpi/platform/aclinuxex.h +++ b/include/acpi/platform/aclinuxex.h @@ -47,26 +47,19 @@ acpi_status acpi_os_terminate(void); * However, boot has (system_state != SYSTEM_RUNNING) * to quiet __might_sleep() in kmalloc() and resume does not. */ -static inline void *acpi_os_allocate(acpi_size size) -{ - return kmalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); -} +#define acpi_os_allocate(_size) \ + kmalloc(_size, irqs_disabled() ? GFP_ATOMIC : GFP_KERNEL) -static inline void *acpi_os_allocate_zeroed(acpi_size size) -{ - return kzalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); -} +#define acpi_os_allocate_zeroed(_size) \ + kzalloc(_size, irqs_disabled() ? GFP_ATOMIC : GFP_KERNEL) static inline void acpi_os_free(void *memory) { kfree(memory); } -static inline void *acpi_os_acquire_object(acpi_cache_t * cache) -{ - return kmem_cache_zalloc(cache, - irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); -} +#define acpi_os_acquire_object(_cache) \ + kmem_cache_zalloc(_cache, irqs_disabled() ? GFP_ATOMIC : GFP_KERNEL) static inline acpi_thread_id acpi_os_get_thread_id(void) { diff --git a/include/asm-generic/pgalloc.h b/include/asm-generic/pgalloc.h index 879e5f8aa5e9..7c48f5fbf8aa 100644 --- a/include/asm-generic/pgalloc.h +++ b/include/asm-generic/pgalloc.h @@ -16,15 +16,16 @@ * * Return: pointer to the allocated memory or %NULL on error */ -static inline pte_t *__pte_alloc_one_kernel(struct mm_struct *mm) +static inline pte_t *__pte_alloc_one_kernel_noprof(struct mm_struct *mm) { - struct ptdesc *ptdesc = pagetable_alloc(GFP_PGTABLE_KERNEL & + struct ptdesc *ptdesc = pagetable_alloc_noprof(GFP_PGTABLE_KERNEL & ~__GFP_HIGHMEM, 0); if (!ptdesc) return NULL; return ptdesc_address(ptdesc); } +#define __pte_alloc_one_kernel(...) alloc_hooks(__pte_alloc_one_kernel_noprof(__VA_ARGS__)) #ifndef __HAVE_ARCH_PTE_ALLOC_ONE_KERNEL /** @@ -33,10 +34,11 @@ static inline pte_t *__pte_alloc_one_kernel(struct mm_struct *mm) * * Return: pointer to the allocated memory or %NULL on error */ -static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm) +static inline pte_t *pte_alloc_one_kernel_noprof(struct mm_struct *mm) { - return __pte_alloc_one_kernel(mm); + return __pte_alloc_one_kernel_noprof(mm); } +#define pte_alloc_one_kernel(...) alloc_hooks(pte_alloc_one_kernel_noprof(__VA_ARGS__)) #endif /** @@ -61,11 +63,11 @@ static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) * * Return: `struct page` referencing the ptdesc or %NULL on error */ -static inline pgtable_t __pte_alloc_one(struct mm_struct *mm, gfp_t gfp) +static inline pgtable_t __pte_alloc_one_noprof(struct mm_struct *mm, gfp_t gfp) { struct ptdesc *ptdesc; - ptdesc = pagetable_alloc(gfp, 0); + ptdesc = pagetable_alloc_noprof(gfp, 0); if (!ptdesc) return NULL; if (!pagetable_pte_ctor(ptdesc)) { @@ -75,6 +77,7 @@ static inline pgtable_t __pte_alloc_one(struct mm_struct *mm, gfp_t gfp) return ptdesc_page(ptdesc); } +#define __pte_alloc_one(...) alloc_hooks(__pte_alloc_one_noprof(__VA_ARGS__)) #ifndef __HAVE_ARCH_PTE_ALLOC_ONE /** @@ -85,10 +88,11 @@ static inline pgtable_t __pte_alloc_one(struct mm_struct *mm, gfp_t gfp) * * Return: `struct page` referencing the ptdesc or %NULL on error */ -static inline pgtable_t pte_alloc_one(struct mm_struct *mm) +static inline pgtable_t pte_alloc_one_noprof(struct mm_struct *mm) { - return __pte_alloc_one(mm, GFP_PGTABLE_USER); + return __pte_alloc_one_noprof(mm, GFP_PGTABLE_USER); } +#define pte_alloc_one(...) alloc_hooks(pte_alloc_one_noprof(__VA_ARGS__)) #endif /* @@ -124,14 +128,14 @@ static inline void pte_free(struct mm_struct *mm, struct page *pte_page) * * Return: pointer to the allocated memory or %NULL on error */ -static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) +static inline pmd_t *pmd_alloc_one_noprof(struct mm_struct *mm, unsigned long addr) { struct ptdesc *ptdesc; gfp_t gfp = GFP_PGTABLE_USER; if (mm == &init_mm) gfp = GFP_PGTABLE_KERNEL; - ptdesc = pagetable_alloc(gfp, 0); + ptdesc = pagetable_alloc_noprof(gfp, 0); if (!ptdesc) return NULL; if (!pagetable_pmd_ctor(ptdesc)) { @@ -140,6 +144,7 @@ static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) } return ptdesc_address(ptdesc); } +#define pmd_alloc_one(...) alloc_hooks(pmd_alloc_one_noprof(__VA_ARGS__)) #endif #ifndef __HAVE_ARCH_PMD_FREE @@ -157,7 +162,7 @@ static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) #if CONFIG_PGTABLE_LEVELS > 3 -static inline pud_t *__pud_alloc_one(struct mm_struct *mm, unsigned long addr) +static inline pud_t *__pud_alloc_one_noprof(struct mm_struct *mm, unsigned long addr) { gfp_t gfp = GFP_PGTABLE_USER; struct ptdesc *ptdesc; @@ -166,13 +171,14 @@ static inline pud_t *__pud_alloc_one(struct mm_struct *mm, unsigned long addr) gfp = GFP_PGTABLE_KERNEL; gfp &= ~__GFP_HIGHMEM; - ptdesc = pagetable_alloc(gfp, 0); + ptdesc = pagetable_alloc_noprof(gfp, 0); if (!ptdesc) return NULL; pagetable_pud_ctor(ptdesc); return ptdesc_address(ptdesc); } +#define __pud_alloc_one(...) alloc_hooks(__pud_alloc_one_noprof(__VA_ARGS__)) #ifndef __HAVE_ARCH_PUD_ALLOC_ONE /** @@ -184,10 +190,11 @@ static inline pud_t *__pud_alloc_one(struct mm_struct *mm, unsigned long addr) * * Return: pointer to the allocated memory or %NULL on error */ -static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long addr) +static inline pud_t *pud_alloc_one_noprof(struct mm_struct *mm, unsigned long addr) { - return __pud_alloc_one(mm, addr); + return __pud_alloc_one_noprof(mm, addr); } +#define pud_alloc_one(...) alloc_hooks(pud_alloc_one_noprof(__VA_ARGS__)) #endif static inline void __pud_free(struct mm_struct *mm, pud_t *pud) diff --git a/include/crypto/hash.h b/include/crypto/hash.h index 5d61f576cfc8..e5181cc9b7c5 100644 --- a/include/crypto/hash.h +++ b/include/crypto/hash.h @@ -578,19 +578,20 @@ static inline void ahash_request_set_tfm(struct ahash_request *req, * * Return: allocated request handle in case of success, or NULL if out of memory */ -static inline struct ahash_request *ahash_request_alloc( +static inline struct ahash_request *ahash_request_alloc_noprof( struct crypto_ahash *tfm, gfp_t gfp) { struct ahash_request *req; - req = kmalloc(sizeof(struct ahash_request) + - crypto_ahash_reqsize(tfm), gfp); + req = kmalloc_noprof(sizeof(struct ahash_request) + + crypto_ahash_reqsize(tfm), gfp); if (likely(req)) ahash_request_set_tfm(req, tfm); return req; } +#define ahash_request_alloc(...) alloc_hooks(ahash_request_alloc_noprof(__VA_ARGS__)) /** * ahash_request_free() - zeroize and free the request data structure diff --git a/include/crypto/internal/acompress.h b/include/crypto/internal/acompress.h index 4ac46bafba9d..2a67793f52ad 100644 --- a/include/crypto/internal/acompress.h +++ b/include/crypto/internal/acompress.h @@ -69,15 +69,16 @@ static inline void acomp_request_complete(struct acomp_req *req, crypto_request_complete(&req->base, err); } -static inline struct acomp_req *__acomp_request_alloc(struct crypto_acomp *tfm) +static inline struct acomp_req *__acomp_request_alloc_noprof(struct crypto_acomp *tfm) { struct acomp_req *req; - req = kzalloc(sizeof(*req) + crypto_acomp_reqsize(tfm), GFP_KERNEL); + req = kzalloc_noprof(sizeof(*req) + crypto_acomp_reqsize(tfm), GFP_KERNEL); if (likely(req)) acomp_request_set_tfm(req, tfm); return req; } +#define __acomp_request_alloc(...) alloc_hooks(__acomp_request_alloc_noprof(__VA_ARGS__)) static inline void __acomp_request_free(struct acomp_req *req) { diff --git a/include/crypto/skcipher.h b/include/crypto/skcipher.h index c8857d7bdb37..6c5330e316b0 100644 --- a/include/crypto/skcipher.h +++ b/include/crypto/skcipher.h @@ -861,19 +861,20 @@ static inline struct skcipher_request *skcipher_request_cast( * * Return: allocated request handle in case of success, or NULL if out of memory */ -static inline struct skcipher_request *skcipher_request_alloc( +static inline struct skcipher_request *skcipher_request_alloc_noprof( struct crypto_skcipher *tfm, gfp_t gfp) { struct skcipher_request *req; - req = kmalloc(sizeof(struct skcipher_request) + - crypto_skcipher_reqsize(tfm), gfp); + req = kmalloc_noprof(sizeof(struct skcipher_request) + + crypto_skcipher_reqsize(tfm), gfp); if (likely(req)) skcipher_request_set_tfm(req, tfm); return req; } +#define skcipher_request_alloc(...) alloc_hooks(skcipher_request_alloc_noprof(__VA_ARGS__)) /** * skcipher_request_free() - zeroize and free request data structure diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 890e152d553e..a86da1f38b3c 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -2244,31 +2244,14 @@ void *bpf_map_kvcalloc(struct bpf_map *map, size_t n, size_t size, void __percpu *bpf_map_alloc_percpu(const struct bpf_map *map, size_t size, size_t align, gfp_t flags); #else -static inline void * -bpf_map_kmalloc_node(const struct bpf_map *map, size_t size, gfp_t flags, - int node) -{ - return kmalloc_node(size, flags, node); -} - -static inline void * -bpf_map_kzalloc(const struct bpf_map *map, size_t size, gfp_t flags) -{ - return kzalloc(size, flags); -} - -static inline void * -bpf_map_kvcalloc(struct bpf_map *map, size_t n, size_t size, gfp_t flags) -{ - return kvcalloc(n, size, flags); -} - -static inline void __percpu * -bpf_map_alloc_percpu(const struct bpf_map *map, size_t size, size_t align, - gfp_t flags) -{ - return __alloc_percpu_gfp(size, align, flags); -} +#define bpf_map_kmalloc_node(_map, _size, _flags, _node) \ + kmalloc_node(_size, _flags, _node) +#define bpf_map_kzalloc(_map, _size, _flags) \ + kzalloc(_size, _flags) +#define bpf_map_kvcalloc(_map, _n, _size, _flags) \ + kvcalloc(_n, _size, _flags) +#define bpf_map_alloc_percpu(_map, _size, _align, _flags) \ + __alloc_percpu_gfp(_size, _align, _flags) #endif static inline int diff --git a/include/linux/bpfptr.h b/include/linux/bpfptr.h index 79b2f78eec1a..1af241525a17 100644 --- a/include/linux/bpfptr.h +++ b/include/linux/bpfptr.h @@ -65,9 +65,9 @@ static inline int copy_to_bpfptr_offset(bpfptr_t dst, size_t offset, return copy_to_sockptr_offset((sockptr_t) dst, offset, src, size); } -static inline void *kvmemdup_bpfptr(bpfptr_t src, size_t len) +static inline void *kvmemdup_bpfptr_noprof(bpfptr_t src, size_t len) { - void *p = kvmalloc(len, GFP_USER | __GFP_NOWARN); + void *p = kvmalloc_noprof(len, GFP_USER | __GFP_NOWARN); if (!p) return ERR_PTR(-ENOMEM); @@ -77,6 +77,7 @@ static inline void *kvmemdup_bpfptr(bpfptr_t src, size_t len) } return p; } +#define kvmemdup_bpfptr(...) alloc_hooks(kvmemdup_bpfptr_noprof(__VA_ARGS__)) static inline long strncpy_from_bpfptr(char *dst, bpfptr_t src, size_t count) { diff --git a/include/linux/dma-fence-chain.h b/include/linux/dma-fence-chain.h index 4bdf0b96da28..ad9e2506c2f4 100644 --- a/include/linux/dma-fence-chain.h +++ b/include/linux/dma-fence-chain.h @@ -86,10 +86,8 @@ dma_fence_chain_contained(struct dma_fence *fence) * * Returns a new struct dma_fence_chain object or NULL on failure. */ -static inline struct dma_fence_chain *dma_fence_chain_alloc(void) -{ - return kmalloc(sizeof(struct dma_fence_chain), GFP_KERNEL); -}; +#define dma_fence_chain_alloc() \ + ((struct dma_fence_chain *)kmalloc(sizeof(struct dma_fence_chain), GFP_KERNEL)) /** * dma_fence_chain_free diff --git a/include/linux/hid_bpf.h b/include/linux/hid_bpf.h index 7118ac28d468..ca70eeb6393e 100644 --- a/include/linux/hid_bpf.h +++ b/include/linux/hid_bpf.h @@ -149,10 +149,8 @@ static inline int hid_bpf_connect_device(struct hid_device *hdev) { return 0; } static inline void hid_bpf_disconnect_device(struct hid_device *hdev) {} static inline void hid_bpf_destroy_device(struct hid_device *hid) {} static inline void hid_bpf_device_init(struct hid_device *hid) {} -static inline u8 *call_hid_bpf_rdesc_fixup(struct hid_device *hdev, u8 *rdesc, unsigned int *size) -{ - return kmemdup(rdesc, *size, GFP_KERNEL); -} +#define call_hid_bpf_rdesc_fixup(_hdev, _rdesc, _size) \ + ((u8 *)kmemdup(_rdesc, *(_size), GFP_KERNEL)) #endif /* CONFIG_HID_BPF */ diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 971f3e826e15..9512fe332668 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1586,10 +1586,8 @@ void jbd2_journal_put_journal_head(struct journal_head *jh); */ extern struct kmem_cache *jbd2_handle_cache; -static inline handle_t *jbd2_alloc_handle(gfp_t gfp_flags) -{ - return kmem_cache_zalloc(jbd2_handle_cache, gfp_flags); -} +#define jbd2_alloc_handle(_gfp_flags) \ + ((handle_t *)kmem_cache_zalloc(jbd2_handle_cache, _gfp_flags)) static inline void jbd2_free_handle(handle_t *handle) { @@ -1602,10 +1600,8 @@ static inline void jbd2_free_handle(handle_t *handle) */ extern struct kmem_cache *jbd2_inode_cache; -static inline struct jbd2_inode *jbd2_alloc_inode(gfp_t gfp_flags) -{ - return kmem_cache_alloc(jbd2_inode_cache, gfp_flags); -} +#define jbd2_alloc_inode(_gfp_flags) \ + ((struct jbd2_inode *)kmem_cache_alloc(jbd2_inode_cache, _gfp_flags)) static inline void jbd2_free_inode(struct jbd2_inode *jinode) { diff --git a/include/linux/mm.h b/include/linux/mm.h index 0dbd737cfabd..60eabc6c8e00 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2860,12 +2860,13 @@ static inline bool pagetable_is_reserved(struct ptdesc *pt) * * Return: The ptdesc describing the allocated page tables. */ -static inline struct ptdesc *pagetable_alloc(gfp_t gfp, unsigned int order) +static inline struct ptdesc *pagetable_alloc_noprof(gfp_t gfp, unsigned int order) { - struct page *page = alloc_pages(gfp | __GFP_COMP, order); + struct page *page = alloc_pages_noprof(gfp | __GFP_COMP, order); return page_ptdesc(page); } +#define pagetable_alloc(...) alloc_hooks(pagetable_alloc_noprof(__VA_ARGS__)) /** * pagetable_free - Free pagetables diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 4ae4684d1add..6fef56938a17 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -1170,14 +1170,15 @@ static inline void mm_init_cid(struct mm_struct *mm) cpumask_clear(mm_cidmask(mm)); } -static inline int mm_alloc_cid(struct mm_struct *mm) +static inline int mm_alloc_cid_noprof(struct mm_struct *mm) { - mm->pcpu_cid = alloc_percpu(struct mm_cid); + mm->pcpu_cid = alloc_percpu_noprof(struct mm_cid); if (!mm->pcpu_cid) return -ENOMEM; mm_init_cid(mm); return 0; } +#define mm_alloc_cid(...) alloc_hooks(mm_alloc_cid_noprof(__VA_ARGS__)) static inline void mm_destroy_cid(struct mm_struct *mm) { diff --git a/include/linux/percpu.h b/include/linux/percpu.h index 13a82f11e4fd..03053de557cf 100644 --- a/include/linux/percpu.h +++ b/include/linux/percpu.h @@ -151,6 +151,9 @@ extern size_t pcpu_alloc_size(void __percpu *__pdata); #define alloc_percpu(type) \ (typeof(type) __percpu *)__alloc_percpu(sizeof(type), \ __alignof__(type)) +#define alloc_percpu_noprof(type) \ + ((typeof(type) __percpu *)pcpu_alloc_noprof(sizeof(type), \ + __alignof__(type), false, GFP_KERNEL)) extern void free_percpu(void __percpu *__pdata); diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index 808f9d3ee546..fd037c127bb0 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -464,11 +464,11 @@ static inline int ptr_ring_consume_batched_bh(struct ptr_ring *r, /* Not all gfp_t flags (besides GFP_KERNEL) are allowed. See * documentation for vmalloc for which of them are legal. */ -static inline void **__ptr_ring_init_queue_alloc(unsigned int size, gfp_t gfp) +static inline void **__ptr_ring_init_queue_alloc_noprof(unsigned int size, gfp_t gfp) { if (size > KMALLOC_MAX_SIZE / sizeof(void *)) return NULL; - return kvmalloc_array(size, sizeof(void *), gfp | __GFP_ZERO); + return kvmalloc_array_noprof(size, sizeof(void *), gfp | __GFP_ZERO); } static inline void __ptr_ring_set_size(struct ptr_ring *r, int size) @@ -484,9 +484,9 @@ static inline void __ptr_ring_set_size(struct ptr_ring *r, int size) r->batch = 1; } -static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) +static inline int ptr_ring_init_noprof(struct ptr_ring *r, int size, gfp_t gfp) { - r->queue = __ptr_ring_init_queue_alloc(size, gfp); + r->queue = __ptr_ring_init_queue_alloc_noprof(size, gfp); if (!r->queue) return -ENOMEM; @@ -497,6 +497,7 @@ static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) return 0; } +#define ptr_ring_init(...) alloc_hooks(ptr_ring_init_noprof(__VA_ARGS__)) /* * Return entries into ring. Destroy entries that don't fit. @@ -587,11 +588,11 @@ static inline void **__ptr_ring_swap_queue(struct ptr_ring *r, void **queue, * In particular if you consume ring in interrupt or BH context, you must * disable interrupts/BH when doing so. */ -static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, +static inline int ptr_ring_resize_noprof(struct ptr_ring *r, int size, gfp_t gfp, void (*destroy)(void *)) { unsigned long flags; - void **queue = __ptr_ring_init_queue_alloc(size, gfp); + void **queue = __ptr_ring_init_queue_alloc_noprof(size, gfp); void **old; if (!queue) @@ -609,6 +610,7 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, return 0; } +#define ptr_ring_resize(...) alloc_hooks(ptr_ring_resize_noprof(__VA_ARGS__)) /* * Note: producer lock is nested within consumer lock, so if you @@ -616,21 +618,21 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, * In particular if you consume ring in interrupt or BH context, you must * disable interrupts/BH when doing so. */ -static inline int ptr_ring_resize_multiple(struct ptr_ring **rings, - unsigned int nrings, - int size, - gfp_t gfp, void (*destroy)(void *)) +static inline int ptr_ring_resize_multiple_noprof(struct ptr_ring **rings, + unsigned int nrings, + int size, + gfp_t gfp, void (*destroy)(void *)) { unsigned long flags; void ***queues; int i; - queues = kmalloc_array(nrings, sizeof(*queues), gfp); + queues = kmalloc_array_noprof(nrings, sizeof(*queues), gfp); if (!queues) goto noqueues; for (i = 0; i < nrings; ++i) { - queues[i] = __ptr_ring_init_queue_alloc(size, gfp); + queues[i] = __ptr_ring_init_queue_alloc_noprof(size, gfp); if (!queues[i]) goto nomem; } @@ -660,6 +662,8 @@ nomem: noqueues: return -ENOMEM; } +#define ptr_ring_resize_multiple(...) \ + alloc_hooks(ptr_ring_resize_multiple_noprof(__VA_ARGS__)) static inline void ptr_ring_cleanup(struct ptr_ring *r, void (*destroy)(void *)) { diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index e2d45b7cb619..926496c9cc9c 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -177,10 +177,11 @@ static inline int skb_array_peek_len_any(struct skb_array *a) return PTR_RING_PEEK_CALL_ANY(&a->ring, __skb_array_len_with_tag); } -static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) +static inline int skb_array_init_noprof(struct skb_array *a, int size, gfp_t gfp) { - return ptr_ring_init(&a->ring, size, gfp); + return ptr_ring_init_noprof(&a->ring, size, gfp); } +#define skb_array_init(...) alloc_hooks(skb_array_init_noprof(__VA_ARGS__)) static void __skb_array_destroy_skb(void *ptr) { @@ -198,15 +199,17 @@ static inline int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); } -static inline int skb_array_resize_multiple(struct skb_array **rings, - int nrings, unsigned int size, - gfp_t gfp) +static inline int skb_array_resize_multiple_noprof(struct skb_array **rings, + int nrings, unsigned int size, + gfp_t gfp) { BUILD_BUG_ON(offsetof(struct skb_array, ring)); - return ptr_ring_resize_multiple((struct ptr_ring **)rings, - nrings, size, gfp, - __skb_array_destroy_skb); + return ptr_ring_resize_multiple_noprof((struct ptr_ring **)rings, + nrings, size, gfp, + __skb_array_destroy_skb); } +#define skb_array_resize_multiple(...) \ + alloc_hooks(skb_array_resize_multiple_noprof(__VA_ARGS__)) static inline void skb_array_cleanup(struct skb_array *a) { diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9d24aec064e8..b72920c9887b 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3371,7 +3371,7 @@ void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason); * * %NULL is returned if there is no free memory. */ -static inline struct page *__dev_alloc_pages(gfp_t gfp_mask, +static inline struct page *__dev_alloc_pages_noprof(gfp_t gfp_mask, unsigned int order) { /* This piece of code contains several assumptions. @@ -3384,13 +3384,11 @@ static inline struct page *__dev_alloc_pages(gfp_t gfp_mask, */ gfp_mask |= __GFP_COMP | __GFP_MEMALLOC; - return alloc_pages_node(NUMA_NO_NODE, gfp_mask, order); + return alloc_pages_node_noprof(NUMA_NO_NODE, gfp_mask, order); } +#define __dev_alloc_pages(...) alloc_hooks(__dev_alloc_pages_noprof(__VA_ARGS__)) -static inline struct page *dev_alloc_pages(unsigned int order) -{ - return __dev_alloc_pages(GFP_ATOMIC | __GFP_NOWARN, order); -} +#define dev_alloc_pages(_order) __dev_alloc_pages(GFP_ATOMIC | __GFP_NOWARN, _order) /** * __dev_alloc_page - allocate a page for network Rx @@ -3400,15 +3398,13 @@ static inline struct page *dev_alloc_pages(unsigned int order) * * %NULL is returned if there is no free memory. */ -static inline struct page *__dev_alloc_page(gfp_t gfp_mask) +static inline struct page *__dev_alloc_page_noprof(gfp_t gfp_mask) { - return __dev_alloc_pages(gfp_mask, 0); + return __dev_alloc_pages_noprof(gfp_mask, 0); } +#define __dev_alloc_page(...) alloc_hooks(__dev_alloc_page_noprof(__VA_ARGS__)) -static inline struct page *dev_alloc_page(void) -{ - return dev_alloc_pages(0); -} +#define dev_alloc_page() dev_alloc_pages(0) /** * dev_page_is_reusable - check whether a page can be reused for network Rx diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index e65ec3fd2799..78efc5b20284 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -410,11 +410,9 @@ void sk_psock_stop_verdict(struct sock *sk, struct sk_psock *psock); int sk_psock_msg_verdict(struct sock *sk, struct sk_psock *psock, struct sk_msg *msg); -static inline struct sk_psock_link *sk_psock_init_link(void) -{ - return kzalloc(sizeof(struct sk_psock_link), - GFP_ATOMIC | __GFP_NOWARN); -} +#define sk_psock_init_link() \ + ((struct sk_psock_link *)kzalloc(sizeof(struct sk_psock_link), \ + GFP_ATOMIC | __GFP_NOWARN)) static inline void sk_psock_free_link(struct sk_psock_link *link) { diff --git a/include/linux/slab.h b/include/linux/slab.h index c0be1cd03cf6..4cc37ef22aae 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -744,6 +744,9 @@ void *kmalloc_node_track_caller_noprof(size_t size, gfp_t flags, int node, */ #define kmalloc_track_caller(...) kmalloc_node_track_caller(__VA_ARGS__, NUMA_NO_NODE) +#define kmalloc_track_caller_noprof(...) \ + kmalloc_node_track_caller_noprof(__VA_ARGS__, NUMA_NO_NODE, _RET_IP_) + static inline __alloc_size(1, 2) void *kmalloc_array_node_noprof(size_t n, size_t size, gfp_t flags, int node) { @@ -781,6 +784,7 @@ extern void *kvmalloc_node_noprof(size_t size, gfp_t flags, int node) __alloc_si #define kvmalloc_node(...) alloc_hooks(kvmalloc_node_noprof(__VA_ARGS__)) #define kvmalloc(_size, _flags) kvmalloc_node(_size, _flags, NUMA_NO_NODE) +#define kvmalloc_noprof(_size, _flags) kvmalloc_node_noprof(_size, _flags, NUMA_NO_NODE) #define kvzalloc(_size, _flags) kvmalloc(_size, _flags|__GFP_ZERO) #define kvzalloc_node(_size, _flags, _node) kvmalloc_node(_size, _flags|__GFP_ZERO, _node) @@ -797,6 +801,7 @@ static inline __alloc_size(1, 2) void *kvmalloc_array_noprof(size_t n, size_t si #define kvmalloc_array(...) alloc_hooks(kvmalloc_array_noprof(__VA_ARGS__)) #define kvcalloc(_n, _size, _flags) kvmalloc_array(_n, _size, _flags|__GFP_ZERO) +#define kvcalloc_noprof(_n, _size, _flags) kvmalloc_array_noprof(_n, _size, _flags|__GFP_ZERO) extern void *kvrealloc_noprof(const void *p, size_t oldsize, size_t newsize, gfp_t flags) __realloc_size(3); diff --git a/include/linux/sockptr.h b/include/linux/sockptr.h index 317200cd3a60..fc5a206c4043 100644 --- a/include/linux/sockptr.h +++ b/include/linux/sockptr.h @@ -117,9 +117,9 @@ static inline int copy_to_sockptr(sockptr_t dst, const void *src, size_t size) return copy_to_sockptr_offset(dst, 0, src, size); } -static inline void *memdup_sockptr(sockptr_t src, size_t len) +static inline void *memdup_sockptr_noprof(sockptr_t src, size_t len) { - void *p = kmalloc_track_caller(len, GFP_USER | __GFP_NOWARN); + void *p = kmalloc_track_caller_noprof(len, GFP_USER | __GFP_NOWARN); if (!p) return ERR_PTR(-ENOMEM); @@ -129,10 +129,11 @@ static inline void *memdup_sockptr(sockptr_t src, size_t len) } return p; } +#define memdup_sockptr(...) alloc_hooks(memdup_sockptr_noprof(__VA_ARGS__)) -static inline void *memdup_sockptr_nul(sockptr_t src, size_t len) +static inline void *memdup_sockptr_nul_noprof(sockptr_t src, size_t len) { - char *p = kmalloc_track_caller(len + 1, GFP_KERNEL); + char *p = kmalloc_track_caller_noprof(len + 1, GFP_KERNEL); if (!p) return ERR_PTR(-ENOMEM); @@ -143,6 +144,7 @@ static inline void *memdup_sockptr_nul(sockptr_t src, size_t len) p[len] = '\0'; return p; } +#define memdup_sockptr_nul(...) alloc_hooks(memdup_sockptr_nul_noprof(__VA_ARGS__)) static inline long strncpy_from_sockptr(char *dst, sockptr_t src, size_t count) { diff --git a/include/net/netlabel.h b/include/net/netlabel.h index f3ab0b8a4b18..c31bd96dafdb 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -274,15 +274,17 @@ struct netlbl_calipso_ops { * on success, NULL on failure. * */ -static inline struct netlbl_lsm_cache *netlbl_secattr_cache_alloc(gfp_t flags) +static inline struct netlbl_lsm_cache *netlbl_secattr_cache_alloc_noprof(gfp_t flags) { struct netlbl_lsm_cache *cache; - cache = kzalloc(sizeof(*cache), flags); + cache = kzalloc_noprof(sizeof(*cache), flags); if (cache) refcount_set(&cache->refcount, 1); return cache; } +#define netlbl_secattr_cache_alloc(...) \ + alloc_hooks(netlbl_secattr_cache_alloc_noprof(__VA_ARGS__)) /** * netlbl_secattr_cache_free - Frees a netlbl_lsm_cache struct @@ -311,10 +313,11 @@ static inline void netlbl_secattr_cache_free(struct netlbl_lsm_cache *cache) * on failure. * */ -static inline struct netlbl_lsm_catmap *netlbl_catmap_alloc(gfp_t flags) +static inline struct netlbl_lsm_catmap *netlbl_catmap_alloc_noprof(gfp_t flags) { - return kzalloc(sizeof(struct netlbl_lsm_catmap), flags); + return kzalloc_noprof(sizeof(struct netlbl_lsm_catmap), flags); } +#define netlbl_catmap_alloc(...) alloc_hooks(netlbl_catmap_alloc_noprof(__VA_ARGS__)) /** * netlbl_catmap_free - Free a LSM secattr catmap @@ -376,10 +379,11 @@ static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr) * pointer on success, or NULL on failure. * */ -static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc(gfp_t flags) +static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc_noprof(gfp_t flags) { - return kzalloc(sizeof(struct netlbl_lsm_secattr), flags); + return kzalloc_noprof(sizeof(struct netlbl_lsm_secattr), flags); } +#define netlbl_secattr_alloc(...) alloc_hooks(netlbl_secattr_alloc_noprof(__VA_ARGS__)) /** * netlbl_secattr_free - Frees a netlbl_lsm_secattr struct diff --git a/include/net/netlink.h b/include/net/netlink.h index c19ff921b661..972b5484fa6f 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -1891,10 +1891,11 @@ static inline struct nla_bitfield32 nla_get_bitfield32(const struct nlattr *nla) * @src: netlink attribute to duplicate from * @gfp: GFP mask */ -static inline void *nla_memdup(const struct nlattr *src, gfp_t gfp) +static inline void *nla_memdup_noprof(const struct nlattr *src, gfp_t gfp) { - return kmemdup(nla_data(src), nla_len(src), gfp); + return kmemdup_noprof(nla_data(src), nla_len(src), gfp); } +#define nla_memdup(...) alloc_hooks(nla_memdup_noprof(__VA_ARGS__)) /** * nla_nest_start_noflag - Start a new level of nested attributes diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 004e651e6067..29495c331d20 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -127,12 +127,12 @@ static inline struct sock *skb_steal_sock(struct sk_buff *skb, } static inline struct request_sock * -reqsk_alloc(const struct request_sock_ops *ops, struct sock *sk_listener, +reqsk_alloc_noprof(const struct request_sock_ops *ops, struct sock *sk_listener, bool attach_listener) { struct request_sock *req; - req = kmem_cache_alloc(ops->slab, GFP_ATOMIC | __GFP_NOWARN); + req = kmem_cache_alloc_noprof(ops->slab, GFP_ATOMIC | __GFP_NOWARN); if (!req) return NULL; req->rsk_listener = NULL; @@ -157,6 +157,7 @@ reqsk_alloc(const struct request_sock_ops *ops, struct sock *sk_listener, return req; } +#define reqsk_alloc(...) alloc_hooks(reqsk_alloc_noprof(__VA_ARGS__)) static inline void __reqsk_free(struct request_sock *req) { diff --git a/include/net/tcx.h b/include/net/tcx.h index 04be9377785d..72a3e75e539f 100644 --- a/include/net/tcx.h +++ b/include/net/tcx.h @@ -75,9 +75,9 @@ tcx_entry_fetch(struct net_device *dev, bool ingress) return rcu_dereference_rtnl(dev->tcx_egress); } -static inline struct bpf_mprog_entry *tcx_entry_create(void) +static inline struct bpf_mprog_entry *tcx_entry_create_noprof(void) { - struct tcx_entry *tcx = kzalloc(sizeof(*tcx), GFP_KERNEL); + struct tcx_entry *tcx = kzalloc_noprof(sizeof(*tcx), GFP_KERNEL); if (tcx) { bpf_mprog_bundle_init(&tcx->bundle); @@ -85,6 +85,7 @@ static inline struct bpf_mprog_entry *tcx_entry_create(void) } return NULL; } +#define tcx_entry_create(...) alloc_hooks(tcx_entry_create_noprof(__VA_ARGS__)) static inline void tcx_entry_free(struct bpf_mprog_entry *entry) { diff --git a/net/sunrpc/auth_gss/auth_gss_internal.h b/net/sunrpc/auth_gss/auth_gss_internal.h index c53b329092d4..4ebc1b7043d9 100644 --- a/net/sunrpc/auth_gss/auth_gss_internal.h +++ b/net/sunrpc/auth_gss/auth_gss_internal.h @@ -23,7 +23,7 @@ simple_get_bytes(const void *p, const void *end, void *res, size_t len) } static inline const void * -simple_get_netobj(const void *p, const void *end, struct xdr_netobj *dest) +simple_get_netobj_noprof(const void *p, const void *end, struct xdr_netobj *dest) { const void *q; unsigned int len; @@ -35,7 +35,7 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *dest) if (unlikely(q > end || q < p)) return ERR_PTR(-EFAULT); if (len) { - dest->data = kmemdup(p, len, GFP_KERNEL); + dest->data = kmemdup_noprof(p, len, GFP_KERNEL); if (unlikely(dest->data == NULL)) return ERR_PTR(-ENOMEM); } else @@ -43,3 +43,5 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *dest) dest->len = len; return q; } + +#define simple_get_netobj(...) alloc_hooks(simple_get_netobj_noprof(__VA_ARGS__)) -- cgit v1.2.3-59-g8ed1b From dee3d0bef2b00772be430425832ead6aa9d707f9 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 26 Mar 2024 17:10:32 +0000 Subject: proc: rewrite stable_page_flags() Reduce the usage of PageFlag tests and reduce the number of compound_head() calls. For multi-page folios, we'll now show all pages as having the flags that apply to them, e.g. if it's dirty, all pages will have the dirty flag set instead of just the head page. The mapped flag is still per page, as is the hwpoison flag. [willy@infradead.org: fix up some bits vs masks] Link: https://lkml.kernel.org/r/20240403173112.1450721-1-willy@infradead.org [willy@infradead.org: fix warnings] Link: https://lkml.kernel.org/r/ZhBPtCYfSuFuUMEz@casper.infradead.org Link: https://lkml.kernel.org/r/20240326171045.410737-11-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Svetly Todorov Signed-off-by: Andrew Morton --- fs/proc/page.c | 69 ++++++++++++++++++++++-------------------- include/linux/huge_mm.h | 4 +-- include/linux/page-flags.h | 2 +- tools/cgroup/memcg_slabinfo.py | 5 ++- 4 files changed, 42 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/fs/proc/page.c b/fs/proc/page.c index 9223856c934b..05120263af2a 100644 --- a/fs/proc/page.c +++ b/fs/proc/page.c @@ -107,10 +107,13 @@ static inline u64 kpf_copy_bit(u64 kflags, int ubit, int kbit) return ((kflags >> kbit) & 1) << ubit; } -u64 stable_page_flags(struct page *page) +u64 stable_page_flags(const struct page *page) { - u64 k; - u64 u; + const struct folio *folio; + unsigned long k; + unsigned long mapping; + bool is_anon; + u64 u = 0; /* * pseudo flag: KPF_NOPAGE @@ -118,49 +121,47 @@ u64 stable_page_flags(struct page *page) */ if (!page) return 1 << KPF_NOPAGE; + folio = page_folio(page); - k = page->flags; - u = 0; + k = folio->flags; + mapping = (unsigned long)folio->mapping; + is_anon = mapping & PAGE_MAPPING_ANON; /* * pseudo flags for the well known (anonymous) memory mapped pages */ if (page_mapped(page)) u |= 1 << KPF_MMAP; - if (PageAnon(page)) + if (is_anon) { u |= 1 << KPF_ANON; - if (PageKsm(page)) - u |= 1 << KPF_KSM; + if (mapping & PAGE_MAPPING_KSM) + u |= 1 << KPF_KSM; + } /* * compound pages: export both head/tail info * they together define a compound page's start/end pos and order */ - if (PageHead(page)) - u |= 1 << KPF_COMPOUND_HEAD; - if (PageTail(page)) + if (page == &folio->page) + u |= kpf_copy_bit(k, KPF_COMPOUND_HEAD, PG_head); + else u |= 1 << KPF_COMPOUND_TAIL; - if (PageHuge(page)) + if (folio_test_hugetlb(folio)) u |= 1 << KPF_HUGE; /* - * PageTransCompound can be true for non-huge compound pages (slab - * pages or pages allocated by drivers with __GFP_COMP) because it - * just checks PG_head/PG_tail, so we need to check PageLRU/PageAnon + * We need to check PageLRU/PageAnon * to make sure a given page is a thp, not a non-huge compound page. */ - else if (PageTransCompound(page)) { - struct page *head = compound_head(page); - - if (PageLRU(head) || PageAnon(head)) + else if (folio_test_large(folio)) { + if ((k & (1 << PG_lru)) || is_anon) u |= 1 << KPF_THP; - else if (is_huge_zero_page(head)) { + else if (is_huge_zero_page(&folio->page)) { u |= 1 << KPF_ZERO_PAGE; u |= 1 << KPF_THP; } } else if (is_zero_pfn(page_to_pfn(page))) u |= 1 << KPF_ZERO_PAGE; - /* * Caveats on high order pages: PG_buddy and PG_slab will only be set * on the head page. @@ -174,16 +175,17 @@ u64 stable_page_flags(struct page *page) u |= 1 << KPF_OFFLINE; if (PageTable(page)) u |= 1 << KPF_PGTABLE; + if (folio_test_slab(folio)) + u |= 1 << KPF_SLAB; - if (page_is_idle(page)) +#if defined(CONFIG_PAGE_IDLE_FLAG) && defined(CONFIG_64BIT) + u |= kpf_copy_bit(k, KPF_IDLE, PG_idle); +#else + if (folio_test_idle(folio)) u |= 1 << KPF_IDLE; +#endif u |= kpf_copy_bit(k, KPF_LOCKED, PG_locked); - - u |= kpf_copy_bit(k, KPF_SLAB, PG_slab); - if (PageTail(page) && PageSlab(page)) - u |= 1 << KPF_SLAB; - u |= kpf_copy_bit(k, KPF_ERROR, PG_error); u |= kpf_copy_bit(k, KPF_DIRTY, PG_dirty); u |= kpf_copy_bit(k, KPF_UPTODATE, PG_uptodate); @@ -194,7 +196,8 @@ u64 stable_page_flags(struct page *page) u |= kpf_copy_bit(k, KPF_ACTIVE, PG_active); u |= kpf_copy_bit(k, KPF_RECLAIM, PG_reclaim); - if (PageSwapCache(page)) +#define SWAPCACHE ((1 << PG_swapbacked) | (1 << PG_swapcache)) + if ((k & SWAPCACHE) == SWAPCACHE) u |= 1 << KPF_SWAPCACHE; u |= kpf_copy_bit(k, KPF_SWAPBACKED, PG_swapbacked); @@ -202,7 +205,10 @@ u64 stable_page_flags(struct page *page) u |= kpf_copy_bit(k, KPF_MLOCKED, PG_mlocked); #ifdef CONFIG_MEMORY_FAILURE - u |= kpf_copy_bit(k, KPF_HWPOISON, PG_hwpoison); + if (u & (1 << KPF_HUGE)) + u |= kpf_copy_bit(k, KPF_HWPOISON, PG_hwpoison); + else + u |= kpf_copy_bit(page->flags, KPF_HWPOISON, PG_hwpoison); #endif #ifdef CONFIG_ARCH_USES_PG_UNCACHED @@ -228,7 +234,6 @@ static ssize_t kpageflags_read(struct file *file, char __user *buf, { const unsigned long max_dump_pfn = get_max_dump_pfn(); u64 __user *out = (u64 __user *)buf; - struct page *ppage; unsigned long src = *ppos; unsigned long pfn; ssize_t ret = 0; @@ -245,9 +250,9 @@ static ssize_t kpageflags_read(struct file *file, char __user *buf, * TODO: ZONE_DEVICE support requires to identify * memmaps that were actually initialized. */ - ppage = pfn_to_online_page(pfn); + struct page *page = pfn_to_online_page(pfn); - if (put_user(stable_page_flags(ppage), out)) { + if (put_user(stable_page_flags(page), out)) { ret = -EFAULT; break; } diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 7576025db55d..1540a1481daf 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -351,7 +351,7 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf); extern struct page *huge_zero_page; extern unsigned long huge_zero_pfn; -static inline bool is_huge_zero_page(struct page *page) +static inline bool is_huge_zero_page(const struct page *page) { return READ_ONCE(huge_zero_page) == page; } @@ -480,7 +480,7 @@ static inline vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) return 0; } -static inline bool is_huge_zero_page(struct page *page) +static inline bool is_huge_zero_page(const struct page *page) { return false; } diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index eaecf544039f..888353c209c0 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -734,7 +734,7 @@ static __always_inline bool PageKsm(const struct page *page) TESTPAGEFLAG_FALSE(Ksm, ksm) #endif -u64 stable_page_flags(struct page *page); +u64 stable_page_flags(const struct page *page); /** * folio_xor_flags_has_waiters - Change some folio flags. diff --git a/tools/cgroup/memcg_slabinfo.py b/tools/cgroup/memcg_slabinfo.py index 1d3a90d93fe2..270c28a0d098 100644 --- a/tools/cgroup/memcg_slabinfo.py +++ b/tools/cgroup/memcg_slabinfo.py @@ -146,12 +146,11 @@ def detect_kernel_config(): def for_each_slab(prog): - PGSlab = 1 << prog.constant('PG_slab') - PGHead = 1 << prog.constant('PG_head') + PGSlab = ~prog.constant('PG_slab') for page in for_each_page(prog): try: - if page.flags.value_() & PGSlab: + if page.page_type.value_() == PGSlab: yield cast('struct slab *', page) except FaultError: pass -- cgit v1.2.3-59-g8ed1b From 5beaee54a324ba1fe307e341ec825d5d099f4091 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 26 Mar 2024 20:28:22 +0000 Subject: mm: add is_huge_zero_folio() This is the folio equivalent of is_huge_zero_page(). It doesn't add any efficiency, but it does prevent the caller from passing a tail page and getting confused when the predicate returns false. Link: https://lkml.kernel.org/r/20240326202833.523759-3-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Hildenbrand Signed-off-by: Andrew Morton --- fs/proc/page.c | 2 +- include/linux/huge_mm.h | 10 ++++++++++ mm/huge_memory.c | 6 +++--- mm/mempolicy.c | 2 +- mm/swap.c | 2 +- mm/swap_state.c | 2 +- mm/userfaultfd.c | 2 +- 7 files changed, 18 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/proc/page.c b/fs/proc/page.c index 05120263af2a..2fb64bdb64eb 100644 --- a/fs/proc/page.c +++ b/fs/proc/page.c @@ -155,7 +155,7 @@ u64 stable_page_flags(const struct page *page) else if (folio_test_large(folio)) { if ((k & (1 << PG_lru)) || is_anon) u |= 1 << KPF_THP; - else if (is_huge_zero_page(&folio->page)) { + else if (is_huge_zero_folio(folio)) { u |= 1 << KPF_ZERO_PAGE; u |= 1 << KPF_THP; } diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 1540a1481daf..600c6008262b 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -356,6 +356,11 @@ static inline bool is_huge_zero_page(const struct page *page) return READ_ONCE(huge_zero_page) == page; } +static inline bool is_huge_zero_folio(const struct folio *folio) +{ + return READ_ONCE(huge_zero_page) == &folio->page; +} + static inline bool is_huge_zero_pmd(pmd_t pmd) { return pmd_present(pmd) && READ_ONCE(huge_zero_pfn) == pmd_pfn(pmd); @@ -485,6 +490,11 @@ static inline bool is_huge_zero_page(const struct page *page) return false; } +static inline bool is_huge_zero_folio(const struct folio *folio) +{ + return false; +} + static inline bool is_huge_zero_pmd(pmd_t pmd) { return false; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 75ad971ca45e..5c043c7b5062 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -789,12 +789,12 @@ struct deferred_split *get_deferred_split_queue(struct folio *folio) } #endif -static inline bool is_transparent_hugepage(struct folio *folio) +static inline bool is_transparent_hugepage(const struct folio *folio) { if (!folio_test_large(folio)) return false; - return is_huge_zero_page(&folio->page) || + return is_huge_zero_folio(folio) || folio_test_large_rmappable(folio); } @@ -3085,7 +3085,7 @@ int split_huge_page_to_list_to_order(struct page *page, struct list_head *list, } - is_hzp = is_huge_zero_page(&folio->page); + is_hzp = is_huge_zero_folio(folio); if (is_hzp) { pr_warn_ratelimited("Called split_huge_page for huge zero page\n"); return -EBUSY; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 913cff5da5a3..5743028a63a5 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -510,7 +510,7 @@ static void queue_folios_pmd(pmd_t *pmd, struct mm_walk *walk) return; } folio = pfn_folio(pmd_pfn(*pmd)); - if (is_huge_zero_page(&folio->page)) { + if (is_huge_zero_folio(folio)) { walk->action = ACTION_CONTINUE; return; } diff --git a/mm/swap.c b/mm/swap.c index 500a09a48dfd..f72364e92d5f 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -985,7 +985,7 @@ void folios_put_refs(struct folio_batch *folios, unsigned int *refs) struct folio *folio = folios->folios[i]; unsigned int nr_refs = refs ? refs[i] : 1; - if (is_huge_zero_page(&folio->page)) + if (is_huge_zero_folio(folio)) continue; if (folio_is_zone_device(folio)) { diff --git a/mm/swap_state.c b/mm/swap_state.c index bfc7e8c58a6d..2deac23633cd 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -301,7 +301,7 @@ void free_page_and_swap_cache(struct page *page) struct folio *folio = page_folio(page); free_swap_cache(folio); - if (!is_huge_zero_page(page)) + if (!is_huge_zero_folio(folio)) folio_put(folio); } diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index 3c3539c573e7..a0ec14553fbe 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -1664,7 +1664,7 @@ ssize_t move_pages(struct userfaultfd_ctx *ctx, unsigned long dst_start, !pmd_none(dst_pmdval)) { struct folio *folio = pfn_folio(pmd_pfn(*src_pmd)); - if (!folio || (!is_huge_zero_page(&folio->page) && + if (!folio || (!is_huge_zero_folio(folio) && !PageAnonExclusive(&folio->page))) { spin_unlock(ptl); err = -EBUSY; -- cgit v1.2.3-59-g8ed1b From c93012d849c9e3fab9540458623b98794c4bf0a1 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 26 Mar 2024 20:28:27 +0000 Subject: dax: use huge_zero_folio Convert from huge_zero_page to huge_zero_folio. Link: https://lkml.kernel.org/r/20240326202833.523759-8-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Hildenbrand Signed-off-by: Andrew Morton --- fs/dax.c | 14 +++++++------- include/trace/events/fs_dax.h | 16 ++++++++-------- 2 files changed, 15 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/dax.c b/fs/dax.c index 423fc1607dfa..becb4a6920c6 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -1207,17 +1207,17 @@ static vm_fault_t dax_pmd_load_hole(struct xa_state *xas, struct vm_fault *vmf, struct vm_area_struct *vma = vmf->vma; struct inode *inode = mapping->host; pgtable_t pgtable = NULL; - struct page *zero_page; + struct folio *zero_folio; spinlock_t *ptl; pmd_t pmd_entry; pfn_t pfn; - zero_page = mm_get_huge_zero_page(vmf->vma->vm_mm); + zero_folio = mm_get_huge_zero_folio(vmf->vma->vm_mm); - if (unlikely(!zero_page)) + if (unlikely(!zero_folio)) goto fallback; - pfn = page_to_pfn_t(zero_page); + pfn = page_to_pfn_t(&zero_folio->page); *entry = dax_insert_entry(xas, vmf, iter, *entry, pfn, DAX_PMD | DAX_ZERO_PAGE); @@ -1237,17 +1237,17 @@ static vm_fault_t dax_pmd_load_hole(struct xa_state *xas, struct vm_fault *vmf, pgtable_trans_huge_deposit(vma->vm_mm, vmf->pmd, pgtable); mm_inc_nr_ptes(vma->vm_mm); } - pmd_entry = mk_pmd(zero_page, vmf->vma->vm_page_prot); + pmd_entry = mk_pmd(&zero_folio->page, vmf->vma->vm_page_prot); pmd_entry = pmd_mkhuge(pmd_entry); set_pmd_at(vmf->vma->vm_mm, pmd_addr, vmf->pmd, pmd_entry); spin_unlock(ptl); - trace_dax_pmd_load_hole(inode, vmf, zero_page, *entry); + trace_dax_pmd_load_hole(inode, vmf, zero_folio, *entry); return VM_FAULT_NOPAGE; fallback: if (pgtable) pte_free(vma->vm_mm, pgtable); - trace_dax_pmd_load_hole_fallback(inode, vmf, zero_page, *entry); + trace_dax_pmd_load_hole_fallback(inode, vmf, zero_folio, *entry); return VM_FAULT_FALLBACK; } #else diff --git a/include/trace/events/fs_dax.h b/include/trace/events/fs_dax.h index 97b09fcf7e52..86fe6aecff1e 100644 --- a/include/trace/events/fs_dax.h +++ b/include/trace/events/fs_dax.h @@ -62,14 +62,14 @@ DEFINE_PMD_FAULT_EVENT(dax_pmd_fault_done); DECLARE_EVENT_CLASS(dax_pmd_load_hole_class, TP_PROTO(struct inode *inode, struct vm_fault *vmf, - struct page *zero_page, + struct folio *zero_folio, void *radix_entry), - TP_ARGS(inode, vmf, zero_page, radix_entry), + TP_ARGS(inode, vmf, zero_folio, radix_entry), TP_STRUCT__entry( __field(unsigned long, ino) __field(unsigned long, vm_flags) __field(unsigned long, address) - __field(struct page *, zero_page) + __field(struct folio *, zero_folio) __field(void *, radix_entry) __field(dev_t, dev) ), @@ -78,17 +78,17 @@ DECLARE_EVENT_CLASS(dax_pmd_load_hole_class, __entry->ino = inode->i_ino; __entry->vm_flags = vmf->vma->vm_flags; __entry->address = vmf->address; - __entry->zero_page = zero_page; + __entry->zero_folio = zero_folio; __entry->radix_entry = radix_entry; ), - TP_printk("dev %d:%d ino %#lx %s address %#lx zero_page %p " + TP_printk("dev %d:%d ino %#lx %s address %#lx zero_folio %p " "radix_entry %#lx", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->vm_flags & VM_SHARED ? "shared" : "private", __entry->address, - __entry->zero_page, + __entry->zero_folio, (unsigned long)__entry->radix_entry ) ) @@ -96,8 +96,8 @@ DECLARE_EVENT_CLASS(dax_pmd_load_hole_class, #define DEFINE_PMD_LOAD_HOLE_EVENT(name) \ DEFINE_EVENT(dax_pmd_load_hole_class, name, \ TP_PROTO(struct inode *inode, struct vm_fault *vmf, \ - struct page *zero_page, void *radix_entry), \ - TP_ARGS(inode, vmf, zero_page, radix_entry)) + struct folio *zero_folio, void *radix_entry), \ + TP_ARGS(inode, vmf, zero_folio, radix_entry)) DEFINE_PMD_LOAD_HOLE_EVENT(dax_pmd_load_hole); DEFINE_PMD_LOAD_HOLE_EVENT(dax_pmd_load_hole_fallback); -- cgit v1.2.3-59-g8ed1b From afd584398b51eba12d0ffa4e0e0a493323993103 Mon Sep 17 00:00:00 2001 From: ZhangPeng Date: Wed, 27 Mar 2024 17:08:35 +0800 Subject: userfaultfd: early return in dup_userfaultfd() When vma->vm_userfaultfd_ctx.ctx is NULL, vma->vm_flags should have cleared __VM_UFFD_FLAGS. Therefore, there is no need to down_write or clear the flag, which will affect fork performance. Fix this by returning early if octx is NULL in dup_userfaultfd(). By applying this patch we can get a 1.3% performance improvement for lmbench fork_prot. Results are as follows: base early return Process fork+exit: 419.1106 413.4804 Link: https://lkml.kernel.org/r/20240327090835.3232629-1-zhangpeng362@huawei.com Signed-off-by: ZhangPeng Reviewed-by: Peter Xu Cc: Axel Rasmussen Cc: David Hildenbrand Cc: Kefeng Wang Cc: Liam R. Howlett Cc: Lokesh Gidra Cc: Nanyong Sun Cc: Suren Baghdasaryan Signed-off-by: Andrew Morton --- fs/userfaultfd.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 60dcfafdc11a..3e6ddda6f159 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -657,7 +657,10 @@ int dup_userfaultfd(struct vm_area_struct *vma, struct list_head *fcs) struct userfaultfd_fork_ctx *fctx; octx = vma->vm_userfaultfd_ctx.ctx; - if (!octx || !(octx->features & UFFD_FEATURE_EVENT_FORK)) { + if (!octx) + return 0; + + if (!(octx->features & UFFD_FEATURE_EVENT_FORK)) { vma_start_write(vma); vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX; userfaultfd_set_vm_flags(vma, vma->vm_flags & ~__VM_UFFD_FLAGS); -- cgit v1.2.3-59-g8ed1b From 5def1e0f476da713141269d86815aba7c2817cb5 Mon Sep 17 00:00:00 2001 From: Rick Edgecombe Date: Mon, 25 Mar 2024 19:16:43 -0700 Subject: proc: refactor pde_get_unmapped_area as prep MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Patch series "Cover a guard gap corner case", v4. In working on x86’s shadow stack feature, I came across some limitations around the kernel’s handling of guard gaps. AFAICT these limitations are not too important for the traditional stack usage of guard gaps, but have bigger impact on shadow stack’s usage. And now in addition to x86, we have two other architectures implementing shadow stack like features that plan to use guard gaps. I wanted to see about addressing them, but I have not worked on mmap() placement related code before, so would greatly appreciate if people could take a look and point me in the right direction. The nature of the limitations of concern is as follows. In order to ensure guard gaps between mappings, mmap() would need to consider two things: 1. That the new mapping isn’t placed in an any existing mapping’s guard gap. 2. That the new mapping isn’t placed such that any existing mappings are not in *its* guard gaps Currently mmap never considers (2), and (1) is not considered in some situations. When not passing an address hint, or passing one without MAP_FIXED_NOREPLACE, (1) is enforced. With MAP_FIXED_NOREPLACE, (1) is not enforced. With MAP_FIXED, (1) is not considered, but this seems to be expected since MAP_FIXED can already clobber existing mappings. For MAP_FIXED_NOREPLACE I would have guessed it should respect the guard gaps of existing mappings, but it is probably a little ambiguous. In this series I just tried to add enforcement of (2) for the normal (no address hint) case and only for the newer shadow stack memory (not stacks). The reason is that with the no-address-hint situation, landing next to a guard gap could come up naturally and so be more influencable by attackers such that two shadow stacks could be adjacent without a guard gap. Where as the address-hint scenarios would require more control - being able to call mmap() with specific arguments. As for why not just fix the other corner cases anyway, I thought it might have some greater possibility of affecting existing apps. This patch (of 14): Future changes will perform a treewide change to remove the indirect branch that is involved in calling mm->get_unmapped_area(). After doing this, the function will no longer be able to be handled as a function pointer. To make the treewide change diff cleaner and easier to review, refactor pde_get_unmapped_area() such that mm->get_unmapped_area() is called without being stored in a local function pointer. With this in refactoring, follow on changes will be able to simply replace the call site with a future function that calls it directly. Link: https://lkml.kernel.org/r/20240326021656.202649-1-rick.p.edgecombe@intel.com Link: https://lkml.kernel.org/r/20240326021656.202649-2-rick.p.edgecombe@intel.com Signed-off-by: Rick Edgecombe Cc: Andy Lutomirski Cc: Borislav Petkov (AMD) Cc: Christophe Leroy Cc: Dave Hansen Cc: Deepak Gupta Cc: H. Peter Anvin (Intel) Cc: Ingo Molnar Cc: Kees Cook Cc: Kirill A. Shutemov Cc: Liam R. Howlett Cc: Mark Brown Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Alexei Starovoitov Cc: Aneesh Kumar K.V Cc: Dan Williams Cc: Guo Ren Cc: Helge Deller Cc: "James E.J. Bottomley" Cc: Michael Ellerman Cc: Naveen N. Rao Cc: Nicholas Piggin Signed-off-by: Andrew Morton --- fs/proc/inode.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/proc/inode.c b/fs/proc/inode.c index dcd513dccf55..75396a24fd8c 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -451,15 +451,12 @@ pde_get_unmapped_area(struct proc_dir_entry *pde, struct file *file, unsigned lo unsigned long len, unsigned long pgoff, unsigned long flags) { - typeof_member(struct proc_ops, proc_get_unmapped_area) get_area; + if (pde->proc_ops->proc_get_unmapped_area) + return pde->proc_ops->proc_get_unmapped_area(file, orig_addr, len, pgoff, flags); - get_area = pde->proc_ops->proc_get_unmapped_area; #ifdef CONFIG_MMU - if (!get_area) - get_area = current->mm->get_unmapped_area; + return current->mm->get_unmapped_area(file, orig_addr, len, pgoff, flags); #endif - if (get_area) - return get_area(file, orig_addr, len, pgoff, flags); return orig_addr; } -- cgit v1.2.3-59-g8ed1b From 529ce23a764f25d172198b4c6ba90f1e2ad17f93 Mon Sep 17 00:00:00 2001 From: Rick Edgecombe Date: Mon, 25 Mar 2024 19:16:44 -0700 Subject: mm: switch mm->get_unmapped_area() to a flag The mm_struct contains a function pointer *get_unmapped_area(), which is set to either arch_get_unmapped_area() or arch_get_unmapped_area_topdown() during the initialization of the mm. Since the function pointer only ever points to two functions that are named the same across all arch's, a function pointer is not really required. In addition future changes will want to add versions of the functions that take additional arguments. So to save a pointers worth of bytes in mm_struct, and prevent adding additional function pointers to mm_struct in future changes, remove it and keep the information about which get_unmapped_area() to use in a flag. Add the new flag to MMF_INIT_MASK so it doesn't get clobbered on fork by mmf_init_flags(). Most MM flags get clobbered on fork. In the pre-existing behavior mm->get_unmapped_area() would get copied to the new mm in dup_mm(), so not clobbering the flag preserves the existing behavior around inheriting the topdown-ness. Introduce a helper, mm_get_unmapped_area(), to easily convert code that refers to the old function pointer to instead select and call either arch_get_unmapped_area() or arch_get_unmapped_area_topdown() based on the flag. Then drop the mm->get_unmapped_area() function pointer. Leave the get_unmapped_area() pointer in struct file_operations alone. The main purpose of this change is to reorganize in preparation for future changes, but it also converts the calls of mm->get_unmapped_area() from indirect branches into a direct ones. The stress-ng bigheap benchmark calls realloc a lot, which calls through get_unmapped_area() in the kernel. On x86, the change yielded a ~1% improvement there on a retpoline config. In testing a few x86 configs, removing the pointer unfortunately didn't result in any actual size reductions in the compiled layout of mm_struct. But depending on compiler or arch alignment requirements, the change could shrink the size of mm_struct. Link: https://lkml.kernel.org/r/20240326021656.202649-3-rick.p.edgecombe@intel.com Signed-off-by: Rick Edgecombe Acked-by: Dave Hansen Acked-by: Liam R. Howlett Reviewed-by: Kirill A. Shutemov Acked-by: Alexei Starovoitov Cc: Dan Williams Cc: Andy Lutomirski Cc: Aneesh Kumar K.V Cc: Borislav Petkov (AMD) Cc: Christophe Leroy Cc: Deepak Gupta Cc: Guo Ren Cc: Helge Deller Cc: H. Peter Anvin (Intel) Cc: Ingo Molnar Cc: "James E.J. Bottomley" Cc: Kees Cook Cc: Mark Brown Cc: Michael Ellerman Cc: Naveen N. Rao Cc: Nicholas Piggin Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- arch/s390/mm/hugetlbpage.c | 2 +- arch/s390/mm/mmap.c | 4 ++-- arch/sparc/kernel/sys_sparc_64.c | 15 ++++++--------- arch/sparc/mm/hugetlbpage.c | 2 +- arch/x86/kernel/cpu/sgx/driver.c | 2 +- arch/x86/mm/hugetlbpage.c | 2 +- arch/x86/mm/mmap.c | 4 ++-- drivers/char/mem.c | 2 +- drivers/dax/device.c | 6 +++--- fs/hugetlbfs/inode.c | 4 ++-- fs/proc/inode.c | 3 ++- fs/ramfs/file-mmu.c | 2 +- include/linux/mm_types.h | 6 +----- include/linux/sched/coredump.h | 5 ++++- include/linux/sched/mm.h | 5 +++++ io_uring/io_uring.c | 2 +- kernel/bpf/arena.c | 2 +- kernel/bpf/syscall.c | 2 +- mm/debug.c | 6 ------ mm/huge_memory.c | 9 ++++----- mm/mmap.c | 21 ++++++++++++++++++--- mm/shmem.c | 11 +++++------ mm/util.c | 6 +++--- 23 files changed, 66 insertions(+), 57 deletions(-) (limited to 'fs') diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index ca43b6fce71c..7d948e243f4b 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -318,7 +318,7 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, goto check_asce_limit; } - if (mm->get_unmapped_area == arch_get_unmapped_area) + if (!test_bit(MMF_TOPDOWN, &mm->flags)) addr = hugetlb_get_unmapped_area_bottomup(file, addr, len, pgoff, flags); else diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c index b14fc0887654..6b2e4436ad4a 100644 --- a/arch/s390/mm/mmap.c +++ b/arch/s390/mm/mmap.c @@ -185,10 +185,10 @@ void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack) */ if (mmap_is_legacy(rlim_stack)) { mm->mmap_base = mmap_base_legacy(random_factor); - mm->get_unmapped_area = arch_get_unmapped_area; + clear_bit(MMF_TOPDOWN, &mm->flags); } else { mm->mmap_base = mmap_base(random_factor, rlim_stack); - mm->get_unmapped_area = arch_get_unmapped_area_topdown; + set_bit(MMF_TOPDOWN, &mm->flags); } } diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c index 1e9a9e016237..1dbf7211666e 100644 --- a/arch/sparc/kernel/sys_sparc_64.c +++ b/arch/sparc/kernel/sys_sparc_64.c @@ -218,14 +218,10 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, unsigned long get_fb_unmapped_area(struct file *filp, unsigned long orig_addr, unsigned long len, unsigned long pgoff, unsigned long flags) { unsigned long align_goal, addr = -ENOMEM; - unsigned long (*get_area)(struct file *, unsigned long, - unsigned long, unsigned long, unsigned long); - - get_area = current->mm->get_unmapped_area; if (flags & MAP_FIXED) { /* Ok, don't mess with it. */ - return get_area(NULL, orig_addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, NULL, orig_addr, len, pgoff, flags); } flags &= ~MAP_SHARED; @@ -238,7 +234,8 @@ unsigned long get_fb_unmapped_area(struct file *filp, unsigned long orig_addr, u align_goal = (64UL * 1024); do { - addr = get_area(NULL, orig_addr, len + (align_goal - PAGE_SIZE), pgoff, flags); + addr = mm_get_unmapped_area(current->mm, NULL, orig_addr, + len + (align_goal - PAGE_SIZE), pgoff, flags); if (!(addr & ~PAGE_MASK)) { addr = (addr + (align_goal - 1UL)) & ~(align_goal - 1UL); break; @@ -256,7 +253,7 @@ unsigned long get_fb_unmapped_area(struct file *filp, unsigned long orig_addr, u * be obtained. */ if (addr & ~PAGE_MASK) - addr = get_area(NULL, orig_addr, len, pgoff, flags); + addr = mm_get_unmapped_area(current->mm, NULL, orig_addr, len, pgoff, flags); return addr; } @@ -292,7 +289,7 @@ void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack) gap == RLIM_INFINITY || sysctl_legacy_va_layout) { mm->mmap_base = TASK_UNMAPPED_BASE + random_factor; - mm->get_unmapped_area = arch_get_unmapped_area; + clear_bit(MMF_TOPDOWN, &mm->flags); } else { /* We know it's 32-bit */ unsigned long task_size = STACK_TOP32; @@ -303,7 +300,7 @@ void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack) gap = (task_size / 6 * 5); mm->mmap_base = PAGE_ALIGN(task_size - gap - random_factor); - mm->get_unmapped_area = arch_get_unmapped_area_topdown; + set_bit(MMF_TOPDOWN, &mm->flags); } } diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c index 8ed5bdf95d25..c23012e3a353 100644 --- a/arch/sparc/mm/hugetlbpage.c +++ b/arch/sparc/mm/hugetlbpage.c @@ -123,7 +123,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, (!vma || addr + len <= vm_start_gap(vma))) return addr; } - if (mm->get_unmapped_area == arch_get_unmapped_area) + if (!test_bit(MMF_TOPDOWN, &mm->flags)) return hugetlb_get_unmapped_area_bottomup(file, addr, len, pgoff, flags); else diff --git a/arch/x86/kernel/cpu/sgx/driver.c b/arch/x86/kernel/cpu/sgx/driver.c index 262f5fb18d74..22b65a5f5ec6 100644 --- a/arch/x86/kernel/cpu/sgx/driver.c +++ b/arch/x86/kernel/cpu/sgx/driver.c @@ -113,7 +113,7 @@ static unsigned long sgx_get_unmapped_area(struct file *file, if (flags & MAP_FIXED) return addr; - return current->mm->get_unmapped_area(file, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, file, addr, len, pgoff, flags); } #ifdef CONFIG_COMPAT diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index dab6db288e5b..06ca9a60bac2 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -115,7 +115,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, } get_unmapped_area: - if (mm->get_unmapped_area == arch_get_unmapped_area) + if (!test_bit(MMF_TOPDOWN, &mm->flags)) return hugetlb_get_unmapped_area_bottomup(file, addr, len, pgoff, flags); else diff --git a/arch/x86/mm/mmap.c b/arch/x86/mm/mmap.c index c90c20904a60..a2cabb1c81e1 100644 --- a/arch/x86/mm/mmap.c +++ b/arch/x86/mm/mmap.c @@ -129,9 +129,9 @@ static void arch_pick_mmap_base(unsigned long *base, unsigned long *legacy_base, void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack) { if (mmap_is_legacy()) - mm->get_unmapped_area = arch_get_unmapped_area; + clear_bit(MMF_TOPDOWN, &mm->flags); else - mm->get_unmapped_area = arch_get_unmapped_area_topdown; + set_bit(MMF_TOPDOWN, &mm->flags); arch_pick_mmap_base(&mm->mmap_base, &mm->mmap_legacy_base, arch_rnd(mmap64_rnd_bits), task_size_64bit(0), diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 3c6670cf905f..9b80e622ae80 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -544,7 +544,7 @@ static unsigned long get_unmapped_area_zero(struct file *file, } /* Otherwise flags & MAP_PRIVATE: with no shmem object beneath it */ - return current->mm->get_unmapped_area(file, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, file, addr, len, pgoff, flags); #else return -ENOSYS; #endif diff --git a/drivers/dax/device.c b/drivers/dax/device.c index 93ebedc5ec8c..47c126d37b59 100644 --- a/drivers/dax/device.c +++ b/drivers/dax/device.c @@ -329,14 +329,14 @@ static unsigned long dax_get_unmapped_area(struct file *filp, if ((off + len_align) < off) goto out; - addr_align = current->mm->get_unmapped_area(filp, addr, len_align, - pgoff, flags); + addr_align = mm_get_unmapped_area(current->mm, filp, addr, len_align, + pgoff, flags); if (!IS_ERR_VALUE(addr_align)) { addr_align += (off - addr_align) & (align - 1); return addr_align; } out: - return current->mm->get_unmapped_area(filp, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, filp, addr, len, pgoff, flags); } static const struct address_space_operations dev_dax_aops = { diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6502c7e776d1..3dee18bf47ed 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -249,11 +249,11 @@ generic_hugetlb_get_unmapped_area(struct file *file, unsigned long addr, } /* - * Use mm->get_unmapped_area value as a hint to use topdown routine. + * Use MMF_TOPDOWN flag as a hint to use topdown routine. * If architectures have special needs, they should define their own * version of hugetlb_get_unmapped_area. */ - if (mm->get_unmapped_area == arch_get_unmapped_area_topdown) + if (test_bit(MMF_TOPDOWN, &mm->flags)) return hugetlb_get_unmapped_area_topdown(file, addr, len, pgoff, flags); return hugetlb_get_unmapped_area_bottomup(file, addr, len, diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 75396a24fd8c..d19434e2a58e 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -455,8 +455,9 @@ pde_get_unmapped_area(struct proc_dir_entry *pde, struct file *file, unsigned lo return pde->proc_ops->proc_get_unmapped_area(file, orig_addr, len, pgoff, flags); #ifdef CONFIG_MMU - return current->mm->get_unmapped_area(file, orig_addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, file, orig_addr, len, pgoff, flags); #endif + return orig_addr; } diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c index c7a1aa3c882b..b45c7edc3225 100644 --- a/fs/ramfs/file-mmu.c +++ b/fs/ramfs/file-mmu.c @@ -35,7 +35,7 @@ static unsigned long ramfs_mmu_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { - return current->mm->get_unmapped_area(file, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, file, addr, len, pgoff, flags); } const struct file_operations ramfs_file_operations = { diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index fee6561feb7e..fa0d6995706f 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -777,11 +777,7 @@ struct mm_struct { } ____cacheline_aligned_in_smp; struct maple_tree mm_mt; -#ifdef CONFIG_MMU - unsigned long (*get_unmapped_area) (struct file *filp, - unsigned long addr, unsigned long len, - unsigned long pgoff, unsigned long flags); -#endif + unsigned long mmap_base; /* base of mmap area */ unsigned long mmap_legacy_base; /* base of mmap area in bottom-up allocations */ #ifdef CONFIG_HAVE_ARCH_COMPAT_MMAP_BASES diff --git a/include/linux/sched/coredump.h b/include/linux/sched/coredump.h index 02f5090ffea2..e62ff805cfc9 100644 --- a/include/linux/sched/coredump.h +++ b/include/linux/sched/coredump.h @@ -92,9 +92,12 @@ static inline int get_dumpable(struct mm_struct *mm) #define MMF_VM_MERGE_ANY 30 #define MMF_VM_MERGE_ANY_MASK (1 << MMF_VM_MERGE_ANY) +#define MMF_TOPDOWN 31 /* mm searches top down by default */ +#define MMF_TOPDOWN_MASK (1 << MMF_TOPDOWN) + #define MMF_INIT_MASK (MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK |\ MMF_DISABLE_THP_MASK | MMF_HAS_MDWE_MASK |\ - MMF_VM_MERGE_ANY_MASK) + MMF_VM_MERGE_ANY_MASK | MMF_TOPDOWN_MASK) static inline unsigned long mmf_init_flags(unsigned long flags) { diff --git a/include/linux/sched/mm.h b/include/linux/sched/mm.h index b6543f9d78d6..ed1caa26c8be 100644 --- a/include/linux/sched/mm.h +++ b/include/linux/sched/mm.h @@ -8,6 +8,7 @@ #include #include #include +#include /* * Routines for handling mm_structs @@ -186,6 +187,10 @@ arch_get_unmapped_area_topdown(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags); +unsigned long mm_get_unmapped_area(struct mm_struct *mm, struct file *filp, + unsigned long addr, unsigned long len, + unsigned long pgoff, unsigned long flags); + unsigned long generic_get_unmapped_area(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index c170a2b8d2cf..d5edfb8444d7 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -3525,7 +3525,7 @@ static unsigned long io_uring_mmu_get_unmapped_area(struct file *filp, #else addr = 0UL; #endif - return current->mm->get_unmapped_area(filp, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, filp, addr, len, pgoff, flags); } #else /* !CONFIG_MMU */ diff --git a/kernel/bpf/arena.c b/kernel/bpf/arena.c index 343c3456c8dd..16dbf4f6b77f 100644 --- a/kernel/bpf/arena.c +++ b/kernel/bpf/arena.c @@ -314,7 +314,7 @@ static unsigned long arena_get_unmapped_area(struct file *filp, unsigned long ad return -EINVAL; } - ret = current->mm->get_unmapped_area(filp, addr, len * 2, 0, flags); + ret = mm_get_unmapped_area(current->mm, filp, addr, len * 2, 0, flags); if (IS_ERR_VALUE(ret)) return ret; if ((ret >> 32) == ((ret + len - 1) >> 32)) diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index c287925471f6..9cb89e875f0d 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -980,7 +980,7 @@ static unsigned long bpf_get_unmapped_area(struct file *filp, unsigned long addr if (map->ops->map_get_unmapped_area) return map->ops->map_get_unmapped_area(filp, addr, len, pgoff, flags); #ifdef CONFIG_MMU - return current->mm->get_unmapped_area(filp, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, filp, addr, len, pgoff, flags); #else return addr; #endif diff --git a/mm/debug.c b/mm/debug.c index e8a96b8b7197..b71186f1fb0b 100644 --- a/mm/debug.c +++ b/mm/debug.c @@ -177,9 +177,6 @@ EXPORT_SYMBOL(dump_vma); void dump_mm(const struct mm_struct *mm) { pr_emerg("mm %px task_size %lu\n" -#ifdef CONFIG_MMU - "get_unmapped_area %px\n" -#endif "mmap_base %lu mmap_legacy_base %lu\n" "pgd %px mm_users %d mm_count %d pgtables_bytes %lu map_count %d\n" "hiwater_rss %lx hiwater_vm %lx total_vm %lx locked_vm %lx\n" @@ -205,9 +202,6 @@ void dump_mm(const struct mm_struct *mm) "def_flags: %#lx(%pGv)\n", mm, mm->task_size, -#ifdef CONFIG_MMU - mm->get_unmapped_area, -#endif mm->mmap_base, mm->mmap_legacy_base, mm->pgd, atomic_read(&mm->mm_users), atomic_read(&mm->mm_count), diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 157cee64850c..7fdb2275a1e0 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -816,8 +816,8 @@ static unsigned long __thp_get_unmapped_area(struct file *filp, if (len_pad < len || (off + len_pad) < off) return 0; - ret = current->mm->get_unmapped_area(filp, addr, len_pad, - off >> PAGE_SHIFT, flags); + ret = mm_get_unmapped_area(current->mm, filp, addr, len_pad, + off >> PAGE_SHIFT, flags); /* * The failure might be due to length padding. The caller will retry @@ -835,8 +835,7 @@ static unsigned long __thp_get_unmapped_area(struct file *filp, off_sub = (off - ret) & (size - 1); - if (current->mm->get_unmapped_area == arch_get_unmapped_area_topdown && - !off_sub) + if (test_bit(MMF_TOPDOWN, ¤t->mm->flags) && !off_sub) return ret + size; ret += off_sub; @@ -853,7 +852,7 @@ unsigned long thp_get_unmapped_area(struct file *filp, unsigned long addr, if (ret) return ret; - return current->mm->get_unmapped_area(filp, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, filp, addr, len, pgoff, flags); } EXPORT_SYMBOL_GPL(thp_get_unmapped_area); diff --git a/mm/mmap.c b/mm/mmap.c index 77a625e13ec1..dfb8a518e8c9 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1812,7 +1812,8 @@ get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { unsigned long (*get_area)(struct file *, unsigned long, - unsigned long, unsigned long, unsigned long); + unsigned long, unsigned long, unsigned long) + = NULL; unsigned long error = arch_mmap_check(addr, len, flags); if (error) @@ -1822,7 +1823,6 @@ get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, if (len > TASK_SIZE) return -ENOMEM; - get_area = current->mm->get_unmapped_area; if (file) { if (file->f_op->get_unmapped_area) get_area = file->f_op->get_unmapped_area; @@ -1841,7 +1841,11 @@ get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, if (!file) pgoff = 0; - addr = get_area(file, addr, len, pgoff, flags); + if (get_area) + addr = get_area(file, addr, len, pgoff, flags); + else + addr = mm_get_unmapped_area(current->mm, file, addr, len, + pgoff, flags); if (IS_ERR_VALUE(addr)) return addr; @@ -1856,6 +1860,17 @@ get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, EXPORT_SYMBOL(get_unmapped_area); +unsigned long +mm_get_unmapped_area(struct mm_struct *mm, struct file *file, + unsigned long addr, unsigned long len, + unsigned long pgoff, unsigned long flags) +{ + if (test_bit(MMF_TOPDOWN, &mm->flags)) + return arch_get_unmapped_area_topdown(file, addr, len, pgoff, flags); + return arch_get_unmapped_area(file, addr, len, pgoff, flags); +} +EXPORT_SYMBOL(mm_get_unmapped_area); + /** * find_vma_intersection() - Look up the first VMA which intersects the interval * @mm: The process address space. diff --git a/mm/shmem.c b/mm/shmem.c index 98985179f495..fa2a0ed97507 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2267,8 +2267,6 @@ unsigned long shmem_get_unmapped_area(struct file *file, unsigned long uaddr, unsigned long len, unsigned long pgoff, unsigned long flags) { - unsigned long (*get_area)(struct file *, - unsigned long, unsigned long, unsigned long, unsigned long); unsigned long addr; unsigned long offset; unsigned long inflated_len; @@ -2278,8 +2276,8 @@ unsigned long shmem_get_unmapped_area(struct file *file, if (len > TASK_SIZE) return -ENOMEM; - get_area = current->mm->get_unmapped_area; - addr = get_area(file, uaddr, len, pgoff, flags); + addr = mm_get_unmapped_area(current->mm, file, uaddr, len, pgoff, + flags); if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) return addr; @@ -2336,7 +2334,8 @@ unsigned long shmem_get_unmapped_area(struct file *file, if (inflated_len < len) return addr; - inflated_addr = get_area(NULL, uaddr, inflated_len, 0, flags); + inflated_addr = mm_get_unmapped_area(current->mm, NULL, uaddr, + inflated_len, 0, flags); if (IS_ERR_VALUE(inflated_addr)) return addr; if (inflated_addr & ~PAGE_MASK) @@ -4801,7 +4800,7 @@ unsigned long shmem_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { - return current->mm->get_unmapped_area(file, addr, len, pgoff, flags); + return mm_get_unmapped_area(current->mm, file, addr, len, pgoff, flags); } #endif diff --git a/mm/util.c b/mm/util.c index a9e911b22b99..c9e519e6811f 100644 --- a/mm/util.c +++ b/mm/util.c @@ -469,17 +469,17 @@ void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack) if (mmap_is_legacy(rlim_stack)) { mm->mmap_base = TASK_UNMAPPED_BASE + random_factor; - mm->get_unmapped_area = arch_get_unmapped_area; + clear_bit(MMF_TOPDOWN, &mm->flags); } else { mm->mmap_base = mmap_base(random_factor, rlim_stack); - mm->get_unmapped_area = arch_get_unmapped_area_topdown; + set_bit(MMF_TOPDOWN, &mm->flags); } } #elif defined(CONFIG_MMU) && !defined(HAVE_ARCH_PICK_MMAP_LAYOUT) void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack) { mm->mmap_base = TASK_UNMAPPED_BASE; - mm->get_unmapped_area = arch_get_unmapped_area; + clear_bit(MMF_TOPDOWN, &mm->flags); } #endif -- cgit v1.2.3-59-g8ed1b From b80fa3cbb78c0fbe5039682919d97a0dbe05ae7c Mon Sep 17 00:00:00 2001 From: Rick Edgecombe Date: Mon, 25 Mar 2024 19:16:52 -0700 Subject: treewide: use initializer for struct vm_unmapped_area_info Future changes will need to add a new member to struct vm_unmapped_area_info. This would cause trouble for any call site that doesn't initialize the struct. Currently every caller sets each member manually, so if new ones are added they will be uninitialized and the core code parsing the struct will see garbage in the new member. It could be possible to initialize the new member manually to 0 at each call site. This and a couple other options were discussed. Having some struct vm_unmapped_area_info instances not zero initialized will put those sites at risk of feeding garbage into vm_unmapped_area(), if the convention is to zero initialize the struct and any new field addition missed a call site that initializes each field manually. So it is useful to do things similar across the kernel. The consensus (see links) was that in general the best way to accomplish taking into account both code cleanliness and minimizing the chance of introducing bugs, was to do C99 static initialization. As in: struct vm_unmapped_area_info info = {}; With this method of initialization, the whole struct will be zero initialized, and any statements setting fields to zero will be unneeded. The change should not leave cleanup at the call sides. While iterating though the possible solutions a few archs kindly acked other variations that still zero initialized the struct. These sites have been modified in previous changes using the pattern acked by the respective arch. So to be reduce the chance of bugs via uninitialized fields, perform a tree wide change using the consensus for the best general way to do this change. Use C99 static initializing to zero the struct and remove and statements that simply set members to zero. Link: https://lkml.kernel.org/r/20240326021656.202649-11-rick.p.edgecombe@intel.com Link: https://lore.kernel.org/lkml/202402280912.33AEE7A9CF@keescook/#t Link: https://lore.kernel.org/lkml/j7bfvig3gew3qruouxrh7z7ehjjafrgkbcmg6tcghhfh3rhmzi@wzlcoecgy5rs/ Link: https://lore.kernel.org/lkml/ec3e377a-c0a0-4dd3-9cb9-96517e54d17e@csgroup.eu/ Signed-off-by: Rick Edgecombe Reviewed-by: Kees Cook Cc: Alexei Starovoitov Cc: Andy Lutomirski Cc: Aneesh Kumar K.V Cc: Borislav Petkov (AMD) Cc: Christophe Leroy Cc: Dan Williams Cc: Dave Hansen Cc: Deepak Gupta Cc: Guo Ren Cc: Helge Deller Cc: H. Peter Anvin (Intel) Cc: Ingo Molnar Cc: "James E.J. Bottomley" Cc: Kirill A. Shutemov Cc: Liam R. Howlett Cc: Mark Brown Cc: Michael Ellerman Cc: Naveen N. Rao Cc: Nicholas Piggin Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- arch/alpha/kernel/osf_sys.c | 5 +---- arch/arc/mm/mmap.c | 4 +--- arch/arm/mm/mmap.c | 5 ++--- arch/loongarch/mm/mmap.c | 3 +-- arch/mips/mm/mmap.c | 3 +-- arch/s390/mm/hugetlbpage.c | 7 ++----- arch/s390/mm/mmap.c | 5 ++--- arch/sh/mm/mmap.c | 5 ++--- arch/sparc/kernel/sys_sparc_32.c | 3 +-- arch/sparc/kernel/sys_sparc_64.c | 5 ++--- arch/sparc/mm/hugetlbpage.c | 7 ++----- arch/x86/kernel/sys_x86_64.c | 7 ++----- arch/x86/mm/hugetlbpage.c | 7 ++----- fs/hugetlbfs/inode.c | 7 ++----- mm/mmap.c | 9 ++------- 15 files changed, 25 insertions(+), 57 deletions(-) (limited to 'fs') diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 5db88b627439..e5f881bc8288 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -1218,14 +1218,11 @@ static unsigned long arch_get_unmapped_area_1(unsigned long addr, unsigned long len, unsigned long limit) { - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; - info.flags = 0; info.length = len; info.low_limit = addr; info.high_limit = limit; - info.align_mask = 0; - info.align_offset = 0; return vm_unmapped_area(&info); } diff --git a/arch/arc/mm/mmap.c b/arch/arc/mm/mmap.c index 3c1c7ae73292..69a915297155 100644 --- a/arch/arc/mm/mmap.c +++ b/arch/arc/mm/mmap.c @@ -27,7 +27,7 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, { struct mm_struct *mm = current->mm; struct vm_area_struct *vma; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; /* * We enforce the MAP_FIXED case. @@ -51,11 +51,9 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, return addr; } - info.flags = 0; info.length = len; info.low_limit = mm->mmap_base; info.high_limit = TASK_SIZE; - info.align_mask = 0; info.align_offset = pgoff << PAGE_SHIFT; return vm_unmapped_area(&info); } diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c index a0f8a0ca0788..d65d0e6ed10a 100644 --- a/arch/arm/mm/mmap.c +++ b/arch/arm/mm/mmap.c @@ -34,7 +34,7 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, struct vm_area_struct *vma; int do_align = 0; int aliasing = cache_is_vipt_aliasing(); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; /* * We only need to do colour alignment if either the I or D @@ -68,7 +68,6 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, return addr; } - info.flags = 0; info.length = len; info.low_limit = mm->mmap_base; info.high_limit = TASK_SIZE; @@ -87,7 +86,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, unsigned long addr = addr0; int do_align = 0; int aliasing = cache_is_vipt_aliasing(); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; /* * We only need to do colour alignment if either the I or D diff --git a/arch/loongarch/mm/mmap.c b/arch/loongarch/mm/mmap.c index 89af7c12e8c0..889030985135 100644 --- a/arch/loongarch/mm/mmap.c +++ b/arch/loongarch/mm/mmap.c @@ -25,7 +25,7 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp, struct vm_area_struct *vma; unsigned long addr = addr0; int do_color_align; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (unlikely(len > TASK_SIZE)) return -ENOMEM; @@ -83,7 +83,6 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp, */ } - info.flags = 0; info.low_limit = mm->mmap_base; info.high_limit = TASK_SIZE; return vm_unmapped_area(&info); diff --git a/arch/mips/mm/mmap.c b/arch/mips/mm/mmap.c index 00fe90c6db3e..7e11d7b58761 100644 --- a/arch/mips/mm/mmap.c +++ b/arch/mips/mm/mmap.c @@ -34,7 +34,7 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp, struct vm_area_struct *vma; unsigned long addr = addr0; int do_color_align; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (unlikely(len > TASK_SIZE)) return -ENOMEM; @@ -92,7 +92,6 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp, */ } - info.flags = 0; info.low_limit = mm->mmap_base; info.high_limit = TASK_SIZE; return vm_unmapped_area(&info); diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index 7d948e243f4b..e1e63dc1b23d 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -248,14 +248,12 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file, unsigned long pgoff, unsigned long flags) { struct hstate *h = hstate_file(file); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; - info.flags = 0; info.length = len; info.low_limit = current->mm->mmap_base; info.high_limit = TASK_SIZE; info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; return vm_unmapped_area(&info); } @@ -264,7 +262,7 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, unsigned long pgoff, unsigned long flags) { struct hstate *h = hstate_file(file); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; unsigned long addr; info.flags = VM_UNMAPPED_AREA_TOPDOWN; @@ -272,7 +270,6 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, info.low_limit = PAGE_SIZE; info.high_limit = current->mm->mmap_base; info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; addr = vm_unmapped_area(&info); /* diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c index 6b2e4436ad4a..206756946589 100644 --- a/arch/s390/mm/mmap.c +++ b/arch/s390/mm/mmap.c @@ -86,7 +86,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, { struct mm_struct *mm = current->mm; struct vm_area_struct *vma; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (len > TASK_SIZE - mmap_min_addr) return -ENOMEM; @@ -102,7 +102,6 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, goto check_asce_limit; } - info.flags = 0; info.length = len; info.low_limit = mm->mmap_base; info.high_limit = TASK_SIZE; @@ -122,7 +121,7 @@ unsigned long arch_get_unmapped_area_topdown(struct file *filp, unsigned long ad { struct vm_area_struct *vma; struct mm_struct *mm = current->mm; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; /* requested length too big for entire address space */ if (len > TASK_SIZE - mmap_min_addr) diff --git a/arch/sh/mm/mmap.c b/arch/sh/mm/mmap.c index b82199878b45..bee329d4149a 100644 --- a/arch/sh/mm/mmap.c +++ b/arch/sh/mm/mmap.c @@ -57,7 +57,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, struct mm_struct *mm = current->mm; struct vm_area_struct *vma; int do_colour_align; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (flags & MAP_FIXED) { /* We do not accept a shared mapping if it would violate @@ -88,7 +88,6 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, return addr; } - info.flags = 0; info.length = len; info.low_limit = TASK_UNMAPPED_BASE; info.high_limit = TASK_SIZE; @@ -106,7 +105,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, struct mm_struct *mm = current->mm; unsigned long addr = addr0; int do_colour_align; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (flags & MAP_FIXED) { /* We do not accept a shared mapping if it would violate diff --git a/arch/sparc/kernel/sys_sparc_32.c b/arch/sparc/kernel/sys_sparc_32.c index 082a551897ed..08a19727795c 100644 --- a/arch/sparc/kernel/sys_sparc_32.c +++ b/arch/sparc/kernel/sys_sparc_32.c @@ -41,7 +41,7 @@ SYSCALL_DEFINE0(getpagesize) unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (flags & MAP_FIXED) { /* We do not accept a shared mapping if it would violate @@ -59,7 +59,6 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi if (!addr) addr = TASK_UNMAPPED_BASE; - info.flags = 0; info.length = len; info.low_limit = addr; info.high_limit = TASK_SIZE; diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c index 1dbf7211666e..d9c3b34ca744 100644 --- a/arch/sparc/kernel/sys_sparc_64.c +++ b/arch/sparc/kernel/sys_sparc_64.c @@ -93,7 +93,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi struct vm_area_struct * vma; unsigned long task_size = TASK_SIZE; int do_color_align; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (flags & MAP_FIXED) { /* We do not accept a shared mapping if it would violate @@ -126,7 +126,6 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi return addr; } - info.flags = 0; info.length = len; info.low_limit = TASK_UNMAPPED_BASE; info.high_limit = min(task_size, VA_EXCLUDE_START); @@ -154,7 +153,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, unsigned long task_size = STACK_TOP32; unsigned long addr = addr0; int do_color_align; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; /* This should only ever run for 32-bit processes. */ BUG_ON(!test_thread_flag(TIF_32BIT)); diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c index c23012e3a353..cc91ca7a1e18 100644 --- a/arch/sparc/mm/hugetlbpage.c +++ b/arch/sparc/mm/hugetlbpage.c @@ -31,17 +31,15 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp, { struct hstate *h = hstate_file(filp); unsigned long task_size = TASK_SIZE; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; if (test_thread_flag(TIF_32BIT)) task_size = STACK_TOP32; - info.flags = 0; info.length = len; info.low_limit = TASK_UNMAPPED_BASE; info.high_limit = min(task_size, VA_EXCLUDE_START); info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; addr = vm_unmapped_area(&info); if ((addr & ~PAGE_MASK) && task_size > VA_EXCLUDE_END) { @@ -63,7 +61,7 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, struct hstate *h = hstate_file(filp); struct mm_struct *mm = current->mm; unsigned long addr = addr0; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; /* This should only ever run for 32-bit processes. */ BUG_ON(!test_thread_flag(TIF_32BIT)); @@ -73,7 +71,6 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, info.low_limit = PAGE_SIZE; info.high_limit = mm->mmap_base; info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; addr = vm_unmapped_area(&info); /* diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c index cb9fa1d5c66f..96b9d29aead0 100644 --- a/arch/x86/kernel/sys_x86_64.c +++ b/arch/x86/kernel/sys_x86_64.c @@ -118,7 +118,7 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, { struct mm_struct *mm = current->mm; struct vm_area_struct *vma; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; unsigned long begin, end; if (flags & MAP_FIXED) @@ -137,11 +137,9 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, return addr; } - info.flags = 0; info.length = len; info.low_limit = begin; info.high_limit = end; - info.align_mask = 0; info.align_offset = pgoff << PAGE_SHIFT; if (filp) { info.align_mask = get_align_mask(); @@ -158,7 +156,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, struct vm_area_struct *vma; struct mm_struct *mm = current->mm; unsigned long addr = addr0; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; /* requested length too big for entire address space */ if (len > TASK_SIZE) @@ -203,7 +201,6 @@ get_unmapped_area: if (addr > DEFAULT_MAP_WINDOW && !in_32bit_syscall()) info.high_limit += TASK_SIZE_MAX - DEFAULT_MAP_WINDOW; - info.align_mask = 0; info.align_offset = pgoff << PAGE_SHIFT; if (filp) { info.align_mask = get_align_mask(); diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index 06ca9a60bac2..807a5859a3c4 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -25,9 +25,8 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file, unsigned long pgoff, unsigned long flags) { struct hstate *h = hstate_file(file); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; - info.flags = 0; info.length = len; info.low_limit = get_mmap_base(1); @@ -39,7 +38,6 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file, task_size_32bit() : task_size_64bit(addr > DEFAULT_MAP_WINDOW); info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; return vm_unmapped_area(&info); } @@ -48,7 +46,7 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, unsigned long pgoff, unsigned long flags) { struct hstate *h = hstate_file(file); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; info.flags = VM_UNMAPPED_AREA_TOPDOWN; info.length = len; @@ -63,7 +61,6 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, info.high_limit += TASK_SIZE_MAX - DEFAULT_MAP_WINDOW; info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; addr = vm_unmapped_area(&info); /* diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 3dee18bf47ed..2f4e88552d3f 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -176,14 +176,12 @@ hugetlb_get_unmapped_area_bottomup(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { struct hstate *h = hstate_file(file); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; - info.flags = 0; info.length = len; info.low_limit = current->mm->mmap_base; info.high_limit = arch_get_mmap_end(addr, len, flags); info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; return vm_unmapped_area(&info); } @@ -192,14 +190,13 @@ hugetlb_get_unmapped_area_topdown(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { struct hstate *h = hstate_file(file); - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; info.flags = VM_UNMAPPED_AREA_TOPDOWN; info.length = len; info.low_limit = PAGE_SIZE; info.high_limit = arch_get_mmap_base(addr, current->mm->mmap_base); info.align_mask = PAGE_MASK & ~huge_page_mask(h); - info.align_offset = 0; addr = vm_unmapped_area(&info); /* diff --git a/mm/mmap.c b/mm/mmap.c index 0ef1191e3be5..4ad386f3f63f 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1704,7 +1704,7 @@ generic_get_unmapped_area(struct file *filp, unsigned long addr, { struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; const unsigned long mmap_end = arch_get_mmap_end(addr, len, flags); if (len > mmap_end - mmap_min_addr) @@ -1722,12 +1722,9 @@ generic_get_unmapped_area(struct file *filp, unsigned long addr, return addr; } - info.flags = 0; info.length = len; info.low_limit = mm->mmap_base; info.high_limit = mmap_end; - info.align_mask = 0; - info.align_offset = 0; return vm_unmapped_area(&info); } @@ -1752,7 +1749,7 @@ generic_get_unmapped_area_topdown(struct file *filp, unsigned long addr, { struct vm_area_struct *vma, *prev; struct mm_struct *mm = current->mm; - struct vm_unmapped_area_info info; + struct vm_unmapped_area_info info = {}; const unsigned long mmap_end = arch_get_mmap_end(addr, len, flags); /* requested length too big for entire address space */ @@ -1776,8 +1773,6 @@ generic_get_unmapped_area_topdown(struct file *filp, unsigned long addr, info.length = len; info.low_limit = PAGE_SIZE; info.high_limit = arch_get_mmap_base(addr, mm->mmap_base); - info.align_mask = 0; - info.align_offset = 0; addr = vm_unmapped_area(&info); /* -- cgit v1.2.3-59-g8ed1b From 3a9e567ca45fb5280065283d10d9a11f0db61d2b Mon Sep 17 00:00:00 2001 From: Jinjiang Tu Date: Thu, 28 Mar 2024 19:10:08 +0800 Subject: mm/ksm: fix ksm exec support for prctl Patch series "mm/ksm: fix ksm exec support for prctl", v4. commit 3c6f33b7273a ("mm/ksm: support fork/exec for prctl") inherits MMF_VM_MERGE_ANY flag when a task calls execve(). However, it doesn't create the mm_slot, so ksmd will not try to scan this task. The first patch fixes the issue. The second patch refactors to prepare for the third patch. The third patch extends the selftests of ksm to verfity the deduplication really happens after fork/exec inherits ths KSM setting. This patch (of 3): commit 3c6f33b7273a ("mm/ksm: support fork/exec for prctl") inherits MMF_VM_MERGE_ANY flag when a task calls execve(). Howerver, it doesn't create the mm_slot, so ksmd will not try to scan this task. To fix it, allocate and add the mm_slot to ksm_mm_head in __bprm_mm_init() when the mm has MMF_VM_MERGE_ANY flag. Link: https://lkml.kernel.org/r/20240328111010.1502191-1-tujinjiang@huawei.com Link: https://lkml.kernel.org/r/20240328111010.1502191-2-tujinjiang@huawei.com Fixes: 3c6f33b7273a ("mm/ksm: support fork/exec for prctl") Signed-off-by: Jinjiang Tu Reviewed-by: David Hildenbrand Cc: Johannes Weiner Cc: Kefeng Wang Cc: Nanyong Sun Cc: Rik van Riel Cc: Stefan Roesch Signed-off-by: Andrew Morton --- fs/exec.c | 11 +++++++++++ include/linux/ksm.h | 13 +++++++++++++ 2 files changed, 24 insertions(+) (limited to 'fs') diff --git a/fs/exec.c b/fs/exec.c index cf1df7f16e55..0c5f06d08c35 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -67,6 +67,7 @@ #include #include #include +#include #include #include @@ -267,6 +268,14 @@ static int __bprm_mm_init(struct linux_binprm *bprm) goto err_free; } + /* + * Need to be called with mmap write lock + * held, to avoid race with ksmd. + */ + err = ksm_execve(mm); + if (err) + goto err_ksm; + /* * Place the stack at the largest stack address the architecture * supports. Later, we'll move this to an appropriate place. We don't @@ -288,6 +297,8 @@ static int __bprm_mm_init(struct linux_binprm *bprm) bprm->p = vma->vm_end - sizeof(void *); return 0; err: + ksm_exit(mm); +err_ksm: mmap_write_unlock(mm); err_free: bprm->vma = NULL; diff --git a/include/linux/ksm.h b/include/linux/ksm.h index 401348e9f92b..7e2b1de3996a 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -59,6 +59,14 @@ static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) return 0; } +static inline int ksm_execve(struct mm_struct *mm) +{ + if (test_bit(MMF_VM_MERGE_ANY, &mm->flags)) + return __ksm_enter(mm); + + return 0; +} + static inline void ksm_exit(struct mm_struct *mm) { if (test_bit(MMF_VM_MERGEABLE, &mm->flags)) @@ -107,6 +115,11 @@ static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) return 0; } +static inline int ksm_execve(struct mm_struct *mm) +{ + return 0; +} + static inline void ksm_exit(struct mm_struct *mm) { } -- cgit v1.2.3-59-g8ed1b From 03aa577f3b2809bcaeb9a544307c2d997a972965 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 2 Apr 2024 21:12:48 +0100 Subject: proc: convert clear_refs_pte_range to use a folio Patch series "Remove page_idle and page_young wrappers". There are only a couple of places left using the page wrappers for idle & young tracking. Convert the two users in proc and then we can remove the wrappers. That enables the further simplification of autogenerating the definitions when CONFIG_PAGE_IDLE_FLAG is disabled. This patch (of 4): Replaces four calls to compound_head() with two. Link: https://lkml.kernel.org/r/20240402201252.917342-1-willy@infradead.org Link: https://lkml.kernel.org/r/20240402201252.917342-2-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Hildenbrand Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 23fbab954c20..b94101cd2706 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1161,7 +1161,7 @@ static int clear_refs_pte_range(pmd_t *pmd, unsigned long addr, struct vm_area_struct *vma = walk->vma; pte_t *pte, ptent; spinlock_t *ptl; - struct page *page; + struct folio *folio; ptl = pmd_trans_huge_lock(pmd, vma); if (ptl) { @@ -1173,12 +1173,12 @@ static int clear_refs_pte_range(pmd_t *pmd, unsigned long addr, if (!pmd_present(*pmd)) goto out; - page = pmd_page(*pmd); + folio = pmd_folio(*pmd); /* Clear accessed and referenced bits. */ pmdp_test_and_clear_young(vma, addr, pmd); - test_and_clear_page_young(page); - ClearPageReferenced(page); + folio_test_clear_young(folio); + folio_clear_referenced(folio); out: spin_unlock(ptl); return 0; @@ -1200,14 +1200,14 @@ out: if (!pte_present(ptent)) continue; - page = vm_normal_page(vma, addr, ptent); - if (!page) + folio = vm_normal_folio(vma, addr, ptent); + if (!folio) continue; /* Clear accessed and referenced bits. */ ptep_test_and_clear_young(vma, addr, pte); - test_and_clear_page_young(page); - ClearPageReferenced(page); + folio_test_clear_young(folio); + folio_clear_referenced(folio); } pte_unmap_unlock(pte - 1, ptl); cond_resched(); -- cgit v1.2.3-59-g8ed1b From 6c977f36dc36baec8ad15faccbc6a08315eaa195 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 2 Apr 2024 21:12:49 +0100 Subject: proc: convert smaps_account() to use a folio Replace seven calls to compound_head() with one. Link: https://lkml.kernel.org/r/20240402201252.917342-3-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Hildenbrand Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index b94101cd2706..e8d1008a838d 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -444,6 +444,7 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page, bool compound, bool young, bool dirty, bool locked, bool migration) { + struct folio *folio = page_folio(page); int i, nr = compound ? compound_nr(page) : 1; unsigned long size = nr * PAGE_SIZE; @@ -451,27 +452,28 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page, * First accumulate quantities that depend only on |size| and the type * of the compound page. */ - if (PageAnon(page)) { + if (folio_test_anon(folio)) { mss->anonymous += size; - if (!PageSwapBacked(page) && !dirty && !PageDirty(page)) + if (!folio_test_swapbacked(folio) && !dirty && + !folio_test_dirty(folio)) mss->lazyfree += size; } - if (PageKsm(page)) + if (folio_test_ksm(folio)) mss->ksm += size; mss->resident += size; /* Accumulate the size in pages that have been accessed. */ - if (young || page_is_young(page) || PageReferenced(page)) + if (young || folio_test_young(folio) || folio_test_referenced(folio)) mss->referenced += size; /* * Then accumulate quantities that may depend on sharing, or that may * differ page-by-page. * - * page_count(page) == 1 guarantees the page is mapped exactly once. + * refcount == 1 guarantees the page is mapped exactly once. * If any subpage of the compound page mapped with PTE it would elevate - * page_count(). + * the refcount. * * The page_mapcount() is called to get a snapshot of the mapcount. * Without holding the page lock this snapshot can be slightly wrong as @@ -480,7 +482,7 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page, * especially for migration entries. Treat regular migration entries * as mapcount == 1. */ - if ((page_count(page) == 1) || migration) { + if ((folio_ref_count(folio) == 1) || migration) { smaps_page_accumulate(mss, page, size, size << PSS_SHIFT, dirty, locked, true); return; -- cgit v1.2.3-59-g8ed1b From f1dc623fa0d37eeb02b31264dd2ce2011fec0c9b Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:14:52 +0100 Subject: proc: convert gather_stats to use a folio Patch series "Use folio APIs in procfs". We're down to very few users of the PageFoo macros, with proc being a major user. After this patchset and another patchset I have for khugepaged, we can get rid of PageActive, PageReadahead and PageSwapBacked. This patchset has the usual advantages in its own right of removing hidden calls to compound_head(). We have the page table lock, so the mapcount & refcount are stable and there can't be any races with folios suddenly becoming tail pages. This patch (of 4): Replaces six calls to compound_head() with one. Shrinks the function from 5054 bytes to 1756 bytes in an allmodconfig build. Link: https://lkml.kernel.org/r/20240403171456.1445117-1-willy@infradead.org Link: https://lkml.kernel.org/r/20240403171456.1445117-2-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Cc: Christian Brauner Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index e8d1008a838d..5260a2788f74 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -2549,28 +2549,29 @@ struct numa_maps_private { static void gather_stats(struct page *page, struct numa_maps *md, int pte_dirty, unsigned long nr_pages) { + struct folio *folio = page_folio(page); int count = page_mapcount(page); md->pages += nr_pages; - if (pte_dirty || PageDirty(page)) + if (pte_dirty || folio_test_dirty(folio)) md->dirty += nr_pages; - if (PageSwapCache(page)) + if (folio_test_swapcache(folio)) md->swapcache += nr_pages; - if (PageActive(page) || PageUnevictable(page)) + if (folio_test_active(folio) || folio_test_unevictable(folio)) md->active += nr_pages; - if (PageWriteback(page)) + if (folio_test_writeback(folio)) md->writeback += nr_pages; - if (PageAnon(page)) + if (folio_test_anon(folio)) md->anon += nr_pages; if (count > md->mapcount_max) md->mapcount_max = count; - md->node[page_to_nid(page)] += nr_pages; + md->node[folio_nid(folio)] += nr_pages; } static struct page *can_gather_numa_stats(pte_t pte, struct vm_area_struct *vma, -- cgit v1.2.3-59-g8ed1b From cfc96da432fe5c6b514bdbf60ee036cad39a65af Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:14:53 +0100 Subject: proc: convert smaps_page_accumulate to use a folio Replaces three calls to compound_head() with one. Shrinks the function from 2614 bytes to 1112 bytes in an allmodconfig build. Link: https://lkml.kernel.org/r/20240403171456.1445117-3-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Cc: Christian Brauner Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 5260a2788f74..2a3133dd47b1 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -414,11 +414,12 @@ static void smaps_page_accumulate(struct mem_size_stats *mss, struct page *page, unsigned long size, unsigned long pss, bool dirty, bool locked, bool private) { + struct folio *folio = page_folio(page); mss->pss += pss; - if (PageAnon(page)) + if (folio_test_anon(folio)) mss->pss_anon += pss; - else if (PageSwapBacked(page)) + else if (folio_test_swapbacked(folio)) mss->pss_shmem += pss; else mss->pss_file += pss; @@ -426,7 +427,7 @@ static void smaps_page_accumulate(struct mem_size_stats *mss, if (locked) mss->pss_locked += pss; - if (dirty || PageDirty(page)) { + if (dirty || folio_test_dirty(folio)) { mss->pss_dirty += pss; if (private) mss->private_dirty += size; -- cgit v1.2.3-59-g8ed1b From 27bb0a70e5248683fd940f738a7a53066fb470e5 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:14:54 +0100 Subject: proc: pass a folio to smaps_page_accumulate() Both callers already have a folio; pass it in instead of doing the conversion each time. Link: https://lkml.kernel.org/r/20240403171456.1445117-4-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Cc: Christian Brauner Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 2a3133dd47b1..6d4f60bc8824 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -411,10 +411,9 @@ struct mem_size_stats { }; static void smaps_page_accumulate(struct mem_size_stats *mss, - struct page *page, unsigned long size, unsigned long pss, + struct folio *folio, unsigned long size, unsigned long pss, bool dirty, bool locked, bool private) { - struct folio *folio = page_folio(page); mss->pss += pss; if (folio_test_anon(folio)) @@ -484,8 +483,8 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page, * as mapcount == 1. */ if ((folio_ref_count(folio) == 1) || migration) { - smaps_page_accumulate(mss, page, size, size << PSS_SHIFT, dirty, - locked, true); + smaps_page_accumulate(mss, folio, size, size << PSS_SHIFT, + dirty, locked, true); return; } for (i = 0; i < nr; i++, page++) { @@ -493,8 +492,8 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page, unsigned long pss = PAGE_SIZE << PSS_SHIFT; if (mapcount >= 2) pss /= mapcount; - smaps_page_accumulate(mss, page, PAGE_SIZE, pss, dirty, locked, - mapcount < 2); + smaps_page_accumulate(mss, folio, PAGE_SIZE, pss, + dirty, locked, mapcount < 2); } } -- cgit v1.2.3-59-g8ed1b From 039d26d10d62b71e242cde1092b042ae1fee498d Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:14:55 +0100 Subject: proc: convert smaps_pmd_entry to use a folio Replace two calls to compound_head() with one. Link: https://lkml.kernel.org/r/20240403171456.1445117-5-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Cc: Christian Brauner Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 6d4f60bc8824..8ff79bd427ec 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -578,6 +578,7 @@ static void smaps_pmd_entry(pmd_t *pmd, unsigned long addr, struct vm_area_struct *vma = walk->vma; bool locked = !!(vma->vm_flags & VM_LOCKED); struct page *page = NULL; + struct folio *folio; bool migration = false; if (pmd_present(*pmd)) { @@ -592,11 +593,12 @@ static void smaps_pmd_entry(pmd_t *pmd, unsigned long addr, } if (IS_ERR_OR_NULL(page)) return; - if (PageAnon(page)) + folio = page_folio(page); + if (folio_test_anon(folio)) mss->anonymous_thp += HPAGE_PMD_SIZE; - else if (PageSwapBacked(page)) + else if (folio_test_swapbacked(folio)) mss->shmem_thp += HPAGE_PMD_SIZE; - else if (is_zone_device_page(page)) + else if (folio_is_zone_device(folio)) /* pass */; else mss->file_thp += HPAGE_PMD_SIZE; -- cgit v1.2.3-59-g8ed1b From 30dd3478c3cd7d01cc5afc4952e885ba4eefb730 Mon Sep 17 00:00:00 2001 From: Joseph Qi Date: Thu, 14 Mar 2024 10:17:13 +0800 Subject: ocfs2: correctly use ocfs2_find_next_zero_bit() If no bits are zero, ocfs2_find_next_zero_bit() will return max size, so check the return value with -1 is meaningless. Correct this usage and cleanup the code. Link: https://lkml.kernel.org/r/20240314021713.240796-1-joseph.qi@linux.alibaba.com Signed-off-by: Joseph Qi Reviewed-by: Heming Zhao Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Gang He Cc: Jun Piao Signed-off-by: Andrew Morton --- fs/ocfs2/localalloc.c | 19 ++++++------------- fs/ocfs2/reservations.c | 2 +- fs/ocfs2/suballoc.c | 6 ++---- 3 files changed, 9 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index c803c10dd97e..33aeaaa056d7 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -863,14 +863,8 @@ static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb, numfound = bitoff = startoff = 0; left = le32_to_cpu(alloc->id1.bitmap1.i_total); - while ((bitoff = ocfs2_find_next_zero_bit(bitmap, left, startoff)) != -1) { - if (bitoff == left) { - /* mlog(0, "bitoff (%d) == left", bitoff); */ - break; - } - /* mlog(0, "Found a zero: bitoff = %d, startoff = %d, " - "numfound = %d\n", bitoff, startoff, numfound);*/ - + while ((bitoff = ocfs2_find_next_zero_bit(bitmap, left, startoff)) < + left) { /* Ok, we found a zero bit... is it contig. or do we * start over?*/ if (bitoff == startoff) { @@ -976,9 +970,9 @@ static int ocfs2_sync_local_to_main(struct ocfs2_super *osb, start = count = 0; left = le32_to_cpu(alloc->id1.bitmap1.i_total); - while ((bit_off = ocfs2_find_next_zero_bit(bitmap, left, start)) - != -1) { - if ((bit_off < left) && (bit_off == start)) { + while ((bit_off = ocfs2_find_next_zero_bit(bitmap, left, start)) < + left) { + if (bit_off == start) { count++; start++; continue; @@ -1002,8 +996,7 @@ static int ocfs2_sync_local_to_main(struct ocfs2_super *osb, goto bail; } } - if (bit_off >= left) - break; + count = 1; start = bit_off + 1; } diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c index a9d1296d736d..1fe61974d9f0 100644 --- a/fs/ocfs2/reservations.c +++ b/fs/ocfs2/reservations.c @@ -414,7 +414,7 @@ static int ocfs2_resmap_find_free_bits(struct ocfs2_reservation_map *resmap, start = search_start; while ((offset = ocfs2_find_next_zero_bit(bitmap, resmap->m_bitmap_len, - start)) != -1) { + start)) < resmap->m_bitmap_len) { /* Search reached end of the region */ if (offset >= (search_start + search_len)) break; diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 166c8918c825..961998415308 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -1290,10 +1290,8 @@ static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, found = start = best_offset = best_size = 0; bitmap = bg->bg_bitmap; - while((offset = ocfs2_find_next_zero_bit(bitmap, total_bits, start)) != -1) { - if (offset == total_bits) - break; - + while ((offset = ocfs2_find_next_zero_bit(bitmap, total_bits, start)) < + total_bits) { if (!ocfs2_test_bg_bit_allocatable(bg_bh, offset)) { /* We found a zero, but we can't use it as it * hasn't been put to disk yet! */ -- cgit v1.2.3-59-g8ed1b From c9abe099865b2d0df66346657b76922d8efd43e4 Mon Sep 17 00:00:00 2001 From: Su Yue Date: Mon, 18 Mar 2024 19:56:09 +0800 Subject: ocfs2: update inode ctime in ocfs2_fileattr_set inode ctime should be updated if ocfs2_fileattr_set is called. Link: https://lkml.kernel.org/r/20240318115609.3194-1-l@damenly.org Signed-off-by: Su Yue Reviewed-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Gang He Cc: Jun Piao Signed-off-by: Andrew Morton --- fs/ocfs2/ioctl.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index b1550ba73f96..71beef7f8a60 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -125,6 +125,7 @@ int ocfs2_fileattr_set(struct mnt_idmap *idmap, ocfs2_inode->ip_attr = flags; ocfs2_set_inode_flags(inode); + inode_set_ctime_current(inode); status = ocfs2_mark_inode_dirty(handle, inode, bh); if (status < 0) -- cgit v1.2.3-59-g8ed1b From 4d9784c00a15631b68a5d95be907aa308f269551 Mon Sep 17 00:00:00 2001 From: Yang Li Date: Fri, 22 Mar 2024 15:37:24 +0800 Subject: fs: add kernel-doc comments to fat_parse_long() Add kernel-doc style comments with complete parameter descriptions for fat_parse_long(). Link: https://lkml.kernel.org/r/20240322073724.102332-1-yang.lee@linux.alibaba.com Signed-off-by: Yang Li Acked-by: OGAWA Hirofumi Signed-off-by: Andrew Morton --- fs/fat/dir.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'fs') diff --git a/fs/fat/dir.c b/fs/fat/dir.c index 00235b8a1823..acbec5bdd521 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -269,6 +269,18 @@ enum { PARSE_INVALID = 1, PARSE_NOT_LONGNAME, PARSE_EOF, }; /** * fat_parse_long - Parse extended directory entry. * + * @dir: Pointer to the inode that represents the directory. + * @pos: On input, contains the starting position to read from. + * On output, updated with the new position. + * @bh: Pointer to the buffer head that may be used for reading directory + * entries. May be updated. + * @de: On input, points to the current directory entry. + * On output, points to the next directory entry. + * @unicode: Pointer to a buffer where the parsed Unicode long filename will be + * stored. + * @nr_slots: Pointer to a variable that will store the number of longname + * slots found. + * * This function returns zero on success, negative value on error, or one of * the following: * -- cgit v1.2.3-59-g8ed1b From 6b839b3b76cf17296ebd4a893841f32cae08229c Mon Sep 17 00:00:00 2001 From: Douglas Anderson Date: Mon, 5 Feb 2024 09:26:30 -0800 Subject: regset: use kvzalloc() for regset_get_alloc() While browsing through ChromeOS crash reports, I found one with an allocation failure that looked like this: chrome: page allocation failure: order:7, mode:0x40dc0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO), nodemask=(null),cpuset=urgent,mems_allowed=0 CPU: 7 PID: 3295 Comm: chrome Not tainted 5.15.133-20574-g8044615ac35c #1 (HASH:1162 1) Hardware name: Google Lazor (rev3 - 8) with KB Backlight (DT) Call trace: ... warn_alloc+0x104/0x174 __alloc_pages+0x5f0/0x6e4 kmalloc_order+0x44/0x98 kmalloc_order_trace+0x34/0x124 __kmalloc+0x228/0x36c __regset_get+0x68/0xcc regset_get_alloc+0x1c/0x28 elf_core_dump+0x3d8/0xd8c do_coredump+0xeb8/0x1378 get_signal+0x14c/0x804 ... An order 7 allocation is (1 << 7) contiguous pages, or 512K. It's not a surprise that this allocation failed on a system that's been running for a while. More digging showed that it was fairly easy to see the order 7 allocation by just sending a SIGQUIT to chrome (or other processes) to generate a core dump. The actual amount being allocated was 279,584 bytes and it was for "core_note_type" NT_ARM_SVE. There was quite a bit of discussion [1] on the mailing lists in response to my v1 patch attempting to switch to vmalloc. The overall conclusion was that we could likely reduce the 279,584 byte allocation by quite a bit and Mark Brown has sent a patch to that effect [2]. However even with the 279,584 byte allocation gone there are still 65,552 byte allocations. These are just barely more than the 65,536 bytes and thus would require an order 5 allocation. An order 5 allocation is still something to avoid unless necessary and nothing needs the memory here to be contiguous. Change the allocation to kvzalloc() which should still be efficient for small allocations but doesn't force the memory subsystem to work hard (and maybe fail) at getting a large contiguous chunk. [1] https://lore.kernel.org/r/20240201171159.1.Id9ad163b60d21c9e56c2d686b0cc9083a8ba7924@changeid [2] https://lore.kernel.org/r/20240203-arm64-sve-ptrace-regset-size-v1-1-2c3ba1386b9e@kernel.org Link: https://lkml.kernel.org/r/20240205092626.v2.1.Id9ad163b60d21c9e56c2d686b0cc9083a8ba7924@changeid Signed-off-by: Douglas Anderson Reviewed-by: Catalin Marinas Cc: Al Viro Cc: Christian Brauner Cc: Dave Martin Cc: Eric Biederman Cc: Jan Kara Cc: Kees Cook Cc: Mark Brown Cc: Matthew Wilcox (Oracle) Cc: Oleg Nesterov Cc: Will Deacon Signed-off-by: Andrew Morton --- fs/binfmt_elf.c | 2 +- kernel/regset.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 5397b552fbeb..ac178ad38823 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1928,7 +1928,7 @@ static void free_note_info(struct elf_note_info *info) threads = t->next; WARN_ON(t->notes[0].data && t->notes[0].data != &t->prstatus); for (i = 1; i < info->thread_notes; ++i) - kfree(t->notes[i].data); + kvfree(t->notes[i].data); kfree(t); } kfree(info->psinfo.data); diff --git a/kernel/regset.c b/kernel/regset.c index 586823786f39..b2871fa68b2a 100644 --- a/kernel/regset.c +++ b/kernel/regset.c @@ -16,14 +16,14 @@ static int __regset_get(struct task_struct *target, if (size > regset->n * regset->size) size = regset->n * regset->size; if (!p) { - to_free = p = kzalloc(size, GFP_KERNEL); + to_free = p = kvzalloc(size, GFP_KERNEL); if (!p) return -ENOMEM; } res = regset->regset_get(target, regset, (struct membuf){.p = p, .left = size}); if (res < 0) { - kfree(to_free); + kvfree(to_free); return res; } *data = p; @@ -71,6 +71,6 @@ int copy_regset_to_user(struct task_struct *target, ret = regset_get_alloc(target, regset, size, &buf); if (ret > 0) ret = copy_to_user(data, buf, ret) ? -EFAULT : 0; - kfree(buf); + kvfree(buf); return ret; } -- cgit v1.2.3-59-g8ed1b From 4eb7b93e03101fd3f35e69affe566e4b1e3e3dca Mon Sep 17 00:00:00 2001 From: Heming Zhao Date: Thu, 28 Mar 2024 20:52:00 +0800 Subject: ocfs2: improve write IO performance when fragmentation is high The group_search function ocfs2_cluster_group_search() should bypass groups with insufficient space to avoid unnecessary searches. This patch is particularly useful when ocfs2 is handling huge number small files, and volume fragmentation is very high. In this case, ocfs2 is busy with looking up available la window from //global_bitmap. This patch introduces a new member in the Group Description (gd) struct called 'bg_contig_free_bits', representing the max contigous free bits in this gd. When ocfs2 allocates a new la window from //global_bitmap, 'bg_contig_free_bits' helps expedite the search process. Let's image below path. 1. la state (->local_alloc_state) is set THROTTLED or DISABLED. 2. when user delete a large file and trigger ocfs2_local_alloc_seen_free_bits set osb->local_alloc_state unconditionally. 3. a write IOs thread run and trigger the worst performance path ``` ocfs2_reserve_clusters_with_limit ocfs2_reserve_local_alloc_bits ocfs2_local_alloc_slide_window //[1] + ocfs2_local_alloc_reserve_for_window //[2] + ocfs2_local_alloc_new_window //[3] ocfs2_recalc_la_window ``` [1]: will be called when la window bits used up. [2]: under la state is ENABLED, and this func only check global_bitmap free bits, it will succeed in general. [3]: will use the default la window size to search clusters then fail. ocfs2_recalc_la_window attempts other la window sizes. the timing complexity is O(n^4), resulting in a significant time cost for scanning global bitmap. This leads to a dramatic slowdown in write I/Os (e.g., user space 'dd'). i.e. an ocfs2 partition size: 1.45TB, cluster size: 4KB, la window default size: 106MB. The partition is fragmentation by creating & deleting huge mount of small files. before this patch, the timing of [3] should be (the number got from real world): - la window size change order (size: MB): 106, 53, 26.5, 13, 6.5, 3.25, 1.6, 0.8 only 0.8MB succeed, 0.8MB also triggers la window to disable. ocfs2_local_alloc_new_window retries 8 times, first 7 times totally runs in worst case. - group chain number: 242 ocfs2_claim_suballoc_bits calls for-loop 242 times - each chain has 49 block group ocfs2_search_chain calls while-loop 49 times - each bg has 32256 blocks ocfs2_block_group_find_clear_bits calls while-loop for 32256 bits. for ocfs2_find_next_zero_bit uses ffz() to find zero bit, let's use (32256/64) (this is not worst value) for timing calucation. the loop times: 7*242*49*(32256/64) = 41835024 (~42 million times) In the worst case, user space writes 1MB data will trigger 42M scanning times. under this patch, the timing is '7*242*49 = 83006', reduced by three orders of magnitude. Link: https://lkml.kernel.org/r/20240328125203.20892-2-heming.zhao@suse.com Signed-off-by: Heming Zhao Reviewed-by: Joseph Qi Cc: Changwei Ge Cc: Gang He Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Signed-off-by: Andrew Morton --- fs/ocfs2/move_extents.c | 2 +- fs/ocfs2/ocfs2_fs.h | 3 +- fs/ocfs2/resize.c | 8 ++++ fs/ocfs2/suballoc.c | 100 ++++++++++++++++++++++++++++++++++++++++++++---- fs/ocfs2/suballoc.h | 6 ++- 5 files changed, 108 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c index 1f9ed117e78b..f9d6a4f9ca92 100644 --- a/fs/ocfs2/move_extents.c +++ b/fs/ocfs2/move_extents.c @@ -685,7 +685,7 @@ static int ocfs2_move_extent(struct ocfs2_move_extents_context *context, } ret = ocfs2_block_group_set_bits(handle, gb_inode, gd, gd_bh, - goal_bit, len); + goal_bit, len, 0, 0); if (ret) { ocfs2_rollback_alloc_dinode_counts(gb_inode, gb_bh, len, le16_to_cpu(gd->bg_chain)); diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index 7aebdbf5cc0a..c93689b568fe 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -883,7 +883,8 @@ struct ocfs2_group_desc __le16 bg_free_bits_count; /* Free bits count */ __le16 bg_chain; /* What chain I am in. */ /*10*/ __le32 bg_generation; - __le32 bg_reserved1; + __le16 bg_contig_free_bits; /* max contig free bits length */ + __le16 bg_reserved1; __le64 bg_next_group; /* Next group in my list, in blocks */ /*20*/ __le64 bg_parent_dinode; /* dinode which owns me, in diff --git a/fs/ocfs2/resize.c b/fs/ocfs2/resize.c index d65d43c61857..c4a4016d3866 100644 --- a/fs/ocfs2/resize.c +++ b/fs/ocfs2/resize.c @@ -91,6 +91,8 @@ static int ocfs2_update_last_group_and_inode(handle_t *handle, u16 cl_bpc = le16_to_cpu(cl->cl_bpc); u16 cl_cpg = le16_to_cpu(cl->cl_cpg); u16 old_bg_clusters; + u16 contig_bits; + __le16 old_bg_contig_free_bits; trace_ocfs2_update_last_group_and_inode(new_clusters, first_new_cluster); @@ -122,6 +124,11 @@ static int ocfs2_update_last_group_and_inode(handle_t *handle, le16_add_cpu(&group->bg_free_bits_count, -1 * backups); } + contig_bits = ocfs2_find_max_contig_free_bits(group->bg_bitmap, + le16_to_cpu(group->bg_bits), 0); + old_bg_contig_free_bits = group->bg_contig_free_bits; + group->bg_contig_free_bits = cpu_to_le16(contig_bits); + ocfs2_journal_dirty(handle, group_bh); /* update the inode accordingly. */ @@ -160,6 +167,7 @@ out_rollback: le16_add_cpu(&group->bg_free_bits_count, backups); le16_add_cpu(&group->bg_bits, -1 * num_bits); le16_add_cpu(&group->bg_free_bits_count, -1 * num_bits); + group->bg_contig_free_bits = old_bg_contig_free_bits; } out: if (ret) diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 961998415308..7e30e9d79890 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -50,6 +50,10 @@ struct ocfs2_suballoc_result { u64 sr_blkno; /* The first allocated block */ unsigned int sr_bit_offset; /* The bit in the bg */ unsigned int sr_bits; /* How many bits we claimed */ + unsigned int sr_max_contig_bits; /* The length for contiguous + * free bits, only available + * for cluster group + */ }; static u64 ocfs2_group_from_res(struct ocfs2_suballoc_result *res) @@ -1272,6 +1276,26 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, return ret; } +u16 ocfs2_find_max_contig_free_bits(void *bitmap, + u16 total_bits, u16 start) +{ + u16 offset, free_bits; + u16 contig_bits = 0; + + while (start < total_bits) { + offset = ocfs2_find_next_zero_bit(bitmap, total_bits, start); + if (offset == total_bits) + break; + + start = ocfs2_find_next_bit(bitmap, total_bits, offset); + free_bits = start - offset; + if (contig_bits < free_bits) + contig_bits = free_bits; + } + + return contig_bits; +} + static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, struct buffer_head *bg_bh, unsigned int bits_wanted, @@ -1280,6 +1304,7 @@ static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, { void *bitmap; u16 best_offset, best_size; + u16 prev_best_size = 0; int offset, start, found, status = 0; struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data; @@ -1306,6 +1331,7 @@ static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, /* got a zero after some ones */ found = 1; start = offset + 1; + prev_best_size = best_size; } if (found > best_size) { best_size = found; @@ -1318,6 +1344,8 @@ static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, } } + /* best_size will be allocated, we save prev_best_size */ + res->sr_max_contig_bits = prev_best_size; if (best_size) { res->sr_bit_offset = best_offset; res->sr_bits = best_size; @@ -1335,11 +1363,15 @@ int ocfs2_block_group_set_bits(handle_t *handle, struct ocfs2_group_desc *bg, struct buffer_head *group_bh, unsigned int bit_off, - unsigned int num_bits) + unsigned int num_bits, + unsigned int max_contig_bits, + int fastpath) { int status; void *bitmap = bg->bg_bitmap; int journal_type = OCFS2_JOURNAL_ACCESS_WRITE; + unsigned int start = bit_off + num_bits; + u16 contig_bits; /* All callers get the descriptor via * ocfs2_read_group_descriptor(). Any corruption is a code bug. */ @@ -1371,6 +1403,28 @@ int ocfs2_block_group_set_bits(handle_t *handle, while(num_bits--) ocfs2_set_bit(bit_off++, bitmap); + /* + * this is optimize path, caller set old contig value + * in max_contig_bits to bypass finding action. + */ + if (fastpath) { + bg->bg_contig_free_bits = cpu_to_le16(max_contig_bits); + } else if (ocfs2_is_cluster_bitmap(alloc_inode)) { + /* + * Usually, the block group bitmap allocates only 1 bit + * at a time, while the cluster group allocates n bits + * each time. Therefore, we only save the contig bits for + * the cluster group. + */ + contig_bits = ocfs2_find_max_contig_free_bits(bitmap, + le16_to_cpu(bg->bg_bits), start); + if (contig_bits > max_contig_bits) + max_contig_bits = contig_bits; + bg->bg_contig_free_bits = cpu_to_le16(max_contig_bits); + } else { + bg->bg_contig_free_bits = 0; + } + ocfs2_journal_dirty(handle, group_bh); bail: @@ -1484,7 +1538,12 @@ static int ocfs2_cluster_group_search(struct inode *inode, BUG_ON(!ocfs2_is_cluster_bitmap(inode)); - if (gd->bg_free_bits_count) { + if (le16_to_cpu(gd->bg_contig_free_bits) && + le16_to_cpu(gd->bg_contig_free_bits) < bits_wanted) + return -ENOSPC; + + /* ->bg_contig_free_bits may un-initialized, so compare again */ + if (le16_to_cpu(gd->bg_free_bits_count) >= bits_wanted) { max_bits = le16_to_cpu(gd->bg_bits); /* Tail groups in cluster bitmaps which aren't cpg @@ -1553,7 +1612,7 @@ static int ocfs2_block_group_search(struct inode *inode, BUG_ON(min_bits != 1); BUG_ON(ocfs2_is_cluster_bitmap(inode)); - if (bg->bg_free_bits_count) { + if (le16_to_cpu(bg->bg_free_bits_count) >= bits_wanted) { ret = ocfs2_block_group_find_clear_bits(OCFS2_SB(inode->i_sb), group_bh, bits_wanted, le16_to_cpu(bg->bg_bits), @@ -1713,7 +1772,8 @@ static int ocfs2_search_one_group(struct ocfs2_alloc_context *ac, } ret = ocfs2_block_group_set_bits(handle, alloc_inode, gd, group_bh, - res->sr_bit_offset, res->sr_bits); + res->sr_bit_offset, res->sr_bits, + res->sr_max_contig_bits, 0); if (ret < 0) { ocfs2_rollback_alloc_dinode_counts(alloc_inode, ac->ac_bh, res->sr_bits, @@ -1847,7 +1907,9 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, bg, group_bh, res->sr_bit_offset, - res->sr_bits); + res->sr_bits, + res->sr_max_contig_bits, + 0); if (status < 0) { ocfs2_rollback_alloc_dinode_counts(alloc_inode, ac->ac_bh, res->sr_bits, chain); @@ -2161,7 +2223,9 @@ int ocfs2_claim_new_inode_at_loc(handle_t *handle, bg, bg_bh, res->sr_bit_offset, - res->sr_bits); + res->sr_bits, + res->sr_max_contig_bits, + 0); if (ret < 0) { ocfs2_rollback_alloc_dinode_counts(ac->ac_inode, ac->ac_bh, res->sr_bits, chain); @@ -2380,11 +2444,13 @@ static int ocfs2_block_group_clear_bits(handle_t *handle, struct buffer_head *group_bh, unsigned int bit_off, unsigned int num_bits, + unsigned int max_contig_bits, void (*undo_fn)(unsigned int bit, unsigned long *bmap)) { int status; unsigned int tmp; + u16 contig_bits; struct ocfs2_group_desc *undo_bg = NULL; struct journal_head *jh; @@ -2431,6 +2497,20 @@ static int ocfs2_block_group_clear_bits(handle_t *handle, num_bits); } + /* + * TODO: even 'num_bits == 1' (the worst case, release 1 cluster), + * we still need to rescan whole bitmap. + */ + if (ocfs2_is_cluster_bitmap(alloc_inode)) { + contig_bits = ocfs2_find_max_contig_free_bits(bg->bg_bitmap, + le16_to_cpu(bg->bg_bits), 0); + if (contig_bits > max_contig_bits) + max_contig_bits = contig_bits; + bg->bg_contig_free_bits = cpu_to_le16(max_contig_bits); + } else { + bg->bg_contig_free_bits = 0; + } + if (undo_fn) spin_unlock(&jh->b_state_lock); @@ -2457,6 +2537,7 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle, struct ocfs2_chain_list *cl = &fe->id2.i_chain; struct buffer_head *group_bh = NULL; struct ocfs2_group_desc *group; + __le16 old_bg_contig_free_bits = 0; /* The alloc_bh comes from ocfs2_free_dinode() or * ocfs2_free_clusters(). The callers have all locked the @@ -2481,9 +2562,11 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle, BUG_ON((count + start_bit) > le16_to_cpu(group->bg_bits)); + if (ocfs2_is_cluster_bitmap(alloc_inode)) + old_bg_contig_free_bits = group->bg_contig_free_bits; status = ocfs2_block_group_clear_bits(handle, alloc_inode, group, group_bh, - start_bit, count, undo_fn); + start_bit, count, 0, undo_fn); if (status < 0) { mlog_errno(status); goto bail; @@ -2494,7 +2577,8 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle, if (status < 0) { mlog_errno(status); ocfs2_block_group_set_bits(handle, alloc_inode, group, group_bh, - start_bit, count); + start_bit, count, + le16_to_cpu(old_bg_contig_free_bits), 1); goto bail; } diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h index 9c74eace3adc..b481b834857d 100644 --- a/fs/ocfs2/suballoc.h +++ b/fs/ocfs2/suballoc.h @@ -79,12 +79,16 @@ void ocfs2_rollback_alloc_dinode_counts(struct inode *inode, struct buffer_head *di_bh, u32 num_bits, u16 chain); +u16 ocfs2_find_max_contig_free_bits(void *bitmap, + u16 total_bits, u16 start); int ocfs2_block_group_set_bits(handle_t *handle, struct inode *alloc_inode, struct ocfs2_group_desc *bg, struct buffer_head *group_bh, unsigned int bit_off, - unsigned int num_bits); + unsigned int num_bits, + unsigned int max_contig_bits, + int fastpath); int ocfs2_claim_metadata(handle_t *handle, struct ocfs2_alloc_context *ac, -- cgit v1.2.3-59-g8ed1b From f51dac026f75863004ebfb7885cec98e6d3172bb Mon Sep 17 00:00:00 2001 From: Heming Zhao Date: Thu, 28 Mar 2024 20:52:01 +0800 Subject: ocfs2: adjust enabling place for la window Patch series "improve write IO performance when fragmentation is high", v6. This patch (of 4): After introducing gd->bg_contig_free_bits, the code path 'ocfs2_cluster_group_search() => ocfs2_local_alloc_seen_free_bits()' becomes death when all the gd->bg_contig_free_bits are set to the correct value. This patch relocates ocfs2_local_alloc_seen_free_bits() to a more appropriate location. (The new place being ocfs2_block_group_set_bits().) In ocfs2_local_alloc_seen_free_bits(), the scope of the spin-lock has been adjusted to reduce meaningless lock races. e.g: when userspace creates & deletes 1 cluster_size files in parallel, acquiring the spin-lock in ocfs2_local_alloc_seen_free_bits() is totally pointless and impedes IO performance. Link: https://lkml.kernel.org/r/20240328125203.20892-3-heming.zhao@suse.com Signed-off-by: Heming Zhao Reviewed-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Gang He Cc: Jun Piao Signed-off-by: Andrew Morton --- fs/ocfs2/localalloc.c | 11 ++++++----- fs/ocfs2/suballoc.c | 9 ++------- 2 files changed, 8 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 33aeaaa056d7..c84ce53cdec0 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -212,14 +212,15 @@ static inline int ocfs2_la_state_enabled(struct ocfs2_super *osb) void ocfs2_local_alloc_seen_free_bits(struct ocfs2_super *osb, unsigned int num_clusters) { - spin_lock(&osb->osb_lock); - if (osb->local_alloc_state == OCFS2_LA_DISABLED || - osb->local_alloc_state == OCFS2_LA_THROTTLED) - if (num_clusters >= osb->local_alloc_default_bits) { + if (num_clusters >= osb->local_alloc_default_bits) { + spin_lock(&osb->osb_lock); + if (osb->local_alloc_state == OCFS2_LA_DISABLED || + osb->local_alloc_state == OCFS2_LA_THROTTLED) { cancel_delayed_work(&osb->la_enable_wq); osb->local_alloc_state = OCFS2_LA_ENABLED; } - spin_unlock(&osb->osb_lock); + spin_unlock(&osb->osb_lock); + } } void ocfs2_la_enable_worker(struct work_struct *work) diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 7e30e9d79890..8314ec487cfb 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -1372,6 +1372,7 @@ int ocfs2_block_group_set_bits(handle_t *handle, int journal_type = OCFS2_JOURNAL_ACCESS_WRITE; unsigned int start = bit_off + num_bits; u16 contig_bits; + struct ocfs2_super *osb = OCFS2_SB(alloc_inode->i_sb); /* All callers get the descriptor via * ocfs2_read_group_descriptor(). Any corruption is a code bug. */ @@ -1421,6 +1422,7 @@ int ocfs2_block_group_set_bits(handle_t *handle, if (contig_bits > max_contig_bits) max_contig_bits = contig_bits; bg->bg_contig_free_bits = cpu_to_le16(max_contig_bits); + ocfs2_local_alloc_seen_free_bits(osb, max_contig_bits); } else { bg->bg_contig_free_bits = 0; } @@ -1587,13 +1589,6 @@ static int ocfs2_cluster_group_search(struct inode *inode, * of bits. */ if (min_bits <= res->sr_bits) search = 0; /* success */ - else if (res->sr_bits) { - /* - * Don't show bits which we'll be returning - * for allocation to the local alloc bitmap. - */ - ocfs2_local_alloc_seen_free_bits(osb, res->sr_bits); - } } return search; -- cgit v1.2.3-59-g8ed1b From 525350221beb55bc6795595443d4cdeecb68ebec Mon Sep 17 00:00:00 2001 From: Heming Zhao Date: Thu, 28 Mar 2024 20:52:02 +0800 Subject: ocfs2: speed up chain-list searching Add short-circuit code to speed up searching Link: https://lkml.kernel.org/r/20240328125203.20892-4-heming.zhao@suse.com Signed-off-by: Heming Zhao Reviewed-by: Joseph Qi Cc: Changwei Ge Cc: Gang He Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Signed-off-by: Andrew Morton --- fs/ocfs2/suballoc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 8314ec487cfb..f7b483f0de2a 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -2006,7 +2006,7 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac, for (i = 0; i < le16_to_cpu(cl->cl_next_free_rec); i ++) { if (i == victim) continue; - if (!cl->cl_recs[i].c_free) + if (le32_to_cpu(cl->cl_recs[i].c_free) < bits_wanted) continue; ac->ac_chain = i; -- cgit v1.2.3-59-g8ed1b From fc07d2a2118aa83dca9bdc8fda6ab9fe2f28a8d6 Mon Sep 17 00:00:00 2001 From: Heming Zhao Date: Thu, 28 Mar 2024 20:52:03 +0800 Subject: ocfs2: fix sparse warnings 1. fs/ocfs2/localalloc.c:1224:41: warning: incorrect type in argument 1 (different base types) fs/ocfs2/localalloc.c:1224:41: expected unsigned long long val1 fs/ocfs2/localalloc.c:1224:41: got restricted __le32 [usertype] la_bm_off 2. fs/ocfs2/export.c:258:32: warning: cast to restricted __le32 fs/ocfs2/export.c:259:33: warning: cast to restricted __le32 fs/ocfs2/export.c:260:32: warning: cast to restricted __le32 fs/ocfs2/export.c:272:32: warning: cast to restricted __le32 fs/ocfs2/export.c:273:33: warning: cast to restricted __le32 fs/ocfs2/export.c:274:32: warning: cast to restricted __le32 3. fs/ocfs2/inode.c:1623:13: warning: context imbalance in 'ocfs2_inode_cache_lock' - wrong count at exit fs/ocfs2/inode.c:1630:13: warning: context imbalance in 'ocfs2_inode_cache_unlock' - unexpected unlock 4. fs/ocfs2/refcounttree.c:633:27: warning: incorrect type in assignment (different base types) fs/ocfs2/refcounttree.c:633:27: expected restricted __le32 [usertype] rf_generation fs/ocfs2/refcounttree.c:633:27: got unsigned int 5. fs/ocfs2/dlm/dlmdomain.c:1316:20: warning: context imbalance in 'dlm_query_nodeinfo_handler' - different lock contexts for basic block Link: https://lkml.kernel.org/r/20240328125203.20892-5-heming.zhao@suse.com Signed-off-by: Heming Zhao Reviewed-by: Joseph Qi Cc: Changwei Ge Cc: Gang He Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Signed-off-by: Andrew Morton --- fs/ocfs2/dlm/dlmdomain.c | 11 +++++------ fs/ocfs2/export.c | 12 ++++++------ fs/ocfs2/inode.c | 2 ++ fs/ocfs2/localalloc.c | 4 ++-- fs/ocfs2/refcounttree.c | 2 +- 5 files changed, 16 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 5c04dde99981..2e0a2f338282 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -1274,7 +1274,7 @@ static int dlm_query_nodeinfo_handler(struct o2net_msg *msg, u32 len, { struct dlm_query_nodeinfo *qn; struct dlm_ctxt *dlm = NULL; - int locked = 0, status = -EINVAL; + int status = -EINVAL; qn = (struct dlm_query_nodeinfo *) msg->buf; @@ -1290,12 +1290,11 @@ static int dlm_query_nodeinfo_handler(struct o2net_msg *msg, u32 len, } spin_lock(&dlm->spinlock); - locked = 1; if (dlm->joining_node != qn->qn_nodenum) { mlog(ML_ERROR, "Node %d queried nodes on domain %s but " "joining node is %d\n", qn->qn_nodenum, qn->qn_domain, dlm->joining_node); - goto bail; + goto unlock; } /* Support for node query was added in 1.1 */ @@ -1305,14 +1304,14 @@ static int dlm_query_nodeinfo_handler(struct o2net_msg *msg, u32 len, "but active dlm protocol is %d.%d\n", qn->qn_nodenum, qn->qn_domain, dlm->dlm_locking_proto.pv_major, dlm->dlm_locking_proto.pv_minor); - goto bail; + goto unlock; } status = dlm_match_nodes(dlm, qn); +unlock: + spin_unlock(&dlm->spinlock); bail: - if (locked) - spin_unlock(&dlm->spinlock); spin_unlock(&dlm_domain_lock); return status; diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c index b8b6a191b5cb..96b684763b39 100644 --- a/fs/ocfs2/export.c +++ b/fs/ocfs2/export.c @@ -255,9 +255,9 @@ static struct dentry *ocfs2_fh_to_dentry(struct super_block *sb, if (fh_len < 3 || fh_type > 2) return NULL; - handle.ih_blkno = (u64)le32_to_cpu(fid->raw[0]) << 32; - handle.ih_blkno |= (u64)le32_to_cpu(fid->raw[1]); - handle.ih_generation = le32_to_cpu(fid->raw[2]); + handle.ih_blkno = (u64)le32_to_cpu((__force __le32)fid->raw[0]) << 32; + handle.ih_blkno |= (u64)le32_to_cpu((__force __le32)fid->raw[1]); + handle.ih_generation = le32_to_cpu((__force __le32)fid->raw[2]); return ocfs2_get_dentry(sb, &handle); } @@ -269,9 +269,9 @@ static struct dentry *ocfs2_fh_to_parent(struct super_block *sb, if (fh_type != 2 || fh_len < 6) return NULL; - parent.ih_blkno = (u64)le32_to_cpu(fid->raw[3]) << 32; - parent.ih_blkno |= (u64)le32_to_cpu(fid->raw[4]); - parent.ih_generation = le32_to_cpu(fid->raw[5]); + parent.ih_blkno = (u64)le32_to_cpu((__force __le32)fid->raw[3]) << 32; + parent.ih_blkno |= (u64)le32_to_cpu((__force __le32)fid->raw[4]); + parent.ih_generation = le32_to_cpu((__force __le32)fid->raw[5]); return ocfs2_get_dentry(sb, &parent); } diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index 999111bfc271..2cc5c99fe941 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -1621,6 +1621,7 @@ static struct super_block *ocfs2_inode_cache_get_super(struct ocfs2_caching_info } static void ocfs2_inode_cache_lock(struct ocfs2_caching_info *ci) +__acquires(&oi->ip_lock) { struct ocfs2_inode_info *oi = cache_info_to_inode(ci); @@ -1628,6 +1629,7 @@ static void ocfs2_inode_cache_lock(struct ocfs2_caching_info *ci) } static void ocfs2_inode_cache_unlock(struct ocfs2_caching_info *ci) +__releases(&oi->ip_lock) { struct ocfs2_inode_info *oi = cache_info_to_inode(ci); diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index c84ce53cdec0..5df34561c551 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -336,7 +336,7 @@ int ocfs2_load_local_alloc(struct ocfs2_super *osb) "found = %u, set = %u, taken = %u, off = %u\n", num_used, le32_to_cpu(alloc->id1.bitmap1.i_used), le32_to_cpu(alloc->id1.bitmap1.i_total), - OCFS2_LOCAL_ALLOC(alloc)->la_bm_off); + le32_to_cpu(OCFS2_LOCAL_ALLOC(alloc)->la_bm_off)); status = -EINVAL; goto bail; @@ -1214,7 +1214,7 @@ retry_enospc: OCFS2_LOCAL_ALLOC(alloc)->la_bitmap); trace_ocfs2_local_alloc_new_window_result( - OCFS2_LOCAL_ALLOC(alloc)->la_bm_off, + le32_to_cpu(OCFS2_LOCAL_ALLOC(alloc)->la_bm_off), le32_to_cpu(alloc->id1.bitmap1.i_total)); bail: diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 3f80a56d0d60..1f303b1adf1a 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -630,7 +630,7 @@ static int ocfs2_create_refcount_tree(struct inode *inode, rb->rf_records.rl_count = cpu_to_le16(ocfs2_refcount_recs_per_rb(osb->sb)); spin_lock(&osb->osb_lock); - rb->rf_generation = osb->s_next_generation++; + rb->rf_generation = cpu_to_le32(osb->s_next_generation++); spin_unlock(&osb->osb_lock); ocfs2_journal_dirty(handle, new_bh); -- cgit v1.2.3-59-g8ed1b From 040bf9a717885d3646dcd5b4062dc3a4877ec421 Mon Sep 17 00:00:00 2001 From: Phillip Lougher Date: Wed, 3 Apr 2024 19:33:52 +0100 Subject: Squashfs: remove deprecated strncpy by not copying the string Squashfs copied the passed string (name) into a temporary buffer to ensure it was NUL-terminated. This however is completely unnecessary as the string is already NUL-terminated. So remove the deprecated strncpy() by completely removing the string copy. The background behind this unnecessary string copy is that it dates back to the days when Squashfs was an out of kernel patch. The code deliberately did not assume the string was NUL-terminated in case in future this changed (due to kernel changes). This would mean the out of tree patches would be broken but still compile OK. Link: https://lkml.kernel.org/r/20240403183352.391308-1-phillip@squashfs.org.uk Signed-off-by: Phillip Lougher Reviewed-by: Kees Cook Reviewed-by: Justin Stitt Signed-off-by: Andrew Morton --- fs/squashfs/namei.c | 14 ++++---------- 1 file changed, 4 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/squashfs/namei.c b/fs/squashfs/namei.c index 11e4539b9eae..65aae7e2a859 100644 --- a/fs/squashfs/namei.c +++ b/fs/squashfs/namei.c @@ -62,27 +62,21 @@ */ static int get_dir_index_using_name(struct super_block *sb, u64 *next_block, int *next_offset, u64 index_start, - int index_offset, int i_count, const char *name, - int len) + int index_offset, int i_count, const char *name) { struct squashfs_sb_info *msblk = sb->s_fs_info; int i, length = 0, err; unsigned int size; struct squashfs_dir_index *index; - char *str; TRACE("Entered get_dir_index_using_name, i_count %d\n", i_count); - index = kmalloc(sizeof(*index) + SQUASHFS_NAME_LEN * 2 + 2, GFP_KERNEL); + index = kmalloc(sizeof(*index) + SQUASHFS_NAME_LEN + 1, GFP_KERNEL); if (index == NULL) { ERROR("Failed to allocate squashfs_dir_index\n"); goto out; } - str = &index->name[SQUASHFS_NAME_LEN + 1]; - strncpy(str, name, len); - str[len] = '\0'; - for (i = 0; i < i_count; i++) { err = squashfs_read_metadata(sb, index, &index_start, &index_offset, sizeof(*index)); @@ -101,7 +95,7 @@ static int get_dir_index_using_name(struct super_block *sb, index->name[size] = '\0'; - if (strcmp(index->name, str) > 0) + if (strcmp(index->name, name) > 0) break; length = le32_to_cpu(index->index); @@ -153,7 +147,7 @@ static struct dentry *squashfs_lookup(struct inode *dir, struct dentry *dentry, length = get_dir_index_using_name(dir->i_sb, &block, &offset, squashfs_i(dir)->dir_idx_start, squashfs_i(dir)->dir_idx_offset, - squashfs_i(dir)->dir_idx_cnt, name, len); + squashfs_i(dir)->dir_idx_cnt, name); while (length < i_size_read(dir)) { /* -- cgit v1.2.3-59-g8ed1b From ad5f0eb540d31d95e6eefe6e24af0fe1b1b393fa Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Mon, 1 Apr 2024 18:39:55 +0000 Subject: vmcore: replace strncpy with strscpy_pad strncpy() is in the process of being replaced as it is deprecated [1]. We should move towards safer and less ambiguous string interfaces. Looking at vmcoredd_header's definition: | struct vmcoredd_header { | __u32 n_namesz; /* Name size */ | __u32 n_descsz; /* Content size */ | __u32 n_type; /* NT_VMCOREDD */ | __u8 name[8]; /* LINUX\0\0\0 */ | __u8 dump_name[VMCOREDD_MAX_NAME_BYTES]; /* Device dump's name */ | }; .. we see that @name wants to be NUL-padded. We're copying data->dump_name which is defined as: | char dump_name[VMCOREDD_MAX_NAME_BYTES]; /* Unique name of the dump */ .. which shares the same size as vdd_hdr->dump_name. Let's make sure we NUL-pad this as well. Use strscpy_pad() which NUL-terminates and NUL-pads its destination buffers. Specifically, use the new 2-argument version of strscpy_pad introduced in Commit e6584c3964f2f ("string: Allow 2-argument strscpy()"). Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://github.com/KSPP/linux/issues/90 Link: https://lkml.kernel.org/r/20240401-strncpy-fs-proc-vmcore-c-v2-1-dd0a73f42635@google.com Signed-off-by: Justin Stitt Acked-by: Baoquan He Cc: Dave Young Cc: Vivek Goyal Signed-off-by: Andrew Morton --- fs/proc/vmcore.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 1fb213f379a5..5d08d4d159d3 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -1370,9 +1370,8 @@ static void vmcoredd_write_header(void *buf, struct vmcoredd_data *data, vdd_hdr->n_descsz = size + sizeof(vdd_hdr->dump_name); vdd_hdr->n_type = NT_VMCOREDD; - strncpy((char *)vdd_hdr->name, VMCOREDD_NOTE_NAME, - sizeof(vdd_hdr->name)); - memcpy(vdd_hdr->dump_name, data->dump_name, sizeof(vdd_hdr->dump_name)); + strscpy_pad(vdd_hdr->name, VMCOREDD_NOTE_NAME); + strscpy_pad(vdd_hdr->dump_name, data->dump_name); } /** -- cgit v1.2.3-59-g8ed1b From d11547071a9d916604a65394b169c0d1acf0a6d6 Mon Sep 17 00:00:00 2001 From: Su Yue Date: Mon, 8 Apr 2024 16:20:38 +0800 Subject: ocfs2: return real error code in ocfs2_dio_wr_get_block Patch series "ocfs2 bugs fixes exposed by fstests", v3. The patchset is to fix some wrong behavior of ocfs2 exposed by fstests. Patch 1 makes userspace happy when some error happens when doing direct io. Before the patch, DIO always return -EIO in case of error. After the patch, it returns real error code such like -ENOSPC, EDQUOT... Patch 2 fixes an error case when doing AIO+DIO and hole punching at same file position in parallel. generic/300 Patch 3 fixes inode link count mismatch after power failure. Without the patch, inode link would be wrong even fync was called on the file. tests/generic/040,041,104,107,336 patch 4 fixes wrong atime with mount option realtime. Without the patch, atime of new created file won't be updated in right time. tests/generic/192 For stable kernels, I added fixes to patch 2,3,4. The patch 1 is not recommended to be backported since ocfs2_dio_wr_get_block calls too many functions. It's diffcult to check every git history of ocfs2 for every LTS kernel. This patch (of 4): ocfs2_dio_wr_get_block always returns -EIO in case of errors. However, some programs expect right exit codes while doing dio. For example, tools like fio treat -ENOSPC as expected code while doing stress jobs. And quota tools expect -EDQUOT when disk quota exceeds. -EIO is too strong return code in the dio path. The caller of ocfs2_dio_wr_get_block is __blockdev_direct_IO which is widely used and it handles error codes well. I have checked functions called by ocfs2_dio_wr_get_block and their return codes look good and clear. So I think it's safe to let ocfs2_dio_wr_get_block return real error code. Link: https://lkml.kernel.org/r/20240408082041.20925-1-glass.su@suse.com Link: https://lkml.kernel.org/r/20240408082041.20925-2-glass.su@suse.com Signed-off-by: Su Yue Reviewed-by: Joseph Qi Cc: Changwei Ge Cc: Gang He Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Signed-off-by: Andrew Morton --- fs/ocfs2/aops.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index b82185075de7..f0467d3b3c88 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -2283,8 +2283,6 @@ unlock: ocfs2_inode_unlock(inode, 1); brelse(di_bh); out: - if (ret < 0) - ret = -EIO; return ret; } -- cgit v1.2.3-59-g8ed1b From 952b023f06a24b2ad6ba67304c4c84d45bea2f18 Mon Sep 17 00:00:00 2001 From: Su Yue Date: Mon, 8 Apr 2024 16:20:39 +0800 Subject: ocfs2: fix races between hole punching and AIO+DIO After commit "ocfs2: return real error code in ocfs2_dio_wr_get_block", fstests/generic/300 become from always failed to sometimes failed: ======================================================================== [ 473.293420 ] run fstests generic/300 [ 475.296983 ] JBD2: Ignoring recovery information on journal [ 475.302473 ] ocfs2: Mounting device (253,1) on (node local, slot 0) with ordered data mode. [ 494.290998 ] OCFS2: ERROR (device dm-1): ocfs2_change_extent_flag: Owner 5668 has an extent at cpos 78723 which can no longer be found [ 494.291609 ] On-disk corruption discovered. Please run fsck.ocfs2 once the filesystem is unmounted. [ 494.292018 ] OCFS2: File system is now read-only. [ 494.292224 ] (kworker/19:11,2628,19):ocfs2_mark_extent_written:5272 ERROR: status = -30 [ 494.292602 ] (kworker/19:11,2628,19):ocfs2_dio_end_io_write:2374 ERROR: status = -3 fio: io_u error on file /mnt/scratch/racer: Read-only file system: write offset=460849152, buflen=131072 ========================================================================= In __blockdev_direct_IO, ocfs2_dio_wr_get_block is called to add unwritten extents to a list. extents are also inserted into extent tree in ocfs2_write_begin_nolock. Then another thread call fallocate to puch a hole at one of the unwritten extent. The extent at cpos was removed by ocfs2_remove_extent(). At end io worker thread, ocfs2_search_extent_list found there is no such extent at the cpos. T1 T2 T3 inode lock ... insert extents ... inode unlock ocfs2_fallocate __ocfs2_change_file_space inode lock lock ip_alloc_sem ocfs2_remove_inode_range inode ocfs2_remove_btree_range ocfs2_remove_extent ^---remove the extent at cpos 78723 ... unlock ip_alloc_sem inode unlock ocfs2_dio_end_io ocfs2_dio_end_io_write lock ip_alloc_sem ocfs2_mark_extent_written ocfs2_change_extent_flag ocfs2_search_extent_list ^---failed to find extent ... unlock ip_alloc_sem In most filesystems, fallocate is not compatible with racing with AIO+DIO, so fix it by adding to wait for all dio before fallocate/punch_hole like ext4. Link: https://lkml.kernel.org/r/20240408082041.20925-3-glass.su@suse.com Fixes: b25801038da5 ("ocfs2: Support xfs style space reservation ioctls") Signed-off-by: Su Yue Reviewed-by: Joseph Qi Cc: Changwei Ge Cc: Gang He Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/file.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 0da8e7bd3261..ccc57038a977 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1936,6 +1936,8 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode, inode_lock(inode); + /* Wait all existing dio workers, newcomers will block on i_rwsem */ + inode_dio_wait(inode); /* * This prevents concurrent writes on other nodes */ -- cgit v1.2.3-59-g8ed1b From 8c40984eeb8804cffcd28640f427f4fe829243fc Mon Sep 17 00:00:00 2001 From: Su Yue Date: Mon, 8 Apr 2024 16:20:40 +0800 Subject: ocfs2: update inode fsync transaction id in ocfs2_unlink and ocfs2_link transaction id should be updated in ocfs2_unlink and ocfs2_link. Otherwise, inode link will be wrong after journal replay even fsync was called before power failure: ======================================================================= $ touch testdir/bar $ ln testdir/bar testdir/bar_link $ fsync testdir/bar $ stat -c %h $SCRATCH_MNT/testdir/bar 1 $ stat -c %h $SCRATCH_MNT/testdir/bar 1 ======================================================================= Link: https://lkml.kernel.org/r/20240408082041.20925-4-glass.su@suse.com Fixes: ccd979bdbce9 ("[PATCH] OCFS2: The Second Oracle Cluster Filesystem") Signed-off-by: Su Yue Reviewed-by: Joseph Qi Cc: Changwei Ge Cc: Gang He Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/namei.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 9221a33f917b..55c9d90caaaf 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -797,6 +797,7 @@ static int ocfs2_link(struct dentry *old_dentry, ocfs2_set_links_count(fe, inode->i_nlink); fe->i_ctime = cpu_to_le64(inode_get_ctime_sec(inode)); fe->i_ctime_nsec = cpu_to_le32(inode_get_ctime_nsec(inode)); + ocfs2_update_inode_fsync_trans(handle, inode, 0); ocfs2_journal_dirty(handle, fe_bh); err = ocfs2_add_entry(handle, dentry, inode, @@ -993,6 +994,7 @@ static int ocfs2_unlink(struct inode *dir, drop_nlink(inode); drop_nlink(inode); ocfs2_set_links_count(fe, inode->i_nlink); + ocfs2_update_inode_fsync_trans(handle, inode, 0); ocfs2_journal_dirty(handle, fe_bh); inode_set_mtime_to_ts(dir, inode_set_ctime_current(dir)); -- cgit v1.2.3-59-g8ed1b From b8cb324277ee16f3eca3055b96fce4735a5a41c6 Mon Sep 17 00:00:00 2001 From: Su Yue Date: Mon, 8 Apr 2024 16:20:41 +0800 Subject: ocfs2: use coarse time for new created files The default atime related mount option is '-o realtime' which means file atime should be updated if atime <= ctime or atime <= mtime. atime should be updated in the following scenario, but it is not: ========================================================== $ rm /mnt/testfile; $ echo test > /mnt/testfile $ stat -c "%X %Y %Z" /mnt/testfile 1711881646 1711881646 1711881646 $ sleep 5 $ cat /mnt/testfile > /dev/null $ stat -c "%X %Y %Z" /mnt/testfile 1711881646 1711881646 1711881646 ========================================================== And the reason the atime in the test is not updated is that ocfs2 calls ktime_get_real_ts64() in __ocfs2_mknod_locked during file creation. Then inode_set_ctime_current() is called in inode_set_ctime_current() calls ktime_get_coarse_real_ts64() to get current time. ktime_get_real_ts64() is more accurate than ktime_get_coarse_real_ts64(). In my test box, I saw ctime set by ktime_get_coarse_real_ts64() is less than ktime_get_real_ts64() even ctime is set later. The ctime of the new inode is smaller than atime. The call trace is like: ocfs2_create ocfs2_mknod __ocfs2_mknod_locked .... ktime_get_real_ts64 <------- set atime,ctime,mtime, more accurate ocfs2_populate_inode ... ocfs2_init_acl ocfs2_acl_set_mode inode_set_ctime_current current_time ktime_get_coarse_real_ts64 <-------less accurate ocfs2_file_read_iter ocfs2_inode_lock_atime ocfs2_should_update_atime atime <= ctime ? <-------- false, ctime < atime due to accuracy So here call ktime_get_coarse_real_ts64 to set inode time coarser while creating new files. It may lower the accuracy of file times. But it's not a big deal since we already use coarse time in other places like ocfs2_update_inode_atime and inode_set_ctime_current. Link: https://lkml.kernel.org/r/20240408082041.20925-5-glass.su@suse.com Fixes: c62c38f6b91b ("ocfs2: replace CURRENT_TIME macro") Signed-off-by: Su Yue Reviewed-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Gang He Cc: Jun Piao Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 55c9d90caaaf..4d1ea8703fcd 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -566,7 +566,7 @@ static int __ocfs2_mknod_locked(struct inode *dir, fe->i_last_eb_blk = 0; strcpy(fe->i_signature, OCFS2_INODE_SIGNATURE); fe->i_flags |= cpu_to_le32(OCFS2_VALID_FL); - ktime_get_real_ts64(&ts); + ktime_get_coarse_real_ts64(&ts); fe->i_atime = fe->i_ctime = fe->i_mtime = cpu_to_le64(ts.tv_sec); fe->i_mtime_nsec = fe->i_ctime_nsec = fe->i_atime_nsec = -- cgit v1.2.3-59-g8ed1b From 2725844080d2609c48a51e08f9a0a3fbe7316665 Mon Sep 17 00:00:00 2001 From: Yang Li Date: Wed, 10 Apr 2024 16:56:27 +0900 Subject: nilfs2: add kernel-doc comments to nilfs_do_roll_forward() Patch series "nilfs2: fix missing kernel-doc comments". This commit adds kernel-doc style comments with complete parameter descriptions for the function nilfs_do_roll_forward. Link: https://lkml.kernel.org/r/20240410075629.3441-1-konishi.ryusuke@gmail.com Link: https://lkml.kernel.org/r/20240410075629.3441-2-konishi.ryusuke@gmail.com Signed-off-by: Yang Li Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton --- fs/nilfs2/recovery.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c index 49a70c68bf3c..e48372618ac4 100644 --- a/fs/nilfs2/recovery.c +++ b/fs/nilfs2/recovery.c @@ -563,6 +563,7 @@ static int nilfs_recover_dsync_blocks(struct the_nilfs *nilfs, * checkpoint * @nilfs: nilfs object * @sb: super block instance + * @root: NILFS root instance * @ri: pointer to a nilfs_recovery_info */ static int nilfs_do_roll_forward(struct the_nilfs *nilfs, -- cgit v1.2.3-59-g8ed1b From 3da9b9650acc3a2a0c3d3f4542b93d4abe9da1de Mon Sep 17 00:00:00 2001 From: Yang Li Date: Wed, 10 Apr 2024 16:56:28 +0900 Subject: nilfs2: add kernel-doc comments to nilfs_btree_convert_and_insert() This commit adds kernel-doc style comments with complete parameter descriptions for the function nilfs_btree_convert_and_insert. Link: https://lkml.kernel.org/r/20240410075629.3441-3-konishi.ryusuke@gmail.com Signed-off-by: Yang Li Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton --- fs/nilfs2/btree.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/nilfs2/btree.c b/fs/nilfs2/btree.c index 65659fa0372e..a139970e4804 100644 --- a/fs/nilfs2/btree.c +++ b/fs/nilfs2/btree.c @@ -1857,13 +1857,22 @@ nilfs_btree_commit_convert_and_insert(struct nilfs_bmap *btree, } /** - * nilfs_btree_convert_and_insert - - * @bmap: - * @key: - * @ptr: - * @keys: - * @ptrs: - * @n: + * nilfs_btree_convert_and_insert - Convert and insert entries into a B-tree + * @btree: NILFS B-tree structure + * @key: Key of the new entry to be inserted + * @ptr: Pointer (block number) associated with the key to be inserted + * @keys: Array of keys to be inserted in addition to @key + * @ptrs: Array of pointers associated with @keys + * @n: Number of keys and pointers in @keys and @ptrs + * + * This function is used to insert a new entry specified by @key and @ptr, + * along with additional entries specified by @keys and @ptrs arrays, into a + * NILFS B-tree. + * It prepares the necessary changes by allocating the required blocks and any + * necessary intermediate nodes. It converts configurations from other forms of + * block mapping (the one that currently exists is direct mapping) to a B-tree. + * + * Return: 0 on success or a negative error code on failure. */ int nilfs_btree_convert_and_insert(struct nilfs_bmap *btree, __u64 key, __u64 ptr, -- cgit v1.2.3-59-g8ed1b From 4a458576941c78d905415a4a6edb48d4c04f7444 Mon Sep 17 00:00:00 2001 From: Yang Li Date: Wed, 10 Apr 2024 16:56:29 +0900 Subject: nilfs2: add kernel-doc comments to nilfs_remove_all_gcinodes() This commit adds kernel-doc style comments with complete parameter descriptions for the function nilfs_remove_all_gcinodes. Link: https://lkml.kernel.org/r/20240410075629.3441-4-konishi.ryusuke@gmail.com Signed-off-by: Yang Li Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton --- fs/nilfs2/gcinode.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/nilfs2/gcinode.c b/fs/nilfs2/gcinode.c index bf9a11d58817..1c9ae36a03ab 100644 --- a/fs/nilfs2/gcinode.c +++ b/fs/nilfs2/gcinode.c @@ -175,6 +175,7 @@ int nilfs_init_gcinode(struct inode *inode) /** * nilfs_remove_all_gcinodes() - remove all unprocessed gc inodes + * @nilfs: NILFS filesystem instance */ void nilfs_remove_all_gcinodes(struct the_nilfs *nilfs) { -- cgit v1.2.3-59-g8ed1b From 14ee22fef420c864c0869419e54aa4e88f64b4e6 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 25 Apr 2024 15:16:59 +0200 Subject: xfs: factor out a xfs_dir_lookup_args helper Add a helper to switch between the different directory formats for lookup and to handle the -EEXIST return for a successful lookup. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_dir2.c | 66 +++++++++++++++++++++++++++++------------------- fs/xfs/libxfs/xfs_dir2.h | 2 ++ fs/xfs/scrub/readdir.c | 35 +------------------------ 3 files changed, 43 insertions(+), 60 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index 7634344dc515..b4f935908911 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -352,6 +352,45 @@ xfs_dir_cilookup_result( return -EEXIST; } +int +xfs_dir_lookup_args( + struct xfs_da_args *args) +{ + bool is_block, is_leaf; + int error; + + if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { + error = xfs_dir2_sf_lookup(args); + goto out; + } + + /* dir2 functions require that the data fork is loaded */ + error = xfs_iread_extents(args->trans, args->dp, XFS_DATA_FORK); + if (error) + goto out; + + error = xfs_dir2_isblock(args, &is_block); + if (error) + goto out; + + if (is_block) { + error = xfs_dir2_block_lookup(args); + goto out; + } + + error = xfs_dir2_isleaf(args, &is_leaf); + if (error) + goto out; + if (is_leaf) + error = xfs_dir2_leaf_lookup(args); + else + error = xfs_dir2_node_lookup(args); +out: + if (error != -EEXIST) + return error; + return 0; +} + /* * Lookup a name in a directory, give back the inode number. * If ci_name is not NULL, returns the actual name in ci_name if it differs @@ -368,7 +407,6 @@ xfs_dir_lookup( { struct xfs_da_args *args; int rval; - bool v; int lock_mode; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); @@ -390,30 +428,7 @@ xfs_dir_lookup( args->op_flags |= XFS_DA_OP_CILOOKUP; lock_mode = xfs_ilock_data_map_shared(dp); - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { - rval = xfs_dir2_sf_lookup(args); - goto out_check_rval; - } - - rval = xfs_dir2_isblock(args, &v); - if (rval) - goto out_free; - if (v) { - rval = xfs_dir2_block_lookup(args); - goto out_check_rval; - } - - rval = xfs_dir2_isleaf(args, &v); - if (rval) - goto out_free; - if (v) - rval = xfs_dir2_leaf_lookup(args); - else - rval = xfs_dir2_node_lookup(args); - -out_check_rval: - if (rval == -EEXIST) - rval = 0; + rval = xfs_dir_lookup_args(args); if (!rval) { *inum = args->inumber; if (ci_name) { @@ -421,7 +436,6 @@ out_check_rval: ci_name->len = args->valuelen; } } -out_free: xfs_iunlock(dp, lock_mode); kfree(args); return rval; diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index b580a78bcf4f..982c2249bfa3 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -66,6 +66,8 @@ extern int xfs_dir_replace(struct xfs_trans *tp, struct xfs_inode *dp, extern int xfs_dir_canenter(struct xfs_trans *tp, struct xfs_inode *dp, struct xfs_name *name); +int xfs_dir_lookup_args(struct xfs_da_args *args); + /* * Direct call from the bmap code, bypassing the generic directory layer. */ diff --git a/fs/xfs/scrub/readdir.c b/fs/xfs/scrub/readdir.c index 28a94c78b0b1..0ac77359d8e9 100644 --- a/fs/xfs/scrub/readdir.c +++ b/fs/xfs/scrub/readdir.c @@ -328,7 +328,6 @@ xchk_dir_lookup( .op_flags = XFS_DA_OP_OKNOENT, .owner = dp->i_ino, }; - bool isblock, isleaf; int error; if (xfs_is_shutdown(dp->i_mount)) @@ -344,39 +343,7 @@ xchk_dir_lookup( ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); xfs_assert_ilocked(dp, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL); - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { - error = xfs_dir2_sf_lookup(&args); - goto out_check_rval; - } - - /* dir2 functions require that the data fork is loaded */ - error = xfs_iread_extents(sc->tp, dp, XFS_DATA_FORK); - if (error) - return error; - - error = xfs_dir2_isblock(&args, &isblock); - if (error) - return error; - - if (isblock) { - error = xfs_dir2_block_lookup(&args); - goto out_check_rval; - } - - error = xfs_dir2_isleaf(&args, &isleaf); - if (error) - return error; - - if (isleaf) { - error = xfs_dir2_leaf_lookup(&args); - goto out_check_rval; - } - - error = xfs_dir2_node_lookup(&args); - -out_check_rval: - if (error == -EEXIST) - error = 0; + error = xfs_dir_lookup_args(&args); if (!error) *ino = args.inumber; return error; -- cgit v1.2.3-59-g8ed1b From 4d893a40514e9c4ee6e3be48ed1b77efb38c313b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 25 Apr 2024 15:17:00 +0200 Subject: xfs: factor out a xfs_dir_createname_args helper Add a helper to switch between the different directory formats for creating a directory entry and to handle the XFS_DA_OP_JUSTCHECK flag based on the passed in ino number field. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_dir2.c | 53 +++++++++++++++++++++++++---------------------- fs/xfs/libxfs/xfs_dir2.h | 1 + fs/xfs/scrub/dir_repair.c | 19 +---------------- 3 files changed, 30 insertions(+), 43 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index b4f935908911..e2727602d047 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -256,6 +256,33 @@ xfs_dir_init( return error; } +int +xfs_dir_createname_args( + struct xfs_da_args *args) +{ + bool is_block, is_leaf; + int error; + + if (!args->inumber) + args->op_flags |= XFS_DA_OP_JUSTCHECK; + + if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return xfs_dir2_sf_addname(args); + + error = xfs_dir2_isblock(args, &is_block); + if (error) + return error; + if (is_block) + return xfs_dir2_block_addname(args); + + error = xfs_dir2_isleaf(args, &is_leaf); + if (error) + return error; + if (is_leaf) + return xfs_dir2_leaf_addname(args); + return xfs_dir2_node_addname(args); +} + /* * Enter a name in a directory, or check for available space. * If inum is 0, only the available space test is performed. @@ -270,7 +297,6 @@ xfs_dir_createname( { struct xfs_da_args *args; int rval; - bool v; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); @@ -297,31 +323,8 @@ xfs_dir_createname( args->trans = tp; args->op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; args->owner = dp->i_ino; - if (!inum) - args->op_flags |= XFS_DA_OP_JUSTCHECK; - - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { - rval = xfs_dir2_sf_addname(args); - goto out_free; - } - rval = xfs_dir2_isblock(args, &v); - if (rval) - goto out_free; - if (v) { - rval = xfs_dir2_block_addname(args); - goto out_free; - } - - rval = xfs_dir2_isleaf(args, &v); - if (rval) - goto out_free; - if (v) - rval = xfs_dir2_leaf_addname(args); - else - rval = xfs_dir2_node_addname(args); - -out_free: + rval = xfs_dir_createname_args(args); kfree(args); return rval; } diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index 982c2249bfa3..f5361dd7b90a 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -67,6 +67,7 @@ extern int xfs_dir_canenter(struct xfs_trans *tp, struct xfs_inode *dp, struct xfs_name *name); int xfs_dir_lookup_args(struct xfs_da_args *args); +int xfs_dir_createname_args(struct xfs_da_args *args); /* * Direct call from the bmap code, bypassing the generic directory layer. diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index fa1a9954d48d..a1e31b782788 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -687,7 +687,6 @@ xrep_dir_replay_createname( { struct xfs_scrub *sc = rd->sc; struct xfs_inode *dp = rd->sc->tempip; - bool is_block, is_leaf; int error; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); @@ -702,23 +701,7 @@ xrep_dir_replay_createname( rd->args.inumber = inum; rd->args.total = total; rd->args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; - - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) - return xfs_dir2_sf_addname(&rd->args); - - error = xfs_dir2_isblock(&rd->args, &is_block); - if (error) - return error; - if (is_block) - return xfs_dir2_block_addname(&rd->args); - - error = xfs_dir2_isleaf(&rd->args, &is_leaf); - if (error) - return error; - if (is_leaf) - return xfs_dir2_leaf_addname(&rd->args); - - return xfs_dir2_node_addname(&rd->args); + return xfs_dir_createname_args(&rd->args); } /* Replay a stashed removename onto the temporary directory. */ -- cgit v1.2.3-59-g8ed1b From 3866e6e669e2c1b3eebf580b8779ea55838c3f5a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 25 Apr 2024 15:17:01 +0200 Subject: xfs: factor out a xfs_dir_removename_args helper Add a helper to switch between the different directory formats for removing a directory entry. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_dir2.c | 48 ++++++++++++++++++++++++----------------------- fs/xfs/libxfs/xfs_dir2.h | 1 + fs/xfs/scrub/dir_repair.c | 20 +------------------- 3 files changed, 27 insertions(+), 42 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index e2727602d047..76aa11ade2e9 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -444,6 +444,30 @@ xfs_dir_lookup( return rval; } +int +xfs_dir_removename_args( + struct xfs_da_args *args) +{ + bool is_block, is_leaf; + int error; + + if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return xfs_dir2_sf_removename(args); + + error = xfs_dir2_isblock(args, &is_block); + if (error) + return error; + if (is_block) + return xfs_dir2_block_removename(args); + + error = xfs_dir2_isleaf(args, &is_leaf); + if (error) + return error; + if (is_leaf) + return xfs_dir2_leaf_removename(args); + return xfs_dir2_node_removename(args); +} + /* * Remove an entry from a directory. */ @@ -457,7 +481,6 @@ xfs_dir_removename( { struct xfs_da_args *args; int rval; - bool v; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); XFS_STATS_INC(dp->i_mount, xs_dir_remove); @@ -477,28 +500,7 @@ xfs_dir_removename( args->whichfork = XFS_DATA_FORK; args->trans = tp; args->owner = dp->i_ino; - - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { - rval = xfs_dir2_sf_removename(args); - goto out_free; - } - - rval = xfs_dir2_isblock(args, &v); - if (rval) - goto out_free; - if (v) { - rval = xfs_dir2_block_removename(args); - goto out_free; - } - - rval = xfs_dir2_isleaf(args, &v); - if (rval) - goto out_free; - if (v) - rval = xfs_dir2_leaf_removename(args); - else - rval = xfs_dir2_node_removename(args); -out_free: + rval = xfs_dir_removename_args(args); kfree(args); return rval; } diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index f5361dd7b90a..3db54801d69e 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -68,6 +68,7 @@ extern int xfs_dir_canenter(struct xfs_trans *tp, struct xfs_inode *dp, int xfs_dir_lookup_args(struct xfs_da_args *args); int xfs_dir_createname_args(struct xfs_da_args *args); +int xfs_dir_removename_args(struct xfs_da_args *args); /* * Direct call from the bmap code, bypassing the generic directory layer. diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index a1e31b782788..98e4ed25cc23 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -712,8 +712,6 @@ xrep_dir_replay_removename( xfs_extlen_t total) { struct xfs_inode *dp = rd->args.dp; - bool is_block, is_leaf; - int error; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); @@ -722,23 +720,7 @@ xrep_dir_replay_removename( rd->args.total = total; trace_xrep_dir_replay_removename(dp, name, 0); - - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) - return xfs_dir2_sf_removename(&rd->args); - - error = xfs_dir2_isblock(&rd->args, &is_block); - if (error) - return error; - if (is_block) - return xfs_dir2_block_removename(&rd->args); - - error = xfs_dir2_isleaf(&rd->args, &is_leaf); - if (error) - return error; - if (is_leaf) - return xfs_dir2_leaf_removename(&rd->args); - - return xfs_dir2_node_removename(&rd->args); + return xfs_dir_removename_args(&rd->args); } /* -- cgit v1.2.3-59-g8ed1b From dfe5febe2b6a175d730861441bff4f726fc58a6c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 25 Apr 2024 15:17:02 +0200 Subject: xfs: factor out a xfs_dir_replace_args helper Add a helper to switch between the different directory formats for removing a directory entry. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_dir2.c | 49 +++++++++++++++++++++++++---------------------- fs/xfs/libxfs/xfs_dir2.h | 1 + fs/xfs/scrub/dir_repair.c | 19 +----------------- 3 files changed, 28 insertions(+), 41 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index 76aa11ade2e9..d3d4d80c2098 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -505,6 +505,31 @@ xfs_dir_removename( return rval; } +int +xfs_dir_replace_args( + struct xfs_da_args *args) +{ + bool is_block, is_leaf; + int error; + + if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return xfs_dir2_sf_replace(args); + + error = xfs_dir2_isblock(args, &is_block); + if (error) + return error; + if (is_block) + return xfs_dir2_block_replace(args); + + error = xfs_dir2_isleaf(args, &is_leaf); + if (error) + return error; + if (is_leaf) + return xfs_dir2_leaf_replace(args); + + return xfs_dir2_node_replace(args); +} + /* * Replace the inode number of a directory entry. */ @@ -518,7 +543,6 @@ xfs_dir_replace( { struct xfs_da_args *args; int rval; - bool v; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); @@ -541,28 +565,7 @@ xfs_dir_replace( args->whichfork = XFS_DATA_FORK; args->trans = tp; args->owner = dp->i_ino; - - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { - rval = xfs_dir2_sf_replace(args); - goto out_free; - } - - rval = xfs_dir2_isblock(args, &v); - if (rval) - goto out_free; - if (v) { - rval = xfs_dir2_block_replace(args); - goto out_free; - } - - rval = xfs_dir2_isleaf(args, &v); - if (rval) - goto out_free; - if (v) - rval = xfs_dir2_leaf_replace(args); - else - rval = xfs_dir2_node_replace(args); -out_free: + rval = xfs_dir_replace_args(args); kfree(args); return rval; } diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index 3db54801d69e..6c00fe24a898 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -69,6 +69,7 @@ extern int xfs_dir_canenter(struct xfs_trans *tp, struct xfs_inode *dp, int xfs_dir_lookup_args(struct xfs_da_args *args); int xfs_dir_createname_args(struct xfs_da_args *args); int xfs_dir_removename_args(struct xfs_da_args *args); +int xfs_dir_replace_args(struct xfs_da_args *args); /* * Direct call from the bmap code, bypassing the generic directory layer. diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 98e4ed25cc23..64679fe08446 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -1513,7 +1513,6 @@ xrep_dir_replace( xfs_extlen_t total) { struct xfs_scrub *sc = rd->sc; - bool is_block, is_leaf; int error; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); @@ -1525,23 +1524,7 @@ xrep_dir_replace( xrep_dir_init_args(rd, dp, name); rd->args.inumber = inum; rd->args.total = total; - - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) - return xfs_dir2_sf_replace(&rd->args); - - error = xfs_dir2_isblock(&rd->args, &is_block); - if (error) - return error; - if (is_block) - return xfs_dir2_block_replace(&rd->args); - - error = xfs_dir2_isleaf(&rd->args, &is_leaf); - if (error) - return error; - if (is_leaf) - return xfs_dir2_leaf_replace(&rd->args); - - return xfs_dir2_node_replace(&rd->args); + return xfs_dir_replace_args(&rd->args); } /* -- cgit v1.2.3-59-g8ed1b From e58ac1770ded2a316447ca7608bb7809af82eca6 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 25 Apr 2024 15:17:03 +0200 Subject: xfs: refactor dir format helpers Add a new enum and a xfs_dir2_format helper that returns it to allow the code to switch on the format of a directory in a single operation and switch all helpers of xfs_dir2_isblock and xfs_dir2_isleaf to it. This also removes the explicit xfs_iread_extents call in a few of the call sites given that xfs_bmap_last_offset already takes care of it underneath. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_dir2.c | 188 +++++++++++++++++-------------------------- fs/xfs/libxfs/xfs_dir2.h | 12 ++- fs/xfs/libxfs/xfs_exchmaps.c | 9 +-- fs/xfs/scrub/dir.c | 3 +- fs/xfs/scrub/readdir.c | 24 +++--- fs/xfs/xfs_dir2_readdir.c | 19 +++-- 6 files changed, 105 insertions(+), 150 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index d3d4d80c2098..457f9a38f850 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -256,31 +256,60 @@ xfs_dir_init( return error; } +enum xfs_dir2_fmt +xfs_dir2_format( + struct xfs_da_args *args, + int *error) +{ + struct xfs_inode *dp = args->dp; + struct xfs_mount *mp = dp->i_mount; + struct xfs_da_geometry *geo = mp->m_dir_geo; + xfs_fileoff_t eof; + + xfs_assert_ilocked(dp, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL); + + *error = 0; + if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return XFS_DIR2_FMT_SF; + + *error = xfs_bmap_last_offset(dp, &eof, XFS_DATA_FORK); + if (*error) + return XFS_DIR2_FMT_ERROR; + + if (eof == XFS_B_TO_FSB(mp, geo->blksize)) { + if (XFS_IS_CORRUPT(mp, dp->i_disk_size != geo->blksize)) { + xfs_da_mark_sick(args); + *error = -EFSCORRUPTED; + return XFS_DIR2_FMT_ERROR; + } + return XFS_DIR2_FMT_BLOCK; + } + if (eof == geo->leafblk + geo->fsbcount) + return XFS_DIR2_FMT_LEAF; + return XFS_DIR2_FMT_NODE; +} + int xfs_dir_createname_args( struct xfs_da_args *args) { - bool is_block, is_leaf; int error; if (!args->inumber) args->op_flags |= XFS_DA_OP_JUSTCHECK; - if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + switch (xfs_dir2_format(args, &error)) { + case XFS_DIR2_FMT_SF: return xfs_dir2_sf_addname(args); - - error = xfs_dir2_isblock(args, &is_block); - if (error) - return error; - if (is_block) + case XFS_DIR2_FMT_BLOCK: return xfs_dir2_block_addname(args); - - error = xfs_dir2_isleaf(args, &is_leaf); - if (error) - return error; - if (is_leaf) + case XFS_DIR2_FMT_LEAF: return xfs_dir2_leaf_addname(args); - return xfs_dir2_node_addname(args); + case XFS_DIR2_FMT_NODE: + return xfs_dir2_node_addname(args); + default: + return error; + } } /* @@ -359,36 +388,25 @@ int xfs_dir_lookup_args( struct xfs_da_args *args) { - bool is_block, is_leaf; int error; - if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) { + switch (xfs_dir2_format(args, &error)) { + case XFS_DIR2_FMT_SF: error = xfs_dir2_sf_lookup(args); - goto out; - } - - /* dir2 functions require that the data fork is loaded */ - error = xfs_iread_extents(args->trans, args->dp, XFS_DATA_FORK); - if (error) - goto out; - - error = xfs_dir2_isblock(args, &is_block); - if (error) - goto out; - - if (is_block) { + break; + case XFS_DIR2_FMT_BLOCK: error = xfs_dir2_block_lookup(args); - goto out; - } - - error = xfs_dir2_isleaf(args, &is_leaf); - if (error) - goto out; - if (is_leaf) + break; + case XFS_DIR2_FMT_LEAF: error = xfs_dir2_leaf_lookup(args); - else + break; + case XFS_DIR2_FMT_NODE: error = xfs_dir2_node_lookup(args); -out: + break; + default: + break; + } + if (error != -EEXIST) return error; return 0; @@ -448,24 +466,20 @@ int xfs_dir_removename_args( struct xfs_da_args *args) { - bool is_block, is_leaf; int error; - if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + switch (xfs_dir2_format(args, &error)) { + case XFS_DIR2_FMT_SF: return xfs_dir2_sf_removename(args); - - error = xfs_dir2_isblock(args, &is_block); - if (error) - return error; - if (is_block) + case XFS_DIR2_FMT_BLOCK: return xfs_dir2_block_removename(args); - - error = xfs_dir2_isleaf(args, &is_leaf); - if (error) - return error; - if (is_leaf) + case XFS_DIR2_FMT_LEAF: return xfs_dir2_leaf_removename(args); - return xfs_dir2_node_removename(args); + case XFS_DIR2_FMT_NODE: + return xfs_dir2_node_removename(args); + default: + return error; + } } /* @@ -509,25 +523,20 @@ int xfs_dir_replace_args( struct xfs_da_args *args) { - bool is_block, is_leaf; int error; - if (args->dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + switch (xfs_dir2_format(args, &error)) { + case XFS_DIR2_FMT_SF: return xfs_dir2_sf_replace(args); - - error = xfs_dir2_isblock(args, &is_block); - if (error) - return error; - if (is_block) + case XFS_DIR2_FMT_BLOCK: return xfs_dir2_block_replace(args); - - error = xfs_dir2_isleaf(args, &is_leaf); - if (error) - return error; - if (is_leaf) + case XFS_DIR2_FMT_LEAF: return xfs_dir2_leaf_replace(args); - - return xfs_dir2_node_replace(args); + case XFS_DIR2_FMT_NODE: + return xfs_dir2_node_replace(args); + default: + return error; + } } /* @@ -633,57 +642,6 @@ xfs_dir2_grow_inode( return 0; } -/* - * See if the directory is a single-block form directory. - */ -int -xfs_dir2_isblock( - struct xfs_da_args *args, - bool *isblock) -{ - struct xfs_mount *mp = args->dp->i_mount; - xfs_fileoff_t eof; - int error; - - error = xfs_bmap_last_offset(args->dp, &eof, XFS_DATA_FORK); - if (error) - return error; - - *isblock = false; - if (XFS_FSB_TO_B(mp, eof) != args->geo->blksize) - return 0; - - *isblock = true; - if (XFS_IS_CORRUPT(mp, args->dp->i_disk_size != args->geo->blksize)) { - xfs_da_mark_sick(args); - return -EFSCORRUPTED; - } - return 0; -} - -/* - * See if the directory is a single-leaf form directory. - */ -int -xfs_dir2_isleaf( - struct xfs_da_args *args, - bool *isleaf) -{ - xfs_fileoff_t eof; - int error; - - error = xfs_bmap_last_offset(args->dp, &eof, XFS_DATA_FORK); - if (error) - return error; - - *isleaf = false; - if (eof != args->geo->leafblk + args->geo->fsbcount) - return 0; - - *isleaf = true; - return 0; -} - /* * Remove the given block from the directory. * This routine is used for data and free blocks, leaf/node are done diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index 6c00fe24a898..6dbe6e9ecb49 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -36,6 +36,16 @@ xfs_dir2_samename( return !memcmp(n1->name, n2->name, n1->len); } +enum xfs_dir2_fmt { + XFS_DIR2_FMT_SF, + XFS_DIR2_FMT_BLOCK, + XFS_DIR2_FMT_LEAF, + XFS_DIR2_FMT_NODE, + XFS_DIR2_FMT_ERROR, +}; + +enum xfs_dir2_fmt xfs_dir2_format(struct xfs_da_args *args, int *error); + /* * Convert inode mode to directory entry filetype */ @@ -79,8 +89,6 @@ extern int xfs_dir2_sf_to_block(struct xfs_da_args *args); /* * Interface routines used by userspace utilities */ -extern int xfs_dir2_isblock(struct xfs_da_args *args, bool *isblock); -extern int xfs_dir2_isleaf(struct xfs_da_args *args, bool *isleaf); extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db, struct xfs_buf *bp); diff --git a/fs/xfs/libxfs/xfs_exchmaps.c b/fs/xfs/libxfs/xfs_exchmaps.c index 44ab6a9235c0..2021396651de 100644 --- a/fs/xfs/libxfs/xfs_exchmaps.c +++ b/fs/xfs/libxfs/xfs_exchmaps.c @@ -465,17 +465,12 @@ xfs_exchmaps_dir_to_sf( }; struct xfs_dir2_sf_hdr sfh; struct xfs_buf *bp; - bool isblock; int size; - int error; + int error = 0; - error = xfs_dir2_isblock(&args, &isblock); - if (error) + if (xfs_dir2_format(&args, &error) != XFS_DIR2_FMT_BLOCK) return error; - if (!isblock) - return 0; - error = xfs_dir3_block_read(tp, xmi->xmi_ip2, xmi->xmi_ip2->i_ino, &bp); if (error) return error; diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 62474d0557c4..bf9199e8df63 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -808,7 +808,8 @@ xchk_directory_blocks( free_lblk = XFS_B_TO_FSB(mp, XFS_DIR2_FREE_OFFSET); /* Is this a block dir? */ - error = xfs_dir2_isblock(&args, &is_block); + if (xfs_dir2_format(&args, &error) == XFS_DIR2_FMT_BLOCK) + is_block = true; if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error)) goto out; diff --git a/fs/xfs/scrub/readdir.c b/fs/xfs/scrub/readdir.c index 0ac77359d8e9..01c9a2dc0f2c 100644 --- a/fs/xfs/scrub/readdir.c +++ b/fs/xfs/scrub/readdir.c @@ -276,7 +276,6 @@ xchk_dir_walk( .trans = sc->tp, .owner = dp->i_ino, }; - bool isblock; int error; if (xfs_is_shutdown(dp->i_mount)) @@ -285,22 +284,17 @@ xchk_dir_walk( ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); xfs_assert_ilocked(dp, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL); - if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + switch (xfs_dir2_format(&args, &error)) { + case XFS_DIR2_FMT_SF: return xchk_dir_walk_sf(sc, dp, dirent_fn, priv); - - /* dir2 functions require that the data fork is loaded */ - error = xfs_iread_extents(sc->tp, dp, XFS_DATA_FORK); - if (error) - return error; - - error = xfs_dir2_isblock(&args, &isblock); - if (error) - return error; - - if (isblock) + case XFS_DIR2_FMT_BLOCK: return xchk_dir_walk_block(sc, dp, dirent_fn, priv); - - return xchk_dir_walk_leaf(sc, dp, dirent_fn, priv); + case XFS_DIR2_FMT_LEAF: + case XFS_DIR2_FMT_NODE: + return xchk_dir_walk_leaf(sc, dp, dirent_fn, priv); + default: + return error; + } } /* diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c index b3abad5a6cd8..06ac5a7de60a 100644 --- a/fs/xfs/xfs_dir2_readdir.c +++ b/fs/xfs/xfs_dir2_readdir.c @@ -516,7 +516,6 @@ xfs_readdir( { struct xfs_da_args args = { NULL }; unsigned int lock_mode; - bool isblock; int error; trace_xfs_readdir(dp); @@ -539,18 +538,18 @@ xfs_readdir( return xfs_dir2_sf_getdents(&args, ctx); lock_mode = xfs_ilock_data_map_shared(dp); - error = xfs_dir2_isblock(&args, &isblock); - if (error) - goto out_unlock; - - if (isblock) { + switch (xfs_dir2_format(&args, &error)) { + case XFS_DIR2_FMT_BLOCK: error = xfs_dir2_block_getdents(&args, ctx, &lock_mode); - goto out_unlock; + break; + case XFS_DIR2_FMT_LEAF: + case XFS_DIR2_FMT_NODE: + error = xfs_dir2_leaf_getdents(&args, ctx, bufsize, &lock_mode); + break; + default: + break; } - error = xfs_dir2_leaf_getdents(&args, ctx, bufsize, &lock_mode); - -out_unlock: if (lock_mode) xfs_iunlock(dp, lock_mode); return error; -- cgit v1.2.3-59-g8ed1b From 55394d29c9e164f2e1991352f1dc8f973c4f1589 Mon Sep 17 00:00:00 2001 From: Dawid Osuchowski Date: Fri, 26 Apr 2024 09:58:54 +0200 Subject: fs: Create anon_inode_getfile_fmode() Creates an anon_inode_getfile_fmode() function that works similarly to anon_inode_getfile() with the addition of being able to set the fmode member. Signed-off-by: Dawid Osuchowski Link: https://lore.kernel.org/r/20240426075854.4723-1-linux@osuchow.ski Signed-off-by: Christian Brauner --- fs/anon_inodes.c | 33 +++++++++++++++++++++++++++++++++ include/linux/anon_inodes.h | 5 +++++ 2 files changed, 38 insertions(+) (limited to 'fs') diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 0496cb5b6eab..42bd1cb7c9cd 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -148,6 +148,38 @@ struct file *anon_inode_getfile(const char *name, } EXPORT_SYMBOL_GPL(anon_inode_getfile); +/** + * anon_inode_getfile_fmode - creates a new file instance by hooking it up to an + * anonymous inode, and a dentry that describe the "class" + * of the file + * + * @name: [in] name of the "class" of the new file + * @fops: [in] file operations for the new file + * @priv: [in] private data for the new file (will be file's private_data) + * @flags: [in] flags + * @f_mode: [in] fmode + * + * Creates a new file by hooking it on a single inode. This is useful for files + * that do not need to have a full-fledged inode in order to operate correctly. + * All the files created with anon_inode_getfile() will share a single inode, + * hence saving memory and avoiding code duplication for the file/inode/dentry + * setup. Allows setting the fmode. Returns the newly created file* or an error + * pointer. + */ +struct file *anon_inode_getfile_fmode(const char *name, + const struct file_operations *fops, + void *priv, int flags, fmode_t f_mode) +{ + struct file *file; + + file = __anon_inode_getfile(name, fops, priv, flags, NULL, false); + if (!IS_ERR(file)) + file->f_mode |= f_mode; + + return file; +} +EXPORT_SYMBOL_GPL(anon_inode_getfile_fmode); + /** * anon_inode_create_getfile - Like anon_inode_getfile(), but creates a new * !S_PRIVATE anon inode rather than reuse the @@ -271,6 +303,7 @@ int anon_inode_create_getfd(const char *name, const struct file_operations *fops return __anon_inode_getfd(name, fops, priv, flags, context_inode, true); } + static int __init anon_inode_init(void) { anon_inode_mnt = kern_mount(&anon_inode_fs_type); diff --git a/include/linux/anon_inodes.h b/include/linux/anon_inodes.h index 93a5f16d03f3..edef565c2a1a 100644 --- a/include/linux/anon_inodes.h +++ b/include/linux/anon_inodes.h @@ -9,12 +9,17 @@ #ifndef _LINUX_ANON_INODES_H #define _LINUX_ANON_INODES_H +#include + struct file_operations; struct inode; struct file *anon_inode_getfile(const char *name, const struct file_operations *fops, void *priv, int flags); +struct file *anon_inode_getfile_fmode(const char *name, + const struct file_operations *fops, + void *priv, int flags, fmode_t f_mode); struct file *anon_inode_create_getfile(const char *name, const struct file_operations *fops, void *priv, int flags, -- cgit v1.2.3-59-g8ed1b From c97f59e276d4e93480f29a70accbd0d7273cf3f5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 26 Apr 2024 12:15:15 +0100 Subject: netfs: Fix the pre-flush when appending to a file in writethrough mode In netfs_perform_write(), when the file is marked NETFS_ICTX_WRITETHROUGH or O_*SYNC or RWF_*SYNC was specified, write-through caching is performed on a buffered file. When setting up for write-through, we flush any conflicting writes in the region and wait for the write to complete, failing if there's a write error to return. The issue arises if we're writing at or above the EOF position because we skip the flush and - more importantly - the wait. This becomes a problem if there's a partial folio at the end of the file that is being written out and we want to make a write to it too. Both the already-running write and the write we start both want to clear the writeback mark, but whoever is second causes a warning looking something like: ------------[ cut here ]------------ R=00000012: folio 11 is not under writeback WARNING: CPU: 34 PID: 654 at fs/netfs/write_collect.c:105 ... CPU: 34 PID: 654 Comm: kworker/u386:27 Tainted: G S ... ... Workqueue: events_unbound netfs_write_collection_worker ... RIP: 0010:netfs_writeback_lookup_folio Fix this by making the flush-and-wait unconditional. It will do nothing if there are no folios in the pagecache and will return quickly if there are no folios in the region specified. Further, move the WBC attachment above the flush call as the flush is going to attach a WBC and detach it again if it is not present - and since we need one anyway we might as well share it. Fixes: 41d8e7673a77 ("netfs: Implement a write-through caching option") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202404161031.468b84f-oliver.sang@intel.com Signed-off-by: David Howells Link: https://lore.kernel.org/r/2150448.1714130115@warthog.procyon.org.uk Reviewed-by: Jeffrey Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: Marc Dionne cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org cc: linux-cifs@vger.kernel.org Signed-off-by: Christian Brauner --- fs/netfs/buffered_write.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 8f13ca8fbc74..267b622d923b 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -172,15 +172,14 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, if (unlikely(test_bit(NETFS_ICTX_WRITETHROUGH, &ctx->flags) || iocb->ki_flags & (IOCB_DSYNC | IOCB_SYNC)) ) { - if (pos < i_size_read(inode)) { - ret = filemap_write_and_wait_range(mapping, pos, pos + iter->count); - if (ret < 0) { - goto out; - } - } - wbc_attach_fdatawrite_inode(&wbc, mapping->host); + ret = filemap_write_and_wait_range(mapping, pos, pos + iter->count); + if (ret < 0) { + wbc_detach_inode(&wbc); + goto out; + } + wreq = netfs_begin_writethrough(iocb, iter->count); if (IS_ERR(wreq)) { wbc_detach_inode(&wbc); -- cgit v1.2.3-59-g8ed1b From 17597b1e18d2fafef2230c987479eccaeddb4628 Mon Sep 17 00:00:00 2001 From: Hongbo Li Date: Wed, 24 Apr 2024 16:42:47 +0800 Subject: erofs: modify the error message when prepare_ondemand_read failed When prepare_ondemand_read failed, wrong error message is printed. The prepare_read is also implemented in cachefiles, so we amend it. Reviewed-by: Gao Xiang Signed-off-by: Hongbo Li Reviewed-by: Jingbo Xu Reviewed-by: Chao Yu Link: https://lore.kernel.org/r/20240424084247.759432-1-lihongbo22@huawei.com Signed-off-by: Gao Xiang --- fs/erofs/fscache.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c index 8aff1a724805..62da538d91cb 100644 --- a/fs/erofs/fscache.c +++ b/fs/erofs/fscache.c @@ -151,7 +151,7 @@ static int erofs_fscache_read_io_async(struct fscache_cookie *cookie, if (WARN_ON(len == 0)) source = NETFS_INVALID_READ; if (source != NETFS_READ_FROM_CACHE) { - erofs_err(NULL, "prepare_read failed (source %d)", source); + erofs_err(NULL, "prepare_ondemand_read failed (source %d)", source); return -EIO; } -- cgit v1.2.3-59-g8ed1b From 07abe43a28b2c660f726d66f5470f7f114f9643a Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Fri, 19 Apr 2024 20:36:10 +0800 Subject: erofs: get rid of erofs_fs_context Instead of allocating the erofs_sb_info in fill_super() allocate it during erofs_init_fs_context() and ensure that erofs can always have the info available during erofs_kill_sb(). After this erofs_fs_context is no longer needed, replace ctx with sbi, no functional changes. Suggested-by: Jingbo Xu Signed-off-by: Baokun Li Reviewed-by: Jingbo Xu Reviewed-by: Gao Xiang Reviewed-by: Chao Yu Link: https://lore.kernel.org/r/20240419123611.947084-2-libaokun1@huawei.com Signed-off-by: Gao Xiang --- fs/erofs/internal.h | 7 ---- fs/erofs/super.c | 116 ++++++++++++++++++++++++---------------------------- 2 files changed, 53 insertions(+), 70 deletions(-) (limited to 'fs') diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 39c67119f43b..d28ccfc0352b 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -84,13 +84,6 @@ struct erofs_dev_context { bool flatdev; }; -struct erofs_fs_context { - struct erofs_mount_opts opt; - struct erofs_dev_context *devs; - char *fsid; - char *domain_id; -}; - /* all filesystem-wide lz4 configurations */ struct erofs_sb_lz4_info { /* # of pages needed for EROFS lz4 rolling decompression */ diff --git a/fs/erofs/super.c b/fs/erofs/super.c index c0eb139adb07..21faa49bc970 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -370,18 +370,18 @@ out: return ret; } -static void erofs_default_options(struct erofs_fs_context *ctx) +static void erofs_default_options(struct erofs_sb_info *sbi) { #ifdef CONFIG_EROFS_FS_ZIP - ctx->opt.cache_strategy = EROFS_ZIP_CACHE_READAROUND; - ctx->opt.max_sync_decompress_pages = 3; - ctx->opt.sync_decompress = EROFS_SYNC_DECOMPRESS_AUTO; + sbi->opt.cache_strategy = EROFS_ZIP_CACHE_READAROUND; + sbi->opt.max_sync_decompress_pages = 3; + sbi->opt.sync_decompress = EROFS_SYNC_DECOMPRESS_AUTO; #endif #ifdef CONFIG_EROFS_FS_XATTR - set_opt(&ctx->opt, XATTR_USER); + set_opt(&sbi->opt, XATTR_USER); #endif #ifdef CONFIG_EROFS_FS_POSIX_ACL - set_opt(&ctx->opt, POSIX_ACL); + set_opt(&sbi->opt, POSIX_ACL); #endif } @@ -426,16 +426,16 @@ static const struct fs_parameter_spec erofs_fs_parameters[] = { static bool erofs_fc_set_dax_mode(struct fs_context *fc, unsigned int mode) { #ifdef CONFIG_FS_DAX - struct erofs_fs_context *ctx = fc->fs_private; + struct erofs_sb_info *sbi = fc->s_fs_info; switch (mode) { case EROFS_MOUNT_DAX_ALWAYS: - set_opt(&ctx->opt, DAX_ALWAYS); - clear_opt(&ctx->opt, DAX_NEVER); + set_opt(&sbi->opt, DAX_ALWAYS); + clear_opt(&sbi->opt, DAX_NEVER); return true; case EROFS_MOUNT_DAX_NEVER: - set_opt(&ctx->opt, DAX_NEVER); - clear_opt(&ctx->opt, DAX_ALWAYS); + set_opt(&sbi->opt, DAX_NEVER); + clear_opt(&sbi->opt, DAX_ALWAYS); return true; default: DBG_BUGON(1); @@ -450,7 +450,7 @@ static bool erofs_fc_set_dax_mode(struct fs_context *fc, unsigned int mode) static int erofs_fc_parse_param(struct fs_context *fc, struct fs_parameter *param) { - struct erofs_fs_context *ctx = fc->fs_private; + struct erofs_sb_info *sbi = fc->s_fs_info; struct fs_parse_result result; struct erofs_device_info *dif; int opt, ret; @@ -463,9 +463,9 @@ static int erofs_fc_parse_param(struct fs_context *fc, case Opt_user_xattr: #ifdef CONFIG_EROFS_FS_XATTR if (result.boolean) - set_opt(&ctx->opt, XATTR_USER); + set_opt(&sbi->opt, XATTR_USER); else - clear_opt(&ctx->opt, XATTR_USER); + clear_opt(&sbi->opt, XATTR_USER); #else errorfc(fc, "{,no}user_xattr options not supported"); #endif @@ -473,16 +473,16 @@ static int erofs_fc_parse_param(struct fs_context *fc, case Opt_acl: #ifdef CONFIG_EROFS_FS_POSIX_ACL if (result.boolean) - set_opt(&ctx->opt, POSIX_ACL); + set_opt(&sbi->opt, POSIX_ACL); else - clear_opt(&ctx->opt, POSIX_ACL); + clear_opt(&sbi->opt, POSIX_ACL); #else errorfc(fc, "{,no}acl options not supported"); #endif break; case Opt_cache_strategy: #ifdef CONFIG_EROFS_FS_ZIP - ctx->opt.cache_strategy = result.uint_32; + sbi->opt.cache_strategy = result.uint_32; #else errorfc(fc, "compression not supported, cache_strategy ignored"); #endif @@ -504,27 +504,27 @@ static int erofs_fc_parse_param(struct fs_context *fc, kfree(dif); return -ENOMEM; } - down_write(&ctx->devs->rwsem); - ret = idr_alloc(&ctx->devs->tree, dif, 0, 0, GFP_KERNEL); - up_write(&ctx->devs->rwsem); + down_write(&sbi->devs->rwsem); + ret = idr_alloc(&sbi->devs->tree, dif, 0, 0, GFP_KERNEL); + up_write(&sbi->devs->rwsem); if (ret < 0) { kfree(dif->path); kfree(dif); return ret; } - ++ctx->devs->extra_devices; + ++sbi->devs->extra_devices; break; #ifdef CONFIG_EROFS_FS_ONDEMAND case Opt_fsid: - kfree(ctx->fsid); - ctx->fsid = kstrdup(param->string, GFP_KERNEL); - if (!ctx->fsid) + kfree(sbi->fsid); + sbi->fsid = kstrdup(param->string, GFP_KERNEL); + if (!sbi->fsid) return -ENOMEM; break; case Opt_domain_id: - kfree(ctx->domain_id); - ctx->domain_id = kstrdup(param->string, GFP_KERNEL); - if (!ctx->domain_id) + kfree(sbi->domain_id); + sbi->domain_id = kstrdup(param->string, GFP_KERNEL); + if (!sbi->domain_id) return -ENOMEM; break; #else @@ -581,8 +581,7 @@ static const struct export_operations erofs_export_ops = { static int erofs_fc_fill_super(struct super_block *sb, struct fs_context *fc) { struct inode *inode; - struct erofs_sb_info *sbi; - struct erofs_fs_context *ctx = fc->fs_private; + struct erofs_sb_info *sbi = EROFS_SB(sb); int err; sb->s_magic = EROFS_SUPER_MAGIC; @@ -590,19 +589,6 @@ static int erofs_fc_fill_super(struct super_block *sb, struct fs_context *fc) sb->s_maxbytes = MAX_LFS_FILESIZE; sb->s_op = &erofs_sops; - sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); - if (!sbi) - return -ENOMEM; - - sb->s_fs_info = sbi; - sbi->opt = ctx->opt; - sbi->devs = ctx->devs; - ctx->devs = NULL; - sbi->fsid = ctx->fsid; - ctx->fsid = NULL; - sbi->domain_id = ctx->domain_id; - ctx->domain_id = NULL; - sbi->blkszbits = PAGE_SHIFT; if (erofs_is_fscache_mode(sb)) { sb->s_blocksize = PAGE_SIZE; @@ -706,9 +692,9 @@ static int erofs_fc_fill_super(struct super_block *sb, struct fs_context *fc) static int erofs_fc_get_tree(struct fs_context *fc) { - struct erofs_fs_context *ctx = fc->fs_private; + struct erofs_sb_info *sbi = fc->s_fs_info; - if (IS_ENABLED(CONFIG_EROFS_FS_ONDEMAND) && ctx->fsid) + if (IS_ENABLED(CONFIG_EROFS_FS_ONDEMAND) && sbi->fsid) return get_tree_nodev(fc, erofs_fc_fill_super); return get_tree_bdev(fc, erofs_fc_fill_super); @@ -718,19 +704,19 @@ static int erofs_fc_reconfigure(struct fs_context *fc) { struct super_block *sb = fc->root->d_sb; struct erofs_sb_info *sbi = EROFS_SB(sb); - struct erofs_fs_context *ctx = fc->fs_private; + struct erofs_sb_info *new_sbi = fc->s_fs_info; DBG_BUGON(!sb_rdonly(sb)); - if (ctx->fsid || ctx->domain_id) + if (new_sbi->fsid || new_sbi->domain_id) erofs_info(sb, "ignoring reconfiguration for fsid|domain_id."); - if (test_opt(&ctx->opt, POSIX_ACL)) + if (test_opt(&new_sbi->opt, POSIX_ACL)) fc->sb_flags |= SB_POSIXACL; else fc->sb_flags &= ~SB_POSIXACL; - sbi->opt = ctx->opt; + sbi->opt = new_sbi->opt; fc->sb_flags |= SB_RDONLY; return 0; @@ -761,12 +747,15 @@ static void erofs_free_dev_context(struct erofs_dev_context *devs) static void erofs_fc_free(struct fs_context *fc) { - struct erofs_fs_context *ctx = fc->fs_private; + struct erofs_sb_info *sbi = fc->s_fs_info; + + if (!sbi) + return; - erofs_free_dev_context(ctx->devs); - kfree(ctx->fsid); - kfree(ctx->domain_id); - kfree(ctx); + erofs_free_dev_context(sbi->devs); + kfree(sbi->fsid); + kfree(sbi->domain_id); + kfree(sbi); } static const struct fs_context_operations erofs_context_ops = { @@ -778,21 +767,22 @@ static const struct fs_context_operations erofs_context_ops = { static int erofs_init_fs_context(struct fs_context *fc) { - struct erofs_fs_context *ctx; + struct erofs_sb_info *sbi; - ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); - if (!ctx) + sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); + if (!sbi) return -ENOMEM; - ctx->devs = kzalloc(sizeof(struct erofs_dev_context), GFP_KERNEL); - if (!ctx->devs) { - kfree(ctx); + + sbi->devs = kzalloc(sizeof(struct erofs_dev_context), GFP_KERNEL); + if (!sbi->devs) { + kfree(sbi); return -ENOMEM; } - fc->fs_private = ctx; + fc->s_fs_info = sbi; - idr_init(&ctx->devs->tree); - init_rwsem(&ctx->devs->rwsem); - erofs_default_options(ctx); + idr_init(&sbi->devs->tree); + init_rwsem(&sbi->devs->rwsem); + erofs_default_options(sbi); fc->ops = &erofs_context_ops; return 0; } -- cgit v1.2.3-59-g8ed1b From 7af2ae1b1531feab5d38ec9c8f472dc6cceb4606 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Fri, 19 Apr 2024 20:36:11 +0800 Subject: erofs: reliably distinguish block based and fscache mode When erofs_kill_sb() is called in block dev based mode, s_bdev may not have been initialised yet, and if CONFIG_EROFS_FS_ONDEMAND is enabled, it will be mistaken for fscache mode, and then attempt to free an anon_dev that has never been allocated, triggering the following warning: ============================================ ida_free called for id=0 which is not allocated. WARNING: CPU: 14 PID: 926 at lib/idr.c:525 ida_free+0x134/0x140 Modules linked in: CPU: 14 PID: 926 Comm: mount Not tainted 6.9.0-rc3-dirty #630 RIP: 0010:ida_free+0x134/0x140 Call Trace: erofs_kill_sb+0x81/0x90 deactivate_locked_super+0x35/0x80 get_tree_bdev+0x136/0x1e0 vfs_get_tree+0x2c/0xf0 do_new_mount+0x190/0x2f0 [...] ============================================ Now when erofs_kill_sb() is called, erofs_sb_info must have been initialised, so use sbi->fsid to distinguish between the two modes. Signed-off-by: Christian Brauner Signed-off-by: Baokun Li Reviewed-by: Jingbo Xu Reviewed-by: Gao Xiang Reviewed-by: Chao Yu Link: https://lore.kernel.org/r/20240419123611.947084-3-libaokun1@huawei.com Signed-off-by: Gao Xiang --- fs/erofs/super.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 21faa49bc970..30b49b2eee53 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -789,17 +789,13 @@ static int erofs_init_fs_context(struct fs_context *fc) static void erofs_kill_sb(struct super_block *sb) { - struct erofs_sb_info *sbi; + struct erofs_sb_info *sbi = EROFS_SB(sb); - if (erofs_is_fscache_mode(sb)) + if (IS_ENABLED(CONFIG_EROFS_FS_ONDEMAND) && sbi->fsid) kill_anon_super(sb); else kill_block_super(sb); - sbi = EROFS_SB(sb); - if (!sbi) - return; - erofs_free_dev_context(sbi->devs); fs_put_dax(sbi->dax_dev, NULL); erofs_fscache_unregister_fs(sb); -- cgit v1.2.3-59-g8ed1b From bc642d7bfdac3bfd838a1cd6651955ae2eb8535a Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Wed, 24 Apr 2024 16:57:01 +0900 Subject: ksmbd: fix uninitialized symbol 'share' in smb2_tree_connect() Fix uninitialized symbol 'share' in smb2_tree_connect(). Fixes: e9d8c2f95ab8 ("ksmbd: add continuous availability share parameter") Reported-by: kernel test robot Reported-by: Dan Carpenter Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 355824151c2d..30229161b346 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -1926,7 +1926,7 @@ int smb2_tree_connect(struct ksmbd_work *work) struct ksmbd_session *sess = work->sess; char *treename = NULL, *name = NULL; struct ksmbd_tree_conn_status status; - struct ksmbd_share_config *share; + struct ksmbd_share_config *share = NULL; int rc = -EINVAL; WORK_BUFFERS(work, req, rsp); @@ -1988,7 +1988,7 @@ int smb2_tree_connect(struct ksmbd_work *work) write_unlock(&sess->tree_conns_lock); rsp->StructureSize = cpu_to_le16(16); out_err1: - if (server_conf.flags & KSMBD_GLOBAL_FLAG_DURABLE_HANDLE && + if (server_conf.flags & KSMBD_GLOBAL_FLAG_DURABLE_HANDLE && share && test_share_config_flag(share, KSMBD_SHARE_FLAG_CONTINUOUS_AVAILABILITY)) rsp->Capabilities = SMB2_SHARE_CAP_CONTINUOUS_AVAILABILITY; -- cgit v1.2.3-59-g8ed1b From ae92765373c3bd82575041cf2910c96e1ba03118 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 22 Apr 2024 17:33:20 -0400 Subject: bcachefs: Remove accidental debug assert Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 941401a210f5..82f179258867 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -525,7 +525,6 @@ int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca, "different types of data in same bucket: %s, %s", bch2_data_type_str(g->data_type), bch2_data_type_str(data_type))) { - BUG(); ret = -EIO; goto err; } @@ -629,7 +628,6 @@ int bch2_check_bucket_ref(struct btree_trans *trans, bch2_data_type_str(ptr_data_type), (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf)); - BUG(); ret = -EIO; goto err; } -- cgit v1.2.3-59-g8ed1b From f7c3dc2646584cddae6fedc517cd58d97483e5cc Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 25 Apr 2024 03:55:48 -0400 Subject: bcachefs: btree node scan now fills in sectors_written Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_node_scan.c | 7 +++++-- fs/bcachefs/btree_node_scan_types.h | 1 + 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index c60794264da2..45cb8149d374 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -57,13 +57,14 @@ static void found_btree_node_to_key(struct bkey_i *k, const struct found_btree_n bp->v.seq = cpu_to_le64(f->cookie); bp->v.sectors_written = 0; bp->v.flags = 0; + bp->v.sectors_written = cpu_to_le16(f->sectors_written); bp->v.min_key = f->min_key; SET_BTREE_PTR_RANGE_UPDATED(&bp->v, f->range_updated); memcpy(bp->v.start, f->ptrs, sizeof(struct bch_extent_ptr) * f->nr_ptrs); } static bool found_btree_node_is_readable(struct btree_trans *trans, - const struct found_btree_node *f) + struct found_btree_node *f) { struct { __BKEY_PADDED(k, BKEY_BTREE_PTR_VAL_U64s_MAX); } k; @@ -71,8 +72,10 @@ static bool found_btree_node_is_readable(struct btree_trans *trans, struct btree *b = bch2_btree_node_get_noiter(trans, &k.k, f->btree_id, f->level, false); bool ret = !IS_ERR_OR_NULL(b); - if (ret) + if (ret) { + f->sectors_written = b->written; six_unlock_read(&b->c.lock); + } /* * We might update this node's range; if that happens, we need the node diff --git a/fs/bcachefs/btree_node_scan_types.h b/fs/bcachefs/btree_node_scan_types.h index abb7b27d556a..5cfaeb5ac831 100644 --- a/fs/bcachefs/btree_node_scan_types.h +++ b/fs/bcachefs/btree_node_scan_types.h @@ -9,6 +9,7 @@ struct found_btree_node { bool overwritten:1; u8 btree_id; u8 level; + unsigned sectors_written; u32 seq; u64 cookie; -- cgit v1.2.3-59-g8ed1b From c258c08add1cc8fa7719f112c5db36c08c507f1e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 25 Apr 2024 14:41:17 -0400 Subject: bcachefs: fix integer conversion bug Signed-off-by: Kent Overstreet --- fs/bcachefs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index ca4a066e9a54..0f95d7fb5ec0 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -606,7 +606,7 @@ int bch2_trigger_inode(struct btree_trans *trans, struct bkey_s new, unsigned flags) { - s64 nr = bkey_is_inode(new.k) - bkey_is_inode(old.k); + s64 nr = (s64) bkey_is_inode(new.k) - (s64) bkey_is_inode(old.k); if (flags & BTREE_TRIGGER_TRANSACTIONAL) { if (nr) { -- cgit v1.2.3-59-g8ed1b From fcd63086bc14b1689866d7b0c61958f6b7e48604 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Sun, 7 Apr 2024 12:55:44 +0200 Subject: gfs2: gfs2_freeze_unlock cleanup Function gfs2_freeze_unlock() is always called with &sdp->sd_freeze_gh as its argument, so clean up the code by passing in sdp instead. Signed-off-by: Andreas Gruenbacher --- fs/gfs2/ops_fstype.c | 2 +- fs/gfs2/super.c | 12 ++++++------ fs/gfs2/util.c | 6 +++--- fs/gfs2/util.h | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index db0df091a6a7..a9f7f0d44227 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1289,7 +1289,7 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc) error = gfs2_make_fs_rw(sdp); if (error) { - gfs2_freeze_unlock(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(sdp); gfs2_destroy_threads(sdp); fs_err(sdp, "can't make FS RW: %d\n", error); goto fail_per_node; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 32a37daaccbe..713dd24082c9 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -358,7 +358,7 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp) list_add(&lfcc->list, &list); } - gfs2_freeze_unlock(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(sdp); error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_EXCLUSIVE, LM_FLAG_NOEXP | GL_NOPID, @@ -382,7 +382,7 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp) if (!error) goto out; /* success */ - gfs2_freeze_unlock(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(sdp); relock_shared: error2 = gfs2_freeze_lock_shared(sdp); @@ -621,7 +621,7 @@ restart: /* Release stuff */ - gfs2_freeze_unlock(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(sdp); iput(sdp->sd_jindex); iput(sdp->sd_statfs_inode); @@ -707,7 +707,7 @@ void gfs2_freeze_func(struct work_struct *work) if (error) goto freeze_failed; - gfs2_freeze_unlock(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(sdp); set_bit(SDF_FROZEN, &sdp->sd_flags); error = gfs2_do_thaw(sdp); @@ -812,7 +812,7 @@ static int gfs2_thaw_super(struct super_block *sb, enum freeze_holder who) } atomic_inc(&sb->s_active); - gfs2_freeze_unlock(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(sdp); error = gfs2_do_thaw(sdp); @@ -833,7 +833,7 @@ void gfs2_thaw_freeze_initiator(struct super_block *sb) if (!test_bit(SDF_FREEZE_INITIATOR, &sdp->sd_flags)) goto out; - gfs2_freeze_unlock(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(sdp); out: mutex_unlock(&sdp->sd_freeze_mutex); diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index 09238604d741..af4758d8d894 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -109,10 +109,10 @@ int gfs2_freeze_lock_shared(struct gfs2_sbd *sdp) return error; } -void gfs2_freeze_unlock(struct gfs2_holder *freeze_gh) +void gfs2_freeze_unlock(struct gfs2_sbd *sdp) { - if (gfs2_holder_initialized(freeze_gh)) - gfs2_glock_dq_uninit(freeze_gh); + if (gfs2_holder_initialized(&sdp->sd_freeze_gh)) + gfs2_glock_dq_uninit(&sdp->sd_freeze_gh); } static void signal_our_withdraw(struct gfs2_sbd *sdp) diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index d8d62bf0424d..27d03b641024 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -150,7 +150,7 @@ int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function, int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, bool verbose); int gfs2_freeze_lock_shared(struct gfs2_sbd *sdp); -void gfs2_freeze_unlock(struct gfs2_holder *freeze_gh); +void gfs2_freeze_unlock(struct gfs2_sbd *sdp); #define gfs2_io_error(sdp) \ gfs2_io_error_i((sdp), __func__, __FILE__, __LINE__) -- cgit v1.2.3-59-g8ed1b From f3851fed07327b6a19e7ff8c2106e2b424f44cca Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:23:48 +0100 Subject: gfs2: Convert gfs2_page_mkwrite() to use a folio Convert the incoming page to a folio and use it throughout, saving several calls to compound_head(). Also use 'pos' for file position rather than the ambiguous 'offset' and convert 'length' to type size_t in case we get some truly ridiculous sized folios in the future. This function should now be large-folio safe, but I may have missed something. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Andreas Gruenbacher --- fs/gfs2/file.c | 59 +++++++++++++++++++++++++++++----------------------------- 1 file changed, 30 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 992ca4effb50..649f808db9cc 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -376,23 +376,23 @@ static void gfs2_size_hint(struct file *filep, loff_t offset, size_t size) } /** - * gfs2_allocate_page_backing - Allocate blocks for a write fault - * @page: The (locked) page to allocate backing for + * gfs2_allocate_folio_backing - Allocate blocks for a write fault + * @folio: The (locked) folio to allocate backing for * @length: Size of the allocation * - * We try to allocate all the blocks required for the page in one go. This + * We try to allocate all the blocks required for the folio in one go. This * might fail for various reasons, so we keep trying until all the blocks to - * back this page are allocated. If some of the blocks are already allocated, + * back this folio are allocated. If some of the blocks are already allocated, * that is ok too. */ -static int gfs2_allocate_page_backing(struct page *page, unsigned int length) +static int gfs2_allocate_folio_backing(struct folio *folio, size_t length) { - u64 pos = page_offset(page); + u64 pos = folio_pos(folio); do { struct iomap iomap = { }; - if (gfs2_iomap_alloc(page->mapping->host, pos, length, &iomap)) + if (gfs2_iomap_alloc(folio->mapping->host, pos, length, &iomap)) return -EIO; if (length < iomap.length) @@ -414,16 +414,16 @@ static int gfs2_allocate_page_backing(struct page *page, unsigned int length) static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf) { - struct page *page = vmf->page; + struct folio *folio = page_folio(vmf->page); struct inode *inode = file_inode(vmf->vma->vm_file); struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); struct gfs2_alloc_parms ap = {}; - u64 offset = page_offset(page); + u64 pos = folio_pos(folio); unsigned int data_blocks, ind_blocks, rblocks; vm_fault_t ret = VM_FAULT_LOCKED; struct gfs2_holder gh; - unsigned int length; + size_t length; loff_t size; int err; @@ -436,23 +436,23 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf) goto out_uninit; } - /* Check page index against inode size */ + /* Check folio index against inode size */ size = i_size_read(inode); - if (offset >= size) { + if (pos >= size) { ret = VM_FAULT_SIGBUS; goto out_unlock; } - /* Update file times before taking page lock */ + /* Update file times before taking folio lock */ file_update_time(vmf->vma->vm_file); - /* page is wholly or partially inside EOF */ - if (size - offset < PAGE_SIZE) - length = size - offset; + /* folio is wholly or partially inside EOF */ + if (size - pos < folio_size(folio)) + length = size - pos; else - length = PAGE_SIZE; + length = folio_size(folio); - gfs2_size_hint(vmf->vma->vm_file, offset, length); + gfs2_size_hint(vmf->vma->vm_file, pos, length); set_bit(GLF_DIRTY, &ip->i_gl->gl_flags); set_bit(GIF_SW_PAGED, &ip->i_flags); @@ -463,11 +463,12 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf) */ if (!gfs2_is_stuffed(ip) && - !gfs2_write_alloc_required(ip, offset, length)) { - lock_page(page); - if (!PageUptodate(page) || page->mapping != inode->i_mapping) { + !gfs2_write_alloc_required(ip, pos, length)) { + folio_lock(folio); + if (!folio_test_uptodate(folio) || + folio->mapping != inode->i_mapping) { ret = VM_FAULT_NOPAGE; - unlock_page(page); + folio_unlock(folio); } goto out_unlock; } @@ -504,7 +505,7 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf) goto out_trans_fail; } - /* Unstuff, if required, and allocate backing blocks for page */ + /* Unstuff, if required, and allocate backing blocks for folio */ if (gfs2_is_stuffed(ip)) { err = gfs2_unstuff_dinode(ip); if (err) { @@ -513,22 +514,22 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf) } } - lock_page(page); + folio_lock(folio); /* If truncated, we must retry the operation, we may have raced * with the glock demotion code. */ - if (!PageUptodate(page) || page->mapping != inode->i_mapping) { + if (!folio_test_uptodate(folio) || folio->mapping != inode->i_mapping) { ret = VM_FAULT_NOPAGE; goto out_page_locked; } - err = gfs2_allocate_page_backing(page, length); + err = gfs2_allocate_folio_backing(folio, length); if (err) ret = vmf_fs_error(err); out_page_locked: if (ret != VM_FAULT_LOCKED) - unlock_page(page); + folio_unlock(folio); out_trans_end: gfs2_trans_end(sdp); out_trans_fail: @@ -540,8 +541,8 @@ out_unlock: out_uninit: gfs2_holder_uninit(&gh); if (ret == VM_FAULT_LOCKED) { - set_page_dirty(page); - wait_for_stable_page(page); + folio_mark_dirty(folio); + folio_wait_stable(folio); } sb_end_pagefault(inode->i_sb); return ret; -- cgit v1.2.3-59-g8ed1b From bb712842a85d595525e72f0e378c143e620b3ea2 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Thu, 25 Apr 2024 21:13:27 +0800 Subject: xfs: match lock mode in xfs_buffered_write_iomap_begin() Commit 1aa91d9c9933 ("xfs: Add async buffered write support") replace xfs_ilock(XFS_ILOCK_EXCL) with xfs_ilock_for_iomap() when locking the writing inode, and a new variable lockmode is used to indicate the lock mode. Although the lockmode should always be XFS_ILOCK_EXCL, it's still better to use this variable instead of useing XFS_ILOCK_EXCL directly when unlocking the inode. Fixes: 1aa91d9c9933 ("xfs: Add async buffered write support") Signed-off-by: Zhang Yi Reviewed-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig Signed-off-by: Chandan Babu R --- fs/xfs/xfs_iomap.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 9ce0f6b9df93..555603ef4234 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -1175,13 +1175,13 @@ retry: * them out if the write happens to fail. */ seq = xfs_iomap_inode_sequence(ip, IOMAP_F_NEW); - xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(ip, lockmode); trace_xfs_iomap_alloc(ip, offset, count, allocfork, &imap); return xfs_bmbt_to_iomap(ip, iomap, &imap, flags, IOMAP_F_NEW, seq); found_imap: seq = xfs_iomap_inode_sequence(ip, 0); - xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(ip, lockmode); return xfs_bmbt_to_iomap(ip, iomap, &imap, flags, 0, seq); found_cow: @@ -1191,17 +1191,17 @@ found_cow: if (error) goto out_unlock; seq = xfs_iomap_inode_sequence(ip, IOMAP_F_SHARED); - xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(ip, lockmode); return xfs_bmbt_to_iomap(ip, iomap, &cmap, flags, IOMAP_F_SHARED, seq); } xfs_trim_extent(&cmap, offset_fsb, imap.br_startoff - offset_fsb); - xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(ip, lockmode); return xfs_bmbt_to_iomap(ip, iomap, &cmap, flags, 0, seq); out_unlock: - xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_iunlock(ip, lockmode); return error; } -- cgit v1.2.3-59-g8ed1b From fc8d0ba0ff5fe4700fa02008b7751ec6b84b7677 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Thu, 25 Apr 2024 21:13:28 +0800 Subject: xfs: make the seq argument to xfs_bmapi_convert_delalloc() optional Allow callers to pass a NULLL seq argument if they don't care about the fork sequence number. Signed-off-by: Zhang Yi Reviewed-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 6053f5e5c71e..c9b1dfe63944 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4648,7 +4648,8 @@ xfs_bmapi_convert_delalloc( if (!isnullstartblock(bma.got.br_startblock)) { xfs_bmbt_to_iomap(ip, iomap, &bma.got, 0, flags, xfs_iomap_inode_sequence(ip, flags)); - *seq = READ_ONCE(ifp->if_seq); + if (seq) + *seq = READ_ONCE(ifp->if_seq); goto out_trans_cancel; } @@ -4699,7 +4700,8 @@ xfs_bmapi_convert_delalloc( ASSERT(!isnullstartblock(bma.got.br_startblock)); xfs_bmbt_to_iomap(ip, iomap, &bma.got, 0, flags, xfs_iomap_inode_sequence(ip, flags)); - *seq = READ_ONCE(ifp->if_seq); + if (seq) + *seq = READ_ONCE(ifp->if_seq); if (whichfork == XFS_COW_FORK) xfs_refcount_alloc_cow_extent(tp, bma.blkno, bma.length); -- cgit v1.2.3-59-g8ed1b From 2e08371a83f1c06fd85eea8cd37c87a224cc4cc4 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Thu, 25 Apr 2024 21:13:29 +0800 Subject: xfs: make xfs_bmapi_convert_delalloc() to allocate the target offset Since xfs_bmapi_convert_delalloc() only attempts to allocate the entire delalloc extent and require multiple invocations to allocate the target offset. So xfs_convert_blocks() add a loop to do this job and we call it in the write back path, but xfs_convert_blocks() isn't a common helper. Let's do it in xfs_bmapi_convert_delalloc() and drop xfs_convert_blocks(), preparing for the post EOF delalloc blocks converting in the buffered write begin path. Signed-off-by: Zhang Yi Reviewed-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 34 ++++++++++++++++++++++++++++-- fs/xfs/xfs_aops.c | 54 +++++++++++++----------------------------------- 2 files changed, 46 insertions(+), 42 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index c9b1dfe63944..49e2fcad593d 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4590,8 +4590,8 @@ error0: * invocations to allocate the target offset if a large enough physical extent * is not available. */ -int -xfs_bmapi_convert_delalloc( +static int +xfs_bmapi_convert_one_delalloc( struct xfs_inode *ip, int whichfork, xfs_off_t offset, @@ -4724,6 +4724,36 @@ out_trans_cancel: return error; } +/* + * Pass in a dellalloc extent and convert it to real extents, return the real + * extent that maps offset_fsb in iomap. + */ +int +xfs_bmapi_convert_delalloc( + struct xfs_inode *ip, + int whichfork, + loff_t offset, + struct iomap *iomap, + unsigned int *seq) +{ + int error; + + /* + * Attempt to allocate whatever delalloc extent currently backs offset + * and put the result into iomap. Allocate in a loop because it may + * take several attempts to allocate real blocks for a contiguous + * delalloc extent if free space is sufficiently fragmented. + */ + do { + error = xfs_bmapi_convert_one_delalloc(ip, whichfork, offset, + iomap, seq); + if (error) + return error; + } while (iomap->offset + iomap->length <= offset); + + return 0; +} + int xfs_bmapi_remap( struct xfs_trans *tp, diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 3f428620ebf2..f7520f375cee 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -233,45 +233,6 @@ xfs_imap_valid( return true; } -/* - * Pass in a dellalloc extent and convert it to real extents, return the real - * extent that maps offset_fsb in wpc->iomap. - * - * The current page is held locked so nothing could have removed the block - * backing offset_fsb, although it could have moved from the COW to the data - * fork by another thread. - */ -static int -xfs_convert_blocks( - struct iomap_writepage_ctx *wpc, - struct xfs_inode *ip, - int whichfork, - loff_t offset) -{ - int error; - unsigned *seq; - - if (whichfork == XFS_COW_FORK) - seq = &XFS_WPC(wpc)->cow_seq; - else - seq = &XFS_WPC(wpc)->data_seq; - - /* - * Attempt to allocate whatever delalloc extent currently backs offset - * and put the result into wpc->iomap. Allocate in a loop because it - * may take several attempts to allocate real blocks for a contiguous - * delalloc extent if free space is sufficiently fragmented. - */ - do { - error = xfs_bmapi_convert_delalloc(ip, whichfork, offset, - &wpc->iomap, seq); - if (error) - return error; - } while (wpc->iomap.offset + wpc->iomap.length <= offset); - - return 0; -} - static int xfs_map_blocks( struct iomap_writepage_ctx *wpc, @@ -290,6 +251,7 @@ xfs_map_blocks( struct xfs_iext_cursor icur; int retries = 0; int error = 0; + unsigned int *seq; if (xfs_is_shutdown(mp)) return -EIO; @@ -387,7 +349,19 @@ retry: trace_xfs_map_blocks_found(ip, offset, count, whichfork, &imap); return 0; allocate_blocks: - error = xfs_convert_blocks(wpc, ip, whichfork, offset); + /* + * Convert a dellalloc extent to a real one. The current page is held + * locked so nothing could have removed the block backing offset_fsb, + * although it could have moved from the COW to the data fork by another + * thread. + */ + if (whichfork == XFS_COW_FORK) + seq = &XFS_WPC(wpc)->cow_seq; + else + seq = &XFS_WPC(wpc)->data_seq; + + error = xfs_bmapi_convert_delalloc(ip, whichfork, offset, + &wpc->iomap, seq); if (error) { /* * If we failed to find the extent in the COW fork we might have -- cgit v1.2.3-59-g8ed1b From 5ce5674187c345dc31534d2024c09ad8ef29b7ba Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Thu, 25 Apr 2024 21:13:30 +0800 Subject: xfs: convert delayed extents to unwritten when zeroing post eof blocks Current clone operation could be non-atomic if the destination of a file is beyond EOF, user could get a file with corrupted (zeroed) data on crash. The problem is about preallocations. If you write some data into a file: [A...B) and XFS decides to preallocate some post-eof blocks, then it can create a delayed allocation reservation: [A.........D) The writeback path tries to convert delayed extents to real ones by allocating blocks. If there aren't enough contiguous free space, we can end up with two extents, the first real and the second still delalloc: [A....C)[C.D) After that, both the in-memory and the on-disk file sizes are still B. If we clone into the range [E...F) from another file: [A....C)[C.D) [E...F) then xfs_reflink_zero_posteof() calls iomap_zero_range() to zero out the range [B, E) beyond EOF and flush it. Since [C, D) is still a delalloc extent, its pagecache will be zeroed and both the in-memory and on-disk size will be updated to D after flushing but before cloning. This is wrong, because the user can see the size change and read the zeroes while the clone operation is ongoing. We need to keep the in-memory and on-disk size before the clone operation starts, so instead of writing zeroes through the page cache for delayed ranges beyond EOF, we convert these ranges to unwritten and invalidate any cached data over that range beyond EOF. Suggested-by: Dave Chinner Signed-off-by: Zhang Yi Reviewed-by: "Darrick J. Wong" Reviewed-by: Christoph Hellwig Signed-off-by: Chandan Babu R --- fs/xfs/xfs_iomap.c | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) (limited to 'fs') diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 555603ef4234..ec43aa7cec60 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -1039,6 +1039,24 @@ xfs_buffered_write_iomap_begin( goto out_unlock; } + /* + * For zeroing, trim a delalloc extent that extends beyond the EOF + * block. If it starts beyond the EOF block, convert it to an + * unwritten extent. + */ + if ((flags & IOMAP_ZERO) && imap.br_startoff <= offset_fsb && + isnullstartblock(imap.br_startblock)) { + xfs_fileoff_t eof_fsb = XFS_B_TO_FSB(mp, XFS_ISIZE(ip)); + + if (offset_fsb >= eof_fsb) + goto convert_delay; + if (end_fsb > eof_fsb) { + end_fsb = eof_fsb; + xfs_trim_extent(&imap, offset_fsb, + end_fsb - offset_fsb); + } + } + /* * Search the COW fork extent list even if we did not find a data fork * extent. This serves two purposes: first this implements the @@ -1184,6 +1202,17 @@ found_imap: xfs_iunlock(ip, lockmode); return xfs_bmbt_to_iomap(ip, iomap, &imap, flags, 0, seq); +convert_delay: + xfs_iunlock(ip, lockmode); + truncate_pagecache(inode, offset); + error = xfs_bmapi_convert_delalloc(ip, XFS_DATA_FORK, offset, + iomap, NULL); + if (error) + return error; + + trace_xfs_iomap_alloc(ip, offset, count, XFS_DATA_FORK, &imap); + return 0; + found_cow: seq = xfs_iomap_inode_sequence(ip, 0); if (imap.br_startoff <= offset_fsb) { -- cgit v1.2.3-59-g8ed1b From 5f24162f873f08681804059e6de70d77c3e4cea2 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 23 Feb 2024 08:04:33 +0000 Subject: netfs: Update i_blocks when write committed to pagecache Update i_blocks when i_size is updated when we finish making a write to the pagecache to reflect the amount of space we think will be consumed. This maintains cifs commit dbfdff402d89854126658376cbcb08363194d3cd ("smb3: update allocation size more accurately on write completion") which would otherwise be removed by the cifs part of the netfs writeback rewrite. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/netfs/buffered_write.c | 45 ++++++++++++++++++++++++++++++++++----------- 1 file changed, 34 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 267b622d923b..f7455a579f21 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -130,6 +130,37 @@ static struct folio *netfs_grab_folio_for_write(struct address_space *mapping, mapping_gfp_mask(mapping)); } +/* + * Update i_size and estimate the update to i_blocks to reflect the additional + * data written into the pagecache until we can find out from the server what + * the values actually are. + */ +static void netfs_update_i_size(struct netfs_inode *ctx, struct inode *inode, + loff_t i_size, loff_t pos, size_t copied) +{ + blkcnt_t add; + size_t gap; + + if (ctx->ops->update_i_size) { + ctx->ops->update_i_size(inode, pos); + return; + } + + i_size_write(inode, pos); +#if IS_ENABLED(CONFIG_FSCACHE) + fscache_update_cookie(ctx->cache, NULL, &pos); +#endif + + gap = SECTOR_SIZE - (i_size & (SECTOR_SIZE - 1)); + if (copied > gap) { + add = DIV_ROUND_UP(copied - gap, SECTOR_SIZE); + + inode->i_blocks = min_t(blkcnt_t, + DIV_ROUND_UP(pos, SECTOR_SIZE), + inode->i_blocks + add); + } +} + /** * netfs_perform_write - Copy data into the pagecache. * @iocb: The operation parameters @@ -351,18 +382,10 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, trace_netfs_folio(folio, trace); /* Update the inode size if we moved the EOF marker */ - i_size = i_size_read(inode); pos += copied; - if (pos > i_size) { - if (ctx->ops->update_i_size) { - ctx->ops->update_i_size(inode, pos); - } else { - i_size_write(inode, pos); -#if IS_ENABLED(CONFIG_FSCACHE) - fscache_update_cookie(ctx->cache, NULL, &pos); -#endif - } - } + i_size = i_size_read(inode); + if (pos > i_size) + netfs_update_i_size(ctx, inode, i_size, pos, copied); written += copied; if (likely(!wreq)) { -- cgit v1.2.3-59-g8ed1b From 2ff1e97587f4d398686f52c07afde3faf3da4e5c Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 19 Mar 2024 10:00:09 +0000 Subject: netfs: Replace PG_fscache by setting folio->private and marking dirty When dirty data is being written to the cache, setting/waiting on/clearing the fscache flag is always done in tandem with setting/waiting on/clearing the writeback flag. The netfslib buffered write routines wait on and set both flags and the write request cleanup clears both flags, so the fscache flag is almost superfluous. The reason it isn't superfluous is because the fscache flag is also used to indicate that data just read from the server is being written to the cache. The flag is used to prevent a race involving overlapping direct-I/O writes to the cache. Change this to indicate that a page is in need of being copied to the cache by placing a magic value in folio->private and marking the folios dirty. Then when the writeback code sees a folio marked in this way, it only writes it to the cache and not to the server. If a folio that has this magic value set is modified, the value is just replaced and the folio will then be uplodaded too. With this, PG_fscache is no longer required by the netfslib core, 9p and afs. Ceph and nfs, however, still need to use the old PG_fscache-based tracking. To deal with this, a flag, NETFS_ICTX_USE_PGPRIV2, now has to be set on the flags in the netfs_inode struct for those filesystems. This reenables the use of PG_fscache in that inode. 9p and afs use the netfslib write helpers so get switched over; cifs, for the moment, does page-by-page manual access to the cache, so doesn't use PG_fscache and is unaffected. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Matthew Wilcox (Oracle) cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: Marc Dionne cc: Ilya Dryomov cc: Xiubo Li cc: Steve French cc: Paulo Alcantara cc: Ronnie Sahlberg cc: Shyam Prasad N cc: Tom Talpey cc: Bharath SM cc: Trond Myklebust cc: Anna Schumaker cc: netfs@lists.linux.dev cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org cc: ceph-devel@vger.kernel.org cc: linux-cifs@vger.kernel.org cc: linux-nfs@vger.kernel.org cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/ceph/addr.c | 2 +- fs/ceph/inode.c | 2 + fs/netfs/buffered_read.c | 36 ++++++++++++----- fs/netfs/buffered_write.c | 93 +++++++++++++++++++++----------------------- fs/netfs/fscache_io.c | 12 ++++-- fs/netfs/internal.h | 10 +++-- fs/netfs/io.c | 18 +++++---- fs/netfs/main.c | 1 + fs/netfs/misc.c | 10 +---- fs/netfs/objects.c | 6 ++- fs/nfs/fscache.h | 2 + include/linux/fscache.h | 22 +++++++---- include/linux/netfs.h | 20 ++++++++-- include/trace/events/netfs.h | 6 ++- 14 files changed, 143 insertions(+), 97 deletions(-) (limited to 'fs') diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index ee9caf7916fb..28ae4976a4f9 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -517,7 +517,7 @@ static void ceph_fscache_write_to_cache(struct inode *inode, u64 off, u64 len, b struct fscache_cookie *cookie = ceph_fscache_cookie(ci); fscache_write_to_cache(cookie, inode->i_mapping, off, len, i_size_read(inode), - ceph_fscache_write_terminated, inode, caching); + ceph_fscache_write_terminated, inode, true, caching); } #else static inline void ceph_set_page_fscache(struct page *page) diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 7b2e77517f23..99561fddcb38 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -577,6 +577,8 @@ struct inode *ceph_alloc_inode(struct super_block *sb) /* Set parameters for the netfs library */ netfs_inode_init(&ci->netfs, &ceph_netfs_ops, false); + /* [DEPRECATED] Use PG_private_2 to mark folio being written to the cache. */ + __set_bit(NETFS_ICTX_USE_PGPRIV2, &ci->netfs.flags); spin_lock_init(&ci->i_ceph_lock); diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index 3298c29b5548..6d49319c82c6 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -10,8 +10,11 @@ #include "internal.h" /* - * Unlock the folios in a read operation. We need to set PG_fscache on any + * Unlock the folios in a read operation. We need to set PG_writeback on any * folios we're going to write back before we unlock them. + * + * Note that if the deprecated NETFS_RREQ_USE_PGPRIV2 is set then we use + * PG_private_2 and do a direct write to the cache from here instead. */ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq) { @@ -48,14 +51,14 @@ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq) xas_for_each(&xas, folio, last_page) { loff_t pg_end; bool pg_failed = false; - bool folio_started; + bool wback_to_cache = false; + bool folio_started = false; if (xas_retry(&xas, folio)) continue; pg_end = folio_pos(folio) + folio_size(folio) - 1; - folio_started = false; for (;;) { loff_t sreq_end; @@ -63,10 +66,16 @@ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq) pg_failed = true; break; } - if (!folio_started && test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) { - trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); - folio_start_fscache(folio); - folio_started = true; + if (test_bit(NETFS_RREQ_USE_PGPRIV2, &rreq->flags)) { + if (!folio_started && test_bit(NETFS_SREQ_COPY_TO_CACHE, + &subreq->flags)) { + trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); + folio_start_fscache(folio); + folio_started = true; + } + } else { + wback_to_cache |= + test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags); } pg_failed |= subreq_failed; sreq_end = subreq->start + subreq->len - 1; @@ -98,6 +107,11 @@ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq) kfree(finfo); } folio_mark_uptodate(folio); + if (wback_to_cache && !WARN_ON_ONCE(folio_get_private(folio) != NULL)) { + trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); + folio_attach_private(folio, NETFS_FOLIO_COPY_TO_CACHE); + filemap_dirty_folio(folio->mapping, folio); + } } if (!test_bit(NETFS_RREQ_DONT_UNLOCK_FOLIOS, &rreq->flags)) { @@ -491,9 +505,11 @@ retry: netfs_put_request(rreq, false, netfs_rreq_trace_put_return); have_folio: - ret = folio_wait_fscache_killable(folio); - if (ret < 0) - goto error; + if (test_bit(NETFS_ICTX_USE_PGPRIV2, &ctx->flags)) { + ret = folio_wait_fscache_killable(folio); + if (ret < 0) + goto error; + } have_folio_no_wait: *_folio = folio; _leave(" = 0"); diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index f7455a579f21..57c6eab01261 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -30,21 +30,13 @@ static void netfs_cleanup_buffered_write(struct netfs_io_request *wreq); static void netfs_set_group(struct folio *folio, struct netfs_group *netfs_group) { - if (netfs_group && !folio_get_private(folio)) - folio_attach_private(folio, netfs_get_group(netfs_group)); -} + void *priv = folio_get_private(folio); -#if IS_ENABLED(CONFIG_FSCACHE) -static void netfs_folio_start_fscache(bool caching, struct folio *folio) -{ - if (caching) - folio_start_fscache(folio); -} -#else -static void netfs_folio_start_fscache(bool caching, struct folio *folio) -{ + if (netfs_group && (!priv || priv == NETFS_FOLIO_COPY_TO_CACHE)) + folio_attach_private(folio, netfs_get_group(netfs_group)); + else if (!netfs_group && priv == NETFS_FOLIO_COPY_TO_CACHE) + folio_detach_private(folio); } -#endif /* * Decide how we should modify a folio. We might be attempting to do @@ -63,11 +55,12 @@ static enum netfs_how_to_modify netfs_how_to_modify(struct netfs_inode *ctx, bool maybe_trouble) { struct netfs_folio *finfo = netfs_folio_info(folio); + struct netfs_group *group = netfs_folio_group(folio); loff_t pos = folio_file_pos(folio); _enter(""); - if (netfs_folio_group(folio) != netfs_group) + if (group != netfs_group && group != NETFS_FOLIO_COPY_TO_CACHE) return NETFS_FLUSH_CONTENT; if (folio_test_uptodate(folio)) @@ -396,9 +389,7 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, folio_clear_dirty_for_io(folio); /* We make multiple writes to the folio... */ if (!folio_test_writeback(folio)) { - folio_wait_fscache(folio); folio_start_writeback(folio); - folio_start_fscache(folio); if (wreq->iter.count == 0) trace_netfs_folio(folio, netfs_folio_trace_wthru); else @@ -528,6 +519,7 @@ EXPORT_SYMBOL(netfs_file_write_iter); */ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_group) { + struct netfs_group *group; struct folio *folio = page_folio(vmf->page); struct file *file = vmf->vma->vm_file; struct inode *inode = file_inode(file); @@ -550,7 +542,8 @@ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_gr goto out; } - if (netfs_folio_group(folio) != netfs_group) { + group = netfs_folio_group(folio); + if (group != netfs_group && group != NETFS_FOLIO_COPY_TO_CACHE) { folio_unlock(folio); err = filemap_fdatawait_range(inode->i_mapping, folio_pos(folio), @@ -606,8 +599,6 @@ static void netfs_kill_pages(struct address_space *mapping, trace_netfs_folio(folio, netfs_folio_trace_kill); folio_clear_uptodate(folio); - if (folio_test_fscache(folio)) - folio_end_fscache(folio); folio_end_writeback(folio); folio_lock(folio); generic_error_remove_folio(mapping, folio); @@ -643,8 +634,6 @@ static void netfs_redirty_pages(struct address_space *mapping, next = folio_next_index(folio); trace_netfs_folio(folio, netfs_folio_trace_redirty); filemap_dirty_folio(mapping, folio); - if (folio_test_fscache(folio)) - folio_end_fscache(folio); folio_end_writeback(folio); folio_put(folio); } while (index = next, index <= last); @@ -700,7 +689,11 @@ static void netfs_pages_written_back(struct netfs_io_request *wreq) if (!folio_test_dirty(folio)) { folio_detach_private(folio); gcount++; - trace_netfs_folio(folio, netfs_folio_trace_clear_g); + if (group == NETFS_FOLIO_COPY_TO_CACHE) + trace_netfs_folio(folio, + netfs_folio_trace_end_copy); + else + trace_netfs_folio(folio, netfs_folio_trace_clear_g); } else { trace_netfs_folio(folio, netfs_folio_trace_redirtied); } @@ -724,8 +717,6 @@ static void netfs_pages_written_back(struct netfs_io_request *wreq) trace_netfs_folio(folio, netfs_folio_trace_clear); } end_wb: - if (folio_test_fscache(folio)) - folio_end_fscache(folio); xas_advance(&xas, folio_next_index(folio) - 1); folio_end_writeback(folio); } @@ -795,7 +786,6 @@ static void netfs_extend_writeback(struct address_space *mapping, long *_count, loff_t start, loff_t max_len, - bool caching, size_t *_len, size_t *_top) { @@ -846,8 +836,7 @@ static void netfs_extend_writeback(struct address_space *mapping, break; } if (!folio_test_dirty(folio) || - folio_test_writeback(folio) || - folio_test_fscache(folio)) { + folio_test_writeback(folio)) { folio_unlock(folio); folio_put(folio); xas_reset(xas); @@ -860,7 +849,8 @@ static void netfs_extend_writeback(struct address_space *mapping, if ((const struct netfs_group *)priv != group) { stop = true; finfo = netfs_folio_info(folio); - if (finfo->netfs_group != group || + if (!finfo || + finfo->netfs_group != group || finfo->dirty_offset > 0) { folio_unlock(folio); folio_put(folio); @@ -894,12 +884,14 @@ static void netfs_extend_writeback(struct address_space *mapping, for (i = 0; i < folio_batch_count(&fbatch); i++) { folio = fbatch.folios[i]; - trace_netfs_folio(folio, netfs_folio_trace_store_plus); + if (group == NETFS_FOLIO_COPY_TO_CACHE) + trace_netfs_folio(folio, netfs_folio_trace_copy_plus); + else + trace_netfs_folio(folio, netfs_folio_trace_store_plus); if (!folio_clear_dirty_for_io(folio)) BUG(); folio_start_writeback(folio); - netfs_folio_start_fscache(caching, folio); folio_unlock(folio); } @@ -925,14 +917,14 @@ static ssize_t netfs_write_back_from_locked_folio(struct address_space *mapping, struct netfs_inode *ctx = netfs_inode(mapping->host); unsigned long long i_size = i_size_read(&ctx->inode); size_t len, max_len; - bool caching = netfs_is_cache_enabled(ctx); long count = wbc->nr_to_write; int ret; - _enter(",%lx,%llx-%llx,%u", folio->index, start, end, caching); + _enter(",%lx,%llx-%llx", folio->index, start, end); wreq = netfs_alloc_request(mapping, NULL, start, folio_size(folio), - NETFS_WRITEBACK); + group == NETFS_FOLIO_COPY_TO_CACHE ? + NETFS_COPY_TO_CACHE : NETFS_WRITEBACK); if (IS_ERR(wreq)) { folio_unlock(folio); return PTR_ERR(wreq); @@ -941,7 +933,6 @@ static ssize_t netfs_write_back_from_locked_folio(struct address_space *mapping, if (!folio_clear_dirty_for_io(folio)) BUG(); folio_start_writeback(folio); - netfs_folio_start_fscache(caching, folio); count -= folio_nr_pages(folio); @@ -950,7 +941,10 @@ static ssize_t netfs_write_back_from_locked_folio(struct address_space *mapping, * immediately lockable, is not dirty or is missing, or we reach the * end of the range. */ - trace_netfs_folio(folio, netfs_folio_trace_store); + if (group == NETFS_FOLIO_COPY_TO_CACHE) + trace_netfs_folio(folio, netfs_folio_trace_copy); + else + trace_netfs_folio(folio, netfs_folio_trace_store); len = wreq->len; finfo = netfs_folio_info(folio); @@ -973,7 +967,7 @@ static ssize_t netfs_write_back_from_locked_folio(struct address_space *mapping, if (len < max_len) netfs_extend_writeback(mapping, group, xas, &count, start, - max_len, caching, &len, &wreq->upper_len); + max_len, &len, &wreq->upper_len); } cant_expand: @@ -997,15 +991,18 @@ cant_expand: iov_iter_xarray(&wreq->iter, ITER_SOURCE, &mapping->i_pages, start, wreq->upper_len); - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - ret = netfs_begin_write(wreq, true, netfs_write_trace_writeback); + if (group != NETFS_FOLIO_COPY_TO_CACHE) { + __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); + ret = netfs_begin_write(wreq, true, netfs_write_trace_writeback); + } else { + ret = netfs_begin_write(wreq, true, netfs_write_trace_copy_to_cache); + } if (ret == 0 || ret == -EIOCBQUEUED) wbc->nr_to_write -= len / PAGE_SIZE; } else { _debug("write discard %zx @%llx [%llx]", len, start, i_size); /* The dirty region was entirely beyond the EOF. */ - fscache_clear_page_bits(mapping, start, len, caching); netfs_pages_written_back(wreq); ret = 0; } @@ -1058,9 +1055,11 @@ search_again: /* Skip any dirty folio that's not in the group of interest. */ priv = folio_get_private(folio); - if ((const struct netfs_group *)priv != group) { - finfo = netfs_folio_info(folio); - if (finfo->netfs_group != group) { + if ((const struct netfs_group *)priv == NETFS_FOLIO_COPY_TO_CACHE) { + group = NETFS_FOLIO_COPY_TO_CACHE; + } else if ((const struct netfs_group *)priv != group) { + finfo = __netfs_folio_info(priv); + if (!finfo || finfo->netfs_group != group) { folio_put(folio); continue; } @@ -1099,14 +1098,10 @@ lock_again: goto search_again; } - if (folio_test_writeback(folio) || - folio_test_fscache(folio)) { + if (folio_test_writeback(folio)) { folio_unlock(folio); if (wbc->sync_mode != WB_SYNC_NONE) { folio_wait_writeback(folio); -#ifdef CONFIG_FSCACHE - folio_wait_fscache(folio); -#endif goto lock_again; } @@ -1265,7 +1260,8 @@ int netfs_launder_folio(struct folio *folio) bvec_set_folio(&bvec, folio, len, offset); iov_iter_bvec(&wreq->iter, ITER_SOURCE, &bvec, 1, len); - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); + if (group != NETFS_FOLIO_COPY_TO_CACHE) + __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); ret = netfs_begin_write(wreq, true, netfs_write_trace_launder); out_put: @@ -1274,7 +1270,6 @@ out_put: kfree(finfo); netfs_put_request(wreq, false, netfs_rreq_trace_put_return); out: - folio_wait_fscache(folio); _leave(" = %d", ret); return ret; } diff --git a/fs/netfs/fscache_io.c b/fs/netfs/fscache_io.c index 43a651ed8264..5028f2ae30da 100644 --- a/fs/netfs/fscache_io.c +++ b/fs/netfs/fscache_io.c @@ -166,6 +166,7 @@ struct fscache_write_request { loff_t start; size_t len; bool set_bits; + bool using_pgpriv2; netfs_io_terminated_t term_func; void *term_func_priv; }; @@ -197,8 +198,9 @@ static void fscache_wreq_done(void *priv, ssize_t transferred_or_error, { struct fscache_write_request *wreq = priv; - fscache_clear_page_bits(wreq->mapping, wreq->start, wreq->len, - wreq->set_bits); + if (wreq->using_pgpriv2) + fscache_clear_page_bits(wreq->mapping, wreq->start, wreq->len, + wreq->set_bits); if (wreq->term_func) wreq->term_func(wreq->term_func_priv, transferred_or_error, @@ -212,7 +214,7 @@ void __fscache_write_to_cache(struct fscache_cookie *cookie, loff_t start, size_t len, loff_t i_size, netfs_io_terminated_t term_func, void *term_func_priv, - bool cond) + bool using_pgpriv2, bool cond) { struct fscache_write_request *wreq; struct netfs_cache_resources *cres; @@ -230,6 +232,7 @@ void __fscache_write_to_cache(struct fscache_cookie *cookie, wreq->mapping = mapping; wreq->start = start; wreq->len = len; + wreq->using_pgpriv2 = using_pgpriv2; wreq->set_bits = cond; wreq->term_func = term_func; wreq->term_func_priv = term_func_priv; @@ -257,7 +260,8 @@ abandon_end: abandon_free: kfree(wreq); abandon: - fscache_clear_page_bits(mapping, start, len, cond); + if (using_pgpriv2) + fscache_clear_page_bits(mapping, start, len, cond); if (term_func) term_func(term_func_priv, ret, false); } diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index ec7045d24400..156ab138e224 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -168,7 +168,7 @@ static inline bool netfs_is_cache_enabled(struct netfs_inode *ctx) */ static inline struct netfs_group *netfs_get_group(struct netfs_group *netfs_group) { - if (netfs_group) + if (netfs_group && netfs_group != NETFS_FOLIO_COPY_TO_CACHE) refcount_inc(&netfs_group->ref); return netfs_group; } @@ -178,7 +178,9 @@ static inline struct netfs_group *netfs_get_group(struct netfs_group *netfs_grou */ static inline void netfs_put_group(struct netfs_group *netfs_group) { - if (netfs_group && refcount_dec_and_test(&netfs_group->ref)) + if (netfs_group && + netfs_group != NETFS_FOLIO_COPY_TO_CACHE && + refcount_dec_and_test(&netfs_group->ref)) netfs_group->free(netfs_group); } @@ -187,7 +189,9 @@ static inline void netfs_put_group(struct netfs_group *netfs_group) */ static inline void netfs_put_group_many(struct netfs_group *netfs_group, int nr) { - if (netfs_group && refcount_sub_and_test(nr, &netfs_group->ref)) + if (netfs_group && + netfs_group != NETFS_FOLIO_COPY_TO_CACHE && + refcount_sub_and_test(nr, &netfs_group->ref)) netfs_group->free(netfs_group); } diff --git a/fs/netfs/io.c b/fs/netfs/io.c index 4261ad6c55b6..b3b9827a9709 100644 --- a/fs/netfs/io.c +++ b/fs/netfs/io.c @@ -99,8 +99,9 @@ static void netfs_rreq_completed(struct netfs_io_request *rreq, bool was_async) } /* - * Deal with the completion of writing the data to the cache. We have to clear - * the PG_fscache bits on the folios involved and release the caller's ref. + * [DEPRECATED] Deal with the completion of writing the data to the cache. We + * have to clear the PG_fscache bits on the folios involved and release the + * caller's ref. * * May be called in softirq mode and we inherit a ref from the caller. */ @@ -138,7 +139,7 @@ static void netfs_rreq_unmark_after_write(struct netfs_io_request *rreq, } static void netfs_rreq_copy_terminated(void *priv, ssize_t transferred_or_error, - bool was_async) + bool was_async) /* [DEPRECATED] */ { struct netfs_io_subrequest *subreq = priv; struct netfs_io_request *rreq = subreq->rreq; @@ -161,8 +162,8 @@ static void netfs_rreq_copy_terminated(void *priv, ssize_t transferred_or_error, } /* - * Perform any outstanding writes to the cache. We inherit a ref from the - * caller. + * [DEPRECATED] Perform any outstanding writes to the cache. We inherit a ref + * from the caller. */ static void netfs_rreq_do_write_to_cache(struct netfs_io_request *rreq) { @@ -222,7 +223,7 @@ static void netfs_rreq_do_write_to_cache(struct netfs_io_request *rreq) netfs_rreq_unmark_after_write(rreq, false); } -static void netfs_rreq_write_to_cache_work(struct work_struct *work) +static void netfs_rreq_write_to_cache_work(struct work_struct *work) /* [DEPRECATED] */ { struct netfs_io_request *rreq = container_of(work, struct netfs_io_request, work); @@ -230,7 +231,7 @@ static void netfs_rreq_write_to_cache_work(struct work_struct *work) netfs_rreq_do_write_to_cache(rreq); } -static void netfs_rreq_write_to_cache(struct netfs_io_request *rreq) +static void netfs_rreq_write_to_cache(struct netfs_io_request *rreq) /* [DEPRECATED] */ { rreq->work.func = netfs_rreq_write_to_cache_work; if (!queue_work(system_unbound_wq, &rreq->work)) @@ -409,7 +410,8 @@ again: clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &rreq->flags); wake_up_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS); - if (test_bit(NETFS_RREQ_COPY_TO_CACHE, &rreq->flags)) + if (test_bit(NETFS_RREQ_COPY_TO_CACHE, &rreq->flags) && + test_bit(NETFS_RREQ_USE_PGPRIV2, &rreq->flags)) return netfs_rreq_write_to_cache(rreq); netfs_rreq_completed(rreq, was_async); diff --git a/fs/netfs/main.c b/fs/netfs/main.c index 5e77618a7940..c5a73c9ed126 100644 --- a/fs/netfs/main.c +++ b/fs/netfs/main.c @@ -31,6 +31,7 @@ static const char *netfs_origins[nr__netfs_io_origin] = { [NETFS_READAHEAD] = "RA", [NETFS_READPAGE] = "RP", [NETFS_READ_FOR_WRITE] = "RW", + [NETFS_COPY_TO_CACHE] = "CC", [NETFS_WRITEBACK] = "WB", [NETFS_WRITETHROUGH] = "WT", [NETFS_LAUNDER_WRITE] = "LW", diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c index 90051ced8e2a..bc1fc54fb724 100644 --- a/fs/netfs/misc.c +++ b/fs/netfs/misc.c @@ -177,13 +177,11 @@ EXPORT_SYMBOL(netfs_clear_inode_writeback); */ void netfs_invalidate_folio(struct folio *folio, size_t offset, size_t length) { - struct netfs_folio *finfo = NULL; + struct netfs_folio *finfo; size_t flen = folio_size(folio); _enter("{%lx},%zx,%zx", folio->index, offset, length); - folio_wait_fscache(folio); - if (!folio_test_private(folio)) return; @@ -248,12 +246,6 @@ bool netfs_release_folio(struct folio *folio, gfp_t gfp) if (folio_test_private(folio)) return false; - if (folio_test_fscache(folio)) { - if (current_is_kswapd() || !(gfp & __GFP_FS)) - return false; - folio_wait_fscache(folio); - } - fscache_note_page_release(netfs_i_cookie(ctx)); return true; } diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c index 610ceb5bd86c..72b52f070270 100644 --- a/fs/netfs/objects.c +++ b/fs/netfs/objects.c @@ -45,8 +45,12 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping, refcount_set(&rreq->ref, 1); __set_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags); - if (cached) + if (cached) { __set_bit(NETFS_RREQ_WRITE_TO_CACHE, &rreq->flags); + if (test_bit(NETFS_ICTX_USE_PGPRIV2, &ctx->flags)) + /* Filesystem uses deprecated PG_private_2 marking. */ + __set_bit(NETFS_RREQ_USE_PGPRIV2, &rreq->flags); + } if (file && file->f_flags & O_NONBLOCK) __set_bit(NETFS_RREQ_NONBLOCK, &rreq->flags); if (rreq->netfs_ops->init_request) { diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h index e3cb4923316b..814363d1d7c7 100644 --- a/fs/nfs/fscache.h +++ b/fs/nfs/fscache.h @@ -81,6 +81,8 @@ static inline void nfs_netfs_put(struct nfs_netfs_io_data *netfs) static inline void nfs_netfs_inode_init(struct nfs_inode *nfsi) { netfs_inode_init(&nfsi->netfs, &nfs_netfs_ops, false); + /* [DEPRECATED] Use PG_private_2 to mark folio being written to the cache. */ + __set_bit(NETFS_ICTX_USE_PGPRIV2, &nfsi->netfs.flags); } extern void nfs_netfs_initiate_read(struct nfs_pgio_header *hdr); extern void nfs_netfs_read_completion(struct nfs_pgio_header *hdr); diff --git a/include/linux/fscache.h b/include/linux/fscache.h index 6e8562cbcc43..9de27643607f 100644 --- a/include/linux/fscache.h +++ b/include/linux/fscache.h @@ -172,9 +172,12 @@ extern void __fscache_invalidate(struct fscache_cookie *, const void *, loff_t, extern int __fscache_begin_read_operation(struct netfs_cache_resources *, struct fscache_cookie *); extern int __fscache_begin_write_operation(struct netfs_cache_resources *, struct fscache_cookie *); -extern void __fscache_write_to_cache(struct fscache_cookie *, struct address_space *, - loff_t, size_t, loff_t, netfs_io_terminated_t, void *, - bool); +void __fscache_write_to_cache(struct fscache_cookie *cookie, + struct address_space *mapping, + loff_t start, size_t len, loff_t i_size, + netfs_io_terminated_t term_func, + void *term_func_priv, + bool using_pgpriv2, bool cond); extern void __fscache_clear_page_bits(struct address_space *, loff_t, size_t); /** @@ -597,7 +600,8 @@ static inline void fscache_clear_page_bits(struct address_space *mapping, * @i_size: The new size of the inode * @term_func: The function to call upon completion * @term_func_priv: The private data for @term_func - * @caching: If PG_fscache has been set + * @using_pgpriv2: If we're using PG_private_2 to mark in-progress write + * @caching: If we actually want to do the caching * * Helper function for a netfs to write dirty data from an inode into the cache * object that's backing it. @@ -608,19 +612,21 @@ static inline void fscache_clear_page_bits(struct address_space *mapping, * marked with PG_fscache. * * If given, @term_func will be called upon completion and supplied with - * @term_func_priv. Note that the PG_fscache flags will have been cleared by - * this point, so the netfs must retain its own pin on the mapping. + * @term_func_priv. Note that if @using_pgpriv2 is set, the PG_private_2 flags + * will have been cleared by this point, so the netfs must retain its own pin + * on the mapping. */ static inline void fscache_write_to_cache(struct fscache_cookie *cookie, struct address_space *mapping, loff_t start, size_t len, loff_t i_size, netfs_io_terminated_t term_func, void *term_func_priv, - bool caching) + bool using_pgpriv2, bool caching) { if (caching) __fscache_write_to_cache(cookie, mapping, start, len, i_size, - term_func, term_func_priv, caching); + term_func, term_func_priv, + using_pgpriv2, caching); else if (term_func) term_func(term_func_priv, -ENOBUFS, false); diff --git a/include/linux/netfs.h b/include/linux/netfs.h index 100cbb261269..f5e9c5f84a0c 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -143,6 +143,8 @@ struct netfs_inode { #define NETFS_ICTX_UNBUFFERED 1 /* I/O should not use the pagecache */ #define NETFS_ICTX_WRITETHROUGH 2 /* Write-through caching */ #define NETFS_ICTX_NO_WRITE_STREAMING 3 /* Don't engage in write-streaming */ +#define NETFS_ICTX_USE_PGPRIV2 31 /* [DEPRECATED] Use PG_private_2 to mark + * write to cache on read */ }; /* @@ -165,16 +167,25 @@ struct netfs_folio { unsigned int dirty_len; /* Write-streaming dirty data length */ }; #define NETFS_FOLIO_INFO 0x1UL /* OR'd with folio->private. */ +#define NETFS_FOLIO_COPY_TO_CACHE ((struct netfs_group *)0x356UL) /* Write to the cache only */ -static inline struct netfs_folio *netfs_folio_info(struct folio *folio) +static inline bool netfs_is_folio_info(const void *priv) { - void *priv = folio_get_private(folio); + return (unsigned long)priv & NETFS_FOLIO_INFO; +} - if ((unsigned long)priv & NETFS_FOLIO_INFO) +static inline struct netfs_folio *__netfs_folio_info(const void *priv) +{ + if (netfs_is_folio_info(priv)) return (struct netfs_folio *)((unsigned long)priv & ~NETFS_FOLIO_INFO); return NULL; } +static inline struct netfs_folio *netfs_folio_info(struct folio *folio) +{ + return __netfs_folio_info(folio_get_private(folio)); +} + static inline struct netfs_group *netfs_folio_group(struct folio *folio) { struct netfs_folio *finfo; @@ -230,6 +241,7 @@ enum netfs_io_origin { NETFS_READAHEAD, /* This read was triggered by readahead */ NETFS_READPAGE, /* This read is a synchronous read */ NETFS_READ_FOR_WRITE, /* This read is to prepare a write */ + NETFS_COPY_TO_CACHE, /* This write is to copy a read to the cache */ NETFS_WRITEBACK, /* This write was triggered by writepages */ NETFS_WRITETHROUGH, /* This write was made by netfs_perform_write() */ NETFS_LAUNDER_WRITE, /* This is triggered by ->launder_folio() */ @@ -287,6 +299,8 @@ struct netfs_io_request { #define NETFS_RREQ_UPLOAD_TO_SERVER 8 /* Need to write to the server */ #define NETFS_RREQ_NONBLOCK 9 /* Don't block if possible (O_NONBLOCK) */ #define NETFS_RREQ_BLOCKED 10 /* We blocked */ +#define NETFS_RREQ_USE_PGPRIV2 31 /* [DEPRECATED] Use PG_private_2 to mark + * write to cache on read */ const struct netfs_request_ops *netfs_ops; void (*cleanup)(struct netfs_io_request *req); }; diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h index 447a8c21cf57..e03fafb0c1e3 100644 --- a/include/trace/events/netfs.h +++ b/include/trace/events/netfs.h @@ -24,6 +24,7 @@ E_(netfs_read_trace_write_begin, "WRITEBEGN") #define netfs_write_traces \ + EM(netfs_write_trace_copy_to_cache, "COPY2CACH") \ EM(netfs_write_trace_dio_write, "DIO-WRITE") \ EM(netfs_write_trace_launder, "LAUNDER ") \ EM(netfs_write_trace_unbuffered_write, "UNB-WRITE") \ @@ -34,6 +35,7 @@ EM(NETFS_READAHEAD, "RA") \ EM(NETFS_READPAGE, "RP") \ EM(NETFS_READ_FOR_WRITE, "RW") \ + EM(NETFS_COPY_TO_CACHE, "CC") \ EM(NETFS_WRITEBACK, "WB") \ EM(NETFS_WRITETHROUGH, "WT") \ EM(NETFS_LAUNDER_WRITE, "LW") \ @@ -127,7 +129,9 @@ EM(netfs_folio_trace_clear, "clear") \ EM(netfs_folio_trace_clear_s, "clear-s") \ EM(netfs_folio_trace_clear_g, "clear-g") \ - EM(netfs_folio_trace_copy_to_cache, "copy") \ + EM(netfs_folio_trace_copy, "copy") \ + EM(netfs_folio_trace_copy_plus, "copy+") \ + EM(netfs_folio_trace_copy_to_cache, "mark-copy") \ EM(netfs_folio_trace_end_copy, "end-copy") \ EM(netfs_folio_trace_filled_gaps, "filled-gaps") \ EM(netfs_folio_trace_kill, "kill") \ -- cgit v1.2.3-59-g8ed1b From 2e9d7e4b984a61823c41ba65e1b58b98ca9912bb Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 19 Mar 2024 11:13:26 +0000 Subject: mm: Remove the PG_fscache alias for PG_private_2 Remove the PG_fscache alias for PG_private_2 and use the latter directly. Use of this flag for marking pages undergoing writing to the cache should be considered deprecated and the folios should be marked dirty instead and the write done in ->writepages(). Note that PG_private_2 itself should be considered deprecated and up for future removal by the MM folks too. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Matthew Wilcox (Oracle) cc: Ilya Dryomov cc: Xiubo Li cc: Steve French cc: Paulo Alcantara cc: Ronnie Sahlberg cc: Shyam Prasad N cc: Tom Talpey cc: Bharath SM cc: Trond Myklebust cc: Anna Schumaker cc: netfs@lists.linux.dev cc: ceph-devel@vger.kernel.org cc: linux-cifs@vger.kernel.org cc: linux-nfs@vger.kernel.org cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/ceph/addr.c | 11 ++++--- fs/netfs/buffered_read.c | 4 +-- fs/netfs/fscache_io.c | 2 +- fs/netfs/io.c | 2 +- fs/nfs/file.c | 8 ++--- fs/nfs/fscache.h | 4 +-- fs/nfs/write.c | 4 +-- fs/smb/client/file.c | 16 +++++----- include/linux/netfs.h | 80 +++--------------------------------------------- mm/filemap.c | 6 ++-- 10 files changed, 33 insertions(+), 104 deletions(-) (limited to 'fs') diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 28ae4976a4f9..18ddacb00511 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -500,7 +500,7 @@ const struct netfs_request_ops ceph_netfs_ops = { #ifdef CONFIG_CEPH_FSCACHE static void ceph_set_page_fscache(struct page *page) { - set_page_fscache(page); + folio_start_private_2(page_folio(page)); /* [DEPRECATED] */ } static void ceph_fscache_write_terminated(void *priv, ssize_t error, bool was_async) @@ -800,7 +800,7 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc) return AOP_WRITEPAGE_ACTIVATE; } - wait_on_page_fscache(page); + folio_wait_private_2(page_folio(page)); /* [DEPRECATED] */ err = writepage_nounlock(page, wbc); if (err == -ERESTARTSYS) { @@ -1075,7 +1075,8 @@ get_more_pages: unlock_page(page); break; } - if (PageWriteback(page) || PageFsCache(page)) { + if (PageWriteback(page) || + PagePrivate2(page) /* [DEPRECATED] */) { if (wbc->sync_mode == WB_SYNC_NONE) { doutc(cl, "%p under writeback\n", page); unlock_page(page); @@ -1083,7 +1084,7 @@ get_more_pages: } doutc(cl, "waiting on writeback %p\n", page); wait_on_page_writeback(page); - wait_on_page_fscache(page); + folio_wait_private_2(page_folio(page)); /* [DEPRECATED] */ } if (!clear_page_dirty_for_io(page)) { @@ -1513,7 +1514,7 @@ static int ceph_write_begin(struct file *file, struct address_space *mapping, if (r < 0) return r; - folio_wait_fscache(folio); + folio_wait_private_2(folio); /* [DEPRECATED] */ WARN_ON_ONCE(!folio_test_locked(folio)); *pagep = &folio->page; return 0; diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index 6d49319c82c6..b3fd6e1fa322 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -70,7 +70,7 @@ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq) if (!folio_started && test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) { trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); - folio_start_fscache(folio); + folio_start_private_2(folio); folio_started = true; } } else { @@ -506,7 +506,7 @@ retry: have_folio: if (test_bit(NETFS_ICTX_USE_PGPRIV2, &ctx->flags)) { - ret = folio_wait_fscache_killable(folio); + ret = folio_wait_private_2_killable(folio); if (ret < 0) goto error; } diff --git a/fs/netfs/fscache_io.c b/fs/netfs/fscache_io.c index 5028f2ae30da..38637e5c9b57 100644 --- a/fs/netfs/fscache_io.c +++ b/fs/netfs/fscache_io.c @@ -183,7 +183,7 @@ void __fscache_clear_page_bits(struct address_space *mapping, rcu_read_lock(); xas_for_each(&xas, page, last) { - end_page_fscache(page); + folio_end_private_2(page_folio(page)); } rcu_read_unlock(); } diff --git a/fs/netfs/io.c b/fs/netfs/io.c index b3b9827a9709..60a19f96e0ce 100644 --- a/fs/netfs/io.c +++ b/fs/netfs/io.c @@ -129,7 +129,7 @@ static void netfs_rreq_unmark_after_write(struct netfs_io_request *rreq, continue; unlocked = folio_next_index(folio) - 1; trace_netfs_folio(folio, netfs_folio_trace_end_copy); - folio_end_fscache(folio); + folio_end_private_2(folio); have_unlocked = true; } } diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 407c6e15afe2..6bd127e6683d 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -433,7 +433,7 @@ static void nfs_invalidate_folio(struct folio *folio, size_t offset, return; /* Cancel any unstarted writes on this page */ nfs_wb_folio_cancel(inode, folio); - folio_wait_fscache(folio); + folio_wait_private_2(folio); /* [DEPRECATED] */ trace_nfs_invalidate_folio(inode, folio); } @@ -500,7 +500,7 @@ static int nfs_launder_folio(struct folio *folio) dfprintk(PAGECACHE, "NFS: launder_folio(%ld, %llu)\n", inode->i_ino, folio_pos(folio)); - folio_wait_fscache(folio); + folio_wait_private_2(folio); /* [DEPRECATED] */ ret = nfs_wb_folio(inode, folio); trace_nfs_launder_folio_done(inode, folio, ret); return ret; @@ -593,8 +593,8 @@ static vm_fault_t nfs_vm_page_mkwrite(struct vm_fault *vmf) sb_start_pagefault(inode->i_sb); /* make sure the cache has finished storing the page */ - if (folio_test_fscache(folio) && - folio_wait_fscache_killable(folio) < 0) { + if (folio_test_private_2(folio) && /* [DEPRECATED] */ + folio_wait_private_2_killable(folio) < 0) { ret = VM_FAULT_RETRY; goto out; } diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h index 814363d1d7c7..fbed0027996f 100644 --- a/fs/nfs/fscache.h +++ b/fs/nfs/fscache.h @@ -103,10 +103,10 @@ extern int nfs_netfs_read_folio(struct file *file, struct folio *folio); static inline bool nfs_fscache_release_folio(struct folio *folio, gfp_t gfp) { - if (folio_test_fscache(folio)) { + if (folio_test_private_2(folio)) { /* [DEPRECATED] */ if (current_is_kswapd() || !(gfp & __GFP_FS)) return false; - folio_wait_fscache(folio); + folio_wait_private_2(folio); } fscache_note_page_release(netfs_i_cookie(netfs_inode(folio->mapping->host))); return true; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 5de85d725fb9..2329cbb0e446 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -2120,10 +2120,10 @@ int nfs_migrate_folio(struct address_space *mapping, struct folio *dst, if (folio_test_private(src)) return -EBUSY; - if (folio_test_fscache(src)) { + if (folio_test_private_2(src)) { /* [DEPRECATED] */ if (mode == MIGRATE_ASYNC) return -EBUSY; - folio_wait_fscache(src); + folio_wait_private_2(src); } return migrate_folio(mapping, dst, src, mode); diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 9be37d0fe724..b39caae652f6 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -3032,12 +3032,12 @@ lock_again: } if (folio_test_writeback(folio) || - folio_test_fscache(folio)) { + folio_test_private_2(folio)) { /* [DEPRECATED] */ folio_unlock(folio); if (wbc->sync_mode != WB_SYNC_NONE) { folio_wait_writeback(folio); #ifdef CONFIG_CIFS_FSCACHE - folio_wait_fscache(folio); + folio_wait_private_2(folio); #endif goto lock_again; } @@ -4510,8 +4510,8 @@ static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf) * be modified. We then assume the entire folio will need writing back. */ #ifdef CONFIG_CIFS_FSCACHE - if (folio_test_fscache(folio) && - folio_wait_fscache_killable(folio) < 0) + if (folio_test_private_2(folio) && /* [DEPRECATED] */ + folio_wait_private_2_killable(folio) < 0) return VM_FAULT_RETRY; #endif @@ -4977,10 +4977,10 @@ static bool cifs_release_folio(struct folio *folio, gfp_t gfp) { if (folio_test_private(folio)) return 0; - if (folio_test_fscache(folio)) { + if (folio_test_private_2(folio)) { /* [DEPRECATED] */ if (current_is_kswapd() || !(gfp & __GFP_FS)) return false; - folio_wait_fscache(folio); + folio_wait_private_2(folio); } fscache_note_page_release(cifs_inode_cookie(folio->mapping->host)); return true; @@ -4989,7 +4989,7 @@ static bool cifs_release_folio(struct folio *folio, gfp_t gfp) static void cifs_invalidate_folio(struct folio *folio, size_t offset, size_t length) { - folio_wait_fscache(folio); + folio_wait_private_2(folio); /* [DEPRECATED] */ } static int cifs_launder_folio(struct folio *folio) @@ -5009,7 +5009,7 @@ static int cifs_launder_folio(struct folio *folio) if (folio_clear_dirty_for_io(folio)) rc = cifs_writepage_locked(&folio->page, &wbc); - folio_wait_fscache(folio); + folio_wait_private_2(folio); /* [DEPRECATED] */ return rc; } diff --git a/include/linux/netfs.h b/include/linux/netfs.h index f5e9c5f84a0c..f36a6d8163d1 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -21,94 +21,22 @@ enum netfs_sreq_ref_trace; -/* - * Overload PG_private_2 to give us PG_fscache - this is used to indicate that - * a page is currently backed by a local disk cache - */ -#define folio_test_fscache(folio) folio_test_private_2(folio) -#define PageFsCache(page) PagePrivate2((page)) -#define SetPageFsCache(page) SetPagePrivate2((page)) -#define ClearPageFsCache(page) ClearPagePrivate2((page)) -#define TestSetPageFsCache(page) TestSetPagePrivate2((page)) -#define TestClearPageFsCache(page) TestClearPagePrivate2((page)) - /** - * folio_start_fscache - Start an fscache write on a folio. + * folio_start_private_2 - Start an fscache write on a folio. [DEPRECATED] * @folio: The folio. * * Call this function before writing a folio to a local cache. Starting a * second write before the first one finishes is not allowed. + * + * Note that this should no longer be used. */ -static inline void folio_start_fscache(struct folio *folio) +static inline void folio_start_private_2(struct folio *folio) { VM_BUG_ON_FOLIO(folio_test_private_2(folio), folio); folio_get(folio); folio_set_private_2(folio); } -/** - * folio_end_fscache - End an fscache write on a folio. - * @folio: The folio. - * - * Call this function after the folio has been written to the local cache. - * This will wake any sleepers waiting on this folio. - */ -static inline void folio_end_fscache(struct folio *folio) -{ - folio_end_private_2(folio); -} - -/** - * folio_wait_fscache - Wait for an fscache write on this folio to end. - * @folio: The folio. - * - * If this folio is currently being written to a local cache, wait for - * the write to finish. Another write may start after this one finishes, - * unless the caller holds the folio lock. - */ -static inline void folio_wait_fscache(struct folio *folio) -{ - folio_wait_private_2(folio); -} - -/** - * folio_wait_fscache_killable - Wait for an fscache write on this folio to end. - * @folio: The folio. - * - * If this folio is currently being written to a local cache, wait - * for the write to finish or for a fatal signal to be received. - * Another write may start after this one finishes, unless the caller - * holds the folio lock. - * - * Return: - * - 0 if successful. - * - -EINTR if a fatal signal was encountered. - */ -static inline int folio_wait_fscache_killable(struct folio *folio) -{ - return folio_wait_private_2_killable(folio); -} - -static inline void set_page_fscache(struct page *page) -{ - folio_start_fscache(page_folio(page)); -} - -static inline void end_page_fscache(struct page *page) -{ - folio_end_private_2(page_folio(page)); -} - -static inline void wait_on_page_fscache(struct page *page) -{ - folio_wait_private_2(page_folio(page)); -} - -static inline int wait_on_page_fscache_killable(struct page *page) -{ - return folio_wait_private_2_killable(page_folio(page)); -} - /* Marks used on xarray-based buffers */ #define NETFS_BUF_PUT_MARK XA_MARK_0 /* - Page needs putting */ #define NETFS_BUF_PAGECACHE_MARK XA_MARK_1 /* - Page needs wb/dirty flag wrangling */ diff --git a/mm/filemap.c b/mm/filemap.c index 30de18c4fd28..9a2e28bf298a 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1540,7 +1540,7 @@ EXPORT_SYMBOL(folio_end_private_2); * folio_wait_private_2 - Wait for PG_private_2 to be cleared on a folio. * @folio: The folio to wait on. * - * Wait for PG_private_2 (aka PG_fscache) to be cleared on a folio. + * Wait for PG_private_2 to be cleared on a folio. */ void folio_wait_private_2(struct folio *folio) { @@ -1553,8 +1553,8 @@ EXPORT_SYMBOL(folio_wait_private_2); * folio_wait_private_2_killable - Wait for PG_private_2 to be cleared on a folio. * @folio: The folio to wait on. * - * Wait for PG_private_2 (aka PG_fscache) to be cleared on a folio or until a - * fatal signal is received by the calling task. + * Wait for PG_private_2 to be cleared on a folio or until a fatal signal is + * received by the calling task. * * Return: * - 0 if successful. -- cgit v1.2.3-59-g8ed1b From ae678317b95e760607c7b20b97c9cd4ca9ed6e1a Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 24 Nov 2023 13:02:55 +0000 Subject: netfs: Remove deprecated use of PG_private_2 as a second writeback flag Remove the deprecated use of PG_private_2 in netfslib. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Matthew Wilcox (Oracle) cc: linux-cachefs@redhat.com cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/ceph/addr.c | 19 +------ fs/netfs/buffered_read.c | 8 +-- fs/netfs/io.c | 144 ----------------------------------------------- 3 files changed, 2 insertions(+), 169 deletions(-) (limited to 'fs') diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 18ddacb00511..74bfd10b1b1a 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -498,11 +498,6 @@ const struct netfs_request_ops ceph_netfs_ops = { }; #ifdef CONFIG_CEPH_FSCACHE -static void ceph_set_page_fscache(struct page *page) -{ - folio_start_private_2(page_folio(page)); /* [DEPRECATED] */ -} - static void ceph_fscache_write_terminated(void *priv, ssize_t error, bool was_async) { struct inode *inode = priv; @@ -520,10 +515,6 @@ static void ceph_fscache_write_to_cache(struct inode *inode, u64 off, u64 len, b ceph_fscache_write_terminated, inode, true, caching); } #else -static inline void ceph_set_page_fscache(struct page *page) -{ -} - static inline void ceph_fscache_write_to_cache(struct inode *inode, u64 off, u64 len, bool caching) { } @@ -715,8 +706,6 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) len = wlen; set_page_writeback(page); - if (caching) - ceph_set_page_fscache(page); ceph_fscache_write_to_cache(inode, page_off, len, caching); if (IS_ENCRYPTED(inode)) { @@ -800,8 +789,6 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc) return AOP_WRITEPAGE_ACTIVATE; } - folio_wait_private_2(page_folio(page)); /* [DEPRECATED] */ - err = writepage_nounlock(page, wbc); if (err == -ERESTARTSYS) { /* direct memory reclaimer was killed by SIGKILL. return 0 @@ -1075,8 +1062,7 @@ get_more_pages: unlock_page(page); break; } - if (PageWriteback(page) || - PagePrivate2(page) /* [DEPRECATED] */) { + if (PageWriteback(page)) { if (wbc->sync_mode == WB_SYNC_NONE) { doutc(cl, "%p under writeback\n", page); unlock_page(page); @@ -1084,7 +1070,6 @@ get_more_pages: } doutc(cl, "waiting on writeback %p\n", page); wait_on_page_writeback(page); - folio_wait_private_2(page_folio(page)); /* [DEPRECATED] */ } if (!clear_page_dirty_for_io(page)) { @@ -1269,8 +1254,6 @@ new_request: } set_page_writeback(page); - if (caching) - ceph_set_page_fscache(page); len += thp_size(page); } ceph_fscache_write_to_cache(inode, offset, len, caching); diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index b3fd6e1fa322..1622cce535a3 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -464,7 +464,7 @@ retry: if (!netfs_is_cache_enabled(ctx) && netfs_skip_folio_read(folio, pos, len, false)) { netfs_stat(&netfs_n_rh_write_zskip); - goto have_folio_no_wait; + goto have_folio; } rreq = netfs_alloc_request(mapping, file, @@ -505,12 +505,6 @@ retry: netfs_put_request(rreq, false, netfs_rreq_trace_put_return); have_folio: - if (test_bit(NETFS_ICTX_USE_PGPRIV2, &ctx->flags)) { - ret = folio_wait_private_2_killable(folio); - if (ret < 0) - goto error; - } -have_folio_no_wait: *_folio = folio; _leave(" = 0"); return 0; diff --git a/fs/netfs/io.c b/fs/netfs/io.c index 60a19f96e0ce..2641238aae82 100644 --- a/fs/netfs/io.c +++ b/fs/netfs/io.c @@ -98,146 +98,6 @@ static void netfs_rreq_completed(struct netfs_io_request *rreq, bool was_async) netfs_put_request(rreq, was_async, netfs_rreq_trace_put_complete); } -/* - * [DEPRECATED] Deal with the completion of writing the data to the cache. We - * have to clear the PG_fscache bits on the folios involved and release the - * caller's ref. - * - * May be called in softirq mode and we inherit a ref from the caller. - */ -static void netfs_rreq_unmark_after_write(struct netfs_io_request *rreq, - bool was_async) -{ - struct netfs_io_subrequest *subreq; - struct folio *folio; - pgoff_t unlocked = 0; - bool have_unlocked = false; - - rcu_read_lock(); - - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - XA_STATE(xas, &rreq->mapping->i_pages, subreq->start / PAGE_SIZE); - - xas_for_each(&xas, folio, (subreq->start + subreq->len - 1) / PAGE_SIZE) { - if (xas_retry(&xas, folio)) - continue; - - /* We might have multiple writes from the same huge - * folio, but we mustn't unlock a folio more than once. - */ - if (have_unlocked && folio->index <= unlocked) - continue; - unlocked = folio_next_index(folio) - 1; - trace_netfs_folio(folio, netfs_folio_trace_end_copy); - folio_end_private_2(folio); - have_unlocked = true; - } - } - - rcu_read_unlock(); - netfs_rreq_completed(rreq, was_async); -} - -static void netfs_rreq_copy_terminated(void *priv, ssize_t transferred_or_error, - bool was_async) /* [DEPRECATED] */ -{ - struct netfs_io_subrequest *subreq = priv; - struct netfs_io_request *rreq = subreq->rreq; - - if (IS_ERR_VALUE(transferred_or_error)) { - netfs_stat(&netfs_n_rh_write_failed); - trace_netfs_failure(rreq, subreq, transferred_or_error, - netfs_fail_copy_to_cache); - } else { - netfs_stat(&netfs_n_rh_write_done); - } - - trace_netfs_sreq(subreq, netfs_sreq_trace_write_term); - - /* If we decrement nr_copy_ops to 0, the ref belongs to us. */ - if (atomic_dec_and_test(&rreq->nr_copy_ops)) - netfs_rreq_unmark_after_write(rreq, was_async); - - netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); -} - -/* - * [DEPRECATED] Perform any outstanding writes to the cache. We inherit a ref - * from the caller. - */ -static void netfs_rreq_do_write_to_cache(struct netfs_io_request *rreq) -{ - struct netfs_cache_resources *cres = &rreq->cache_resources; - struct netfs_io_subrequest *subreq, *next, *p; - struct iov_iter iter; - int ret; - - trace_netfs_rreq(rreq, netfs_rreq_trace_copy); - - /* We don't want terminating writes trying to wake us up whilst we're - * still going through the list. - */ - atomic_inc(&rreq->nr_copy_ops); - - list_for_each_entry_safe(subreq, p, &rreq->subrequests, rreq_link) { - if (!test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) { - list_del_init(&subreq->rreq_link); - netfs_put_subrequest(subreq, false, - netfs_sreq_trace_put_no_copy); - } - } - - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - /* Amalgamate adjacent writes */ - while (!list_is_last(&subreq->rreq_link, &rreq->subrequests)) { - next = list_next_entry(subreq, rreq_link); - if (next->start != subreq->start + subreq->len) - break; - subreq->len += next->len; - list_del_init(&next->rreq_link); - netfs_put_subrequest(next, false, - netfs_sreq_trace_put_merged); - } - - ret = cres->ops->prepare_write(cres, &subreq->start, &subreq->len, - subreq->len, rreq->i_size, true); - if (ret < 0) { - trace_netfs_failure(rreq, subreq, ret, netfs_fail_prepare_write); - trace_netfs_sreq(subreq, netfs_sreq_trace_write_skip); - continue; - } - - iov_iter_xarray(&iter, ITER_SOURCE, &rreq->mapping->i_pages, - subreq->start, subreq->len); - - atomic_inc(&rreq->nr_copy_ops); - netfs_stat(&netfs_n_rh_write); - netfs_get_subrequest(subreq, netfs_sreq_trace_get_copy_to_cache); - trace_netfs_sreq(subreq, netfs_sreq_trace_write); - cres->ops->write(cres, subreq->start, &iter, - netfs_rreq_copy_terminated, subreq); - } - - /* If we decrement nr_copy_ops to 0, the usage ref belongs to us. */ - if (atomic_dec_and_test(&rreq->nr_copy_ops)) - netfs_rreq_unmark_after_write(rreq, false); -} - -static void netfs_rreq_write_to_cache_work(struct work_struct *work) /* [DEPRECATED] */ -{ - struct netfs_io_request *rreq = - container_of(work, struct netfs_io_request, work); - - netfs_rreq_do_write_to_cache(rreq); -} - -static void netfs_rreq_write_to_cache(struct netfs_io_request *rreq) /* [DEPRECATED] */ -{ - rreq->work.func = netfs_rreq_write_to_cache_work; - if (!queue_work(system_unbound_wq, &rreq->work)) - BUG(); -} - /* * Handle a short read. */ @@ -410,10 +270,6 @@ again: clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &rreq->flags); wake_up_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS); - if (test_bit(NETFS_RREQ_COPY_TO_CACHE, &rreq->flags) && - test_bit(NETFS_RREQ_USE_PGPRIV2, &rreq->flags)) - return netfs_rreq_write_to_cache(rreq); - netfs_rreq_completed(rreq, was_async); } -- cgit v1.2.3-59-g8ed1b From 93bf1cc0096fa1e02244078db3334ca7fa1d88c1 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 17:15:57 +0000 Subject: netfs: Make netfs_io_request::subreq_counter an atomic_t Make the netfs_io_request::subreq_counter, used to generate values for netfs_io_subrequest::debug_index, into an atomic_t so that it can be called from the retry thread at the same time as the app thread issuing writes. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/netfs/output.c | 2 +- include/linux/netfs.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/netfs/output.c b/fs/netfs/output.c index 625eb68f3e5a..fbdbb4f78234 100644 --- a/fs/netfs/output.c +++ b/fs/netfs/output.c @@ -37,7 +37,7 @@ struct netfs_io_subrequest *netfs_create_write_request(struct netfs_io_request * subreq->source = dest; subreq->start = start; subreq->len = len; - subreq->debug_index = wreq->subreq_counter++; + subreq->debug_index = atomic_inc_return(&wreq->subreq_counter); switch (subreq->source) { case NETFS_UPLOAD_TO_SERVER: diff --git a/include/linux/netfs.h b/include/linux/netfs.h index f36a6d8163d1..ddafc6ebff42 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -202,7 +202,7 @@ struct netfs_io_request { unsigned int debug_id; unsigned int rsize; /* Maximum read size (0 for none) */ unsigned int wsize; /* Maximum write size (0 for none) */ - unsigned int subreq_counter; /* Next subreq->debug_index */ + atomic_t subreq_counter; /* Next subreq->debug_index */ atomic_t nr_outstanding; /* Number of ops in progress */ atomic_t nr_copy_ops; /* Number of copy-to-cache ops in progress */ size_t submitted; /* Amount submitted for I/O so far */ -- cgit v1.2.3-59-g8ed1b From 120b878158cb2e98b167fef038f81c05efe2fd14 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 21 Mar 2024 10:16:55 +0000 Subject: netfs: Use subreq_counter to allocate subreq debug_index values Use the subreq_counter in netfs_io_request to allocate subrequest debug_index values in read ops as well as write ops. Signed-off-by: David Howells cc: Jeff Layton cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/netfs/io.c | 7 ++----- fs/netfs/objects.c | 1 + fs/netfs/output.c | 1 - 3 files changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/netfs/io.c b/fs/netfs/io.c index 2641238aae82..8de581ac0cfb 100644 --- a/fs/netfs/io.c +++ b/fs/netfs/io.c @@ -501,8 +501,7 @@ out: * Slice off a piece of a read request and submit an I/O request for it. */ static bool netfs_rreq_submit_slice(struct netfs_io_request *rreq, - struct iov_iter *io_iter, - unsigned int *_debug_index) + struct iov_iter *io_iter) { struct netfs_io_subrequest *subreq; enum netfs_io_source source; @@ -511,7 +510,6 @@ static bool netfs_rreq_submit_slice(struct netfs_io_request *rreq, if (!subreq) return false; - subreq->debug_index = (*_debug_index)++; subreq->start = rreq->start + rreq->submitted; subreq->len = io_iter->count; @@ -565,7 +563,6 @@ subreq_failed: int netfs_begin_read(struct netfs_io_request *rreq, bool sync) { struct iov_iter io_iter; - unsigned int debug_index = 0; int ret; _enter("R=%x %llx-%llx", @@ -596,7 +593,7 @@ int netfs_begin_read(struct netfs_io_request *rreq, bool sync) if (rreq->origin == NETFS_DIO_READ && rreq->start + rreq->submitted >= rreq->i_size) break; - if (!netfs_rreq_submit_slice(rreq, &io_iter, &debug_index)) + if (!netfs_rreq_submit_slice(rreq, &io_iter)) break; if (test_bit(NETFS_RREQ_BLOCKED, &rreq->flags) && test_bit(NETFS_RREQ_NONBLOCK, &rreq->flags)) diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c index 72b52f070270..8acc03a64059 100644 --- a/fs/netfs/objects.c +++ b/fs/netfs/objects.c @@ -152,6 +152,7 @@ struct netfs_io_subrequest *netfs_alloc_subrequest(struct netfs_io_request *rreq INIT_LIST_HEAD(&subreq->rreq_link); refcount_set(&subreq->ref, 2); subreq->rreq = rreq; + subreq->debug_index = atomic_inc_return(&rreq->subreq_counter); netfs_get_request(rreq, netfs_rreq_trace_get_subreq); netfs_stat(&netfs_n_rh_sreq); } diff --git a/fs/netfs/output.c b/fs/netfs/output.c index fbdbb4f78234..e586396d6b72 100644 --- a/fs/netfs/output.c +++ b/fs/netfs/output.c @@ -37,7 +37,6 @@ struct netfs_io_subrequest *netfs_create_write_request(struct netfs_io_request * subreq->source = dest; subreq->start = start; subreq->len = len; - subreq->debug_index = atomic_inc_return(&wreq->subreq_counter); switch (subreq->source) { case NETFS_UPLOAD_TO_SERVER: -- cgit v1.2.3-59-g8ed1b From ecd69be71aad3b9299b37b3d8cc3f0fb6016286f Mon Sep 17 00:00:00 2001 From: Yifan Zhao Date: Thu, 25 Apr 2024 22:55:28 +0800 Subject: f2fs: remove redundant parameter in is_next_segment_free() is_next_segment_free() takes a redundant `type` parameter. Remove it. Signed-off-by: Yifan Zhao Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 5d6dae4861f3..4396df2bceab 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2645,7 +2645,7 @@ static void write_current_sum_page(struct f2fs_sb_info *sbi, } static int is_next_segment_free(struct f2fs_sb_info *sbi, - struct curseg_info *curseg, int type) + struct curseg_info *curseg) { unsigned int segno = curseg->segno + 1; struct free_segmap_info *free_i = FREE_I(sbi); @@ -3073,8 +3073,7 @@ static bool need_new_seg(struct f2fs_sb_info *sbi, int type) if (!is_set_ckpt_flags(sbi, CP_CRC_RECOVERY_FLAG) && curseg->seg_type == CURSEG_WARM_NODE) return true; - if (curseg->alloc_type == LFS && - is_next_segment_free(sbi, curseg, type) && + if (curseg->alloc_type == LFS && is_next_segment_free(sbi, curseg) && likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED))) return true; if (!f2fs_need_SSR(sbi) || !get_ssr_segment(sbi, type, SSR, 0)) -- cgit v1.2.3-59-g8ed1b From a320b2f08b3b26fff0a71777f018553fc1e6873e Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 26 Apr 2024 17:33:48 +0800 Subject: f2fs: fix to avoid allocating WARM_DATA segment for direct IO If active_log is not 6, we never use WARM_DATA segment, let's avoid allocating WARM_DATA segment for direct IO. Signed-off-by: Yunlei He Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 3 ++- fs/f2fs/f2fs.h | 2 +- fs/f2fs/file.c | 5 +++-- fs/f2fs/segment.c | 11 +++++++++-- 4 files changed, 15 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index bee1e45f76b8..0c516c653f05 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -4179,7 +4179,8 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, map.m_lblk = bytes_to_blks(inode, offset); map.m_len = bytes_to_blks(inode, offset + length - 1) - map.m_lblk + 1; map.m_next_pgofs = &next_pgofs; - map.m_seg_type = f2fs_rw_hint_to_seg_type(inode->i_write_hint); + map.m_seg_type = f2fs_rw_hint_to_seg_type(F2FS_I_SB(inode), + inode->i_write_hint); if (flags & IOMAP_WRITE) map.m_may_create = true; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index a0ae99bcca39..3a95c8ccb9e0 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -3745,7 +3745,7 @@ int f2fs_build_segment_manager(struct f2fs_sb_info *sbi); void f2fs_destroy_segment_manager(struct f2fs_sb_info *sbi); int __init f2fs_create_segment_manager_caches(void); void f2fs_destroy_segment_manager_caches(void); -int f2fs_rw_hint_to_seg_type(enum rw_hint hint); +int f2fs_rw_hint_to_seg_type(struct f2fs_sb_info *sbi, enum rw_hint hint); enum rw_hint f2fs_io_type_to_rw_hint(struct f2fs_sb_info *sbi, enum page_type type, enum temp_type temp); unsigned int f2fs_usable_segs_in_sec(struct f2fs_sb_info *sbi, diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index d382f8bc2fbe..f1a6b4da5746 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -4637,7 +4637,8 @@ static int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *iter, map.m_may_create = true; if (dio) { - map.m_seg_type = f2fs_rw_hint_to_seg_type(inode->i_write_hint); + map.m_seg_type = f2fs_rw_hint_to_seg_type(sbi, + inode->i_write_hint); flag = F2FS_GET_BLOCK_PRE_DIO; } else { map.m_seg_type = NO_CHECK_TYPE; @@ -4690,7 +4691,7 @@ static void f2fs_dio_write_submit_io(const struct iomap_iter *iter, { struct inode *inode = iter->inode; struct f2fs_sb_info *sbi = F2FS_I_SB(inode); - int seg_type = f2fs_rw_hint_to_seg_type(inode->i_write_hint); + int seg_type = f2fs_rw_hint_to_seg_type(sbi, inode->i_write_hint); enum temp_type temp = f2fs_get_segment_temp(seg_type); bio->bi_write_hint = f2fs_io_type_to_rw_hint(sbi, DATA, temp); diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 4396df2bceab..8ddd4f1c665a 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3351,8 +3351,14 @@ out: return err; } -int f2fs_rw_hint_to_seg_type(enum rw_hint hint) +int f2fs_rw_hint_to_seg_type(struct f2fs_sb_info *sbi, enum rw_hint hint) { + if (F2FS_OPTION(sbi).active_logs == 2) + return CURSEG_HOT_DATA; + else if (F2FS_OPTION(sbi).active_logs == 4) + return CURSEG_COLD_DATA; + + /* active_log == 6 */ switch (hint) { case WRITE_LIFE_SHORT: return CURSEG_HOT_DATA; @@ -3492,7 +3498,8 @@ static int __get_segment_type_6(struct f2fs_io_info *fio) is_inode_flag_set(inode, FI_HOT_DATA) || f2fs_is_cow_file(inode)) return CURSEG_HOT_DATA; - return f2fs_rw_hint_to_seg_type(inode->i_write_hint); + return f2fs_rw_hint_to_seg_type(F2FS_I_SB(inode), + inode->i_write_hint); } else { if (IS_DNODE(fio->page)) return is_cold_node(fio->page) ? CURSEG_WARM_NODE : -- cgit v1.2.3-59-g8ed1b From 20faaf30e55522bba2b56d9c46689233205d7717 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Thu, 25 Apr 2024 16:58:38 +0800 Subject: f2fs: fix to do sanity check on i_xattr_nid in sanity_check_inode() syzbot reports a kernel bug as below: F2FS-fs (loop0): Mounted with checkpoint version = 48b305e4 ================================================================== BUG: KASAN: slab-out-of-bounds in f2fs_test_bit fs/f2fs/f2fs.h:2933 [inline] BUG: KASAN: slab-out-of-bounds in current_nat_addr fs/f2fs/node.h:213 [inline] BUG: KASAN: slab-out-of-bounds in f2fs_get_node_info+0xece/0x1200 fs/f2fs/node.c:600 Read of size 1 at addr ffff88807a58c76c by task syz-executor280/5076 CPU: 1 PID: 5076 Comm: syz-executor280 Not tainted 6.9.0-rc5-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114 print_address_description mm/kasan/report.c:377 [inline] print_report+0x169/0x550 mm/kasan/report.c:488 kasan_report+0x143/0x180 mm/kasan/report.c:601 f2fs_test_bit fs/f2fs/f2fs.h:2933 [inline] current_nat_addr fs/f2fs/node.h:213 [inline] f2fs_get_node_info+0xece/0x1200 fs/f2fs/node.c:600 f2fs_xattr_fiemap fs/f2fs/data.c:1848 [inline] f2fs_fiemap+0x55d/0x1ee0 fs/f2fs/data.c:1925 ioctl_fiemap fs/ioctl.c:220 [inline] do_vfs_ioctl+0x1c07/0x2e50 fs/ioctl.c:838 __do_sys_ioctl fs/ioctl.c:902 [inline] __se_sys_ioctl+0x81/0x170 fs/ioctl.c:890 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f The root cause is we missed to do sanity check on i_xattr_nid during f2fs_iget(), so that in fiemap() path, current_nat_addr() will access nat_bitmap w/ offset from invalid i_xattr_nid, result in triggering kasan bug report, fix it. Reported-and-tested-by: syzbot+3694e283cf5c40df6d14@syzkaller.appspotmail.com Closes: https://lore.kernel.org/linux-f2fs-devel/00000000000094036c0616e72a1d@google.com Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/inode.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'fs') diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index d7a5a88a1a5e..7968b14d49f4 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -362,6 +362,12 @@ static bool sanity_check_inode(struct inode *inode, struct page *node_page) return false; } + if (fi->i_xattr_nid && f2fs_check_nid_range(sbi, fi->i_xattr_nid)) { + f2fs_warn(sbi, "%s: inode (ino=%lx) has corrupted i_xattr_nid: %u, run fsck to fix.", + __func__, inode->i_ino, fi->i_xattr_nid); + return false; + } + return true; } -- cgit v1.2.3-59-g8ed1b From 48d180e2bf5a527ba6513a8abddf8b3beb6b7674 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Sun, 28 Apr 2024 09:12:36 +0800 Subject: f2fs: zone: fix to don't trigger OPU on pinfile for direct IO Otherwise, it breaks pinfile's sematics. Cc: Daeho Jeong Signed-off-by: Chao Yu Reviewed-by: Daeho Jeong Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 0c516c653f05..a990236f137e 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1595,8 +1595,9 @@ next_block: } /* use out-place-update for direct IO under LFS mode */ - if (map->m_may_create && - (is_hole || (f2fs_lfs_mode(sbi) && flag == F2FS_GET_BLOCK_DIO))) { + if (map->m_may_create && (is_hole || + (flag == F2FS_GET_BLOCK_DIO && f2fs_lfs_mode(sbi) && + !f2fs_is_pinned_file(inode)))) { if (unlikely(f2fs_cp_error(sbi))) { err = -EIO; goto sync_out; -- cgit v1.2.3-59-g8ed1b From 6773da870ab89123d1b513da63ed59e32a29cb77 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:21 +0200 Subject: xfs: fix error returns from xfs_bmapi_write MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit xfs_bmapi_write can return 0 without actually returning a mapping in mval in two different cases: 1) when there is absolutely no space available to do an allocation 2) when converting delalloc space, and the allocation is so small that it only covers parts of the delalloc extent before the range requested by the caller Callers at best can handle one of these cases, but in many cases can't cope with either one. Switch xfs_bmapi_write to always return a mapping or return an error code instead. For case 1) above ENOSPC is the obvious choice which is very much what the callers expect anyway. For case 2) there is no really good error code, so pick a funky one from the SysV streams portfolio. This fixes the reproducer here: https://lore.kernel.org/linux-xfs/CAEJPjCvT3Uag-pMTYuigEjWZHn1sGMZ0GCjVVCv29tNHK76Cgg@mail.gmail.com0/ which uses reserved blocks to create file systems that are gravely out of space and thus cause at least xfs_file_alloc_space to hang and trigger the lack of ENOSPC handling in xfs_dquot_disk_alloc. Note that this patch does not actually make any caller but xfs_alloc_file_space deal intelligently with case 2) above. Signed-off-by: Christoph Hellwig Reported-by: 刘通 Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_attr_remote.c | 1 - fs/xfs/libxfs/xfs_bmap.c | 46 +++++++++++++++++++++++++++++++++-------- fs/xfs/libxfs/xfs_da_btree.c | 20 +++++------------- fs/xfs/scrub/quota_repair.c | 6 ------ fs/xfs/scrub/rtbitmap_repair.c | 2 -- fs/xfs/xfs_bmap_util.c | 31 ++++++++++++++------------- fs/xfs/xfs_dquot.c | 1 - fs/xfs/xfs_iomap.c | 8 ------- fs/xfs/xfs_reflink.c | 14 ------------- fs/xfs/xfs_rtalloc.c | 2 -- 10 files changed, 57 insertions(+), 74 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index a8de9dc1e998..beb0efdd8f6b 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -625,7 +625,6 @@ xfs_attr_rmtval_set_blk( if (error) return error; - ASSERT(nmap == 1); ASSERT((map->br_startblock != DELAYSTARTBLOCK) && (map->br_startblock != HOLESTARTBLOCK)); diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 49e2fcad593d..14c9781ec0ce 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4217,8 +4217,10 @@ xfs_bmapi_allocate( } else { error = xfs_bmap_alloc_userdata(bma); } - if (error || bma->blkno == NULLFSBLOCK) + if (error) return error; + if (bma->blkno == NULLFSBLOCK) + return -ENOSPC; if (bma->flags & XFS_BMAPI_ZERO) { error = xfs_zero_extent(bma->ip, bma->blkno, bma->length); @@ -4397,6 +4399,15 @@ xfs_bmapi_finish( * extent state if necessary. Details behaviour is controlled by the flags * parameter. Only allocates blocks from a single allocation group, to avoid * locking problems. + * + * Returns 0 on success and places the extent mappings in mval. nmaps is used + * as an input/output parameter where the caller specifies the maximum number + * of mappings that may be returned and xfs_bmapi_write passes back the number + * of mappings (including existing mappings) it found. + * + * Returns a negative error code on failure, including -ENOSPC when it could not + * allocate any blocks and -ENOSR when it did allocate blocks to convert a + * delalloc range, but those blocks were before the passed in range. */ int xfs_bmapi_write( @@ -4525,10 +4536,16 @@ xfs_bmapi_write( ASSERT(len > 0); ASSERT(bma.length > 0); error = xfs_bmapi_allocate(&bma); - if (error) + if (error) { + /* + * If we already allocated space in a previous + * iteration return what we go so far when + * running out of space. + */ + if (error == -ENOSPC && bma.nallocs) + break; goto error0; - if (bma.blkno == NULLFSBLOCK) - break; + } /* * If this is a CoW allocation, record the data in @@ -4566,7 +4583,6 @@ xfs_bmapi_write( if (!xfs_iext_next_extent(ifp, &bma.icur, &bma.got)) eof = true; } - *nmap = n; error = xfs_bmap_btree_to_extents(tp, ip, bma.cur, &bma.logflags, whichfork); @@ -4577,7 +4593,22 @@ xfs_bmapi_write( ifp->if_nextents > XFS_IFORK_MAXEXT(ip, whichfork)); xfs_bmapi_finish(&bma, whichfork, 0); xfs_bmap_validate_ret(orig_bno, orig_len, orig_flags, orig_mval, - orig_nmap, *nmap); + orig_nmap, n); + + /* + * When converting delayed allocations, xfs_bmapi_allocate ignores + * the passed in bno and always converts from the start of the found + * delalloc extent. + * + * To avoid a successful return with *nmap set to 0, return the magic + * -ENOSR error code for this particular case so that the caller can + * handle it. + */ + if (!n) { + ASSERT(bma.nallocs >= *nmap); + return -ENOSR; + } + *nmap = n; return 0; error0: xfs_bmapi_finish(&bma, whichfork, error); @@ -4685,9 +4716,6 @@ xfs_bmapi_convert_one_delalloc( if (error) goto out_finish; - error = -ENOSPC; - if (WARN_ON_ONCE(bma.blkno == NULLFSBLOCK)) - goto out_finish; if (WARN_ON_ONCE(!xfs_valid_startblock(ip, bma.got.br_startblock))) { xfs_bmap_mark_sick(ip, whichfork); error = -EFSCORRUPTED; diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index b13796629e22..16a529a88780 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -2297,8 +2297,8 @@ xfs_da_grow_inode_int( struct xfs_inode *dp = args->dp; int w = args->whichfork; xfs_rfsblock_t nblks = dp->i_nblocks; - struct xfs_bmbt_irec map, *mapp; - int nmap, error, got, i, mapi; + struct xfs_bmbt_irec map, *mapp = ↦ + int nmap, error, got, i, mapi = 1; /* * Find a spot in the file space to put the new block. @@ -2314,14 +2314,7 @@ xfs_da_grow_inode_int( error = xfs_bmapi_write(tp, dp, *bno, count, xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA|XFS_BMAPI_CONTIG, args->total, &map, &nmap); - if (error) - return error; - - ASSERT(nmap <= 1); - if (nmap == 1) { - mapp = ↦ - mapi = 1; - } else if (nmap == 0 && count > 1) { + if (error == -ENOSPC && count > 1) { xfs_fileoff_t b; int c; @@ -2339,16 +2332,13 @@ xfs_da_grow_inode_int( args->total, &mapp[mapi], &nmap); if (error) goto out_free_map; - if (nmap < 1) - break; mapi += nmap; b = mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount; } - } else { - mapi = 0; - mapp = NULL; } + if (error) + goto out_free_map; /* * Count the blocks we got, make sure it matches the total. diff --git a/fs/xfs/scrub/quota_repair.c b/fs/xfs/scrub/quota_repair.c index 0bab4c30cb85..90cd1512bba9 100644 --- a/fs/xfs/scrub/quota_repair.c +++ b/fs/xfs/scrub/quota_repair.c @@ -77,8 +77,6 @@ xrep_quota_item_fill_bmap_hole( irec, &nmaps); if (error) return error; - if (nmaps != 1) - return -ENOSPC; dq->q_blkno = XFS_FSB_TO_DADDR(mp, irec->br_startblock); @@ -444,10 +442,6 @@ xrep_quota_data_fork( XFS_BMAPI_CONVERT, 0, &nrec, &nmap); if (error) goto out; - if (nmap != 1) { - error = -ENOSPC; - goto out; - } ASSERT(nrec.br_startoff == irec.br_startoff); ASSERT(nrec.br_blockcount == irec.br_blockcount); diff --git a/fs/xfs/scrub/rtbitmap_repair.c b/fs/xfs/scrub/rtbitmap_repair.c index 46f5d5f605c9..0fef98e9f834 100644 --- a/fs/xfs/scrub/rtbitmap_repair.c +++ b/fs/xfs/scrub/rtbitmap_repair.c @@ -108,8 +108,6 @@ xrep_rtbitmap_data_mappings( 0, &map, &nmaps); if (error) return error; - if (nmaps != 1) - return -EFSCORRUPTED; /* Commit new extent and all deferred work. */ error = xrep_defer_finish(sc); diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 53aa90a0ee3a..2e6f08198c07 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -721,33 +721,32 @@ xfs_alloc_file_space( if (error) goto error; - error = xfs_bmapi_write(tp, ip, startoffset_fsb, - allocatesize_fsb, XFS_BMAPI_PREALLOC, 0, imapp, - &nimaps); - if (error) - goto error; - - ip->i_diflags |= XFS_DIFLAG_PREALLOC; - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - - error = xfs_trans_commit(tp); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - if (error) - break; - /* * If the allocator cannot find a single free extent large * enough to cover the start block of the requested range, - * xfs_bmapi_write will return 0 but leave *nimaps set to 0. + * xfs_bmapi_write will return -ENOSR. * * In that case we simply need to keep looping with the same * startoffset_fsb so that one of the following allocations * will eventually reach the requested range. */ - if (nimaps) { + error = xfs_bmapi_write(tp, ip, startoffset_fsb, + allocatesize_fsb, XFS_BMAPI_PREALLOC, 0, imapp, + &nimaps); + if (error) { + if (error != -ENOSR) + goto error; + error = 0; + } else { startoffset_fsb += imapp->br_blockcount; allocatesize_fsb -= imapp->br_blockcount; } + + ip->i_diflags |= XFS_DIFLAG_PREALLOC; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + + error = xfs_trans_commit(tp); + xfs_iunlock(ip, XFS_ILOCK_EXCL); } return error; diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 13aba84bd64a..43acb4f0d174 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -357,7 +357,6 @@ xfs_dquot_disk_alloc( goto err_cancel; ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB); - ASSERT(nmaps == 1); ASSERT((map.br_startblock != DELAYSTARTBLOCK) && (map.br_startblock != HOLESTARTBLOCK)); diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index ec43aa7cec60..f5d0ed45721b 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -322,14 +322,6 @@ xfs_iomap_write_direct( if (error) goto out_unlock; - /* - * Copy any maps to caller's array and return any error. - */ - if (nimaps == 0) { - error = -ENOSPC; - goto out_unlock; - } - if (unlikely(!xfs_valid_startblock(ip, imap->br_startblock))) { xfs_bmap_mark_sick(ip, XFS_DATA_FORK); error = xfs_alert_fsblock_zero(ip, imap); diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 7da0e8f961d3..5ecb52a234be 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -430,13 +430,6 @@ xfs_reflink_fill_cow_hole( if (error) return error; - /* - * Allocation succeeded but the requested range was not even partially - * satisfied? Bail out! - */ - if (nimaps == 0) - return -ENOSPC; - convert: return xfs_reflink_convert_unwritten(ip, imap, cmap, convert_now); @@ -499,13 +492,6 @@ xfs_reflink_fill_delalloc( error = xfs_trans_commit(tp); if (error) return error; - - /* - * Allocation succeeded but the requested range was not even - * partially satisfied? Bail out! - */ - if (nimaps == 0) - return -ENOSPC; } while (cmap->br_startoff + cmap->br_blockcount <= imap->br_startoff); return xfs_reflink_convert_unwritten(ip, imap, cmap, convert_now); diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index b476a876478d..150f544445ca 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -709,8 +709,6 @@ xfs_growfs_rt_alloc( nmap = 1; error = xfs_bmapi_write(tp, ip, oblocks, nblocks - oblocks, XFS_BMAPI_METADATA, 0, &map, &nmap); - if (!error && nmap < 1) - error = -ENOSPC; if (error) goto out_trans_cancel; /* -- cgit v1.2.3-59-g8ed1b From b11ed354c9f725ece2b9440dd6343b42cd5d031c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:22 +0200 Subject: xfs: remove the unusued tmp_logflags variable in xfs_bmapi_allocate tmp_logflags is initialized to 0 and then ORed into bma->logflags, which isn't actually doing anything. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 14c9781ec0ce..d59fc8107d69 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4182,7 +4182,6 @@ xfs_bmapi_allocate( struct xfs_mount *mp = bma->ip->i_mount; int whichfork = xfs_bmapi_whichfork(bma->flags); struct xfs_ifork *ifp = xfs_ifork_ptr(bma->ip, whichfork); - int tmp_logflags = 0; int error; ASSERT(bma->length > 0); @@ -4253,8 +4252,6 @@ xfs_bmapi_allocate( error = xfs_bmap_add_extent_hole_real(bma->tp, bma->ip, whichfork, &bma->icur, &bma->cur, &bma->got, &bma->logflags, bma->flags); - - bma->logflags |= tmp_logflags; if (error) return error; -- cgit v1.2.3-59-g8ed1b From 04c609e6e5066294b60329420d3711e990c47abf Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:23 +0200 Subject: xfs: lift a xfs_valid_startblock into xfs_bmapi_allocate xfs_bmapi_convert_delalloc has a xfs_valid_startblock check on the block allocated by xfs_bmapi_allocate. Lift it into xfs_bmapi_allocate as we should assert the same for xfs_bmapi_write. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index d59fc8107d69..eade9689599f 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4221,6 +4221,11 @@ xfs_bmapi_allocate( if (bma->blkno == NULLFSBLOCK) return -ENOSPC; + if (WARN_ON_ONCE(!xfs_valid_startblock(bma->ip, bma->blkno))) { + xfs_bmap_mark_sick(bma->ip, whichfork); + return -EFSCORRUPTED; + } + if (bma->flags & XFS_BMAPI_ZERO) { error = xfs_zero_extent(bma->ip, bma->blkno, bma->length); if (error) @@ -4713,12 +4718,6 @@ xfs_bmapi_convert_one_delalloc( if (error) goto out_finish; - if (WARN_ON_ONCE(!xfs_valid_startblock(ip, bma.got.br_startblock))) { - xfs_bmap_mark_sick(ip, whichfork); - error = -EFSCORRUPTED; - goto out_finish; - } - XFS_STATS_ADD(mp, xs_xstrat_bytes, XFS_FSB_TO_B(mp, bma.length)); XFS_STATS_INC(mp, xs_xstrat_quick); -- cgit v1.2.3-59-g8ed1b From 9d06960341ec5f45d3d65bdead3fbce753455e8a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:24 +0200 Subject: xfs: don't open code XFS_FILBLKS_MIN in xfs_bmapi_write XFS_FILBLKS_MIN uses min_t and thus does the comparison using the correct xfs_filblks_t type. Use it in xfs_bmapi_write and slightly adjust the comment document th potential pitfall to take account of this Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index eade9689599f..6507dcaac438 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4528,14 +4528,11 @@ xfs_bmapi_write( * allocation length request (which can be 64 bits in * length) and the bma length request, which is * xfs_extlen_t and therefore 32 bits. Hence we have to - * check for 32-bit overflows and handle them here. + * be careful and do the min() using the larger type to + * avoid overflows. */ - if (len > (xfs_filblks_t)XFS_MAX_BMBT_EXTLEN) - bma.length = XFS_MAX_BMBT_EXTLEN; - else - bma.length = len; + bma.length = XFS_FILBLKS_MIN(len, XFS_MAX_BMBT_EXTLEN); - ASSERT(len > 0); ASSERT(bma.length > 0); error = xfs_bmapi_allocate(&bma); if (error) { -- cgit v1.2.3-59-g8ed1b From 2a9b99d45be0981536f6d3faf40ae3f58febdd49 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:25 +0200 Subject: xfs: pass the actual offset and len to allocate to xfs_bmapi_allocate xfs_bmapi_allocate currently overwrites offset and len when converting delayed allocations, and duplicates the length cap done for non-delalloc allocations. Move all that logic into the callers to avoid duplication and to make the calling conventions more obvious. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 32 ++++++++++++++++++-------------- 1 file changed, 18 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 6507dcaac438..c4126f37dea3 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4185,21 +4185,11 @@ xfs_bmapi_allocate( int error; ASSERT(bma->length > 0); + ASSERT(bma->length <= XFS_MAX_BMBT_EXTLEN); - /* - * For the wasdelay case, we could also just allocate the stuff asked - * for in this bmap call but that wouldn't be as good. - */ if (bma->wasdel) { - bma->length = (xfs_extlen_t)bma->got.br_blockcount; - bma->offset = bma->got.br_startoff; if (!xfs_iext_peek_prev_extent(ifp, &bma->icur, &bma->prev)) bma->prev.br_startoff = NULLFILEOFF; - } else { - bma->length = XFS_FILBLKS_MIN(bma->length, XFS_MAX_BMBT_EXTLEN); - if (!bma->eof) - bma->length = XFS_FILBLKS_MIN(bma->length, - bma->got.br_startoff - bma->offset); } if (bma->flags & XFS_BMAPI_CONTIG) @@ -4533,6 +4523,15 @@ xfs_bmapi_write( */ bma.length = XFS_FILBLKS_MIN(len, XFS_MAX_BMBT_EXTLEN); + if (wasdelay) { + bma.offset = bma.got.br_startoff; + bma.length = bma.got.br_blockcount; + } else { + if (!eof) + bma.length = XFS_FILBLKS_MIN(bma.length, + bma.got.br_startoff - bno); + } + ASSERT(bma.length > 0); error = xfs_bmapi_allocate(&bma); if (error) { @@ -4686,11 +4685,16 @@ xfs_bmapi_convert_one_delalloc( bma.tp = tp; bma.ip = ip; bma.wasdel = true; - bma.offset = bma.got.br_startoff; - bma.length = max_t(xfs_filblks_t, bma.got.br_blockcount, - XFS_MAX_BMBT_EXTLEN); bma.minleft = xfs_bmapi_minleft(tp, ip, whichfork); + /* + * Always allocate convert from the start of the delalloc extent even if + * that is outside the passed in range to create large contiguous + * extents on disk. + */ + bma.offset = bma.got.br_startoff; + bma.length = bma.got.br_blockcount; + /* * When we're converting the delalloc reservations backing dirty pages * in the page cache, we must be careful about how we create the new -- cgit v1.2.3-59-g8ed1b From a8bb258f703f42c322638022afa16808ca4a7d25 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:26 +0200 Subject: xfs: remove the xfs_iext_peek_prev_extent call in xfs_bmapi_allocate Both callers of xfs_bmapi_allocate already initialize bma->prev, don't redo that in xfs_bmapi_allocate. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index c4126f37dea3..d0aa64b73c78 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4187,11 +4187,6 @@ xfs_bmapi_allocate( ASSERT(bma->length > 0); ASSERT(bma->length <= XFS_MAX_BMBT_EXTLEN); - if (bma->wasdel) { - if (!xfs_iext_peek_prev_extent(ifp, &bma->icur, &bma->prev)) - bma->prev.br_startoff = NULLFILEOFF; - } - if (bma->flags & XFS_BMAPI_CONTIG) bma->minlen = bma->length; else -- cgit v1.2.3-59-g8ed1b From d69bee6a35d3c5e4873b9e164dd1a9711351a97c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:27 +0200 Subject: xfs: fix xfs_bmap_add_extent_delay_real for partial conversions xfs_bmap_add_extent_delay_real takes parts or all of a delalloc extent and converts them to a real extent. It is written to deal with any potential overlap of the to be converted range with the delalloc extent, but it turns out that currently only converting the entire extents, or a part starting at the beginning is actually exercised, as the only caller always tries to convert the entire delalloc extent, and either succeeds or at least progresses partially from the start. If it only converts a tiny part of a delalloc extent, the indirect block calculation for the new delalloc extent (da_new) might be equivalent to that of the existing delalloc extent (da_old). If this extent conversion now requires allocating an indirect block that gets accounted into da_new, leading to the assert that da_new must be smaller or equal to da_new unless we split the extent to trigger. Except for the assert that case is actually handled by just trying to allocate more space, as that already handled for the split case (which currently can't be reached at all), so just reusing it should be fine. Except that without dipping into the reserved block pool that would make it a bit too easy to trigger a fs shutdown due to ENOSPC. So in addition to adjusting the assert, also dip into the reserved block pool. Note that I could only reproduce the assert with a change to only convert the actually asked range instead of the full delalloc extent from xfs_bmapi_write. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index d0aa64b73c78..ec204c430d53 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -1570,6 +1570,7 @@ xfs_bmap_add_extent_delay_real( if (error) goto done; } + ASSERT(da_new <= da_old); break; case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG: @@ -1600,6 +1601,7 @@ xfs_bmap_add_extent_delay_real( if (error) goto done; } + ASSERT(da_new <= da_old); break; case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG: @@ -1634,6 +1636,7 @@ xfs_bmap_add_extent_delay_real( if (error) goto done; } + ASSERT(da_new <= da_old); break; case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING: @@ -1668,6 +1671,7 @@ xfs_bmap_add_extent_delay_real( goto done; } } + ASSERT(da_new <= da_old); break; case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG: @@ -1706,6 +1710,7 @@ xfs_bmap_add_extent_delay_real( if (error) goto done; } + ASSERT(da_new <= da_old); break; case BMAP_LEFT_FILLING: @@ -1796,6 +1801,7 @@ xfs_bmap_add_extent_delay_real( xfs_iext_update_extent(bma->ip, state, &bma->icur, &PREV); xfs_iext_next(ifp, &bma->icur); xfs_iext_update_extent(bma->ip, state, &bma->icur, &RIGHT); + ASSERT(da_new <= da_old); break; case BMAP_RIGHT_FILLING: @@ -1845,6 +1851,7 @@ xfs_bmap_add_extent_delay_real( PREV.br_blockcount = temp; xfs_iext_insert(bma->ip, &bma->icur, &PREV, state); xfs_iext_next(ifp, &bma->icur); + ASSERT(da_new <= da_old); break; case 0: @@ -1967,12 +1974,10 @@ xfs_bmap_add_extent_delay_real( } /* adjust for changes in reserved delayed indirect blocks */ - if (da_new < da_old) { + if (da_new < da_old) xfs_add_fdblocks(mp, da_old - da_new); - } else if (da_new > da_old) { - ASSERT(state == 0); - error = xfs_dec_fdblocks(mp, da_new - da_old, false); - } + else if (da_new > da_old) + error = xfs_dec_fdblocks(mp, da_new - da_old, true); xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork); done: -- cgit v1.2.3-59-g8ed1b From 21255afdd7296f57dd65f815301426bcf911c82d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 29 Apr 2024 08:15:28 +0200 Subject: xfs: do not allocate the entire delalloc extent in xfs_bmapi_write MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit While trying to convert the entire delalloc extent is a good decision for regular writeback as it leads to larger contigous on-disk extents, but for other callers of xfs_bmapi_write is is rather questionable as it forced them to loop creating new transactions just in case there is no large enough contiguous extent to cover the whole delalloc reservation. Change xfs_bmapi_write to only allocate the passed in range instead, whіle the writeback path through xfs_bmapi_convert_delalloc and xfs_bmapi_allocate still always converts the full extents. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index ec204c430d53..be515bac8ea1 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4524,8 +4524,9 @@ xfs_bmapi_write( bma.length = XFS_FILBLKS_MIN(len, XFS_MAX_BMBT_EXTLEN); if (wasdelay) { - bma.offset = bma.got.br_startoff; - bma.length = bma.got.br_blockcount; + bma.length = XFS_FILBLKS_MIN(bma.length, + bma.got.br_blockcount - + (bno - bma.got.br_startoff)); } else { if (!eof) bma.length = XFS_FILBLKS_MIN(bma.length, -- cgit v1.2.3-59-g8ed1b From 63a6ce5a1a6261e4c70bad2b55c4e0de8da4762e Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Tue, 16 Apr 2024 08:07:00 +0930 Subject: btrfs: set correct ram_bytes when splitting ordered extent [BUG] When running generic/287, the following file extent items can be generated: item 16 key (258 EXTENT_DATA 2682880) itemoff 15305 itemsize 53 generation 9 type 1 (regular) extent data disk byte 1378414592 nr 462848 extent data offset 0 nr 462848 ram 2097152 extent compression 0 (none) Note that file extent item is not a compressed one, but its ram_bytes is way larger than its disk_num_bytes. According to btrfs on-disk scheme, ram_bytes should match disk_num_bytes if it's not a compressed one. [CAUSE] Since commit b73a6fd1b1ef ("btrfs: split partial dio bios before submit"), for partial dio writes, we would split the ordered extent. However the function btrfs_split_ordered_extent() doesn't update the ram_bytes even it has already shrunk the disk_num_bytes. Originally the function btrfs_split_ordered_extent() is only introduced for zoned devices in commit d22002fd37bd ("btrfs: zoned: split ordered extent when bio is sent"), but later commit b73a6fd1b1ef ("btrfs: split partial dio bios before submit") makes non-zoned btrfs affected. Thankfully for un-compressed file extent, we do not really utilize the ram_bytes member, thus it won't cause any real problem. [FIX] Also update btrfs_ordered_extent::ram_bytes inside btrfs_split_ordered_extent(). Fixes: d22002fd37bd ("btrfs: zoned: split ordered extent when bio is sent") CC: stable@vger.kernel.org # 5.15+ Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ordered-data.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index b749ba45da2b..c2a42bcde98e 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -1188,6 +1188,7 @@ struct btrfs_ordered_extent *btrfs_split_ordered_extent( ordered->disk_bytenr += len; ordered->num_bytes -= len; ordered->disk_num_bytes -= len; + ordered->ram_bytes -= len; if (test_bit(BTRFS_ORDERED_IO_DONE, &ordered->flags)) { ASSERT(ordered->bytes_left == 0); -- cgit v1.2.3-59-g8ed1b From 74e797d79cf131d9b1a54001cf64a0b492a83e60 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Mar 2024 08:51:38 +0000 Subject: mm: Provide a means of invalidation without using launder_folio Implement a replacement for launder_folio. The key feature of invalidate_inode_pages2() is that it locks each folio individually, unmaps it to prevent mmap'd accesses interfering and calls the ->launder_folio() address_space op to flush it. This has problems: firstly, each folio is written individually as one or more small writes; secondly, adjacent folios cannot be added so easily into the laundry; thirdly, it's yet another op to implement. Instead, use the invalidate lock to cause anyone wanting to add a folio to the inode to wait, then unmap all the folios if we have mmaps, then, conditionally, use ->writepages() to flush any dirty data back and then discard all pages. The invalidate lock prevents ->read_iter(), ->write_iter() and faulting through mmap all from adding pages for the duration. This is then used from netfslib to handle the flusing in unbuffered and direct writes. Signed-off-by: David Howells cc: Matthew Wilcox cc: Miklos Szeredi cc: Trond Myklebust cc: Christoph Hellwig cc: Andrew Morton cc: Alexander Viro cc: Christian Brauner cc: Jeff Layton cc: linux-mm@kvack.org cc: linux-fsdevel@vger.kernel.org cc: netfs@lists.linux.dev cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org cc: ceph-devel@vger.kernel.org cc: linux-cifs@vger.kernel.org cc: linux-nfs@vger.kernel.org cc: devel@lists.orangefs.org --- fs/netfs/direct_write.c | 28 +++++++++++++++++++++---- include/linux/pagemap.h | 2 ++ mm/filemap.c | 54 +++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 80 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c index bee047e20f5d..2b81cd4aae6e 100644 --- a/fs/netfs/direct_write.c +++ b/fs/netfs/direct_write.c @@ -132,12 +132,14 @@ out: ssize_t netfs_unbuffered_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; - struct inode *inode = file->f_mapping->host; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; struct netfs_inode *ictx = netfs_inode(inode); - unsigned long long end; ssize_t ret; + loff_t pos = iocb->ki_pos; + unsigned long long end = pos + iov_iter_count(from) - 1; - _enter("%llx,%zx,%llx", iocb->ki_pos, iov_iter_count(from), i_size_read(inode)); + _enter("%llx,%zx,%llx", pos, iov_iter_count(from), i_size_read(inode)); if (!iov_iter_count(from)) return 0; @@ -157,7 +159,25 @@ ssize_t netfs_unbuffered_write_iter(struct kiocb *iocb, struct iov_iter *from) ret = file_update_time(file); if (ret < 0) goto out; - ret = kiocb_invalidate_pages(iocb, iov_iter_count(from)); + if (iocb->ki_flags & IOCB_NOWAIT) { + /* We could block if there are any pages in the range. */ + ret = -EAGAIN; + if (filemap_range_has_page(mapping, pos, end)) + if (filemap_invalidate_inode(inode, true, pos, end)) + goto out; + } else { + ret = filemap_write_and_wait_range(mapping, pos, end); + if (ret < 0) + goto out; + } + + /* + * After a write we want buffered reads to be sure to go to disk to get + * the new data. We invalidate clean cached page from the region we're + * about to write. We do this *before* the write so that we can return + * without clobbering -EIOCBQUEUED from ->direct_IO(). + */ + ret = filemap_invalidate_inode(inode, true, pos, end); if (ret < 0) goto out; end = iocb->ki_pos + iov_iter_count(from); diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index 2df35e65557d..c5e33e2ca48a 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -40,6 +40,8 @@ int filemap_fdatawait_keep_errors(struct address_space *mapping); int filemap_fdatawait_range(struct address_space *, loff_t lstart, loff_t lend); int filemap_fdatawait_range_keep_errors(struct address_space *mapping, loff_t start_byte, loff_t end_byte); +int filemap_invalidate_inode(struct inode *inode, bool flush, + loff_t start, loff_t end); static inline int filemap_fdatawait(struct address_space *mapping) { diff --git a/mm/filemap.c b/mm/filemap.c index 9a2e28bf298a..1d6b3a369077 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -4134,6 +4134,60 @@ bool filemap_release_folio(struct folio *folio, gfp_t gfp) } EXPORT_SYMBOL(filemap_release_folio); +/** + * filemap_invalidate_inode - Invalidate/forcibly write back a range of an inode's pagecache + * @inode: The inode to flush + * @flush: Set to write back rather than simply invalidate. + * @start: First byte to in range. + * @end: Last byte in range (inclusive), or LLONG_MAX for everything from start + * onwards. + * + * Invalidate all the folios on an inode that contribute to the specified + * range, possibly writing them back first. Whilst the operation is + * undertaken, the invalidate lock is held to prevent new folios from being + * installed. + */ +int filemap_invalidate_inode(struct inode *inode, bool flush, + loff_t start, loff_t end) +{ + struct address_space *mapping = inode->i_mapping; + pgoff_t first = start >> PAGE_SHIFT; + pgoff_t last = end >> PAGE_SHIFT; + pgoff_t nr = end == LLONG_MAX ? ULONG_MAX : last - first + 1; + + if (!mapping || !mapping->nrpages || end < start) + goto out; + + /* Prevent new folios from being added to the inode. */ + filemap_invalidate_lock(mapping); + + if (!mapping->nrpages) + goto unlock; + + unmap_mapping_pages(mapping, first, nr, false); + + /* Write back the data if we're asked to. */ + if (flush) { + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = LONG_MAX, + .range_start = start, + .range_end = end, + }; + + filemap_fdatawrite_wbc(mapping, &wbc); + } + + /* Wait for writeback to complete on all folios and discard. */ + truncate_inode_pages_range(mapping, start, end); + +unlock: + filemap_invalidate_unlock(mapping); +out: + return filemap_check_errors(mapping); +} +EXPORT_SYMBOL_GPL(filemap_invalidate_inode); + #ifdef CONFIG_CACHESTAT_SYSCALL /** * filemap_cachestat() - compute the page cache statistics of a mapping -- cgit v1.2.3-59-g8ed1b From 40fb4828d5f6fc9dfe549e4fd8c9d705dea1315a Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Mar 2024 21:17:18 +0000 Subject: 9p: Use alternative invalidation to using launder_folio Use writepages-based flushing invalidation instead of invalidate_inode_pages2() and ->launder_folio(). This will allow ->launder_folio() to be removed eventually. Signed-off-by: David Howells cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: Jeff Layton cc: v9fs@lists.linux.dev cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/9p/vfs_addr.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 047855033d32..5a943c122d83 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -89,7 +89,6 @@ static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file) bool writing = (rreq->origin == NETFS_READ_FOR_WRITE || rreq->origin == NETFS_WRITEBACK || rreq->origin == NETFS_WRITETHROUGH || - rreq->origin == NETFS_LAUNDER_WRITE || rreq->origin == NETFS_UNBUFFERED_WRITE || rreq->origin == NETFS_DIO_WRITE); @@ -141,7 +140,6 @@ const struct address_space_operations v9fs_addr_operations = { .dirty_folio = netfs_dirty_folio, .release_folio = netfs_release_folio, .invalidate_folio = netfs_invalidate_folio, - .launder_folio = netfs_launder_folio, .direct_IO = noop_direct_IO, .writepages = netfs_writepages, }; -- cgit v1.2.3-59-g8ed1b From d73065e60dcc89c5cae9346ce651d83ac333898f Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Mar 2024 21:19:17 +0000 Subject: afs: Use alternative invalidation to using launder_folio Use writepages-based flushing invalidation instead of invalidate_inode_pages2() and ->launder_folio(). This will allow ->launder_folio() to be removed eventually. Signed-off-by: David Howells cc: Marc Dionne cc: Jeff Layton cc: linux-afs@lists.infradead.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/afs/file.c | 1 - fs/afs/internal.h | 1 - fs/afs/validation.c | 4 ++-- fs/afs/write.c | 10 +++------- 4 files changed, 5 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/afs/file.c b/fs/afs/file.c index ef2cc8f565d2..dfd8f60f5e1f 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -54,7 +54,6 @@ const struct address_space_operations afs_file_aops = { .read_folio = netfs_read_folio, .readahead = netfs_readahead, .dirty_folio = netfs_dirty_folio, - .launder_folio = netfs_launder_folio, .release_folio = netfs_release_folio, .invalidate_folio = netfs_invalidate_folio, .migrate_folio = filemap_migrate_folio, diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 6ce5a612937c..b93aa026daa4 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -916,7 +916,6 @@ struct afs_operation { loff_t pos; loff_t size; loff_t i_size; - bool laundering; /* Laundering page, PG_writeback not set */ } store; struct { struct iattr *attr; diff --git a/fs/afs/validation.c b/fs/afs/validation.c index 32a53fc8dfb2..bef8af12ebe2 100644 --- a/fs/afs/validation.c +++ b/fs/afs/validation.c @@ -365,9 +365,9 @@ static void afs_zap_data(struct afs_vnode *vnode) * written back in a regular file and completely discard the pages in a * directory or symlink */ if (S_ISREG(vnode->netfs.inode.i_mode)) - invalidate_remote_inode(&vnode->netfs.inode); + filemap_invalidate_inode(&vnode->netfs.inode, true, 0, LLONG_MAX); else - invalidate_inode_pages2(vnode->netfs.inode.i_mapping); + filemap_invalidate_inode(&vnode->netfs.inode, false, 0, LLONG_MAX); } /* diff --git a/fs/afs/write.c b/fs/afs/write.c index 74402d95a884..1bc26466eb72 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -75,8 +75,7 @@ static void afs_store_data_success(struct afs_operation *op) op->ctime = op->file[0].scb.status.mtime_client; afs_vnode_commit_status(op, &op->file[0]); if (!afs_op_error(op)) { - if (!op->store.laundering) - afs_pages_written_back(vnode, op->store.pos, op->store.size); + afs_pages_written_back(vnode, op->store.pos, op->store.size); afs_stat_v(vnode, n_stores); atomic_long_add(op->store.size, &afs_v2net(vnode)->n_store_bytes); } @@ -91,8 +90,7 @@ static const struct afs_operation_ops afs_store_data_operation = { /* * write to a file */ -static int afs_store_data(struct afs_vnode *vnode, struct iov_iter *iter, loff_t pos, - bool laundering) +static int afs_store_data(struct afs_vnode *vnode, struct iov_iter *iter, loff_t pos) { struct afs_operation *op; struct afs_wb_key *wbk = NULL; @@ -123,7 +121,6 @@ static int afs_store_data(struct afs_vnode *vnode, struct iov_iter *iter, loff_t op->file[0].modification = true; op->store.pos = pos; op->store.size = size; - op->store.laundering = laundering; op->flags |= AFS_OPERATION_UNINTR; op->ops = &afs_store_data_operation; @@ -168,8 +165,7 @@ static void afs_upload_to_server(struct netfs_io_subrequest *subreq) subreq->rreq->debug_id, subreq->debug_index, subreq->io_iter.count); trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - ret = afs_store_data(vnode, &subreq->io_iter, subreq->start, - subreq->rreq->origin == NETFS_LAUNDER_WRITE); + ret = afs_store_data(vnode, &subreq->io_iter, subreq->start); netfs_write_subrequest_terminated(subreq, ret < 0 ? ret : subreq->len, false); } -- cgit v1.2.3-59-g8ed1b From b4ff7b178bda0ce4ec9f799c6a85579ba17f0df3 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Mar 2024 17:58:39 +0000 Subject: netfs: Remove ->launder_folio() support Remove support for ->launder_folio() from netfslib and expect filesystems to use filemap_invalidate_inode() instead. netfs_launder_folio() can then be got rid of. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: David Howells cc: Marc Dionne cc: Steve French cc: Matthew Wilcox cc: linux-mm@kvack.org cc: linux-fsdevel@vger.kernel.org cc: netfs@lists.linux.dev cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org cc: ceph-devel@vger.kernel.org cc: linux-cifs@vger.kernel.org cc: devel@lists.orangefs.org --- fs/netfs/buffered_write.c | 74 -------------------------------------------- fs/netfs/main.c | 1 - include/linux/netfs.h | 2 -- include/trace/events/netfs.h | 3 -- 4 files changed, 80 deletions(-) (limited to 'fs') diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 57c6eab01261..d8f66ce94575 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -1200,77 +1200,3 @@ out: return ret; } EXPORT_SYMBOL(netfs_writepages); - -/* - * Deal with the disposition of a laundered folio. - */ -static void netfs_cleanup_launder_folio(struct netfs_io_request *wreq) -{ - if (wreq->error) { - pr_notice("R=%08x Laundering error %d\n", wreq->debug_id, wreq->error); - mapping_set_error(wreq->mapping, wreq->error); - } -} - -/** - * netfs_launder_folio - Clean up a dirty folio that's being invalidated - * @folio: The folio to clean - * - * This is called to write back a folio that's being invalidated when an inode - * is getting torn down. Ideally, writepages would be used instead. - */ -int netfs_launder_folio(struct folio *folio) -{ - struct netfs_io_request *wreq; - struct address_space *mapping = folio->mapping; - struct netfs_folio *finfo = netfs_folio_info(folio); - struct netfs_group *group = netfs_folio_group(folio); - struct bio_vec bvec; - unsigned long long i_size = i_size_read(mapping->host); - unsigned long long start = folio_pos(folio); - size_t offset = 0, len; - int ret = 0; - - if (finfo) { - offset = finfo->dirty_offset; - start += offset; - len = finfo->dirty_len; - } else { - len = folio_size(folio); - } - len = min_t(unsigned long long, len, i_size - start); - - wreq = netfs_alloc_request(mapping, NULL, start, len, NETFS_LAUNDER_WRITE); - if (IS_ERR(wreq)) { - ret = PTR_ERR(wreq); - goto out; - } - - if (!folio_clear_dirty_for_io(folio)) - goto out_put; - - trace_netfs_folio(folio, netfs_folio_trace_launder); - - _debug("launder %llx-%llx", start, start + len - 1); - - /* Speculatively write to the cache. We have to fix this up later if - * the store fails. - */ - wreq->cleanup = netfs_cleanup_launder_folio; - - bvec_set_folio(&bvec, folio, len, offset); - iov_iter_bvec(&wreq->iter, ITER_SOURCE, &bvec, 1, len); - if (group != NETFS_FOLIO_COPY_TO_CACHE) - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - ret = netfs_begin_write(wreq, true, netfs_write_trace_launder); - -out_put: - folio_detach_private(folio); - netfs_put_group(group); - kfree(finfo); - netfs_put_request(wreq, false, netfs_rreq_trace_put_return); -out: - _leave(" = %d", ret); - return ret; -} -EXPORT_SYMBOL(netfs_launder_folio); diff --git a/fs/netfs/main.c b/fs/netfs/main.c index c5a73c9ed126..844efbb2e7a2 100644 --- a/fs/netfs/main.c +++ b/fs/netfs/main.c @@ -34,7 +34,6 @@ static const char *netfs_origins[nr__netfs_io_origin] = { [NETFS_COPY_TO_CACHE] = "CC", [NETFS_WRITEBACK] = "WB", [NETFS_WRITETHROUGH] = "WT", - [NETFS_LAUNDER_WRITE] = "LW", [NETFS_UNBUFFERED_WRITE] = "UW", [NETFS_DIO_READ] = "DR", [NETFS_DIO_WRITE] = "DW", diff --git a/include/linux/netfs.h b/include/linux/netfs.h index ddafc6ebff42..3af589dabd7f 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -172,7 +172,6 @@ enum netfs_io_origin { NETFS_COPY_TO_CACHE, /* This write is to copy a read to the cache */ NETFS_WRITEBACK, /* This write was triggered by writepages */ NETFS_WRITETHROUGH, /* This write was made by netfs_perform_write() */ - NETFS_LAUNDER_WRITE, /* This is triggered by ->launder_folio() */ NETFS_UNBUFFERED_WRITE, /* This is an unbuffered write */ NETFS_DIO_READ, /* This is a direct I/O read */ NETFS_DIO_WRITE, /* This is a direct I/O write */ @@ -352,7 +351,6 @@ int netfs_unpin_writeback(struct inode *inode, struct writeback_control *wbc); void netfs_clear_inode_writeback(struct inode *inode, const void *aux); void netfs_invalidate_folio(struct folio *folio, size_t offset, size_t length); bool netfs_release_folio(struct folio *folio, gfp_t gfp); -int netfs_launder_folio(struct folio *folio); /* VMA operations API. */ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_group); diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h index e03fafb0c1e3..30769103638f 100644 --- a/include/trace/events/netfs.h +++ b/include/trace/events/netfs.h @@ -26,7 +26,6 @@ #define netfs_write_traces \ EM(netfs_write_trace_copy_to_cache, "COPY2CACH") \ EM(netfs_write_trace_dio_write, "DIO-WRITE") \ - EM(netfs_write_trace_launder, "LAUNDER ") \ EM(netfs_write_trace_unbuffered_write, "UNB-WRITE") \ EM(netfs_write_trace_writeback, "WRITEBACK") \ E_(netfs_write_trace_writethrough, "WRITETHRU") @@ -38,7 +37,6 @@ EM(NETFS_COPY_TO_CACHE, "CC") \ EM(NETFS_WRITEBACK, "WB") \ EM(NETFS_WRITETHROUGH, "WT") \ - EM(NETFS_LAUNDER_WRITE, "LW") \ EM(NETFS_UNBUFFERED_WRITE, "UW") \ EM(NETFS_DIO_READ, "DR") \ E_(NETFS_DIO_WRITE, "DW") @@ -135,7 +133,6 @@ EM(netfs_folio_trace_end_copy, "end-copy") \ EM(netfs_folio_trace_filled_gaps, "filled-gaps") \ EM(netfs_folio_trace_kill, "kill") \ - EM(netfs_folio_trace_launder, "launder") \ EM(netfs_folio_trace_mkwrite, "mkwrite") \ EM(netfs_folio_trace_mkwrite_plus, "mkwrite+") \ EM(netfs_folio_trace_read_gaps, "read-gaps") \ -- cgit v1.2.3-59-g8ed1b From d9f85a04fb0eee0171f451fb4c4875b8a00eeaec Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 15 Mar 2024 14:37:18 +0000 Subject: netfs: Use mempools for allocating requests and subrequests Use mempools for allocating requests and subrequests in an effort to make sure that allocation always succeeds so that when performing writeback we can always make progress. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/netfs/internal.h | 2 ++ fs/netfs/main.c | 51 ++++++++++++++++++++++++++++++++++++++------ fs/netfs/objects.c | 59 ++++++++++++++++++++++++++++++++++----------------- include/linux/netfs.h | 5 +++-- 4 files changed, 89 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index 156ab138e224..c67da478cd2b 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -37,6 +37,8 @@ int netfs_begin_read(struct netfs_io_request *rreq, bool sync); extern unsigned int netfs_debug; extern struct list_head netfs_io_requests; extern spinlock_t netfs_proc_lock; +extern mempool_t netfs_request_pool; +extern mempool_t netfs_subrequest_pool; #ifdef CONFIG_PROC_FS static inline void netfs_proc_add_rreq(struct netfs_io_request *rreq) diff --git a/fs/netfs/main.c b/fs/netfs/main.c index 844efbb2e7a2..4805b9377364 100644 --- a/fs/netfs/main.c +++ b/fs/netfs/main.c @@ -7,6 +7,7 @@ #include #include +#include #include #include #include "internal.h" @@ -23,6 +24,11 @@ unsigned netfs_debug; module_param_named(debug, netfs_debug, uint, S_IWUSR | S_IRUGO); MODULE_PARM_DESC(netfs_debug, "Netfs support debugging mask"); +static struct kmem_cache *netfs_request_slab; +static struct kmem_cache *netfs_subrequest_slab; +mempool_t netfs_request_pool; +mempool_t netfs_subrequest_pool; + #ifdef CONFIG_PROC_FS LIST_HEAD(netfs_io_requests); DEFINE_SPINLOCK(netfs_proc_lock); @@ -98,25 +104,54 @@ static int __init netfs_init(void) { int ret = -ENOMEM; + netfs_request_slab = kmem_cache_create("netfs_request", + sizeof(struct netfs_io_request), 0, + SLAB_HWCACHE_ALIGN | SLAB_ACCOUNT, + NULL); + if (!netfs_request_slab) + goto error_req; + + if (mempool_init_slab_pool(&netfs_request_pool, 100, netfs_request_slab) < 0) + goto error_reqpool; + + netfs_subrequest_slab = kmem_cache_create("netfs_subrequest", + sizeof(struct netfs_io_subrequest), 0, + SLAB_HWCACHE_ALIGN | SLAB_ACCOUNT, + NULL); + if (!netfs_subrequest_slab) + goto error_subreq; + + if (mempool_init_slab_pool(&netfs_subrequest_pool, 100, netfs_subrequest_slab) < 0) + goto error_subreqpool; + if (!proc_mkdir("fs/netfs", NULL)) - goto error; + goto error_proc; if (!proc_create_seq("fs/netfs/requests", S_IFREG | 0444, NULL, &netfs_requests_seq_ops)) - goto error_proc; + goto error_procfile; #ifdef CONFIG_FSCACHE_STATS if (!proc_create_single("fs/netfs/stats", S_IFREG | 0444, NULL, netfs_stats_show)) - goto error_proc; + goto error_procfile; #endif ret = fscache_init(); if (ret < 0) - goto error_proc; + goto error_fscache; return 0; -error_proc: +error_fscache: +error_procfile: remove_proc_entry("fs/netfs", NULL); -error: +error_proc: + mempool_exit(&netfs_subrequest_pool); +error_subreqpool: + kmem_cache_destroy(netfs_subrequest_slab); +error_subreq: + mempool_exit(&netfs_request_pool); +error_reqpool: + kmem_cache_destroy(netfs_request_slab); +error_req: return ret; } fs_initcall(netfs_init); @@ -125,5 +160,9 @@ static void __exit netfs_exit(void) { fscache_exit(); remove_proc_entry("fs/netfs", NULL); + mempool_exit(&netfs_subrequest_pool); + kmem_cache_destroy(netfs_subrequest_slab); + mempool_exit(&netfs_request_pool); + kmem_cache_destroy(netfs_request_slab); } module_exit(netfs_exit); diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c index 8acc03a64059..1a4e2ce735ce 100644 --- a/fs/netfs/objects.c +++ b/fs/netfs/objects.c @@ -6,6 +6,8 @@ */ #include +#include +#include #include "internal.h" /* @@ -20,17 +22,22 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping, struct inode *inode = file ? file_inode(file) : mapping->host; struct netfs_inode *ctx = netfs_inode(inode); struct netfs_io_request *rreq; + mempool_t *mempool = ctx->ops->request_pool ?: &netfs_request_pool; + struct kmem_cache *cache = mempool->pool_data; bool is_unbuffered = (origin == NETFS_UNBUFFERED_WRITE || origin == NETFS_DIO_READ || origin == NETFS_DIO_WRITE); bool cached = !is_unbuffered && netfs_is_cache_enabled(ctx); int ret; - rreq = kzalloc(ctx->ops->io_request_size ?: sizeof(struct netfs_io_request), - GFP_KERNEL); - if (!rreq) - return ERR_PTR(-ENOMEM); + for (;;) { + rreq = mempool_alloc(mempool, GFP_KERNEL); + if (rreq) + break; + msleep(10); + } + memset(rreq, 0, kmem_cache_size(cache)); rreq->start = start; rreq->len = len; rreq->upper_len = len; @@ -56,7 +63,7 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping, if (rreq->netfs_ops->init_request) { ret = rreq->netfs_ops->init_request(rreq, file); if (ret < 0) { - kfree(rreq); + mempool_free(rreq, rreq->netfs_ops->request_pool ?: &netfs_request_pool); return ERR_PTR(ret); } } @@ -88,6 +95,14 @@ void netfs_clear_subrequests(struct netfs_io_request *rreq, bool was_async) } } +static void netfs_free_request_rcu(struct rcu_head *rcu) +{ + struct netfs_io_request *rreq = container_of(rcu, struct netfs_io_request, rcu); + + mempool_free(rreq, rreq->netfs_ops->request_pool ?: &netfs_request_pool); + netfs_stat_d(&netfs_n_rh_rreq); +} + static void netfs_free_request(struct work_struct *work) { struct netfs_io_request *rreq = @@ -110,8 +125,7 @@ static void netfs_free_request(struct work_struct *work) } kvfree(rreq->direct_bv); } - kfree_rcu(rreq, rcu); - netfs_stat_d(&netfs_n_rh_rreq); + call_rcu(&rreq->rcu, netfs_free_request_rcu); } void netfs_put_request(struct netfs_io_request *rreq, bool was_async, @@ -143,20 +157,25 @@ void netfs_put_request(struct netfs_io_request *rreq, bool was_async, struct netfs_io_subrequest *netfs_alloc_subrequest(struct netfs_io_request *rreq) { struct netfs_io_subrequest *subreq; - - subreq = kzalloc(rreq->netfs_ops->io_subrequest_size ?: - sizeof(struct netfs_io_subrequest), - GFP_KERNEL); - if (subreq) { - INIT_WORK(&subreq->work, NULL); - INIT_LIST_HEAD(&subreq->rreq_link); - refcount_set(&subreq->ref, 2); - subreq->rreq = rreq; - subreq->debug_index = atomic_inc_return(&rreq->subreq_counter); - netfs_get_request(rreq, netfs_rreq_trace_get_subreq); - netfs_stat(&netfs_n_rh_sreq); + mempool_t *mempool = rreq->netfs_ops->subrequest_pool ?: &netfs_subrequest_pool; + struct kmem_cache *cache = mempool->pool_data; + + for (;;) { + subreq = mempool_alloc(rreq->netfs_ops->subrequest_pool ?: &netfs_subrequest_pool, + GFP_KERNEL); + if (subreq) + break; + msleep(10); } + memset(subreq, 0, kmem_cache_size(cache)); + INIT_WORK(&subreq->work, NULL); + INIT_LIST_HEAD(&subreq->rreq_link); + refcount_set(&subreq->ref, 2); + subreq->rreq = rreq; + subreq->debug_index = atomic_inc_return(&rreq->subreq_counter); + netfs_get_request(rreq, netfs_rreq_trace_get_subreq); + netfs_stat(&netfs_n_rh_sreq); return subreq; } @@ -178,7 +197,7 @@ static void netfs_free_subrequest(struct netfs_io_subrequest *subreq, trace_netfs_sreq(subreq, netfs_sreq_trace_free); if (rreq->netfs_ops->free_subrequest) rreq->netfs_ops->free_subrequest(subreq); - kfree(subreq); + mempool_free(subreq, rreq->netfs_ops->subrequest_pool ?: &netfs_subrequest_pool); netfs_stat_d(&netfs_n_rh_sreq); netfs_put_request(rreq, was_async, netfs_rreq_trace_put_subreq); } diff --git a/include/linux/netfs.h b/include/linux/netfs.h index 3af589dabd7f..0b6c2c2d3c23 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -20,6 +20,7 @@ #include enum netfs_sreq_ref_trace; +typedef struct mempool_s mempool_t; /** * folio_start_private_2 - Start an fscache write on a folio. [DEPRECATED] @@ -236,8 +237,8 @@ struct netfs_io_request { * Operations the network filesystem can/must provide to the helpers. */ struct netfs_request_ops { - unsigned int io_request_size; /* Alloc size for netfs_io_request struct */ - unsigned int io_subrequest_size; /* Alloc size for netfs_io_subrequest struct */ + mempool_t *request_pool; + mempool_t *subrequest_pool; int (*init_request)(struct netfs_io_request *rreq, struct file *file); void (*free_request)(struct netfs_io_request *rreq); void (*free_subrequest)(struct netfs_io_subrequest *rreq); -- cgit v1.2.3-59-g8ed1b From 7ba167c4c73ed96eb002c98a9d7d49317dfb0191 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 16:57:31 +0000 Subject: netfs: Switch to using unsigned long long rather than loff_t Switch to using unsigned long long rather than loff_t in netfslib to avoid problems with the sign flipping in the maths when we're dealing with the byte at position 0x7fffffffffffffff. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Ilya Dryomov cc: Xiubo Li cc: netfs@lists.linux.dev cc: ceph-devel@vger.kernel.org cc: linux-fsdevel@vger.kernel.org --- fs/cachefiles/io.c | 2 +- fs/ceph/addr.c | 2 +- fs/netfs/buffered_read.c | 4 +++- fs/netfs/buffered_write.c | 2 +- fs/netfs/io.c | 6 +++--- fs/netfs/main.c | 2 +- fs/netfs/output.c | 4 ++-- include/linux/netfs.h | 16 +++++++++------- include/trace/events/netfs.h | 6 +++--- 9 files changed, 24 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/cachefiles/io.c b/fs/cachefiles/io.c index 1d685357e67f..5ba5c7814fe4 100644 --- a/fs/cachefiles/io.c +++ b/fs/cachefiles/io.c @@ -493,7 +493,7 @@ out_no_object: * boundary as appropriate. */ static enum netfs_io_source cachefiles_prepare_read(struct netfs_io_subrequest *subreq, - loff_t i_size) + unsigned long long i_size) { return cachefiles_do_prepare_read(&subreq->rreq->cache_resources, subreq->start, &subreq->len, i_size, diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 74bfd10b1b1a..8c16bc5250ef 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -193,7 +193,7 @@ static void ceph_netfs_expand_readahead(struct netfs_io_request *rreq) * block, but do not exceed the file size, unless the original * request already exceeds it. */ - new_end = min(round_up(end, lo->stripe_unit), rreq->i_size); + new_end = umin(round_up(end, lo->stripe_unit), rreq->i_size); if (new_end > end && new_end <= rreq->start + max_len) rreq->len = new_end - rreq->start; diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index 1622cce535a3..47603f08680e 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -130,7 +130,9 @@ void netfs_rreq_unlock_folios(struct netfs_io_request *rreq) } static void netfs_cache_expand_readahead(struct netfs_io_request *rreq, - loff_t *_start, size_t *_len, loff_t i_size) + unsigned long long *_start, + unsigned long long *_len, + unsigned long long i_size) { struct netfs_cache_resources *cres = &rreq->cache_resources; diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index d8f66ce94575..eba49bfafe64 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -664,7 +664,7 @@ static void netfs_pages_written_back(struct netfs_io_request *wreq) last = (wreq->start + wreq->len - 1) / PAGE_SIZE; xas_for_each(&xas, folio, last) { WARN(!folio_test_writeback(folio), - "bad %zx @%llx page %lx %lx\n", + "bad %llx @%llx page %lx %lx\n", wreq->len, wreq->start, folio->index, last); if ((finfo = netfs_folio_info(folio))) { diff --git a/fs/netfs/io.c b/fs/netfs/io.c index 8de581ac0cfb..6cfecfcd02e1 100644 --- a/fs/netfs/io.c +++ b/fs/netfs/io.c @@ -476,7 +476,7 @@ netfs_rreq_prepare_read(struct netfs_io_request *rreq, set: if (subreq->len > rreq->len) - pr_warn("R=%08x[%u] SREQ>RREQ %zx > %zx\n", + pr_warn("R=%08x[%u] SREQ>RREQ %zx > %llx\n", rreq->debug_id, subreq->debug_index, subreq->len, rreq->len); @@ -513,7 +513,7 @@ static bool netfs_rreq_submit_slice(struct netfs_io_request *rreq, subreq->start = rreq->start + rreq->submitted; subreq->len = io_iter->count; - _debug("slice %llx,%zx,%zx", subreq->start, subreq->len, rreq->submitted); + _debug("slice %llx,%zx,%llx", subreq->start, subreq->len, rreq->submitted); list_add_tail(&subreq->rreq_link, &rreq->subrequests); /* Call out to the cache to find out what it can do with the remaining @@ -588,7 +588,7 @@ int netfs_begin_read(struct netfs_io_request *rreq, bool sync) atomic_set(&rreq->nr_outstanding, 1); io_iter = rreq->io_iter; do { - _debug("submit %llx + %zx >= %llx", + _debug("submit %llx + %llx >= %llx", rreq->start, rreq->submitted, rreq->i_size); if (rreq->origin == NETFS_DIO_READ && rreq->start + rreq->submitted >= rreq->i_size) diff --git a/fs/netfs/main.c b/fs/netfs/main.c index 4805b9377364..5f0f438e5d21 100644 --- a/fs/netfs/main.c +++ b/fs/netfs/main.c @@ -62,7 +62,7 @@ static int netfs_requests_seq_show(struct seq_file *m, void *v) rreq = list_entry(v, struct netfs_io_request, proc_link); seq_printf(m, - "%08x %s %3d %2lx %4d %3d @%04llx %zx/%zx", + "%08x %s %3d %2lx %4d %3d @%04llx %llx/%llx", rreq->debug_id, netfs_origins[rreq->origin], refcount_read(&rreq->ref), diff --git a/fs/netfs/output.c b/fs/netfs/output.c index e586396d6b72..85374322f10f 100644 --- a/fs/netfs/output.c +++ b/fs/netfs/output.c @@ -439,7 +439,7 @@ static void netfs_submit_writethrough(struct netfs_io_request *wreq, bool final) */ int netfs_advance_writethrough(struct netfs_io_request *wreq, size_t copied, bool to_page_end) { - _enter("ic=%zu sb=%zu ws=%u cp=%zu tp=%u", + _enter("ic=%zu sb=%llu ws=%u cp=%zu tp=%u", wreq->iter.count, wreq->submitted, wreq->wsize, copied, to_page_end); wreq->iter.count += copied; @@ -457,7 +457,7 @@ int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb) { int ret = -EIOCBQUEUED; - _enter("ic=%zu sb=%zu ws=%u", + _enter("ic=%zu sb=%llu ws=%u", wreq->iter.count, wreq->submitted, wreq->wsize); if (wreq->submitted < wreq->io_iter.count) diff --git a/include/linux/netfs.h b/include/linux/netfs.h index 0b6c2c2d3c23..88269681d4fc 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -149,7 +149,7 @@ struct netfs_io_subrequest { struct work_struct work; struct list_head rreq_link; /* Link in rreq->subrequests */ struct iov_iter io_iter; /* Iterator for this subrequest */ - loff_t start; /* Where to start the I/O */ + unsigned long long start; /* Where to start the I/O */ size_t len; /* Size of the I/O */ size_t transferred; /* Amount of data transferred */ refcount_t ref; @@ -205,15 +205,15 @@ struct netfs_io_request { atomic_t subreq_counter; /* Next subreq->debug_index */ atomic_t nr_outstanding; /* Number of ops in progress */ atomic_t nr_copy_ops; /* Number of copy-to-cache ops in progress */ - size_t submitted; /* Amount submitted for I/O so far */ - size_t len; /* Length of the request */ size_t upper_len; /* Length can be extended to here */ + unsigned long long submitted; /* Amount submitted for I/O so far */ + unsigned long long len; /* Length of the request */ size_t transferred; /* Amount to be indicated as transferred */ short error; /* 0 or error that occurred */ enum netfs_io_origin origin; /* Origin of the request */ bool direct_bv_unpin; /* T if direct_bv[] must be unpinned */ - loff_t i_size; /* Size of the file */ - loff_t start; /* Start position */ + unsigned long long i_size; /* Size of the file */ + unsigned long long start; /* Start position */ pgoff_t no_unlock_folio; /* Don't unlock this folio after read */ refcount_t ref; unsigned long flags; @@ -294,13 +294,15 @@ struct netfs_cache_ops { /* Expand readahead request */ void (*expand_readahead)(struct netfs_cache_resources *cres, - loff_t *_start, size_t *_len, loff_t i_size); + unsigned long long *_start, + unsigned long long *_len, + unsigned long long i_size); /* Prepare a read operation, shortening it to a cached/uncached * boundary as appropriate. */ enum netfs_io_source (*prepare_read)(struct netfs_io_subrequest *subreq, - loff_t i_size); + unsigned long long i_size); /* Prepare a write operation, working out what part of the write we can * actually do. diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h index 30769103638f..7126d2ea459c 100644 --- a/include/trace/events/netfs.h +++ b/include/trace/events/netfs.h @@ -280,7 +280,7 @@ TRACE_EVENT(netfs_sreq, __entry->start = sreq->start; ), - TP_printk("R=%08x[%u] %s %s f=%02x s=%llx %zx/%zx e=%d", + TP_printk("R=%08x[%x] %s %s f=%02x s=%llx %zx/%zx e=%d", __entry->rreq, __entry->index, __print_symbolic(__entry->source, netfs_sreq_sources), __print_symbolic(__entry->what, netfs_sreq_traces), @@ -320,7 +320,7 @@ TRACE_EVENT(netfs_failure, __entry->start = sreq ? sreq->start : 0; ), - TP_printk("R=%08x[%d] %s f=%02x s=%llx %zx/%zx %s e=%d", + TP_printk("R=%08x[%x] %s f=%02x s=%llx %zx/%zx %s e=%d", __entry->rreq, __entry->index, __print_symbolic(__entry->source, netfs_sreq_sources), __entry->flags, @@ -436,7 +436,7 @@ TRACE_EVENT(netfs_write, __field(unsigned int, cookie ) __field(enum netfs_write_trace, what ) __field(unsigned long long, start ) - __field(size_t, len ) + __field(unsigned long long, len ) ), TP_fast_assign( -- cgit v1.2.3-59-g8ed1b From 288ace2f57c9d06dd2e42bd80d03747d879a4068 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 16:52:05 +0000 Subject: netfs: New writeback implementation The current netfslib writeback implementation creates writeback requests of contiguous folio data and then separately tiles subrequests over the space twice, once for the server and once for the cache. This creates a few issues: (1) Every time there's a discontiguity or a change between writing to only one destination or writing to both, it must create a new request. This makes it harder to do vectored writes. (2) The folios don't have the writeback mark removed until the end of the request - and a request could be hundreds of megabytes. (3) In future, I want to support a larger cache granularity, which will require aggregation of some folios that contain unmodified data (which only need to go to the cache) and some which contain modifications (which need to be uploaded and stored to the cache) - but, currently, these are treated as discontiguous. There's also a move to get everyone to use writeback_iter() to extract writable folios from the pagecache. That said, currently writeback_iter() has some issues that make it less than ideal: (1) there's no way to cancel the iteration, even if you find a "temporary" error that means the current folio and all subsequent folios are going to fail; (2) there's no way to filter the folios being written back - something that will impact Ceph with it's ordered snap system; (3) and if you get a folio you can't immediately deal with (say you need to flush the preceding writes), you are left with a folio hanging in the locked state for the duration, when really we should unlock it and relock it later. In this new implementation, I use writeback_iter() to pump folios, progressively creating two parallel, but separate streams and cleaning up the finished folios as the subrequests complete. Either or both streams can contain gaps, and the subrequests in each stream can be of variable size, don't need to align with each other and don't need to align with the folios. Indeed, subrequests can cross folio boundaries, may cover several folios or a folio may be spanned by multiple folios, e.g.: +---+---+-----+-----+---+----------+ Folios: | | | | | | | +---+---+-----+-----+---+----------+ +------+------+ +----+----+ Upload: | | |.....| | | +------+------+ +----+----+ +------+------+------+------+------+ Cache: | | | | | | +------+------+------+------+------+ The progressive subrequest construction permits the algorithm to be preparing both the next upload to the server and the next write to the cache whilst the previous ones are already in progress. Throttling can be applied to control the rate of production of subrequests - and, in any case, we probably want to write them to the server in ascending order, particularly if the file will be extended. Content crypto can also be prepared at the same time as the subrequests and run asynchronously, with the prepped requests being stalled until the crypto catches up with them. This might also be useful for transport crypto, but that happens at a lower layer, so probably would be harder to pull off. The algorithm is split into three parts: (1) The issuer. This walks through the data, packaging it up, encrypting it and creating subrequests. The part of this that generates subrequests only deals with file positions and spans and so is usable for DIO/unbuffered writes as well as buffered writes. (2) The collector. This asynchronously collects completed subrequests, unlocks folios, frees crypto buffers and performs any retries. This runs in a work queue so that the issuer can return to the caller for writeback (so that the VM can have its kswapd thread back) or async writes. (3) The retryer. This pauses the issuer, waits for all outstanding subrequests to complete and then goes through the failed subrequests to reissue them. This may involve reprepping them (with cifs, the credits must be renegotiated, and a subrequest may need splitting), and doing RMW for content crypto if there's a conflicting change on the server. [!] Note that some of the functions are prefixed with "new_" to avoid clashes with existing functions. These will be renamed in a later patch that cuts over to the new algorithm. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: Marc Dionne cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/netfs/Makefile | 4 +- fs/netfs/buffered_write.c | 4 - fs/netfs/internal.h | 27 ++ fs/netfs/objects.c | 17 + fs/netfs/write_collect.c | 803 +++++++++++++++++++++++++++++++++++++++++++ fs/netfs/write_issue.c | 683 ++++++++++++++++++++++++++++++++++++ include/linux/netfs.h | 68 +++- include/trace/events/netfs.h | 232 ++++++++++++- 8 files changed, 1829 insertions(+), 9 deletions(-) create mode 100644 fs/netfs/write_collect.c create mode 100644 fs/netfs/write_issue.c (limited to 'fs') diff --git a/fs/netfs/Makefile b/fs/netfs/Makefile index d4d1d799819e..1eb86e34b5a9 100644 --- a/fs/netfs/Makefile +++ b/fs/netfs/Makefile @@ -11,7 +11,9 @@ netfs-y := \ main.o \ misc.o \ objects.o \ - output.o + output.o \ + write_collect.o \ + write_issue.o netfs-$(CONFIG_NETFS_STATS) += stats.o diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index eba49bfafe64..84ac95ee4b4d 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -74,16 +74,12 @@ static enum netfs_how_to_modify netfs_how_to_modify(struct netfs_inode *ctx, if (file->f_mode & FMODE_READ) goto no_write_streaming; - if (test_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags)) - goto no_write_streaming; if (netfs_is_cache_enabled(ctx)) { /* We don't want to get a streaming write on a file that loses * caching service temporarily because the backing store got * culled. */ - if (!test_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags)) - set_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags); goto no_write_streaming; } diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index c67da478cd2b..dc11d1f67363 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -150,6 +150,33 @@ static inline void netfs_stat_d(atomic_t *stat) #define netfs_stat_d(x) do {} while(0) #endif +/* + * write_collect.c + */ +int netfs_folio_written_back(struct folio *folio); +void netfs_write_collection_worker(struct work_struct *work); +void netfs_wake_write_collector(struct netfs_io_request *wreq, bool was_async); + +/* + * write_issue.c + */ +struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, + struct file *file, + loff_t start, + enum netfs_io_origin origin); +void netfs_reissue_write(struct netfs_io_stream *stream, + struct netfs_io_subrequest *subreq); +int netfs_advance_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start, size_t len, bool to_eof); +struct netfs_io_request *new_netfs_begin_writethrough(struct kiocb *iocb, size_t len); +int new_netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *folio, size_t copied, bool to_page_end, + struct folio **writethrough_cache); +int new_netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *writethrough_cache); +int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len); + /* * Miscellaneous functions. */ diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c index 1a4e2ce735ce..c90d482b1650 100644 --- a/fs/netfs/objects.c +++ b/fs/netfs/objects.c @@ -47,6 +47,10 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping, rreq->inode = inode; rreq->i_size = i_size_read(inode); rreq->debug_id = atomic_inc_return(&debug_ids); + rreq->wsize = INT_MAX; + spin_lock_init(&rreq->lock); + INIT_LIST_HEAD(&rreq->io_streams[0].subrequests); + INIT_LIST_HEAD(&rreq->io_streams[1].subrequests); INIT_LIST_HEAD(&rreq->subrequests); INIT_WORK(&rreq->work, NULL); refcount_set(&rreq->ref, 1); @@ -85,6 +89,8 @@ void netfs_get_request(struct netfs_io_request *rreq, enum netfs_rreq_ref_trace void netfs_clear_subrequests(struct netfs_io_request *rreq, bool was_async) { struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream; + int s; while (!list_empty(&rreq->subrequests)) { subreq = list_first_entry(&rreq->subrequests, @@ -93,6 +99,17 @@ void netfs_clear_subrequests(struct netfs_io_request *rreq, bool was_async) netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_clear); } + + for (s = 0; s < ARRAY_SIZE(rreq->io_streams); s++) { + stream = &rreq->io_streams[s]; + while (!list_empty(&stream->subrequests)) { + subreq = list_first_entry(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + list_del(&subreq->rreq_link); + netfs_put_subrequest(subreq, was_async, + netfs_sreq_trace_put_clear); + } + } } static void netfs_free_request_rcu(struct rcu_head *rcu) diff --git a/fs/netfs/write_collect.c b/fs/netfs/write_collect.c new file mode 100644 index 000000000000..b8c1d3ca724a --- /dev/null +++ b/fs/netfs/write_collect.c @@ -0,0 +1,803 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem write subrequest result collection, assessment + * and retrying. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include +#include +#include +#include +#include +#include "internal.h" + +/* Notes made in the collector */ +#define HIT_PENDING 0x01 /* A front op was still pending */ +#define SOME_EMPTY 0x02 /* One of more streams are empty */ +#define ALL_EMPTY 0x04 /* All streams are empty */ +#define MAYBE_DISCONTIG 0x08 /* A front op may be discontiguous (rounded to PAGE_SIZE) */ +#define NEED_REASSESS 0x10 /* Need to loop round and reassess */ +#define REASSESS_DISCONTIG 0x20 /* Reassess discontiguity if contiguity advances */ +#define MADE_PROGRESS 0x40 /* Made progress cleaning up a stream or the folio set */ +#define BUFFERED 0x80 /* The pagecache needs cleaning up */ +#define NEED_RETRY 0x100 /* A front op requests retrying */ +#define SAW_FAILURE 0x200 /* One stream or hit a permanent failure */ + +/* + * Successful completion of write of a folio to the server and/or cache. Note + * that we are not allowed to lock the folio here on pain of deadlocking with + * truncate. + */ +int netfs_folio_written_back(struct folio *folio) +{ + enum netfs_folio_trace why = netfs_folio_trace_clear; + struct netfs_folio *finfo; + struct netfs_group *group = NULL; + int gcount = 0; + + if ((finfo = netfs_folio_info(folio))) { + /* Streaming writes cannot be redirtied whilst under writeback, + * so discard the streaming record. + */ + folio_detach_private(folio); + group = finfo->netfs_group; + gcount++; + kfree(finfo); + why = netfs_folio_trace_clear_s; + goto end_wb; + } + + if ((group = netfs_folio_group(folio))) { + if (group == NETFS_FOLIO_COPY_TO_CACHE) { + why = netfs_folio_trace_clear_cc; + folio_detach_private(folio); + goto end_wb; + } + + /* Need to detach the group pointer if the page didn't get + * redirtied. If it has been redirtied, then it must be within + * the same group. + */ + why = netfs_folio_trace_redirtied; + if (!folio_test_dirty(folio)) { + folio_detach_private(folio); + gcount++; + why = netfs_folio_trace_clear_g; + } + } + +end_wb: + trace_netfs_folio(folio, why); + folio_end_writeback(folio); + return gcount; +} + +/* + * Get hold of a folio we have under writeback. We don't want to get the + * refcount on it. + */ +static struct folio *netfs_writeback_lookup_folio(struct netfs_io_request *wreq, loff_t pos) +{ + XA_STATE(xas, &wreq->mapping->i_pages, pos / PAGE_SIZE); + struct folio *folio; + + rcu_read_lock(); + + for (;;) { + xas_reset(&xas); + folio = xas_load(&xas); + if (xas_retry(&xas, folio)) + continue; + + if (!folio || xa_is_value(folio)) + kdebug("R=%08x: folio %lx (%llx) not present", + wreq->debug_id, xas.xa_index, pos / PAGE_SIZE); + BUG_ON(!folio || xa_is_value(folio)); + + if (folio == xas_reload(&xas)) + break; + } + + rcu_read_unlock(); + + if (WARN_ONCE(!folio_test_writeback(folio), + "R=%08x: folio %lx is not under writeback\n", + wreq->debug_id, folio->index)) { + trace_netfs_folio(folio, netfs_folio_trace_not_under_wback); + } + return folio; +} + +/* + * Unlock any folios we've finished with. + */ +static void netfs_writeback_unlock_folios(struct netfs_io_request *wreq, + unsigned long long collected_to, + unsigned int *notes) +{ + for (;;) { + struct folio *folio; + struct netfs_folio *finfo; + unsigned long long fpos, fend; + size_t fsize, flen; + + folio = netfs_writeback_lookup_folio(wreq, wreq->cleaned_to); + + fpos = folio_pos(folio); + fsize = folio_size(folio); + finfo = netfs_folio_info(folio); + flen = finfo ? finfo->dirty_offset + finfo->dirty_len : fsize; + + fend = min_t(unsigned long long, fpos + flen, wreq->i_size); + + trace_netfs_collect_folio(wreq, folio, fend, collected_to); + + if (fpos + fsize > wreq->contiguity) { + trace_netfs_collect_contig(wreq, fpos + fsize, + netfs_contig_trace_unlock); + wreq->contiguity = fpos + fsize; + } + + /* Unlock any folio we've transferred all of. */ + if (collected_to < fend) + break; + + wreq->nr_group_rel += netfs_folio_written_back(folio); + wreq->cleaned_to = fpos + fsize; + *notes |= MADE_PROGRESS; + + if (fpos + fsize >= collected_to) + break; + } +} + +/* + * Perform retries on the streams that need it. + */ +static void netfs_retry_write_stream(struct netfs_io_request *wreq, + struct netfs_io_stream *stream) +{ + struct list_head *next; + + _enter("R=%x[%x:]", wreq->debug_id, stream->stream_nr); + + if (unlikely(stream->failed)) + return; + + /* If there's no renegotiation to do, just resend each failed subreq. */ + if (!stream->prepare_write) { + struct netfs_io_subrequest *subreq; + + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) + break; + if (__test_and_clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + __set_bit(NETFS_SREQ_RETRYING, &subreq->flags); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_write(stream, subreq); + } + } + return; + } + + if (list_empty(&stream->subrequests)) + return; + next = stream->subrequests.next; + + do { + struct netfs_io_subrequest *subreq = NULL, *from, *to, *tmp; + unsigned long long start, len; + size_t part; + bool boundary = false; + + /* Go through the stream and find the next span of contiguous + * data that we then rejig (cifs, for example, needs the wsize + * renegotiating) and reissue. + */ + from = list_entry(next, struct netfs_io_subrequest, rreq_link); + to = from; + start = from->start + from->transferred; + len = from->len - from->transferred; + + if (test_bit(NETFS_SREQ_FAILED, &from->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &from->flags)) + return; + + list_for_each_continue(next, &stream->subrequests) { + subreq = list_entry(next, struct netfs_io_subrequest, rreq_link); + if (subreq->start + subreq->transferred != start + len || + test_bit(NETFS_SREQ_BOUNDARY, &subreq->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) + break; + to = subreq; + len += to->len; + } + + /* Work through the sublist. */ + subreq = from; + list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { + if (!len) + break; + /* Renegotiate max_len (wsize) */ + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + __set_bit(NETFS_SREQ_RETRYING, &subreq->flags); + stream->prepare_write(subreq); + + part = min(len, subreq->max_len); + subreq->len = part; + subreq->start = start; + subreq->transferred = 0; + len -= part; + start += part; + if (len && subreq == to && + __test_and_clear_bit(NETFS_SREQ_BOUNDARY, &to->flags)) + boundary = true; + + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_write(stream, subreq); + if (subreq == to) + break; + } + + /* If we managed to use fewer subreqs, we can discard the + * excess; if we used the same number, then we're done. + */ + if (!len) { + if (subreq == to) + continue; + list_for_each_entry_safe_from(subreq, tmp, + &stream->subrequests, rreq_link) { + trace_netfs_sreq(subreq, netfs_sreq_trace_discard); + list_del(&subreq->rreq_link); + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); + if (subreq == to) + break; + } + continue; + } + + /* We ran out of subrequests, so we need to allocate some more + * and insert them after. + */ + do { + subreq = netfs_alloc_subrequest(wreq); + subreq->source = to->source; + subreq->start = start; + subreq->max_len = len; + subreq->max_nr_segs = INT_MAX; + subreq->debug_index = atomic_inc_return(&wreq->subreq_counter); + subreq->stream_nr = to->stream_nr; + __set_bit(NETFS_SREQ_RETRYING, &subreq->flags); + + trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), + netfs_sreq_trace_new); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + + list_add(&subreq->rreq_link, &to->rreq_link); + to = list_next_entry(to, rreq_link); + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + switch (stream->source) { + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload); + subreq->max_len = min(len, wreq->wsize); + break; + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write); + break; + default: + WARN_ON_ONCE(1); + } + + stream->prepare_write(subreq); + + part = min(len, subreq->max_len); + subreq->len = subreq->transferred + part; + len -= part; + start += part; + if (!len && boundary) { + __set_bit(NETFS_SREQ_BOUNDARY, &to->flags); + boundary = false; + } + + netfs_reissue_write(stream, subreq); + if (!len) + break; + + } while (len); + + } while (!list_is_head(next, &stream->subrequests)); +} + +/* + * Perform retries on the streams that need it. If we're doing content + * encryption and the server copy changed due to a third-party write, we may + * need to do an RMW cycle and also rewrite the data to the cache. + */ +static void netfs_retry_writes(struct netfs_io_request *wreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream; + int s; + + /* Wait for all outstanding I/O to quiesce before performing retries as + * we may need to renegotiate the I/O sizes. + */ + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (!stream->active) + continue; + + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + wait_on_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS, + TASK_UNINTERRUPTIBLE); + } + } + + // TODO: Enc: Fetch changed partial pages + // TODO: Enc: Reencrypt content if needed. + // TODO: Enc: Wind back transferred point. + // TODO: Enc: Mark cache pages for retry. + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->need_retry) { + stream->need_retry = false; + netfs_retry_write_stream(wreq, stream); + } + } +} + +/* + * Collect and assess the results of various write subrequests. We may need to + * retry some of the results - or even do an RMW cycle for content crypto. + * + * Note that we have a number of parallel, overlapping lists of subrequests, + * one to the server and one to the local cache for example, which may not be + * the same size or starting position and may not even correspond in boundary + * alignment. + */ +static void netfs_collect_write_results(struct netfs_io_request *wreq) +{ + struct netfs_io_subrequest *front, *remove; + struct netfs_io_stream *stream; + unsigned long long collected_to; + unsigned int notes; + int s; + + _enter("%llx-%llx", wreq->start, wreq->start + wreq->len); + trace_netfs_collect(wreq); + trace_netfs_rreq(wreq, netfs_rreq_trace_collect); + +reassess_streams: + smp_rmb(); + collected_to = ULLONG_MAX; + if (wreq->origin == NETFS_WRITEBACK) + notes = ALL_EMPTY | BUFFERED | MAYBE_DISCONTIG; + else if (wreq->origin == NETFS_WRITETHROUGH) + notes = ALL_EMPTY | BUFFERED; + else + notes = ALL_EMPTY; + + /* Remove completed subrequests from the front of the streams and + * advance the completion point on each stream. We stop when we hit + * something that's in progress. The issuer thread may be adding stuff + * to the tail whilst we're doing this. + * + * We must not, however, merge in discontiguities that span whole + * folios that aren't under writeback. This is made more complicated + * by the folios in the gap being of unpredictable sizes - if they even + * exist - but we don't want to look them up. + */ + for (s = 0; s < NR_IO_STREAMS; s++) { + loff_t rstart, rend; + + stream = &wreq->io_streams[s]; + /* Read active flag before list pointers */ + if (!smp_load_acquire(&stream->active)) + continue; + + front = stream->front; + while (front) { + trace_netfs_collect_sreq(wreq, front); + //_debug("sreq [%x] %llx %zx/%zx", + // front->debug_index, front->start, front->transferred, front->len); + + /* Stall if there may be a discontinuity. */ + rstart = round_down(front->start, PAGE_SIZE); + if (rstart > wreq->contiguity) { + if (wreq->contiguity > stream->collected_to) { + trace_netfs_collect_gap(wreq, stream, + wreq->contiguity, 'D'); + stream->collected_to = wreq->contiguity; + } + notes |= REASSESS_DISCONTIG; + break; + } + rend = round_up(front->start + front->len, PAGE_SIZE); + if (rend > wreq->contiguity) { + trace_netfs_collect_contig(wreq, rend, + netfs_contig_trace_collect); + wreq->contiguity = rend; + if (notes & REASSESS_DISCONTIG) + notes |= NEED_REASSESS; + } + notes &= ~MAYBE_DISCONTIG; + + /* Stall if the front is still undergoing I/O. */ + if (test_bit(NETFS_SREQ_IN_PROGRESS, &front->flags)) { + notes |= HIT_PENDING; + break; + } + smp_rmb(); /* Read counters after I-P flag. */ + + if (stream->failed) { + stream->collected_to = front->start + front->len; + notes |= MADE_PROGRESS | SAW_FAILURE; + goto cancel; + } + if (front->start + front->transferred > stream->collected_to) { + stream->collected_to = front->start + front->transferred; + stream->transferred = stream->collected_to - wreq->start; + notes |= MADE_PROGRESS; + } + if (test_bit(NETFS_SREQ_FAILED, &front->flags)) { + stream->failed = true; + stream->error = front->error; + if (stream->source == NETFS_UPLOAD_TO_SERVER) + mapping_set_error(wreq->mapping, front->error); + notes |= NEED_REASSESS | SAW_FAILURE; + break; + } + if (front->transferred < front->len) { + stream->need_retry = true; + notes |= NEED_RETRY | MADE_PROGRESS; + break; + } + + cancel: + /* Remove if completely consumed. */ + spin_lock(&wreq->lock); + + remove = front; + list_del_init(&front->rreq_link); + front = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + stream->front = front; + if (!front) { + unsigned long long jump_to = atomic64_read(&wreq->issued_to); + + if (stream->collected_to < jump_to) { + trace_netfs_collect_gap(wreq, stream, jump_to, 'A'); + stream->collected_to = jump_to; + } + } + + spin_unlock(&wreq->lock); + netfs_put_subrequest(remove, false, + notes & SAW_FAILURE ? + netfs_sreq_trace_put_cancel : + netfs_sreq_trace_put_done); + } + + if (front) + notes &= ~ALL_EMPTY; + else + notes |= SOME_EMPTY; + + if (stream->collected_to < collected_to) + collected_to = stream->collected_to; + } + + if (collected_to != ULLONG_MAX && collected_to > wreq->collected_to) + wreq->collected_to = collected_to; + + /* If we have an empty stream, we need to jump it forward over any gap + * otherwise the collection point will never advance. + * + * Note that the issuer always adds to the stream with the lowest + * so-far submitted start, so if we see two consecutive subreqs in one + * stream with nothing between then in another stream, then the second + * stream has a gap that can be jumped. + */ + if (notes & SOME_EMPTY) { + unsigned long long jump_to = wreq->start + wreq->len; + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->active && + stream->front && + stream->front->start < jump_to) + jump_to = stream->front->start; + } + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->active && + !stream->front && + stream->collected_to < jump_to) { + trace_netfs_collect_gap(wreq, stream, jump_to, 'B'); + stream->collected_to = jump_to; + } + } + } + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->active) + trace_netfs_collect_stream(wreq, stream); + } + + trace_netfs_collect_state(wreq, wreq->collected_to, notes); + + /* Unlock any folios that we have now finished with. */ + if (notes & BUFFERED) { + unsigned long long clean_to = min(wreq->collected_to, wreq->contiguity); + + if (wreq->cleaned_to < clean_to) + netfs_writeback_unlock_folios(wreq, clean_to, ¬es); + } else { + wreq->cleaned_to = wreq->collected_to; + } + + // TODO: Discard encryption buffers + + /* If all streams are discontiguous with the last folio we cleared, we + * may need to skip a set of folios. + */ + if ((notes & (MAYBE_DISCONTIG | ALL_EMPTY)) == MAYBE_DISCONTIG) { + unsigned long long jump_to = ULLONG_MAX; + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->active && stream->front && + stream->front->start < jump_to) + jump_to = stream->front->start; + } + + trace_netfs_collect_contig(wreq, jump_to, netfs_contig_trace_jump); + wreq->contiguity = jump_to; + wreq->cleaned_to = jump_to; + wreq->collected_to = jump_to; + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->collected_to < jump_to) + stream->collected_to = jump_to; + } + //cond_resched(); + notes |= MADE_PROGRESS; + goto reassess_streams; + } + + if (notes & NEED_RETRY) + goto need_retry; + if ((notes & MADE_PROGRESS) && test_bit(NETFS_RREQ_PAUSE, &wreq->flags)) { + trace_netfs_rreq(wreq, netfs_rreq_trace_unpause); + clear_bit_unlock(NETFS_RREQ_PAUSE, &wreq->flags); + wake_up_bit(&wreq->flags, NETFS_RREQ_PAUSE); + } + + if (notes & NEED_REASSESS) { + //cond_resched(); + goto reassess_streams; + } + if (notes & MADE_PROGRESS) { + //cond_resched(); + goto reassess_streams; + } + +out: + netfs_put_group_many(wreq->group, wreq->nr_group_rel); + wreq->nr_group_rel = 0; + _leave(" = %x", notes); + return; + +need_retry: + /* Okay... We're going to have to retry one or both streams. Note + * that any partially completed op will have had any wholly transferred + * folios removed from it. + */ + _debug("retry"); + netfs_retry_writes(wreq); + goto out; +} + +/* + * Perform the collection of subrequests, folios and encryption buffers. + */ +void netfs_write_collection_worker(struct work_struct *work) +{ + struct netfs_io_request *wreq = container_of(work, struct netfs_io_request, work); + struct netfs_inode *ictx = netfs_inode(wreq->inode); + size_t transferred; + int s; + + _enter("R=%x", wreq->debug_id); + + netfs_see_request(wreq, netfs_rreq_trace_see_work); + if (!test_bit(NETFS_RREQ_IN_PROGRESS, &wreq->flags)) { + netfs_put_request(wreq, false, netfs_rreq_trace_put_work); + return; + } + + netfs_collect_write_results(wreq); + + /* We're done when the app thread has finished posting subreqs and all + * the queues in all the streams are empty. + */ + if (!test_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags)) { + netfs_put_request(wreq, false, netfs_rreq_trace_put_work); + return; + } + smp_rmb(); /* Read ALL_QUEUED before lists. */ + + transferred = LONG_MAX; + for (s = 0; s < NR_IO_STREAMS; s++) { + struct netfs_io_stream *stream = &wreq->io_streams[s]; + if (!stream->active) + continue; + if (!list_empty(&stream->subrequests)) { + netfs_put_request(wreq, false, netfs_rreq_trace_put_work); + return; + } + if (stream->transferred < transferred) + transferred = stream->transferred; + } + + /* Okay, declare that all I/O is complete. */ + wreq->transferred = transferred; + trace_netfs_rreq(wreq, netfs_rreq_trace_write_done); + + if (wreq->io_streams[1].active && + wreq->io_streams[1].failed) { + /* Cache write failure doesn't prevent writeback completion + * unless we're in disconnected mode. + */ + ictx->ops->invalidate_cache(wreq); + } + + if (wreq->cleanup) + wreq->cleanup(wreq); + + if (wreq->origin == NETFS_DIO_WRITE && + wreq->mapping->nrpages) { + /* mmap may have got underfoot and we may now have folios + * locally covering the region we just wrote. Attempt to + * discard the folios, but leave in place any modified locally. + * ->write_iter() is prevented from interfering by the DIO + * counter. + */ + pgoff_t first = wreq->start >> PAGE_SHIFT; + pgoff_t last = (wreq->start + wreq->transferred - 1) >> PAGE_SHIFT; + invalidate_inode_pages2_range(wreq->mapping, first, last); + } + + if (wreq->origin == NETFS_DIO_WRITE) + inode_dio_end(wreq->inode); + + _debug("finished"); + trace_netfs_rreq(wreq, netfs_rreq_trace_wake_ip); + clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &wreq->flags); + wake_up_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS); + + if (wreq->iocb) { + wreq->iocb->ki_pos += wreq->transferred; + if (wreq->iocb->ki_complete) + wreq->iocb->ki_complete( + wreq->iocb, wreq->error ? wreq->error : wreq->transferred); + wreq->iocb = VFS_PTR_POISON; + } + + netfs_clear_subrequests(wreq, false); + netfs_put_request(wreq, false, netfs_rreq_trace_put_work_complete); +} + +/* + * Wake the collection work item. + */ +void netfs_wake_write_collector(struct netfs_io_request *wreq, bool was_async) +{ + if (!work_pending(&wreq->work)) { + netfs_get_request(wreq, netfs_rreq_trace_get_work); + if (!queue_work(system_unbound_wq, &wreq->work)) + netfs_put_request(wreq, was_async, netfs_rreq_trace_put_work_nq); + } +} + +/** + * new_netfs_write_subrequest_terminated - Note the termination of a write operation. + * @_op: The I/O request that has terminated. + * @transferred_or_error: The amount of data transferred or an error code. + * @was_async: The termination was asynchronous + * + * This tells the library that a contributory write I/O operation has + * terminated, one way or another, and that it should collect the results. + * + * The caller indicates in @transferred_or_error the outcome of the operation, + * supplying a positive value to indicate the number of bytes transferred or a + * negative error code. The library will look after reissuing I/O operations + * as appropriate and writing downloaded data to the cache. + * + * If @was_async is true, the caller might be running in softirq or interrupt + * context and we can't sleep. + * + * When this is called, ownership of the subrequest is transferred back to the + * library, along with a ref. + * + * Note that %_op is a void* so that the function can be passed to + * kiocb::term_func without the need for a casting wrapper. + */ +void new_netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, + bool was_async) +{ + struct netfs_io_subrequest *subreq = _op; + struct netfs_io_request *wreq = subreq->rreq; + struct netfs_io_stream *stream = &wreq->io_streams[subreq->stream_nr]; + + _enter("%x[%x] %zd", wreq->debug_id, subreq->debug_index, transferred_or_error); + + switch (subreq->source) { + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload_done); + break; + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write_done); + break; + case NETFS_INVALID_WRITE: + break; + default: + BUG(); + } + + if (IS_ERR_VALUE(transferred_or_error)) { + subreq->error = transferred_or_error; + if (subreq->error == -EAGAIN) + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + else + set_bit(NETFS_SREQ_FAILED, &subreq->flags); + trace_netfs_failure(wreq, subreq, transferred_or_error, netfs_fail_write); + + switch (subreq->source) { + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write_failed); + break; + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload_failed); + break; + default: + break; + } + trace_netfs_rreq(wreq, netfs_rreq_trace_set_pause); + set_bit(NETFS_RREQ_PAUSE, &wreq->flags); + } else { + if (WARN(transferred_or_error > subreq->len - subreq->transferred, + "Subreq excess write: R=%x[%x] %zd > %zu - %zu", + wreq->debug_id, subreq->debug_index, + transferred_or_error, subreq->len, subreq->transferred)) + transferred_or_error = subreq->len - subreq->transferred; + + subreq->error = 0; + subreq->transferred += transferred_or_error; + + if (subreq->transferred < subreq->len) + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + } + + trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); + + clear_bit_unlock(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + wake_up_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS); + + /* If we are at the head of the queue, wake up the collector, + * transferring a ref to it if we were the ones to do so. + */ + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) + netfs_wake_write_collector(wreq, was_async); + + netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); +} +EXPORT_SYMBOL(new_netfs_write_subrequest_terminated); diff --git a/fs/netfs/write_issue.c b/fs/netfs/write_issue.c new file mode 100644 index 000000000000..1b2e69934fbd --- /dev/null +++ b/fs/netfs/write_issue.c @@ -0,0 +1,683 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem high-level (buffered) writeback. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * + * To support network filesystems with local caching, we manage a situation + * that can be envisioned like the following: + * + * +---+---+-----+-----+---+----------+ + * Folios: | | | | | | | + * +---+---+-----+-----+---+----------+ + * + * +------+------+ +----+----+ + * Upload: | | |.....| | | + * (Stream 0) +------+------+ +----+----+ + * + * +------+------+------+------+------+ + * Cache: | | | | | | + * (Stream 1) +------+------+------+------+------+ + * + * Where we have a sequence of folios of varying sizes that we need to overlay + * with multiple parallel streams of I/O requests, where the I/O requests in a + * stream may also be of various sizes (in cifs, for example, the sizes are + * negotiated with the server; in something like ceph, they may represent the + * sizes of storage objects). + * + * The sequence in each stream may contain gaps and noncontiguous subrequests + * may be glued together into single vectored write RPCs. + */ + +#include +#include +#include +#include +#include "internal.h" + +/* + * Kill all dirty folios in the event of an unrecoverable error, starting with + * a locked folio we've already obtained from writeback_iter(). + */ +static void netfs_kill_dirty_pages(struct address_space *mapping, + struct writeback_control *wbc, + struct folio *folio) +{ + int error = 0; + + do { + enum netfs_folio_trace why = netfs_folio_trace_kill; + struct netfs_group *group = NULL; + struct netfs_folio *finfo = NULL; + void *priv; + + priv = folio_detach_private(folio); + if (priv) { + finfo = __netfs_folio_info(priv); + if (finfo) { + /* Kill folio from streaming write. */ + group = finfo->netfs_group; + why = netfs_folio_trace_kill_s; + } else { + group = priv; + if (group == NETFS_FOLIO_COPY_TO_CACHE) { + /* Kill copy-to-cache folio */ + why = netfs_folio_trace_kill_cc; + group = NULL; + } else { + /* Kill folio with group */ + why = netfs_folio_trace_kill_g; + } + } + } + + trace_netfs_folio(folio, why); + + folio_start_writeback(folio); + folio_unlock(folio); + folio_end_writeback(folio); + + netfs_put_group(group); + kfree(finfo); + + } while ((folio = writeback_iter(mapping, wbc, folio, &error))); +} + +/* + * Create a write request and set it up appropriately for the origin type. + */ +struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, + struct file *file, + loff_t start, + enum netfs_io_origin origin) +{ + struct netfs_io_request *wreq; + struct netfs_inode *ictx; + + wreq = netfs_alloc_request(mapping, file, start, 0, origin); + if (IS_ERR(wreq)) + return wreq; + + _enter("R=%x", wreq->debug_id); + + ictx = netfs_inode(wreq->inode); + if (test_bit(NETFS_RREQ_WRITE_TO_CACHE, &wreq->flags)) + fscache_begin_write_operation(&wreq->cache_resources, netfs_i_cookie(ictx)); + + wreq->contiguity = wreq->start; + wreq->cleaned_to = wreq->start; + INIT_WORK(&wreq->work, netfs_write_collection_worker); + + wreq->io_streams[0].stream_nr = 0; + wreq->io_streams[0].source = NETFS_UPLOAD_TO_SERVER; + wreq->io_streams[0].prepare_write = ictx->ops->prepare_write; + wreq->io_streams[0].issue_write = ictx->ops->issue_write; + wreq->io_streams[0].collected_to = start; + wreq->io_streams[0].transferred = LONG_MAX; + + wreq->io_streams[1].stream_nr = 1; + wreq->io_streams[1].source = NETFS_WRITE_TO_CACHE; + wreq->io_streams[1].collected_to = start; + wreq->io_streams[1].transferred = LONG_MAX; + if (fscache_resources_valid(&wreq->cache_resources)) { + wreq->io_streams[1].avail = true; + wreq->io_streams[1].prepare_write = wreq->cache_resources.ops->prepare_write_subreq; + wreq->io_streams[1].issue_write = wreq->cache_resources.ops->issue_write; + } + + return wreq; +} + +/** + * netfs_prepare_write_failed - Note write preparation failed + * @subreq: The subrequest to mark + * + * Mark a subrequest to note that preparation for write failed. + */ +void netfs_prepare_write_failed(struct netfs_io_subrequest *subreq) +{ + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + trace_netfs_sreq(subreq, netfs_sreq_trace_prep_failed); +} +EXPORT_SYMBOL(netfs_prepare_write_failed); + +/* + * Prepare a write subrequest. We need to allocate a new subrequest + * if we don't have one. + */ +static void netfs_prepare_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start) +{ + struct netfs_io_subrequest *subreq; + + subreq = netfs_alloc_subrequest(wreq); + subreq->source = stream->source; + subreq->start = start; + subreq->max_len = ULONG_MAX; + subreq->max_nr_segs = INT_MAX; + subreq->stream_nr = stream->stream_nr; + + _enter("R=%x[%x]", wreq->debug_id, subreq->debug_index); + + trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), + netfs_sreq_trace_new); + + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + + switch (stream->source) { + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload); + subreq->max_len = wreq->wsize; + break; + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write); + break; + default: + WARN_ON_ONCE(1); + break; + } + + if (stream->prepare_write) + stream->prepare_write(subreq); + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + /* We add to the end of the list whilst the collector may be walking + * the list. The collector only goes nextwards and uses the lock to + * remove entries off of the front. + */ + spin_lock(&wreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) { + stream->front = subreq; + if (!stream->active) { + stream->collected_to = stream->front->start; + /* Write list pointers before active flag */ + smp_store_release(&stream->active, true); + } + } + + spin_unlock(&wreq->lock); + + stream->construct = subreq; +} + +/* + * Set the I/O iterator for the filesystem/cache to use and dispatch the I/O + * operation. The operation may be asynchronous and should call + * netfs_write_subrequest_terminated() when complete. + */ +static void netfs_do_issue_write(struct netfs_io_stream *stream, + struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *wreq = subreq->rreq; + + _enter("R=%x[%x],%zx", wreq->debug_id, subreq->debug_index, subreq->len); + + if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) + return netfs_write_subrequest_terminated(subreq, subreq->error, false); + + // TODO: Use encrypted buffer + if (test_bit(NETFS_RREQ_USE_IO_ITER, &wreq->flags)) { + subreq->io_iter = wreq->io_iter; + iov_iter_advance(&subreq->io_iter, + subreq->start + subreq->transferred - wreq->start); + iov_iter_truncate(&subreq->io_iter, + subreq->len - subreq->transferred); + } else { + iov_iter_xarray(&subreq->io_iter, ITER_SOURCE, &wreq->mapping->i_pages, + subreq->start + subreq->transferred, + subreq->len - subreq->transferred); + } + + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + stream->issue_write(subreq); +} + +void netfs_reissue_write(struct netfs_io_stream *stream, + struct netfs_io_subrequest *subreq) +{ + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + netfs_do_issue_write(stream, subreq); +} + +static void netfs_issue_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream) +{ + struct netfs_io_subrequest *subreq = stream->construct; + + if (!subreq) + return; + stream->construct = NULL; + + if (subreq->start + subreq->len > wreq->start + wreq->submitted) + wreq->len = wreq->submitted = subreq->start + subreq->len - wreq->start; + netfs_do_issue_write(stream, subreq); +} + +/* + * Add data to the write subrequest, dispatching each as we fill it up or if it + * is discontiguous with the previous. We only fill one part at a time so that + * we can avoid overrunning the credits obtained (cifs) and try to parallelise + * content-crypto preparation with network writes. + */ +int netfs_advance_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start, size_t len, bool to_eof) +{ + struct netfs_io_subrequest *subreq = stream->construct; + size_t part; + + if (!stream->avail) { + _leave("no write"); + return len; + } + + _enter("R=%x[%x]", wreq->debug_id, subreq ? subreq->debug_index : 0); + + if (subreq && start != subreq->start + subreq->len) { + netfs_issue_write(wreq, stream); + subreq = NULL; + } + + if (!stream->construct) + netfs_prepare_write(wreq, stream, start); + subreq = stream->construct; + + part = min(subreq->max_len - subreq->len, len); + _debug("part %zx/%zx %zx/%zx", subreq->len, subreq->max_len, part, len); + subreq->len += part; + subreq->nr_segs++; + + if (subreq->len >= subreq->max_len || + subreq->nr_segs >= subreq->max_nr_segs || + to_eof) { + netfs_issue_write(wreq, stream); + subreq = NULL; + } + + return part; +} + +/* + * Write some of a pending folio data back to the server. + */ +static int netfs_write_folio(struct netfs_io_request *wreq, + struct writeback_control *wbc, + struct folio *folio) +{ + struct netfs_io_stream *upload = &wreq->io_streams[0]; + struct netfs_io_stream *cache = &wreq->io_streams[1]; + struct netfs_io_stream *stream; + struct netfs_group *fgroup; /* TODO: Use this with ceph */ + struct netfs_folio *finfo; + size_t fsize = folio_size(folio), flen = fsize, foff = 0; + loff_t fpos = folio_pos(folio), i_size; + bool to_eof = false, streamw = false; + bool debug = false; + + _enter(""); + + /* netfs_perform_write() may shift i_size around the page or from out + * of the page to beyond it, but cannot move i_size into or through the + * page since we have it locked. + */ + i_size = i_size_read(wreq->inode); + + if (fpos >= i_size) { + /* mmap beyond eof. */ + _debug("beyond eof"); + folio_start_writeback(folio); + folio_unlock(folio); + wreq->nr_group_rel += netfs_folio_written_back(folio); + netfs_put_group_many(wreq->group, wreq->nr_group_rel); + wreq->nr_group_rel = 0; + return 0; + } + + if (fpos + fsize > wreq->i_size) + wreq->i_size = i_size; + + fgroup = netfs_folio_group(folio); + finfo = netfs_folio_info(folio); + if (finfo) { + foff = finfo->dirty_offset; + flen = foff + finfo->dirty_len; + streamw = true; + } + + if (wreq->origin == NETFS_WRITETHROUGH) { + to_eof = false; + if (flen > i_size - fpos) + flen = i_size - fpos; + } else if (flen > i_size - fpos) { + flen = i_size - fpos; + if (!streamw) + folio_zero_segment(folio, flen, fsize); + to_eof = true; + } else if (flen == i_size - fpos) { + to_eof = true; + } + flen -= foff; + + _debug("folio %zx %zx %zx", foff, flen, fsize); + + /* Deal with discontinuities in the stream of dirty pages. These can + * arise from a number of sources: + * + * (1) Intervening non-dirty pages from random-access writes, multiple + * flushers writing back different parts simultaneously and manual + * syncing. + * + * (2) Partially-written pages from write-streaming. + * + * (3) Pages that belong to a different write-back group (eg. Ceph + * snapshots). + * + * (4) Actually-clean pages that were marked for write to the cache + * when they were read. Note that these appear as a special + * write-back group. + */ + if (fgroup == NETFS_FOLIO_COPY_TO_CACHE) { + netfs_issue_write(wreq, upload); + } else if (fgroup != wreq->group) { + /* We can't write this page to the server yet. */ + kdebug("wrong group"); + folio_redirty_for_writepage(wbc, folio); + folio_unlock(folio); + netfs_issue_write(wreq, upload); + netfs_issue_write(wreq, cache); + return 0; + } + + if (foff > 0) + netfs_issue_write(wreq, upload); + if (streamw) + netfs_issue_write(wreq, cache); + + /* Flip the page to the writeback state and unlock. If we're called + * from write-through, then the page has already been put into the wb + * state. + */ + if (wreq->origin == NETFS_WRITEBACK) + folio_start_writeback(folio); + folio_unlock(folio); + + if (fgroup == NETFS_FOLIO_COPY_TO_CACHE) { + if (!fscache_resources_valid(&wreq->cache_resources)) { + trace_netfs_folio(folio, netfs_folio_trace_cancel_copy); + netfs_issue_write(wreq, upload); + netfs_folio_written_back(folio); + return 0; + } + trace_netfs_folio(folio, netfs_folio_trace_store_copy); + } else if (!upload->construct) { + trace_netfs_folio(folio, netfs_folio_trace_store); + } else { + trace_netfs_folio(folio, netfs_folio_trace_store_plus); + } + + /* Move the submission point forward to allow for write-streaming data + * not starting at the front of the page. We don't do write-streaming + * with the cache as the cache requires DIO alignment. + * + * Also skip uploading for data that's been read and just needs copying + * to the cache. + */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + stream->submit_max_len = fsize; + stream->submit_off = foff; + stream->submit_len = flen; + if ((stream->source == NETFS_WRITE_TO_CACHE && streamw) || + (stream->source == NETFS_UPLOAD_TO_SERVER && + fgroup == NETFS_FOLIO_COPY_TO_CACHE)) { + stream->submit_off = UINT_MAX; + stream->submit_len = 0; + stream->submit_max_len = 0; + } + } + + /* Attach the folio to one or more subrequests. For a big folio, we + * could end up with thousands of subrequests if the wsize is small - + * but we might need to wait during the creation of subrequests for + * network resources (eg. SMB credits). + */ + for (;;) { + ssize_t part; + size_t lowest_off = ULONG_MAX; + int choose_s = -1; + + /* Always add to the lowest-submitted stream first. */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->submit_len > 0 && + stream->submit_off < lowest_off) { + lowest_off = stream->submit_off; + choose_s = s; + } + } + + if (choose_s < 0) + break; + stream = &wreq->io_streams[choose_s]; + + part = netfs_advance_write(wreq, stream, fpos + stream->submit_off, + stream->submit_len, to_eof); + atomic64_set(&wreq->issued_to, fpos + stream->submit_off); + stream->submit_off += part; + stream->submit_max_len -= part; + if (part > stream->submit_len) + stream->submit_len = 0; + else + stream->submit_len -= part; + if (part > 0) + debug = true; + } + + atomic64_set(&wreq->issued_to, fpos + fsize); + + if (!debug) + kdebug("R=%x: No submit", wreq->debug_id); + + if (flen < fsize) + for (int s = 0; s < NR_IO_STREAMS; s++) + netfs_issue_write(wreq, &wreq->io_streams[s]); + + _leave(" = 0"); + return 0; +} + +/* + * Write some of the pending data back to the server + */ +int new_netfs_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct netfs_inode *ictx = netfs_inode(mapping->host); + struct netfs_io_request *wreq = NULL; + struct folio *folio; + int error = 0; + + if (wbc->sync_mode == WB_SYNC_ALL) + mutex_lock(&ictx->wb_lock); + else if (!mutex_trylock(&ictx->wb_lock)) + return 0; + + /* Need the first folio to be able to set up the op. */ + folio = writeback_iter(mapping, wbc, NULL, &error); + if (!folio) + goto out; + + wreq = netfs_create_write_req(mapping, NULL, folio_pos(folio), NETFS_WRITEBACK); + if (IS_ERR(wreq)) { + error = PTR_ERR(wreq); + goto couldnt_start; + } + + trace_netfs_write(wreq, netfs_write_trace_writeback); + + do { + _debug("wbiter %lx %llx", folio->index, wreq->start + wreq->submitted); + + /* It appears we don't have to handle cyclic writeback wrapping. */ + WARN_ON_ONCE(wreq && folio_pos(folio) < wreq->start + wreq->submitted); + + if (netfs_folio_group(folio) != NETFS_FOLIO_COPY_TO_CACHE && + unlikely(!test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags))) { + set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); + wreq->netfs_ops->begin_writeback(wreq); + } + + error = netfs_write_folio(wreq, wbc, folio); + if (error < 0) + break; + } while ((folio = writeback_iter(mapping, wbc, folio, &error))); + + for (int s = 0; s < NR_IO_STREAMS; s++) + netfs_issue_write(wreq, &wreq->io_streams[s]); + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags); + + mutex_unlock(&ictx->wb_lock); + + netfs_put_request(wreq, false, netfs_rreq_trace_put_return); + _leave(" = %d", error); + return error; + +couldnt_start: + netfs_kill_dirty_pages(mapping, wbc, folio); +out: + mutex_unlock(&ictx->wb_lock); + _leave(" = %d", error); + return error; +} +EXPORT_SYMBOL(new_netfs_writepages); + +/* + * Begin a write operation for writing through the pagecache. + */ +struct netfs_io_request *new_netfs_begin_writethrough(struct kiocb *iocb, size_t len) +{ + struct netfs_io_request *wreq = NULL; + struct netfs_inode *ictx = netfs_inode(file_inode(iocb->ki_filp)); + + mutex_lock(&ictx->wb_lock); + + wreq = netfs_create_write_req(iocb->ki_filp->f_mapping, iocb->ki_filp, + iocb->ki_pos, NETFS_WRITETHROUGH); + if (IS_ERR(wreq)) { + mutex_unlock(&ictx->wb_lock); + return wreq; + } + + wreq->io_streams[0].avail = true; + trace_netfs_write(wreq, netfs_write_trace_writethrough); + return wreq; +} + +/* + * Advance the state of the write operation used when writing through the + * pagecache. Data has been copied into the pagecache that we need to append + * to the request. If we've added more than wsize then we need to create a new + * subrequest. + */ +int new_netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *folio, size_t copied, bool to_page_end, + struct folio **writethrough_cache) +{ + _enter("R=%x ic=%zu ws=%u cp=%zu tp=%u", + wreq->debug_id, wreq->iter.count, wreq->wsize, copied, to_page_end); + + if (!*writethrough_cache) { + if (folio_test_dirty(folio)) + /* Sigh. mmap. */ + folio_clear_dirty_for_io(folio); + + /* We can make multiple writes to the folio... */ + folio_start_writeback(folio); + if (wreq->len == 0) + trace_netfs_folio(folio, netfs_folio_trace_wthru); + else + trace_netfs_folio(folio, netfs_folio_trace_wthru_plus); + *writethrough_cache = folio; + } + + wreq->len += copied; + if (!to_page_end) + return 0; + + *writethrough_cache = NULL; + return netfs_write_folio(wreq, wbc, folio); +} + +/* + * End a write operation used when writing through the pagecache. + */ +int new_netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *writethrough_cache) +{ + struct netfs_inode *ictx = netfs_inode(wreq->inode); + int ret; + + _enter("R=%x", wreq->debug_id); + + if (writethrough_cache) + netfs_write_folio(wreq, wbc, writethrough_cache); + + netfs_issue_write(wreq, &wreq->io_streams[0]); + netfs_issue_write(wreq, &wreq->io_streams[1]); + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags); + + mutex_unlock(&ictx->wb_lock); + + ret = wreq->error; + netfs_put_request(wreq, false, netfs_rreq_trace_put_return); + return ret; +} + +/* + * Write data to the server without going through the pagecache and without + * writing it to the local cache. + */ +int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len) +{ + struct netfs_io_stream *upload = &wreq->io_streams[0]; + ssize_t part; + loff_t start = wreq->start; + int error = 0; + + _enter("%zx", len); + + if (wreq->origin == NETFS_DIO_WRITE) + inode_dio_begin(wreq->inode); + + while (len) { + // TODO: Prepare content encryption + + _debug("unbuffered %zx", len); + part = netfs_advance_write(wreq, upload, start, len, false); + start += part; + len -= part; + if (test_bit(NETFS_RREQ_PAUSE, &wreq->flags)) { + trace_netfs_rreq(wreq, netfs_rreq_trace_wait_pause); + wait_on_bit(&wreq->flags, NETFS_RREQ_PAUSE, TASK_UNINTERRUPTIBLE); + } + if (test_bit(NETFS_RREQ_FAILED, &wreq->flags)) + break; + } + + netfs_issue_write(wreq, upload); + + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags); + if (list_empty(&upload->subrequests)) + netfs_wake_write_collector(wreq, false); + + _leave(" = %d", error); + return error; +} diff --git a/include/linux/netfs.h b/include/linux/netfs.h index 88269681d4fc..42dba05a428b 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -64,6 +64,7 @@ struct netfs_inode { #if IS_ENABLED(CONFIG_FSCACHE) struct fscache_cookie *cache; #endif + struct mutex wb_lock; /* Writeback serialisation */ loff_t remote_i_size; /* Size of the remote file */ loff_t zero_point; /* Size after which we assume there's no data * on the server */ @@ -71,7 +72,6 @@ struct netfs_inode { #define NETFS_ICTX_ODIRECT 0 /* The file has DIO in progress */ #define NETFS_ICTX_UNBUFFERED 1 /* I/O should not use the pagecache */ #define NETFS_ICTX_WRITETHROUGH 2 /* Write-through caching */ -#define NETFS_ICTX_NO_WRITE_STREAMING 3 /* Don't engage in write-streaming */ #define NETFS_ICTX_USE_PGPRIV2 31 /* [DEPRECATED] Use PG_private_2 to mark * write to cache on read */ }; @@ -126,6 +126,33 @@ static inline struct netfs_group *netfs_folio_group(struct folio *folio) return priv; } +/* + * Stream of I/O subrequests going to a particular destination, such as the + * server or the local cache. This is mainly intended for writing where we may + * have to write to multiple destinations concurrently. + */ +struct netfs_io_stream { + /* Submission tracking */ + struct netfs_io_subrequest *construct; /* Op being constructed */ + unsigned int submit_off; /* Folio offset we're submitting from */ + unsigned int submit_len; /* Amount of data left to submit */ + unsigned int submit_max_len; /* Amount I/O can be rounded up to */ + void (*prepare_write)(struct netfs_io_subrequest *subreq); + void (*issue_write)(struct netfs_io_subrequest *subreq); + /* Collection tracking */ + struct list_head subrequests; /* Contributory I/O operations */ + struct netfs_io_subrequest *front; /* Op being collected */ + unsigned long long collected_to; /* Position we've collected results to */ + size_t transferred; /* The amount transferred from this stream */ + enum netfs_io_source source; /* Where to read from/write to */ + unsigned short error; /* Aggregate error for the stream */ + unsigned char stream_nr; /* Index of stream in parent table */ + bool avail; /* T if stream is available */ + bool active; /* T if stream is active */ + bool need_retry; /* T if this stream needs retrying */ + bool failed; /* T if this stream failed */ +}; + /* * Resources required to do operations on a cache. */ @@ -150,13 +177,16 @@ struct netfs_io_subrequest { struct list_head rreq_link; /* Link in rreq->subrequests */ struct iov_iter io_iter; /* Iterator for this subrequest */ unsigned long long start; /* Where to start the I/O */ + size_t max_len; /* Maximum size of the I/O */ size_t len; /* Size of the I/O */ size_t transferred; /* Amount of data transferred */ refcount_t ref; short error; /* 0 or error that occurred */ unsigned short debug_index; /* Index in list (for debugging output) */ + unsigned int nr_segs; /* Number of segs in io_iter */ unsigned int max_nr_segs; /* 0 or max number of segments in an iterator */ enum netfs_io_source source; /* Where to read from/write to */ + unsigned char stream_nr; /* I/O stream this belongs to */ unsigned long flags; #define NETFS_SREQ_COPY_TO_CACHE 0 /* Set if should copy the data to the cache */ #define NETFS_SREQ_CLEAR_TAIL 1 /* Set if the rest of the read should be cleared */ @@ -164,6 +194,11 @@ struct netfs_io_subrequest { #define NETFS_SREQ_SEEK_DATA_READ 3 /* Set if ->read() should SEEK_DATA first */ #define NETFS_SREQ_NO_PROGRESS 4 /* Set if we didn't manage to read any data */ #define NETFS_SREQ_ONDEMAND 5 /* Set if it's from on-demand read mode */ +#define NETFS_SREQ_BOUNDARY 6 /* Set if ends on hard boundary (eg. ceph object) */ +#define NETFS_SREQ_IN_PROGRESS 8 /* Unlocked when the subrequest completes */ +#define NETFS_SREQ_NEED_RETRY 9 /* Set if the filesystem requests a retry */ +#define NETFS_SREQ_RETRYING 10 /* Set if we're retrying */ +#define NETFS_SREQ_FAILED 11 /* Set if the subreq failed unretryably */ }; enum netfs_io_origin { @@ -194,6 +229,9 @@ struct netfs_io_request { struct netfs_cache_resources cache_resources; struct list_head proc_link; /* Link in netfs_iorequests */ struct list_head subrequests; /* Contributory I/O operations */ + struct netfs_io_stream io_streams[2]; /* Streams of parallel I/O operations */ +#define NR_IO_STREAMS 2 //wreq->nr_io_streams + struct netfs_group *group; /* Writeback group being written back */ struct iov_iter iter; /* Unencrypted-side iterator */ struct iov_iter io_iter; /* I/O (Encrypted-side) iterator */ void *netfs_priv; /* Private data for the netfs */ @@ -203,6 +241,8 @@ struct netfs_io_request { unsigned int rsize; /* Maximum read size (0 for none) */ unsigned int wsize; /* Maximum write size (0 for none) */ atomic_t subreq_counter; /* Next subreq->debug_index */ + unsigned int nr_group_rel; /* Number of refs to release on ->group */ + spinlock_t lock; /* Lock for queuing subreqs */ atomic_t nr_outstanding; /* Number of ops in progress */ atomic_t nr_copy_ops; /* Number of copy-to-cache ops in progress */ size_t upper_len; /* Length can be extended to here */ @@ -214,6 +254,10 @@ struct netfs_io_request { bool direct_bv_unpin; /* T if direct_bv[] must be unpinned */ unsigned long long i_size; /* Size of the file */ unsigned long long start; /* Start position */ + atomic64_t issued_to; /* Write issuer folio cursor */ + unsigned long long contiguity; /* Tracking for gaps in the writeback sequence */ + unsigned long long collected_to; /* Point we've collected to */ + unsigned long long cleaned_to; /* Position we've cleaned folios to */ pgoff_t no_unlock_folio; /* Don't unlock this folio after read */ refcount_t ref; unsigned long flags; @@ -227,6 +271,9 @@ struct netfs_io_request { #define NETFS_RREQ_UPLOAD_TO_SERVER 8 /* Need to write to the server */ #define NETFS_RREQ_NONBLOCK 9 /* Don't block if possible (O_NONBLOCK) */ #define NETFS_RREQ_BLOCKED 10 /* We blocked */ +#define NETFS_RREQ_PAUSE 11 /* Pause subrequest generation */ +#define NETFS_RREQ_USE_IO_ITER 12 /* Use ->io_iter rather than ->i_pages */ +#define NETFS_RREQ_ALL_QUEUED 13 /* All subreqs are now queued */ #define NETFS_RREQ_USE_PGPRIV2 31 /* [DEPRECATED] Use PG_private_2 to mark * write to cache on read */ const struct netfs_request_ops *netfs_ops; @@ -258,6 +305,9 @@ struct netfs_request_ops { /* Write request handling */ void (*create_write_requests)(struct netfs_io_request *wreq, loff_t start, size_t len); + void (*begin_writeback)(struct netfs_io_request *wreq); + void (*prepare_write)(struct netfs_io_subrequest *subreq); + void (*issue_write)(struct netfs_io_subrequest *subreq); void (*invalidate_cache)(struct netfs_io_request *wreq); }; @@ -292,6 +342,9 @@ struct netfs_cache_ops { netfs_io_terminated_t term_func, void *term_func_priv); + /* Write data to the cache from a netfs subrequest. */ + void (*issue_write)(struct netfs_io_subrequest *subreq); + /* Expand readahead request */ void (*expand_readahead)(struct netfs_cache_resources *cres, unsigned long long *_start, @@ -304,6 +357,13 @@ struct netfs_cache_ops { enum netfs_io_source (*prepare_read)(struct netfs_io_subrequest *subreq, unsigned long long i_size); + /* Prepare a write subrequest, working out if we're allowed to do it + * and finding out the maximum amount of data to gather before + * attempting to submit. If we're not permitted to do it, the + * subrequest should be marked failed. + */ + void (*prepare_write_subreq)(struct netfs_io_subrequest *subreq); + /* Prepare a write operation, working out what part of the write we can * actually do. */ @@ -349,6 +409,8 @@ int netfs_write_begin(struct netfs_inode *, struct file *, struct folio **, void **fsdata); int netfs_writepages(struct address_space *mapping, struct writeback_control *wbc); +int new_netfs_writepages(struct address_space *mapping, + struct writeback_control *wbc); bool netfs_dirty_folio(struct address_space *mapping, struct folio *folio); int netfs_unpin_writeback(struct inode *inode, struct writeback_control *wbc); void netfs_clear_inode_writeback(struct inode *inode, const void *aux); @@ -372,8 +434,11 @@ size_t netfs_limit_iter(const struct iov_iter *iter, size_t start_offset, struct netfs_io_subrequest *netfs_create_write_request( struct netfs_io_request *wreq, enum netfs_io_source dest, loff_t start, size_t len, work_func_t worker); +void netfs_prepare_write_failed(struct netfs_io_subrequest *subreq); void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, bool was_async); +void new_netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, + bool was_async); void netfs_queue_write_request(struct netfs_io_subrequest *subreq); int netfs_start_io_read(struct inode *inode); @@ -415,6 +480,7 @@ static inline void netfs_inode_init(struct netfs_inode *ctx, #if IS_ENABLED(CONFIG_FSCACHE) ctx->cache = NULL; #endif + mutex_init(&ctx->wb_lock); /* ->releasepage() drives zero_point */ if (use_zero_point) { ctx->zero_point = ctx->remote_i_size; diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h index 7126d2ea459c..e7700172ae7e 100644 --- a/include/trace/events/netfs.h +++ b/include/trace/events/netfs.h @@ -44,14 +44,18 @@ #define netfs_rreq_traces \ EM(netfs_rreq_trace_assess, "ASSESS ") \ EM(netfs_rreq_trace_copy, "COPY ") \ + EM(netfs_rreq_trace_collect, "COLLECT") \ EM(netfs_rreq_trace_done, "DONE ") \ EM(netfs_rreq_trace_free, "FREE ") \ EM(netfs_rreq_trace_redirty, "REDIRTY") \ EM(netfs_rreq_trace_resubmit, "RESUBMT") \ + EM(netfs_rreq_trace_set_pause, "PAUSE ") \ EM(netfs_rreq_trace_unlock, "UNLOCK ") \ EM(netfs_rreq_trace_unmark, "UNMARK ") \ EM(netfs_rreq_trace_wait_ip, "WAIT-IP") \ + EM(netfs_rreq_trace_wait_pause, "WT-PAUS") \ EM(netfs_rreq_trace_wake_ip, "WAKE-IP") \ + EM(netfs_rreq_trace_unpause, "UNPAUSE") \ E_(netfs_rreq_trace_write_done, "WR-DONE") #define netfs_sreq_sources \ @@ -64,11 +68,15 @@ E_(NETFS_INVALID_WRITE, "INVL") #define netfs_sreq_traces \ + EM(netfs_sreq_trace_discard, "DSCRD") \ EM(netfs_sreq_trace_download_instead, "RDOWN") \ + EM(netfs_sreq_trace_fail, "FAIL ") \ EM(netfs_sreq_trace_free, "FREE ") \ EM(netfs_sreq_trace_limited, "LIMIT") \ EM(netfs_sreq_trace_prepare, "PREP ") \ + EM(netfs_sreq_trace_prep_failed, "PRPFL") \ EM(netfs_sreq_trace_resubmit_short, "SHORT") \ + EM(netfs_sreq_trace_retry, "RETRY") \ EM(netfs_sreq_trace_submit, "SUBMT") \ EM(netfs_sreq_trace_terminated, "TERM ") \ EM(netfs_sreq_trace_write, "WRITE") \ @@ -88,6 +96,7 @@ #define netfs_rreq_ref_traces \ EM(netfs_rreq_trace_get_for_outstanding,"GET OUTSTND") \ EM(netfs_rreq_trace_get_subreq, "GET SUBREQ ") \ + EM(netfs_rreq_trace_get_work, "GET WORK ") \ EM(netfs_rreq_trace_put_complete, "PUT COMPLT ") \ EM(netfs_rreq_trace_put_discard, "PUT DISCARD") \ EM(netfs_rreq_trace_put_failed, "PUT FAILED ") \ @@ -95,6 +104,8 @@ EM(netfs_rreq_trace_put_return, "PUT RETURN ") \ EM(netfs_rreq_trace_put_subreq, "PUT SUBREQ ") \ EM(netfs_rreq_trace_put_work, "PUT WORK ") \ + EM(netfs_rreq_trace_put_work_complete, "PUT WORK CP") \ + EM(netfs_rreq_trace_put_work_nq, "PUT WORK NQ") \ EM(netfs_rreq_trace_see_work, "SEE WORK ") \ E_(netfs_rreq_trace_new, "NEW ") @@ -103,11 +114,14 @@ EM(netfs_sreq_trace_get_resubmit, "GET RESUBMIT") \ EM(netfs_sreq_trace_get_short_read, "GET SHORTRD") \ EM(netfs_sreq_trace_new, "NEW ") \ + EM(netfs_sreq_trace_put_cancel, "PUT CANCEL ") \ EM(netfs_sreq_trace_put_clear, "PUT CLEAR ") \ EM(netfs_sreq_trace_put_discard, "PUT DISCARD") \ + EM(netfs_sreq_trace_put_done, "PUT DONE ") \ EM(netfs_sreq_trace_put_failed, "PUT FAILED ") \ EM(netfs_sreq_trace_put_merged, "PUT MERGED ") \ EM(netfs_sreq_trace_put_no_copy, "PUT NO COPY") \ + EM(netfs_sreq_trace_put_oom, "PUT OOM ") \ EM(netfs_sreq_trace_put_wip, "PUT WIP ") \ EM(netfs_sreq_trace_put_work, "PUT WORK ") \ E_(netfs_sreq_trace_put_terminated, "PUT TERM ") @@ -124,7 +138,9 @@ EM(netfs_streaming_filled_page, "mod-streamw-f") \ EM(netfs_streaming_cont_filled_page, "mod-streamw-f+") \ /* The rest are for writeback */ \ + EM(netfs_folio_trace_cancel_copy, "cancel-copy") \ EM(netfs_folio_trace_clear, "clear") \ + EM(netfs_folio_trace_clear_cc, "clear-cc") \ EM(netfs_folio_trace_clear_s, "clear-s") \ EM(netfs_folio_trace_clear_g, "clear-g") \ EM(netfs_folio_trace_copy, "copy") \ @@ -133,16 +149,26 @@ EM(netfs_folio_trace_end_copy, "end-copy") \ EM(netfs_folio_trace_filled_gaps, "filled-gaps") \ EM(netfs_folio_trace_kill, "kill") \ + EM(netfs_folio_trace_kill_cc, "kill-cc") \ + EM(netfs_folio_trace_kill_g, "kill-g") \ + EM(netfs_folio_trace_kill_s, "kill-s") \ EM(netfs_folio_trace_mkwrite, "mkwrite") \ EM(netfs_folio_trace_mkwrite_plus, "mkwrite+") \ + EM(netfs_folio_trace_not_under_wback, "!wback") \ EM(netfs_folio_trace_read_gaps, "read-gaps") \ EM(netfs_folio_trace_redirty, "redirty") \ EM(netfs_folio_trace_redirtied, "redirtied") \ EM(netfs_folio_trace_store, "store") \ + EM(netfs_folio_trace_store_copy, "store-copy") \ EM(netfs_folio_trace_store_plus, "store+") \ EM(netfs_folio_trace_wthru, "wthru") \ E_(netfs_folio_trace_wthru_plus, "wthru+") +#define netfs_collect_contig_traces \ + EM(netfs_contig_trace_collect, "Collect") \ + EM(netfs_contig_trace_jump, "-->JUMP-->") \ + E_(netfs_contig_trace_unlock, "Unlock") + #ifndef __NETFS_DECLARE_TRACE_ENUMS_ONCE_ONLY #define __NETFS_DECLARE_TRACE_ENUMS_ONCE_ONLY @@ -159,6 +185,7 @@ enum netfs_failure { netfs_failures } __mode(byte); enum netfs_rreq_ref_trace { netfs_rreq_ref_traces } __mode(byte); enum netfs_sreq_ref_trace { netfs_sreq_ref_traces } __mode(byte); enum netfs_folio_trace { netfs_folio_traces } __mode(byte); +enum netfs_collect_contig_trace { netfs_collect_contig_traces } __mode(byte); #endif @@ -180,6 +207,7 @@ netfs_failures; netfs_rreq_ref_traces; netfs_sreq_ref_traces; netfs_folio_traces; +netfs_collect_contig_traces; /* * Now redefine the EM() and E_() macros to map the enums to the strings that @@ -413,16 +441,18 @@ TRACE_EVENT(netfs_write_iter, __field(unsigned long long, start ) __field(size_t, len ) __field(unsigned int, flags ) + __field(unsigned int, ino ) ), TP_fast_assign( __entry->start = iocb->ki_pos; __entry->len = iov_iter_count(from); + __entry->ino = iocb->ki_filp->f_inode->i_ino; __entry->flags = iocb->ki_flags; ), - TP_printk("WRITE-ITER s=%llx l=%zx f=%x", - __entry->start, __entry->len, __entry->flags) + TP_printk("WRITE-ITER i=%x s=%llx l=%zx f=%x", + __entry->ino, __entry->start, __entry->len, __entry->flags) ); TRACE_EVENT(netfs_write, @@ -434,6 +464,7 @@ TRACE_EVENT(netfs_write, TP_STRUCT__entry( __field(unsigned int, wreq ) __field(unsigned int, cookie ) + __field(unsigned int, ino ) __field(enum netfs_write_trace, what ) __field(unsigned long long, start ) __field(unsigned long long, len ) @@ -444,18 +475,213 @@ TRACE_EVENT(netfs_write, struct fscache_cookie *__cookie = netfs_i_cookie(__ctx); __entry->wreq = wreq->debug_id; __entry->cookie = __cookie ? __cookie->debug_id : 0; + __entry->ino = wreq->inode->i_ino; __entry->what = what; __entry->start = wreq->start; __entry->len = wreq->len; ), - TP_printk("R=%08x %s c=%08x by=%llx-%llx", + TP_printk("R=%08x %s c=%08x i=%x by=%llx-%llx", __entry->wreq, __print_symbolic(__entry->what, netfs_write_traces), __entry->cookie, + __entry->ino, __entry->start, __entry->start + __entry->len - 1) ); +TRACE_EVENT(netfs_collect, + TP_PROTO(const struct netfs_io_request *wreq), + + TP_ARGS(wreq), + + TP_STRUCT__entry( + __field(unsigned int, wreq ) + __field(unsigned int, len ) + __field(unsigned long long, transferred ) + __field(unsigned long long, start ) + ), + + TP_fast_assign( + __entry->wreq = wreq->debug_id; + __entry->start = wreq->start; + __entry->len = wreq->len; + __entry->transferred = wreq->transferred; + ), + + TP_printk("R=%08x s=%llx-%llx", + __entry->wreq, + __entry->start + __entry->transferred, + __entry->start + __entry->len) + ); + +TRACE_EVENT(netfs_collect_contig, + TP_PROTO(const struct netfs_io_request *wreq, unsigned long long to, + enum netfs_collect_contig_trace type), + + TP_ARGS(wreq, to, type), + + TP_STRUCT__entry( + __field(unsigned int, wreq) + __field(enum netfs_collect_contig_trace, type) + __field(unsigned long long, contiguity) + __field(unsigned long long, to) + ), + + TP_fast_assign( + __entry->wreq = wreq->debug_id; + __entry->type = type; + __entry->contiguity = wreq->contiguity; + __entry->to = to; + ), + + TP_printk("R=%08x %llx -> %llx %s", + __entry->wreq, + __entry->contiguity, + __entry->to, + __print_symbolic(__entry->type, netfs_collect_contig_traces)) + ); + +TRACE_EVENT(netfs_collect_sreq, + TP_PROTO(const struct netfs_io_request *wreq, + const struct netfs_io_subrequest *subreq), + + TP_ARGS(wreq, subreq), + + TP_STRUCT__entry( + __field(unsigned int, wreq ) + __field(unsigned int, subreq ) + __field(unsigned int, stream ) + __field(unsigned int, len ) + __field(unsigned int, transferred ) + __field(unsigned long long, start ) + ), + + TP_fast_assign( + __entry->wreq = wreq->debug_id; + __entry->subreq = subreq->debug_index; + __entry->stream = subreq->stream_nr; + __entry->start = subreq->start; + __entry->len = subreq->len; + __entry->transferred = subreq->transferred; + ), + + TP_printk("R=%08x[%u:%02x] s=%llx t=%x/%x", + __entry->wreq, __entry->stream, __entry->subreq, + __entry->start, __entry->transferred, __entry->len) + ); + +TRACE_EVENT(netfs_collect_folio, + TP_PROTO(const struct netfs_io_request *wreq, + const struct folio *folio, + unsigned long long fend, + unsigned long long collected_to), + + TP_ARGS(wreq, folio, fend, collected_to), + + TP_STRUCT__entry( + __field(unsigned int, wreq ) + __field(unsigned long, index ) + __field(unsigned long long, fend ) + __field(unsigned long long, cleaned_to ) + __field(unsigned long long, collected_to ) + ), + + TP_fast_assign( + __entry->wreq = wreq->debug_id; + __entry->index = folio->index; + __entry->fend = fend; + __entry->cleaned_to = wreq->cleaned_to; + __entry->collected_to = collected_to; + ), + + TP_printk("R=%08x ix=%05lx r=%llx-%llx t=%llx/%llx", + __entry->wreq, __entry->index, + (unsigned long long)__entry->index * PAGE_SIZE, __entry->fend, + __entry->cleaned_to, __entry->collected_to) + ); + +TRACE_EVENT(netfs_collect_state, + TP_PROTO(const struct netfs_io_request *wreq, + unsigned long long collected_to, + unsigned int notes), + + TP_ARGS(wreq, collected_to, notes), + + TP_STRUCT__entry( + __field(unsigned int, wreq ) + __field(unsigned int, notes ) + __field(unsigned long long, collected_to ) + __field(unsigned long long, cleaned_to ) + __field(unsigned long long, contiguity ) + ), + + TP_fast_assign( + __entry->wreq = wreq->debug_id; + __entry->notes = notes; + __entry->collected_to = collected_to; + __entry->cleaned_to = wreq->cleaned_to; + __entry->contiguity = wreq->contiguity; + ), + + TP_printk("R=%08x cto=%llx fto=%llx ctg=%llx n=%x", + __entry->wreq, __entry->collected_to, + __entry->cleaned_to, __entry->contiguity, + __entry->notes) + ); + +TRACE_EVENT(netfs_collect_gap, + TP_PROTO(const struct netfs_io_request *wreq, + const struct netfs_io_stream *stream, + unsigned long long jump_to, char type), + + TP_ARGS(wreq, stream, jump_to, type), + + TP_STRUCT__entry( + __field(unsigned int, wreq) + __field(unsigned char, stream) + __field(unsigned char, type) + __field(unsigned long long, from) + __field(unsigned long long, to) + ), + + TP_fast_assign( + __entry->wreq = wreq->debug_id; + __entry->stream = stream->stream_nr; + __entry->from = stream->collected_to; + __entry->to = jump_to; + __entry->type = type; + ), + + TP_printk("R=%08x[%x:] %llx->%llx %c", + __entry->wreq, __entry->stream, + __entry->from, __entry->to, __entry->type) + ); + +TRACE_EVENT(netfs_collect_stream, + TP_PROTO(const struct netfs_io_request *wreq, + const struct netfs_io_stream *stream), + + TP_ARGS(wreq, stream), + + TP_STRUCT__entry( + __field(unsigned int, wreq) + __field(unsigned char, stream) + __field(unsigned long long, collected_to) + __field(unsigned long long, front) + ), + + TP_fast_assign( + __entry->wreq = wreq->debug_id; + __entry->stream = stream->stream_nr; + __entry->collected_to = stream->collected_to; + __entry->front = stream->front ? stream->front->start : UINT_MAX; + ), + + TP_printk("R=%08x[%x:] cto=%llx frn=%llx", + __entry->wreq, __entry->stream, + __entry->collected_to, __entry->front) + ); + #undef EM #undef E_ #endif /* _TRACE_NETFS_H */ -- cgit v1.2.3-59-g8ed1b From 4824e5917f907c163d001c753c050bc547a72b71 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 26 Mar 2024 08:48:44 +0000 Subject: netfs: Add some write-side stats and clean up some stat names Add some write-side stats to count buffered writes, buffered writethrough, and writepages calls. Whilst we're at it, clean up the naming on some of the existing stats counters and organise the output into two sets. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/netfs/buffered_read.c | 2 +- fs/netfs/buffered_write.c | 3 +++ fs/netfs/direct_write.c | 2 +- fs/netfs/internal.h | 7 +++++-- fs/netfs/stats.c | 17 ++++++++++++----- fs/netfs/write_issue.c | 1 + 6 files changed, 23 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index 47603f08680e..a6bb03bea920 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -282,7 +282,7 @@ int netfs_read_folio(struct file *file, struct folio *folio) if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS) goto discard; - netfs_stat(&netfs_n_rh_readpage); + netfs_stat(&netfs_n_rh_read_folio); trace_netfs_read(rreq, rreq->start, rreq->len, netfs_read_trace_readpage); /* Set up the output buffer */ diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 84ac95ee4b4d..33ea4c20e7e7 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -210,6 +210,9 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, if (!is_sync_kiocb(iocb)) wreq->iocb = iocb; wreq->cleanup = netfs_cleanup_buffered_write; + netfs_stat(&netfs_n_wh_writethrough); + } else { + netfs_stat(&netfs_n_wh_buffered_write); } do { diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c index 2b81cd4aae6e..36b6db504500 100644 --- a/fs/netfs/direct_write.c +++ b/fs/netfs/direct_write.c @@ -145,7 +145,7 @@ ssize_t netfs_unbuffered_write_iter(struct kiocb *iocb, struct iov_iter *from) return 0; trace_netfs_write_iter(iocb, from); - netfs_stat(&netfs_n_rh_dio_write); + netfs_stat(&netfs_n_wh_dio_write); ret = netfs_start_io_direct(inode); if (ret < 0) diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index dc11d1f67363..5d3f74a70fa7 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -106,9 +106,8 @@ int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb); */ #ifdef CONFIG_NETFS_STATS extern atomic_t netfs_n_rh_dio_read; -extern atomic_t netfs_n_rh_dio_write; extern atomic_t netfs_n_rh_readahead; -extern atomic_t netfs_n_rh_readpage; +extern atomic_t netfs_n_rh_read_folio; extern atomic_t netfs_n_rh_rreq; extern atomic_t netfs_n_rh_sreq; extern atomic_t netfs_n_rh_download; @@ -125,6 +124,10 @@ extern atomic_t netfs_n_rh_write_begin; extern atomic_t netfs_n_rh_write_done; extern atomic_t netfs_n_rh_write_failed; extern atomic_t netfs_n_rh_write_zskip; +extern atomic_t netfs_n_wh_buffered_write; +extern atomic_t netfs_n_wh_writethrough; +extern atomic_t netfs_n_wh_dio_write; +extern atomic_t netfs_n_wh_writepages; extern atomic_t netfs_n_wh_wstream_conflict; extern atomic_t netfs_n_wh_upload; extern atomic_t netfs_n_wh_upload_done; diff --git a/fs/netfs/stats.c b/fs/netfs/stats.c index deeba9f9dcf5..0892768eea32 100644 --- a/fs/netfs/stats.c +++ b/fs/netfs/stats.c @@ -10,9 +10,8 @@ #include "internal.h" atomic_t netfs_n_rh_dio_read; -atomic_t netfs_n_rh_dio_write; atomic_t netfs_n_rh_readahead; -atomic_t netfs_n_rh_readpage; +atomic_t netfs_n_rh_read_folio; atomic_t netfs_n_rh_rreq; atomic_t netfs_n_rh_sreq; atomic_t netfs_n_rh_download; @@ -29,6 +28,10 @@ atomic_t netfs_n_rh_write_begin; atomic_t netfs_n_rh_write_done; atomic_t netfs_n_rh_write_failed; atomic_t netfs_n_rh_write_zskip; +atomic_t netfs_n_wh_buffered_write; +atomic_t netfs_n_wh_writethrough; +atomic_t netfs_n_wh_dio_write; +atomic_t netfs_n_wh_writepages; atomic_t netfs_n_wh_wstream_conflict; atomic_t netfs_n_wh_upload; atomic_t netfs_n_wh_upload_done; @@ -39,13 +42,17 @@ atomic_t netfs_n_wh_write_failed; int netfs_stats_show(struct seq_file *m, void *v) { - seq_printf(m, "Netfs : DR=%u DW=%u RA=%u RP=%u WB=%u WBZ=%u\n", + seq_printf(m, "Netfs : DR=%u RA=%u RF=%u WB=%u WBZ=%u\n", atomic_read(&netfs_n_rh_dio_read), - atomic_read(&netfs_n_rh_dio_write), atomic_read(&netfs_n_rh_readahead), - atomic_read(&netfs_n_rh_readpage), + atomic_read(&netfs_n_rh_read_folio), atomic_read(&netfs_n_rh_write_begin), atomic_read(&netfs_n_rh_write_zskip)); + seq_printf(m, "Netfs : BW=%u WT=%u DW=%u WP=%u\n", + atomic_read(&netfs_n_wh_buffered_write), + atomic_read(&netfs_n_wh_writethrough), + atomic_read(&netfs_n_wh_dio_write), + atomic_read(&netfs_n_wh_writepages)); seq_printf(m, "Netfs : ZR=%u sh=%u sk=%u\n", atomic_read(&netfs_n_rh_zero), atomic_read(&netfs_n_rh_short_read), diff --git a/fs/netfs/write_issue.c b/fs/netfs/write_issue.c index 1b2e69934fbd..7ea86e33382c 100644 --- a/fs/netfs/write_issue.c +++ b/fs/netfs/write_issue.c @@ -519,6 +519,7 @@ int new_netfs_writepages(struct address_space *mapping, } trace_netfs_write(wreq, netfs_write_trace_writeback); + netfs_stat(&netfs_n_wh_writepages); do { _debug("wbiter %lx %llx", folio->index, wreq->start + wreq->submitted); -- cgit v1.2.3-59-g8ed1b From ed22e1dbf831bbc747a726b7c1f924c18c1ad350 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 20:25:53 +0000 Subject: netfs, afs: Implement helpers for new write code Implement the helpers for the new write code in afs. There's now an optional ->prepare_write() that allows the filesystem to set the parameters for the next write, such as maximum size and maximum segment count, and an ->issue_write() that is called to initiate an (asynchronous) write operation. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Marc Dionne cc: linux-afs@lists.infradead.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/afs/file.c | 3 +++ fs/afs/internal.h | 3 +++ fs/afs/write.c | 54 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 60 insertions(+) (limited to 'fs') diff --git a/fs/afs/file.c b/fs/afs/file.c index dfd8f60f5e1f..db9ebae84fa2 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -400,6 +400,9 @@ const struct netfs_request_ops afs_req_ops = { .update_i_size = afs_update_i_size, .invalidate_cache = afs_netfs_invalidate_cache, .create_write_requests = afs_create_write_requests, + .begin_writeback = afs_begin_writeback, + .prepare_write = afs_prepare_write, + .issue_write = afs_issue_write, }; static void afs_add_open_mmap(struct afs_vnode *vnode) diff --git a/fs/afs/internal.h b/fs/afs/internal.h index b93aa026daa4..dcf0ae0323d3 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -1598,6 +1598,9 @@ extern int afs_check_volume_status(struct afs_volume *, struct afs_operation *); /* * write.c */ +void afs_prepare_write(struct netfs_io_subrequest *subreq); +void afs_issue_write(struct netfs_io_subrequest *subreq); +void afs_begin_writeback(struct netfs_io_request *wreq); extern int afs_writepages(struct address_space *, struct writeback_control *); extern int afs_fsync(struct file *, loff_t, loff_t, int); extern vm_fault_t afs_page_mkwrite(struct vm_fault *vmf); diff --git a/fs/afs/write.c b/fs/afs/write.c index 1bc26466eb72..34595f482718 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -194,6 +194,60 @@ void afs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size netfs_queue_write_request(subreq); } +/* + * Writeback calls this when it finds a folio that needs uploading. This isn't + * called if writeback only has copy-to-cache to deal with. + */ +void afs_begin_writeback(struct netfs_io_request *wreq) +{ + wreq->io_streams[0].avail = true; +} + +/* + * Prepare a subrequest to write to the server. This sets the max_len + * parameter. + */ +void afs_prepare_write(struct netfs_io_subrequest *subreq) +{ + //if (test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) + // subreq->max_len = 512 * 1024; + //else + subreq->max_len = 256 * 1024 * 1024; +} + +/* + * Issue a subrequest to write to the server. + */ +static void afs_issue_write_worker(struct work_struct *work) +{ + struct netfs_io_subrequest *subreq = container_of(work, struct netfs_io_subrequest, work); + struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode); + ssize_t ret; + + _enter("%x[%x],%zx", + subreq->rreq->debug_id, subreq->debug_index, subreq->io_iter.count); + +#if 0 // Error injection + if (subreq->debug_index == 3) + return netfs_write_subrequest_terminated(subreq, -ENOANO, false); + + if (!test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) { + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + return netfs_write_subrequest_terminated(subreq, -EAGAIN, false); + } +#endif + + ret = afs_store_data(vnode, &subreq->io_iter, subreq->start); + netfs_write_subrequest_terminated(subreq, ret < 0 ? ret : subreq->len, false); +} + +void afs_issue_write(struct netfs_io_subrequest *subreq) +{ + subreq->work.func = afs_issue_write_worker; + if (!queue_work(system_unbound_wq, &subreq->work)) + WARN_ON_ONCE(1); +} + /* * write some of the pending data back to the server */ -- cgit v1.2.3-59-g8ed1b From 5fb70e7275a61dd404f684370e1add7fe0ebe9c5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 20:29:53 +0000 Subject: netfs, 9p: Implement helpers for new write code Implement the helpers for the new write code in 9p. There's now an optional ->prepare_write() that allows the filesystem to set the parameters for the next write, such as maximum size and maximum segment count, and an ->issue_write() that is called to initiate an (asynchronous) write operation. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: v9fs@lists.linux.dev cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/9p/vfs_addr.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++ include/net/9p/client.h | 2 ++ net/9p/Kconfig | 1 + net/9p/client.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 100 insertions(+) (limited to 'fs') diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 5a943c122d83..07d03efdd594 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -26,6 +26,40 @@ #include "cache.h" #include "fid.h" +/* + * Writeback calls this when it finds a folio that needs uploading. This isn't + * called if writeback only has copy-to-cache to deal with. + */ +static void v9fs_begin_writeback(struct netfs_io_request *wreq) +{ + struct p9_fid *fid; + + fid = v9fs_fid_find_inode(wreq->inode, true, INVALID_UID, true); + if (!fid) { + WARN_ONCE(1, "folio expected an open fid inode->i_ino=%lx\n", + wreq->inode->i_ino); + return; + } + + wreq->wsize = fid->clnt->msize - P9_IOHDRSZ; + if (fid->iounit) + wreq->wsize = min(wreq->wsize, fid->iounit); + wreq->netfs_priv = fid; + wreq->io_streams[0].avail = true; +} + +/* + * Issue a subrequest to write to the server. + */ +static void v9fs_issue_write(struct netfs_io_subrequest *subreq) +{ + struct p9_fid *fid = subreq->rreq->netfs_priv; + int err, len; + + len = p9_client_write(fid, subreq->start, &subreq->io_iter, &err); + netfs_write_subrequest_terminated(subreq, len ?: err, false); +} + static void v9fs_upload_to_server(struct netfs_io_subrequest *subreq) { struct p9_fid *fid = subreq->rreq->netfs_priv; @@ -92,6 +126,14 @@ static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file) rreq->origin == NETFS_UNBUFFERED_WRITE || rreq->origin == NETFS_DIO_WRITE); +#if 0 // TODO: Cut over + if (rreq->origin == NETFS_WRITEBACK) + return 0; /* We don't get the write handle until we find we + * have actually dirty data and not just + * copy-to-cache data. + */ +#endif + if (file) { fid = file->private_data; if (!fid) @@ -103,6 +145,10 @@ static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file) goto no_fid; } + rreq->wsize = fid->clnt->msize - P9_IOHDRSZ; + if (fid->iounit) + rreq->wsize = min(rreq->wsize, fid->iounit); + /* we might need to read from a fid that was opened write-only * for read-modify-write of page cache, use the writeback fid * for that */ @@ -131,6 +177,8 @@ const struct netfs_request_ops v9fs_req_ops = { .init_request = v9fs_init_request, .free_request = v9fs_free_request, .issue_read = v9fs_issue_read, + .begin_writeback = v9fs_begin_writeback, + .issue_write = v9fs_issue_write, .create_write_requests = v9fs_create_write_requests, }; diff --git a/include/net/9p/client.h b/include/net/9p/client.h index 78ebcf782ce5..4f785098c67a 100644 --- a/include/net/9p/client.h +++ b/include/net/9p/client.h @@ -207,6 +207,8 @@ int p9_client_read(struct p9_fid *fid, u64 offset, struct iov_iter *to, int *err int p9_client_read_once(struct p9_fid *fid, u64 offset, struct iov_iter *to, int *err); int p9_client_write(struct p9_fid *fid, u64 offset, struct iov_iter *from, int *err); +struct netfs_io_subrequest; +void p9_client_write_subreq(struct netfs_io_subrequest *subreq); int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset); int p9dirent_read(struct p9_client *clnt, char *buf, int len, struct p9_dirent *dirent); diff --git a/net/9p/Kconfig b/net/9p/Kconfig index 00ebce9e5a65..bcdab9c23b40 100644 --- a/net/9p/Kconfig +++ b/net/9p/Kconfig @@ -5,6 +5,7 @@ menuconfig NET_9P tristate "Plan 9 Resource Sharing Support (9P2000)" + select NETFS_SUPPORT help If you say Y here, you will get experimental support for Plan 9 resource sharing via the 9P2000 protocol. diff --git a/net/9p/client.c b/net/9p/client.c index f7e90b4769bb..00774656eeac 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -1661,6 +1662,54 @@ p9_client_write(struct p9_fid *fid, u64 offset, struct iov_iter *from, int *err) } EXPORT_SYMBOL(p9_client_write); +void +p9_client_write_subreq(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *wreq = subreq->rreq; + struct p9_fid *fid = wreq->netfs_priv; + struct p9_client *clnt = fid->clnt; + struct p9_req_t *req; + unsigned long long start = subreq->start + subreq->transferred; + int written, len = subreq->len - subreq->transferred; + int err; + + p9_debug(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu len %d\n", + fid->fid, start, len); + + /* Don't bother zerocopy for small IO (< 1024) */ + if (clnt->trans_mod->zc_request && len > 1024) { + req = p9_client_zc_rpc(clnt, P9_TWRITE, NULL, &subreq->io_iter, + 0, wreq->len, P9_ZC_HDR_SZ, "dqd", + fid->fid, start, len); + } else { + req = p9_client_rpc(clnt, P9_TWRITE, "dqV", fid->fid, + start, len, &subreq->io_iter); + } + if (IS_ERR(req)) { + netfs_write_subrequest_terminated(subreq, PTR_ERR(req), false); + return; + } + + err = p9pdu_readf(&req->rc, clnt->proto_version, "d", &written); + if (err) { + trace_9p_protocol_dump(clnt, &req->rc); + p9_req_put(clnt, req); + netfs_write_subrequest_terminated(subreq, err, false); + return; + } + + if (written > len) { + pr_err("bogus RWRITE count (%d > %u)\n", written, len); + written = len; + } + + p9_debug(P9_DEBUG_9P, "<<< RWRITE count %d\n", len); + + p9_req_put(clnt, req); + netfs_write_subrequest_terminated(subreq, written, false); +} +EXPORT_SYMBOL(p9_client_write_subreq); + struct p9_wstat *p9_client_stat(struct p9_fid *fid) { int err; -- cgit v1.2.3-59-g8ed1b From 64e64e6c18c6bc7767ea6f2762c87c9ac981f2d1 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 20:31:13 +0000 Subject: netfs, cachefiles: Implement helpers for new write code Implement the helpers for the new write code in cachefiles. There's now an optional ->prepare_write() that allows the filesystem to set the parameters for the next write, such as maximum size and maximum segment count, and an ->issue_write() that is called to initiate an (asynchronous) write operation. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: netfs@lists.linux.dev cc: linux-erofs@lists.ozlabs.org cc: linux-fsdevel@vger.kernel.org --- fs/cachefiles/io.c | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 74 insertions(+) (limited to 'fs') diff --git a/fs/cachefiles/io.c b/fs/cachefiles/io.c index 5ba5c7814fe4..e667dbcd20e8 100644 --- a/fs/cachefiles/io.c +++ b/fs/cachefiles/io.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -622,6 +623,77 @@ static int cachefiles_prepare_write(struct netfs_cache_resources *cres, return ret; } +static void cachefiles_prepare_write_subreq(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *wreq = subreq->rreq; + struct netfs_cache_resources *cres = &wreq->cache_resources; + + _enter("W=%x[%x] %llx", wreq->debug_id, subreq->debug_index, subreq->start); + + subreq->max_len = ULONG_MAX; + subreq->max_nr_segs = BIO_MAX_VECS; + + if (!cachefiles_cres_file(cres)) { + if (!fscache_wait_for_operation(cres, FSCACHE_WANT_WRITE)) + return netfs_prepare_write_failed(subreq); + if (!cachefiles_cres_file(cres)) + return netfs_prepare_write_failed(subreq); + } +} + +static void cachefiles_issue_write(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *wreq = subreq->rreq; + struct netfs_cache_resources *cres = &wreq->cache_resources; + struct cachefiles_object *object = cachefiles_cres_object(cres); + struct cachefiles_cache *cache = object->volume->cache; + const struct cred *saved_cred; + size_t off, pre, post, len = subreq->len; + loff_t start = subreq->start; + int ret; + + _enter("W=%x[%x] %llx-%llx", + wreq->debug_id, subreq->debug_index, start, start + len - 1); + + /* We need to start on the cache granularity boundary */ + off = start & (CACHEFILES_DIO_BLOCK_SIZE - 1); + if (off) { + pre = CACHEFILES_DIO_BLOCK_SIZE - off; + if (pre >= len) { + netfs_write_subrequest_terminated(subreq, len, false); + return; + } + subreq->transferred += pre; + start += pre; + len -= pre; + iov_iter_advance(&subreq->io_iter, pre); + } + + /* We also need to end on the cache granularity boundary */ + post = len & (CACHEFILES_DIO_BLOCK_SIZE - 1); + if (post) { + len -= post; + if (len == 0) { + netfs_write_subrequest_terminated(subreq, post, false); + return; + } + iov_iter_truncate(&subreq->io_iter, len); + } + + cachefiles_begin_secure(cache, &saved_cred); + ret = __cachefiles_prepare_write(object, cachefiles_cres_file(cres), + &start, &len, len, true); + cachefiles_end_secure(cache, saved_cred); + if (ret < 0) { + netfs_write_subrequest_terminated(subreq, ret, false); + return; + } + + cachefiles_write(&subreq->rreq->cache_resources, + subreq->start, &subreq->io_iter, + netfs_write_subrequest_terminated, subreq); +} + /* * Clean up an operation. */ @@ -638,8 +710,10 @@ static const struct netfs_cache_ops cachefiles_netfs_cache_ops = { .end_operation = cachefiles_end_operation, .read = cachefiles_read, .write = cachefiles_write, + .issue_write = cachefiles_issue_write, .prepare_read = cachefiles_prepare_read, .prepare_write = cachefiles_prepare_write, + .prepare_write_subreq = cachefiles_prepare_write_subreq, .prepare_ondemand_read = cachefiles_prepare_ondemand_read, .query_occupancy = cachefiles_query_occupancy, }; -- cgit v1.2.3-59-g8ed1b From 2df86547b23dabcd02ab000a24ed7813606c269f Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 8 Mar 2024 12:36:05 +0000 Subject: netfs: Cut over to using new writeback code Cut over to using the new writeback code. The old code is #ifdef'd out or otherwise removed from compilation to avoid conflicts and will be removed in a future patch. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: Marc Dionne cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/9p/vfs_addr.c | 6 ++---- fs/afs/file.c | 3 +-- fs/afs/internal.h | 1 - fs/afs/write.c | 2 ++ fs/netfs/Makefile | 1 - fs/netfs/buffered_write.c | 45 ++++++++++++++++++++++++--------------------- fs/netfs/direct_write.c | 26 ++++++++++++++------------ fs/netfs/internal.h | 21 ++++++--------------- fs/netfs/write_collect.c | 8 ++++---- fs/netfs/write_issue.c | 18 +++++++++--------- include/linux/netfs.h | 9 --------- 11 files changed, 62 insertions(+), 78 deletions(-) (limited to 'fs') diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 07d03efdd594..4845e655bc39 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -60,6 +60,7 @@ static void v9fs_issue_write(struct netfs_io_subrequest *subreq) netfs_write_subrequest_terminated(subreq, len ?: err, false); } +#if 0 // TODO: Remove static void v9fs_upload_to_server(struct netfs_io_subrequest *subreq) { struct p9_fid *fid = subreq->rreq->netfs_priv; @@ -91,6 +92,7 @@ static void v9fs_create_write_requests(struct netfs_io_request *wreq, loff_t sta if (subreq) netfs_queue_write_request(subreq); } +#endif /** * v9fs_issue_read - Issue a read from 9P @@ -121,18 +123,15 @@ static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file) { struct p9_fid *fid; bool writing = (rreq->origin == NETFS_READ_FOR_WRITE || - rreq->origin == NETFS_WRITEBACK || rreq->origin == NETFS_WRITETHROUGH || rreq->origin == NETFS_UNBUFFERED_WRITE || rreq->origin == NETFS_DIO_WRITE); -#if 0 // TODO: Cut over if (rreq->origin == NETFS_WRITEBACK) return 0; /* We don't get the write handle until we find we * have actually dirty data and not just * copy-to-cache data. */ -#endif if (file) { fid = file->private_data; @@ -179,7 +178,6 @@ const struct netfs_request_ops v9fs_req_ops = { .issue_read = v9fs_issue_read, .begin_writeback = v9fs_begin_writeback, .issue_write = v9fs_issue_write, - .create_write_requests = v9fs_create_write_requests, }; const struct address_space_operations v9fs_addr_operations = { diff --git a/fs/afs/file.c b/fs/afs/file.c index db9ebae84fa2..8f983e3ecae7 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -353,7 +353,7 @@ static int afs_init_request(struct netfs_io_request *rreq, struct file *file) if (file) rreq->netfs_priv = key_get(afs_file_key(file)); rreq->rsize = 256 * 1024; - rreq->wsize = 256 * 1024; + rreq->wsize = 256 * 1024 * 1024; return 0; } @@ -399,7 +399,6 @@ const struct netfs_request_ops afs_req_ops = { .issue_read = afs_issue_read, .update_i_size = afs_update_i_size, .invalidate_cache = afs_netfs_invalidate_cache, - .create_write_requests = afs_create_write_requests, .begin_writeback = afs_begin_writeback, .prepare_write = afs_prepare_write, .issue_write = afs_issue_write, diff --git a/fs/afs/internal.h b/fs/afs/internal.h index dcf0ae0323d3..887245f9336d 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -1605,7 +1605,6 @@ extern int afs_writepages(struct address_space *, struct writeback_control *); extern int afs_fsync(struct file *, loff_t, loff_t, int); extern vm_fault_t afs_page_mkwrite(struct vm_fault *vmf); extern void afs_prune_wb_keys(struct afs_vnode *); -void afs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size_t len); /* * xattr.c diff --git a/fs/afs/write.c b/fs/afs/write.c index 34595f482718..35db74627563 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -156,6 +156,7 @@ try_next_key: return afs_put_operation(op); } +#if 0 // TODO: Remove static void afs_upload_to_server(struct netfs_io_subrequest *subreq) { struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode); @@ -193,6 +194,7 @@ void afs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size if (subreq) netfs_queue_write_request(subreq); } +#endif /* * Writeback calls this when it finds a folio that needs uploading. This isn't diff --git a/fs/netfs/Makefile b/fs/netfs/Makefile index 1eb86e34b5a9..8e6781e0b10b 100644 --- a/fs/netfs/Makefile +++ b/fs/netfs/Makefile @@ -11,7 +11,6 @@ netfs-y := \ main.o \ misc.o \ objects.o \ - output.o \ write_collect.o \ write_issue.o diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 33ea4c20e7e7..ee8d9e3216bd 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -26,8 +26,6 @@ enum netfs_how_to_modify { NETFS_FLUSH_CONTENT, /* Flush incompatible content. */ }; -static void netfs_cleanup_buffered_write(struct netfs_io_request *wreq); - static void netfs_set_group(struct folio *folio, struct netfs_group *netfs_group) { void *priv = folio_get_private(folio); @@ -180,7 +178,7 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, }; struct netfs_io_request *wreq = NULL; struct netfs_folio *finfo; - struct folio *folio; + struct folio *folio, *writethrough = NULL; enum netfs_how_to_modify howto; enum netfs_folio_trace trace; unsigned int bdp_flags = (iocb->ki_flags & IOCB_SYNC) ? 0: BDP_ASYNC; @@ -209,7 +207,6 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, } if (!is_sync_kiocb(iocb)) wreq->iocb = iocb; - wreq->cleanup = netfs_cleanup_buffered_write; netfs_stat(&netfs_n_wh_writethrough); } else { netfs_stat(&netfs_n_wh_buffered_write); @@ -253,6 +250,16 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, offset = pos & (flen - 1); part = min_t(size_t, flen - offset, part); + /* Wait for writeback to complete. The writeback engine owns + * the info in folio->private and may change it until it + * removes the WB mark. + */ + if (folio_get_private(folio) && + folio_wait_writeback_killable(folio)) { + ret = written ? -EINTR : -ERESTARTSYS; + goto error_folio_unlock; + } + if (signal_pending(current)) { ret = written ? -EINTR : -ERESTARTSYS; goto error_folio_unlock; @@ -327,6 +334,7 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, maybe_trouble = true; iov_iter_revert(iter, copied); copied = 0; + folio_unlock(folio); goto retry; } netfs_set_group(folio, netfs_group); @@ -382,23 +390,14 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, if (likely(!wreq)) { folio_mark_dirty(folio); + folio_unlock(folio); } else { - if (folio_test_dirty(folio)) - /* Sigh. mmap. */ - folio_clear_dirty_for_io(folio); - /* We make multiple writes to the folio... */ - if (!folio_test_writeback(folio)) { - folio_start_writeback(folio); - if (wreq->iter.count == 0) - trace_netfs_folio(folio, netfs_folio_trace_wthru); - else - trace_netfs_folio(folio, netfs_folio_trace_wthru_plus); - } - netfs_advance_writethrough(wreq, copied, - offset + copied == flen); + netfs_advance_writethrough(wreq, &wbc, folio, copied, + offset + copied == flen, + &writethrough); + /* Folio unlocked */ } retry: - folio_unlock(folio); folio_put(folio); folio = NULL; @@ -407,7 +406,7 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, out: if (unlikely(wreq)) { - ret2 = netfs_end_writethrough(wreq, iocb); + ret2 = netfs_end_writethrough(wreq, &wbc, writethrough); wbc_detach_inode(&wbc); if (ret2 == -EIOCBQUEUED) return ret2; @@ -529,11 +528,13 @@ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_gr sb_start_pagefault(inode->i_sb); - if (folio_wait_writeback_killable(folio)) + if (folio_lock_killable(folio) < 0) goto out; - if (folio_lock_killable(folio) < 0) + if (folio_wait_writeback_killable(folio)) { + ret = VM_FAULT_LOCKED; goto out; + } /* Can we see a streaming write here? */ if (WARN_ON(!folio_test_uptodate(folio))) { @@ -573,6 +574,7 @@ out: } EXPORT_SYMBOL(netfs_page_mkwrite); +#if 0 // TODO: Remove /* * Kill all the pages in the given range */ @@ -1199,3 +1201,4 @@ out: return ret; } EXPORT_SYMBOL(netfs_writepages); +#endif diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c index 36b6db504500..608ba6416919 100644 --- a/fs/netfs/direct_write.c +++ b/fs/netfs/direct_write.c @@ -34,6 +34,7 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov unsigned long long start = iocb->ki_pos; unsigned long long end = start + iov_iter_count(iter); ssize_t ret, n; + size_t len = iov_iter_count(iter); bool async = !is_sync_kiocb(iocb); _enter(""); @@ -46,13 +47,17 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov _debug("uw %llx-%llx", start, end); - wreq = netfs_alloc_request(iocb->ki_filp->f_mapping, iocb->ki_filp, - start, end - start, - iocb->ki_flags & IOCB_DIRECT ? - NETFS_DIO_WRITE : NETFS_UNBUFFERED_WRITE); + wreq = netfs_create_write_req(iocb->ki_filp->f_mapping, iocb->ki_filp, start, + iocb->ki_flags & IOCB_DIRECT ? + NETFS_DIO_WRITE : NETFS_UNBUFFERED_WRITE); if (IS_ERR(wreq)) return PTR_ERR(wreq); + wreq->io_streams[0].avail = true; + trace_netfs_write(wreq, (iocb->ki_flags & IOCB_DIRECT ? + netfs_write_trace_dio_write : + netfs_write_trace_unbuffered_write)); + { /* If this is an async op and we're not using a bounce buffer, * we have to save the source buffer as the iterator is only @@ -63,7 +68,7 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov * request. */ if (async || user_backed_iter(iter)) { - n = netfs_extract_user_iter(iter, wreq->len, &wreq->iter, 0); + n = netfs_extract_user_iter(iter, len, &wreq->iter, 0); if (n < 0) { ret = n; goto out; @@ -71,7 +76,6 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov wreq->direct_bv = (struct bio_vec *)wreq->iter.bvec; wreq->direct_bv_count = n; wreq->direct_bv_unpin = iov_iter_extract_will_pin(iter); - wreq->len = iov_iter_count(&wreq->iter); } else { wreq->iter = *iter; } @@ -79,6 +83,8 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov wreq->io_iter = wreq->iter; } + __set_bit(NETFS_RREQ_USE_IO_ITER, &wreq->flags); + /* Copy the data into the bounce buffer and encrypt it. */ // TODO @@ -87,10 +93,7 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov if (async) wreq->iocb = iocb; wreq->cleanup = netfs_cleanup_dio_write; - ret = netfs_begin_write(wreq, is_sync_kiocb(iocb), - iocb->ki_flags & IOCB_DIRECT ? - netfs_write_trace_dio_write : - netfs_write_trace_unbuffered_write); + ret = netfs_unbuffered_write(wreq, is_sync_kiocb(iocb), iov_iter_count(&wreq->io_iter)); if (ret < 0) { _debug("begin = %zd", ret); goto out; @@ -100,9 +103,8 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov trace_netfs_rreq(wreq, netfs_rreq_trace_wait_ip); wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, TASK_UNINTERRUPTIBLE); - + smp_rmb(); /* Read error/transferred after RIP flag */ ret = wreq->error; - _debug("waited = %zd", ret); if (ret == 0) { ret = wreq->transferred; iocb->ki_pos += ret; diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index 5d3f74a70fa7..95e281a8af78 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -92,15 +92,6 @@ static inline void netfs_see_request(struct netfs_io_request *rreq, trace_netfs_rreq_ref(rreq->debug_id, refcount_read(&rreq->ref), what); } -/* - * output.c - */ -int netfs_begin_write(struct netfs_io_request *wreq, bool may_wait, - enum netfs_write_trace what); -struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len); -int netfs_advance_writethrough(struct netfs_io_request *wreq, size_t copied, bool to_page_end); -int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb); - /* * stats.c */ @@ -172,12 +163,12 @@ void netfs_reissue_write(struct netfs_io_stream *stream, int netfs_advance_write(struct netfs_io_request *wreq, struct netfs_io_stream *stream, loff_t start, size_t len, bool to_eof); -struct netfs_io_request *new_netfs_begin_writethrough(struct kiocb *iocb, size_t len); -int new_netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, - struct folio *folio, size_t copied, bool to_page_end, - struct folio **writethrough_cache); -int new_netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, - struct folio *writethrough_cache); +struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len); +int netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *folio, size_t copied, bool to_page_end, + struct folio **writethrough_cache); +int netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *writethrough_cache); int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len); /* diff --git a/fs/netfs/write_collect.c b/fs/netfs/write_collect.c index b8c1d3ca724a..f14c08bf605d 100644 --- a/fs/netfs/write_collect.c +++ b/fs/netfs/write_collect.c @@ -709,7 +709,7 @@ void netfs_wake_write_collector(struct netfs_io_request *wreq, bool was_async) } /** - * new_netfs_write_subrequest_terminated - Note the termination of a write operation. + * netfs_write_subrequest_terminated - Note the termination of a write operation. * @_op: The I/O request that has terminated. * @transferred_or_error: The amount of data transferred or an error code. * @was_async: The termination was asynchronous @@ -731,8 +731,8 @@ void netfs_wake_write_collector(struct netfs_io_request *wreq, bool was_async) * Note that %_op is a void* so that the function can be passed to * kiocb::term_func without the need for a casting wrapper. */ -void new_netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, - bool was_async) +void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, + bool was_async) { struct netfs_io_subrequest *subreq = _op; struct netfs_io_request *wreq = subreq->rreq; @@ -800,4 +800,4 @@ void new_netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_err netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); } -EXPORT_SYMBOL(new_netfs_write_subrequest_terminated); +EXPORT_SYMBOL(netfs_write_subrequest_terminated); diff --git a/fs/netfs/write_issue.c b/fs/netfs/write_issue.c index 7ea86e33382c..e190043bc0da 100644 --- a/fs/netfs/write_issue.c +++ b/fs/netfs/write_issue.c @@ -494,8 +494,8 @@ static int netfs_write_folio(struct netfs_io_request *wreq, /* * Write some of the pending data back to the server */ -int new_netfs_writepages(struct address_space *mapping, - struct writeback_control *wbc) +int netfs_writepages(struct address_space *mapping, + struct writeback_control *wbc) { struct netfs_inode *ictx = netfs_inode(mapping->host); struct netfs_io_request *wreq = NULL; @@ -556,12 +556,12 @@ out: _leave(" = %d", error); return error; } -EXPORT_SYMBOL(new_netfs_writepages); +EXPORT_SYMBOL(netfs_writepages); /* * Begin a write operation for writing through the pagecache. */ -struct netfs_io_request *new_netfs_begin_writethrough(struct kiocb *iocb, size_t len) +struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len) { struct netfs_io_request *wreq = NULL; struct netfs_inode *ictx = netfs_inode(file_inode(iocb->ki_filp)); @@ -586,9 +586,9 @@ struct netfs_io_request *new_netfs_begin_writethrough(struct kiocb *iocb, size_t * to the request. If we've added more than wsize then we need to create a new * subrequest. */ -int new_netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, - struct folio *folio, size_t copied, bool to_page_end, - struct folio **writethrough_cache) +int netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *folio, size_t copied, bool to_page_end, + struct folio **writethrough_cache) { _enter("R=%x ic=%zu ws=%u cp=%zu tp=%u", wreq->debug_id, wreq->iter.count, wreq->wsize, copied, to_page_end); @@ -618,8 +618,8 @@ int new_netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeba /* * End a write operation used when writing through the pagecache. */ -int new_netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, - struct folio *writethrough_cache) +int netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *writethrough_cache) { struct netfs_inode *ictx = netfs_inode(wreq->inode); int ret; diff --git a/include/linux/netfs.h b/include/linux/netfs.h index 42dba05a428b..c2ba364041b0 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -303,8 +303,6 @@ struct netfs_request_ops { void (*update_i_size)(struct inode *inode, loff_t i_size); /* Write request handling */ - void (*create_write_requests)(struct netfs_io_request *wreq, - loff_t start, size_t len); void (*begin_writeback)(struct netfs_io_request *wreq); void (*prepare_write)(struct netfs_io_subrequest *subreq); void (*issue_write)(struct netfs_io_subrequest *subreq); @@ -409,8 +407,6 @@ int netfs_write_begin(struct netfs_inode *, struct file *, struct folio **, void **fsdata); int netfs_writepages(struct address_space *mapping, struct writeback_control *wbc); -int new_netfs_writepages(struct address_space *mapping, - struct writeback_control *wbc); bool netfs_dirty_folio(struct address_space *mapping, struct folio *folio); int netfs_unpin_writeback(struct inode *inode, struct writeback_control *wbc); void netfs_clear_inode_writeback(struct inode *inode, const void *aux); @@ -431,14 +427,9 @@ ssize_t netfs_extract_user_iter(struct iov_iter *orig, size_t orig_len, iov_iter_extraction_t extraction_flags); size_t netfs_limit_iter(const struct iov_iter *iter, size_t start_offset, size_t max_size, size_t max_segs); -struct netfs_io_subrequest *netfs_create_write_request( - struct netfs_io_request *wreq, enum netfs_io_source dest, - loff_t start, size_t len, work_func_t worker); void netfs_prepare_write_failed(struct netfs_io_subrequest *subreq); void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, bool was_async); -void new_netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, - bool was_async); void netfs_queue_write_request(struct netfs_io_subrequest *subreq); int netfs_start_io_read(struct inode *inode); -- cgit v1.2.3-59-g8ed1b From c245868524cc6e4954dd26588aade08e2cc405d0 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 8 Mar 2024 12:36:05 +0000 Subject: netfs: Remove the old writeback code Remove the old writeback code. Signed-off-by: David Howells cc: Jeff Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: Marc Dionne cc: v9fs@lists.linux.dev cc: linux-afs@lists.infradead.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/9p/vfs_addr.c | 34 --- fs/afs/write.c | 40 --- fs/netfs/buffered_write.c | 629 ---------------------------------------------- fs/netfs/output.c | 477 ----------------------------------- 4 files changed, 1180 deletions(-) delete mode 100644 fs/netfs/output.c (limited to 'fs') diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 4845e655bc39..a97ceb105cd8 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -60,40 +60,6 @@ static void v9fs_issue_write(struct netfs_io_subrequest *subreq) netfs_write_subrequest_terminated(subreq, len ?: err, false); } -#if 0 // TODO: Remove -static void v9fs_upload_to_server(struct netfs_io_subrequest *subreq) -{ - struct p9_fid *fid = subreq->rreq->netfs_priv; - int err, len; - - trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - len = p9_client_write(fid, subreq->start, &subreq->io_iter, &err); - netfs_write_subrequest_terminated(subreq, len ?: err, false); -} - -static void v9fs_upload_to_server_worker(struct work_struct *work) -{ - struct netfs_io_subrequest *subreq = - container_of(work, struct netfs_io_subrequest, work); - - v9fs_upload_to_server(subreq); -} - -/* - * Set up write requests for a writeback slice. We need to add a write request - * for each write we want to make. - */ -static void v9fs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size_t len) -{ - struct netfs_io_subrequest *subreq; - - subreq = netfs_create_write_request(wreq, NETFS_UPLOAD_TO_SERVER, - start, len, v9fs_upload_to_server_worker); - if (subreq) - netfs_queue_write_request(subreq); -} -#endif - /** * v9fs_issue_read - Issue a read from 9P * @subreq: The read to make diff --git a/fs/afs/write.c b/fs/afs/write.c index 35db74627563..b8505a8b622a 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -156,46 +156,6 @@ try_next_key: return afs_put_operation(op); } -#if 0 // TODO: Remove -static void afs_upload_to_server(struct netfs_io_subrequest *subreq) -{ - struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode); - ssize_t ret; - - _enter("%x[%x],%zx", - subreq->rreq->debug_id, subreq->debug_index, subreq->io_iter.count); - - trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - ret = afs_store_data(vnode, &subreq->io_iter, subreq->start); - netfs_write_subrequest_terminated(subreq, ret < 0 ? ret : subreq->len, - false); -} - -static void afs_upload_to_server_worker(struct work_struct *work) -{ - struct netfs_io_subrequest *subreq = - container_of(work, struct netfs_io_subrequest, work); - - afs_upload_to_server(subreq); -} - -/* - * Set up write requests for a writeback slice. We need to add a write request - * for each write we want to make. - */ -void afs_create_write_requests(struct netfs_io_request *wreq, loff_t start, size_t len) -{ - struct netfs_io_subrequest *subreq; - - _enter("%x,%llx-%llx", wreq->debug_id, start, start + len); - - subreq = netfs_create_write_request(wreq, NETFS_UPLOAD_TO_SERVER, - start, len, afs_upload_to_server_worker); - if (subreq) - netfs_queue_write_request(subreq); -} -#endif - /* * Writeback calls this when it finds a folio that needs uploading. This isn't * called if writeback only has copy-to-cache to deal with. diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index ee8d9e3216bd..b21d331eaa1e 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -573,632 +573,3 @@ out: return ret; } EXPORT_SYMBOL(netfs_page_mkwrite); - -#if 0 // TODO: Remove -/* - * Kill all the pages in the given range - */ -static void netfs_kill_pages(struct address_space *mapping, - loff_t start, loff_t len) -{ - struct folio *folio; - pgoff_t index = start / PAGE_SIZE; - pgoff_t last = (start + len - 1) / PAGE_SIZE, next; - - _enter("%llx-%llx", start, start + len - 1); - - do { - _debug("kill %lx (to %lx)", index, last); - - folio = filemap_get_folio(mapping, index); - if (IS_ERR(folio)) { - next = index + 1; - continue; - } - - next = folio_next_index(folio); - - trace_netfs_folio(folio, netfs_folio_trace_kill); - folio_clear_uptodate(folio); - folio_end_writeback(folio); - folio_lock(folio); - generic_error_remove_folio(mapping, folio); - folio_unlock(folio); - folio_put(folio); - - } while (index = next, index <= last); - - _leave(""); -} - -/* - * Redirty all the pages in a given range. - */ -static void netfs_redirty_pages(struct address_space *mapping, - loff_t start, loff_t len) -{ - struct folio *folio; - pgoff_t index = start / PAGE_SIZE; - pgoff_t last = (start + len - 1) / PAGE_SIZE, next; - - _enter("%llx-%llx", start, start + len - 1); - - do { - _debug("redirty %llx @%llx", len, start); - - folio = filemap_get_folio(mapping, index); - if (IS_ERR(folio)) { - next = index + 1; - continue; - } - - next = folio_next_index(folio); - trace_netfs_folio(folio, netfs_folio_trace_redirty); - filemap_dirty_folio(mapping, folio); - folio_end_writeback(folio); - folio_put(folio); - } while (index = next, index <= last); - - balance_dirty_pages_ratelimited(mapping); - - _leave(""); -} - -/* - * Completion of write to server - */ -static void netfs_pages_written_back(struct netfs_io_request *wreq) -{ - struct address_space *mapping = wreq->mapping; - struct netfs_folio *finfo; - struct netfs_group *group = NULL; - struct folio *folio; - pgoff_t last; - int gcount = 0; - - XA_STATE(xas, &mapping->i_pages, wreq->start / PAGE_SIZE); - - _enter("%llx-%llx", wreq->start, wreq->start + wreq->len); - - rcu_read_lock(); - - last = (wreq->start + wreq->len - 1) / PAGE_SIZE; - xas_for_each(&xas, folio, last) { - WARN(!folio_test_writeback(folio), - "bad %llx @%llx page %lx %lx\n", - wreq->len, wreq->start, folio->index, last); - - if ((finfo = netfs_folio_info(folio))) { - /* Streaming writes cannot be redirtied whilst under - * writeback, so discard the streaming record. - */ - folio_detach_private(folio); - group = finfo->netfs_group; - gcount++; - trace_netfs_folio(folio, netfs_folio_trace_clear_s); - kfree(finfo); - } else if ((group = netfs_folio_group(folio))) { - /* Need to detach the group pointer if the page didn't - * get redirtied. If it has been redirtied, then it - * must be within the same group. - */ - if (folio_test_dirty(folio)) { - trace_netfs_folio(folio, netfs_folio_trace_redirtied); - goto end_wb; - } - if (folio_trylock(folio)) { - if (!folio_test_dirty(folio)) { - folio_detach_private(folio); - gcount++; - if (group == NETFS_FOLIO_COPY_TO_CACHE) - trace_netfs_folio(folio, - netfs_folio_trace_end_copy); - else - trace_netfs_folio(folio, netfs_folio_trace_clear_g); - } else { - trace_netfs_folio(folio, netfs_folio_trace_redirtied); - } - folio_unlock(folio); - goto end_wb; - } - - xas_pause(&xas); - rcu_read_unlock(); - folio_lock(folio); - if (!folio_test_dirty(folio)) { - folio_detach_private(folio); - gcount++; - trace_netfs_folio(folio, netfs_folio_trace_clear_g); - } else { - trace_netfs_folio(folio, netfs_folio_trace_redirtied); - } - folio_unlock(folio); - rcu_read_lock(); - } else { - trace_netfs_folio(folio, netfs_folio_trace_clear); - } - end_wb: - xas_advance(&xas, folio_next_index(folio) - 1); - folio_end_writeback(folio); - } - - rcu_read_unlock(); - netfs_put_group_many(group, gcount); - _leave(""); -} - -/* - * Deal with the disposition of the folios that are under writeback to close - * out the operation. - */ -static void netfs_cleanup_buffered_write(struct netfs_io_request *wreq) -{ - struct address_space *mapping = wreq->mapping; - - _enter(""); - - switch (wreq->error) { - case 0: - netfs_pages_written_back(wreq); - break; - - default: - pr_notice("R=%08x Unexpected error %d\n", wreq->debug_id, wreq->error); - fallthrough; - case -EACCES: - case -EPERM: - case -ENOKEY: - case -EKEYEXPIRED: - case -EKEYREJECTED: - case -EKEYREVOKED: - case -ENETRESET: - case -EDQUOT: - case -ENOSPC: - netfs_redirty_pages(mapping, wreq->start, wreq->len); - break; - - case -EROFS: - case -EIO: - case -EREMOTEIO: - case -EFBIG: - case -ENOENT: - case -ENOMEDIUM: - case -ENXIO: - netfs_kill_pages(mapping, wreq->start, wreq->len); - break; - } - - if (wreq->error) - mapping_set_error(mapping, wreq->error); - if (wreq->netfs_ops->done) - wreq->netfs_ops->done(wreq); -} - -/* - * Extend the region to be written back to include subsequent contiguously - * dirty pages if possible, but don't sleep while doing so. - * - * If this page holds new content, then we can include filler zeros in the - * writeback. - */ -static void netfs_extend_writeback(struct address_space *mapping, - struct netfs_group *group, - struct xa_state *xas, - long *_count, - loff_t start, - loff_t max_len, - size_t *_len, - size_t *_top) -{ - struct netfs_folio *finfo; - struct folio_batch fbatch; - struct folio *folio; - unsigned int i; - pgoff_t index = (start + *_len) / PAGE_SIZE; - size_t len; - void *priv; - bool stop = true; - - folio_batch_init(&fbatch); - - do { - /* Firstly, we gather up a batch of contiguous dirty pages - * under the RCU read lock - but we can't clear the dirty flags - * there if any of those pages are mapped. - */ - rcu_read_lock(); - - xas_for_each(xas, folio, ULONG_MAX) { - stop = true; - if (xas_retry(xas, folio)) - continue; - if (xa_is_value(folio)) - break; - if (folio->index != index) { - xas_reset(xas); - break; - } - - if (!folio_try_get_rcu(folio)) { - xas_reset(xas); - continue; - } - - /* Has the folio moved or been split? */ - if (unlikely(folio != xas_reload(xas))) { - folio_put(folio); - xas_reset(xas); - break; - } - - if (!folio_trylock(folio)) { - folio_put(folio); - xas_reset(xas); - break; - } - if (!folio_test_dirty(folio) || - folio_test_writeback(folio)) { - folio_unlock(folio); - folio_put(folio); - xas_reset(xas); - break; - } - - stop = false; - len = folio_size(folio); - priv = folio_get_private(folio); - if ((const struct netfs_group *)priv != group) { - stop = true; - finfo = netfs_folio_info(folio); - if (!finfo || - finfo->netfs_group != group || - finfo->dirty_offset > 0) { - folio_unlock(folio); - folio_put(folio); - xas_reset(xas); - break; - } - len = finfo->dirty_len; - } - - *_top += folio_size(folio); - index += folio_nr_pages(folio); - *_count -= folio_nr_pages(folio); - *_len += len; - if (*_len >= max_len || *_count <= 0) - stop = true; - - if (!folio_batch_add(&fbatch, folio)) - break; - if (stop) - break; - } - - xas_pause(xas); - rcu_read_unlock(); - - /* Now, if we obtained any folios, we can shift them to being - * writable and mark them for caching. - */ - if (!folio_batch_count(&fbatch)) - break; - - for (i = 0; i < folio_batch_count(&fbatch); i++) { - folio = fbatch.folios[i]; - if (group == NETFS_FOLIO_COPY_TO_CACHE) - trace_netfs_folio(folio, netfs_folio_trace_copy_plus); - else - trace_netfs_folio(folio, netfs_folio_trace_store_plus); - - if (!folio_clear_dirty_for_io(folio)) - BUG(); - folio_start_writeback(folio); - folio_unlock(folio); - } - - folio_batch_release(&fbatch); - cond_resched(); - } while (!stop); -} - -/* - * Synchronously write back the locked page and any subsequent non-locked dirty - * pages. - */ -static ssize_t netfs_write_back_from_locked_folio(struct address_space *mapping, - struct writeback_control *wbc, - struct netfs_group *group, - struct xa_state *xas, - struct folio *folio, - unsigned long long start, - unsigned long long end) -{ - struct netfs_io_request *wreq; - struct netfs_folio *finfo; - struct netfs_inode *ctx = netfs_inode(mapping->host); - unsigned long long i_size = i_size_read(&ctx->inode); - size_t len, max_len; - long count = wbc->nr_to_write; - int ret; - - _enter(",%lx,%llx-%llx", folio->index, start, end); - - wreq = netfs_alloc_request(mapping, NULL, start, folio_size(folio), - group == NETFS_FOLIO_COPY_TO_CACHE ? - NETFS_COPY_TO_CACHE : NETFS_WRITEBACK); - if (IS_ERR(wreq)) { - folio_unlock(folio); - return PTR_ERR(wreq); - } - - if (!folio_clear_dirty_for_io(folio)) - BUG(); - folio_start_writeback(folio); - - count -= folio_nr_pages(folio); - - /* Find all consecutive lockable dirty pages that have contiguous - * written regions, stopping when we find a page that is not - * immediately lockable, is not dirty or is missing, or we reach the - * end of the range. - */ - if (group == NETFS_FOLIO_COPY_TO_CACHE) - trace_netfs_folio(folio, netfs_folio_trace_copy); - else - trace_netfs_folio(folio, netfs_folio_trace_store); - - len = wreq->len; - finfo = netfs_folio_info(folio); - if (finfo) { - start += finfo->dirty_offset; - if (finfo->dirty_offset + finfo->dirty_len != len) { - len = finfo->dirty_len; - goto cant_expand; - } - len = finfo->dirty_len; - } - - if (start < i_size) { - /* Trim the write to the EOF; the extra data is ignored. Also - * put an upper limit on the size of a single storedata op. - */ - max_len = 65536 * 4096; - max_len = min_t(unsigned long long, max_len, end - start + 1); - max_len = min_t(unsigned long long, max_len, i_size - start); - - if (len < max_len) - netfs_extend_writeback(mapping, group, xas, &count, start, - max_len, &len, &wreq->upper_len); - } - -cant_expand: - len = min_t(unsigned long long, len, i_size - start); - - /* We now have a contiguous set of dirty pages, each with writeback - * set; the first page is still locked at this point, but all the rest - * have been unlocked. - */ - folio_unlock(folio); - wreq->start = start; - wreq->len = len; - - if (start < i_size) { - _debug("write back %zx @%llx [%llx]", len, start, i_size); - - /* Speculatively write to the cache. We have to fix this up - * later if the store fails. - */ - wreq->cleanup = netfs_cleanup_buffered_write; - - iov_iter_xarray(&wreq->iter, ITER_SOURCE, &mapping->i_pages, start, - wreq->upper_len); - if (group != NETFS_FOLIO_COPY_TO_CACHE) { - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - ret = netfs_begin_write(wreq, true, netfs_write_trace_writeback); - } else { - ret = netfs_begin_write(wreq, true, netfs_write_trace_copy_to_cache); - } - if (ret == 0 || ret == -EIOCBQUEUED) - wbc->nr_to_write -= len / PAGE_SIZE; - } else { - _debug("write discard %zx @%llx [%llx]", len, start, i_size); - - /* The dirty region was entirely beyond the EOF. */ - netfs_pages_written_back(wreq); - ret = 0; - } - - netfs_put_request(wreq, false, netfs_rreq_trace_put_return); - _leave(" = 1"); - return 1; -} - -/* - * Write a region of pages back to the server - */ -static ssize_t netfs_writepages_begin(struct address_space *mapping, - struct writeback_control *wbc, - struct netfs_group *group, - struct xa_state *xas, - unsigned long long *_start, - unsigned long long end) -{ - const struct netfs_folio *finfo; - struct folio *folio; - unsigned long long start = *_start; - ssize_t ret; - void *priv; - int skips = 0; - - _enter("%llx,%llx,", start, end); - -search_again: - /* Find the first dirty page in the group. */ - rcu_read_lock(); - - for (;;) { - folio = xas_find_marked(xas, end / PAGE_SIZE, PAGECACHE_TAG_DIRTY); - if (xas_retry(xas, folio) || xa_is_value(folio)) - continue; - if (!folio) - break; - - if (!folio_try_get_rcu(folio)) { - xas_reset(xas); - continue; - } - - if (unlikely(folio != xas_reload(xas))) { - folio_put(folio); - xas_reset(xas); - continue; - } - - /* Skip any dirty folio that's not in the group of interest. */ - priv = folio_get_private(folio); - if ((const struct netfs_group *)priv == NETFS_FOLIO_COPY_TO_CACHE) { - group = NETFS_FOLIO_COPY_TO_CACHE; - } else if ((const struct netfs_group *)priv != group) { - finfo = __netfs_folio_info(priv); - if (!finfo || finfo->netfs_group != group) { - folio_put(folio); - continue; - } - } - - xas_pause(xas); - break; - } - rcu_read_unlock(); - if (!folio) - return 0; - - start = folio_pos(folio); /* May regress with THPs */ - - _debug("wback %lx", folio->index); - - /* At this point we hold neither the i_pages lock nor the page lock: - * the page may be truncated or invalidated (changing page->mapping to - * NULL), or even swizzled back from swapper_space to tmpfs file - * mapping - */ -lock_again: - if (wbc->sync_mode != WB_SYNC_NONE) { - ret = folio_lock_killable(folio); - if (ret < 0) - return ret; - } else { - if (!folio_trylock(folio)) - goto search_again; - } - - if (folio->mapping != mapping || - !folio_test_dirty(folio)) { - start += folio_size(folio); - folio_unlock(folio); - goto search_again; - } - - if (folio_test_writeback(folio)) { - folio_unlock(folio); - if (wbc->sync_mode != WB_SYNC_NONE) { - folio_wait_writeback(folio); - goto lock_again; - } - - start += folio_size(folio); - if (wbc->sync_mode == WB_SYNC_NONE) { - if (skips >= 5 || need_resched()) { - ret = 0; - goto out; - } - skips++; - } - goto search_again; - } - - ret = netfs_write_back_from_locked_folio(mapping, wbc, group, xas, - folio, start, end); -out: - if (ret > 0) - *_start = start + ret; - _leave(" = %zd [%llx]", ret, *_start); - return ret; -} - -/* - * Write a region of pages back to the server - */ -static int netfs_writepages_region(struct address_space *mapping, - struct writeback_control *wbc, - struct netfs_group *group, - unsigned long long *_start, - unsigned long long end) -{ - ssize_t ret; - - XA_STATE(xas, &mapping->i_pages, *_start / PAGE_SIZE); - - do { - ret = netfs_writepages_begin(mapping, wbc, group, &xas, - _start, end); - if (ret > 0 && wbc->nr_to_write > 0) - cond_resched(); - } while (ret > 0 && wbc->nr_to_write > 0); - - return ret > 0 ? 0 : ret; -} - -/* - * write some of the pending data back to the server - */ -int netfs_writepages(struct address_space *mapping, - struct writeback_control *wbc) -{ - struct netfs_group *group = NULL; - loff_t start, end; - int ret; - - _enter(""); - - /* We have to be careful as we can end up racing with setattr() - * truncating the pagecache since the caller doesn't take a lock here - * to prevent it. - */ - - if (wbc->range_cyclic && mapping->writeback_index) { - start = mapping->writeback_index * PAGE_SIZE; - ret = netfs_writepages_region(mapping, wbc, group, - &start, LLONG_MAX); - if (ret < 0) - goto out; - - if (wbc->nr_to_write <= 0) { - mapping->writeback_index = start / PAGE_SIZE; - goto out; - } - - start = 0; - end = mapping->writeback_index * PAGE_SIZE; - mapping->writeback_index = 0; - ret = netfs_writepages_region(mapping, wbc, group, &start, end); - if (ret == 0) - mapping->writeback_index = start / PAGE_SIZE; - } else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) { - start = 0; - ret = netfs_writepages_region(mapping, wbc, group, - &start, LLONG_MAX); - if (wbc->nr_to_write > 0 && ret == 0) - mapping->writeback_index = start / PAGE_SIZE; - } else { - start = wbc->range_start; - ret = netfs_writepages_region(mapping, wbc, group, - &start, wbc->range_end); - } - -out: - _leave(" = %d", ret); - return ret; -} -EXPORT_SYMBOL(netfs_writepages); -#endif diff --git a/fs/netfs/output.c b/fs/netfs/output.c deleted file mode 100644 index 85374322f10f..000000000000 --- a/fs/netfs/output.c +++ /dev/null @@ -1,477 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* Network filesystem high-level write support. - * - * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - */ - -#include -#include -#include -#include -#include -#include -#include "internal.h" - -/** - * netfs_create_write_request - Create a write operation. - * @wreq: The write request this is storing from. - * @dest: The destination type - * @start: Start of the region this write will modify - * @len: Length of the modification - * @worker: The worker function to handle the write(s) - * - * Allocate a write operation, set it up and add it to the list on a write - * request. - */ -struct netfs_io_subrequest *netfs_create_write_request(struct netfs_io_request *wreq, - enum netfs_io_source dest, - loff_t start, size_t len, - work_func_t worker) -{ - struct netfs_io_subrequest *subreq; - - subreq = netfs_alloc_subrequest(wreq); - if (subreq) { - INIT_WORK(&subreq->work, worker); - subreq->source = dest; - subreq->start = start; - subreq->len = len; - - switch (subreq->source) { - case NETFS_UPLOAD_TO_SERVER: - netfs_stat(&netfs_n_wh_upload); - break; - case NETFS_WRITE_TO_CACHE: - netfs_stat(&netfs_n_wh_write); - break; - default: - BUG(); - } - - subreq->io_iter = wreq->io_iter; - iov_iter_advance(&subreq->io_iter, subreq->start - wreq->start); - iov_iter_truncate(&subreq->io_iter, subreq->len); - - trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, - refcount_read(&subreq->ref), - netfs_sreq_trace_new); - atomic_inc(&wreq->nr_outstanding); - list_add_tail(&subreq->rreq_link, &wreq->subrequests); - trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); - } - - return subreq; -} -EXPORT_SYMBOL(netfs_create_write_request); - -/* - * Process a completed write request once all the component operations have - * been completed. - */ -static void netfs_write_terminated(struct netfs_io_request *wreq, bool was_async) -{ - struct netfs_io_subrequest *subreq; - struct netfs_inode *ctx = netfs_inode(wreq->inode); - size_t transferred = 0; - - _enter("R=%x[]", wreq->debug_id); - - trace_netfs_rreq(wreq, netfs_rreq_trace_write_done); - - list_for_each_entry(subreq, &wreq->subrequests, rreq_link) { - if (subreq->error || subreq->transferred == 0) - break; - transferred += subreq->transferred; - if (subreq->transferred < subreq->len) - break; - } - wreq->transferred = transferred; - - list_for_each_entry(subreq, &wreq->subrequests, rreq_link) { - if (!subreq->error) - continue; - switch (subreq->source) { - case NETFS_UPLOAD_TO_SERVER: - /* Depending on the type of failure, this may prevent - * writeback completion unless we're in disconnected - * mode. - */ - if (!wreq->error) - wreq->error = subreq->error; - break; - - case NETFS_WRITE_TO_CACHE: - /* Failure doesn't prevent writeback completion unless - * we're in disconnected mode. - */ - if (subreq->error != -ENOBUFS) - ctx->ops->invalidate_cache(wreq); - break; - - default: - WARN_ON_ONCE(1); - if (!wreq->error) - wreq->error = -EIO; - return; - } - } - - wreq->cleanup(wreq); - - if (wreq->origin == NETFS_DIO_WRITE && - wreq->mapping->nrpages) { - pgoff_t first = wreq->start >> PAGE_SHIFT; - pgoff_t last = (wreq->start + wreq->transferred - 1) >> PAGE_SHIFT; - invalidate_inode_pages2_range(wreq->mapping, first, last); - } - - if (wreq->origin == NETFS_DIO_WRITE) - inode_dio_end(wreq->inode); - - _debug("finished"); - trace_netfs_rreq(wreq, netfs_rreq_trace_wake_ip); - clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &wreq->flags); - wake_up_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS); - - if (wreq->iocb) { - wreq->iocb->ki_pos += transferred; - if (wreq->iocb->ki_complete) - wreq->iocb->ki_complete( - wreq->iocb, wreq->error ? wreq->error : transferred); - } - - netfs_clear_subrequests(wreq, was_async); - netfs_put_request(wreq, was_async, netfs_rreq_trace_put_complete); -} - -/* - * Deal with the completion of writing the data to the cache. - */ -void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, - bool was_async) -{ - struct netfs_io_subrequest *subreq = _op; - struct netfs_io_request *wreq = subreq->rreq; - unsigned int u; - - _enter("%x[%x] %zd", wreq->debug_id, subreq->debug_index, transferred_or_error); - - switch (subreq->source) { - case NETFS_UPLOAD_TO_SERVER: - netfs_stat(&netfs_n_wh_upload_done); - break; - case NETFS_WRITE_TO_CACHE: - netfs_stat(&netfs_n_wh_write_done); - break; - case NETFS_INVALID_WRITE: - break; - default: - BUG(); - } - - if (IS_ERR_VALUE(transferred_or_error)) { - subreq->error = transferred_or_error; - trace_netfs_failure(wreq, subreq, transferred_or_error, - netfs_fail_write); - goto failed; - } - - if (WARN(transferred_or_error > subreq->len - subreq->transferred, - "Subreq excess write: R%x[%x] %zd > %zu - %zu", - wreq->debug_id, subreq->debug_index, - transferred_or_error, subreq->len, subreq->transferred)) - transferred_or_error = subreq->len - subreq->transferred; - - subreq->error = 0; - subreq->transferred += transferred_or_error; - - if (iov_iter_count(&subreq->io_iter) != subreq->len - subreq->transferred) - pr_warn("R=%08x[%u] ITER POST-MISMATCH %zx != %zx-%zx %x\n", - wreq->debug_id, subreq->debug_index, - iov_iter_count(&subreq->io_iter), subreq->len, - subreq->transferred, subreq->io_iter.iter_type); - - if (subreq->transferred < subreq->len) - goto incomplete; - - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); -out: - trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); - - /* If we decrement nr_outstanding to 0, the ref belongs to us. */ - u = atomic_dec_return(&wreq->nr_outstanding); - if (u == 0) - netfs_write_terminated(wreq, was_async); - else if (u == 1) - wake_up_var(&wreq->nr_outstanding); - - netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); - return; - -incomplete: - if (transferred_or_error == 0) { - if (__test_and_set_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags)) { - subreq->error = -ENODATA; - goto failed; - } - } else { - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); - } - - __set_bit(NETFS_SREQ_SHORT_IO, &subreq->flags); - set_bit(NETFS_RREQ_INCOMPLETE_IO, &wreq->flags); - goto out; - -failed: - switch (subreq->source) { - case NETFS_WRITE_TO_CACHE: - netfs_stat(&netfs_n_wh_write_failed); - set_bit(NETFS_RREQ_INCOMPLETE_IO, &wreq->flags); - break; - case NETFS_UPLOAD_TO_SERVER: - netfs_stat(&netfs_n_wh_upload_failed); - set_bit(NETFS_RREQ_FAILED, &wreq->flags); - wreq->error = subreq->error; - break; - default: - break; - } - goto out; -} -EXPORT_SYMBOL(netfs_write_subrequest_terminated); - -static void netfs_write_to_cache_op(struct netfs_io_subrequest *subreq) -{ - struct netfs_io_request *wreq = subreq->rreq; - struct netfs_cache_resources *cres = &wreq->cache_resources; - - trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - - cres->ops->write(cres, subreq->start, &subreq->io_iter, - netfs_write_subrequest_terminated, subreq); -} - -static void netfs_write_to_cache_op_worker(struct work_struct *work) -{ - struct netfs_io_subrequest *subreq = - container_of(work, struct netfs_io_subrequest, work); - - netfs_write_to_cache_op(subreq); -} - -/** - * netfs_queue_write_request - Queue a write request for attention - * @subreq: The write request to be queued - * - * Queue the specified write request for processing by a worker thread. We - * pass the caller's ref on the request to the worker thread. - */ -void netfs_queue_write_request(struct netfs_io_subrequest *subreq) -{ - if (!queue_work(system_unbound_wq, &subreq->work)) - netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_wip); -} -EXPORT_SYMBOL(netfs_queue_write_request); - -/* - * Set up a op for writing to the cache. - */ -static void netfs_set_up_write_to_cache(struct netfs_io_request *wreq) -{ - struct netfs_cache_resources *cres = &wreq->cache_resources; - struct netfs_io_subrequest *subreq; - struct netfs_inode *ctx = netfs_inode(wreq->inode); - struct fscache_cookie *cookie = netfs_i_cookie(ctx); - loff_t start = wreq->start; - size_t len = wreq->len; - int ret; - - if (!fscache_cookie_enabled(cookie)) { - clear_bit(NETFS_RREQ_WRITE_TO_CACHE, &wreq->flags); - return; - } - - _debug("write to cache"); - ret = fscache_begin_write_operation(cres, cookie); - if (ret < 0) - return; - - ret = cres->ops->prepare_write(cres, &start, &len, wreq->upper_len, - i_size_read(wreq->inode), true); - if (ret < 0) - return; - - subreq = netfs_create_write_request(wreq, NETFS_WRITE_TO_CACHE, start, len, - netfs_write_to_cache_op_worker); - if (!subreq) - return; - - netfs_write_to_cache_op(subreq); -} - -/* - * Begin the process of writing out a chunk of data. - * - * We are given a write request that holds a series of dirty regions and - * (partially) covers a sequence of folios, all of which are present. The - * pages must have been marked as writeback as appropriate. - * - * We need to perform the following steps: - * - * (1) If encrypting, create an output buffer and encrypt each block of the - * data into it, otherwise the output buffer will point to the original - * folios. - * - * (2) If the data is to be cached, set up a write op for the entire output - * buffer to the cache, if the cache wants to accept it. - * - * (3) If the data is to be uploaded (ie. not merely cached): - * - * (a) If the data is to be compressed, create a compression buffer and - * compress the data into it. - * - * (b) For each destination we want to upload to, set up write ops to write - * to that destination. We may need multiple writes if the data is not - * contiguous or the span exceeds wsize for a server. - */ -int netfs_begin_write(struct netfs_io_request *wreq, bool may_wait, - enum netfs_write_trace what) -{ - struct netfs_inode *ctx = netfs_inode(wreq->inode); - - _enter("R=%x %llx-%llx f=%lx", - wreq->debug_id, wreq->start, wreq->start + wreq->len - 1, - wreq->flags); - - trace_netfs_write(wreq, what); - if (wreq->len == 0 || wreq->iter.count == 0) { - pr_err("Zero-sized write [R=%x]\n", wreq->debug_id); - return -EIO; - } - - if (wreq->origin == NETFS_DIO_WRITE) - inode_dio_begin(wreq->inode); - - wreq->io_iter = wreq->iter; - - /* ->outstanding > 0 carries a ref */ - netfs_get_request(wreq, netfs_rreq_trace_get_for_outstanding); - atomic_set(&wreq->nr_outstanding, 1); - - /* Start the encryption/compression going. We can do that in the - * background whilst we generate a list of write ops that we want to - * perform. - */ - // TODO: Encrypt or compress the region as appropriate - - /* We need to write all of the region to the cache */ - if (test_bit(NETFS_RREQ_WRITE_TO_CACHE, &wreq->flags)) - netfs_set_up_write_to_cache(wreq); - - /* However, we don't necessarily write all of the region to the server. - * Caching of reads is being managed this way also. - */ - if (test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags)) - ctx->ops->create_write_requests(wreq, wreq->start, wreq->len); - - if (atomic_dec_and_test(&wreq->nr_outstanding)) - netfs_write_terminated(wreq, false); - - if (!may_wait) - return -EIOCBQUEUED; - - wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); - return wreq->error; -} - -/* - * Begin a write operation for writing through the pagecache. - */ -struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len) -{ - struct netfs_io_request *wreq; - struct file *file = iocb->ki_filp; - - wreq = netfs_alloc_request(file->f_mapping, file, iocb->ki_pos, len, - NETFS_WRITETHROUGH); - if (IS_ERR(wreq)) - return wreq; - - trace_netfs_write(wreq, netfs_write_trace_writethrough); - - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - iov_iter_xarray(&wreq->iter, ITER_SOURCE, &wreq->mapping->i_pages, wreq->start, 0); - wreq->io_iter = wreq->iter; - - /* ->outstanding > 0 carries a ref */ - netfs_get_request(wreq, netfs_rreq_trace_get_for_outstanding); - atomic_set(&wreq->nr_outstanding, 1); - return wreq; -} - -static void netfs_submit_writethrough(struct netfs_io_request *wreq, bool final) -{ - struct netfs_inode *ictx = netfs_inode(wreq->inode); - unsigned long long start; - size_t len; - - if (!test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags)) - return; - - start = wreq->start + wreq->submitted; - len = wreq->iter.count - wreq->submitted; - if (!final) { - len /= wreq->wsize; /* Round to number of maximum packets */ - len *= wreq->wsize; - } - - ictx->ops->create_write_requests(wreq, start, len); - wreq->submitted += len; -} - -/* - * Advance the state of the write operation used when writing through the - * pagecache. Data has been copied into the pagecache that we need to append - * to the request. If we've added more than wsize then we need to create a new - * subrequest. - */ -int netfs_advance_writethrough(struct netfs_io_request *wreq, size_t copied, bool to_page_end) -{ - _enter("ic=%zu sb=%llu ws=%u cp=%zu tp=%u", - wreq->iter.count, wreq->submitted, wreq->wsize, copied, to_page_end); - - wreq->iter.count += copied; - wreq->io_iter.count += copied; - if (to_page_end && wreq->io_iter.count - wreq->submitted >= wreq->wsize) - netfs_submit_writethrough(wreq, false); - - return wreq->error; -} - -/* - * End a write operation used when writing through the pagecache. - */ -int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb) -{ - int ret = -EIOCBQUEUED; - - _enter("ic=%zu sb=%llu ws=%u", - wreq->iter.count, wreq->submitted, wreq->wsize); - - if (wreq->submitted < wreq->io_iter.count) - netfs_submit_writethrough(wreq, true); - - if (atomic_dec_and_test(&wreq->nr_outstanding)) - netfs_write_terminated(wreq, false); - - if (is_sync_kiocb(iocb)) { - wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); - ret = wreq->error; - } - - netfs_put_request(wreq, false, netfs_rreq_trace_put_return); - return ret; -} -- cgit v1.2.3-59-g8ed1b From d41ca44c20c3578154f30b07aa85ed4a951f34ab Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 20:14:32 +0000 Subject: netfs: Miscellaneous tidy ups Do a couple of miscellaneous tidy ups: (1) Add a qualifier into a file banner comment. (2) Put the writeback folio traces back into alphabetical order. (3) Remove some unused folio traces. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/netfs/buffered_write.c | 2 +- include/trace/events/netfs.h | 6 +----- 2 files changed, 2 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index b21d331eaa1e..825e6632ee4f 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -1,5 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only -/* Network filesystem high-level write support. +/* Network filesystem high-level buffered write support. * * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h index e7700172ae7e..4ba553a6d71b 100644 --- a/include/trace/events/netfs.h +++ b/include/trace/events/netfs.h @@ -141,12 +141,9 @@ EM(netfs_folio_trace_cancel_copy, "cancel-copy") \ EM(netfs_folio_trace_clear, "clear") \ EM(netfs_folio_trace_clear_cc, "clear-cc") \ - EM(netfs_folio_trace_clear_s, "clear-s") \ EM(netfs_folio_trace_clear_g, "clear-g") \ - EM(netfs_folio_trace_copy, "copy") \ - EM(netfs_folio_trace_copy_plus, "copy+") \ + EM(netfs_folio_trace_clear_s, "clear-s") \ EM(netfs_folio_trace_copy_to_cache, "mark-copy") \ - EM(netfs_folio_trace_end_copy, "end-copy") \ EM(netfs_folio_trace_filled_gaps, "filled-gaps") \ EM(netfs_folio_trace_kill, "kill") \ EM(netfs_folio_trace_kill_cc, "kill-cc") \ @@ -156,7 +153,6 @@ EM(netfs_folio_trace_mkwrite_plus, "mkwrite+") \ EM(netfs_folio_trace_not_under_wback, "!wback") \ EM(netfs_folio_trace_read_gaps, "read-gaps") \ - EM(netfs_folio_trace_redirty, "redirty") \ EM(netfs_folio_trace_redirtied, "redirtied") \ EM(netfs_folio_trace_store, "store") \ EM(netfs_folio_trace_store_copy, "store-copy") \ -- cgit v1.2.3-59-g8ed1b From 1ecb146f7cd82e44277de448d4f736b98741f3cb Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 15 Mar 2024 15:15:44 +0000 Subject: netfs, afs: Use writeback retry to deal with alternate keys Use a hook in the new writeback code's retry algorithm to rotate the keys once all the outstanding subreqs have failed rather than doing it separately on each subreq. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Marc Dionne cc: linux-afs@lists.infradead.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/afs/file.c | 1 + fs/afs/internal.h | 1 + fs/afs/write.c | 191 +++++++++++++++++++++++------------------------ fs/netfs/write_collect.c | 9 ++- include/linux/netfs.h | 2 + 5 files changed, 104 insertions(+), 100 deletions(-) (limited to 'fs') diff --git a/fs/afs/file.c b/fs/afs/file.c index 8f983e3ecae7..c3f0c45ae9a9 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -368,6 +368,7 @@ static int afs_check_write_begin(struct file *file, loff_t pos, unsigned len, static void afs_free_request(struct netfs_io_request *rreq) { key_put(rreq->netfs_priv); + afs_put_wb_key(rreq->netfs_priv2); } static void afs_update_i_size(struct inode *inode, loff_t new_i_size) diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 887245f9336d..6e1d3c4daf72 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -1601,6 +1601,7 @@ extern int afs_check_volume_status(struct afs_volume *, struct afs_operation *); void afs_prepare_write(struct netfs_io_subrequest *subreq); void afs_issue_write(struct netfs_io_subrequest *subreq); void afs_begin_writeback(struct netfs_io_request *wreq); +void afs_retry_request(struct netfs_io_request *wreq, struct netfs_io_stream *stream); extern int afs_writepages(struct address_space *, struct writeback_control *); extern int afs_fsync(struct file *, loff_t, loff_t, int); extern vm_fault_t afs_page_mkwrite(struct vm_fault *vmf); diff --git a/fs/afs/write.c b/fs/afs/write.c index b8505a8b622a..e959640694c2 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -29,43 +29,39 @@ static void afs_pages_written_back(struct afs_vnode *vnode, loff_t start, unsign /* * Find a key to use for the writeback. We cached the keys used to author the - * writes on the vnode. *_wbk will contain the last writeback key used or NULL - * and we need to start from there if it's set. + * writes on the vnode. wreq->netfs_priv2 will contain the last writeback key + * record used or NULL and we need to start from there if it's set. + * wreq->netfs_priv will be set to the key itself or NULL. */ -static int afs_get_writeback_key(struct afs_vnode *vnode, - struct afs_wb_key **_wbk) +static void afs_get_writeback_key(struct netfs_io_request *wreq) { - struct afs_wb_key *wbk = NULL; - struct list_head *p; - int ret = -ENOKEY, ret2; + struct afs_wb_key *wbk, *old = wreq->netfs_priv2; + struct afs_vnode *vnode = AFS_FS_I(wreq->inode); + + key_put(wreq->netfs_priv); + wreq->netfs_priv = NULL; + wreq->netfs_priv2 = NULL; spin_lock(&vnode->wb_lock); - if (*_wbk) - p = (*_wbk)->vnode_link.next; + if (old) + wbk = list_next_entry(old, vnode_link); else - p = vnode->wb_keys.next; + wbk = list_first_entry(&vnode->wb_keys, struct afs_wb_key, vnode_link); - while (p != &vnode->wb_keys) { - wbk = list_entry(p, struct afs_wb_key, vnode_link); + list_for_each_entry_from(wbk, &vnode->wb_keys, vnode_link) { _debug("wbk %u", key_serial(wbk->key)); - ret2 = key_validate(wbk->key); - if (ret2 == 0) { + if (key_validate(wbk->key) == 0) { refcount_inc(&wbk->usage); + wreq->netfs_priv = key_get(wbk->key); + wreq->netfs_priv2 = wbk; _debug("USE WB KEY %u", key_serial(wbk->key)); break; } - - wbk = NULL; - if (ret == -ENOKEY) - ret = ret2; - p = p->next; } spin_unlock(&vnode->wb_lock); - if (*_wbk) - afs_put_wb_key(*_wbk); - *_wbk = wbk; - return 0; + + afs_put_wb_key(old); } static void afs_store_data_success(struct afs_operation *op) @@ -88,72 +84,91 @@ static const struct afs_operation_ops afs_store_data_operation = { }; /* - * write to a file + * Prepare a subrequest to write to the server. This sets the max_len + * parameter. */ -static int afs_store_data(struct afs_vnode *vnode, struct iov_iter *iter, loff_t pos) +void afs_prepare_write(struct netfs_io_subrequest *subreq) { + //if (test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) + // subreq->max_len = 512 * 1024; + //else + subreq->max_len = 256 * 1024 * 1024; +} + +/* + * Issue a subrequest to write to the server. + */ +static void afs_issue_write_worker(struct work_struct *work) +{ + struct netfs_io_subrequest *subreq = container_of(work, struct netfs_io_subrequest, work); + struct netfs_io_request *wreq = subreq->rreq; struct afs_operation *op; - struct afs_wb_key *wbk = NULL; - loff_t size = iov_iter_count(iter); + struct afs_vnode *vnode = AFS_FS_I(wreq->inode); + unsigned long long pos = subreq->start + subreq->transferred; + size_t len = subreq->len - subreq->transferred; int ret = -ENOKEY; - _enter("%s{%llx:%llu.%u},%llx,%llx", + _enter("R=%x[%x],%s{%llx:%llu.%u},%llx,%zx", + wreq->debug_id, subreq->debug_index, vnode->volume->name, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, - size, pos); + pos, len); - ret = afs_get_writeback_key(vnode, &wbk); - if (ret) { - _leave(" = %d [no keys]", ret); - return ret; - } +#if 0 // Error injection + if (subreq->debug_index == 3) + return netfs_write_subrequest_terminated(subreq, -ENOANO, false); - op = afs_alloc_operation(wbk->key, vnode->volume); - if (IS_ERR(op)) { - afs_put_wb_key(wbk); - return -ENOMEM; + if (!test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) { + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + return netfs_write_subrequest_terminated(subreq, -EAGAIN, false); } +#endif + + op = afs_alloc_operation(wreq->netfs_priv, vnode->volume); + if (IS_ERR(op)) + return netfs_write_subrequest_terminated(subreq, -EAGAIN, false); afs_op_set_vnode(op, 0, vnode); - op->file[0].dv_delta = 1; + op->file[0].dv_delta = 1; op->file[0].modification = true; - op->store.pos = pos; - op->store.size = size; - op->flags |= AFS_OPERATION_UNINTR; - op->ops = &afs_store_data_operation; + op->store.pos = pos; + op->store.size = len; + op->flags |= AFS_OPERATION_UNINTR; + op->ops = &afs_store_data_operation; -try_next_key: afs_begin_vnode_operation(op); - op->store.write_iter = iter; - op->store.i_size = max(pos + size, vnode->netfs.remote_i_size); - op->mtime = inode_get_mtime(&vnode->netfs.inode); + op->store.write_iter = &subreq->io_iter; + op->store.i_size = umax(pos + len, vnode->netfs.remote_i_size); + op->mtime = inode_get_mtime(&vnode->netfs.inode); afs_wait_for_operation(op); - - switch (afs_op_error(op)) { + ret = afs_put_operation(op); + switch (ret) { case -EACCES: case -EPERM: case -ENOKEY: case -EKEYEXPIRED: case -EKEYREJECTED: case -EKEYREVOKED: - _debug("next"); - - ret = afs_get_writeback_key(vnode, &wbk); - if (ret == 0) { - key_put(op->key); - op->key = key_get(wbk->key); - goto try_next_key; - } + /* If there are more keys we can try, use the retry algorithm + * to rotate the keys. + */ + if (wreq->netfs_priv2) + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); break; } - afs_put_wb_key(wbk); - _leave(" = %d", afs_op_error(op)); - return afs_put_operation(op); + netfs_write_subrequest_terminated(subreq, ret < 0 ? ret : subreq->len, false); +} + +void afs_issue_write(struct netfs_io_subrequest *subreq) +{ + subreq->work.func = afs_issue_write_worker; + if (!queue_work(system_unbound_wq, &subreq->work)) + WARN_ON_ONCE(1); } /* @@ -162,52 +177,32 @@ try_next_key: */ void afs_begin_writeback(struct netfs_io_request *wreq) { + afs_get_writeback_key(wreq); wreq->io_streams[0].avail = true; } /* - * Prepare a subrequest to write to the server. This sets the max_len - * parameter. - */ -void afs_prepare_write(struct netfs_io_subrequest *subreq) -{ - //if (test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) - // subreq->max_len = 512 * 1024; - //else - subreq->max_len = 256 * 1024 * 1024; -} - -/* - * Issue a subrequest to write to the server. + * Prepare to retry the writes in request. Use this to try rotating the + * available writeback keys. */ -static void afs_issue_write_worker(struct work_struct *work) +void afs_retry_request(struct netfs_io_request *wreq, struct netfs_io_stream *stream) { - struct netfs_io_subrequest *subreq = container_of(work, struct netfs_io_subrequest, work); - struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode); - ssize_t ret; - - _enter("%x[%x],%zx", - subreq->rreq->debug_id, subreq->debug_index, subreq->io_iter.count); - -#if 0 // Error injection - if (subreq->debug_index == 3) - return netfs_write_subrequest_terminated(subreq, -ENOANO, false); + struct netfs_io_subrequest *subreq = + list_first_entry(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); - if (!test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) { - set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); - return netfs_write_subrequest_terminated(subreq, -EAGAIN, false); + switch (subreq->error) { + case -EACCES: + case -EPERM: + case -ENOKEY: + case -EKEYEXPIRED: + case -EKEYREJECTED: + case -EKEYREVOKED: + afs_get_writeback_key(wreq); + if (!wreq->netfs_priv) + stream->failed = true; + break; } -#endif - - ret = afs_store_data(vnode, &subreq->io_iter, subreq->start); - netfs_write_subrequest_terminated(subreq, ret < 0 ? ret : subreq->len, false); -} - -void afs_issue_write(struct netfs_io_subrequest *subreq) -{ - subreq->work.func = afs_issue_write_worker; - if (!queue_work(system_unbound_wq, &subreq->work)) - WARN_ON_ONCE(1); } /* diff --git a/fs/netfs/write_collect.c b/fs/netfs/write_collect.c index f14c08bf605d..60112e4b2c5e 100644 --- a/fs/netfs/write_collect.c +++ b/fs/netfs/write_collect.c @@ -163,6 +163,13 @@ static void netfs_retry_write_stream(struct netfs_io_request *wreq, _enter("R=%x[%x:]", wreq->debug_id, stream->stream_nr); + if (list_empty(&stream->subrequests)) + return; + + if (stream->source == NETFS_UPLOAD_TO_SERVER && + wreq->netfs_ops->retry_request) + wreq->netfs_ops->retry_request(wreq, stream); + if (unlikely(stream->failed)) return; @@ -182,8 +189,6 @@ static void netfs_retry_write_stream(struct netfs_io_request *wreq, return; } - if (list_empty(&stream->subrequests)) - return; next = stream->subrequests.next; do { diff --git a/include/linux/netfs.h b/include/linux/netfs.h index c2ba364041b0..298552f5122c 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -235,6 +235,7 @@ struct netfs_io_request { struct iov_iter iter; /* Unencrypted-side iterator */ struct iov_iter io_iter; /* I/O (Encrypted-side) iterator */ void *netfs_priv; /* Private data for the netfs */ + void *netfs_priv2; /* Private data for the netfs */ struct bio_vec *direct_bv; /* DIO buffer list (when handling iovec-iter) */ unsigned int direct_bv_count; /* Number of elements in direct_bv[] */ unsigned int debug_id; @@ -306,6 +307,7 @@ struct netfs_request_ops { void (*begin_writeback)(struct netfs_io_request *wreq); void (*prepare_write)(struct netfs_io_subrequest *subreq); void (*issue_write)(struct netfs_io_subrequest *subreq); + void (*retry_request)(struct netfs_io_request *wreq, struct netfs_io_stream *stream); void (*invalidate_cache)(struct netfs_io_request *wreq); }; -- cgit v1.2.3-59-g8ed1b From 0f7c0f3f51501a472a08d16315903d17012325ca Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Mar 2024 08:51:38 +0000 Subject: cifs: Use alternative invalidation to using launder_folio Use writepages-based flushing invalidation instead of invalidate_inode_pages2() and ->launder_folio(). This will allow ->launder_folio() to be removed eventually. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/smb/client/cifsfs.h | 1 - fs/smb/client/file.c | 122 ------------------------------------------------- fs/smb/client/inode.c | 25 ++-------- 3 files changed, 5 insertions(+), 143 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.h b/fs/smb/client/cifsfs.h index ca55d01117c8..1ab7e5998c58 100644 --- a/fs/smb/client/cifsfs.h +++ b/fs/smb/client/cifsfs.h @@ -69,7 +69,6 @@ extern int cifs_revalidate_file_attr(struct file *filp); extern int cifs_revalidate_dentry_attr(struct dentry *); extern int cifs_revalidate_file(struct file *filp); extern int cifs_revalidate_dentry(struct dentry *); -extern int cifs_invalidate_mapping(struct inode *inode); extern int cifs_revalidate_mapping(struct inode *inode); extern int cifs_zap_mapping(struct inode *inode); extern int cifs_getattr(struct mnt_idmap *, const struct path *, diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index b39caae652f6..2d5488a67965 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -2655,64 +2655,6 @@ struct cifs_writedata *cifs_writedata_alloc(work_func_t complete) return wdata; } -static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) -{ - struct address_space *mapping = page->mapping; - loff_t offset = (loff_t)page->index << PAGE_SHIFT; - char *write_data; - int rc = -EFAULT; - int bytes_written = 0; - struct inode *inode; - struct cifsFileInfo *open_file; - - if (!mapping || !mapping->host) - return -EFAULT; - - inode = page->mapping->host; - - offset += (loff_t)from; - write_data = kmap(page); - write_data += from; - - if ((to > PAGE_SIZE) || (from > to)) { - kunmap(page); - return -EIO; - } - - /* racing with truncate? */ - if (offset > mapping->host->i_size) { - kunmap(page); - return 0; /* don't care */ - } - - /* check to make sure that we are not extending the file */ - if (mapping->host->i_size - offset < (loff_t)to) - to = (unsigned)(mapping->host->i_size - offset); - - rc = cifs_get_writable_file(CIFS_I(mapping->host), FIND_WR_ANY, - &open_file); - if (!rc) { - bytes_written = cifs_write(open_file, open_file->pid, - write_data, to - from, &offset); - cifsFileInfo_put(open_file); - /* Does mm or vfs already set times? */ - simple_inode_init_ts(inode); - if ((bytes_written > 0) && (offset)) - rc = 0; - else if (bytes_written < 0) - rc = bytes_written; - else - rc = -EFAULT; - } else { - cifs_dbg(FYI, "No writable handle for write page rc=%d\n", rc); - if (!is_retryable_error(rc)) - rc = -EIO; - } - - kunmap(page); - return rc; -} - /* * Extend the region to be written back to include subsequent contiguously * dirty pages if possible, but don't sleep while doing so. @@ -3126,47 +3068,6 @@ out: return ret; } -static int -cifs_writepage_locked(struct page *page, struct writeback_control *wbc) -{ - int rc; - unsigned int xid; - - xid = get_xid(); -/* BB add check for wbc flags */ - get_page(page); - if (!PageUptodate(page)) - cifs_dbg(FYI, "ppw - page not up to date\n"); - - /* - * Set the "writeback" flag, and clear "dirty" in the radix tree. - * - * A writepage() implementation always needs to do either this, - * or re-dirty the page with "redirty_page_for_writepage()" in - * the case of a failure. - * - * Just unlocking the page will cause the radix tree tag-bits - * to fail to update with the state of the page correctly. - */ - set_page_writeback(page); -retry_write: - rc = cifs_partialpagewrite(page, 0, PAGE_SIZE); - if (is_retryable_error(rc)) { - if (wbc->sync_mode == WB_SYNC_ALL && rc == -EAGAIN) - goto retry_write; - redirty_page_for_writepage(wbc, page); - } else if (rc != 0) { - SetPageError(page); - mapping_set_error(page->mapping, rc); - } else { - SetPageUptodate(page); - } - end_page_writeback(page); - put_page(page); - free_xid(xid); - return rc; -} - static int cifs_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct page *page, void *fsdata) @@ -4992,27 +4893,6 @@ static void cifs_invalidate_folio(struct folio *folio, size_t offset, folio_wait_private_2(folio); /* [DEPRECATED] */ } -static int cifs_launder_folio(struct folio *folio) -{ - int rc = 0; - loff_t range_start = folio_pos(folio); - loff_t range_end = range_start + folio_size(folio); - struct writeback_control wbc = { - .sync_mode = WB_SYNC_ALL, - .nr_to_write = 0, - .range_start = range_start, - .range_end = range_end, - }; - - cifs_dbg(FYI, "Launder page: %lu\n", folio->index); - - if (folio_clear_dirty_for_io(folio)) - rc = cifs_writepage_locked(&folio->page, &wbc); - - folio_wait_private_2(folio); /* [DEPRECATED] */ - return rc; -} - void cifs_oplock_break(struct work_struct *work) { struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo, @@ -5191,7 +5071,6 @@ const struct address_space_operations cifs_addr_ops = { .release_folio = cifs_release_folio, .direct_IO = cifs_direct_io, .invalidate_folio = cifs_invalidate_folio, - .launder_folio = cifs_launder_folio, .migrate_folio = filemap_migrate_folio, /* * TODO: investigate and if useful we could add an is_dirty_writeback @@ -5214,6 +5093,5 @@ const struct address_space_operations cifs_addr_ops_smallbuf = { .dirty_folio = netfs_dirty_folio, .release_folio = cifs_release_folio, .invalidate_folio = cifs_invalidate_folio, - .launder_folio = cifs_launder_folio, .migrate_folio = filemap_migrate_folio, }; diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index 60afab5c83d4..033948d34c5a 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -2431,24 +2431,6 @@ cifs_dentry_needs_reval(struct dentry *dentry) return false; } -/* - * Zap the cache. Called when invalid_mapping flag is set. - */ -int -cifs_invalidate_mapping(struct inode *inode) -{ - int rc = 0; - - if (inode->i_mapping && inode->i_mapping->nrpages != 0) { - rc = invalidate_inode_pages2(inode->i_mapping); - if (rc) - cifs_dbg(VFS, "%s: invalidate inode %p failed with rc %d\n", - __func__, inode, rc); - } - - return rc; -} - /** * cifs_wait_bit_killable - helper for functions that are sleeping on bit locks * @@ -2485,9 +2467,12 @@ cifs_revalidate_mapping(struct inode *inode) if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE) goto skip_invalidate; - rc = cifs_invalidate_mapping(inode); - if (rc) + rc = filemap_invalidate_inode(inode, true, 0, LLONG_MAX); + if (rc) { + cifs_dbg(VFS, "%s: invalidate inode %p failed with rc %d\n", + __func__, inode, rc); set_bit(CIFS_INO_INVALID_MAPPING, flags); + } } skip_invalidate: -- cgit v1.2.3-59-g8ed1b From 753b67eb630db34e36ec4ae1e86c75e243ea4fc9 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 9 Mar 2022 11:01:12 +0000 Subject: cifs: Replace cifs_readdata with a wrapper around netfs_io_subrequest Netfslib has a facility whereby the allocation for netfs_io_subrequest can be increased to so that filesystem-specific data can be tagged on the end. Prepare to use this by making a struct, cifs_io_subrequest, that wraps netfs_io_subrequest, and absorb struct cifs_readdata into it. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/cifsglob.h | 22 ++++++++++++---------- fs/smb/client/cifsproto.h | 13 +++++++++++-- fs/smb/client/cifssmb.c | 11 ++++------- fs/smb/client/file.c | 44 ++++++++++++++++++++------------------------ fs/smb/client/smb2ops.c | 2 +- fs/smb/client/smb2pdu.c | 12 +++++++----- fs/smb/client/smb2proto.h | 2 +- fs/smb/client/transport.c | 4 ++-- 8 files changed, 58 insertions(+), 52 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 6ff35570db81..405110052e17 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -268,7 +268,7 @@ struct dfs_info3_param; struct cifs_fattr; struct smb3_fs_context; struct cifs_fid; -struct cifs_readdata; +struct cifs_io_subrequest; struct cifs_writedata; struct cifs_io_parms; struct cifs_search_info; @@ -450,7 +450,7 @@ struct smb_version_operations { /* send a flush request to the server */ int (*flush)(const unsigned int, struct cifs_tcon *, struct cifs_fid *); /* async read from the server */ - int (*async_readv)(struct cifs_readdata *); + int (*async_readv)(struct cifs_io_subrequest *); /* async write to the server */ int (*async_writev)(struct cifs_writedata *, void (*release)(struct kref *)); @@ -1493,26 +1493,28 @@ struct cifs_aio_ctx { }; /* asynchronous read support */ -struct cifs_readdata { - struct kref refcount; - struct list_head list; - struct completion done; +struct cifs_io_subrequest { + struct netfs_io_subrequest subreq; struct cifsFileInfo *cfile; struct address_space *mapping; struct cifs_aio_ctx *ctx; - __u64 offset; ssize_t got_bytes; - unsigned int bytes; pid_t pid; int result; - struct work_struct work; - struct iov_iter iter; struct kvec iov[2]; struct TCP_Server_Info *server; #ifdef CONFIG_CIFS_SMB_DIRECT struct smbd_mr *mr; #endif struct cifs_credits credits; + + // TODO: Remove following elements + struct list_head list; + struct completion done; + struct work_struct work; + struct iov_iter iter; + __u64 offset; + unsigned int bytes; }; /* asynchronous write support */ diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index fbc358c09da3..5712f789eaa0 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -599,8 +599,17 @@ void __cifs_put_smb_ses(struct cifs_ses *ses); extern struct cifs_ses * cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx); -void cifs_readdata_release(struct kref *refcount); -int cifs_async_readv(struct cifs_readdata *rdata); +void cifs_readdata_release(struct cifs_io_subrequest *rdata); +static inline void cifs_get_readdata(struct cifs_io_subrequest *rdata) +{ + refcount_inc(&rdata->subreq.ref); +} +static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata) +{ + if (refcount_dec_and_test(&rdata->subreq.ref)) + cifs_readdata_release(rdata); +} +int cifs_async_readv(struct cifs_io_subrequest *rdata); int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid); int cifs_async_writev(struct cifs_writedata *wdata, diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index 23b5709ddc31..2aaeae6c5766 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -24,6 +24,8 @@ #include #include #include +#include +#include #include "cifspdu.h" #include "cifsfs.h" #include "cifsglob.h" @@ -1262,12 +1264,11 @@ openRetry: static void cifs_readv_callback(struct mid_q_entry *mid) { - struct cifs_readdata *rdata = mid->callback_data; + struct cifs_io_subrequest *rdata = mid->callback_data; struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); struct TCP_Server_Info *server = tcon->ses->server; struct smb_rqst rqst = { .rq_iov = rdata->iov, .rq_nvec = 2, - .rq_iter_size = iov_iter_count(&rdata->iter), .rq_iter = rdata->iter }; struct cifs_credits credits = { .value = 1, .instance = 0 }; @@ -1312,7 +1313,7 @@ cifs_readv_callback(struct mid_q_entry *mid) /* cifs_async_readv - send an async write, and set up mid to handle result */ int -cifs_async_readv(struct cifs_readdata *rdata) +cifs_async_readv(struct cifs_io_subrequest *rdata) { int rc; READ_REQ *smb = NULL; @@ -1364,15 +1365,11 @@ cifs_async_readv(struct cifs_readdata *rdata) rdata->iov[1].iov_base = (char *)smb + 4; rdata->iov[1].iov_len = get_rfc1002_length(smb); - kref_get(&rdata->refcount); rc = cifs_call_async(tcon->ses->server, &rqst, cifs_readv_receive, cifs_readv_callback, NULL, rdata, 0, NULL); if (rc == 0) cifs_stats_inc(&tcon->stats.cifs_stats.num_reads); - else - kref_put(&rdata->refcount, cifs_readdata_release); - cifs_small_buf_release(smb); return rc; } diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 2d5488a67965..0bea058099d4 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -3822,13 +3822,13 @@ out: return written; } -static struct cifs_readdata *cifs_readdata_alloc(work_func_t complete) +static struct cifs_io_subrequest *cifs_readdata_alloc(work_func_t complete) { - struct cifs_readdata *rdata; + struct cifs_io_subrequest *rdata; rdata = kzalloc(sizeof(*rdata), GFP_KERNEL); if (rdata) { - kref_init(&rdata->refcount); + refcount_set(&rdata->subreq.ref, 1); INIT_LIST_HEAD(&rdata->list); init_completion(&rdata->done); INIT_WORK(&rdata->work, complete); @@ -3838,11 +3838,8 @@ static struct cifs_readdata *cifs_readdata_alloc(work_func_t complete) } void -cifs_readdata_release(struct kref *refcount) +cifs_readdata_release(struct cifs_io_subrequest *rdata) { - struct cifs_readdata *rdata = container_of(refcount, - struct cifs_readdata, refcount); - if (rdata->ctx) kref_put(&rdata->ctx->refcount, cifs_aio_ctx_release); #ifdef CONFIG_CIFS_SMB_DIRECT @@ -3862,16 +3859,16 @@ static void collect_uncached_read_data(struct cifs_aio_ctx *ctx); static void cifs_uncached_readv_complete(struct work_struct *work) { - struct cifs_readdata *rdata = container_of(work, - struct cifs_readdata, work); + struct cifs_io_subrequest *rdata = + container_of(work, struct cifs_io_subrequest, work); complete(&rdata->done); collect_uncached_read_data(rdata->ctx); /* the below call can possibly free the last ref to aio ctx */ - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); } -static int cifs_resend_rdata(struct cifs_readdata *rdata, +static int cifs_resend_rdata(struct cifs_io_subrequest *rdata, struct list_head *rdata_list, struct cifs_aio_ctx *ctx) { @@ -3939,7 +3936,7 @@ static int cifs_resend_rdata(struct cifs_readdata *rdata, } while (rc == -EAGAIN); fail: - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); return rc; } @@ -3948,7 +3945,7 @@ cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file, struct cifs_sb_info *cifs_sb, struct list_head *rdata_list, struct cifs_aio_ctx *ctx) { - struct cifs_readdata *rdata; + struct cifs_io_subrequest *rdata; unsigned int rsize, nsegs, max_segs = INT_MAX; struct cifs_credits credits_on_stack; struct cifs_credits *credits = &credits_on_stack; @@ -4030,7 +4027,7 @@ cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file, if (rc) { add_credits_and_wake_if(server, &rdata->credits, 0); - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); if (rc == -EAGAIN) continue; break; @@ -4048,7 +4045,7 @@ cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file, static void collect_uncached_read_data(struct cifs_aio_ctx *ctx) { - struct cifs_readdata *rdata, *tmp; + struct cifs_io_subrequest *rdata, *tmp; struct cifs_sb_info *cifs_sb; int rc; @@ -4094,8 +4091,7 @@ again: rdata->cfile, cifs_sb, &tmp_list, ctx); - kref_put(&rdata->refcount, - cifs_readdata_release); + cifs_put_readdata(rdata); } list_splice(&tmp_list, &ctx->list); @@ -4111,7 +4107,7 @@ again: ctx->total_len += rdata->got_bytes; } list_del_init(&rdata->list); - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); } /* mask nodata case */ @@ -4483,8 +4479,8 @@ static void cifs_unlock_folios(struct address_space *mapping, pgoff_t first, pgo static void cifs_readahead_complete(struct work_struct *work) { - struct cifs_readdata *rdata = container_of(work, - struct cifs_readdata, work); + struct cifs_io_subrequest *rdata = container_of(work, + struct cifs_io_subrequest, work); struct folio *folio; pgoff_t last; bool good = rdata->result == 0 || (rdata->result == -EAGAIN && rdata->got_bytes); @@ -4510,7 +4506,7 @@ static void cifs_readahead_complete(struct work_struct *work) } rcu_read_unlock(); - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); } static void cifs_readahead(struct readahead_control *ractl) @@ -4550,7 +4546,7 @@ static void cifs_readahead(struct readahead_control *ractl) */ while ((nr_pages = ra_pages)) { unsigned int i, rsize; - struct cifs_readdata *rdata; + struct cifs_io_subrequest *rdata; struct cifs_credits credits_on_stack; struct cifs_credits *credits = &credits_on_stack; struct folio *folio; @@ -4669,11 +4665,11 @@ static void cifs_readahead(struct readahead_control *ractl) rdata->offset / PAGE_SIZE, (rdata->offset + rdata->bytes - 1) / PAGE_SIZE); /* Fallback to the readpage in error/reconnect cases */ - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); break; } - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); } free_xid(xid); diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 28f0b7d19d53..56026f4cbc13 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -4490,7 +4490,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid, unsigned int cur_off; unsigned int cur_page_idx; unsigned int pad_len; - struct cifs_readdata *rdata = mid->callback_data; + struct cifs_io_subrequest *rdata = mid->callback_data; struct smb2_hdr *shdr = (struct smb2_hdr *)buf; int length; bool use_rdma_mr = false; diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index a5efce03cb58..0957dd4c07e7 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -23,6 +23,8 @@ #include #include #include +#include +#include #include "cifsglob.h" #include "cifsacl.h" #include "cifsproto.h" @@ -4391,7 +4393,7 @@ static inline bool smb3_use_rdma_offload(struct cifs_io_parms *io_parms) */ static int smb2_new_read_req(void **buf, unsigned int *total_len, - struct cifs_io_parms *io_parms, struct cifs_readdata *rdata, + struct cifs_io_parms *io_parms, struct cifs_io_subrequest *rdata, unsigned int remaining_bytes, int request_type) { int rc = -EACCES; @@ -4483,7 +4485,7 @@ smb2_new_read_req(void **buf, unsigned int *total_len, static void smb2_readv_callback(struct mid_q_entry *mid) { - struct cifs_readdata *rdata = mid->callback_data; + struct cifs_io_subrequest *rdata = mid->callback_data; struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); struct TCP_Server_Info *server = rdata->server; struct smb2_hdr *shdr = @@ -4570,7 +4572,7 @@ smb2_readv_callback(struct mid_q_entry *mid) /* smb2_async_readv - send an async read, and set up mid to handle result */ int -smb2_async_readv(struct cifs_readdata *rdata) +smb2_async_readv(struct cifs_io_subrequest *rdata) { int rc, flags = 0; char *buf; @@ -4628,13 +4630,13 @@ smb2_async_readv(struct cifs_readdata *rdata) flags |= CIFS_HAS_CREDITS; } - kref_get(&rdata->refcount); + cifs_get_readdata(rdata); rc = cifs_call_async(server, &rqst, cifs_readv_receive, smb2_readv_callback, smb3_handle_read_data, rdata, flags, &rdata->credits); if (rc) { - kref_put(&rdata->refcount, cifs_readdata_release); + cifs_put_readdata(rdata); cifs_stats_fail_inc(io_parms.tcon, SMB2_READ_HE); trace_smb3_read_err(0 /* xid */, io_parms.persistent_fid, io_parms.tcon->tid, diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h index 732169d8a67a..740053c0118e 100644 --- a/fs/smb/client/smb2proto.h +++ b/fs/smb/client/smb2proto.h @@ -210,7 +210,7 @@ extern int SMB2_query_acl(const unsigned int xid, struct cifs_tcon *tcon, extern int SMB2_get_srv_num(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, u64 volatile_fid, __le64 *uniqueid); -extern int smb2_async_readv(struct cifs_readdata *rdata); +extern int smb2_async_readv(struct cifs_io_subrequest *rdata); extern int SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, char **buf, int *buf_type); extern int smb2_async_writev(struct cifs_writedata *wdata, diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index ddf1a3aafee5..dcc10a105b56 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -1692,7 +1692,7 @@ __cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid, static int cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid) { - struct cifs_readdata *rdata = mid->callback_data; + struct cifs_io_subrequest *rdata = mid->callback_data; return __cifs_readv_discard(server, mid, rdata->result); } @@ -1702,7 +1702,7 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) { int length, len; unsigned int data_offset, data_len; - struct cifs_readdata *rdata = mid->callback_data; + struct cifs_io_subrequest *rdata = mid->callback_data; char *buf = server->smallbuf; unsigned int buflen = server->pdu_size + HEADER_PREAMBLE_SIZE(server); bool use_rdma_mr = false; -- cgit v1.2.3-59-g8ed1b From a975a2f22cdce7ec0c678ce8d73d2f6616cb281c Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 16:27:41 +0100 Subject: cifs: Replace cifs_writedata with a wrapper around netfs_io_subrequest Replace the cifs_writedata struct with the same wrapper around netfs_io_subrequest that was used to replace cifs_readdata. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/cifsglob.h | 32 ++++--------------- fs/smb/client/cifsproto.h | 16 +++++++--- fs/smb/client/cifssmb.c | 9 +++--- fs/smb/client/file.c | 79 +++++++++++++++++++---------------------------- fs/smb/client/smb2pdu.c | 9 +++--- fs/smb/client/smb2proto.h | 3 +- 6 files changed, 59 insertions(+), 89 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 405110052e17..7c07cf8cc29d 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -269,7 +269,6 @@ struct cifs_fattr; struct smb3_fs_context; struct cifs_fid; struct cifs_io_subrequest; -struct cifs_writedata; struct cifs_io_parms; struct cifs_search_info; struct cifsInodeInfo; @@ -452,8 +451,7 @@ struct smb_version_operations { /* async read from the server */ int (*async_readv)(struct cifs_io_subrequest *); /* async write to the server */ - int (*async_writev)(struct cifs_writedata *, - void (*release)(struct kref *)); + int (*async_writev)(struct cifs_io_subrequest *); /* sync read from the server */ int (*sync_read)(const unsigned int, struct cifs_fid *, struct cifs_io_parms *, unsigned int *, char **, @@ -1508,36 +1506,18 @@ struct cifs_io_subrequest { #endif struct cifs_credits credits; - // TODO: Remove following elements - struct list_head list; - struct completion done; - struct work_struct work; - struct iov_iter iter; - __u64 offset; - unsigned int bytes; -}; + enum writeback_sync_modes sync_mode; + bool uncached; + bool replay; + struct bio_vec *bv; -/* asynchronous write support */ -struct cifs_writedata { - struct kref refcount; + // TODO: Remove following elements struct list_head list; struct completion done; - enum writeback_sync_modes sync_mode; struct work_struct work; - struct cifsFileInfo *cfile; - struct cifs_aio_ctx *ctx; struct iov_iter iter; - struct bio_vec *bv; __u64 offset; - pid_t pid; unsigned int bytes; - int result; - struct TCP_Server_Info *server; -#ifdef CONFIG_CIFS_SMB_DIRECT - struct smbd_mr *mr; -#endif - struct cifs_credits credits; - bool replay; }; /* diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index 5712f789eaa0..cb5cc5711a84 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -612,11 +612,19 @@ static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata) int cifs_async_readv(struct cifs_io_subrequest *rdata); int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid); -int cifs_async_writev(struct cifs_writedata *wdata, - void (*release)(struct kref *kref)); +int cifs_async_writev(struct cifs_io_subrequest *wdata); void cifs_writev_complete(struct work_struct *work); -struct cifs_writedata *cifs_writedata_alloc(work_func_t complete); -void cifs_writedata_release(struct kref *refcount); +struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete); +void cifs_writedata_release(struct cifs_io_subrequest *rdata); +static inline void cifs_get_writedata(struct cifs_io_subrequest *wdata) +{ + refcount_inc(&wdata->subreq.ref); +} +static inline void cifs_put_writedata(struct cifs_io_subrequest *wdata) +{ + if (refcount_dec_and_test(&wdata->subreq.ref)) + cifs_writedata_release(wdata); +} int cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, const unsigned char *path, char *pbuf, diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index 2aaeae6c5766..30d24b4fbdf6 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -1612,7 +1612,7 @@ CIFSSMBWrite(const unsigned int xid, struct cifs_io_parms *io_parms, static void cifs_writev_callback(struct mid_q_entry *mid) { - struct cifs_writedata *wdata = mid->callback_data; + struct cifs_io_subrequest *wdata = mid->callback_data; struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); unsigned int written; WRITE_RSP *smb = (WRITE_RSP *)mid->resp_buf; @@ -1657,8 +1657,7 @@ cifs_writev_callback(struct mid_q_entry *mid) /* cifs_async_writev - send an async write, and set up mid to handle result */ int -cifs_async_writev(struct cifs_writedata *wdata, - void (*release)(struct kref *kref)) +cifs_async_writev(struct cifs_io_subrequest *wdata) { int rc = -EACCES; WRITE_REQ *smb = NULL; @@ -1725,14 +1724,14 @@ cifs_async_writev(struct cifs_writedata *wdata, iov[1].iov_len += 4; /* pad bigger by four bytes */ } - kref_get(&wdata->refcount); + cifs_get_writedata(wdata); rc = cifs_call_async(tcon->ses->server, &rqst, NULL, cifs_writev_callback, NULL, wdata, 0, NULL); if (rc == 0) cifs_stats_inc(&tcon->stats.cifs_stats.num_writes); else - kref_put(&wdata->refcount, release); + cifs_put_writedata(wdata); async_writev_out: cifs_small_buf_release(smb); diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 0bea058099d4..2a0982b2d577 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -2510,10 +2510,10 @@ cifs_get_readable_path(struct cifs_tcon *tcon, const char *name, } void -cifs_writedata_release(struct kref *refcount) +cifs_writedata_release(struct cifs_io_subrequest *wdata) { - struct cifs_writedata *wdata = container_of(refcount, - struct cifs_writedata, refcount); + if (wdata->uncached) + kref_put(&wdata->ctx->refcount, cifs_aio_ctx_release); #ifdef CONFIG_CIFS_SMB_DIRECT if (wdata->mr) { smbd_deregister_mr(wdata->mr); @@ -2532,7 +2532,7 @@ cifs_writedata_release(struct kref *refcount) * possible that the page was redirtied so re-clean the page. */ static void -cifs_writev_requeue(struct cifs_writedata *wdata) +cifs_writev_requeue(struct cifs_io_subrequest *wdata) { int rc = 0; struct inode *inode = d_inode(wdata->cfile->dentry); @@ -2542,7 +2542,7 @@ cifs_writev_requeue(struct cifs_writedata *wdata) server = tlink_tcon(wdata->cfile->tlink)->ses->server; do { - struct cifs_writedata *wdata2; + struct cifs_io_subrequest *wdata2; unsigned int wsize, cur_len; wsize = server->ops->wp_retry_size(inode); @@ -2565,7 +2565,7 @@ cifs_writev_requeue(struct cifs_writedata *wdata) wdata2->sync_mode = wdata->sync_mode; wdata2->offset = fpos; wdata2->bytes = cur_len; - wdata2->iter = wdata->iter; + wdata2->iter = wdata->iter; iov_iter_advance(&wdata2->iter, fpos - wdata->offset); iov_iter_truncate(&wdata2->iter, wdata2->bytes); @@ -2587,11 +2587,10 @@ cifs_writev_requeue(struct cifs_writedata *wdata) rc = -EBADF; } else { wdata2->pid = wdata2->cfile->pid; - rc = server->ops->async_writev(wdata2, - cifs_writedata_release); + rc = server->ops->async_writev(wdata2); } - kref_put(&wdata2->refcount, cifs_writedata_release); + cifs_put_writedata(wdata2); if (rc) { if (is_retryable_error(rc)) continue; @@ -2610,14 +2609,14 @@ cifs_writev_requeue(struct cifs_writedata *wdata) if (rc != 0 && !is_retryable_error(rc)) mapping_set_error(inode->i_mapping, rc); - kref_put(&wdata->refcount, cifs_writedata_release); + cifs_put_writedata(wdata); } void cifs_writev_complete(struct work_struct *work) { - struct cifs_writedata *wdata = container_of(work, - struct cifs_writedata, work); + struct cifs_io_subrequest *wdata = container_of(work, + struct cifs_io_subrequest, work); struct inode *inode = d_inode(wdata->cfile->dentry); if (wdata->result == 0) { @@ -2638,16 +2637,16 @@ cifs_writev_complete(struct work_struct *work) if (wdata->result != -EAGAIN) mapping_set_error(inode->i_mapping, wdata->result); - kref_put(&wdata->refcount, cifs_writedata_release); + cifs_put_writedata(wdata); } -struct cifs_writedata *cifs_writedata_alloc(work_func_t complete) +struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete) { - struct cifs_writedata *wdata; + struct cifs_io_subrequest *wdata; wdata = kzalloc(sizeof(*wdata), GFP_NOFS); if (wdata != NULL) { - kref_init(&wdata->refcount); + refcount_set(&wdata->subreq.ref, 1); INIT_LIST_HEAD(&wdata->list); init_completion(&wdata->done); INIT_WORK(&wdata->work, complete); @@ -2778,7 +2777,7 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, { struct inode *inode = mapping->host; struct TCP_Server_Info *server; - struct cifs_writedata *wdata; + struct cifs_io_subrequest *wdata; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifs_credits credits_on_stack; struct cifs_credits *credits = &credits_on_stack; @@ -2871,10 +2870,9 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, if (wdata->cfile->invalidHandle) rc = -EAGAIN; else - rc = wdata->server->ops->async_writev(wdata, - cifs_writedata_release); + rc = wdata->server->ops->async_writev(wdata); if (rc >= 0) { - kref_put(&wdata->refcount, cifs_writedata_release); + cifs_put_writedata(wdata); goto err_close; } } else { @@ -2884,7 +2882,7 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, } err_wdata: - kref_put(&wdata->refcount, cifs_writedata_release); + cifs_put_writedata(wdata); err_uncredit: add_credits_and_wake_if(server, credits, 0); err_close: @@ -3261,23 +3259,13 @@ int cifs_flush(struct file *file, fl_owner_t id) return rc; } -static void -cifs_uncached_writedata_release(struct kref *refcount) -{ - struct cifs_writedata *wdata = container_of(refcount, - struct cifs_writedata, refcount); - - kref_put(&wdata->ctx->refcount, cifs_aio_ctx_release); - cifs_writedata_release(refcount); -} - static void collect_uncached_write_data(struct cifs_aio_ctx *ctx); static void cifs_uncached_writev_complete(struct work_struct *work) { - struct cifs_writedata *wdata = container_of(work, - struct cifs_writedata, work); + struct cifs_io_subrequest *wdata = container_of(work, + struct cifs_io_subrequest, work); struct inode *inode = d_inode(wdata->cfile->dentry); struct cifsInodeInfo *cifsi = CIFS_I(inode); @@ -3290,11 +3278,11 @@ cifs_uncached_writev_complete(struct work_struct *work) complete(&wdata->done); collect_uncached_write_data(wdata->ctx); /* the below call can possibly free the last ref to aio ctx */ - kref_put(&wdata->refcount, cifs_uncached_writedata_release); + cifs_put_writedata(wdata); } static int -cifs_resend_wdata(struct cifs_writedata *wdata, struct list_head *wdata_list, +cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list, struct cifs_aio_ctx *ctx) { unsigned int wsize; @@ -3344,8 +3332,7 @@ cifs_resend_wdata(struct cifs_writedata *wdata, struct list_head *wdata_list, wdata->mr = NULL; } #endif - rc = server->ops->async_writev(wdata, - cifs_uncached_writedata_release); + rc = server->ops->async_writev(wdata); } } @@ -3360,7 +3347,7 @@ cifs_resend_wdata(struct cifs_writedata *wdata, struct list_head *wdata_list, } while (rc == -EAGAIN); fail: - kref_put(&wdata->refcount, cifs_uncached_writedata_release); + cifs_put_writedata(wdata); return rc; } @@ -3412,7 +3399,7 @@ cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from, { int rc = 0; size_t cur_len, max_len; - struct cifs_writedata *wdata; + struct cifs_io_subrequest *wdata; pid_t pid; struct TCP_Server_Info *server; unsigned int xid, max_segs = INT_MAX; @@ -3476,6 +3463,7 @@ cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from, break; } + wdata->uncached = true; wdata->sync_mode = WB_SYNC_ALL; wdata->offset = (__u64)fpos; wdata->cfile = cifsFileInfo_get(open_file); @@ -3495,14 +3483,12 @@ cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from, if (wdata->cfile->invalidHandle) rc = -EAGAIN; else - rc = server->ops->async_writev(wdata, - cifs_uncached_writedata_release); + rc = server->ops->async_writev(wdata); } if (rc) { add_credits_and_wake_if(server, &wdata->credits, 0); - kref_put(&wdata->refcount, - cifs_uncached_writedata_release); + cifs_put_writedata(wdata); if (rc == -EAGAIN) continue; break; @@ -3520,7 +3506,7 @@ cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from, static void collect_uncached_write_data(struct cifs_aio_ctx *ctx) { - struct cifs_writedata *wdata, *tmp; + struct cifs_io_subrequest *wdata, *tmp; struct cifs_tcon *tcon; struct cifs_sb_info *cifs_sb; struct dentry *dentry = ctx->cfile->dentry; @@ -3575,8 +3561,7 @@ restart_loop: ctx->cfile, cifs_sb, &tmp_list, ctx); - kref_put(&wdata->refcount, - cifs_uncached_writedata_release); + cifs_put_writedata(wdata); } list_splice(&tmp_list, &ctx->list); @@ -3584,7 +3569,7 @@ restart_loop: } } list_del_init(&wdata->list); - kref_put(&wdata->refcount, cifs_uncached_writedata_release); + cifs_put_writedata(wdata); } cifs_stats_bytes_written(tcon, ctx->total_len); diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 0957dd4c07e7..8143611ba336 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -4737,7 +4737,7 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, static void smb2_writev_callback(struct mid_q_entry *mid) { - struct cifs_writedata *wdata = mid->callback_data; + struct cifs_io_subrequest *wdata = mid->callback_data; struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); struct TCP_Server_Info *server = wdata->server; unsigned int written; @@ -4818,8 +4818,7 @@ smb2_writev_callback(struct mid_q_entry *mid) /* smb2_async_writev - send an async write, and set up mid to handle result */ int -smb2_async_writev(struct cifs_writedata *wdata, - void (*release)(struct kref *kref)) +smb2_async_writev(struct cifs_io_subrequest *wdata) { int rc = -EACCES, flags = 0; struct smb2_write_req *req = NULL; @@ -4953,7 +4952,7 @@ smb2_async_writev(struct cifs_writedata *wdata, flags |= CIFS_HAS_CREDITS; } - kref_get(&wdata->refcount); + cifs_get_writedata(wdata); rc = cifs_call_async(server, &rqst, NULL, smb2_writev_callback, NULL, wdata, flags, &wdata->credits); @@ -4965,7 +4964,7 @@ smb2_async_writev(struct cifs_writedata *wdata, io_parms->offset, io_parms->length, rc); - kref_put(&wdata->refcount, release); + cifs_put_writedata(wdata); cifs_stats_fail_inc(tcon, SMB2_WRITE_HE); } diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h index 740053c0118e..c781eaa0cba8 100644 --- a/fs/smb/client/smb2proto.h +++ b/fs/smb/client/smb2proto.h @@ -213,8 +213,7 @@ extern int SMB2_get_srv_num(const unsigned int xid, struct cifs_tcon *tcon, extern int smb2_async_readv(struct cifs_io_subrequest *rdata); extern int SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, char **buf, int *buf_type); -extern int smb2_async_writev(struct cifs_writedata *wdata, - void (*release)(struct kref *kref)); +extern int smb2_async_writev(struct cifs_io_subrequest *wdata); extern int SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, struct kvec *iov, int n_vec); extern int SMB2_echo(struct TCP_Server_Info *server); -- cgit v1.2.3-59-g8ed1b From ab58fbdeebc7f9fe8b9bc202660eae3a10e5e678 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 16:55:20 +0100 Subject: cifs: Use more fields from netfs_io_subrequest Use more fields from netfs_io_subrequest instead of those incorporated into cifs_io_subrequest from cifs_readdata and cifs_writedata. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/cifsglob.h | 3 -- fs/smb/client/cifssmb.c | 52 ++++++++++----------- fs/smb/client/file.c | 112 +++++++++++++++++++++++----------------------- fs/smb/client/smb2ops.c | 4 +- fs/smb/client/smb2pdu.c | 52 ++++++++++----------- fs/smb/client/transport.c | 6 +-- 6 files changed, 113 insertions(+), 116 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 7c07cf8cc29d..5100aa047ed6 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1515,9 +1515,6 @@ struct cifs_io_subrequest { struct list_head list; struct completion done; struct work_struct work; - struct iov_iter iter; - __u64 offset; - unsigned int bytes; }; /* diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index 30d24b4fbdf6..99b75102ba6b 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -1269,12 +1269,12 @@ cifs_readv_callback(struct mid_q_entry *mid) struct TCP_Server_Info *server = tcon->ses->server; struct smb_rqst rqst = { .rq_iov = rdata->iov, .rq_nvec = 2, - .rq_iter = rdata->iter }; + .rq_iter = rdata->subreq.io_iter }; struct cifs_credits credits = { .value = 1, .instance = 0 }; - cifs_dbg(FYI, "%s: mid=%llu state=%d result=%d bytes=%u\n", + cifs_dbg(FYI, "%s: mid=%llu state=%d result=%d bytes=%zu\n", __func__, mid->mid, mid->mid_state, rdata->result, - rdata->bytes); + rdata->subreq.len); switch (mid->mid_state) { case MID_RESPONSE_RECEIVED: @@ -1322,14 +1322,14 @@ cifs_async_readv(struct cifs_io_subrequest *rdata) struct smb_rqst rqst = { .rq_iov = rdata->iov, .rq_nvec = 2 }; - cifs_dbg(FYI, "%s: offset=%llu bytes=%u\n", - __func__, rdata->offset, rdata->bytes); + cifs_dbg(FYI, "%s: offset=%llu bytes=%zu\n", + __func__, rdata->subreq.start, rdata->subreq.len); if (tcon->ses->capabilities & CAP_LARGE_FILES) wct = 12; else { wct = 10; /* old style read */ - if ((rdata->offset >> 32) > 0) { + if ((rdata->subreq.start >> 32) > 0) { /* can not handle this big offset for old */ return -EIO; } @@ -1344,12 +1344,12 @@ cifs_async_readv(struct cifs_io_subrequest *rdata) smb->AndXCommand = 0xFF; /* none */ smb->Fid = rdata->cfile->fid.netfid; - smb->OffsetLow = cpu_to_le32(rdata->offset & 0xFFFFFFFF); + smb->OffsetLow = cpu_to_le32(rdata->subreq.start & 0xFFFFFFFF); if (wct == 12) - smb->OffsetHigh = cpu_to_le32(rdata->offset >> 32); + smb->OffsetHigh = cpu_to_le32(rdata->subreq.start >> 32); smb->Remaining = 0; - smb->MaxCount = cpu_to_le16(rdata->bytes & 0xFFFF); - smb->MaxCountHigh = cpu_to_le32(rdata->bytes >> 16); + smb->MaxCount = cpu_to_le16(rdata->subreq.len & 0xFFFF); + smb->MaxCountHigh = cpu_to_le32(rdata->subreq.len >> 16); if (wct == 12) smb->ByteCount = 0; else { @@ -1633,13 +1633,13 @@ cifs_writev_callback(struct mid_q_entry *mid) * client. OS/2 servers are known to set incorrect * CountHigh values. */ - if (written > wdata->bytes) + if (written > wdata->subreq.len) written &= 0xFFFF; - if (written < wdata->bytes) + if (written < wdata->subreq.len) wdata->result = -ENOSPC; else - wdata->bytes = written; + wdata->subreq.len = written; break; case MID_REQUEST_SUBMITTED: case MID_RETRY_NEEDED: @@ -1670,7 +1670,7 @@ cifs_async_writev(struct cifs_io_subrequest *wdata) wct = 14; } else { wct = 12; - if (wdata->offset >> 32 > 0) { + if (wdata->subreq.start >> 32 > 0) { /* can not handle big offset for old srv */ return -EIO; } @@ -1685,9 +1685,9 @@ cifs_async_writev(struct cifs_io_subrequest *wdata) smb->AndXCommand = 0xFF; /* none */ smb->Fid = wdata->cfile->fid.netfid; - smb->OffsetLow = cpu_to_le32(wdata->offset & 0xFFFFFFFF); + smb->OffsetLow = cpu_to_le32(wdata->subreq.start & 0xFFFFFFFF); if (wct == 14) - smb->OffsetHigh = cpu_to_le32(wdata->offset >> 32); + smb->OffsetHigh = cpu_to_le32(wdata->subreq.start >> 32); smb->Reserved = 0xFFFFFFFF; smb->WriteMode = 0; smb->Remaining = 0; @@ -1703,24 +1703,24 @@ cifs_async_writev(struct cifs_io_subrequest *wdata) rqst.rq_iov = iov; rqst.rq_nvec = 2; - rqst.rq_iter = wdata->iter; - rqst.rq_iter_size = iov_iter_count(&wdata->iter); + rqst.rq_iter = wdata->subreq.io_iter; + rqst.rq_iter_size = iov_iter_count(&wdata->subreq.io_iter); - cifs_dbg(FYI, "async write at %llu %u bytes\n", - wdata->offset, wdata->bytes); + cifs_dbg(FYI, "async write at %llu %zu bytes\n", + wdata->subreq.start, wdata->subreq.len); - smb->DataLengthLow = cpu_to_le16(wdata->bytes & 0xFFFF); - smb->DataLengthHigh = cpu_to_le16(wdata->bytes >> 16); + smb->DataLengthLow = cpu_to_le16(wdata->subreq.len & 0xFFFF); + smb->DataLengthHigh = cpu_to_le16(wdata->subreq.len >> 16); if (wct == 14) { - inc_rfc1001_len(&smb->hdr, wdata->bytes + 1); - put_bcc(wdata->bytes + 1, &smb->hdr); + inc_rfc1001_len(&smb->hdr, wdata->subreq.len + 1); + put_bcc(wdata->subreq.len + 1, &smb->hdr); } else { /* wct == 12 */ struct smb_com_writex_req *smbw = (struct smb_com_writex_req *)smb; - inc_rfc1001_len(&smbw->hdr, wdata->bytes + 5); - put_bcc(wdata->bytes + 5, &smbw->hdr); + inc_rfc1001_len(&smbw->hdr, wdata->subreq.len + 5); + put_bcc(wdata->subreq.len + 5, &smbw->hdr); iov[1].iov_len += 4; /* pad bigger by four bytes */ } diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 2a0982b2d577..7528f6436ec4 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -2537,8 +2537,8 @@ cifs_writev_requeue(struct cifs_io_subrequest *wdata) int rc = 0; struct inode *inode = d_inode(wdata->cfile->dentry); struct TCP_Server_Info *server; - unsigned int rest_len = wdata->bytes; - loff_t fpos = wdata->offset; + unsigned int rest_len = wdata->subreq.len; + loff_t fpos = wdata->subreq.start; server = tlink_tcon(wdata->cfile->tlink)->ses->server; do { @@ -2563,14 +2563,14 @@ cifs_writev_requeue(struct cifs_io_subrequest *wdata) } wdata2->sync_mode = wdata->sync_mode; - wdata2->offset = fpos; - wdata2->bytes = cur_len; - wdata2->iter = wdata->iter; + wdata2->subreq.start = fpos; + wdata2->subreq.len = cur_len; + wdata2->subreq.io_iter = wdata->subreq.io_iter; - iov_iter_advance(&wdata2->iter, fpos - wdata->offset); - iov_iter_truncate(&wdata2->iter, wdata2->bytes); + iov_iter_advance(&wdata2->subreq.io_iter, fpos - wdata->subreq.start); + iov_iter_truncate(&wdata2->subreq.io_iter, wdata2->subreq.len); - if (iov_iter_is_xarray(&wdata2->iter)) + if (iov_iter_is_xarray(&wdata2->subreq.io_iter)) /* Check for pages having been redirtied and clean * them. We can do this by walking the xarray. If * it's not an xarray, then it's a DIO and we shouldn't @@ -2604,7 +2604,7 @@ cifs_writev_requeue(struct cifs_io_subrequest *wdata) } while (rest_len > 0); /* Clean up remaining pages from the original wdata */ - if (iov_iter_is_xarray(&wdata->iter)) + if (iov_iter_is_xarray(&wdata->subreq.io_iter)) cifs_pages_write_failed(inode, fpos, rest_len); if (rc != 0 && !is_retryable_error(rc)) @@ -2621,19 +2621,19 @@ cifs_writev_complete(struct work_struct *work) if (wdata->result == 0) { spin_lock(&inode->i_lock); - cifs_update_eof(CIFS_I(inode), wdata->offset, wdata->bytes); + cifs_update_eof(CIFS_I(inode), wdata->subreq.start, wdata->subreq.len); spin_unlock(&inode->i_lock); cifs_stats_bytes_written(tlink_tcon(wdata->cfile->tlink), - wdata->bytes); + wdata->subreq.len); } else if (wdata->sync_mode == WB_SYNC_ALL && wdata->result == -EAGAIN) return cifs_writev_requeue(wdata); if (wdata->result == -EAGAIN) - cifs_pages_write_redirty(inode, wdata->offset, wdata->bytes); + cifs_pages_write_redirty(inode, wdata->subreq.start, wdata->subreq.len); else if (wdata->result < 0) - cifs_pages_write_failed(inode, wdata->offset, wdata->bytes); + cifs_pages_write_failed(inode, wdata->subreq.start, wdata->subreq.len); else - cifs_pages_written_back(inode, wdata->offset, wdata->bytes); + cifs_pages_written_back(inode, wdata->subreq.start, wdata->subreq.len); if (wdata->result != -EAGAIN) mapping_set_error(inode->i_mapping, wdata->result); @@ -2816,7 +2816,7 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, } wdata->sync_mode = wbc->sync_mode; - wdata->offset = folio_pos(folio); + wdata->subreq.start = folio_pos(folio); wdata->pid = cfile->pid; wdata->credits = credits_on_stack; wdata->cfile = cfile; @@ -2857,13 +2857,13 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, * have been unlocked. */ folio_unlock(folio); - wdata->bytes = len; + wdata->subreq.len = len; if (start < i_size) { - iov_iter_xarray(&wdata->iter, ITER_SOURCE, &mapping->i_pages, + iov_iter_xarray(&wdata->subreq.io_iter, ITER_SOURCE, &mapping->i_pages, start, len); - rc = adjust_credits(wdata->server, &wdata->credits, wdata->bytes); + rc = adjust_credits(wdata->server, &wdata->credits, wdata->subreq.len); if (rc) goto err_wdata; @@ -3270,7 +3270,7 @@ cifs_uncached_writev_complete(struct work_struct *work) struct cifsInodeInfo *cifsi = CIFS_I(inode); spin_lock(&inode->i_lock); - cifs_update_eof(cifsi, wdata->offset, wdata->bytes); + cifs_update_eof(cifsi, wdata->subreq.start, wdata->subreq.len); if (cifsi->netfs.remote_i_size > inode->i_size) i_size_write(inode, cifsi->netfs.remote_i_size); spin_unlock(&inode->i_lock); @@ -3306,19 +3306,19 @@ cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list * segments */ do { - rc = server->ops->wait_mtu_credits(server, wdata->bytes, + rc = server->ops->wait_mtu_credits(server, wdata->subreq.len, &wsize, &credits); if (rc) goto fail; - if (wsize < wdata->bytes) { + if (wsize < wdata->subreq.len) { add_credits_and_wake_if(server, &credits, 0); msleep(1000); } - } while (wsize < wdata->bytes); + } while (wsize < wdata->subreq.len); wdata->credits = credits; - rc = adjust_credits(server, &wdata->credits, wdata->bytes); + rc = adjust_credits(server, &wdata->credits, wdata->subreq.len); if (!rc) { if (wdata->cfile->invalidHandle) @@ -3465,19 +3465,19 @@ cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from, wdata->uncached = true; wdata->sync_mode = WB_SYNC_ALL; - wdata->offset = (__u64)fpos; + wdata->subreq.start = (__u64)fpos; wdata->cfile = cifsFileInfo_get(open_file); wdata->server = server; wdata->pid = pid; - wdata->bytes = cur_len; + wdata->subreq.len = cur_len; wdata->credits = credits_on_stack; - wdata->iter = *from; + wdata->subreq.io_iter = *from; wdata->ctx = ctx; kref_get(&ctx->refcount); - iov_iter_truncate(&wdata->iter, cur_len); + iov_iter_truncate(&wdata->subreq.io_iter, cur_len); - rc = adjust_credits(server, &wdata->credits, wdata->bytes); + rc = adjust_credits(server, &wdata->credits, wdata->subreq.len); if (!rc) { if (wdata->cfile->invalidHandle) @@ -3539,7 +3539,7 @@ restart_loop: if (wdata->result) rc = wdata->result; else - ctx->total_len += wdata->bytes; + ctx->total_len += wdata->subreq.len; /* resend call if it's a retryable error */ if (rc == -EAGAIN) { @@ -3554,10 +3554,10 @@ restart_loop: wdata, &tmp_list, ctx); else { iov_iter_advance(&tmp_from, - wdata->offset - ctx->pos); + wdata->subreq.start - ctx->pos); - rc = cifs_write_from_iter(wdata->offset, - wdata->bytes, &tmp_from, + rc = cifs_write_from_iter(wdata->subreq.start, + wdata->subreq.len, &tmp_from, ctx->cfile, cifs_sb, &tmp_list, ctx); @@ -3880,20 +3880,20 @@ static int cifs_resend_rdata(struct cifs_io_subrequest *rdata, * segments */ do { - rc = server->ops->wait_mtu_credits(server, rdata->bytes, + rc = server->ops->wait_mtu_credits(server, rdata->subreq.len, &rsize, &credits); if (rc) goto fail; - if (rsize < rdata->bytes) { + if (rsize < rdata->subreq.len) { add_credits_and_wake_if(server, &credits, 0); msleep(1000); } - } while (rsize < rdata->bytes); + } while (rsize < rdata->subreq.len); rdata->credits = credits; - rc = adjust_credits(server, &rdata->credits, rdata->bytes); + rc = adjust_credits(server, &rdata->credits, rdata->subreq.len); if (!rc) { if (rdata->cfile->invalidHandle) rc = -EAGAIN; @@ -3991,17 +3991,17 @@ cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file, rdata->server = server; rdata->cfile = cifsFileInfo_get(open_file); - rdata->offset = fpos; - rdata->bytes = cur_len; + rdata->subreq.start = fpos; + rdata->subreq.len = cur_len; rdata->pid = pid; rdata->credits = credits_on_stack; rdata->ctx = ctx; kref_get(&ctx->refcount); - rdata->iter = ctx->iter; - iov_iter_truncate(&rdata->iter, cur_len); + rdata->subreq.io_iter = ctx->iter; + iov_iter_truncate(&rdata->subreq.io_iter, cur_len); - rc = adjust_credits(server, &rdata->credits, rdata->bytes); + rc = adjust_credits(server, &rdata->credits, rdata->subreq.len); if (!rc) { if (rdata->cfile->invalidHandle) @@ -4071,8 +4071,8 @@ again: &tmp_list, ctx); } else { rc = cifs_send_async_read( - rdata->offset + got_bytes, - rdata->bytes - got_bytes, + rdata->subreq.start + got_bytes, + rdata->subreq.len - got_bytes, rdata->cfile, cifs_sb, &tmp_list, ctx); @@ -4086,7 +4086,7 @@ again: rc = rdata->result; /* if there was a short read -- discard anything left */ - if (rdata->got_bytes && rdata->got_bytes < rdata->bytes) + if (rdata->got_bytes && rdata->got_bytes < rdata->subreq.len) rc = -ENODATA; ctx->total_len += rdata->got_bytes; @@ -4470,16 +4470,16 @@ static void cifs_readahead_complete(struct work_struct *work) pgoff_t last; bool good = rdata->result == 0 || (rdata->result == -EAGAIN && rdata->got_bytes); - XA_STATE(xas, &rdata->mapping->i_pages, rdata->offset / PAGE_SIZE); + XA_STATE(xas, &rdata->mapping->i_pages, rdata->subreq.start / PAGE_SIZE); if (good) cifs_readahead_to_fscache(rdata->mapping->host, - rdata->offset, rdata->bytes); + rdata->subreq.start, rdata->subreq.len); - if (iov_iter_count(&rdata->iter) > 0) - iov_iter_zero(iov_iter_count(&rdata->iter), &rdata->iter); + if (iov_iter_count(&rdata->subreq.io_iter) > 0) + iov_iter_zero(iov_iter_count(&rdata->subreq.io_iter), &rdata->subreq.io_iter); - last = (rdata->offset + rdata->bytes - 1) / PAGE_SIZE; + last = (rdata->subreq.start + rdata->subreq.len - 1) / PAGE_SIZE; rcu_read_lock(); xas_for_each(&xas, folio, last) { @@ -4618,8 +4618,8 @@ static void cifs_readahead(struct readahead_control *ractl) break; } - rdata->offset = ra_index * PAGE_SIZE; - rdata->bytes = nr_pages * PAGE_SIZE; + rdata->subreq.start = ra_index * PAGE_SIZE; + rdata->subreq.len = nr_pages * PAGE_SIZE; rdata->cfile = cifsFileInfo_get(open_file); rdata->server = server; rdata->mapping = ractl->mapping; @@ -4633,10 +4633,10 @@ static void cifs_readahead(struct readahead_control *ractl) ra_pages -= nr_pages; ra_index += nr_pages; - iov_iter_xarray(&rdata->iter, ITER_DEST, &rdata->mapping->i_pages, - rdata->offset, rdata->bytes); + iov_iter_xarray(&rdata->subreq.io_iter, ITER_DEST, &rdata->mapping->i_pages, + rdata->subreq.start, rdata->subreq.len); - rc = adjust_credits(server, &rdata->credits, rdata->bytes); + rc = adjust_credits(server, &rdata->credits, rdata->subreq.len); if (!rc) { if (rdata->cfile->invalidHandle) rc = -EAGAIN; @@ -4647,8 +4647,8 @@ static void cifs_readahead(struct readahead_control *ractl) if (rc) { add_credits_and_wake_if(server, &rdata->credits, 0); cifs_unlock_folios(rdata->mapping, - rdata->offset / PAGE_SIZE, - (rdata->offset + rdata->bytes - 1) / PAGE_SIZE); + rdata->subreq.start / PAGE_SIZE, + (rdata->subreq.start + rdata->subreq.len - 1) / PAGE_SIZE); /* Fallback to the readpage in error/reconnect cases */ cifs_put_readdata(rdata); break; diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 56026f4cbc13..c13705c0e47f 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -4592,7 +4592,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid, /* Copy the data to the output I/O iterator. */ rdata->result = cifs_copy_pages_to_iter(pages, pages_len, - cur_off, &rdata->iter); + cur_off, &rdata->subreq.io_iter); if (rdata->result != 0) { if (is_offloaded) mid->mid_state = MID_RESPONSE_MALFORMED; @@ -4606,7 +4606,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid, /* read response payload is in buf */ WARN_ONCE(pages && !xa_empty(pages), "read data can be either in buf or in pages"); - length = copy_to_iter(buf + data_offset, data_len, &rdata->iter); + length = copy_to_iter(buf + data_offset, data_len, &rdata->subreq.io_iter); if (length < 0) return length; rdata->got_bytes = data_len; diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 8143611ba336..ca8ba6878f9f 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -4434,7 +4434,7 @@ smb2_new_read_req(void **buf, unsigned int *total_len, struct smbd_buffer_descriptor_v1 *v1; bool need_invalidate = server->dialect == SMB30_PROT_ID; - rdata->mr = smbd_register_mr(server->smbd_conn, &rdata->iter, + rdata->mr = smbd_register_mr(server->smbd_conn, &rdata->subreq.io_iter, true, need_invalidate); if (!rdata->mr) return -EAGAIN; @@ -4494,17 +4494,17 @@ smb2_readv_callback(struct mid_q_entry *mid) struct smb_rqst rqst = { .rq_iov = &rdata->iov[1], .rq_nvec = 1 }; if (rdata->got_bytes) { - rqst.rq_iter = rdata->iter; - rqst.rq_iter_size = iov_iter_count(&rdata->iter); + rqst.rq_iter = rdata->subreq.io_iter; + rqst.rq_iter_size = iov_iter_count(&rdata->subreq.io_iter); } WARN_ONCE(rdata->server != mid->server, "rdata server %p != mid server %p", rdata->server, mid->server); - cifs_dbg(FYI, "%s: mid=%llu state=%d result=%d bytes=%u\n", + cifs_dbg(FYI, "%s: mid=%llu state=%d result=%d bytes=%zu\n", __func__, mid->mid, mid->mid_state, rdata->result, - rdata->bytes); + rdata->subreq.len); switch (mid->mid_state) { case MID_RESPONSE_RECEIVED: @@ -4557,13 +4557,13 @@ smb2_readv_callback(struct mid_q_entry *mid) cifs_stats_fail_inc(tcon, SMB2_READ_HE); trace_smb3_read_err(0 /* xid */, rdata->cfile->fid.persistent_fid, - tcon->tid, tcon->ses->Suid, rdata->offset, - rdata->bytes, rdata->result); + tcon->tid, tcon->ses->Suid, rdata->subreq.start, + rdata->subreq.len, rdata->result); } else trace_smb3_read_done(0 /* xid */, rdata->cfile->fid.persistent_fid, tcon->tid, tcon->ses->Suid, - rdata->offset, rdata->got_bytes); + rdata->subreq.start, rdata->got_bytes); queue_work(cifsiod_wq, &rdata->work); release_mid(mid); @@ -4585,16 +4585,16 @@ smb2_async_readv(struct cifs_io_subrequest *rdata) unsigned int total_len; int credit_request; - cifs_dbg(FYI, "%s: offset=%llu bytes=%u\n", - __func__, rdata->offset, rdata->bytes); + cifs_dbg(FYI, "%s: offset=%llu bytes=%zu\n", + __func__, rdata->subreq.start, rdata->subreq.len); if (!rdata->server) rdata->server = cifs_pick_channel(tcon->ses); io_parms.tcon = tlink_tcon(rdata->cfile->tlink); io_parms.server = server = rdata->server; - io_parms.offset = rdata->offset; - io_parms.length = rdata->bytes; + io_parms.offset = rdata->subreq.start; + io_parms.length = rdata->subreq.len; io_parms.persistent_fid = rdata->cfile->fid.persistent_fid; io_parms.volatile_fid = rdata->cfile->fid.volatile_fid; io_parms.pid = rdata->pid; @@ -4613,7 +4613,7 @@ smb2_async_readv(struct cifs_io_subrequest *rdata) shdr = (struct smb2_hdr *)buf; if (rdata->credits.value > 0) { - shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(rdata->bytes, + shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(rdata->subreq.len, SMB2_MAX_BUFFER_SIZE)); credit_request = le16_to_cpu(shdr->CreditCharge) + 8; if (server->credits >= server->max_credits) @@ -4623,7 +4623,7 @@ smb2_async_readv(struct cifs_io_subrequest *rdata) min_t(int, server->max_credits - server->credits, credit_request)); - rc = adjust_credits(server, &rdata->credits, rdata->bytes); + rc = adjust_credits(server, &rdata->credits, rdata->subreq.len); if (rc) goto async_readv_out; @@ -4763,13 +4763,13 @@ smb2_writev_callback(struct mid_q_entry *mid) * client. OS/2 servers are known to set incorrect * CountHigh values. */ - if (written > wdata->bytes) + if (written > wdata->subreq.len) written &= 0xFFFF; - if (written < wdata->bytes) + if (written < wdata->subreq.len) wdata->result = -ENOSPC; else - wdata->bytes = written; + wdata->subreq.len = written; break; case MID_REQUEST_SUBMITTED: case MID_RETRY_NEEDED: @@ -4800,8 +4800,8 @@ smb2_writev_callback(struct mid_q_entry *mid) cifs_stats_fail_inc(tcon, SMB2_WRITE_HE); trace_smb3_write_err(0 /* no xid */, wdata->cfile->fid.persistent_fid, - tcon->tid, tcon->ses->Suid, wdata->offset, - wdata->bytes, wdata->result); + tcon->tid, tcon->ses->Suid, wdata->subreq.start, + wdata->subreq.len, wdata->result); if (wdata->result == -ENOSPC) pr_warn_once("Out of space writing to %s\n", tcon->tree_name); @@ -4809,7 +4809,7 @@ smb2_writev_callback(struct mid_q_entry *mid) trace_smb3_write_done(0 /* no xid */, wdata->cfile->fid.persistent_fid, tcon->tid, tcon->ses->Suid, - wdata->offset, wdata->bytes); + wdata->subreq.start, wdata->subreq.len); queue_work(cifsiod_wq, &wdata->work); release_mid(mid); @@ -4842,8 +4842,8 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) _io_parms = (struct cifs_io_parms) { .tcon = tcon, .server = server, - .offset = wdata->offset, - .length = wdata->bytes, + .offset = wdata->subreq.start, + .length = wdata->subreq.len, .persistent_fid = wdata->cfile->fid.persistent_fid, .volatile_fid = wdata->cfile->fid.volatile_fid, .pid = wdata->pid, @@ -4885,10 +4885,10 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) */ if (smb3_use_rdma_offload(io_parms)) { struct smbd_buffer_descriptor_v1 *v1; - size_t data_size = iov_iter_count(&wdata->iter); + size_t data_size = iov_iter_count(&wdata->subreq.io_iter); bool need_invalidate = server->dialect == SMB30_PROT_ID; - wdata->mr = smbd_register_mr(server->smbd_conn, &wdata->iter, + wdata->mr = smbd_register_mr(server->smbd_conn, &wdata->subreq.io_iter, false, need_invalidate); if (!wdata->mr) { rc = -EAGAIN; @@ -4915,7 +4915,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) rqst.rq_iov = iov; rqst.rq_nvec = 1; - rqst.rq_iter = wdata->iter; + rqst.rq_iter = wdata->subreq.io_iter; rqst.rq_iter_size = iov_iter_count(&rqst.rq_iter); if (wdata->replay) smb2_set_replay(server, &rqst); @@ -4935,7 +4935,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) #endif if (wdata->credits.value > 0) { - shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(wdata->bytes, + shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(wdata->subreq.len, SMB2_MAX_BUFFER_SIZE)); credit_request = le16_to_cpu(shdr->CreditCharge) + 8; if (server->credits >= server->max_credits) diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index dcc10a105b56..cccc1248c48a 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -1707,8 +1707,8 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) unsigned int buflen = server->pdu_size + HEADER_PREAMBLE_SIZE(server); bool use_rdma_mr = false; - cifs_dbg(FYI, "%s: mid=%llu offset=%llu bytes=%u\n", - __func__, mid->mid, rdata->offset, rdata->bytes); + cifs_dbg(FYI, "%s: mid=%llu offset=%llu bytes=%zu\n", + __func__, mid->mid, rdata->subreq.start, rdata->subreq.len); /* * read the rest of READ_RSP header (sans Data array), or whatever we @@ -1813,7 +1813,7 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) length = data_len; /* An RDMA read is already done. */ else #endif - length = cifs_read_iter_from_socket(server, &rdata->iter, + length = cifs_read_iter_from_socket(server, &rdata->subreq.io_iter, data_len); if (length > 0) rdata->got_bytes += length; -- cgit v1.2.3-59-g8ed1b From 56257334e8e0075515aedc44044a5585dcf7f465 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 18:20:39 +0100 Subject: cifs: Make wait_mtu_credits take size_t args Make the wait_mtu_credits functions use size_t for the size and num arguments rather than unsigned int as netfslib uses size_t/ssize_t for arguments and return values to allow for extra capacity. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: linux-cachefs@redhat.com cc: netfs@lists.linux.dev cc: linux-mm@kvack.org --- fs/smb/client/cifsglob.h | 4 ++-- fs/smb/client/cifsproto.h | 2 +- fs/smb/client/file.c | 17 ++++++++++------- fs/smb/client/smb2ops.c | 4 ++-- fs/smb/client/transport.c | 4 ++-- 5 files changed, 17 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 5100aa047ed6..13161e0b5cc0 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -546,8 +546,8 @@ struct smb_version_operations { /* writepages retry size */ unsigned int (*wp_retry_size)(struct inode *); /* get mtu credits */ - int (*wait_mtu_credits)(struct TCP_Server_Info *, unsigned int, - unsigned int *, struct cifs_credits *); + int (*wait_mtu_credits)(struct TCP_Server_Info *, size_t, + size_t *, struct cifs_credits *); /* adjust previously taken mtu credits to request size */ int (*adjust_credits)(struct TCP_Server_Info *server, struct cifs_credits *credits, diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index cb5cc5711a84..103f264b42d7 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -121,7 +121,7 @@ extern struct mid_q_entry *cifs_setup_async_request(struct TCP_Server_Info *, extern int cifs_check_receive(struct mid_q_entry *mid, struct TCP_Server_Info *server, bool log_error); extern int cifs_wait_mtu_credits(struct TCP_Server_Info *server, - unsigned int size, unsigned int *num, + size_t size, size_t *num, struct cifs_credits *credits); extern int SendReceive2(const unsigned int /* xid */ , struct cifs_ses *, struct kvec *, int /* nvec to send */, diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 7528f6436ec4..4d8cd0d82104 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -2783,7 +2783,8 @@ static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, struct cifs_credits *credits = &credits_on_stack; struct cifsFileInfo *cfile = NULL; unsigned long long i_size = i_size_read(inode), max_len; - unsigned int xid, wsize; + unsigned int xid; + size_t wsize; size_t len = folio_size(folio); long count = wbc->nr_to_write; int rc; @@ -3285,7 +3286,7 @@ static int cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list, struct cifs_aio_ctx *ctx) { - unsigned int wsize; + size_t wsize; struct cifs_credits credits; int rc; struct TCP_Server_Info *server = wdata->server; @@ -3420,7 +3421,8 @@ cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from, do { struct cifs_credits credits_on_stack; struct cifs_credits *credits = &credits_on_stack; - unsigned int wsize, nsegs = 0; + unsigned int nsegs = 0; + size_t wsize; if (signal_pending(current)) { rc = -EINTR; @@ -3857,7 +3859,7 @@ static int cifs_resend_rdata(struct cifs_io_subrequest *rdata, struct list_head *rdata_list, struct cifs_aio_ctx *ctx) { - unsigned int rsize; + size_t rsize; struct cifs_credits credits; int rc; struct TCP_Server_Info *server; @@ -3931,10 +3933,10 @@ cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file, struct cifs_aio_ctx *ctx) { struct cifs_io_subrequest *rdata; - unsigned int rsize, nsegs, max_segs = INT_MAX; + unsigned int nsegs, max_segs = INT_MAX; struct cifs_credits credits_on_stack; struct cifs_credits *credits = &credits_on_stack; - size_t cur_len, max_len; + size_t cur_len, max_len, rsize; int rc; pid_t pid; struct TCP_Server_Info *server; @@ -4530,12 +4532,13 @@ static void cifs_readahead(struct readahead_control *ractl) * Chop the readahead request up into rsize-sized read requests. */ while ((nr_pages = ra_pages)) { - unsigned int i, rsize; + unsigned int i; struct cifs_io_subrequest *rdata; struct cifs_credits credits_on_stack; struct cifs_credits *credits = &credits_on_stack; struct folio *folio; pgoff_t fsize; + size_t rsize; /* * Find out if we have anything cached in the range of diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index c13705c0e47f..ef18cd30f66c 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -217,8 +217,8 @@ smb2_get_credits(struct mid_q_entry *mid) } static int -smb2_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size, - unsigned int *num, struct cifs_credits *credits) +smb2_wait_mtu_credits(struct TCP_Server_Info *server, size_t size, + size_t *num, struct cifs_credits *credits) { int rc = 0; unsigned int scredits, in_flight; diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index cccc1248c48a..4c8af995c60b 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -691,8 +691,8 @@ wait_for_compound_request(struct TCP_Server_Info *server, int num, } int -cifs_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size, - unsigned int *num, struct cifs_credits *credits) +cifs_wait_mtu_credits(struct TCP_Server_Info *server, size_t size, + size_t *num, struct cifs_credits *credits) { *num = size; credits->value = 0; -- cgit v1.2.3-59-g8ed1b From dc5939de82f149633d6ec1c403003538442ec9ef Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 5 Feb 2024 09:08:57 +0000 Subject: cifs: Replace the writedata replay bool with a netfs sreq flag Replace the 'replay' bool in cifs_writedata (now cifs_io_subrequest) with a flag in the netfs_io_subrequest flags. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/file.c | 2 +- fs/smb/client/smb2pdu.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 4d8cd0d82104..7f2ce68d2a2a 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -3325,7 +3325,7 @@ cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list if (wdata->cfile->invalidHandle) rc = -EAGAIN; else { - wdata->replay = true; + set_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags); #ifdef CONFIG_CIFS_SMB_DIRECT if (wdata->mr) { wdata->mr->need_invalidate = true; diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index ca8ba6878f9f..b063f5e95f5d 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -4832,7 +4832,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) struct cifs_io_parms *io_parms = NULL; int credit_request; - if (!wdata->server || wdata->replay) + if (!wdata->server || test_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags)) server = wdata->server = cifs_pick_channel(tcon->ses); /* @@ -4917,7 +4917,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) rqst.rq_nvec = 1; rqst.rq_iter = wdata->subreq.io_iter; rqst.rq_iter_size = iov_iter_count(&rqst.rq_iter); - if (wdata->replay) + if (test_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags)) smb2_set_replay(server, &rqst); #ifdef CONFIG_CIFS_SMB_DIRECT if (wdata->mr) -- cgit v1.2.3-59-g8ed1b From 1a5b4edd97cee40922ca8bfb91008338d3a1de60 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 12 Oct 2023 08:36:44 +0100 Subject: cifs: Move cifs_loose_read_iter() and cifs_file_write_iter() to file.c Move cifs_loose_read_iter() and cifs_file_write_iter() to file.c so that they are colocated with similar functions rather than being split with cifsfs.c. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/cifsfs.c | 55 -------------------------------------------------- fs/smb/client/cifsfs.h | 2 ++ fs/smb/client/file.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 55 insertions(+), 55 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 39277c37185c..847df5152835 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -986,61 +986,6 @@ out: return root; } - -static ssize_t -cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter) -{ - ssize_t rc; - struct inode *inode = file_inode(iocb->ki_filp); - - if (iocb->ki_flags & IOCB_DIRECT) - return cifs_user_readv(iocb, iter); - - rc = cifs_revalidate_mapping(inode); - if (rc) - return rc; - - return generic_file_read_iter(iocb, iter); -} - -static ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) -{ - struct inode *inode = file_inode(iocb->ki_filp); - struct cifsInodeInfo *cinode = CIFS_I(inode); - ssize_t written; - int rc; - - if (iocb->ki_filp->f_flags & O_DIRECT) { - written = cifs_user_writev(iocb, from); - if (written > 0 && CIFS_CACHE_READ(cinode)) { - cifs_zap_mapping(inode); - cifs_dbg(FYI, - "Set no oplock for inode=%p after a write operation\n", - inode); - cinode->oplock = 0; - } - return written; - } - - written = cifs_get_writer(cinode); - if (written) - return written; - - written = generic_file_write_iter(iocb, from); - - if (CIFS_CACHE_WRITE(CIFS_I(inode))) - goto out; - - rc = filemap_fdatawrite(inode->i_mapping); - if (rc) - cifs_dbg(FYI, "cifs_file_write_iter: %d rc on %p inode\n", - rc, inode); - -out: - cifs_put_writer(cinode); - return written; -} - static loff_t cifs_llseek(struct file *file, loff_t offset, int whence) { struct cifsFileInfo *cfile = file->private_data; diff --git a/fs/smb/client/cifsfs.h b/fs/smb/client/cifsfs.h index 1ab7e5998c58..1acf6bfc06de 100644 --- a/fs/smb/client/cifsfs.h +++ b/fs/smb/client/cifsfs.h @@ -99,6 +99,8 @@ extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to); extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from); extern ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from); extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from); +ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from); +ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter); extern int cifs_flock(struct file *pfile, int cmd, struct file_lock *plock); extern int cifs_lock(struct file *, int, struct file_lock *); extern int cifs_fsync(struct file *, loff_t, loff_t, int); diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 7f2ce68d2a2a..331446fc3d22 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -4250,6 +4250,59 @@ ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to) return __cifs_readv(iocb, to, false); } +ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter) +{ + ssize_t rc; + struct inode *inode = file_inode(iocb->ki_filp); + + if (iocb->ki_flags & IOCB_DIRECT) + return cifs_user_readv(iocb, iter); + + rc = cifs_revalidate_mapping(inode); + if (rc) + return rc; + + return generic_file_read_iter(iocb, iter); +} + +ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) +{ + struct inode *inode = file_inode(iocb->ki_filp); + struct cifsInodeInfo *cinode = CIFS_I(inode); + ssize_t written; + int rc; + + if (iocb->ki_filp->f_flags & O_DIRECT) { + written = cifs_user_writev(iocb, from); + if (written > 0 && CIFS_CACHE_READ(cinode)) { + cifs_zap_mapping(inode); + cifs_dbg(FYI, + "Set no oplock for inode=%p after a write operation\n", + inode); + cinode->oplock = 0; + } + return written; + } + + written = cifs_get_writer(cinode); + if (written) + return written; + + written = generic_file_write_iter(iocb, from); + + if (CIFS_CACHE_WRITE(CIFS_I(inode))) + goto out; + + rc = filemap_fdatawrite(inode->i_mapping); + if (rc) + cifs_dbg(FYI, "cifs_file_write_iter: %d rc on %p inode\n", + rc, inode); + +out: + cifs_put_writer(cinode); + return written; +} + ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) { -- cgit v1.2.3-59-g8ed1b From 3758c485f6c9124d8ad76b88382004cbc28a0892 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 28 Nov 2023 16:26:44 +0000 Subject: cifs: Set zero_point in the copy_file_range() and remap_file_range() Set zero_point in the copy_file_range() and remap_file_range() implementations so that we don't skip reading data modified on a server-side copy. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/cifsfs.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 847df5152835..9d656a08f617 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -1287,6 +1287,8 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off, rc = cifs_flush_folio(target_inode, destend, &fstart, &fend, false); if (rc) goto unlock; + if (fend > target_cifsi->netfs.zero_point) + target_cifsi->netfs.zero_point = fend + 1; /* Discard all the folios that overlap the destination region. */ cifs_dbg(FYI, "about to discard pages %llx-%llx\n", fstart, fend); @@ -1305,6 +1307,8 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off, fscache_resize_cookie(cifs_inode_cookie(target_inode), new_size); } + if (rc == 0 && new_size > target_cifsi->netfs.zero_point) + target_cifsi->netfs.zero_point = new_size; } /* force revalidate of size and timestamps of target file now @@ -1396,6 +1400,8 @@ ssize_t cifs_file_copychunk_range(unsigned int xid, rc = cifs_flush_folio(target_inode, destend, &fstart, &fend, false); if (rc) goto unlock; + if (fend > target_cifsi->netfs.zero_point) + target_cifsi->netfs.zero_point = fend + 1; /* Discard all the folios that overlap the destination region. */ truncate_inode_pages_range(&target_inode->i_data, fstart, fend); -- cgit v1.2.3-59-g8ed1b From edea94a69730b74a8867bbafe742c3fc4e580722 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 15 Mar 2024 18:03:30 +0000 Subject: cifs: Add mempools for cifs_io_request and cifs_io_subrequest structs Add mempools for the allocation of cifs_io_request and cifs_io_subrequest structs for netfslib to use so that it can guarantee eventual allocation in writeback. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/cifsfs.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++- fs/smb/client/cifsglob.h | 2 ++ 2 files changed, 56 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 9d656a08f617..763d17870e0b 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -371,9 +371,13 @@ static struct kmem_cache *cifs_inode_cachep; static struct kmem_cache *cifs_req_cachep; static struct kmem_cache *cifs_mid_cachep; static struct kmem_cache *cifs_sm_req_cachep; +static struct kmem_cache *cifs_io_request_cachep; +static struct kmem_cache *cifs_io_subrequest_cachep; mempool_t *cifs_sm_req_poolp; mempool_t *cifs_req_poolp; mempool_t *cifs_mid_poolp; +mempool_t cifs_io_request_pool; +mempool_t cifs_io_subrequest_pool; static struct inode * cifs_alloc_inode(struct super_block *sb) @@ -1750,6 +1754,48 @@ static void destroy_mids(void) kmem_cache_destroy(cifs_mid_cachep); } +static int cifs_init_netfs(void) +{ + cifs_io_request_cachep = + kmem_cache_create("cifs_io_request", + sizeof(struct netfs_io_request), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!cifs_io_request_cachep) + goto nomem_req; + + if (mempool_init_slab_pool(&cifs_io_request_pool, 100, cifs_io_request_cachep) < 0) + goto nomem_reqpool; + + cifs_io_subrequest_cachep = + kmem_cache_create("cifs_io_subrequest", + sizeof(struct cifs_io_subrequest), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!cifs_io_subrequest_cachep) + goto nomem_subreq; + + if (mempool_init_slab_pool(&cifs_io_subrequest_pool, 100, cifs_io_subrequest_cachep) < 0) + goto nomem_subreqpool; + + return 0; + +nomem_subreqpool: + kmem_cache_destroy(cifs_io_subrequest_cachep); +nomem_subreq: + mempool_destroy(&cifs_io_request_pool); +nomem_reqpool: + kmem_cache_destroy(cifs_io_request_cachep); +nomem_req: + return -ENOMEM; +} + +static void cifs_destroy_netfs(void) +{ + mempool_destroy(&cifs_io_subrequest_pool); + kmem_cache_destroy(cifs_io_subrequest_cachep); + mempool_destroy(&cifs_io_request_pool); + kmem_cache_destroy(cifs_io_request_cachep); +} + static int __init init_cifs(void) { @@ -1854,10 +1900,14 @@ init_cifs(void) if (rc) goto out_destroy_deferredclose_wq; - rc = init_mids(); + rc = cifs_init_netfs(); if (rc) goto out_destroy_inodecache; + rc = init_mids(); + if (rc) + goto out_destroy_netfs; + rc = cifs_init_request_bufs(); if (rc) goto out_destroy_mids; @@ -1912,6 +1962,8 @@ out_destroy_request_bufs: cifs_destroy_request_bufs(); out_destroy_mids: destroy_mids(); +out_destroy_netfs: + cifs_destroy_netfs(); out_destroy_inodecache: cifs_destroy_inodecache(); out_destroy_deferredclose_wq: @@ -1950,6 +2002,7 @@ exit_cifs(void) #endif cifs_destroy_request_bufs(); destroy_mids(); + cifs_destroy_netfs(); cifs_destroy_inodecache(); destroy_workqueue(deferredclose_wq); destroy_workqueue(cifsoplockd_wq); diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 13161e0b5cc0..29c9ee2dd304 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -2094,6 +2094,8 @@ extern __u32 cifs_lock_secret; extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; extern mempool_t *cifs_mid_poolp; +extern mempool_t cifs_io_request_pool; +extern mempool_t cifs_io_subrequest_pool; /* Operations for different SMB versions */ #define SMB1_VERSION_STRING "1.0" -- cgit v1.2.3-59-g8ed1b From c20c0d7325abd9a8bf985a934591d75d514a3d4d Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 28 Mar 2024 07:59:18 +0000 Subject: cifs: Make add_credits_and_wake_if() clear deducted credits Make add_credits_and_wake_if() clear the amount of credits in the cifs_credits struct after it has returned them to the overall counter. This allows add_credits_and_wake_if() to be called multiple times during the error handling and cleanup without accidentally returning the credits again and again. Note that the wake_up() in add_credits_and_wake_if() may also be superfluous as ->add_credits() also does a wake on the request_q. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/smb/client/cifsglob.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 29c9ee2dd304..611f59c6d2c0 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -881,11 +881,12 @@ add_credits(struct TCP_Server_Info *server, const struct cifs_credits *credits, static inline void add_credits_and_wake_if(struct TCP_Server_Info *server, - const struct cifs_credits *credits, const int optype) + struct cifs_credits *credits, const int optype) { if (credits->value) { server->ops->add_credits(server, credits, optype); wake_up(&server->request_q); + credits->value = 0; } } -- cgit v1.2.3-59-g8ed1b From 69c3c023af25edb5433a2db824d3e7cc328f0183 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 18:16:15 +0100 Subject: cifs: Implement netfslib hooks Provide implementation of the netfslib hooks that will be used by netfslib to ask cifs to set up and perform operations. Of particular note are (*) cifs_clamp_length() - This is used to negotiate the size of the next subrequest in a read request, taking into account the credit available and the rsize. The credits are attached to the subrequest. (*) cifs_req_issue_read() - This is used to issue a subrequest that has been set up and clamped. (*) cifs_prepare_write() - This prepares to fill a subrequest by picking a channel, reopening the file and requesting credits so that we can set the maximum size of the subrequest and also sets the maximum number of segments if we're doing RDMA. (*) cifs_issue_write() - This releases any unneeded credits and issues an asynchronous data write for the contiguous slice of file covered by the subrequest. This should possibly be folded in to all ->async_writev() ops and that called directly. (*) cifs_begin_writeback() - This gets the cached writable handle through which we do writeback (this does not affect writethrough, unbuffered or direct writes). At this point, cifs is not wired up to actually *use* netfslib; that will be done in a subsequent patch. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/netfs/buffered_write.c | 6 + fs/smb/client/Kconfig | 1 + fs/smb/client/cifsfs.c | 2 +- fs/smb/client/cifsfs.h | 1 + fs/smb/client/cifsglob.h | 28 ++-- fs/smb/client/file.c | 315 +++++++++++++++++++++++++++++++++++++++++++ include/linux/netfs.h | 1 + include/trace/events/netfs.h | 1 + 8 files changed, 345 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 825e6632ee4f..1121601536d1 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -405,6 +405,9 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, } while (iov_iter_count(iter)); out: + if (likely(written) && ctx->ops->post_modify) + ctx->ops->post_modify(inode); + if (unlikely(wreq)) { ret2 = netfs_end_writethrough(wreq, &wbc, writethrough); wbc_detach_inode(&wbc); @@ -521,6 +524,7 @@ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_gr struct folio *folio = page_folio(vmf->page); struct file *file = vmf->vma->vm_file; struct inode *inode = file_inode(file); + struct netfs_inode *ictx = netfs_inode(inode); vm_fault_t ret = VM_FAULT_RETRY; int err; @@ -567,6 +571,8 @@ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_gr trace_netfs_folio(folio, netfs_folio_trace_mkwrite); netfs_set_group(folio, netfs_group); file_update_time(file); + if (ictx->ops->post_modify) + ictx->ops->post_modify(inode); ret = VM_FAULT_LOCKED; out: sb_end_pagefault(inode->i_sb); diff --git a/fs/smb/client/Kconfig b/fs/smb/client/Kconfig index 2927bd174a88..2517dc242386 100644 --- a/fs/smb/client/Kconfig +++ b/fs/smb/client/Kconfig @@ -2,6 +2,7 @@ config CIFS tristate "SMB3 and CIFS support (advanced network filesystem)" depends on INET + select NETFS_SUPPORT select NLS select NLS_UCS2_UTILS select CRYPTO diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 763d17870e0b..8f7165567be7 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -1758,7 +1758,7 @@ static int cifs_init_netfs(void) { cifs_io_request_cachep = kmem_cache_create("cifs_io_request", - sizeof(struct netfs_io_request), 0, + sizeof(struct cifs_io_request), 0, SLAB_HWCACHE_ALIGN, NULL); if (!cifs_io_request_cachep) goto nomem_req; diff --git a/fs/smb/client/cifsfs.h b/fs/smb/client/cifsfs.h index 1acf6bfc06de..922c10d7cfdd 100644 --- a/fs/smb/client/cifsfs.h +++ b/fs/smb/client/cifsfs.h @@ -84,6 +84,7 @@ extern const struct inode_operations cifs_namespace_inode_operations; /* Functions related to files and directories */ +extern const struct netfs_request_ops cifs_req_ops; extern const struct file_operations cifs_file_ops; extern const struct file_operations cifs_file_direct_ops; /* if directio mnt */ extern const struct file_operations cifs_file_strict_ops; /* if strictio mnt */ diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 611f59c6d2c0..4e9033b2f191 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1491,15 +1491,24 @@ struct cifs_aio_ctx { bool direct_io; }; +struct cifs_io_request { + struct netfs_io_request rreq; + struct cifsFileInfo *cfile; +}; + /* asynchronous read support */ struct cifs_io_subrequest { - struct netfs_io_subrequest subreq; - struct cifsFileInfo *cfile; - struct address_space *mapping; - struct cifs_aio_ctx *ctx; + union { + struct netfs_io_subrequest subreq; + struct netfs_io_request *rreq; + struct cifs_io_request *req; + }; ssize_t got_bytes; pid_t pid; + unsigned int xid; int result; + bool have_xid; + bool replay; struct kvec iov[2]; struct TCP_Server_Info *server; #ifdef CONFIG_CIFS_SMB_DIRECT @@ -1507,15 +1516,16 @@ struct cifs_io_subrequest { #endif struct cifs_credits credits; - enum writeback_sync_modes sync_mode; - bool uncached; - bool replay; - struct bio_vec *bv; - // TODO: Remove following elements struct list_head list; struct completion done; struct work_struct work; + struct cifsFileInfo *cfile; + struct address_space *mapping; + struct cifs_aio_ctx *ctx; + enum writeback_sync_modes sync_mode; + bool uncached; + struct bio_vec *bv; }; /* diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 331446fc3d22..c9c33b3a54f3 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -36,6 +36,321 @@ #include "fs_context.h" #include "cifs_ioctl.h" #include "cached_dir.h" +#include + +static int cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush); + +/* + * Prepare a subrequest to upload to the server. We need to allocate credits + * so that we know the maximum amount of data that we can include in it. + */ +static void cifs_prepare_write(struct netfs_io_subrequest *subreq) +{ + struct cifs_io_subrequest *wdata = + container_of(subreq, struct cifs_io_subrequest, subreq); + struct cifs_io_request *req = wdata->req; + struct TCP_Server_Info *server; + struct cifsFileInfo *open_file = req->cfile; + size_t wsize = req->rreq.wsize; + int rc; + + if (!wdata->have_xid) { + wdata->xid = get_xid(); + wdata->have_xid = true; + } + + server = cifs_pick_channel(tlink_tcon(open_file->tlink)->ses); + wdata->server = server; + +retry: + if (open_file->invalidHandle) { + rc = cifs_reopen_file(open_file, false); + if (rc < 0) { + if (rc == -EAGAIN) + goto retry; + subreq->error = rc; + return netfs_prepare_write_failed(subreq); + } + } + + rc = server->ops->wait_mtu_credits(server, wsize, &wdata->subreq.max_len, + &wdata->credits); + if (rc < 0) { + subreq->error = rc; + return netfs_prepare_write_failed(subreq); + } + +#ifdef CONFIG_CIFS_SMB_DIRECT + if (server->smbd_conn) + subreq->max_nr_segs = server->smbd_conn->max_frmr_depth; +#endif +} + +/* + * Issue a subrequest to upload to the server. + */ +static void cifs_issue_write(struct netfs_io_subrequest *subreq) +{ + struct cifs_io_subrequest *wdata = + container_of(subreq, struct cifs_io_subrequest, subreq); + struct cifs_sb_info *sbi = CIFS_SB(subreq->rreq->inode->i_sb); + int rc; + + if (cifs_forced_shutdown(sbi)) { + rc = -EIO; + goto fail; + } + + rc = adjust_credits(wdata->server, &wdata->credits, wdata->subreq.len); + if (rc) + goto fail; + + rc = -EAGAIN; + if (wdata->req->cfile->invalidHandle) + goto fail; + + wdata->server->ops->async_writev(wdata); +out: + return; + +fail: + if (rc == -EAGAIN) + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + else + trace_netfs_sreq(subreq, netfs_sreq_trace_fail); + add_credits_and_wake_if(wdata->server, &wdata->credits, 0); + netfs_write_subrequest_terminated(wdata, rc, false); + goto out; +} + +/* + * Split the read up according to how many credits we can get for each piece. + * It's okay to sleep here if we need to wait for more credit to become + * available. + * + * We also choose the server and allocate an operation ID to be cleaned up + * later. + */ +static bool cifs_clamp_length(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *rreq = subreq->rreq; + struct TCP_Server_Info *server; + struct cifs_io_subrequest *rdata = container_of(subreq, struct cifs_io_subrequest, subreq); + struct cifs_io_request *req = container_of(subreq->rreq, struct cifs_io_request, rreq); + struct cifs_sb_info *cifs_sb = CIFS_SB(rreq->inode->i_sb); + size_t rsize = 0; + int rc; + + rdata->xid = get_xid(); + rdata->have_xid = true; + + server = cifs_pick_channel(tlink_tcon(req->cfile->tlink)->ses); + rdata->server = server; + + if (cifs_sb->ctx->rsize == 0) + cifs_sb->ctx->rsize = + server->ops->negotiate_rsize(tlink_tcon(req->cfile->tlink), + cifs_sb->ctx); + + + rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->rsize, &rsize, + &rdata->credits); + if (rc) { + subreq->error = rc; + return false; + } + + subreq->len = min_t(size_t, subreq->len, rsize); +#ifdef CONFIG_CIFS_SMB_DIRECT + if (server->smbd_conn) + subreq->max_nr_segs = server->smbd_conn->max_frmr_depth; +#endif + return true; +} + +/* + * Issue a read operation on behalf of the netfs helper functions. We're asked + * to make a read of a certain size at a point in the file. We are permitted + * to only read a portion of that, but as long as we read something, the netfs + * helper will call us again so that we can issue another read. + */ +static void cifs_req_issue_read(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *rreq = subreq->rreq; + struct cifs_io_subrequest *rdata = container_of(subreq, struct cifs_io_subrequest, subreq); + struct cifs_io_request *req = container_of(subreq->rreq, struct cifs_io_request, rreq); + struct cifs_sb_info *cifs_sb = CIFS_SB(rreq->inode->i_sb); + pid_t pid; + int rc = 0; + + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + pid = req->cfile->pid; + else + pid = current->tgid; // Ummm... This may be a workqueue + + cifs_dbg(FYI, "%s: op=%08x[%x] mapping=%p len=%zu/%zu\n", + __func__, rreq->debug_id, subreq->debug_index, rreq->mapping, + subreq->transferred, subreq->len); + + if (req->cfile->invalidHandle) { + do { + rc = cifs_reopen_file(req->cfile, true); + } while (rc == -EAGAIN); + if (rc) + goto out; + } + + __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); + rdata->pid = pid; + + rc = adjust_credits(rdata->server, &rdata->credits, rdata->subreq.len); + if (!rc) { + if (rdata->req->cfile->invalidHandle) + rc = -EAGAIN; + else + rc = rdata->server->ops->async_readv(rdata); + } + +out: + if (rc) + netfs_subreq_terminated(subreq, rc, false); +} + +/* + * Writeback calls this when it finds a folio that needs uploading. This isn't + * called if writeback only has copy-to-cache to deal with. + */ +static void cifs_begin_writeback(struct netfs_io_request *wreq) +{ + struct cifs_io_request *req = container_of(wreq, struct cifs_io_request, rreq); + int ret; + + ret = cifs_get_writable_file(CIFS_I(wreq->inode), FIND_WR_ANY, &req->cfile); + if (ret) { + cifs_dbg(VFS, "No writable handle in writepages ret=%d\n", ret); + return; + } + + wreq->io_streams[0].avail = true; +} + +/* + * Initialise a request. + */ +static int cifs_init_request(struct netfs_io_request *rreq, struct file *file) +{ + struct cifs_io_request *req = container_of(rreq, struct cifs_io_request, rreq); + struct cifs_sb_info *cifs_sb = CIFS_SB(rreq->inode->i_sb); + struct cifsFileInfo *open_file = NULL; + + rreq->rsize = cifs_sb->ctx->rsize; + rreq->wsize = cifs_sb->ctx->wsize; + + if (file) { + open_file = file->private_data; + rreq->netfs_priv = file->private_data; + req->cfile = cifsFileInfo_get(open_file); + } else if (rreq->origin != NETFS_WRITEBACK) { + WARN_ON_ONCE(1); + return -EIO; + } + + return 0; +} + +/* + * Expand the size of a readahead to the size of the rsize, if at least as + * large as a page, allowing for the possibility that rsize is not pow-2 + * aligned. + */ +static void cifs_expand_readahead(struct netfs_io_request *rreq) +{ + unsigned int rsize = rreq->rsize; + loff_t misalignment, i_size = i_size_read(rreq->inode); + + if (rsize < PAGE_SIZE) + return; + + if (rsize < INT_MAX) + rsize = roundup_pow_of_two(rsize); + else + rsize = ((unsigned int)INT_MAX + 1) / 2; + + misalignment = rreq->start & (rsize - 1); + if (misalignment) { + rreq->start -= misalignment; + rreq->len += misalignment; + } + + rreq->len = round_up(rreq->len, rsize); + if (rreq->start < i_size && rreq->len > i_size - rreq->start) + rreq->len = i_size - rreq->start; +} + +/* + * Completion of a request operation. + */ +static void cifs_rreq_done(struct netfs_io_request *rreq) +{ + struct timespec64 atime, mtime; + struct inode *inode = rreq->inode; + + /* we do not want atime to be less than mtime, it broke some apps */ + atime = inode_set_atime_to_ts(inode, current_time(inode)); + mtime = inode_get_mtime(inode); + if (timespec64_compare(&atime, &mtime)) + inode_set_atime_to_ts(inode, inode_get_mtime(inode)); +} + +static void cifs_post_modify(struct inode *inode) +{ + /* Indication to update ctime and mtime as close is deferred */ + set_bit(CIFS_INO_MODIFIED_ATTR, &CIFS_I(inode)->flags); +} + +static void cifs_free_request(struct netfs_io_request *rreq) +{ + struct cifs_io_request *req = container_of(rreq, struct cifs_io_request, rreq); + + if (req->cfile) + cifsFileInfo_put(req->cfile); +} + +static void cifs_free_subrequest(struct netfs_io_subrequest *subreq) +{ + struct cifs_io_subrequest *rdata = + container_of(subreq, struct cifs_io_subrequest, subreq); + int rc = subreq->error; + + if (rdata->subreq.source == NETFS_DOWNLOAD_FROM_SERVER) { +#ifdef CONFIG_CIFS_SMB_DIRECT + if (rdata->mr) { + smbd_deregister_mr(rdata->mr); + rdata->mr = NULL; + } +#endif + } + + add_credits_and_wake_if(rdata->server, &rdata->credits, 0); + if (rdata->have_xid) + free_xid(rdata->xid); +} + +const struct netfs_request_ops cifs_req_ops = { + .request_pool = &cifs_io_request_pool, + .subrequest_pool = &cifs_io_subrequest_pool, + .init_request = cifs_init_request, + .free_request = cifs_free_request, + .free_subrequest = cifs_free_subrequest, + .expand_readahead = cifs_expand_readahead, + .clamp_length = cifs_clamp_length, + .issue_read = cifs_req_issue_read, + .done = cifs_rreq_done, + .post_modify = cifs_post_modify, + .begin_writeback = cifs_begin_writeback, + .prepare_write = cifs_prepare_write, + .issue_write = cifs_issue_write, +}; /* * Remove the dirty flags from a span of pages. diff --git a/include/linux/netfs.h b/include/linux/netfs.h index 298552f5122c..f45d06284f2f 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -302,6 +302,7 @@ struct netfs_request_ops { /* Modification handling */ void (*update_i_size)(struct inode *inode, loff_t i_size); + void (*post_modify)(struct inode *inode); /* Write request handling */ void (*begin_writeback)(struct netfs_io_request *wreq); diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h index 4ba553a6d71b..da23484268df 100644 --- a/include/trace/events/netfs.h +++ b/include/trace/events/netfs.h @@ -112,6 +112,7 @@ #define netfs_sreq_ref_traces \ EM(netfs_sreq_trace_get_copy_to_cache, "GET COPY2C ") \ EM(netfs_sreq_trace_get_resubmit, "GET RESUBMIT") \ + EM(netfs_sreq_trace_get_submit, "GET SUBMIT") \ EM(netfs_sreq_trace_get_short_read, "GET SHORTRD") \ EM(netfs_sreq_trace_new, "NEW ") \ EM(netfs_sreq_trace_put_cancel, "PUT CANCEL ") \ -- cgit v1.2.3-59-g8ed1b From 3ee1a1fc39819906f04d6c62c180e760cd3a689d Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 18:29:59 +0100 Subject: cifs: Cut over to using netfslib Make the cifs filesystem use netfslib to handle reading and writing on behalf of cifs. The changes include: (1) Various read_iter/write_iter type functions are turned into wrappers around netfslib API functions or are pointed directly at those functions: cifs_file_direct{,_nobrl}_ops switch to use netfs_unbuffered_read_iter and netfs_unbuffered_write_iter. Large pieces of code that will be removed are #if'd out and will be removed in subsequent patches. [?] Why does cifs mark the page dirty in the destination buffer of a DIO read? Should that happen automatically? Does netfs need to do that? Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/netfs/io.c | 7 +- fs/smb/client/cifsfs.c | 8 +-- fs/smb/client/cifsfs.h | 7 -- fs/smb/client/cifsglob.h | 5 +- fs/smb/client/cifsproto.h | 8 ++- fs/smb/client/cifssmb.c | 54 +++++++++------- fs/smb/client/file.c | 161 +++++++++++++++++++++++++--------------------- fs/smb/client/fscache.c | 2 + fs/smb/client/fscache.h | 4 ++ fs/smb/client/inode.c | 19 +++++- fs/smb/client/smb2pdu.c | 117 +++++++++++++++++++-------------- fs/smb/client/smb2proto.h | 2 +- fs/smb/client/trace.h | 144 +++++++++++++++++++++++++++++++++++------ fs/smb/client/transport.c | 3 + 14 files changed, 361 insertions(+), 180 deletions(-) (limited to 'fs') diff --git a/fs/netfs/io.c b/fs/netfs/io.c index 6cfecfcd02e1..c93851b98368 100644 --- a/fs/netfs/io.c +++ b/fs/netfs/io.c @@ -213,8 +213,13 @@ static void netfs_rreq_assess_dio(struct netfs_io_request *rreq) unsigned int i; size_t transferred = 0; - for (i = 0; i < rreq->direct_bv_count; i++) + for (i = 0; i < rreq->direct_bv_count; i++) { flush_dcache_page(rreq->direct_bv[i].bv_page); + // TODO: cifs marks pages in the destination buffer + // dirty under some circumstances after a read. Do we + // need to do that too? + set_page_dirty(rreq->direct_bv[i].bv_page); + } list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { if (subreq->error || subreq->transferred == 0) diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 8f7165567be7..6e1698614745 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -1522,8 +1522,8 @@ const struct file_operations cifs_file_strict_ops = { }; const struct file_operations cifs_file_direct_ops = { - .read_iter = cifs_direct_readv, - .write_iter = cifs_direct_writev, + .read_iter = netfs_unbuffered_read_iter, + .write_iter = netfs_file_write_iter, .open = cifs_open, .release = cifs_close, .lock = cifs_lock, @@ -1578,8 +1578,8 @@ const struct file_operations cifs_file_strict_nobrl_ops = { }; const struct file_operations cifs_file_direct_nobrl_ops = { - .read_iter = cifs_direct_readv, - .write_iter = cifs_direct_writev, + .read_iter = netfs_unbuffered_read_iter, + .write_iter = netfs_file_write_iter, .open = cifs_open, .release = cifs_close, .fsync = cifs_fsync, diff --git a/fs/smb/client/cifsfs.h b/fs/smb/client/cifsfs.h index 922c10d7cfdd..87310f05d397 100644 --- a/fs/smb/client/cifsfs.h +++ b/fs/smb/client/cifsfs.h @@ -94,11 +94,7 @@ extern const struct file_operations cifs_file_strict_nobrl_ops; extern int cifs_open(struct inode *inode, struct file *file); extern int cifs_close(struct inode *inode, struct file *file); extern int cifs_closedir(struct inode *inode, struct file *file); -extern ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to); -extern ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to); extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to); -extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from); -extern ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from); extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from); ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from); ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter); @@ -112,9 +108,6 @@ extern int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma); extern const struct file_operations cifs_dir_ops; extern int cifs_dir_open(struct inode *inode, struct file *file); extern int cifs_readdir(struct file *file, struct dir_context *ctx); -extern void cifs_pages_written_back(struct inode *inode, loff_t start, unsigned int len); -extern void cifs_pages_write_failed(struct inode *inode, loff_t start, unsigned int len); -extern void cifs_pages_write_redirty(struct inode *inode, loff_t start, unsigned int len); /* Functions related to dir entries */ extern const struct dentry_operations cifs_dentry_ops; diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 4e9033b2f191..983860bf5fbb 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -451,7 +451,7 @@ struct smb_version_operations { /* async read from the server */ int (*async_readv)(struct cifs_io_subrequest *); /* async write to the server */ - int (*async_writev)(struct cifs_io_subrequest *); + void (*async_writev)(struct cifs_io_subrequest *); /* sync read from the server */ int (*sync_read)(const unsigned int, struct cifs_fid *, struct cifs_io_parms *, unsigned int *, char **, @@ -1516,7 +1516,7 @@ struct cifs_io_subrequest { #endif struct cifs_credits credits; - // TODO: Remove following elements +#if 0 // TODO: Remove following elements struct list_head list; struct completion done; struct work_struct work; @@ -1526,6 +1526,7 @@ struct cifs_io_subrequest { enum writeback_sync_modes sync_mode; bool uncached; struct bio_vec *bv; +#endif }; /* diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index 103f264b42d7..d46ad86150cd 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -148,6 +148,8 @@ extern bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 eof, bool from_readdir); extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, unsigned int bytes_written); +void cifs_write_subrequest_terminated(struct cifs_io_subrequest *wdata, ssize_t result, + bool was_async); extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, int); extern int cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags, @@ -599,6 +601,7 @@ void __cifs_put_smb_ses(struct cifs_ses *ses); extern struct cifs_ses * cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx); +#if 0 // TODO Remove void cifs_readdata_release(struct cifs_io_subrequest *rdata); static inline void cifs_get_readdata(struct cifs_io_subrequest *rdata) { @@ -609,11 +612,13 @@ static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata) if (refcount_dec_and_test(&rdata->subreq.ref)) cifs_readdata_release(rdata); } +#endif int cifs_async_readv(struct cifs_io_subrequest *rdata); int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid); -int cifs_async_writev(struct cifs_io_subrequest *wdata); +void cifs_async_writev(struct cifs_io_subrequest *wdata); void cifs_writev_complete(struct work_struct *work); +#if 0 // TODO Remove struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete); void cifs_writedata_release(struct cifs_io_subrequest *rdata); static inline void cifs_get_writedata(struct cifs_io_subrequest *wdata) @@ -625,6 +630,7 @@ static inline void cifs_put_writedata(struct cifs_io_subrequest *wdata) if (refcount_dec_and_test(&wdata->subreq.ref)) cifs_writedata_release(wdata); } +#endif int cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, const unsigned char *path, char *pbuf, diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index 99b75102ba6b..25e9ab947c17 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -1265,7 +1265,7 @@ static void cifs_readv_callback(struct mid_q_entry *mid) { struct cifs_io_subrequest *rdata = mid->callback_data; - struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(rdata->req->cfile->tlink); struct TCP_Server_Info *server = tcon->ses->server; struct smb_rqst rqst = { .rq_iov = rdata->iov, .rq_nvec = 2, @@ -1306,7 +1306,13 @@ cifs_readv_callback(struct mid_q_entry *mid) rdata->result = -EIO; } - queue_work(cifsiod_wq, &rdata->work); + if (rdata->result == 0 || rdata->result == -EAGAIN) + iov_iter_advance(&rdata->subreq.io_iter, rdata->got_bytes); + rdata->credits.value = 0; + netfs_subreq_terminated(&rdata->subreq, + (rdata->result == 0 || rdata->result == -EAGAIN) ? + rdata->got_bytes : rdata->result, + false); release_mid(mid); add_credits(server, &credits, 0); } @@ -1318,7 +1324,7 @@ cifs_async_readv(struct cifs_io_subrequest *rdata) int rc; READ_REQ *smb = NULL; int wct; - struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(rdata->req->cfile->tlink); struct smb_rqst rqst = { .rq_iov = rdata->iov, .rq_nvec = 2 }; @@ -1343,7 +1349,7 @@ cifs_async_readv(struct cifs_io_subrequest *rdata) smb->hdr.PidHigh = cpu_to_le16((__u16)(rdata->pid >> 16)); smb->AndXCommand = 0xFF; /* none */ - smb->Fid = rdata->cfile->fid.netfid; + smb->Fid = rdata->req->cfile->fid.netfid; smb->OffsetLow = cpu_to_le32(rdata->subreq.start & 0xFFFFFFFF); if (wct == 12) smb->OffsetHigh = cpu_to_le32(rdata->subreq.start >> 32); @@ -1613,15 +1619,16 @@ static void cifs_writev_callback(struct mid_q_entry *mid) { struct cifs_io_subrequest *wdata = mid->callback_data; - struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); - unsigned int written; + struct cifs_tcon *tcon = tlink_tcon(wdata->req->cfile->tlink); WRITE_RSP *smb = (WRITE_RSP *)mid->resp_buf; struct cifs_credits credits = { .value = 1, .instance = 0 }; + ssize_t result; + size_t written; switch (mid->mid_state) { case MID_RESPONSE_RECEIVED: - wdata->result = cifs_check_receive(mid, tcon->ses->server, 0); - if (wdata->result != 0) + result = cifs_check_receive(mid, tcon->ses->server, 0); + if (result != 0) break; written = le16_to_cpu(smb->CountHigh); @@ -1637,32 +1644,33 @@ cifs_writev_callback(struct mid_q_entry *mid) written &= 0xFFFF; if (written < wdata->subreq.len) - wdata->result = -ENOSPC; + result = -ENOSPC; else - wdata->subreq.len = written; + result = written; break; case MID_REQUEST_SUBMITTED: case MID_RETRY_NEEDED: - wdata->result = -EAGAIN; + result = -EAGAIN; break; default: - wdata->result = -EIO; + result = -EIO; break; } - queue_work(cifsiod_wq, &wdata->work); + wdata->credits.value = 0; + cifs_write_subrequest_terminated(wdata, result, true); release_mid(mid); add_credits(tcon->ses->server, &credits, 0); } /* cifs_async_writev - send an async write, and set up mid to handle result */ -int +void cifs_async_writev(struct cifs_io_subrequest *wdata) { int rc = -EACCES; WRITE_REQ *smb = NULL; int wct; - struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(wdata->req->cfile->tlink); struct kvec iov[2]; struct smb_rqst rqst = { }; @@ -1672,7 +1680,8 @@ cifs_async_writev(struct cifs_io_subrequest *wdata) wct = 12; if (wdata->subreq.start >> 32 > 0) { /* can not handle big offset for old srv */ - return -EIO; + rc = -EIO; + goto out; } } @@ -1684,7 +1693,7 @@ cifs_async_writev(struct cifs_io_subrequest *wdata) smb->hdr.PidHigh = cpu_to_le16((__u16)(wdata->pid >> 16)); smb->AndXCommand = 0xFF; /* none */ - smb->Fid = wdata->cfile->fid.netfid; + smb->Fid = wdata->req->cfile->fid.netfid; smb->OffsetLow = cpu_to_le32(wdata->subreq.start & 0xFFFFFFFF); if (wct == 14) smb->OffsetHigh = cpu_to_le32(wdata->subreq.start >> 32); @@ -1724,18 +1733,19 @@ cifs_async_writev(struct cifs_io_subrequest *wdata) iov[1].iov_len += 4; /* pad bigger by four bytes */ } - cifs_get_writedata(wdata); rc = cifs_call_async(tcon->ses->server, &rqst, NULL, cifs_writev_callback, NULL, wdata, 0, NULL); - + /* Can't touch wdata if rc == 0 */ if (rc == 0) cifs_stats_inc(&tcon->stats.cifs_stats.num_writes); - else - cifs_put_writedata(wdata); async_writev_out: cifs_small_buf_release(smb); - return rc; +out: + if (rc) { + add_credits_and_wake_if(wdata->server, &wdata->credits, 0); + cifs_write_subrequest_terminated(wdata, rc, false); + } } int diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index c9c33b3a54f3..bb3adde9343e 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -119,7 +119,7 @@ fail: else trace_netfs_sreq(subreq, netfs_sreq_trace_fail); add_credits_and_wake_if(wdata->server, &wdata->credits, 0); - netfs_write_subrequest_terminated(wdata, rc, false); + cifs_write_subrequest_terminated(wdata, rc, false); goto out; } @@ -352,6 +352,7 @@ const struct netfs_request_ops cifs_req_ops = { .issue_write = cifs_issue_write, }; +#if 0 // TODO remove 397 /* * Remove the dirty flags from a span of pages. */ @@ -476,6 +477,7 @@ void cifs_pages_write_redirty(struct inode *inode, loff_t start, unsigned int le rcu_read_unlock(); } +#endif // end netfslib remove 397 /* * Mark as invalid, all open files on tree connections since they @@ -2522,20 +2524,23 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *flock) return rc; } -/* - * update the file size (if needed) after a write. Should be called with - * the inode->i_lock held - */ -void -cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, - unsigned int bytes_written) +void cifs_write_subrequest_terminated(struct cifs_io_subrequest *wdata, ssize_t result, + bool was_async) { - loff_t end_of_write = offset + bytes_written; + struct netfs_io_request *wreq = wdata->rreq; + loff_t new_server_eof; - if (end_of_write > cifsi->netfs.remote_i_size) - netfs_resize_file(&cifsi->netfs, end_of_write, true); + if (result > 0) { + new_server_eof = wdata->subreq.start + wdata->subreq.transferred + result; + + if (new_server_eof > netfs_inode(wreq->inode)->remote_i_size) + netfs_resize_file(netfs_inode(wreq->inode), new_server_eof, true); + } + + netfs_write_subrequest_terminated(&wdata->subreq, result, was_async); } +#if 0 // TODO remove 2483 static ssize_t cifs_write(struct cifsFileInfo *open_file, __u32 pid, const char *write_data, size_t write_size, loff_t *offset) @@ -2619,6 +2624,7 @@ cifs_write(struct cifsFileInfo *open_file, __u32 pid, const char *write_data, free_xid(xid); return total_written; } +#endif // end netfslib remove 2483 struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) @@ -2824,6 +2830,7 @@ cifs_get_readable_path(struct cifs_tcon *tcon, const char *name, return -ENOENT; } +#if 0 // TODO remove 2773 void cifs_writedata_release(struct cifs_io_subrequest *wdata) { @@ -3454,7 +3461,11 @@ static int cifs_write_end(struct file *file, struct address_space *mapping, return rc; } +#endif // End netfs removal 2773 +/* + * Flush data on a strict file. + */ int cifs_strict_fsync(struct file *file, loff_t start, loff_t end, int datasync) { @@ -3509,6 +3520,9 @@ strict_fsync_exit: return rc; } +/* + * Flush data on a non-strict data. + */ int cifs_fsync(struct file *file, loff_t start, loff_t end, int datasync) { unsigned int xid; @@ -3575,6 +3589,7 @@ int cifs_flush(struct file *file, fl_owner_t id) return rc; } +#if 0 // TODO remove 3594 static void collect_uncached_write_data(struct cifs_aio_ctx *ctx); static void @@ -4037,6 +4052,7 @@ ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) { return __cifs_writev(iocb, from, false); } +#endif // TODO remove 3594 static ssize_t cifs_writev(struct kiocb *iocb, struct iov_iter *from) @@ -4048,7 +4064,10 @@ cifs_writev(struct kiocb *iocb, struct iov_iter *from) struct TCP_Server_Info *server = tlink_tcon(cfile->tlink)->ses->server; ssize_t rc; - inode_lock(inode); + rc = netfs_start_io_write(inode); + if (rc < 0) + return rc; + /* * We need to hold the sem to be sure nobody modifies lock list * with a brlock that prevents writing. @@ -4062,13 +4081,12 @@ cifs_writev(struct kiocb *iocb, struct iov_iter *from) if (!cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(from), server->vals->exclusive_lock_type, 0, NULL, CIFS_WRITE_OP)) - rc = __generic_file_write_iter(iocb, from); + rc = netfs_buffered_write_iter_locked(iocb, from, NULL); else rc = -EACCES; out: up_read(&cinode->lock_sem); - inode_unlock(inode); - + netfs_end_io_write(inode); if (rc > 0) rc = generic_write_sync(iocb, rc); return rc; @@ -4091,9 +4109,9 @@ cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from) if (CIFS_CACHE_WRITE(cinode)) { if (cap_unix(tcon->ses) && - (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) - && ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) { - written = generic_file_write_iter(iocb, from); + (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) && + ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) { + written = netfs_file_write_iter(iocb, from); goto out; } written = cifs_writev(iocb, from); @@ -4105,7 +4123,7 @@ cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from) * affected pages because it may cause a error with mandatory locks on * these pages but not on the region from pos to ppos+len-1. */ - written = cifs_user_writev(iocb, from); + written = netfs_file_write_iter(iocb, from); if (CIFS_CACHE_READ(cinode)) { /* * We have read level caching and we have just sent a write @@ -4124,6 +4142,7 @@ out: return written; } +#if 0 // TODO remove 4143 static struct cifs_io_subrequest *cifs_readdata_alloc(work_func_t complete) { struct cifs_io_subrequest *rdata; @@ -4563,7 +4582,9 @@ ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to) ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to) { return __cifs_readv(iocb, to, false); + } +#endif // end netfslib removal 4143 ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter) { @@ -4571,13 +4592,13 @@ ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter) struct inode *inode = file_inode(iocb->ki_filp); if (iocb->ki_flags & IOCB_DIRECT) - return cifs_user_readv(iocb, iter); + return netfs_unbuffered_read_iter(iocb, iter); rc = cifs_revalidate_mapping(inode); if (rc) return rc; - return generic_file_read_iter(iocb, iter); + return netfs_file_read_iter(iocb, iter); } ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) @@ -4588,7 +4609,7 @@ ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) int rc; if (iocb->ki_filp->f_flags & O_DIRECT) { - written = cifs_user_writev(iocb, from); + written = netfs_unbuffered_write_iter(iocb, from); if (written > 0 && CIFS_CACHE_READ(cinode)) { cifs_zap_mapping(inode); cifs_dbg(FYI, @@ -4603,17 +4624,15 @@ ssize_t cifs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (written) return written; - written = generic_file_write_iter(iocb, from); - - if (CIFS_CACHE_WRITE(CIFS_I(inode))) - goto out; + written = netfs_file_write_iter(iocb, from); - rc = filemap_fdatawrite(inode->i_mapping); - if (rc) - cifs_dbg(FYI, "cifs_file_write_iter: %d rc on %p inode\n", - rc, inode); + if (!CIFS_CACHE_WRITE(CIFS_I(inode))) { + rc = filemap_fdatawrite(inode->i_mapping); + if (rc) + cifs_dbg(FYI, "cifs_file_write_iter: %d rc on %p inode\n", + rc, inode); + } -out: cifs_put_writer(cinode); return written; } @@ -4638,12 +4657,15 @@ cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) * pos+len-1. */ if (!CIFS_CACHE_READ(cinode)) - return cifs_user_readv(iocb, to); + return netfs_unbuffered_read_iter(iocb, to); if (cap_unix(tcon->ses) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) && - ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) - return generic_file_read_iter(iocb, to); + ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) { + if (iocb->ki_flags & IOCB_DIRECT) + return netfs_unbuffered_read_iter(iocb, to); + return netfs_buffered_read_iter(iocb, to); + } /* * We need to hold the sem to be sure nobody modifies lock list @@ -4652,12 +4674,17 @@ cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) down_read(&cinode->lock_sem); if (!cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(to), tcon->ses->server->vals->shared_lock_type, - 0, NULL, CIFS_READ_OP)) - rc = generic_file_read_iter(iocb, to); + 0, NULL, CIFS_READ_OP)) { + if (iocb->ki_flags & IOCB_DIRECT) + rc = netfs_unbuffered_read_iter(iocb, to); + else + rc = netfs_buffered_read_iter(iocb, to); + } up_read(&cinode->lock_sem); return rc; } +#if 0 // TODO remove 4633 static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset) { @@ -4749,29 +4776,11 @@ cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset) free_xid(xid); return total_read; } +#endif // end netfslib remove 4633 -/* - * If the page is mmap'ed into a process' page tables, then we need to make - * sure that it doesn't change while being written back. - */ static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf) { - struct folio *folio = page_folio(vmf->page); - - /* Wait for the folio to be written to the cache before we allow it to - * be modified. We then assume the entire folio will need writing back. - */ -#ifdef CONFIG_CIFS_FSCACHE - if (folio_test_private_2(folio) && /* [DEPRECATED] */ - folio_wait_private_2_killable(folio) < 0) - return VM_FAULT_RETRY; -#endif - - folio_wait_writeback(folio); - - if (folio_lock_killable(folio) < 0) - return VM_FAULT_RETRY; - return VM_FAULT_LOCKED; + return netfs_page_mkwrite(vmf, NULL); } static const struct vm_operations_struct cifs_file_vm_ops = { @@ -4817,6 +4826,7 @@ int cifs_file_mmap(struct file *file, struct vm_area_struct *vma) return rc; } +#if 0 // TODO remove 4794 /* * Unlock a bunch of folios in the pagecache. */ @@ -5101,6 +5111,7 @@ static int cifs_read_folio(struct file *file, struct folio *folio) free_xid(xid); return rc; } +#endif // end netfslib remove 4794 static int is_inode_writable(struct cifsInodeInfo *cifs_inode) { @@ -5149,6 +5160,7 @@ bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file, return true; } +#if 0 // TODO remove 5152 static int cifs_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, struct page **pagep, void **fsdata) @@ -5244,6 +5256,7 @@ static void cifs_invalidate_folio(struct folio *folio, size_t offset, { folio_wait_private_2(folio); /* [DEPRECATED] */ } +#endif // end netfslib remove 5152 void cifs_oplock_break(struct work_struct *work) { @@ -5334,6 +5347,7 @@ out: cifs_done_oplock_break(cinode); } +#if 0 // TODO remove 5333 /* * The presence of cifs_direct_io() in the address space ops vector * allowes open() O_DIRECT flags which would have failed otherwise. @@ -5352,6 +5366,7 @@ cifs_direct_io(struct kiocb *iocb, struct iov_iter *iter) */ return -EINVAL; } +#endif // netfs end remove 5333 static int cifs_swap_activate(struct swap_info_struct *sis, struct file *swap_file, sector_t *span) @@ -5414,21 +5429,19 @@ static void cifs_swap_deactivate(struct file *file) } const struct address_space_operations cifs_addr_ops = { - .read_folio = cifs_read_folio, - .readahead = cifs_readahead, - .writepages = cifs_writepages, - .write_begin = cifs_write_begin, - .write_end = cifs_write_end, - .dirty_folio = netfs_dirty_folio, - .release_folio = cifs_release_folio, - .direct_IO = cifs_direct_io, - .invalidate_folio = cifs_invalidate_folio, - .migrate_folio = filemap_migrate_folio, + .read_folio = netfs_read_folio, + .readahead = netfs_readahead, + .writepages = netfs_writepages, + .dirty_folio = netfs_dirty_folio, + .release_folio = netfs_release_folio, + .direct_IO = noop_direct_IO, + .invalidate_folio = netfs_invalidate_folio, + .migrate_folio = filemap_migrate_folio, /* * TODO: investigate and if useful we could add an is_dirty_writeback * helper if needed */ - .swap_activate = cifs_swap_activate, + .swap_activate = cifs_swap_activate, .swap_deactivate = cifs_swap_deactivate, }; @@ -5438,12 +5451,10 @@ const struct address_space_operations cifs_addr_ops = { * to leave cifs_readahead out of the address space operations. */ const struct address_space_operations cifs_addr_ops_smallbuf = { - .read_folio = cifs_read_folio, - .writepages = cifs_writepages, - .write_begin = cifs_write_begin, - .write_end = cifs_write_end, - .dirty_folio = netfs_dirty_folio, - .release_folio = cifs_release_folio, - .invalidate_folio = cifs_invalidate_folio, - .migrate_folio = filemap_migrate_folio, + .read_folio = netfs_read_folio, + .writepages = netfs_writepages, + .dirty_folio = netfs_dirty_folio, + .release_folio = netfs_release_folio, + .invalidate_folio = netfs_invalidate_folio, + .migrate_folio = filemap_migrate_folio, }; diff --git a/fs/smb/client/fscache.c b/fs/smb/client/fscache.c index 1a895e6243ee..b36c493f4c56 100644 --- a/fs/smb/client/fscache.c +++ b/fs/smb/client/fscache.c @@ -171,6 +171,7 @@ void cifs_fscache_release_inode_cookie(struct inode *inode) } } +#if 0 // TODO remove /* * Fallback page reading interface. */ @@ -279,3 +280,4 @@ int __cifs_fscache_query_occupancy(struct inode *inode, fscache_end_operation(&cres); return ret; } +#endif diff --git a/fs/smb/client/fscache.h b/fs/smb/client/fscache.h index 1f2ea9f5cc9a..08b30f79d4cd 100644 --- a/fs/smb/client/fscache.h +++ b/fs/smb/client/fscache.h @@ -74,6 +74,7 @@ static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags i_size_read(inode), flags); } +#if 0 // TODO remove extern int __cifs_fscache_query_occupancy(struct inode *inode, pgoff_t first, unsigned int nr_pages, pgoff_t *_data_first, @@ -108,6 +109,7 @@ static inline void cifs_readahead_to_fscache(struct inode *inode, if (cifs_inode_cookie(inode)) __cifs_readahead_to_fscache(inode, pos, len); } +#endif static inline bool cifs_fscache_enabled(struct inode *inode) { @@ -131,6 +133,7 @@ static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { re static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) {} static inline bool cifs_fscache_enabled(struct inode *inode) { return false; } +#if 0 // TODO remove static inline int cifs_fscache_query_occupancy(struct inode *inode, pgoff_t first, unsigned int nr_pages, pgoff_t *_data_first, @@ -149,6 +152,7 @@ cifs_readpage_from_fscache(struct inode *inode, struct page *page) static inline void cifs_readahead_to_fscache(struct inode *inode, loff_t pos, size_t len) {} +#endif #endif /* CONFIG_CIFS_FSCACHE */ diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index 033948d34c5a..5239c823de86 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -28,14 +28,29 @@ #include "cached_dir.h" #include "reparse.h" +/* + * Set parameters for the netfs library + */ +static void cifs_set_netfs_context(struct inode *inode) +{ + struct cifsInodeInfo *cifs_i = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + + netfs_inode_init(&cifs_i->netfs, &cifs_req_ops, true); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) + __set_bit(NETFS_ICTX_WRITETHROUGH, &cifs_i->netfs.flags); +} + static void cifs_set_ops(struct inode *inode) { struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct netfs_inode *ictx = netfs_inode(inode); switch (inode->i_mode & S_IFMT) { case S_IFREG: inode->i_op = &cifs_file_inode_ops; if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { + set_bit(NETFS_ICTX_UNBUFFERED, &ictx->flags); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) inode->i_fop = &cifs_file_direct_nobrl_ops; else @@ -221,8 +236,10 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr, if (fattr->cf_flags & CIFS_FATTR_JUNCTION) inode->i_flags |= S_AUTOMOUNT; - if (inode->i_state & I_NEW) + if (inode->i_state & I_NEW) { + cifs_set_netfs_context(inode); cifs_set_ops(inode); + } return 0; } diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index b063f5e95f5d..993ac36c3d58 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -4421,10 +4421,12 @@ smb2_new_read_req(void **buf, unsigned int *total_len, req->Length = cpu_to_le32(io_parms->length); req->Offset = cpu_to_le64(io_parms->offset); - trace_smb3_read_enter(0 /* xid */, - io_parms->persistent_fid, - io_parms->tcon->tid, io_parms->tcon->ses->Suid, - io_parms->offset, io_parms->length); + trace_smb3_read_enter(rdata ? rdata->rreq->debug_id : 0, + rdata ? rdata->subreq.debug_index : 0, + rdata ? rdata->xid : 0, + io_parms->persistent_fid, + io_parms->tcon->tid, io_parms->tcon->ses->Suid, + io_parms->offset, io_parms->length); #ifdef CONFIG_CIFS_SMB_DIRECT /* * If we want to do a RDMA write, fill in and append @@ -4486,7 +4488,7 @@ static void smb2_readv_callback(struct mid_q_entry *mid) { struct cifs_io_subrequest *rdata = mid->callback_data; - struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(rdata->req->cfile->tlink); struct TCP_Server_Info *server = rdata->server; struct smb2_hdr *shdr = (struct smb2_hdr *)rdata->iov[0].iov_base; @@ -4514,7 +4516,6 @@ smb2_readv_callback(struct mid_q_entry *mid) if (server->sign && !mid->decrypted) { int rc; - iov_iter_revert(&rqst.rq_iter, rdata->got_bytes); iov_iter_truncate(&rqst.rq_iter, rdata->got_bytes); rc = smb2_verify_signature(&rqst, server); if (rc) @@ -4555,17 +4556,33 @@ smb2_readv_callback(struct mid_q_entry *mid) #endif if (rdata->result && rdata->result != -ENODATA) { cifs_stats_fail_inc(tcon, SMB2_READ_HE); - trace_smb3_read_err(0 /* xid */, - rdata->cfile->fid.persistent_fid, + trace_smb3_read_err(rdata->rreq->debug_id, + rdata->subreq.debug_index, + rdata->xid, + rdata->req->cfile->fid.persistent_fid, tcon->tid, tcon->ses->Suid, rdata->subreq.start, rdata->subreq.len, rdata->result); } else - trace_smb3_read_done(0 /* xid */, - rdata->cfile->fid.persistent_fid, + trace_smb3_read_done(rdata->rreq->debug_id, + rdata->subreq.debug_index, + rdata->xid, + rdata->req->cfile->fid.persistent_fid, tcon->tid, tcon->ses->Suid, rdata->subreq.start, rdata->got_bytes); - queue_work(cifsiod_wq, &rdata->work); + if (rdata->result == -ENODATA) { + /* We may have got an EOF error because fallocate + * failed to enlarge the file. + */ + if (rdata->subreq.start < rdata->subreq.rreq->i_size) + rdata->result = 0; + } + if (rdata->result == 0 || rdata->result == -EAGAIN) + iov_iter_advance(&rdata->subreq.io_iter, rdata->got_bytes); + rdata->credits.value = 0; + netfs_subreq_terminated(&rdata->subreq, + (rdata->result == 0 || rdata->result == -EAGAIN) ? + rdata->got_bytes : rdata->result, true); release_mid(mid); add_credits(server, &credits, 0); } @@ -4581,7 +4598,7 @@ smb2_async_readv(struct cifs_io_subrequest *rdata) struct smb_rqst rqst = { .rq_iov = rdata->iov, .rq_nvec = 1 }; struct TCP_Server_Info *server; - struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(rdata->req->cfile->tlink); unsigned int total_len; int credit_request; @@ -4591,12 +4608,12 @@ smb2_async_readv(struct cifs_io_subrequest *rdata) if (!rdata->server) rdata->server = cifs_pick_channel(tcon->ses); - io_parms.tcon = tlink_tcon(rdata->cfile->tlink); + io_parms.tcon = tlink_tcon(rdata->req->cfile->tlink); io_parms.server = server = rdata->server; io_parms.offset = rdata->subreq.start; io_parms.length = rdata->subreq.len; - io_parms.persistent_fid = rdata->cfile->fid.persistent_fid; - io_parms.volatile_fid = rdata->cfile->fid.volatile_fid; + io_parms.persistent_fid = rdata->req->cfile->fid.persistent_fid; + io_parms.volatile_fid = rdata->req->cfile->fid.volatile_fid; io_parms.pid = rdata->pid; rc = smb2_new_read_req( @@ -4630,15 +4647,15 @@ smb2_async_readv(struct cifs_io_subrequest *rdata) flags |= CIFS_HAS_CREDITS; } - cifs_get_readdata(rdata); rc = cifs_call_async(server, &rqst, cifs_readv_receive, smb2_readv_callback, smb3_handle_read_data, rdata, flags, &rdata->credits); if (rc) { - cifs_put_readdata(rdata); cifs_stats_fail_inc(io_parms.tcon, SMB2_READ_HE); - trace_smb3_read_err(0 /* xid */, io_parms.persistent_fid, + trace_smb3_read_err(rdata->rreq->debug_id, + rdata->subreq.debug_index, + rdata->xid, io_parms.persistent_fid, io_parms.tcon->tid, io_parms.tcon->ses->Suid, io_parms.offset, io_parms.length, rc); @@ -4689,22 +4706,23 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, if (rc != -ENODATA) { cifs_stats_fail_inc(io_parms->tcon, SMB2_READ_HE); cifs_dbg(VFS, "Send error in read = %d\n", rc); - trace_smb3_read_err(xid, + trace_smb3_read_err(0, 0, xid, req->PersistentFileId, io_parms->tcon->tid, ses->Suid, io_parms->offset, io_parms->length, rc); } else - trace_smb3_read_done(xid, req->PersistentFileId, io_parms->tcon->tid, + trace_smb3_read_done(0, 0, xid, + req->PersistentFileId, io_parms->tcon->tid, ses->Suid, io_parms->offset, 0); free_rsp_buf(resp_buftype, rsp_iov.iov_base); cifs_small_buf_release(req); return rc == -ENODATA ? 0 : rc; } else - trace_smb3_read_done(xid, - req->PersistentFileId, - io_parms->tcon->tid, ses->Suid, - io_parms->offset, io_parms->length); + trace_smb3_read_done(0, 0, xid, + req->PersistentFileId, + io_parms->tcon->tid, ses->Suid, + io_parms->offset, io_parms->length); cifs_small_buf_release(req); @@ -4738,11 +4756,12 @@ static void smb2_writev_callback(struct mid_q_entry *mid) { struct cifs_io_subrequest *wdata = mid->callback_data; - struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(wdata->req->cfile->tlink); struct TCP_Server_Info *server = wdata->server; - unsigned int written; struct smb2_write_rsp *rsp = (struct smb2_write_rsp *)mid->resp_buf; struct cifs_credits credits = { .value = 0, .instance = 0 }; + ssize_t result = 0; + size_t written; WARN_ONCE(wdata->server != mid->server, "wdata server %p != mid server %p", @@ -4752,8 +4771,8 @@ smb2_writev_callback(struct mid_q_entry *mid) case MID_RESPONSE_RECEIVED: credits.value = le16_to_cpu(rsp->hdr.CreditRequest); credits.instance = server->reconnect_instance; - wdata->result = smb2_check_receive(mid, server, 0); - if (wdata->result != 0) + result = smb2_check_receive(mid, server, 0); + if (result != 0) break; written = le32_to_cpu(rsp->DataLength); @@ -4770,17 +4789,18 @@ smb2_writev_callback(struct mid_q_entry *mid) wdata->result = -ENOSPC; else wdata->subreq.len = written; + iov_iter_advance(&wdata->subreq.io_iter, written); break; case MID_REQUEST_SUBMITTED: case MID_RETRY_NEEDED: - wdata->result = -EAGAIN; + result = -EAGAIN; break; case MID_RESPONSE_MALFORMED: credits.value = le16_to_cpu(rsp->hdr.CreditRequest); credits.instance = server->reconnect_instance; fallthrough; default: - wdata->result = -EIO; + result = -EIO; break; } #ifdef CONFIG_CIFS_SMB_DIRECT @@ -4796,10 +4816,10 @@ smb2_writev_callback(struct mid_q_entry *mid) wdata->mr = NULL; } #endif - if (wdata->result) { + if (result) { cifs_stats_fail_inc(tcon, SMB2_WRITE_HE); - trace_smb3_write_err(0 /* no xid */, - wdata->cfile->fid.persistent_fid, + trace_smb3_write_err(wdata->xid, + wdata->req->cfile->fid.persistent_fid, tcon->tid, tcon->ses->Suid, wdata->subreq.start, wdata->subreq.len, wdata->result); if (wdata->result == -ENOSPC) @@ -4807,27 +4827,28 @@ smb2_writev_callback(struct mid_q_entry *mid) tcon->tree_name); } else trace_smb3_write_done(0 /* no xid */, - wdata->cfile->fid.persistent_fid, + wdata->req->cfile->fid.persistent_fid, tcon->tid, tcon->ses->Suid, wdata->subreq.start, wdata->subreq.len); - queue_work(cifsiod_wq, &wdata->work); + wdata->credits.value = 0; + cifs_write_subrequest_terminated(wdata, result ?: written, true); release_mid(mid); add_credits(server, &credits, 0); } /* smb2_async_writev - send an async write, and set up mid to handle result */ -int +void smb2_async_writev(struct cifs_io_subrequest *wdata) { int rc = -EACCES, flags = 0; struct smb2_write_req *req = NULL; struct smb2_hdr *shdr; - struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(wdata->req->cfile->tlink); struct TCP_Server_Info *server = wdata->server; struct kvec iov[1]; struct smb_rqst rqst = { }; - unsigned int total_len; + unsigned int total_len, xid = wdata->xid; struct cifs_io_parms _io_parms; struct cifs_io_parms *io_parms = NULL; int credit_request; @@ -4844,8 +4865,8 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) .server = server, .offset = wdata->subreq.start, .length = wdata->subreq.len, - .persistent_fid = wdata->cfile->fid.persistent_fid, - .volatile_fid = wdata->cfile->fid.volatile_fid, + .persistent_fid = wdata->req->cfile->fid.persistent_fid, + .volatile_fid = wdata->req->cfile->fid.volatile_fid, .pid = wdata->pid, }; io_parms = &_io_parms; @@ -4853,7 +4874,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) rc = smb2_plain_req_init(SMB2_WRITE, tcon, server, (void **) &req, &total_len); if (rc) - return rc; + goto out; if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; @@ -4871,7 +4892,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) offsetof(struct smb2_write_req, Buffer)); req->RemainingBytes = 0; - trace_smb3_write_enter(0 /* xid */, + trace_smb3_write_enter(wdata->xid, io_parms->persistent_fid, io_parms->tcon->tid, io_parms->tcon->ses->Suid, @@ -4952,25 +4973,27 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) flags |= CIFS_HAS_CREDITS; } - cifs_get_writedata(wdata); rc = cifs_call_async(server, &rqst, NULL, smb2_writev_callback, NULL, wdata, flags, &wdata->credits); - + /* Can't touch wdata if rc == 0 */ if (rc) { - trace_smb3_write_err(0 /* no xid */, + trace_smb3_write_err(xid, io_parms->persistent_fid, io_parms->tcon->tid, io_parms->tcon->ses->Suid, io_parms->offset, io_parms->length, rc); - cifs_put_writedata(wdata); cifs_stats_fail_inc(tcon, SMB2_WRITE_HE); } async_writev_out: cifs_small_buf_release(req); - return rc; +out: + if (rc) { + add_credits_and_wake_if(wdata->server, &wdata->credits, 0); + cifs_write_subrequest_terminated(wdata, rc, true); + } } /* diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h index c781eaa0cba8..b208232b12a2 100644 --- a/fs/smb/client/smb2proto.h +++ b/fs/smb/client/smb2proto.h @@ -213,7 +213,7 @@ extern int SMB2_get_srv_num(const unsigned int xid, struct cifs_tcon *tcon, extern int smb2_async_readv(struct cifs_io_subrequest *rdata); extern int SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, char **buf, int *buf_type); -extern int smb2_async_writev(struct cifs_io_subrequest *wdata); +extern void smb2_async_writev(struct cifs_io_subrequest *wdata); extern int SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, struct kvec *iov, int n_vec); extern int SMB2_echo(struct TCP_Server_Info *server); diff --git a/fs/smb/client/trace.h b/fs/smb/client/trace.h index 604e52876cd2..af97389e983e 100644 --- a/fs/smb/client/trace.h +++ b/fs/smb/client/trace.h @@ -85,6 +85,62 @@ smb3_tcon_ref_traces; /* For logging errors in read or write */ DECLARE_EVENT_CLASS(smb3_rw_err_class, + TP_PROTO(unsigned int rreq_debug_id, + unsigned int rreq_debug_index, + unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + __u64 offset, + __u32 len, + int rc), + TP_ARGS(rreq_debug_id, rreq_debug_index, + xid, fid, tid, sesid, offset, len, rc), + TP_STRUCT__entry( + __field(unsigned int, rreq_debug_id) + __field(unsigned int, rreq_debug_index) + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u64, offset) + __field(__u32, len) + __field(int, rc) + ), + TP_fast_assign( + __entry->rreq_debug_id = rreq_debug_id; + __entry->rreq_debug_index = rreq_debug_index; + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->offset = offset; + __entry->len = len; + __entry->rc = rc; + ), + TP_printk("\tR=%08x[%x] xid=%u sid=0x%llx tid=0x%x fid=0x%llx offset=0x%llx len=0x%x rc=%d", + __entry->rreq_debug_id, __entry->rreq_debug_index, + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->offset, __entry->len, __entry->rc) +) + +#define DEFINE_SMB3_RW_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_rw_err_class, smb3_##name, \ + TP_PROTO(unsigned int rreq_debug_id, \ + unsigned int rreq_debug_index, \ + unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + __u64 offset, \ + __u32 len, \ + int rc), \ + TP_ARGS(rreq_debug_id, rreq_debug_index, xid, fid, tid, sesid, offset, len, rc)) + +DEFINE_SMB3_RW_ERR_EVENT(read_err); + +/* For logging errors in other file I/O ops */ +DECLARE_EVENT_CLASS(smb3_other_err_class, TP_PROTO(unsigned int xid, __u64 fid, __u32 tid, @@ -116,8 +172,8 @@ DECLARE_EVENT_CLASS(smb3_rw_err_class, __entry->offset, __entry->len, __entry->rc) ) -#define DEFINE_SMB3_RW_ERR_EVENT(name) \ -DEFINE_EVENT(smb3_rw_err_class, smb3_##name, \ +#define DEFINE_SMB3_OTHER_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_other_err_class, smb3_##name, \ TP_PROTO(unsigned int xid, \ __u64 fid, \ __u32 tid, \ @@ -127,15 +183,67 @@ DEFINE_EVENT(smb3_rw_err_class, smb3_##name, \ int rc), \ TP_ARGS(xid, fid, tid, sesid, offset, len, rc)) -DEFINE_SMB3_RW_ERR_EVENT(write_err); -DEFINE_SMB3_RW_ERR_EVENT(read_err); -DEFINE_SMB3_RW_ERR_EVENT(query_dir_err); -DEFINE_SMB3_RW_ERR_EVENT(zero_err); -DEFINE_SMB3_RW_ERR_EVENT(falloc_err); +DEFINE_SMB3_OTHER_ERR_EVENT(write_err); +DEFINE_SMB3_OTHER_ERR_EVENT(query_dir_err); +DEFINE_SMB3_OTHER_ERR_EVENT(zero_err); +DEFINE_SMB3_OTHER_ERR_EVENT(falloc_err); /* For logging successful read or write */ DECLARE_EVENT_CLASS(smb3_rw_done_class, + TP_PROTO(unsigned int rreq_debug_id, + unsigned int rreq_debug_index, + unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + __u64 offset, + __u32 len), + TP_ARGS(rreq_debug_id, rreq_debug_index, + xid, fid, tid, sesid, offset, len), + TP_STRUCT__entry( + __field(unsigned int, rreq_debug_id) + __field(unsigned int, rreq_debug_index) + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u64, offset) + __field(__u32, len) + ), + TP_fast_assign( + __entry->rreq_debug_id = rreq_debug_id; + __entry->rreq_debug_index = rreq_debug_index; + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->offset = offset; + __entry->len = len; + ), + TP_printk("R=%08x[%x] xid=%u sid=0x%llx tid=0x%x fid=0x%llx offset=0x%llx len=0x%x", + __entry->rreq_debug_id, __entry->rreq_debug_index, + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->offset, __entry->len) +) + +#define DEFINE_SMB3_RW_DONE_EVENT(name) \ +DEFINE_EVENT(smb3_rw_done_class, smb3_##name, \ + TP_PROTO(unsigned int rreq_debug_id, \ + unsigned int rreq_debug_index, \ + unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + __u64 offset, \ + __u32 len), \ + TP_ARGS(rreq_debug_id, rreq_debug_index, xid, fid, tid, sesid, offset, len)) + +DEFINE_SMB3_RW_DONE_EVENT(read_enter); +DEFINE_SMB3_RW_DONE_EVENT(read_done); + +/* For logging successful other op */ +DECLARE_EVENT_CLASS(smb3_other_done_class, TP_PROTO(unsigned int xid, __u64 fid, __u32 tid, @@ -164,8 +272,8 @@ DECLARE_EVENT_CLASS(smb3_rw_done_class, __entry->offset, __entry->len) ) -#define DEFINE_SMB3_RW_DONE_EVENT(name) \ -DEFINE_EVENT(smb3_rw_done_class, smb3_##name, \ +#define DEFINE_SMB3_OTHER_DONE_EVENT(name) \ +DEFINE_EVENT(smb3_other_done_class, smb3_##name, \ TP_PROTO(unsigned int xid, \ __u64 fid, \ __u32 tid, \ @@ -174,16 +282,14 @@ DEFINE_EVENT(smb3_rw_done_class, smb3_##name, \ __u32 len), \ TP_ARGS(xid, fid, tid, sesid, offset, len)) -DEFINE_SMB3_RW_DONE_EVENT(write_enter); -DEFINE_SMB3_RW_DONE_EVENT(read_enter); -DEFINE_SMB3_RW_DONE_EVENT(query_dir_enter); -DEFINE_SMB3_RW_DONE_EVENT(zero_enter); -DEFINE_SMB3_RW_DONE_EVENT(falloc_enter); -DEFINE_SMB3_RW_DONE_EVENT(write_done); -DEFINE_SMB3_RW_DONE_EVENT(read_done); -DEFINE_SMB3_RW_DONE_EVENT(query_dir_done); -DEFINE_SMB3_RW_DONE_EVENT(zero_done); -DEFINE_SMB3_RW_DONE_EVENT(falloc_done); +DEFINE_SMB3_OTHER_DONE_EVENT(write_enter); +DEFINE_SMB3_OTHER_DONE_EVENT(query_dir_enter); +DEFINE_SMB3_OTHER_DONE_EVENT(zero_enter); +DEFINE_SMB3_OTHER_DONE_EVENT(falloc_enter); +DEFINE_SMB3_OTHER_DONE_EVENT(write_done); +DEFINE_SMB3_OTHER_DONE_EVENT(query_dir_done); +DEFINE_SMB3_OTHER_DONE_EVENT(zero_done); +DEFINE_SMB3_OTHER_DONE_EVENT(falloc_done); /* For logging successful set EOF (truncate) */ DECLARE_EVENT_CLASS(smb3_eof_class, diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index 4c8af995c60b..012b9bd06995 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -1813,8 +1813,11 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) length = data_len; /* An RDMA read is already done. */ else #endif + { length = cifs_read_iter_from_socket(server, &rdata->subreq.io_iter, data_len); + iov_iter_revert(&rdata->subreq.io_iter, data_len); + } if (length > 0) rdata->got_bytes += length; server->total_read += length; -- cgit v1.2.3-59-g8ed1b From 742b3443e23104fc81edbeccfb993ae350aae981 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 22:27:15 +0100 Subject: cifs: Remove some code that's no longer used, part 1 Remove some code that was #if'd out with the netfslib conversion. This is split into parts for file.c as the diff generator otherwise produces a hard to read diff for part of it where a big chunk is cut out. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/cifsglob.h | 12 - fs/smb/client/cifsproto.h | 25 -- fs/smb/client/file.c | 619 ---------------------------------------------- fs/smb/client/fscache.c | 111 --------- fs/smb/client/fscache.h | 58 ----- 5 files changed, 825 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 983860bf5fbb..65574e69ba4f 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1515,18 +1515,6 @@ struct cifs_io_subrequest { struct smbd_mr *mr; #endif struct cifs_credits credits; - -#if 0 // TODO: Remove following elements - struct list_head list; - struct completion done; - struct work_struct work; - struct cifsFileInfo *cfile; - struct address_space *mapping; - struct cifs_aio_ctx *ctx; - enum writeback_sync_modes sync_mode; - bool uncached; - struct bio_vec *bv; -#endif }; /* diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index d46ad86150cd..c15bb5ee7eb7 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -601,36 +601,11 @@ void __cifs_put_smb_ses(struct cifs_ses *ses); extern struct cifs_ses * cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx); -#if 0 // TODO Remove -void cifs_readdata_release(struct cifs_io_subrequest *rdata); -static inline void cifs_get_readdata(struct cifs_io_subrequest *rdata) -{ - refcount_inc(&rdata->subreq.ref); -} -static inline void cifs_put_readdata(struct cifs_io_subrequest *rdata) -{ - if (refcount_dec_and_test(&rdata->subreq.ref)) - cifs_readdata_release(rdata); -} -#endif int cifs_async_readv(struct cifs_io_subrequest *rdata); int cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid); void cifs_async_writev(struct cifs_io_subrequest *wdata); void cifs_writev_complete(struct work_struct *work); -#if 0 // TODO Remove -struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete); -void cifs_writedata_release(struct cifs_io_subrequest *rdata); -static inline void cifs_get_writedata(struct cifs_io_subrequest *wdata) -{ - refcount_inc(&wdata->subreq.ref); -} -static inline void cifs_put_writedata(struct cifs_io_subrequest *wdata) -{ - if (refcount_dec_and_test(&wdata->subreq.ref)) - cifs_writedata_release(wdata); -} -#endif int cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, const unsigned char *path, char *pbuf, diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index bb3adde9343e..6088d0f80522 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -352,133 +352,6 @@ const struct netfs_request_ops cifs_req_ops = { .issue_write = cifs_issue_write, }; -#if 0 // TODO remove 397 -/* - * Remove the dirty flags from a span of pages. - */ -static void cifs_undirty_folios(struct inode *inode, loff_t start, unsigned int len) -{ - struct address_space *mapping = inode->i_mapping; - struct folio *folio; - pgoff_t end; - - XA_STATE(xas, &mapping->i_pages, start / PAGE_SIZE); - - rcu_read_lock(); - - end = (start + len - 1) / PAGE_SIZE; - xas_for_each_marked(&xas, folio, end, PAGECACHE_TAG_DIRTY) { - if (xas_retry(&xas, folio)) - continue; - xas_pause(&xas); - rcu_read_unlock(); - folio_lock(folio); - folio_clear_dirty_for_io(folio); - folio_unlock(folio); - rcu_read_lock(); - } - - rcu_read_unlock(); -} - -/* - * Completion of write to server. - */ -void cifs_pages_written_back(struct inode *inode, loff_t start, unsigned int len) -{ - struct address_space *mapping = inode->i_mapping; - struct folio *folio; - pgoff_t end; - - XA_STATE(xas, &mapping->i_pages, start / PAGE_SIZE); - - if (!len) - return; - - rcu_read_lock(); - - end = (start + len - 1) / PAGE_SIZE; - xas_for_each(&xas, folio, end) { - if (xas_retry(&xas, folio)) - continue; - if (!folio_test_writeback(folio)) { - WARN_ONCE(1, "bad %x @%llx page %lx %lx\n", - len, start, folio->index, end); - continue; - } - - folio_detach_private(folio); - folio_end_writeback(folio); - } - - rcu_read_unlock(); -} - -/* - * Failure of write to server. - */ -void cifs_pages_write_failed(struct inode *inode, loff_t start, unsigned int len) -{ - struct address_space *mapping = inode->i_mapping; - struct folio *folio; - pgoff_t end; - - XA_STATE(xas, &mapping->i_pages, start / PAGE_SIZE); - - if (!len) - return; - - rcu_read_lock(); - - end = (start + len - 1) / PAGE_SIZE; - xas_for_each(&xas, folio, end) { - if (xas_retry(&xas, folio)) - continue; - if (!folio_test_writeback(folio)) { - WARN_ONCE(1, "bad %x @%llx page %lx %lx\n", - len, start, folio->index, end); - continue; - } - - folio_set_error(folio); - folio_end_writeback(folio); - } - - rcu_read_unlock(); -} - -/* - * Redirty pages after a temporary failure. - */ -void cifs_pages_write_redirty(struct inode *inode, loff_t start, unsigned int len) -{ - struct address_space *mapping = inode->i_mapping; - struct folio *folio; - pgoff_t end; - - XA_STATE(xas, &mapping->i_pages, start / PAGE_SIZE); - - if (!len) - return; - - rcu_read_lock(); - - end = (start + len - 1) / PAGE_SIZE; - xas_for_each(&xas, folio, end) { - if (!folio_test_writeback(folio)) { - WARN_ONCE(1, "bad %x @%llx page %lx %lx\n", - len, start, folio->index, end); - continue; - } - - filemap_dirty_folio(folio->mapping, folio); - folio_end_writeback(folio); - } - - rcu_read_unlock(); -} -#endif // end netfslib remove 397 - /* * Mark as invalid, all open files on tree connections since they * were closed when session to server was lost. @@ -2540,92 +2413,6 @@ void cifs_write_subrequest_terminated(struct cifs_io_subrequest *wdata, ssize_t netfs_write_subrequest_terminated(&wdata->subreq, result, was_async); } -#if 0 // TODO remove 2483 -static ssize_t -cifs_write(struct cifsFileInfo *open_file, __u32 pid, const char *write_data, - size_t write_size, loff_t *offset) -{ - int rc = 0; - unsigned int bytes_written = 0; - unsigned int total_written; - struct cifs_tcon *tcon; - struct TCP_Server_Info *server; - unsigned int xid; - struct dentry *dentry = open_file->dentry; - struct cifsInodeInfo *cifsi = CIFS_I(d_inode(dentry)); - struct cifs_io_parms io_parms = {0}; - - cifs_dbg(FYI, "write %zd bytes to offset %lld of %pd\n", - write_size, *offset, dentry); - - tcon = tlink_tcon(open_file->tlink); - server = tcon->ses->server; - - if (!server->ops->sync_write) - return -ENOSYS; - - xid = get_xid(); - - for (total_written = 0; write_size > total_written; - total_written += bytes_written) { - rc = -EAGAIN; - while (rc == -EAGAIN) { - struct kvec iov[2]; - unsigned int len; - - if (open_file->invalidHandle) { - /* we could deadlock if we called - filemap_fdatawait from here so tell - reopen_file not to flush data to - server now */ - rc = cifs_reopen_file(open_file, false); - if (rc != 0) - break; - } - - len = min(server->ops->wp_retry_size(d_inode(dentry)), - (unsigned int)write_size - total_written); - /* iov[0] is reserved for smb header */ - iov[1].iov_base = (char *)write_data + total_written; - iov[1].iov_len = len; - io_parms.pid = pid; - io_parms.tcon = tcon; - io_parms.offset = *offset; - io_parms.length = len; - rc = server->ops->sync_write(xid, &open_file->fid, - &io_parms, &bytes_written, iov, 1); - } - if (rc || (bytes_written == 0)) { - if (total_written) - break; - else { - free_xid(xid); - return rc; - } - } else { - spin_lock(&d_inode(dentry)->i_lock); - cifs_update_eof(cifsi, *offset, bytes_written); - spin_unlock(&d_inode(dentry)->i_lock); - *offset += bytes_written; - } - } - - cifs_stats_bytes_written(tcon, total_written); - - if (total_written > 0) { - spin_lock(&d_inode(dentry)->i_lock); - if (*offset > d_inode(dentry)->i_size) { - i_size_write(d_inode(dentry), *offset); - d_inode(dentry)->i_blocks = (512 - 1 + *offset) >> 9; - } - spin_unlock(&d_inode(dentry)->i_lock); - } - mark_inode_dirty_sync(d_inode(dentry)); - free_xid(xid); - return total_written; -} -#endif // end netfslib remove 2483 - struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) { @@ -4826,293 +4613,6 @@ int cifs_file_mmap(struct file *file, struct vm_area_struct *vma) return rc; } -#if 0 // TODO remove 4794 -/* - * Unlock a bunch of folios in the pagecache. - */ -static void cifs_unlock_folios(struct address_space *mapping, pgoff_t first, pgoff_t last) -{ - struct folio *folio; - XA_STATE(xas, &mapping->i_pages, first); - - rcu_read_lock(); - xas_for_each(&xas, folio, last) { - folio_unlock(folio); - } - rcu_read_unlock(); -} - -static void cifs_readahead_complete(struct work_struct *work) -{ - struct cifs_io_subrequest *rdata = container_of(work, - struct cifs_io_subrequest, work); - struct folio *folio; - pgoff_t last; - bool good = rdata->result == 0 || (rdata->result == -EAGAIN && rdata->got_bytes); - - XA_STATE(xas, &rdata->mapping->i_pages, rdata->subreq.start / PAGE_SIZE); - - if (good) - cifs_readahead_to_fscache(rdata->mapping->host, - rdata->subreq.start, rdata->subreq.len); - - if (iov_iter_count(&rdata->subreq.io_iter) > 0) - iov_iter_zero(iov_iter_count(&rdata->subreq.io_iter), &rdata->subreq.io_iter); - - last = (rdata->subreq.start + rdata->subreq.len - 1) / PAGE_SIZE; - - rcu_read_lock(); - xas_for_each(&xas, folio, last) { - if (good) { - flush_dcache_folio(folio); - folio_mark_uptodate(folio); - } - folio_unlock(folio); - } - rcu_read_unlock(); - - cifs_put_readdata(rdata); -} - -static void cifs_readahead(struct readahead_control *ractl) -{ - struct cifsFileInfo *open_file = ractl->file->private_data; - struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(ractl->file); - struct TCP_Server_Info *server; - unsigned int xid, nr_pages, cache_nr_pages = 0; - unsigned int ra_pages; - pgoff_t next_cached = ULONG_MAX, ra_index; - bool caching = fscache_cookie_enabled(cifs_inode_cookie(ractl->mapping->host)) && - cifs_inode_cookie(ractl->mapping->host)->cache_priv; - bool check_cache = caching; - pid_t pid; - int rc = 0; - - /* Note that readahead_count() lags behind our dequeuing of pages from - * the ractl, wo we have to keep track for ourselves. - */ - ra_pages = readahead_count(ractl); - ra_index = readahead_index(ractl); - - xid = get_xid(); - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) - pid = open_file->pid; - else - pid = current->tgid; - - server = cifs_pick_channel(tlink_tcon(open_file->tlink)->ses); - - cifs_dbg(FYI, "%s: file=%p mapping=%p num_pages=%u\n", - __func__, ractl->file, ractl->mapping, ra_pages); - - /* - * Chop the readahead request up into rsize-sized read requests. - */ - while ((nr_pages = ra_pages)) { - unsigned int i; - struct cifs_io_subrequest *rdata; - struct cifs_credits credits_on_stack; - struct cifs_credits *credits = &credits_on_stack; - struct folio *folio; - pgoff_t fsize; - size_t rsize; - - /* - * Find out if we have anything cached in the range of - * interest, and if so, where the next chunk of cached data is. - */ - if (caching) { - if (check_cache) { - rc = cifs_fscache_query_occupancy( - ractl->mapping->host, ra_index, nr_pages, - &next_cached, &cache_nr_pages); - if (rc < 0) - caching = false; - check_cache = false; - } - - if (ra_index == next_cached) { - /* - * TODO: Send a whole batch of pages to be read - * by the cache. - */ - folio = readahead_folio(ractl); - fsize = folio_nr_pages(folio); - ra_pages -= fsize; - ra_index += fsize; - if (cifs_readpage_from_fscache(ractl->mapping->host, - &folio->page) < 0) { - /* - * TODO: Deal with cache read failure - * here, but for the moment, delegate - * that to readpage. - */ - caching = false; - } - folio_unlock(folio); - next_cached += fsize; - cache_nr_pages -= fsize; - if (cache_nr_pages == 0) - check_cache = true; - continue; - } - } - - if (open_file->invalidHandle) { - rc = cifs_reopen_file(open_file, true); - if (rc) { - if (rc == -EAGAIN) - continue; - break; - } - } - - if (cifs_sb->ctx->rsize == 0) - cifs_sb->ctx->rsize = - server->ops->negotiate_rsize(tlink_tcon(open_file->tlink), - cifs_sb->ctx); - - rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->rsize, - &rsize, credits); - if (rc) - break; - nr_pages = min_t(size_t, rsize / PAGE_SIZE, ra_pages); - if (next_cached != ULONG_MAX) - nr_pages = min_t(size_t, nr_pages, next_cached - ra_index); - - /* - * Give up immediately if rsize is too small to read an entire - * page. The VFS will fall back to readpage. We should never - * reach this point however since we set ra_pages to 0 when the - * rsize is smaller than a cache page. - */ - if (unlikely(!nr_pages)) { - add_credits_and_wake_if(server, credits, 0); - break; - } - - rdata = cifs_readdata_alloc(cifs_readahead_complete); - if (!rdata) { - /* best to give up if we're out of mem */ - add_credits_and_wake_if(server, credits, 0); - break; - } - - rdata->subreq.start = ra_index * PAGE_SIZE; - rdata->subreq.len = nr_pages * PAGE_SIZE; - rdata->cfile = cifsFileInfo_get(open_file); - rdata->server = server; - rdata->mapping = ractl->mapping; - rdata->pid = pid; - rdata->credits = credits_on_stack; - - for (i = 0; i < nr_pages; i++) { - if (!readahead_folio(ractl)) - WARN_ON(1); - } - ra_pages -= nr_pages; - ra_index += nr_pages; - - iov_iter_xarray(&rdata->subreq.io_iter, ITER_DEST, &rdata->mapping->i_pages, - rdata->subreq.start, rdata->subreq.len); - - rc = adjust_credits(server, &rdata->credits, rdata->subreq.len); - if (!rc) { - if (rdata->cfile->invalidHandle) - rc = -EAGAIN; - else - rc = server->ops->async_readv(rdata); - } - - if (rc) { - add_credits_and_wake_if(server, &rdata->credits, 0); - cifs_unlock_folios(rdata->mapping, - rdata->subreq.start / PAGE_SIZE, - (rdata->subreq.start + rdata->subreq.len - 1) / PAGE_SIZE); - /* Fallback to the readpage in error/reconnect cases */ - cifs_put_readdata(rdata); - break; - } - - cifs_put_readdata(rdata); - } - - free_xid(xid); -} - -/* - * cifs_readpage_worker must be called with the page pinned - */ -static int cifs_readpage_worker(struct file *file, struct page *page, - loff_t *poffset) -{ - struct inode *inode = file_inode(file); - struct timespec64 atime, mtime; - char *read_data; - int rc; - - /* Is the page cached? */ - rc = cifs_readpage_from_fscache(inode, page); - if (rc == 0) - goto read_complete; - - read_data = kmap(page); - /* for reads over a certain size could initiate async read ahead */ - - rc = cifs_read(file, read_data, PAGE_SIZE, poffset); - - if (rc < 0) - goto io_error; - else - cifs_dbg(FYI, "Bytes read %d\n", rc); - - /* we do not want atime to be less than mtime, it broke some apps */ - atime = inode_set_atime_to_ts(inode, current_time(inode)); - mtime = inode_get_mtime(inode); - if (timespec64_compare(&atime, &mtime) < 0) - inode_set_atime_to_ts(inode, inode_get_mtime(inode)); - - if (PAGE_SIZE > rc) - memset(read_data + rc, 0, PAGE_SIZE - rc); - - flush_dcache_page(page); - SetPageUptodate(page); - rc = 0; - -io_error: - kunmap(page); - -read_complete: - unlock_page(page); - return rc; -} - -static int cifs_read_folio(struct file *file, struct folio *folio) -{ - struct page *page = &folio->page; - loff_t offset = page_file_offset(page); - int rc = -EACCES; - unsigned int xid; - - xid = get_xid(); - - if (file->private_data == NULL) { - rc = -EBADF; - free_xid(xid); - return rc; - } - - cifs_dbg(FYI, "read_folio %p at offset %d 0x%x\n", - page, (int)offset, (int)offset); - - rc = cifs_readpage_worker(file, page, &offset); - - free_xid(xid); - return rc; -} -#endif // end netfslib remove 4794 - static int is_inode_writable(struct cifsInodeInfo *cifs_inode) { struct cifsFileInfo *open_file; @@ -5160,104 +4660,6 @@ bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file, return true; } -#if 0 // TODO remove 5152 -static int cifs_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, - struct page **pagep, void **fsdata) -{ - int oncethru = 0; - pgoff_t index = pos >> PAGE_SHIFT; - loff_t offset = pos & (PAGE_SIZE - 1); - loff_t page_start = pos & PAGE_MASK; - loff_t i_size; - struct page *page; - int rc = 0; - - cifs_dbg(FYI, "write_begin from %lld len %d\n", (long long)pos, len); - -start: - page = grab_cache_page_write_begin(mapping, index); - if (!page) { - rc = -ENOMEM; - goto out; - } - - if (PageUptodate(page)) - goto out; - - /* - * If we write a full page it will be up to date, no need to read from - * the server. If the write is short, we'll end up doing a sync write - * instead. - */ - if (len == PAGE_SIZE) - goto out; - - /* - * optimize away the read when we have an oplock, and we're not - * expecting to use any of the data we'd be reading in. That - * is, when the page lies beyond the EOF, or straddles the EOF - * and the write will cover all of the existing data. - */ - if (CIFS_CACHE_READ(CIFS_I(mapping->host))) { - i_size = i_size_read(mapping->host); - if (page_start >= i_size || - (offset == 0 && (pos + len) >= i_size)) { - zero_user_segments(page, 0, offset, - offset + len, - PAGE_SIZE); - /* - * PageChecked means that the parts of the page - * to which we're not writing are considered up - * to date. Once the data is copied to the - * page, it can be set uptodate. - */ - SetPageChecked(page); - goto out; - } - } - - if ((file->f_flags & O_ACCMODE) != O_WRONLY && !oncethru) { - /* - * might as well read a page, it is fast enough. If we get - * an error, we don't need to return it. cifs_write_end will - * do a sync write instead since PG_uptodate isn't set. - */ - cifs_readpage_worker(file, page, &page_start); - put_page(page); - oncethru = 1; - goto start; - } else { - /* we could try using another file handle if there is one - - but how would we lock it to prevent close of that handle - racing with this read? In any case - this will be written out by write_end so is fine */ - } -out: - *pagep = page; - return rc; -} - -static bool cifs_release_folio(struct folio *folio, gfp_t gfp) -{ - if (folio_test_private(folio)) - return 0; - if (folio_test_private_2(folio)) { /* [DEPRECATED] */ - if (current_is_kswapd() || !(gfp & __GFP_FS)) - return false; - folio_wait_private_2(folio); - } - fscache_note_page_release(cifs_inode_cookie(folio->mapping->host)); - return true; -} - -static void cifs_invalidate_folio(struct folio *folio, size_t offset, - size_t length) -{ - folio_wait_private_2(folio); /* [DEPRECATED] */ -} -#endif // end netfslib remove 5152 - void cifs_oplock_break(struct work_struct *work) { struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo, @@ -5347,27 +4749,6 @@ out: cifs_done_oplock_break(cinode); } -#if 0 // TODO remove 5333 -/* - * The presence of cifs_direct_io() in the address space ops vector - * allowes open() O_DIRECT flags which would have failed otherwise. - * - * In the non-cached mode (mount with cache=none), we shunt off direct read and write requests - * so this method should never be called. - * - * Direct IO is not yet supported in the cached mode. - */ -static ssize_t -cifs_direct_io(struct kiocb *iocb, struct iov_iter *iter) -{ - /* - * FIXME - * Eventually need to support direct IO for non forcedirectio mounts - */ - return -EINVAL; -} -#endif // netfs end remove 5333 - static int cifs_swap_activate(struct swap_info_struct *sis, struct file *swap_file, sector_t *span) { diff --git a/fs/smb/client/fscache.c b/fs/smb/client/fscache.c index b36c493f4c56..01424a5cdb99 100644 --- a/fs/smb/client/fscache.c +++ b/fs/smb/client/fscache.c @@ -170,114 +170,3 @@ void cifs_fscache_release_inode_cookie(struct inode *inode) cifsi->netfs.cache = NULL; } } - -#if 0 // TODO remove -/* - * Fallback page reading interface. - */ -static int fscache_fallback_read_page(struct inode *inode, struct page *page) -{ - struct netfs_cache_resources cres; - struct fscache_cookie *cookie = cifs_inode_cookie(inode); - struct iov_iter iter; - struct bio_vec bvec; - int ret; - - memset(&cres, 0, sizeof(cres)); - bvec_set_page(&bvec, page, PAGE_SIZE, 0); - iov_iter_bvec(&iter, ITER_DEST, &bvec, 1, PAGE_SIZE); - - ret = fscache_begin_read_operation(&cres, cookie); - if (ret < 0) - return ret; - - ret = fscache_read(&cres, page_offset(page), &iter, NETFS_READ_HOLE_FAIL, - NULL, NULL); - fscache_end_operation(&cres); - return ret; -} - -/* - * Fallback page writing interface. - */ -static int fscache_fallback_write_pages(struct inode *inode, loff_t start, size_t len, - bool no_space_allocated_yet) -{ - struct netfs_cache_resources cres; - struct fscache_cookie *cookie = cifs_inode_cookie(inode); - struct iov_iter iter; - int ret; - - memset(&cres, 0, sizeof(cres)); - iov_iter_xarray(&iter, ITER_SOURCE, &inode->i_mapping->i_pages, start, len); - - ret = fscache_begin_write_operation(&cres, cookie); - if (ret < 0) - return ret; - - ret = cres.ops->prepare_write(&cres, &start, &len, len, i_size_read(inode), - no_space_allocated_yet); - if (ret == 0) - ret = fscache_write(&cres, start, &iter, NULL, NULL); - fscache_end_operation(&cres); - return ret; -} - -/* - * Retrieve a page from FS-Cache - */ -int __cifs_readpage_from_fscache(struct inode *inode, struct page *page) -{ - int ret; - - cifs_dbg(FYI, "%s: (fsc:%p, p:%p, i:0x%p\n", - __func__, cifs_inode_cookie(inode), page, inode); - - ret = fscache_fallback_read_page(inode, page); - if (ret < 0) - return ret; - - /* Read completed synchronously */ - SetPageUptodate(page); - return 0; -} - -void __cifs_readahead_to_fscache(struct inode *inode, loff_t pos, size_t len) -{ - cifs_dbg(FYI, "%s: (fsc: %p, p: %llx, l: %zx, i: %p)\n", - __func__, cifs_inode_cookie(inode), pos, len, inode); - - fscache_fallback_write_pages(inode, pos, len, true); -} - -/* - * Query the cache occupancy. - */ -int __cifs_fscache_query_occupancy(struct inode *inode, - pgoff_t first, unsigned int nr_pages, - pgoff_t *_data_first, - unsigned int *_data_nr_pages) -{ - struct netfs_cache_resources cres; - struct fscache_cookie *cookie = cifs_inode_cookie(inode); - loff_t start, data_start; - size_t len, data_len; - int ret; - - ret = fscache_begin_read_operation(&cres, cookie); - if (ret < 0) - return ret; - - start = first * PAGE_SIZE; - len = nr_pages * PAGE_SIZE; - ret = cres.ops->query_occupancy(&cres, start, len, PAGE_SIZE, - &data_start, &data_len); - if (ret == 0) { - *_data_first = data_start / PAGE_SIZE; - *_data_nr_pages = len / PAGE_SIZE; - } - - fscache_end_operation(&cres); - return ret; -} -#endif diff --git a/fs/smb/client/fscache.h b/fs/smb/client/fscache.h index 08b30f79d4cd..f06cb24f5f3c 100644 --- a/fs/smb/client/fscache.h +++ b/fs/smb/client/fscache.h @@ -74,43 +74,6 @@ static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags i_size_read(inode), flags); } -#if 0 // TODO remove -extern int __cifs_fscache_query_occupancy(struct inode *inode, - pgoff_t first, unsigned int nr_pages, - pgoff_t *_data_first, - unsigned int *_data_nr_pages); - -static inline int cifs_fscache_query_occupancy(struct inode *inode, - pgoff_t first, unsigned int nr_pages, - pgoff_t *_data_first, - unsigned int *_data_nr_pages) -{ - if (!cifs_inode_cookie(inode)) - return -ENOBUFS; - return __cifs_fscache_query_occupancy(inode, first, nr_pages, - _data_first, _data_nr_pages); -} - -extern int __cifs_readpage_from_fscache(struct inode *pinode, struct page *ppage); -extern void __cifs_readahead_to_fscache(struct inode *pinode, loff_t pos, size_t len); - - -static inline int cifs_readpage_from_fscache(struct inode *inode, - struct page *page) -{ - if (cifs_inode_cookie(inode)) - return __cifs_readpage_from_fscache(inode, page); - return -ENOBUFS; -} - -static inline void cifs_readahead_to_fscache(struct inode *inode, - loff_t pos, size_t len) -{ - if (cifs_inode_cookie(inode)) - __cifs_readahead_to_fscache(inode, pos, len); -} -#endif - static inline bool cifs_fscache_enabled(struct inode *inode) { return fscache_cookie_enabled(cifs_inode_cookie(inode)); @@ -133,27 +96,6 @@ static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { re static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) {} static inline bool cifs_fscache_enabled(struct inode *inode) { return false; } -#if 0 // TODO remove -static inline int cifs_fscache_query_occupancy(struct inode *inode, - pgoff_t first, unsigned int nr_pages, - pgoff_t *_data_first, - unsigned int *_data_nr_pages) -{ - *_data_first = ULONG_MAX; - *_data_nr_pages = 0; - return -ENOBUFS; -} - -static inline int -cifs_readpage_from_fscache(struct inode *inode, struct page *page) -{ - return -ENOBUFS; -} - -static inline -void cifs_readahead_to_fscache(struct inode *inode, loff_t pos, size_t len) {} -#endif - #endif /* CONFIG_CIFS_FSCACHE */ #endif /* _CIFS_FSCACHE_H */ -- cgit v1.2.3-59-g8ed1b From 2f99c0bce6d30851313c150369cabfd74e594a45 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 18:29:59 +0100 Subject: cifs: Remove some code that's no longer used, part 2 Remove some code that was #if'd out with the netfslib conversion. This is split into parts for file.c as the diff generator otherwise produces a hard to read diff for part of it where a big chunk is cut out. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/file.c | 634 +-------------------------------------------------- 1 file changed, 1 insertion(+), 633 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 6088d0f80522..f504cda886f0 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -2617,639 +2617,6 @@ cifs_get_readable_path(struct cifs_tcon *tcon, const char *name, return -ENOENT; } -#if 0 // TODO remove 2773 -void -cifs_writedata_release(struct cifs_io_subrequest *wdata) -{ - if (wdata->uncached) - kref_put(&wdata->ctx->refcount, cifs_aio_ctx_release); -#ifdef CONFIG_CIFS_SMB_DIRECT - if (wdata->mr) { - smbd_deregister_mr(wdata->mr); - wdata->mr = NULL; - } -#endif - - if (wdata->cfile) - cifsFileInfo_put(wdata->cfile); - - kfree(wdata); -} - -/* - * Write failed with a retryable error. Resend the write request. It's also - * possible that the page was redirtied so re-clean the page. - */ -static void -cifs_writev_requeue(struct cifs_io_subrequest *wdata) -{ - int rc = 0; - struct inode *inode = d_inode(wdata->cfile->dentry); - struct TCP_Server_Info *server; - unsigned int rest_len = wdata->subreq.len; - loff_t fpos = wdata->subreq.start; - - server = tlink_tcon(wdata->cfile->tlink)->ses->server; - do { - struct cifs_io_subrequest *wdata2; - unsigned int wsize, cur_len; - - wsize = server->ops->wp_retry_size(inode); - if (wsize < rest_len) { - if (wsize < PAGE_SIZE) { - rc = -EOPNOTSUPP; - break; - } - cur_len = min(round_down(wsize, PAGE_SIZE), rest_len); - } else { - cur_len = rest_len; - } - - wdata2 = cifs_writedata_alloc(cifs_writev_complete); - if (!wdata2) { - rc = -ENOMEM; - break; - } - - wdata2->sync_mode = wdata->sync_mode; - wdata2->subreq.start = fpos; - wdata2->subreq.len = cur_len; - wdata2->subreq.io_iter = wdata->subreq.io_iter; - - iov_iter_advance(&wdata2->subreq.io_iter, fpos - wdata->subreq.start); - iov_iter_truncate(&wdata2->subreq.io_iter, wdata2->subreq.len); - - if (iov_iter_is_xarray(&wdata2->subreq.io_iter)) - /* Check for pages having been redirtied and clean - * them. We can do this by walking the xarray. If - * it's not an xarray, then it's a DIO and we shouldn't - * be mucking around with the page bits. - */ - cifs_undirty_folios(inode, fpos, cur_len); - - rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY, - &wdata2->cfile); - if (!wdata2->cfile) { - cifs_dbg(VFS, "No writable handle to retry writepages rc=%d\n", - rc); - if (!is_retryable_error(rc)) - rc = -EBADF; - } else { - wdata2->pid = wdata2->cfile->pid; - rc = server->ops->async_writev(wdata2); - } - - cifs_put_writedata(wdata2); - if (rc) { - if (is_retryable_error(rc)) - continue; - fpos += cur_len; - rest_len -= cur_len; - break; - } - - fpos += cur_len; - rest_len -= cur_len; - } while (rest_len > 0); - - /* Clean up remaining pages from the original wdata */ - if (iov_iter_is_xarray(&wdata->subreq.io_iter)) - cifs_pages_write_failed(inode, fpos, rest_len); - - if (rc != 0 && !is_retryable_error(rc)) - mapping_set_error(inode->i_mapping, rc); - cifs_put_writedata(wdata); -} - -void -cifs_writev_complete(struct work_struct *work) -{ - struct cifs_io_subrequest *wdata = container_of(work, - struct cifs_io_subrequest, work); - struct inode *inode = d_inode(wdata->cfile->dentry); - - if (wdata->result == 0) { - spin_lock(&inode->i_lock); - cifs_update_eof(CIFS_I(inode), wdata->subreq.start, wdata->subreq.len); - spin_unlock(&inode->i_lock); - cifs_stats_bytes_written(tlink_tcon(wdata->cfile->tlink), - wdata->subreq.len); - } else if (wdata->sync_mode == WB_SYNC_ALL && wdata->result == -EAGAIN) - return cifs_writev_requeue(wdata); - - if (wdata->result == -EAGAIN) - cifs_pages_write_redirty(inode, wdata->subreq.start, wdata->subreq.len); - else if (wdata->result < 0) - cifs_pages_write_failed(inode, wdata->subreq.start, wdata->subreq.len); - else - cifs_pages_written_back(inode, wdata->subreq.start, wdata->subreq.len); - - if (wdata->result != -EAGAIN) - mapping_set_error(inode->i_mapping, wdata->result); - cifs_put_writedata(wdata); -} - -struct cifs_io_subrequest *cifs_writedata_alloc(work_func_t complete) -{ - struct cifs_io_subrequest *wdata; - - wdata = kzalloc(sizeof(*wdata), GFP_NOFS); - if (wdata != NULL) { - refcount_set(&wdata->subreq.ref, 1); - INIT_LIST_HEAD(&wdata->list); - init_completion(&wdata->done); - INIT_WORK(&wdata->work, complete); - } - return wdata; -} - -/* - * Extend the region to be written back to include subsequent contiguously - * dirty pages if possible, but don't sleep while doing so. - */ -static void cifs_extend_writeback(struct address_space *mapping, - struct xa_state *xas, - long *_count, - loff_t start, - int max_pages, - loff_t max_len, - size_t *_len) -{ - struct folio_batch batch; - struct folio *folio; - unsigned int nr_pages; - pgoff_t index = (start + *_len) / PAGE_SIZE; - size_t len; - bool stop = true; - unsigned int i; - - folio_batch_init(&batch); - - do { - /* Firstly, we gather up a batch of contiguous dirty pages - * under the RCU read lock - but we can't clear the dirty flags - * there if any of those pages are mapped. - */ - rcu_read_lock(); - - xas_for_each(xas, folio, ULONG_MAX) { - stop = true; - if (xas_retry(xas, folio)) - continue; - if (xa_is_value(folio)) - break; - if (folio->index != index) { - xas_reset(xas); - break; - } - - if (!folio_try_get_rcu(folio)) { - xas_reset(xas); - continue; - } - nr_pages = folio_nr_pages(folio); - if (nr_pages > max_pages) { - xas_reset(xas); - break; - } - - /* Has the page moved or been split? */ - if (unlikely(folio != xas_reload(xas))) { - folio_put(folio); - xas_reset(xas); - break; - } - - if (!folio_trylock(folio)) { - folio_put(folio); - xas_reset(xas); - break; - } - if (!folio_test_dirty(folio) || - folio_test_writeback(folio)) { - folio_unlock(folio); - folio_put(folio); - xas_reset(xas); - break; - } - - max_pages -= nr_pages; - len = folio_size(folio); - stop = false; - - index += nr_pages; - *_count -= nr_pages; - *_len += len; - if (max_pages <= 0 || *_len >= max_len || *_count <= 0) - stop = true; - - if (!folio_batch_add(&batch, folio)) - break; - if (stop) - break; - } - - xas_pause(xas); - rcu_read_unlock(); - - /* Now, if we obtained any pages, we can shift them to being - * writable and mark them for caching. - */ - if (!folio_batch_count(&batch)) - break; - - for (i = 0; i < folio_batch_count(&batch); i++) { - folio = batch.folios[i]; - /* The folio should be locked, dirty and not undergoing - * writeback from the loop above. - */ - if (!folio_clear_dirty_for_io(folio)) - WARN_ON(1); - folio_start_writeback(folio); - folio_unlock(folio); - } - - folio_batch_release(&batch); - cond_resched(); - } while (!stop); -} - -/* - * Write back the locked page and any subsequent non-locked dirty pages. - */ -static ssize_t cifs_write_back_from_locked_folio(struct address_space *mapping, - struct writeback_control *wbc, - struct xa_state *xas, - struct folio *folio, - unsigned long long start, - unsigned long long end) -{ - struct inode *inode = mapping->host; - struct TCP_Server_Info *server; - struct cifs_io_subrequest *wdata; - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); - struct cifs_credits credits_on_stack; - struct cifs_credits *credits = &credits_on_stack; - struct cifsFileInfo *cfile = NULL; - unsigned long long i_size = i_size_read(inode), max_len; - unsigned int xid; - size_t wsize; - size_t len = folio_size(folio); - long count = wbc->nr_to_write; - int rc; - - /* The folio should be locked, dirty and not undergoing writeback. */ - if (!folio_clear_dirty_for_io(folio)) - WARN_ON_ONCE(1); - folio_start_writeback(folio); - - count -= folio_nr_pages(folio); - - xid = get_xid(); - server = cifs_pick_channel(cifs_sb_master_tcon(cifs_sb)->ses); - - rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY, &cfile); - if (rc) { - cifs_dbg(VFS, "No writable handle in writepages rc=%d\n", rc); - goto err_xid; - } - - rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->wsize, - &wsize, credits); - if (rc != 0) - goto err_close; - - wdata = cifs_writedata_alloc(cifs_writev_complete); - if (!wdata) { - rc = -ENOMEM; - goto err_uncredit; - } - - wdata->sync_mode = wbc->sync_mode; - wdata->subreq.start = folio_pos(folio); - wdata->pid = cfile->pid; - wdata->credits = credits_on_stack; - wdata->cfile = cfile; - wdata->server = server; - cfile = NULL; - - /* Find all consecutive lockable dirty pages that have contiguous - * written regions, stopping when we find a page that is not - * immediately lockable, is not dirty or is missing, or we reach the - * end of the range. - */ - if (start < i_size) { - /* Trim the write to the EOF; the extra data is ignored. Also - * put an upper limit on the size of a single storedata op. - */ - max_len = wsize; - max_len = min_t(unsigned long long, max_len, end - start + 1); - max_len = min_t(unsigned long long, max_len, i_size - start); - - if (len < max_len) { - int max_pages = INT_MAX; - -#ifdef CONFIG_CIFS_SMB_DIRECT - if (server->smbd_conn) - max_pages = server->smbd_conn->max_frmr_depth; -#endif - max_pages -= folio_nr_pages(folio); - - if (max_pages > 0) - cifs_extend_writeback(mapping, xas, &count, start, - max_pages, max_len, &len); - } - } - len = min_t(unsigned long long, len, i_size - start); - - /* We now have a contiguous set of dirty pages, each with writeback - * set; the first page is still locked at this point, but all the rest - * have been unlocked. - */ - folio_unlock(folio); - wdata->subreq.len = len; - - if (start < i_size) { - iov_iter_xarray(&wdata->subreq.io_iter, ITER_SOURCE, &mapping->i_pages, - start, len); - - rc = adjust_credits(wdata->server, &wdata->credits, wdata->subreq.len); - if (rc) - goto err_wdata; - - if (wdata->cfile->invalidHandle) - rc = -EAGAIN; - else - rc = wdata->server->ops->async_writev(wdata); - if (rc >= 0) { - cifs_put_writedata(wdata); - goto err_close; - } - } else { - /* The dirty region was entirely beyond the EOF. */ - cifs_pages_written_back(inode, start, len); - rc = 0; - } - -err_wdata: - cifs_put_writedata(wdata); -err_uncredit: - add_credits_and_wake_if(server, credits, 0); -err_close: - if (cfile) - cifsFileInfo_put(cfile); -err_xid: - free_xid(xid); - if (rc == 0) { - wbc->nr_to_write = count; - rc = len; - } else if (is_retryable_error(rc)) { - cifs_pages_write_redirty(inode, start, len); - } else { - cifs_pages_write_failed(inode, start, len); - mapping_set_error(mapping, rc); - } - /* Indication to update ctime and mtime as close is deferred */ - set_bit(CIFS_INO_MODIFIED_ATTR, &CIFS_I(inode)->flags); - return rc; -} - -/* - * write a region of pages back to the server - */ -static ssize_t cifs_writepages_begin(struct address_space *mapping, - struct writeback_control *wbc, - struct xa_state *xas, - unsigned long long *_start, - unsigned long long end) -{ - struct folio *folio; - unsigned long long start = *_start; - ssize_t ret; - int skips = 0; - -search_again: - /* Find the first dirty page. */ - rcu_read_lock(); - - for (;;) { - folio = xas_find_marked(xas, end / PAGE_SIZE, PAGECACHE_TAG_DIRTY); - if (xas_retry(xas, folio) || xa_is_value(folio)) - continue; - if (!folio) - break; - - if (!folio_try_get_rcu(folio)) { - xas_reset(xas); - continue; - } - - if (unlikely(folio != xas_reload(xas))) { - folio_put(folio); - xas_reset(xas); - continue; - } - - xas_pause(xas); - break; - } - rcu_read_unlock(); - if (!folio) - return 0; - - start = folio_pos(folio); /* May regress with THPs */ - - /* At this point we hold neither the i_pages lock nor the page lock: - * the page may be truncated or invalidated (changing page->mapping to - * NULL), or even swizzled back from swapper_space to tmpfs file - * mapping - */ -lock_again: - if (wbc->sync_mode != WB_SYNC_NONE) { - ret = folio_lock_killable(folio); - if (ret < 0) - return ret; - } else { - if (!folio_trylock(folio)) - goto search_again; - } - - if (folio->mapping != mapping || - !folio_test_dirty(folio)) { - start += folio_size(folio); - folio_unlock(folio); - goto search_again; - } - - if (folio_test_writeback(folio) || - folio_test_private_2(folio)) { /* [DEPRECATED] */ - folio_unlock(folio); - if (wbc->sync_mode != WB_SYNC_NONE) { - folio_wait_writeback(folio); -#ifdef CONFIG_CIFS_FSCACHE - folio_wait_private_2(folio); -#endif - goto lock_again; - } - - start += folio_size(folio); - if (wbc->sync_mode == WB_SYNC_NONE) { - if (skips >= 5 || need_resched()) { - ret = 0; - goto out; - } - skips++; - } - goto search_again; - } - - ret = cifs_write_back_from_locked_folio(mapping, wbc, xas, folio, start, end); -out: - if (ret > 0) - *_start = start + ret; - return ret; -} - -/* - * Write a region of pages back to the server - */ -static int cifs_writepages_region(struct address_space *mapping, - struct writeback_control *wbc, - unsigned long long *_start, - unsigned long long end) -{ - ssize_t ret; - - XA_STATE(xas, &mapping->i_pages, *_start / PAGE_SIZE); - - do { - ret = cifs_writepages_begin(mapping, wbc, &xas, _start, end); - if (ret > 0 && wbc->nr_to_write > 0) - cond_resched(); - } while (ret > 0 && wbc->nr_to_write > 0); - - return ret > 0 ? 0 : ret; -} - -/* - * Write some of the pending data back to the server - */ -static int cifs_writepages(struct address_space *mapping, - struct writeback_control *wbc) -{ - loff_t start, end; - int ret; - - /* We have to be careful as we can end up racing with setattr() - * truncating the pagecache since the caller doesn't take a lock here - * to prevent it. - */ - - if (wbc->range_cyclic && mapping->writeback_index) { - start = mapping->writeback_index * PAGE_SIZE; - ret = cifs_writepages_region(mapping, wbc, &start, LLONG_MAX); - if (ret < 0) - goto out; - - if (wbc->nr_to_write <= 0) { - mapping->writeback_index = start / PAGE_SIZE; - goto out; - } - - start = 0; - end = mapping->writeback_index * PAGE_SIZE; - mapping->writeback_index = 0; - ret = cifs_writepages_region(mapping, wbc, &start, end); - if (ret == 0) - mapping->writeback_index = start / PAGE_SIZE; - } else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) { - start = 0; - ret = cifs_writepages_region(mapping, wbc, &start, LLONG_MAX); - if (wbc->nr_to_write > 0 && ret == 0) - mapping->writeback_index = start / PAGE_SIZE; - } else { - start = wbc->range_start; - ret = cifs_writepages_region(mapping, wbc, &start, wbc->range_end); - } - -out: - return ret; -} - -static int cifs_write_end(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) -{ - int rc; - struct inode *inode = mapping->host; - struct cifsFileInfo *cfile = file->private_data; - struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb); - struct folio *folio = page_folio(page); - __u32 pid; - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) - pid = cfile->pid; - else - pid = current->tgid; - - cifs_dbg(FYI, "write_end for page %p from pos %lld with %d bytes\n", - page, pos, copied); - - if (folio_test_checked(folio)) { - if (copied == len) - folio_mark_uptodate(folio); - folio_clear_checked(folio); - } else if (!folio_test_uptodate(folio) && copied == PAGE_SIZE) - folio_mark_uptodate(folio); - - if (!folio_test_uptodate(folio)) { - char *page_data; - unsigned offset = pos & (PAGE_SIZE - 1); - unsigned int xid; - - xid = get_xid(); - /* this is probably better than directly calling - partialpage_write since in this function the file handle is - known which we might as well leverage */ - /* BB check if anything else missing out of ppw - such as updating last write time */ - page_data = kmap(page); - rc = cifs_write(cfile, pid, page_data + offset, copied, &pos); - /* if (rc < 0) should we set writebehind rc? */ - kunmap(page); - - free_xid(xid); - } else { - rc = copied; - pos += copied; - set_page_dirty(page); - } - - if (rc > 0) { - spin_lock(&inode->i_lock); - if (pos > inode->i_size) { - loff_t additional_blocks = (512 - 1 + copied) >> 9; - - i_size_write(inode, pos); - /* - * Estimate new allocation size based on the amount written. - * This will be updated from server on close (and on queryinfo) - */ - inode->i_blocks = min_t(blkcnt_t, (512 - 1 + pos) >> 9, - inode->i_blocks + additional_blocks); - } - spin_unlock(&inode->i_lock); - } - - unlock_page(page); - put_page(page); - /* Indication to update ctime and mtime as close is deferred */ - set_bit(CIFS_INO_MODIFIED_ATTR, &CIFS_I(inode)->flags); - - return rc; -} -#endif // End netfs removal 2773 - /* * Flush data on a strict file. */ @@ -4565,6 +3932,7 @@ cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset) } #endif // end netfslib remove 4633 + static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf) { return netfs_page_mkwrite(vmf, NULL); -- cgit v1.2.3-59-g8ed1b From b593634424d4ff1319226eb0187c634b0b819224 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 6 Oct 2023 22:36:30 +0100 Subject: cifs: Remove some code that's no longer used, part 3 Remove some code that was #if'd out with the netfslib conversion. This is split into parts for file.c as the diff generator otherwise produces a hard to read diff for part of it where a big chunk is cut out. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/file.c | 1004 -------------------------------------------------- 1 file changed, 1004 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index f504cda886f0..4c981ce89f8a 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -2743,471 +2743,6 @@ int cifs_flush(struct file *file, fl_owner_t id) return rc; } -#if 0 // TODO remove 3594 -static void collect_uncached_write_data(struct cifs_aio_ctx *ctx); - -static void -cifs_uncached_writev_complete(struct work_struct *work) -{ - struct cifs_io_subrequest *wdata = container_of(work, - struct cifs_io_subrequest, work); - struct inode *inode = d_inode(wdata->cfile->dentry); - struct cifsInodeInfo *cifsi = CIFS_I(inode); - - spin_lock(&inode->i_lock); - cifs_update_eof(cifsi, wdata->subreq.start, wdata->subreq.len); - if (cifsi->netfs.remote_i_size > inode->i_size) - i_size_write(inode, cifsi->netfs.remote_i_size); - spin_unlock(&inode->i_lock); - - complete(&wdata->done); - collect_uncached_write_data(wdata->ctx); - /* the below call can possibly free the last ref to aio ctx */ - cifs_put_writedata(wdata); -} - -static int -cifs_resend_wdata(struct cifs_io_subrequest *wdata, struct list_head *wdata_list, - struct cifs_aio_ctx *ctx) -{ - size_t wsize; - struct cifs_credits credits; - int rc; - struct TCP_Server_Info *server = wdata->server; - - do { - if (wdata->cfile->invalidHandle) { - rc = cifs_reopen_file(wdata->cfile, false); - if (rc == -EAGAIN) - continue; - else if (rc) - break; - } - - - /* - * Wait for credits to resend this wdata. - * Note: we are attempting to resend the whole wdata not in - * segments - */ - do { - rc = server->ops->wait_mtu_credits(server, wdata->subreq.len, - &wsize, &credits); - if (rc) - goto fail; - - if (wsize < wdata->subreq.len) { - add_credits_and_wake_if(server, &credits, 0); - msleep(1000); - } - } while (wsize < wdata->subreq.len); - wdata->credits = credits; - - rc = adjust_credits(server, &wdata->credits, wdata->subreq.len); - - if (!rc) { - if (wdata->cfile->invalidHandle) - rc = -EAGAIN; - else { - set_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags); -#ifdef CONFIG_CIFS_SMB_DIRECT - if (wdata->mr) { - wdata->mr->need_invalidate = true; - smbd_deregister_mr(wdata->mr); - wdata->mr = NULL; - } -#endif - rc = server->ops->async_writev(wdata); - } - } - - /* If the write was successfully sent, we are done */ - if (!rc) { - list_add_tail(&wdata->list, wdata_list); - return 0; - } - - /* Roll back credits and retry if needed */ - add_credits_and_wake_if(server, &wdata->credits, 0); - } while (rc == -EAGAIN); - -fail: - cifs_put_writedata(wdata); - return rc; -} - -/* - * Select span of a bvec iterator we're going to use. Limit it by both maximum - * size and maximum number of segments. - */ -static size_t cifs_limit_bvec_subset(const struct iov_iter *iter, size_t max_size, - size_t max_segs, unsigned int *_nsegs) -{ - const struct bio_vec *bvecs = iter->bvec; - unsigned int nbv = iter->nr_segs, ix = 0, nsegs = 0; - size_t len, span = 0, n = iter->count; - size_t skip = iter->iov_offset; - - if (WARN_ON(!iov_iter_is_bvec(iter)) || n == 0) - return 0; - - while (n && ix < nbv && skip) { - len = bvecs[ix].bv_len; - if (skip < len) - break; - skip -= len; - n -= len; - ix++; - } - - while (n && ix < nbv) { - len = min3(n, bvecs[ix].bv_len - skip, max_size); - span += len; - max_size -= len; - nsegs++; - ix++; - if (max_size == 0 || nsegs >= max_segs) - break; - skip = 0; - n -= len; - } - - *_nsegs = nsegs; - return span; -} - -static int -cifs_write_from_iter(loff_t fpos, size_t len, struct iov_iter *from, - struct cifsFileInfo *open_file, - struct cifs_sb_info *cifs_sb, struct list_head *wdata_list, - struct cifs_aio_ctx *ctx) -{ - int rc = 0; - size_t cur_len, max_len; - struct cifs_io_subrequest *wdata; - pid_t pid; - struct TCP_Server_Info *server; - unsigned int xid, max_segs = INT_MAX; - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) - pid = open_file->pid; - else - pid = current->tgid; - - server = cifs_pick_channel(tlink_tcon(open_file->tlink)->ses); - xid = get_xid(); - -#ifdef CONFIG_CIFS_SMB_DIRECT - if (server->smbd_conn) - max_segs = server->smbd_conn->max_frmr_depth; -#endif - - do { - struct cifs_credits credits_on_stack; - struct cifs_credits *credits = &credits_on_stack; - unsigned int nsegs = 0; - size_t wsize; - - if (signal_pending(current)) { - rc = -EINTR; - break; - } - - if (open_file->invalidHandle) { - rc = cifs_reopen_file(open_file, false); - if (rc == -EAGAIN) - continue; - else if (rc) - break; - } - - rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->wsize, - &wsize, credits); - if (rc) - break; - - max_len = min_t(const size_t, len, wsize); - if (!max_len) { - rc = -EAGAIN; - add_credits_and_wake_if(server, credits, 0); - break; - } - - cur_len = cifs_limit_bvec_subset(from, max_len, max_segs, &nsegs); - cifs_dbg(FYI, "write_from_iter len=%zx/%zx nsegs=%u/%lu/%u\n", - cur_len, max_len, nsegs, from->nr_segs, max_segs); - if (cur_len == 0) { - rc = -EIO; - add_credits_and_wake_if(server, credits, 0); - break; - } - - wdata = cifs_writedata_alloc(cifs_uncached_writev_complete); - if (!wdata) { - rc = -ENOMEM; - add_credits_and_wake_if(server, credits, 0); - break; - } - - wdata->uncached = true; - wdata->sync_mode = WB_SYNC_ALL; - wdata->subreq.start = (__u64)fpos; - wdata->cfile = cifsFileInfo_get(open_file); - wdata->server = server; - wdata->pid = pid; - wdata->subreq.len = cur_len; - wdata->credits = credits_on_stack; - wdata->subreq.io_iter = *from; - wdata->ctx = ctx; - kref_get(&ctx->refcount); - - iov_iter_truncate(&wdata->subreq.io_iter, cur_len); - - rc = adjust_credits(server, &wdata->credits, wdata->subreq.len); - - if (!rc) { - if (wdata->cfile->invalidHandle) - rc = -EAGAIN; - else - rc = server->ops->async_writev(wdata); - } - - if (rc) { - add_credits_and_wake_if(server, &wdata->credits, 0); - cifs_put_writedata(wdata); - if (rc == -EAGAIN) - continue; - break; - } - - list_add_tail(&wdata->list, wdata_list); - iov_iter_advance(from, cur_len); - fpos += cur_len; - len -= cur_len; - } while (len > 0); - - free_xid(xid); - return rc; -} - -static void collect_uncached_write_data(struct cifs_aio_ctx *ctx) -{ - struct cifs_io_subrequest *wdata, *tmp; - struct cifs_tcon *tcon; - struct cifs_sb_info *cifs_sb; - struct dentry *dentry = ctx->cfile->dentry; - ssize_t rc; - - tcon = tlink_tcon(ctx->cfile->tlink); - cifs_sb = CIFS_SB(dentry->d_sb); - - mutex_lock(&ctx->aio_mutex); - - if (list_empty(&ctx->list)) { - mutex_unlock(&ctx->aio_mutex); - return; - } - - rc = ctx->rc; - /* - * Wait for and collect replies for any successful sends in order of - * increasing offset. Once an error is hit, then return without waiting - * for any more replies. - */ -restart_loop: - list_for_each_entry_safe(wdata, tmp, &ctx->list, list) { - if (!rc) { - if (!try_wait_for_completion(&wdata->done)) { - mutex_unlock(&ctx->aio_mutex); - return; - } - - if (wdata->result) - rc = wdata->result; - else - ctx->total_len += wdata->subreq.len; - - /* resend call if it's a retryable error */ - if (rc == -EAGAIN) { - struct list_head tmp_list; - struct iov_iter tmp_from = ctx->iter; - - INIT_LIST_HEAD(&tmp_list); - list_del_init(&wdata->list); - - if (ctx->direct_io) - rc = cifs_resend_wdata( - wdata, &tmp_list, ctx); - else { - iov_iter_advance(&tmp_from, - wdata->subreq.start - ctx->pos); - - rc = cifs_write_from_iter(wdata->subreq.start, - wdata->subreq.len, &tmp_from, - ctx->cfile, cifs_sb, &tmp_list, - ctx); - - cifs_put_writedata(wdata); - } - - list_splice(&tmp_list, &ctx->list); - goto restart_loop; - } - } - list_del_init(&wdata->list); - cifs_put_writedata(wdata); - } - - cifs_stats_bytes_written(tcon, ctx->total_len); - set_bit(CIFS_INO_INVALID_MAPPING, &CIFS_I(dentry->d_inode)->flags); - - ctx->rc = (rc == 0) ? ctx->total_len : rc; - - mutex_unlock(&ctx->aio_mutex); - - if (ctx->iocb && ctx->iocb->ki_complete) - ctx->iocb->ki_complete(ctx->iocb, ctx->rc); - else - complete(&ctx->done); -} - -static ssize_t __cifs_writev( - struct kiocb *iocb, struct iov_iter *from, bool direct) -{ - struct file *file = iocb->ki_filp; - ssize_t total_written = 0; - struct cifsFileInfo *cfile; - struct cifs_tcon *tcon; - struct cifs_sb_info *cifs_sb; - struct cifs_aio_ctx *ctx; - int rc; - - rc = generic_write_checks(iocb, from); - if (rc <= 0) - return rc; - - cifs_sb = CIFS_FILE_SB(file); - cfile = file->private_data; - tcon = tlink_tcon(cfile->tlink); - - if (!tcon->ses->server->ops->async_writev) - return -ENOSYS; - - ctx = cifs_aio_ctx_alloc(); - if (!ctx) - return -ENOMEM; - - ctx->cfile = cifsFileInfo_get(cfile); - - if (!is_sync_kiocb(iocb)) - ctx->iocb = iocb; - - ctx->pos = iocb->ki_pos; - ctx->direct_io = direct; - ctx->nr_pinned_pages = 0; - - if (user_backed_iter(from)) { - /* - * Extract IOVEC/UBUF-type iterators to a BVEC-type iterator as - * they contain references to the calling process's virtual - * memory layout which won't be available in an async worker - * thread. This also takes a pin on every folio involved. - */ - rc = netfs_extract_user_iter(from, iov_iter_count(from), - &ctx->iter, 0); - if (rc < 0) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return rc; - } - - ctx->nr_pinned_pages = rc; - ctx->bv = (void *)ctx->iter.bvec; - ctx->bv_need_unpin = iov_iter_extract_will_pin(from); - } else if ((iov_iter_is_bvec(from) || iov_iter_is_kvec(from)) && - !is_sync_kiocb(iocb)) { - /* - * If the op is asynchronous, we need to copy the list attached - * to a BVEC/KVEC-type iterator, but we assume that the storage - * will be pinned by the caller; in any case, we may or may not - * be able to pin the pages, so we don't try. - */ - ctx->bv = (void *)dup_iter(&ctx->iter, from, GFP_KERNEL); - if (!ctx->bv) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return -ENOMEM; - } - } else { - /* - * Otherwise, we just pass the iterator down as-is and rely on - * the caller to make sure the pages referred to by the - * iterator don't evaporate. - */ - ctx->iter = *from; - } - - ctx->len = iov_iter_count(&ctx->iter); - - /* grab a lock here due to read response handlers can access ctx */ - mutex_lock(&ctx->aio_mutex); - - rc = cifs_write_from_iter(iocb->ki_pos, ctx->len, &ctx->iter, - cfile, cifs_sb, &ctx->list, ctx); - - /* - * If at least one write was successfully sent, then discard any rc - * value from the later writes. If the other write succeeds, then - * we'll end up returning whatever was written. If it fails, then - * we'll get a new rc value from that. - */ - if (!list_empty(&ctx->list)) - rc = 0; - - mutex_unlock(&ctx->aio_mutex); - - if (rc) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return rc; - } - - if (!is_sync_kiocb(iocb)) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return -EIOCBQUEUED; - } - - rc = wait_for_completion_killable(&ctx->done); - if (rc) { - mutex_lock(&ctx->aio_mutex); - ctx->rc = rc = -EINTR; - total_written = ctx->total_len; - mutex_unlock(&ctx->aio_mutex); - } else { - rc = ctx->rc; - total_written = ctx->total_len; - } - - kref_put(&ctx->refcount, cifs_aio_ctx_release); - - if (unlikely(!total_written)) - return rc; - - iocb->ki_pos += total_written; - return total_written; -} - -ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from) -{ - struct file *file = iocb->ki_filp; - - cifs_revalidate_mapping(file->f_inode); - return __cifs_writev(iocb, from, true); -} - -ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) -{ - return __cifs_writev(iocb, from, false); -} -#endif // TODO remove 3594 - static ssize_t cifs_writev(struct kiocb *iocb, struct iov_iter *from) { @@ -3296,450 +2831,6 @@ out: return written; } -#if 0 // TODO remove 4143 -static struct cifs_io_subrequest *cifs_readdata_alloc(work_func_t complete) -{ - struct cifs_io_subrequest *rdata; - - rdata = kzalloc(sizeof(*rdata), GFP_KERNEL); - if (rdata) { - refcount_set(&rdata->subreq.ref, 1); - INIT_LIST_HEAD(&rdata->list); - init_completion(&rdata->done); - INIT_WORK(&rdata->work, complete); - } - - return rdata; -} - -void -cifs_readdata_release(struct cifs_io_subrequest *rdata) -{ - if (rdata->ctx) - kref_put(&rdata->ctx->refcount, cifs_aio_ctx_release); -#ifdef CONFIG_CIFS_SMB_DIRECT - if (rdata->mr) { - smbd_deregister_mr(rdata->mr); - rdata->mr = NULL; - } -#endif - if (rdata->cfile) - cifsFileInfo_put(rdata->cfile); - - kfree(rdata); -} - -static void collect_uncached_read_data(struct cifs_aio_ctx *ctx); - -static void -cifs_uncached_readv_complete(struct work_struct *work) -{ - struct cifs_io_subrequest *rdata = - container_of(work, struct cifs_io_subrequest, work); - - complete(&rdata->done); - collect_uncached_read_data(rdata->ctx); - /* the below call can possibly free the last ref to aio ctx */ - cifs_put_readdata(rdata); -} - -static int cifs_resend_rdata(struct cifs_io_subrequest *rdata, - struct list_head *rdata_list, - struct cifs_aio_ctx *ctx) -{ - size_t rsize; - struct cifs_credits credits; - int rc; - struct TCP_Server_Info *server; - - /* XXX: should we pick a new channel here? */ - server = rdata->server; - - do { - if (rdata->cfile->invalidHandle) { - rc = cifs_reopen_file(rdata->cfile, true); - if (rc == -EAGAIN) - continue; - else if (rc) - break; - } - - /* - * Wait for credits to resend this rdata. - * Note: we are attempting to resend the whole rdata not in - * segments - */ - do { - rc = server->ops->wait_mtu_credits(server, rdata->subreq.len, - &rsize, &credits); - - if (rc) - goto fail; - - if (rsize < rdata->subreq.len) { - add_credits_and_wake_if(server, &credits, 0); - msleep(1000); - } - } while (rsize < rdata->subreq.len); - rdata->credits = credits; - - rc = adjust_credits(server, &rdata->credits, rdata->subreq.len); - if (!rc) { - if (rdata->cfile->invalidHandle) - rc = -EAGAIN; - else { -#ifdef CONFIG_CIFS_SMB_DIRECT - if (rdata->mr) { - rdata->mr->need_invalidate = true; - smbd_deregister_mr(rdata->mr); - rdata->mr = NULL; - } -#endif - rc = server->ops->async_readv(rdata); - } - } - - /* If the read was successfully sent, we are done */ - if (!rc) { - /* Add to aio pending list */ - list_add_tail(&rdata->list, rdata_list); - return 0; - } - - /* Roll back credits and retry if needed */ - add_credits_and_wake_if(server, &rdata->credits, 0); - } while (rc == -EAGAIN); - -fail: - cifs_put_readdata(rdata); - return rc; -} - -static int -cifs_send_async_read(loff_t fpos, size_t len, struct cifsFileInfo *open_file, - struct cifs_sb_info *cifs_sb, struct list_head *rdata_list, - struct cifs_aio_ctx *ctx) -{ - struct cifs_io_subrequest *rdata; - unsigned int nsegs, max_segs = INT_MAX; - struct cifs_credits credits_on_stack; - struct cifs_credits *credits = &credits_on_stack; - size_t cur_len, max_len, rsize; - int rc; - pid_t pid; - struct TCP_Server_Info *server; - - server = cifs_pick_channel(tlink_tcon(open_file->tlink)->ses); - -#ifdef CONFIG_CIFS_SMB_DIRECT - if (server->smbd_conn) - max_segs = server->smbd_conn->max_frmr_depth; -#endif - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) - pid = open_file->pid; - else - pid = current->tgid; - - do { - if (open_file->invalidHandle) { - rc = cifs_reopen_file(open_file, true); - if (rc == -EAGAIN) - continue; - else if (rc) - break; - } - - if (cifs_sb->ctx->rsize == 0) - cifs_sb->ctx->rsize = - server->ops->negotiate_rsize(tlink_tcon(open_file->tlink), - cifs_sb->ctx); - - rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->rsize, - &rsize, credits); - if (rc) - break; - - max_len = min_t(size_t, len, rsize); - - cur_len = cifs_limit_bvec_subset(&ctx->iter, max_len, - max_segs, &nsegs); - cifs_dbg(FYI, "read-to-iter len=%zx/%zx nsegs=%u/%lu/%u\n", - cur_len, max_len, nsegs, ctx->iter.nr_segs, max_segs); - if (cur_len == 0) { - rc = -EIO; - add_credits_and_wake_if(server, credits, 0); - break; - } - - rdata = cifs_readdata_alloc(cifs_uncached_readv_complete); - if (!rdata) { - add_credits_and_wake_if(server, credits, 0); - rc = -ENOMEM; - break; - } - - rdata->server = server; - rdata->cfile = cifsFileInfo_get(open_file); - rdata->subreq.start = fpos; - rdata->subreq.len = cur_len; - rdata->pid = pid; - rdata->credits = credits_on_stack; - rdata->ctx = ctx; - kref_get(&ctx->refcount); - - rdata->subreq.io_iter = ctx->iter; - iov_iter_truncate(&rdata->subreq.io_iter, cur_len); - - rc = adjust_credits(server, &rdata->credits, rdata->subreq.len); - - if (!rc) { - if (rdata->cfile->invalidHandle) - rc = -EAGAIN; - else - rc = server->ops->async_readv(rdata); - } - - if (rc) { - add_credits_and_wake_if(server, &rdata->credits, 0); - cifs_put_readdata(rdata); - if (rc == -EAGAIN) - continue; - break; - } - - list_add_tail(&rdata->list, rdata_list); - iov_iter_advance(&ctx->iter, cur_len); - fpos += cur_len; - len -= cur_len; - } while (len > 0); - - return rc; -} - -static void -collect_uncached_read_data(struct cifs_aio_ctx *ctx) -{ - struct cifs_io_subrequest *rdata, *tmp; - struct cifs_sb_info *cifs_sb; - int rc; - - cifs_sb = CIFS_SB(ctx->cfile->dentry->d_sb); - - mutex_lock(&ctx->aio_mutex); - - if (list_empty(&ctx->list)) { - mutex_unlock(&ctx->aio_mutex); - return; - } - - rc = ctx->rc; - /* the loop below should proceed in the order of increasing offsets */ -again: - list_for_each_entry_safe(rdata, tmp, &ctx->list, list) { - if (!rc) { - if (!try_wait_for_completion(&rdata->done)) { - mutex_unlock(&ctx->aio_mutex); - return; - } - - if (rdata->result == -EAGAIN) { - /* resend call if it's a retryable error */ - struct list_head tmp_list; - unsigned int got_bytes = rdata->got_bytes; - - list_del_init(&rdata->list); - INIT_LIST_HEAD(&tmp_list); - - if (ctx->direct_io) { - /* - * Re-use rdata as this is a - * direct I/O - */ - rc = cifs_resend_rdata( - rdata, - &tmp_list, ctx); - } else { - rc = cifs_send_async_read( - rdata->subreq.start + got_bytes, - rdata->subreq.len - got_bytes, - rdata->cfile, cifs_sb, - &tmp_list, ctx); - - cifs_put_readdata(rdata); - } - - list_splice(&tmp_list, &ctx->list); - - goto again; - } else if (rdata->result) - rc = rdata->result; - - /* if there was a short read -- discard anything left */ - if (rdata->got_bytes && rdata->got_bytes < rdata->subreq.len) - rc = -ENODATA; - - ctx->total_len += rdata->got_bytes; - } - list_del_init(&rdata->list); - cifs_put_readdata(rdata); - } - - /* mask nodata case */ - if (rc == -ENODATA) - rc = 0; - - ctx->rc = (rc == 0) ? (ssize_t)ctx->total_len : rc; - - mutex_unlock(&ctx->aio_mutex); - - if (ctx->iocb && ctx->iocb->ki_complete) - ctx->iocb->ki_complete(ctx->iocb, ctx->rc); - else - complete(&ctx->done); -} - -static ssize_t __cifs_readv( - struct kiocb *iocb, struct iov_iter *to, bool direct) -{ - size_t len; - struct file *file = iocb->ki_filp; - struct cifs_sb_info *cifs_sb; - struct cifsFileInfo *cfile; - struct cifs_tcon *tcon; - ssize_t rc, total_read = 0; - loff_t offset = iocb->ki_pos; - struct cifs_aio_ctx *ctx; - - len = iov_iter_count(to); - if (!len) - return 0; - - cifs_sb = CIFS_FILE_SB(file); - cfile = file->private_data; - tcon = tlink_tcon(cfile->tlink); - - if (!tcon->ses->server->ops->async_readv) - return -ENOSYS; - - if ((file->f_flags & O_ACCMODE) == O_WRONLY) - cifs_dbg(FYI, "attempting read on write only file instance\n"); - - ctx = cifs_aio_ctx_alloc(); - if (!ctx) - return -ENOMEM; - - ctx->pos = offset; - ctx->direct_io = direct; - ctx->len = len; - ctx->cfile = cifsFileInfo_get(cfile); - ctx->nr_pinned_pages = 0; - - if (!is_sync_kiocb(iocb)) - ctx->iocb = iocb; - - if (user_backed_iter(to)) { - /* - * Extract IOVEC/UBUF-type iterators to a BVEC-type iterator as - * they contain references to the calling process's virtual - * memory layout which won't be available in an async worker - * thread. This also takes a pin on every folio involved. - */ - rc = netfs_extract_user_iter(to, iov_iter_count(to), - &ctx->iter, 0); - if (rc < 0) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return rc; - } - - ctx->nr_pinned_pages = rc; - ctx->bv = (void *)ctx->iter.bvec; - ctx->bv_need_unpin = iov_iter_extract_will_pin(to); - ctx->should_dirty = true; - } else if ((iov_iter_is_bvec(to) || iov_iter_is_kvec(to)) && - !is_sync_kiocb(iocb)) { - /* - * If the op is asynchronous, we need to copy the list attached - * to a BVEC/KVEC-type iterator, but we assume that the storage - * will be retained by the caller; in any case, we may or may - * not be able to pin the pages, so we don't try. - */ - ctx->bv = (void *)dup_iter(&ctx->iter, to, GFP_KERNEL); - if (!ctx->bv) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return -ENOMEM; - } - } else { - /* - * Otherwise, we just pass the iterator down as-is and rely on - * the caller to make sure the pages referred to by the - * iterator don't evaporate. - */ - ctx->iter = *to; - } - - if (direct) { - rc = filemap_write_and_wait_range(file->f_inode->i_mapping, - offset, offset + len - 1); - if (rc) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return -EAGAIN; - } - } - - /* grab a lock here due to read response handlers can access ctx */ - mutex_lock(&ctx->aio_mutex); - - rc = cifs_send_async_read(offset, len, cfile, cifs_sb, &ctx->list, ctx); - - /* if at least one read request send succeeded, then reset rc */ - if (!list_empty(&ctx->list)) - rc = 0; - - mutex_unlock(&ctx->aio_mutex); - - if (rc) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return rc; - } - - if (!is_sync_kiocb(iocb)) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return -EIOCBQUEUED; - } - - rc = wait_for_completion_killable(&ctx->done); - if (rc) { - mutex_lock(&ctx->aio_mutex); - ctx->rc = rc = -EINTR; - total_read = ctx->total_len; - mutex_unlock(&ctx->aio_mutex); - } else { - rc = ctx->rc; - total_read = ctx->total_len; - } - - kref_put(&ctx->refcount, cifs_aio_ctx_release); - - if (total_read) { - iocb->ki_pos += total_read; - return total_read; - } - return rc; -} - -ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to) -{ - return __cifs_readv(iocb, to, true); -} - -ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to) -{ - return __cifs_readv(iocb, to, false); - -} -#endif // end netfslib removal 4143 - ssize_t cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter) { ssize_t rc; @@ -3838,101 +2929,6 @@ cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) return rc; } -#if 0 // TODO remove 4633 -static ssize_t -cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset) -{ - int rc = -EACCES; - unsigned int bytes_read = 0; - unsigned int total_read; - unsigned int current_read_size; - unsigned int rsize; - struct cifs_sb_info *cifs_sb; - struct cifs_tcon *tcon; - struct TCP_Server_Info *server; - unsigned int xid; - char *cur_offset; - struct cifsFileInfo *open_file; - struct cifs_io_parms io_parms = {0}; - int buf_type = CIFS_NO_BUFFER; - __u32 pid; - - xid = get_xid(); - cifs_sb = CIFS_FILE_SB(file); - - /* FIXME: set up handlers for larger reads and/or convert to async */ - rsize = min_t(unsigned int, cifs_sb->ctx->rsize, CIFSMaxBufSize); - - if (file->private_data == NULL) { - rc = -EBADF; - free_xid(xid); - return rc; - } - open_file = file->private_data; - tcon = tlink_tcon(open_file->tlink); - server = cifs_pick_channel(tcon->ses); - - if (!server->ops->sync_read) { - free_xid(xid); - return -ENOSYS; - } - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) - pid = open_file->pid; - else - pid = current->tgid; - - if ((file->f_flags & O_ACCMODE) == O_WRONLY) - cifs_dbg(FYI, "attempting read on write only file instance\n"); - - for (total_read = 0, cur_offset = read_data; read_size > total_read; - total_read += bytes_read, cur_offset += bytes_read) { - do { - current_read_size = min_t(uint, read_size - total_read, - rsize); - /* - * For windows me and 9x we do not want to request more - * than it negotiated since it will refuse the read - * then. - */ - if (!(tcon->ses->capabilities & - tcon->ses->server->vals->cap_large_files)) { - current_read_size = min_t(uint, - current_read_size, CIFSMaxBufSize); - } - if (open_file->invalidHandle) { - rc = cifs_reopen_file(open_file, true); - if (rc != 0) - break; - } - io_parms.pid = pid; - io_parms.tcon = tcon; - io_parms.offset = *offset; - io_parms.length = current_read_size; - io_parms.server = server; - rc = server->ops->sync_read(xid, &open_file->fid, &io_parms, - &bytes_read, &cur_offset, - &buf_type); - } while (rc == -EAGAIN); - - if (rc || (bytes_read == 0)) { - if (total_read) { - break; - } else { - free_xid(xid); - return rc; - } - } else { - cifs_stats_bytes_read(tcon, total_read); - *offset += bytes_read; - } - } - free_xid(xid); - return total_read; -} -#endif // end netfslib remove 4633 - - static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf) { return netfs_page_mkwrite(vmf, NULL); -- cgit v1.2.3-59-g8ed1b From 7c1ac89480e8d5d34d38a868642216c8f05ee602 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 17 Apr 2024 15:00:55 +0100 Subject: cifs: Enable large folio support Now that cifs is using netfslib for its VM interaction, it only sees I/O in terms of iov_iter iterators and does not see pages or folios. This makes large multipage folios transparent to cifs and so we can turn on multipage folios on regular files. Signed-off-by: David Howells cc: Steve French cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/smb/client/inode.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index 5239c823de86..e8bfeea23660 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -72,6 +72,7 @@ static void cifs_set_ops(struct inode *inode) inode->i_data.a_ops = &cifs_addr_ops_smallbuf; else inode->i_data.a_ops = &cifs_addr_ops; + mapping_set_large_folios(inode->i_mapping); break; case S_IFDIR: if (IS_AUTOMOUNT(inode)) { -- cgit v1.2.3-59-g8ed1b From 0a960ba49869ebe8ff859d000351504dd6b93b68 Mon Sep 17 00:00:00 2001 From: "Tyler Hicks (Microsoft)" Date: Tue, 30 Apr 2024 19:56:46 -0500 Subject: proc: Move fdinfo PTRACE_MODE_READ check into the inode .permission operation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The following commits loosened the permissions of /proc//fdinfo/ directory, as well as the files within it, from 0500 to 0555 while also introducing a PTRACE_MODE_READ check between the current task and 's task: - commit 7bc3fa0172a4 ("procfs: allow reading fdinfo with PTRACE_MODE_READ") - commit 1927e498aee1 ("procfs: prevent unprivileged processes accessing fdinfo dir") Before those changes, inode based system calls like inotify_add_watch(2) would fail when the current task didn't have sufficient read permissions: [...] lstat("/proc/1/task/1/fdinfo", {st_mode=S_IFDIR|0500, st_size=0, ...}) = 0 inotify_add_watch(64, "/proc/1/task/1/fdinfo", IN_MODIFY|IN_ATTRIB|IN_MOVED_FROM|IN_MOVED_TO|IN_CREATE|IN_DELETE| IN_ONLYDIR|IN_DONT_FOLLOW|IN_EXCL_UNLINK) = -1 EACCES (Permission denied) [...] This matches the documented behavior in the inotify_add_watch(2) man page: ERRORS EACCES Read access to the given file is not permitted. After those changes, inotify_add_watch(2) started succeeding despite the current task not having PTRACE_MODE_READ privileges on the target task: [...] lstat("/proc/1/task/1/fdinfo", {st_mode=S_IFDIR|0555, st_size=0, ...}) = 0 inotify_add_watch(64, "/proc/1/task/1/fdinfo", IN_MODIFY|IN_ATTRIB|IN_MOVED_FROM|IN_MOVED_TO|IN_CREATE|IN_DELETE| IN_ONLYDIR|IN_DONT_FOLLOW|IN_EXCL_UNLINK) = 1757 openat(AT_FDCWD, "/proc/1/task/1/fdinfo", O_RDONLY|O_NONBLOCK|O_CLOEXEC|O_DIRECTORY) = -1 EACCES (Permission denied) [...] This change in behavior broke .NET prior to v7. See the github link below for the v7 commit that inadvertently/quietly (?) fixed .NET after the kernel changes mentioned above. Return to the old behavior by moving the PTRACE_MODE_READ check out of the file .open operation and into the inode .permission operation: [...] lstat("/proc/1/task/1/fdinfo", {st_mode=S_IFDIR|0555, st_size=0, ...}) = 0 inotify_add_watch(64, "/proc/1/task/1/fdinfo", IN_MODIFY|IN_ATTRIB|IN_MOVED_FROM|IN_MOVED_TO|IN_CREATE|IN_DELETE| IN_ONLYDIR|IN_DONT_FOLLOW|IN_EXCL_UNLINK) = -1 EACCES (Permission denied) [...] Reported-by: Kevin Parsons (Microsoft) Link: https://github.com/dotnet/runtime/commit/89e5469ac591b82d38510fe7de98346cce74ad4f Link: https://stackoverflow.com/questions/75379065/start-self-contained-net6-build-exe-as-service-on-raspbian-system-unauthorizeda Fixes: 7bc3fa0172a4 ("procfs: allow reading fdinfo with PTRACE_MODE_READ") Cc: stable@vger.kernel.org Cc: Christian Brauner Cc: Christian König Cc: Jann Horn Cc: Kalesh Singh Cc: Hardik Garg Cc: Allen Pais Signed-off-by: Tyler Hicks (Microsoft) Link: https://lore.kernel.org/r/20240501005646.745089-1-code@tyhicks.com Signed-off-by: Christian Brauner --- fs/proc/fd.c | 42 ++++++++++++++++++++---------------------- 1 file changed, 20 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/proc/fd.c b/fs/proc/fd.c index 6e72e5ad42bc..f4b1c8b42a51 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -74,7 +74,18 @@ out: return 0; } -static int proc_fdinfo_access_allowed(struct inode *inode) +static int seq_fdinfo_open(struct inode *inode, struct file *file) +{ + return single_open(file, seq_show, inode); +} + +/** + * Shared /proc/pid/fdinfo and /proc/pid/fdinfo/fd permission helper to ensure + * that the current task has PTRACE_MODE_READ in addition to the normal + * POSIX-like checks. + */ +static int proc_fdinfo_permission(struct mnt_idmap *idmap, struct inode *inode, + int mask) { bool allowed = false; struct task_struct *task = get_proc_task(inode); @@ -88,18 +99,13 @@ static int proc_fdinfo_access_allowed(struct inode *inode) if (!allowed) return -EACCES; - return 0; + return generic_permission(idmap, inode, mask); } -static int seq_fdinfo_open(struct inode *inode, struct file *file) -{ - int ret = proc_fdinfo_access_allowed(inode); - - if (ret) - return ret; - - return single_open(file, seq_show, inode); -} +static const struct inode_operations proc_fdinfo_file_inode_operations = { + .permission = proc_fdinfo_permission, + .setattr = proc_setattr, +}; static const struct file_operations proc_fdinfo_file_operations = { .open = seq_fdinfo_open, @@ -388,6 +394,8 @@ static struct dentry *proc_fdinfo_instantiate(struct dentry *dentry, ei = PROC_I(inode); ei->fd = data->fd; + inode->i_op = &proc_fdinfo_file_inode_operations; + inode->i_fop = &proc_fdinfo_file_operations; tid_fd_update_inode(task, inode, 0); @@ -407,23 +415,13 @@ static int proc_readfdinfo(struct file *file, struct dir_context *ctx) proc_fdinfo_instantiate); } -static int proc_open_fdinfo(struct inode *inode, struct file *file) -{ - int ret = proc_fdinfo_access_allowed(inode); - - if (ret) - return ret; - - return 0; -} - const struct inode_operations proc_fdinfo_inode_operations = { .lookup = proc_lookupfdinfo, + .permission = proc_fdinfo_permission, .setattr = proc_setattr, }; const struct file_operations proc_fdinfo_operations = { - .open = proc_open_fdinfo, .read = generic_read_dir, .iterate_shared = proc_readfdinfo, .llseek = generic_file_llseek, -- cgit v1.2.3-59-g8ed1b From 45751097aeb386eb239f6a8ed0ccfd7dabce068e Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Mon, 22 Apr 2024 12:24:06 +0200 Subject: seq_file: Optimize seq_puts() Most of seq_puts() usages are done with a string literal. In such cases, the length of the string car be computed at compile time in order to save a strlen() call at run-time. seq_putc() or seq_write() can then be used instead. This saves a few cycles. To have an estimation of how often this optimization triggers: $ git grep seq_puts.*\" | wc -l 3436 $ git grep seq_puts.*\".\" | wc -l 84 Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/a8589bffe4830dafcb9111e22acf06603fea7132.1713781332.git.christophe.jaillet@wanadoo.fr Signed-off-by: Christian Brauner The output for seq_putc() generation has also be checked and works. --- fs/seq_file.c | 4 ++-- include/linux/seq_file.h | 13 ++++++++++++- 2 files changed, 14 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/seq_file.c b/fs/seq_file.c index f5fdaf3b1572..8ef0a07033ca 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -669,7 +669,7 @@ void seq_putc(struct seq_file *m, char c) } EXPORT_SYMBOL(seq_putc); -void seq_puts(struct seq_file *m, const char *s) +void __seq_puts(struct seq_file *m, const char *s) { int len = strlen(s); @@ -680,7 +680,7 @@ void seq_puts(struct seq_file *m, const char *s) memcpy(m->buf + m->count, s, len); m->count += len; } -EXPORT_SYMBOL(seq_puts); +EXPORT_SYMBOL(__seq_puts); /** * seq_put_decimal_ull_width - A helper routine for putting decimal numbers diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index 234bcdb1fba4..8bd4fda6e027 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -118,7 +118,18 @@ void seq_vprintf(struct seq_file *m, const char *fmt, va_list args); __printf(2, 3) void seq_printf(struct seq_file *m, const char *fmt, ...); void seq_putc(struct seq_file *m, char c); -void seq_puts(struct seq_file *m, const char *s); +void __seq_puts(struct seq_file *m, const char *s); + +static __always_inline void seq_puts(struct seq_file *m, const char *s) +{ + if (!__builtin_constant_p(*s)) + __seq_puts(m, s); + else if (s[0] && !s[1]) + seq_putc(m, s[0]); + else + seq_write(m, s, __builtin_strlen(s)); +} + void seq_put_decimal_ull_width(struct seq_file *m, const char *delimiter, unsigned long long num, unsigned int width); void seq_put_decimal_ull(struct seq_file *m, const char *delimiter, -- cgit v1.2.3-59-g8ed1b From e035af9f6ebacd98774b1be2af58a5afd6d0d291 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Mon, 22 Apr 2024 12:24:07 +0200 Subject: seq_file: Simplify __seq_puts() Change the implementation of the out-of-line __seq_puts() to simply be a seq_write() call instead of duplicating the overflow/memcpy logic. Suggested-by: Rasmus Villemoes Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/7cebc1412d8d1338a7e52cc9291d00f5368c14e4.1713781332.git.christophe.jaillet@wanadoo.fr Signed-off-by: Christian Brauner --- fs/seq_file.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/seq_file.c b/fs/seq_file.c index 8ef0a07033ca..e676c8b0cf5d 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -671,14 +671,7 @@ EXPORT_SYMBOL(seq_putc); void __seq_puts(struct seq_file *m, const char *s) { - int len = strlen(s); - - if (m->count + len >= m->size) { - seq_set_overflow(m); - return; - } - memcpy(m->buf + m->count, s, len); - m->count += len; + seq_write(m, s, strlen(s)); } EXPORT_SYMBOL(__seq_puts); -- cgit v1.2.3-59-g8ed1b From a86f8671d03e6eb31abaefdf6928b92df0a2368c Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Thu, 2 May 2024 07:48:35 -0700 Subject: xfs: use unsigned ints for non-negative quantities in xfs_attr_remote.c In the next few patches we're going to refactor the attr remote code so that we can support headerless remote xattr values for storing merkle tree blocks. For now, let's change the code to use unsigned int to describe quantities of bytes and blocks that cannot be negative. Signed-off-by: Darrick J. Wong Reviewed-by: Andrey Albershteyn Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_remote.c | 61 ++++++++++++++++++++--------------------- fs/xfs/libxfs/xfs_attr_remote.h | 2 +- 2 files changed, 31 insertions(+), 32 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index beb0efdd8f6b..7c38c6feb8c9 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -47,13 +47,13 @@ * Each contiguous block has a header, so it is not just a simple attribute * length to FSB conversion. */ -int +unsigned int xfs_attr3_rmt_blocks( - struct xfs_mount *mp, - int attrlen) + struct xfs_mount *mp, + unsigned int attrlen) { if (xfs_has_crc(mp)) { - int buflen = XFS_ATTR3_RMT_BUF_SPACE(mp, mp->m_sb.sb_blocksize); + unsigned int buflen = XFS_ATTR3_RMT_BUF_SPACE(mp, mp->m_sb.sb_blocksize); return (attrlen + buflen - 1) / buflen; } return XFS_B_TO_FSB(mp, attrlen); @@ -92,7 +92,6 @@ xfs_attr3_rmt_verify( struct xfs_mount *mp, struct xfs_buf *bp, void *ptr, - int fsbsize, xfs_daddr_t bno) { struct xfs_attr3_rmt_hdr *rmt = ptr; @@ -103,7 +102,7 @@ xfs_attr3_rmt_verify( return __this_address; if (be64_to_cpu(rmt->rm_blkno) != bno) return __this_address; - if (be32_to_cpu(rmt->rm_bytes) > fsbsize - sizeof(*rmt)) + if (be32_to_cpu(rmt->rm_bytes) > mp->m_attr_geo->blksize - sizeof(*rmt)) return __this_address; if (be32_to_cpu(rmt->rm_offset) + be32_to_cpu(rmt->rm_bytes) > XFS_XATTR_SIZE_MAX) @@ -122,9 +121,9 @@ __xfs_attr3_rmt_read_verify( { struct xfs_mount *mp = bp->b_mount; char *ptr; - int len; + unsigned int len; xfs_daddr_t bno; - int blksize = mp->m_attr_geo->blksize; + unsigned int blksize = mp->m_attr_geo->blksize; /* no verification of non-crc buffers */ if (!xfs_has_crc(mp)) @@ -141,7 +140,7 @@ __xfs_attr3_rmt_read_verify( *failaddr = __this_address; return -EFSBADCRC; } - *failaddr = xfs_attr3_rmt_verify(mp, bp, ptr, blksize, bno); + *failaddr = xfs_attr3_rmt_verify(mp, bp, ptr, bno); if (*failaddr) return -EFSCORRUPTED; len -= blksize; @@ -186,7 +185,7 @@ xfs_attr3_rmt_write_verify( { struct xfs_mount *mp = bp->b_mount; xfs_failaddr_t fa; - int blksize = mp->m_attr_geo->blksize; + unsigned int blksize = mp->m_attr_geo->blksize; char *ptr; int len; xfs_daddr_t bno; @@ -203,7 +202,7 @@ xfs_attr3_rmt_write_verify( while (len > 0) { struct xfs_attr3_rmt_hdr *rmt = (struct xfs_attr3_rmt_hdr *)ptr; - fa = xfs_attr3_rmt_verify(mp, bp, ptr, blksize, bno); + fa = xfs_attr3_rmt_verify(mp, bp, ptr, bno); if (fa) { xfs_verifier_error(bp, -EFSCORRUPTED, fa); return; @@ -281,20 +280,20 @@ xfs_attr_rmtval_copyout( struct xfs_buf *bp, struct xfs_inode *dp, xfs_ino_t owner, - int *offset, - int *valuelen, + unsigned int *offset, + unsigned int *valuelen, uint8_t **dst) { char *src = bp->b_addr; xfs_daddr_t bno = xfs_buf_daddr(bp); - int len = BBTOB(bp->b_length); - int blksize = mp->m_attr_geo->blksize; + unsigned int len = BBTOB(bp->b_length); + unsigned int blksize = mp->m_attr_geo->blksize; ASSERT(len >= blksize); while (len > 0 && *valuelen > 0) { - int hdr_size = 0; - int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize); + unsigned int hdr_size = 0; + unsigned int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize); byte_cnt = min(*valuelen, byte_cnt); @@ -330,20 +329,20 @@ xfs_attr_rmtval_copyin( struct xfs_mount *mp, struct xfs_buf *bp, xfs_ino_t ino, - int *offset, - int *valuelen, + unsigned int *offset, + unsigned int *valuelen, uint8_t **src) { char *dst = bp->b_addr; xfs_daddr_t bno = xfs_buf_daddr(bp); - int len = BBTOB(bp->b_length); - int blksize = mp->m_attr_geo->blksize; + unsigned int len = BBTOB(bp->b_length); + unsigned int blksize = mp->m_attr_geo->blksize; ASSERT(len >= blksize); while (len > 0 && *valuelen > 0) { - int hdr_size; - int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize); + unsigned int hdr_size; + unsigned int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize); byte_cnt = min(*valuelen, byte_cnt); hdr_size = xfs_attr3_rmt_hdr_set(mp, dst, ino, *offset, @@ -389,12 +388,12 @@ xfs_attr_rmtval_get( struct xfs_buf *bp; xfs_dablk_t lblkno = args->rmtblkno; uint8_t *dst = args->value; - int valuelen; + unsigned int valuelen; int nmap; int error; - int blkcnt = args->rmtblkcnt; + unsigned int blkcnt = args->rmtblkcnt; int i; - int offset = 0; + unsigned int offset = 0; trace_xfs_attr_rmtval_get(args); @@ -452,7 +451,7 @@ xfs_attr_rmt_find_hole( struct xfs_inode *dp = args->dp; struct xfs_mount *mp = dp->i_mount; int error; - int blkcnt; + unsigned int blkcnt; xfs_fileoff_t lfileoff = 0; /* @@ -481,11 +480,11 @@ xfs_attr_rmtval_set_value( struct xfs_bmbt_irec map; xfs_dablk_t lblkno; uint8_t *src = args->value; - int blkcnt; - int valuelen; + unsigned int blkcnt; + unsigned int valuelen; int nmap; int error; - int offset = 0; + unsigned int offset = 0; /* * Roll through the "value", copying the attribute value to the @@ -644,7 +643,7 @@ xfs_attr_rmtval_invalidate( struct xfs_da_args *args) { xfs_dablk_t lblkno; - int blkcnt; + unsigned int blkcnt; int error; /* diff --git a/fs/xfs/libxfs/xfs_attr_remote.h b/fs/xfs/libxfs/xfs_attr_remote.h index d097ec6c4dc3..c64b04f91caf 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.h +++ b/fs/xfs/libxfs/xfs_attr_remote.h @@ -6,7 +6,7 @@ #ifndef __XFS_ATTR_REMOTE_H__ #define __XFS_ATTR_REMOTE_H__ -int xfs_attr3_rmt_blocks(struct xfs_mount *mp, int attrlen); +unsigned int xfs_attr3_rmt_blocks(struct xfs_mount *mp, unsigned int attrlen); int xfs_attr_rmtval_get(struct xfs_da_args *args); int xfs_attr_rmtval_stale(struct xfs_inode *ip, struct xfs_bmbt_irec *map, -- cgit v1.2.3-59-g8ed1b From a5714b67cad586f44777ad834e577037ce6b64fd Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Thu, 2 May 2024 07:48:36 -0700 Subject: xfs: turn XFS_ATTR3_RMT_BUF_SPACE into a function Turn this into a properly typechecked function, and actually use the correct blocksize for extended attributes. The function cannot be static inline because xfsprogs userspace uses it. Signed-off-by: Darrick J. Wong Reviewed-by: Andrey Albershteyn Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_remote.c | 19 ++++++++++++++++--- fs/xfs/libxfs/xfs_da_format.h | 4 +--- 2 files changed, 17 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 7c38c6feb8c9..043b837a3ef7 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -43,6 +43,19 @@ * the logging system and therefore never have a log item. */ +/* How many bytes can be stored in a remote value buffer? */ +inline unsigned int +xfs_attr3_rmt_buf_space( + struct xfs_mount *mp) +{ + unsigned int blocksize = mp->m_attr_geo->blksize; + + if (xfs_has_crc(mp)) + return blocksize - sizeof(struct xfs_attr3_rmt_hdr); + + return blocksize; +} + /* * Each contiguous block has a header, so it is not just a simple attribute * length to FSB conversion. @@ -53,7 +66,7 @@ xfs_attr3_rmt_blocks( unsigned int attrlen) { if (xfs_has_crc(mp)) { - unsigned int buflen = XFS_ATTR3_RMT_BUF_SPACE(mp, mp->m_sb.sb_blocksize); + unsigned int buflen = xfs_attr3_rmt_buf_space(mp); return (attrlen + buflen - 1) / buflen; } return XFS_B_TO_FSB(mp, attrlen); @@ -293,7 +306,7 @@ xfs_attr_rmtval_copyout( while (len > 0 && *valuelen > 0) { unsigned int hdr_size = 0; - unsigned int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize); + unsigned int byte_cnt = xfs_attr3_rmt_buf_space(mp); byte_cnt = min(*valuelen, byte_cnt); @@ -342,7 +355,7 @@ xfs_attr_rmtval_copyin( while (len > 0 && *valuelen > 0) { unsigned int hdr_size; - unsigned int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize); + unsigned int byte_cnt = xfs_attr3_rmt_buf_space(mp); byte_cnt = min(*valuelen, byte_cnt); hdr_size = xfs_attr3_rmt_hdr_set(mp, dst, ino, *offset, diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h index ebde6eb1da65..86de99e2f757 100644 --- a/fs/xfs/libxfs/xfs_da_format.h +++ b/fs/xfs/libxfs/xfs_da_format.h @@ -880,9 +880,7 @@ struct xfs_attr3_rmt_hdr { #define XFS_ATTR3_RMT_CRC_OFF offsetof(struct xfs_attr3_rmt_hdr, rm_crc) -#define XFS_ATTR3_RMT_BUF_SPACE(mp, bufsize) \ - ((bufsize) - (xfs_has_crc((mp)) ? \ - sizeof(struct xfs_attr3_rmt_hdr) : 0)) +unsigned int xfs_attr3_rmt_buf_space(struct xfs_mount *mp); /* Number of bytes in a directory block. */ static inline unsigned int xfs_dir2_dirblock_bytes(struct xfs_sb *sbp) -- cgit v1.2.3-59-g8ed1b From 204a26aa1d5a891154c9275fe4022e28793ca112 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Thu, 2 May 2024 07:48:36 -0700 Subject: xfs: create a helper to compute the blockcount of a max sized remote value Create a helper function to compute the number of fsblocks needed to store a maximally-sized extended attribute value. Signed-off-by: Darrick J. Wong Reviewed-by: Andrey Albershteyn Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr.c | 2 +- fs/xfs/libxfs/xfs_attr_remote.h | 6 ++++++ fs/xfs/scrub/reap.c | 4 ++-- 3 files changed, 9 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 1c2a27fce08a..eb274d4d636d 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -1036,7 +1036,7 @@ xfs_attr_set( break; case XFS_ATTRUPDATE_REMOVE: XFS_STATS_INC(mp, xs_attr_remove); - rmt_blks = xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); + rmt_blks = xfs_attr3_max_rmt_blocks(mp); break; } diff --git a/fs/xfs/libxfs/xfs_attr_remote.h b/fs/xfs/libxfs/xfs_attr_remote.h index c64b04f91caf..e3c6c7d774bf 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.h +++ b/fs/xfs/libxfs/xfs_attr_remote.h @@ -8,6 +8,12 @@ unsigned int xfs_attr3_rmt_blocks(struct xfs_mount *mp, unsigned int attrlen); +/* Number of rmt blocks needed to store the maximally sized attr value */ +static inline unsigned int xfs_attr3_max_rmt_blocks(struct xfs_mount *mp) +{ + return xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); +} + int xfs_attr_rmtval_get(struct xfs_da_args *args); int xfs_attr_rmtval_stale(struct xfs_inode *ip, struct xfs_bmbt_irec *map, xfs_buf_flags_t incore_flags); diff --git a/fs/xfs/scrub/reap.c b/fs/xfs/scrub/reap.c index 01ceaa4efa16..be283153c254 100644 --- a/fs/xfs/scrub/reap.c +++ b/fs/xfs/scrub/reap.c @@ -223,7 +223,7 @@ xrep_bufscan_max_sectors( int max_fsbs; /* Remote xattr values are the largest buffers that we support. */ - max_fsbs = xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); + max_fsbs = xfs_attr3_max_rmt_blocks(mp); return XFS_FSB_TO_BB(mp, min_t(xfs_extlen_t, fsblocks, max_fsbs)); } @@ -806,7 +806,7 @@ xreap_bmapi_binval( * of the next hole. */ off = imap->br_startoff + imap->br_blockcount; - max_off = off + xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); + max_off = off + xfs_attr3_max_rmt_blocks(mp); while (off < max_off) { struct xfs_bmbt_irec hmap; int nhmaps = 1; -- cgit v1.2.3-59-g8ed1b From 3791a053294b037a6bf62df03480f5c5ddfd4d1b Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Thu, 2 May 2024 07:48:37 -0700 Subject: xfs: minor cleanups of xfs_attr3_rmt_blocks Clean up the type signature of this function since we don't have negative attr lengths or block counts. Signed-off-by: Darrick J. Wong Reviewed-by: Andrey Albershteyn Reviewed-by: Christoph Hellwig --- fs/xfs/libxfs/xfs_attr_remote.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 043b837a3ef7..4c44ce1c8a64 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -56,19 +56,19 @@ xfs_attr3_rmt_buf_space( return blocksize; } -/* - * Each contiguous block has a header, so it is not just a simple attribute - * length to FSB conversion. - */ +/* Compute number of fsblocks needed to store a remote attr value */ unsigned int xfs_attr3_rmt_blocks( struct xfs_mount *mp, unsigned int attrlen) { - if (xfs_has_crc(mp)) { - unsigned int buflen = xfs_attr3_rmt_buf_space(mp); - return (attrlen + buflen - 1) / buflen; - } + /* + * Each contiguous block has a header, so it is not just a simple + * attribute length to FSB conversion. + */ + if (xfs_has_crc(mp)) + return howmany(attrlen, xfs_attr3_rmt_buf_space(mp)); + return XFS_B_TO_FSB(mp, attrlen); } -- cgit v1.2.3-59-g8ed1b From 1a3f1afb2532028c7dc5552b3aa39423c2621062 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Thu, 2 May 2024 07:48:37 -0700 Subject: xfs: widen flags argument to the xfs_iflags_* helpers xfs_inode.i_flags is an unsigned long, so make these helpers take that as the flags argument instead of unsigned short. This is needed for the next patch. While we're at it, remove the iflags variable from xfs_iget_cache_miss because we no longer need it. Signed-off-by: Darrick J. Wong Reviewed-by: Andrey Albershteyn --- fs/xfs/xfs_icache.c | 4 +--- fs/xfs/xfs_inode.h | 14 +++++++------- 2 files changed, 8 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 74f1812b03cb..0953163a2d84 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -613,7 +613,6 @@ xfs_iget_cache_miss( struct xfs_inode *ip; int error; xfs_agino_t agino = XFS_INO_TO_AGINO(mp, ino); - int iflags; ip = xfs_inode_alloc(mp, ino); if (!ip) @@ -693,13 +692,12 @@ xfs_iget_cache_miss( * memory barrier that ensures this detection works correctly at lookup * time. */ - iflags = XFS_INEW; if (flags & XFS_IGET_DONTCACHE) d_mark_dontcache(VFS_I(ip)); ip->i_udquot = NULL; ip->i_gdquot = NULL; ip->i_pdquot = NULL; - xfs_iflags_set(ip, iflags); + xfs_iflags_set(ip, XFS_INEW); /* insert the new inode */ spin_lock(&pag->pag_ici_lock); diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 9fd4d29a5713..292b90b5f2ac 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -207,13 +207,13 @@ xfs_new_eof(struct xfs_inode *ip, xfs_fsize_t new_size) * i_flags helper functions */ static inline void -__xfs_iflags_set(xfs_inode_t *ip, unsigned short flags) +__xfs_iflags_set(xfs_inode_t *ip, unsigned long flags) { ip->i_flags |= flags; } static inline void -xfs_iflags_set(xfs_inode_t *ip, unsigned short flags) +xfs_iflags_set(xfs_inode_t *ip, unsigned long flags) { spin_lock(&ip->i_flags_lock); __xfs_iflags_set(ip, flags); @@ -221,7 +221,7 @@ xfs_iflags_set(xfs_inode_t *ip, unsigned short flags) } static inline void -xfs_iflags_clear(xfs_inode_t *ip, unsigned short flags) +xfs_iflags_clear(xfs_inode_t *ip, unsigned long flags) { spin_lock(&ip->i_flags_lock); ip->i_flags &= ~flags; @@ -229,13 +229,13 @@ xfs_iflags_clear(xfs_inode_t *ip, unsigned short flags) } static inline int -__xfs_iflags_test(xfs_inode_t *ip, unsigned short flags) +__xfs_iflags_test(xfs_inode_t *ip, unsigned long flags) { return (ip->i_flags & flags); } static inline int -xfs_iflags_test(xfs_inode_t *ip, unsigned short flags) +xfs_iflags_test(xfs_inode_t *ip, unsigned long flags) { int ret; spin_lock(&ip->i_flags_lock); @@ -245,7 +245,7 @@ xfs_iflags_test(xfs_inode_t *ip, unsigned short flags) } static inline int -xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags) +xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned long flags) { int ret; @@ -258,7 +258,7 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags) } static inline int -xfs_iflags_test_and_set(xfs_inode_t *ip, unsigned short flags) +xfs_iflags_test_and_set(xfs_inode_t *ip, unsigned long flags) { int ret; -- cgit v1.2.3-59-g8ed1b From c2a09f3d782de952f09a3962d03b939e7fa7ffa4 Mon Sep 17 00:00:00 2001 From: "Ritesh Harjani (IBM)" Date: Thu, 29 Feb 2024 11:40:13 +0530 Subject: ext4: Fixes len calculation in mpage_journal_page_buffers Truncate operation can race with writeback, in which inode->i_size can get truncated and therefore size - folio_pos() can be negative. This fixes the len calculation. However this path doesn't get easily triggered even with data journaling. Cc: stable@kernel.org # v6.5 Fixes: 80be8c5cc925 ("Fixes: ext4: Make mpage_journal_page_buffers use folio") Signed-off-by: Ritesh Harjani (IBM) Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/cff4953b5c9306aba71e944ab176a5d396b9a1b7.1709182250.git.ritesh.list@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 537803250ca9..bab9223d94ac 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2334,7 +2334,7 @@ static int mpage_journal_page_buffers(handle_t *handle, if (folio_pos(folio) + len > size && !ext4_verity_in_progress(inode)) - len = size - folio_pos(folio); + len = size & (len - 1); return ext4_journal_folio_buffers(handle, folio, len); } -- cgit v1.2.3-59-g8ed1b From 53c17fe55a06cbb405b94d96759611d725d2c47a Mon Sep 17 00:00:00 2001 From: "Ritesh Harjani (IBM)" Date: Thu, 29 Feb 2024 11:40:14 +0530 Subject: ext4: Remove PAGE_MASK dependency on mpage_submit_folio This patch simply removes the PAGE_MASK dependency since mpage_submit_folio() is already converted to work with folio. Signed-off-by: Ritesh Harjani (IBM) Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/d6eadb090334ea49ceef4e643b371fabfcea328f.1709182251.git.ritesh.list@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index bab9223d94ac..e8b0773e5d2d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1865,7 +1865,7 @@ static int mpage_submit_folio(struct mpage_da_data *mpd, struct folio *folio) len = folio_size(folio); if (folio_pos(folio) + len > size && !ext4_verity_in_progress(mpd->inode)) - len = size & ~PAGE_MASK; + len = size & (len - 1); err = ext4_bio_write_folio(&mpd->io_submit, folio, len); if (!err) mpd->wbc->nr_to_write--; -- cgit v1.2.3-59-g8ed1b From a0c7cce824a54dbb83bb722df19f1ddcfa5f8d25 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 29 Feb 2024 19:54:12 +0530 Subject: ext4: set FMODE_CAN_ODIRECT instead of a dummy direct_IO method Since commit a2ad63daa88b ("VFS: add FMODE_CAN_ODIRECT file flag") file systems can just set the FMODE_CAN_ODIRECT flag at open time instead of wiring up a dummy direct_IO method to indicate support for direct I/O. Signed-off-by: Christoph Hellwig [RH: Rebased to upstream] Signed-off-by: Ritesh Harjani (IBM) Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/e5797bb597219a49043e53e4e90aa494b97dc328.1709215665.git.ritesh.list@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/file.c | 2 +- fs/ext4/inode.c | 4 ---- 2 files changed, 1 insertion(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 54d6ff22585c..965febab1d04 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -886,7 +886,7 @@ static int ext4_file_open(struct inode *inode, struct file *filp) } filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC | - FMODE_DIO_PARALLEL_WRITE; + FMODE_DIO_PARALLEL_WRITE | FMODE_CAN_ODIRECT; return dquot_file_open(inode, filp); } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index e8b0773e5d2d..989e28c24a8c 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3530,7 +3530,6 @@ static const struct address_space_operations ext4_aops = { .bmap = ext4_bmap, .invalidate_folio = ext4_invalidate_folio, .release_folio = ext4_release_folio, - .direct_IO = noop_direct_IO, .migrate_folio = buffer_migrate_folio, .is_partially_uptodate = block_is_partially_uptodate, .error_remove_folio = generic_error_remove_folio, @@ -3547,7 +3546,6 @@ static const struct address_space_operations ext4_journalled_aops = { .bmap = ext4_bmap, .invalidate_folio = ext4_journalled_invalidate_folio, .release_folio = ext4_release_folio, - .direct_IO = noop_direct_IO, .migrate_folio = buffer_migrate_folio_norefs, .is_partially_uptodate = block_is_partially_uptodate, .error_remove_folio = generic_error_remove_folio, @@ -3564,7 +3562,6 @@ static const struct address_space_operations ext4_da_aops = { .bmap = ext4_bmap, .invalidate_folio = ext4_invalidate_folio, .release_folio = ext4_release_folio, - .direct_IO = noop_direct_IO, .migrate_folio = buffer_migrate_folio, .is_partially_uptodate = block_is_partially_uptodate, .error_remove_folio = generic_error_remove_folio, @@ -3573,7 +3570,6 @@ static const struct address_space_operations ext4_da_aops = { static const struct address_space_operations ext4_dax_aops = { .writepages = ext4_dax_writepages, - .direct_IO = noop_direct_IO, .dirty_folio = noop_dirty_folio, .bmap = ext4_bmap, .swap_activate = ext4_iomap_swap_activate, -- cgit v1.2.3-59-g8ed1b From 75377ae754c93a312e8430e9c159db3273bb679c Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:23:50 +0100 Subject: gfs2: Simplify gfs2_read_super Use submit_bio_wait() instead of hand-rolling our own synchronous wait. Also allocate the BIO on the stack since we're not deep in the call stack at this point. There's no need to kmap the page, since it isn't allocated from HIGHMEM. Turn the GFP_NOFS allocation into GFP_KERNEL; if the page allocator enters reclaim, we cannot be called as the filesystem has not yet been initialised and so has no pages to reclaim. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Andreas Gruenbacher --- fs/gfs2/ops_fstype.c | 46 +++++++++++++--------------------------------- 1 file changed, 13 insertions(+), 33 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index a9f7f0d44227..10bf46867adc 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -185,22 +185,10 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent) return 0; } -static void end_bio_io_page(struct bio *bio) -{ - struct page *page = bio->bi_private; - - if (!bio->bi_status) - SetPageUptodate(page); - else - pr_warn("error %d reading superblock\n", bio->bi_status); - unlock_page(page); -} - -static void gfs2_sb_in(struct gfs2_sbd *sdp, const void *buf) +static void gfs2_sb_in(struct gfs2_sbd *sdp, const struct gfs2_sb *str) { struct gfs2_sb_host *sb = &sdp->sd_sb; struct super_block *s = sdp->sd_vfs; - const struct gfs2_sb *str = buf; sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic); sb->sb_type = be32_to_cpu(str->sb_header.mh_type); @@ -240,34 +228,26 @@ static void gfs2_sb_in(struct gfs2_sbd *sdp, const void *buf) static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent) { struct super_block *sb = sdp->sd_vfs; - struct gfs2_sb *p; struct page *page; - struct bio *bio; + struct bio_vec bvec; + struct bio bio; + int err; - page = alloc_page(GFP_NOFS); + page = alloc_page(GFP_KERNEL); if (unlikely(!page)) return -ENOMEM; - ClearPageUptodate(page); - ClearPageDirty(page); - lock_page(page); - - bio = bio_alloc(sb->s_bdev, 1, REQ_OP_READ | REQ_META, GFP_NOFS); - bio->bi_iter.bi_sector = sector * (sb->s_blocksize >> 9); - __bio_add_page(bio, page, PAGE_SIZE, 0); + bio_init(&bio, sb->s_bdev, &bvec, 1, REQ_OP_READ | REQ_META); + bio.bi_iter.bi_sector = sector * (sb->s_blocksize >> 9); + __bio_add_page(&bio, page, PAGE_SIZE, 0); - bio->bi_end_io = end_bio_io_page; - bio->bi_private = page; - submit_bio(bio); - wait_on_page_locked(page); - bio_put(bio); - if (!PageUptodate(page)) { + err = submit_bio_wait(&bio); + if (err) { + pr_warn("error %d reading superblock\n", err); __free_page(page); - return -EIO; + return err; } - p = kmap(page); - gfs2_sb_in(sdp, p); - kunmap(page); + gfs2_sb_in(sdp, page_address(page)); __free_page(page); return gfs2_check_sb(sdp, silent); } -- cgit v1.2.3-59-g8ed1b From 9a87907de3597a339cc129229d1a20bc7365ea5f Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Thu, 2 May 2024 20:35:57 +0200 Subject: ovl: implement tmpfile Combine inode creation with opening a file. There are six separate objects that are being set up: the backing inode, dentry and file, and the overlay inode, dentry and file. Cleanup in case of an error is a bit of a challenge and is difficult to test, so careful review is needed. All tmpfile testcases except generic/509 now run/pass, and no regressions are observed with full xfstests. Signed-off-by: Miklos Szeredi Reviewed-by: Amir Goldstein --- fs/backing-file.c | 23 +++++++ fs/internal.h | 3 + fs/namei.c | 6 +- fs/overlayfs/dir.c | 149 +++++++++++++++++++++++++++++++++++++------ fs/overlayfs/file.c | 3 - fs/overlayfs/overlayfs.h | 3 + include/linux/backing-file.h | 3 + 7 files changed, 165 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/backing-file.c b/fs/backing-file.c index 740185198db3..afb557446c27 100644 --- a/fs/backing-file.c +++ b/fs/backing-file.c @@ -52,6 +52,29 @@ struct file *backing_file_open(const struct path *user_path, int flags, } EXPORT_SYMBOL_GPL(backing_file_open); +struct file *backing_tmpfile_open(const struct path *user_path, int flags, + const struct path *real_parentpath, + umode_t mode, const struct cred *cred) +{ + struct mnt_idmap *real_idmap = mnt_idmap(real_parentpath->mnt); + struct file *f; + int error; + + f = alloc_empty_backing_file(flags, cred); + if (IS_ERR(f)) + return f; + + path_get(user_path); + *backing_file_user_path(f) = *user_path; + error = vfs_tmpfile(real_idmap, real_parentpath, f, mode); + if (error) { + fput(f); + f = ERR_PTR(error); + } + return f; +} +EXPORT_SYMBOL(backing_tmpfile_open); + struct backing_aio { struct kiocb iocb; refcount_t ref; diff --git a/fs/internal.h b/fs/internal.h index 7ca738904e34..ab2225136f60 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -62,6 +62,9 @@ int do_mkdirat(int dfd, struct filename *name, umode_t mode); int do_symlinkat(struct filename *from, int newdfd, struct filename *to); int do_linkat(int olddfd, struct filename *old, int newdfd, struct filename *new, int flags); +int vfs_tmpfile(struct mnt_idmap *idmap, + const struct path *parentpath, + struct file *file, umode_t mode); /* * namespace.c diff --git a/fs/namei.c b/fs/namei.c index c5b2a25be7d0..13e50b0a49d2 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -3668,9 +3668,9 @@ static int do_open(struct nameidata *nd, * On non-idmapped mounts or if permission checking is to be performed on the * raw inode simply pass @nop_mnt_idmap. */ -static int vfs_tmpfile(struct mnt_idmap *idmap, - const struct path *parentpath, - struct file *file, umode_t mode) +int vfs_tmpfile(struct mnt_idmap *idmap, + const struct path *parentpath, + struct file *file, umode_t mode) { struct dentry *child; struct inode *dir = d_inode(parentpath->dentry); diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c index 0f8b4a719237..c614e923b143 100644 --- a/fs/overlayfs/dir.c +++ b/fs/overlayfs/dir.c @@ -14,6 +14,7 @@ #include #include #include +#include #include "overlayfs.h" static unsigned short ovl_redirect_max = 256; @@ -260,14 +261,13 @@ static int ovl_set_opaque(struct dentry *dentry, struct dentry *upperdentry) * may not use to instantiate the new dentry. */ static int ovl_instantiate(struct dentry *dentry, struct inode *inode, - struct dentry *newdentry, bool hardlink) + struct dentry *newdentry, bool hardlink, struct file *tmpfile) { struct ovl_inode_params oip = { .upperdentry = newdentry, .newinode = inode, }; - ovl_dir_modified(dentry->d_parent, false); ovl_dentry_set_upper_alias(dentry); ovl_dentry_init_reval(dentry, newdentry, NULL); @@ -295,6 +295,9 @@ static int ovl_instantiate(struct dentry *dentry, struct inode *inode, inc_nlink(inode); } + if (tmpfile) + d_mark_tmpfile(tmpfile, inode); + d_instantiate(dentry, inode); if (inode != oip.newinode) { pr_warn_ratelimited("newly created inode found in cache (%pd2)\n", @@ -345,7 +348,8 @@ static int ovl_create_upper(struct dentry *dentry, struct inode *inode, ovl_set_opaque(dentry, newdentry); } - err = ovl_instantiate(dentry, inode, newdentry, !!attr->hardlink); + ovl_dir_modified(dentry->d_parent, false); + err = ovl_instantiate(dentry, inode, newdentry, !!attr->hardlink, NULL); if (err) goto out_cleanup; out_unlock: @@ -529,7 +533,8 @@ static int ovl_create_over_whiteout(struct dentry *dentry, struct inode *inode, if (err) goto out_cleanup; } - err = ovl_instantiate(dentry, inode, newdentry, hardlink); + ovl_dir_modified(dentry->d_parent, false); + err = ovl_instantiate(dentry, inode, newdentry, hardlink, NULL); if (err) { ovl_cleanup(ofs, udir, newdentry); dput(newdentry); @@ -551,12 +556,35 @@ out_cleanup: goto out_dput; } +static int ovl_setup_cred_for_create(struct dentry *dentry, struct inode *inode, + umode_t mode, const struct cred *old_cred) +{ + int err; + struct cred *override_cred; + + override_cred = prepare_creds(); + if (!override_cred) + return -ENOMEM; + + override_cred->fsuid = inode->i_uid; + override_cred->fsgid = inode->i_gid; + err = security_dentry_create_files_as(dentry, mode, &dentry->d_name, + old_cred, override_cred); + if (err) { + put_cred(override_cred); + return err; + } + put_cred(override_creds(override_cred)); + put_cred(override_cred); + + return 0; +} + static int ovl_create_or_link(struct dentry *dentry, struct inode *inode, struct ovl_cattr *attr, bool origin) { int err; const struct cred *old_cred; - struct cred *override_cred; struct dentry *parent = dentry->d_parent; old_cred = ovl_override_creds(dentry->d_sb); @@ -572,10 +600,6 @@ static int ovl_create_or_link(struct dentry *dentry, struct inode *inode, } if (!attr->hardlink) { - err = -ENOMEM; - override_cred = prepare_creds(); - if (!override_cred) - goto out_revert_creds; /* * In the creation cases(create, mkdir, mknod, symlink), * ovl should transfer current's fs{u,g}id to underlying @@ -589,17 +613,9 @@ static int ovl_create_or_link(struct dentry *dentry, struct inode *inode, * create a new inode, so just use the ovl mounter's * fs{u,g}id. */ - override_cred->fsuid = inode->i_uid; - override_cred->fsgid = inode->i_gid; - err = security_dentry_create_files_as(dentry, - attr->mode, &dentry->d_name, old_cred, - override_cred); - if (err) { - put_cred(override_cred); + err = ovl_setup_cred_for_create(dentry, inode, attr->mode, old_cred); + if (err) goto out_revert_creds; - } - put_cred(override_creds(override_cred)); - put_cred(override_cred); } if (!ovl_dentry_is_whiteout(dentry)) @@ -1290,6 +1306,100 @@ out: return err; } +static int ovl_create_tmpfile(struct file *file, struct dentry *dentry, + struct inode *inode, umode_t mode) +{ + const struct cred *old_cred; + struct path realparentpath; + struct file *realfile; + struct dentry *newdentry; + /* It's okay to set O_NOATIME, since the owner will be current fsuid */ + int flags = file->f_flags | OVL_OPEN_FLAGS; + int err; + + err = ovl_copy_up(dentry->d_parent); + if (err) + return err; + + old_cred = ovl_override_creds(dentry->d_sb); + err = ovl_setup_cred_for_create(dentry, inode, mode, old_cred); + if (err) + goto out_revert_creds; + + ovl_path_upper(dentry->d_parent, &realparentpath); + realfile = backing_tmpfile_open(&file->f_path, flags, &realparentpath, + mode, current_cred()); + err = PTR_ERR_OR_ZERO(realfile); + pr_debug("tmpfile/open(%pd2, 0%o) = %i\n", realparentpath.dentry, mode, err); + if (err) + goto out_revert_creds; + + /* ovl_instantiate() consumes the newdentry reference on success */ + newdentry = dget(realfile->f_path.dentry); + err = ovl_instantiate(dentry, inode, newdentry, false, file); + if (!err) { + file->private_data = realfile; + } else { + dput(newdentry); + fput(realfile); + } +out_revert_creds: + revert_creds(old_cred); + return err; +} + +static int ovl_dummy_open(struct inode *inode, struct file *file) +{ + return 0; +} + +static int ovl_tmpfile(struct mnt_idmap *idmap, struct inode *dir, + struct file *file, umode_t mode) +{ + int err; + struct dentry *dentry = file->f_path.dentry; + struct inode *inode; + + if (!OVL_FS(dentry->d_sb)->tmpfile) + return -EOPNOTSUPP; + + err = ovl_want_write(dentry); + if (err) + return err; + + err = -ENOMEM; + inode = ovl_new_inode(dentry->d_sb, mode, 0); + if (!inode) + goto drop_write; + + inode_init_owner(&nop_mnt_idmap, inode, dir, mode); + err = ovl_create_tmpfile(file, dentry, inode, inode->i_mode); + if (err) + goto put_inode; + + /* + * Check if the preallocated inode was actually used. Having something + * else assigned to the dentry shouldn't happen as that would indicate + * that the backing tmpfile "leaked" out of overlayfs. + */ + err = -EIO; + if (WARN_ON(inode != d_inode(dentry))) + goto put_realfile; + + /* inode reference was transferred to dentry */ + inode = NULL; + err = finish_open(file, dentry, ovl_dummy_open); +put_realfile: + /* Without FMODE_OPENED ->release() won't be called on @file */ + if (!(file->f_mode & FMODE_OPENED)) + fput(file->private_data); +put_inode: + iput(inode); +drop_write: + ovl_drop_write(dentry); + return err; +} + const struct inode_operations ovl_dir_inode_operations = { .lookup = ovl_lookup, .mkdir = ovl_mkdir, @@ -1310,4 +1420,5 @@ const struct inode_operations ovl_dir_inode_operations = { .update_time = ovl_update_time, .fileattr_get = ovl_fileattr_get, .fileattr_set = ovl_fileattr_set, + .tmpfile = ovl_tmpfile, }; diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c index 05536964d37f..1a411cae57ed 100644 --- a/fs/overlayfs/file.c +++ b/fs/overlayfs/file.c @@ -24,9 +24,6 @@ static char ovl_whatisit(struct inode *inode, struct inode *realinode) return 'm'; } -/* No atime modification on underlying */ -#define OVL_OPEN_FLAGS (O_NOATIME) - static struct file *ovl_open_realfile(const struct file *file, const struct path *realpath) { diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h index ee949f3e7c77..0bfe35da4b7b 100644 --- a/fs/overlayfs/overlayfs.h +++ b/fs/overlayfs/overlayfs.h @@ -175,6 +175,9 @@ static inline int ovl_metadata_digest_size(const struct ovl_metacopy *metacopy) return (int)metacopy->len - OVL_METACOPY_MIN_SIZE; } +/* No atime modification on underlying */ +#define OVL_OPEN_FLAGS (O_NOATIME) + extern const char *const ovl_xattr_table[][2]; static inline const char *ovl_xattr(struct ovl_fs *ofs, enum ovl_xattr ox) { diff --git a/include/linux/backing-file.h b/include/linux/backing-file.h index 3f1fe1774f1b..4b61b0e57720 100644 --- a/include/linux/backing-file.h +++ b/include/linux/backing-file.h @@ -22,6 +22,9 @@ struct backing_file_ctx { struct file *backing_file_open(const struct path *user_path, int flags, const struct path *real_path, const struct cred *cred); +struct file *backing_tmpfile_open(const struct path *user_path, int flags, + const struct path *real_parentpath, + umode_t mode, const struct cred *cred); ssize_t backing_file_read_iter(struct file *file, struct iov_iter *iter, struct kiocb *iocb, int flags, struct backing_file_ctx *ctx); -- cgit v1.2.3-59-g8ed1b From 096802748ea1dea8b476938e0a8dc16f4bd2f1ad Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Thu, 2 May 2024 20:35:57 +0200 Subject: ovl: remove upper umask handling from ovl_create_upper() This is already done by vfs_prepare_mode() when creating the upper object by vfs_create(), vfs_mkdir() and vfs_mknod(). No regressions have been observed in xfstests run with posix acls turned off for the upper filesystem. Fixes: 1639a49ccdce ("fs: move S_ISGID stripping into the vfs_*() helpers") Signed-off-by: Miklos Szeredi --- fs/overlayfs/dir.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c index c614e923b143..116f542442dd 100644 --- a/fs/overlayfs/dir.c +++ b/fs/overlayfs/dir.c @@ -330,9 +330,6 @@ static int ovl_create_upper(struct dentry *dentry, struct inode *inode, struct dentry *newdentry; int err; - if (!attr->hardlink && !IS_POSIXACL(udir)) - attr->mode &= ~current_umask(); - inode_lock_nested(udir, I_MUTEX_PARENT); newdentry = ovl_create_real(ofs, udir, ovl_lookup_upper(ofs, dentry->d_name.name, -- cgit v1.2.3-59-g8ed1b From ea7d09ad7c280122a322f408672ab8d75c1a0e30 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Sun, 17 Mar 2024 16:36:39 +0100 Subject: ext4: remove unneeded if checks before kfree kfree already checks if its argument is NULL. This fixes two Coccinelle/coccicheck warnings reported by ifnullfree.cocci. Signed-off-by: Thorsten Blum Reviewed-by: Ritesh Harjani (IBM) Link: https://lore.kernel.org/r/20240317153638.2136-2-thorsten.blum@toblux.com Signed-off-by: Theodore Ts'o --- fs/ext4/super.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 044135796f2b..28cbe0f1996f 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2078,8 +2078,7 @@ static int unnote_qf_name(struct fs_context *fc, int qtype) { struct ext4_fs_context *ctx = fc->fs_private; - if (ctx->s_qf_names[qtype]) - kfree(ctx->s_qf_names[qtype]); + kfree(ctx->s_qf_names[qtype]); ctx->s_qf_names[qtype] = NULL; ctx->qname_spec |= 1 << qtype; @@ -2484,8 +2483,7 @@ static int parse_options(struct fs_context *fc, char *options) param.size = v_len; ret = ext4_parse_param(fc, ¶m); - if (param.string) - kfree(param.string); + kfree(param.string); if (ret < 0) return ret; } -- cgit v1.2.3-59-g8ed1b From 35a1f12f0ca857fee1d7a04ef52cbd5f1f84de13 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 7 Mar 2024 12:53:20 +0100 Subject: ext4: avoid excessive credit estimate in ext4_tmpfile() A user with minimum journal size (1024 blocks these days) complained about the following error triggered by generic/697 test in ext4_tmpfile(): run fstests generic/697 at 2024-02-28 05:34:46 JBD2: vfstest wants too many credits credits:260 rsv_credits:0 max:256 EXT4-fs error (device loop0) in __ext4_new_inode:1083: error 28 Indeed the credit estimate in ext4_tmpfile() is huge. EXT4_MAXQUOTAS_INIT_BLOCKS() is 219, then 10 credits from ext4_tmpfile() itself and then ext4_xattr_credits_for_new_inode() adds more credits needed for security attributes and ACLs. Now the EXT4_MAXQUOTAS_INIT_BLOCKS() is in fact unnecessary because we've already initialized quotas with dquot_init() shortly before and so EXT4_MAXQUOTAS_TRANS_BLOCKS() is enough (which boils down to 3 credits). Fixes: af51a2ac36d1 ("ext4: ->tmpfile() support") Signed-off-by: Jan Kara Tested-by: Luis Henriques Tested-by: Disha Goel Link: https://lore.kernel.org/r/20240307115320.28949-1-jack@suse.cz Signed-off-by: Theodore Ts'o --- fs/ext4/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 5e4f65c14dfb..a630b27a4cc6 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2897,7 +2897,7 @@ retry: inode = ext4_new_inode_start_handle(idmap, dir, mode, NULL, 0, NULL, EXT4_HT_DIR, - EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb) + + EXT4_MAXQUOTAS_TRANS_BLOCKS(dir->i_sb) + 4 + EXT4_XATTR_TRANS_BLOCKS); handle = ext4_journal_current_handle(); err = PTR_ERR(inode); -- cgit v1.2.3-59-g8ed1b From fb092d407262eb4278f3d1ca24da54396a038c62 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 14 Mar 2024 23:53:02 -0400 Subject: ext4: add support for FS_IOC_GETFSSYSFSPATH The new sysfs path ioctl lets us get the /sys/fs/ path for a given filesystem in a fs agnostic way, potentially nudging us towards standarizing some of our reporting. Signed-off-by: Kent Overstreet Cc: Theodore Ts'o Cc: Andreas Dilger Cc: linux-ext4@vger.kernel.org Link: https://lore.kernel.org/r/20240315035308.3563511-4-kent.overstreet@linux.dev Signed-off-by: Theodore Ts'o --- fs/ext4/super.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 28cbe0f1996f..9bde300b1551 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -5340,6 +5340,7 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP | QTYPE_MASK_PRJ; #endif super_set_uuid(sb, es->s_uuid, sizeof(es->s_uuid)); + super_set_sysfs_name_bdev(sb); INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */ mutex_init(&sbi->s_orphan_lock); -- cgit v1.2.3-59-g8ed1b From b5357cb268c41b4e2b7383d2759fc562f5b58c33 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Sat, 20 Apr 2024 17:20:27 +0930 Subject: btrfs: qgroup: do not check qgroup inherit if qgroup is disabled [BUG] After kernel commit 86211eea8ae1 ("btrfs: qgroup: validate btrfs_qgroup_inherit parameter"), user space tool snapper will fail to create snapshot using its timeline feature. [CAUSE] It turns out that, if using timeline snapper would unconditionally pass btrfs_qgroup_inherit parameter (assigning the new snapshot to qgroup 1/0) for snapshot creation. In that case, since qgroup is disabled there would be no qgroup 1/0, and btrfs_qgroup_check_inherit() would return -ENOENT and fail the whole snapshot creation. [FIX] Just skip the check if qgroup is not enabled. This is to keep the older behavior for user space tools, as if the kernel behavior changed for user space, it is a regression of kernel. Thankfully snapper is also fixing the behavior by detecting if qgroup is running in the first place, so the effect should not be that huge. Link: https://github.com/openSUSE/snapper/issues/894 Fixes: 86211eea8ae1 ("btrfs: qgroup: validate btrfs_qgroup_inherit parameter") CC: stable@vger.kernel.org # 6.8+ Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 364acc9bbe73..40e5f7f2fcb7 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3045,6 +3045,8 @@ int btrfs_qgroup_check_inherit(struct btrfs_fs_info *fs_info, struct btrfs_qgroup_inherit *inherit, size_t size) { + if (!btrfs_qgroup_enabled(fs_info)) + return 0; if (inherit->flags & ~BTRFS_QGROUP_INHERIT_FLAGS_SUPP) return -EOPNOTSUPP; if (size < sizeof(*inherit) || size > PAGE_SIZE) -- cgit v1.2.3-59-g8ed1b From e03418abde871314e1a3a550f4c8afb7b89cb273 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 29 Apr 2024 09:03:35 -0400 Subject: btrfs: make sure that WRITTEN is set on all metadata blocks We previously would call btrfs_check_leaf() if we had the check integrity code enabled, which meant that we could only run the extended leaf checks if we had WRITTEN set on the header flags. This leaves a gap in our checking, because we could end up with corruption on disk where WRITTEN isn't set on the leaf, and then the extended leaf checks don't get run which we rely on to validate all of the item pointers to make sure we don't access memory outside of the extent buffer. However, since 732fab95abe2 ("btrfs: check-integrity: remove CONFIG_BTRFS_FS_CHECK_INTEGRITY option") we no longer call btrfs_check_leaf() from btrfs_mark_buffer_dirty(), which means we only ever call it on blocks that are being written out, and thus have WRITTEN set, or that are being read in, which should have WRITTEN set. Add checks to make sure we have WRITTEN set appropriately, and then make sure __btrfs_check_leaf() always does the item checking. This will protect us from file systems that have been corrupted and no longer have WRITTEN set on some of the blocks. This was hit on a crafted image tweaking the WRITTEN bit and reported by KASAN as out-of-bound access in the eb accessors. The example is a dir item at the end of an eb. [2.042] BTRFS warning (device loop1): bad eb member start: ptr 0x3fff start 30572544 member offset 16410 size 2 [2.040] general protection fault, probably for non-canonical address 0xe0009d1000000003: 0000 [#1] PREEMPT SMP KASAN NOPTI [2.537] KASAN: maybe wild-memory-access in range [0x0005088000000018-0x000508800000001f] [2.729] CPU: 0 PID: 2587 Comm: mount Not tainted 6.8.2 #1 [2.729] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 [2.621] RIP: 0010:btrfs_get_16+0x34b/0x6d0 [2.621] RSP: 0018:ffff88810871fab8 EFLAGS: 00000206 [2.621] RAX: 0000a11000000003 RBX: ffff888104ff8720 RCX: ffff88811b2288c0 [2.621] RDX: dffffc0000000000 RSI: ffffffff81dd8aca RDI: ffff88810871f748 [2.621] RBP: 000000000000401a R08: 0000000000000001 R09: ffffed10210e3ee9 [2.621] R10: ffff88810871f74f R11: 205d323430333737 R12: 000000000000001a [2.621] R13: 000508800000001a R14: 1ffff110210e3f5d R15: ffffffff850011e8 [2.621] FS: 00007f56ea275840(0000) GS:ffff88811b200000(0000) knlGS:0000000000000000 [2.621] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [2.621] CR2: 00007febd13b75c0 CR3: 000000010bb50000 CR4: 00000000000006f0 [2.621] Call Trace: [2.621] [2.621] ? show_regs+0x74/0x80 [2.621] ? die_addr+0x46/0xc0 [2.621] ? exc_general_protection+0x161/0x2a0 [2.621] ? asm_exc_general_protection+0x26/0x30 [2.621] ? btrfs_get_16+0x33a/0x6d0 [2.621] ? btrfs_get_16+0x34b/0x6d0 [2.621] ? btrfs_get_16+0x33a/0x6d0 [2.621] ? __pfx_btrfs_get_16+0x10/0x10 [2.621] ? __pfx_mutex_unlock+0x10/0x10 [2.621] btrfs_match_dir_item_name+0x101/0x1a0 [2.621] btrfs_lookup_dir_item+0x1f3/0x280 [2.621] ? __pfx_btrfs_lookup_dir_item+0x10/0x10 [2.621] btrfs_get_tree+0xd25/0x1910 Reported-by: lei lu CC: stable@vger.kernel.org # 6.7+ Reviewed-by: Qu Wenruo Signed-off-by: Josef Bacik Reviewed-by: David Sterba [ copy more details from report ] Signed-off-by: David Sterba --- fs/btrfs/tree-checker.c | 30 +++++++++++++++--------------- fs/btrfs/tree-checker.h | 1 + 2 files changed, 16 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index c8fbcae4e88e..32604e9b31c3 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1797,6 +1797,11 @@ enum btrfs_tree_block_status __btrfs_check_leaf(struct extent_buffer *leaf) return BTRFS_TREE_BLOCK_INVALID_LEVEL; } + if (unlikely(!btrfs_header_flag(leaf, BTRFS_HEADER_FLAG_WRITTEN))) { + generic_err(leaf, 0, "invalid flag for leaf, WRITTEN not set"); + return BTRFS_TREE_BLOCK_WRITTEN_NOT_SET; + } + /* * Extent buffers from a relocation tree have a owner field that * corresponds to the subvolume tree they are based on. So just from an @@ -1858,6 +1863,7 @@ enum btrfs_tree_block_status __btrfs_check_leaf(struct extent_buffer *leaf) for (slot = 0; slot < nritems; slot++) { u32 item_end_expected; u64 item_data_end; + enum btrfs_tree_block_status ret; btrfs_item_key_to_cpu(leaf, &key, slot); @@ -1913,21 +1919,10 @@ enum btrfs_tree_block_status __btrfs_check_leaf(struct extent_buffer *leaf) return BTRFS_TREE_BLOCK_INVALID_OFFSETS; } - /* - * We only want to do this if WRITTEN is set, otherwise the leaf - * may be in some intermediate state and won't appear valid. - */ - if (btrfs_header_flag(leaf, BTRFS_HEADER_FLAG_WRITTEN)) { - enum btrfs_tree_block_status ret; - - /* - * Check if the item size and content meet other - * criteria - */ - ret = check_leaf_item(leaf, &key, slot, &prev_key); - if (unlikely(ret != BTRFS_TREE_BLOCK_CLEAN)) - return ret; - } + /* Check if the item size and content meet other criteria. */ + ret = check_leaf_item(leaf, &key, slot, &prev_key); + if (unlikely(ret != BTRFS_TREE_BLOCK_CLEAN)) + return ret; prev_key.objectid = key.objectid; prev_key.type = key.type; @@ -1957,6 +1952,11 @@ enum btrfs_tree_block_status __btrfs_check_node(struct extent_buffer *node) int level = btrfs_header_level(node); u64 bytenr; + if (unlikely(!btrfs_header_flag(node, BTRFS_HEADER_FLAG_WRITTEN))) { + generic_err(node, 0, "invalid flag for node, WRITTEN not set"); + return BTRFS_TREE_BLOCK_WRITTEN_NOT_SET; + } + if (unlikely(level <= 0 || level >= BTRFS_MAX_LEVEL)) { generic_err(node, 0, "invalid level for node, have %d expect [1, %d]", diff --git a/fs/btrfs/tree-checker.h b/fs/btrfs/tree-checker.h index 5c809b50b2d0..01669cfa6578 100644 --- a/fs/btrfs/tree-checker.h +++ b/fs/btrfs/tree-checker.h @@ -53,6 +53,7 @@ enum btrfs_tree_block_status { BTRFS_TREE_BLOCK_INVALID_BLOCKPTR, BTRFS_TREE_BLOCK_INVALID_ITEM, BTRFS_TREE_BLOCK_INVALID_OWNER, + BTRFS_TREE_BLOCK_WRITTEN_NOT_SET, }; /* -- cgit v1.2.3-59-g8ed1b From b85c42981ac4abeeb15d16437c40f52a8a64787d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 18 Apr 2024 00:21:25 -0400 Subject: btrfs_get_bdev_and_sb(): call set_blocksize() only for exclusive opens btrfs_get_bdev_and_sb() has two callers - btrfs_open_one_device(), which asks for open to be exclusive and btrfs_get_dev_args_from_path(), which doesn't. Currently it does set_blocksize() in all cases. I'm rather dubious about the need to do set_blocksize() anywhere in btrfs, to be honest - there's some access to page cache of underlying block devices in there, but it's nowhere near the hot paths, AFAICT. In any case, btrfs_get_dev_args_from_path() only needs to read the on-disk superblock and copy several fields out of it; all callers are only interested in devices that are already opened and brought into per-filesystem set, so setting the block size is redundant for those and actively harmful if we are given a pathname of unrelated device. So we only need btrfs_get_bdev_and_sb() to call set_blocksize() when it's asked to open exclusive. Reviewed-by: Christoph Hellwig Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/btrfs/volumes.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index f15591f3e54f..43af5a9fb547 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -482,10 +482,12 @@ btrfs_get_bdev_and_sb(const char *device_path, blk_mode_t flags, void *holder, if (flush) sync_blockdev(bdev); - ret = set_blocksize(bdev, BTRFS_BDEV_BLOCKSIZE); - if (ret) { - fput(*bdev_file); - goto error; + if (holder) { + ret = set_blocksize(bdev, BTRFS_BDEV_BLOCKSIZE); + if (ret) { + fput(*bdev_file); + goto error; + } } invalidate_bdev(bdev); *disk_super = btrfs_read_dev_super(bdev); @@ -498,6 +500,7 @@ btrfs_get_bdev_and_sb(const char *device_path, blk_mode_t flags, void *holder, return 0; error: + *disk_super = NULL; *bdev_file = NULL; return ret; } -- cgit v1.2.3-59-g8ed1b From ead083aeeed9df44fab9227e47688f7305c3a233 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 18 Apr 2024 00:34:31 -0400 Subject: set_blocksize(): switch to passing struct file * Signed-off-by: Al Viro --- block/bdev.c | 11 +++++++---- block/ioctl.c | 21 ++++++++++++--------- drivers/block/pktcdvd.c | 2 +- fs/btrfs/dev-replace.c | 2 +- fs/btrfs/volumes.c | 4 ++-- fs/ext4/super.c | 2 +- fs/reiserfs/journal.c | 5 ++--- fs/xfs/xfs_buf.c | 2 +- include/linux/blkdev.h | 2 +- 9 files changed, 28 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/block/bdev.c b/block/bdev.c index b8e32d933a63..a329ff9be11d 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -144,8 +144,11 @@ static void set_init_blocksize(struct block_device *bdev) bdev->bd_inode->i_blkbits = blksize_bits(bsize); } -int set_blocksize(struct block_device *bdev, int size) +int set_blocksize(struct file *file, int size) { + struct inode *inode = file->f_mapping->host; + struct block_device *bdev = I_BDEV(inode); + /* Size must be a power of two, and between 512 and PAGE_SIZE */ if (size > PAGE_SIZE || size < 512 || !is_power_of_2(size)) return -EINVAL; @@ -155,9 +158,9 @@ int set_blocksize(struct block_device *bdev, int size) return -EINVAL; /* Don't change the size if it is same as current */ - if (bdev->bd_inode->i_blkbits != blksize_bits(size)) { + if (inode->i_blkbits != blksize_bits(size)) { sync_blockdev(bdev); - bdev->bd_inode->i_blkbits = blksize_bits(size); + inode->i_blkbits = blksize_bits(size); kill_bdev(bdev); } return 0; @@ -167,7 +170,7 @@ EXPORT_SYMBOL(set_blocksize); int sb_set_blocksize(struct super_block *sb, int size) { - if (set_blocksize(sb->s_bdev, size)) + if (set_blocksize(sb->s_bdev_file, size)) return 0; /* If we get here, we know size is power of two * and it's value is between 512 and PAGE_SIZE */ diff --git a/block/ioctl.c b/block/ioctl.c index a9028a2c2db5..1c800364bc70 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -473,11 +473,14 @@ static int compat_hdio_getgeo(struct block_device *bdev, #endif /* set the logical block size */ -static int blkdev_bszset(struct block_device *bdev, blk_mode_t mode, +static int blkdev_bszset(struct file *file, blk_mode_t mode, int __user *argp) { + // this one might be file_inode(file)->i_rdev - a rare valid + // use of file_inode() for those. + dev_t dev = I_BDEV(file->f_mapping->host)->bd_dev; + struct file *excl_file; int ret, n; - struct file *file; if (!capable(CAP_SYS_ADMIN)) return -EACCES; @@ -487,13 +490,13 @@ static int blkdev_bszset(struct block_device *bdev, blk_mode_t mode, return -EFAULT; if (mode & BLK_OPEN_EXCL) - return set_blocksize(bdev, n); + return set_blocksize(file, n); - file = bdev_file_open_by_dev(bdev->bd_dev, mode, &bdev, NULL); - if (IS_ERR(file)) + excl_file = bdev_file_open_by_dev(dev, mode, &dev, NULL); + if (IS_ERR(excl_file)) return -EBUSY; - ret = set_blocksize(bdev, n); - fput(file); + ret = set_blocksize(excl_file, n); + fput(excl_file); return ret; } @@ -621,7 +624,7 @@ long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */ return put_int(argp, block_size(bdev)); case BLKBSZSET: - return blkdev_bszset(bdev, mode, argp); + return blkdev_bszset(file, mode, argp); case BLKGETSIZE64: return put_u64(argp, bdev_nr_bytes(bdev)); @@ -681,7 +684,7 @@ long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) case BLKBSZGET_32: /* get the logical block size (cf. BLKSSZGET) */ return put_int(argp, bdev_logical_block_size(bdev)); case BLKBSZSET_32: - return blkdev_bszset(bdev, mode, argp); + return blkdev_bszset(file, mode, argp); case BLKGETSIZE64_32: return put_u64(argp, bdev_nr_bytes(bdev)); diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 05933f25b397..8a2ce8070010 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -2215,7 +2215,7 @@ static int pkt_open_dev(struct pktcdvd_device *pd, bool write) } dev_info(ddev, "%lukB available on disc\n", lba << 1); } - set_blocksize(file_bdev(bdev_file), CD_FRAMESIZE); + set_blocksize(bdev_file, CD_FRAMESIZE); return 0; diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index 7696beec4c21..7130040d92ab 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -316,7 +316,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info, set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state); device->dev_stats_valid = 1; - set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE); + set_blocksize(bdev_file, BTRFS_BDEV_BLOCKSIZE); device->fs_devices = fs_devices; ret = btrfs_get_dev_zone_info(device, false); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 43af5a9fb547..65c03ddecc59 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -483,7 +483,7 @@ btrfs_get_bdev_and_sb(const char *device_path, blk_mode_t flags, void *holder, if (flush) sync_blockdev(bdev); if (holder) { - ret = set_blocksize(bdev, BTRFS_BDEV_BLOCKSIZE); + ret = set_blocksize(*bdev_file, BTRFS_BDEV_BLOCKSIZE); if (ret) { fput(*bdev_file); goto error; @@ -2717,7 +2717,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); clear_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state); device->dev_stats_valid = 1; - set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE); + set_blocksize(device->bdev_file, BTRFS_BDEV_BLOCKSIZE); if (seeding_dev) { btrfs_clear_sb_rdonly(sb); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 044135796f2b..9988b3a40b42 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -5873,7 +5873,7 @@ static struct file *ext4_get_journal_blkdev(struct super_block *sb, sb_block = EXT4_MIN_BLOCK_SIZE / blocksize; offset = EXT4_MIN_BLOCK_SIZE % blocksize; - set_blocksize(bdev, blocksize); + set_blocksize(bdev_file, blocksize); bh = __bread(bdev, sb_block, blocksize); if (!bh) { ext4_msg(sb, KERN_ERR, "couldn't read superblock of " diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index e539ccd39e1e..e477ee0ff35d 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2626,8 +2626,7 @@ static int journal_init_dev(struct super_block *super, MAJOR(jdev), MINOR(jdev), result); return result; } else if (jdev != super->s_dev) - set_blocksize(file_bdev(journal->j_bdev_file), - super->s_blocksize); + set_blocksize(journal->j_bdev_file, super->s_blocksize); return 0; } @@ -2643,7 +2642,7 @@ static int journal_init_dev(struct super_block *super, return result; } - set_blocksize(file_bdev(journal->j_bdev_file), super->s_blocksize); + set_blocksize(journal->j_bdev_file, super->s_blocksize); reiserfs_info(super, "journal_init_dev: journal device: %pg\n", file_bdev(journal->j_bdev_file)); diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index f0fa02264eda..2dc0eacb0999 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -2043,7 +2043,7 @@ xfs_setsize_buftarg( btp->bt_meta_sectorsize = sectorsize; btp->bt_meta_sectormask = sectorsize - 1; - if (set_blocksize(btp->bt_bdev, sectorsize)) { + if (set_blocksize(btp->bt_bdev_file, sectorsize)) { xfs_warn(btp->bt_mount, "Cannot set_blocksize to %u on device %pg", sectorsize, btp->bt_bdev); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 172c91879999..20c749b2ebc2 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1474,7 +1474,7 @@ static inline void bio_end_io_acct(struct bio *bio, unsigned long start_time) } int bdev_read_only(struct block_device *bdev); -int set_blocksize(struct block_device *bdev, int size); +int set_blocksize(struct file *file, int size); int lookup_bdev(const char *pathname, dev_t *dev); -- cgit v1.2.3-59-g8ed1b From c77194965dd0dcc26f9c1671d2e74e4eb1248af5 Mon Sep 17 00:00:00 2001 From: Max Kellermann Date: Fri, 15 Mar 2024 15:29:56 +0100 Subject: Revert "ext4: apply umask if ACL support is disabled" This reverts commit 484fd6c1de13b336806a967908a927cc0356e312. The commit caused a regression because now the umask was applied to symlinks and the fix is unnecessary because the umask/O_TMPFILE bug has been fixed somewhere else already. Fixes: https://lore.kernel.org/lkml/28DSITL9912E1.2LSZUVTGTO52Q@mforney.org/ Signed-off-by: Max Kellermann Reviewed-by: Christian Brauner Tested-by: Michael Forney Link: https://lore.kernel.org/r/20240315142956.2420360-1-max.kellermann@ionos.com Signed-off-by: Theodore Ts'o --- fs/ext4/acl.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'fs') diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h index ef4c19e5f570..0c5a79c3b5d4 100644 --- a/fs/ext4/acl.h +++ b/fs/ext4/acl.h @@ -68,11 +68,6 @@ extern int ext4_init_acl(handle_t *, struct inode *, struct inode *); static inline int ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) { - /* usually, the umask is applied by posix_acl_create(), but if - ext4 ACL support is disabled at compile time, we need to do - it here, because posix_acl_create() will never be called */ - inode->i_mode &= ~current_umask(); - return 0; } #endif /* CONFIG_EXT4_FS_POSIX_ACL */ -- cgit v1.2.3-59-g8ed1b From 9e8e819f8f272c4e5dcd0bd6c7450e36481ed139 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:17 +0800 Subject: ext4: avoid overflow when setting values via sysfs When setting values of type unsigned int through sysfs, we use kstrtoul() to parse it and then truncate part of it as the final set value, when the set value is greater than UINT_MAX, the set value will not match what we see because of the truncation. As follows: $ echo 4294967296 > /sys/fs/ext4/sda/mb_max_linear_groups $ cat /sys/fs/ext4/sda/mb_max_linear_groups 0 So we use kstrtouint() to parse the attr_pointer_ui type to avoid the inconsistency described above. In addition, a judgment is added to avoid setting s_resv_clusters less than 0. Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-2-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/sysfs.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 6d332dff79dd..ca820620b974 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -104,7 +104,7 @@ static ssize_t reserved_clusters_store(struct ext4_sb_info *sbi, int ret; ret = kstrtoull(skip_spaces(buf), 0, &val); - if (ret || val >= clusters) + if (ret || val >= clusters || (s64)val < 0) return -EINVAL; atomic64_set(&sbi->s_resv_clusters, val); @@ -451,7 +451,8 @@ static ssize_t ext4_attr_store(struct kobject *kobj, s_kobj); struct ext4_attr *a = container_of(attr, struct ext4_attr, attr); void *ptr = calc_ptr(a, sbi); - unsigned long t; + unsigned int t; + unsigned long lt; int ret; switch (a->attr_id) { @@ -460,7 +461,7 @@ static ssize_t ext4_attr_store(struct kobject *kobj, case attr_pointer_ui: if (!ptr) return 0; - ret = kstrtoul(skip_spaces(buf), 0, &t); + ret = kstrtouint(skip_spaces(buf), 0, &t); if (ret) return ret; if (a->attr_ptr == ptr_ext4_super_block_offset) @@ -471,10 +472,10 @@ static ssize_t ext4_attr_store(struct kobject *kobj, case attr_pointer_ul: if (!ptr) return 0; - ret = kstrtoul(skip_spaces(buf), 0, &t); + ret = kstrtoul(skip_spaces(buf), 0, <); if (ret) return ret; - *((unsigned long *) ptr) = t; + *((unsigned long *) ptr) = lt; return len; case attr_inode_readahead: return inode_readahead_blks_store(sbi, buf, len); -- cgit v1.2.3-59-g8ed1b From f536808adcc37a546bf9cc819c349bd55a28159b Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:18 +0800 Subject: ext4: refactor out ext4_generic_attr_store() Refactor out the function ext4_generic_attr_store() to handle the setting of values of various common types, with no functional changes. Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-3-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/sysfs.c | 40 +++++++++++++++++++++++++--------------- 1 file changed, 25 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index ca820620b974..2b1c529b7fdf 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -443,24 +443,20 @@ static ssize_t ext4_attr_show(struct kobject *kobj, return 0; } -static ssize_t ext4_attr_store(struct kobject *kobj, - struct attribute *attr, - const char *buf, size_t len) +static ssize_t ext4_generic_attr_store(struct ext4_attr *a, + struct ext4_sb_info *sbi, + const char *buf, size_t len) { - struct ext4_sb_info *sbi = container_of(kobj, struct ext4_sb_info, - s_kobj); - struct ext4_attr *a = container_of(attr, struct ext4_attr, attr); - void *ptr = calc_ptr(a, sbi); + int ret; unsigned int t; unsigned long lt; - int ret; + void *ptr = calc_ptr(a, sbi); + + if (!ptr) + return 0; switch (a->attr_id) { - case attr_reserved_clusters: - return reserved_clusters_store(sbi, buf, len); case attr_pointer_ui: - if (!ptr) - return 0; ret = kstrtouint(skip_spaces(buf), 0, &t); if (ret) return ret; @@ -470,19 +466,33 @@ static ssize_t ext4_attr_store(struct kobject *kobj, *((unsigned int *) ptr) = t; return len; case attr_pointer_ul: - if (!ptr) - return 0; ret = kstrtoul(skip_spaces(buf), 0, <); if (ret) return ret; *((unsigned long *) ptr) = lt; return len; + } + return 0; +} + +static ssize_t ext4_attr_store(struct kobject *kobj, + struct attribute *attr, + const char *buf, size_t len) +{ + struct ext4_sb_info *sbi = container_of(kobj, struct ext4_sb_info, + s_kobj); + struct ext4_attr *a = container_of(attr, struct ext4_attr, attr); + + switch (a->attr_id) { + case attr_reserved_clusters: + return reserved_clusters_store(sbi, buf, len); case attr_inode_readahead: return inode_readahead_blks_store(sbi, buf, len); case attr_trigger_test_error: return trigger_test_error(sbi, buf, len); + default: + return ext4_generic_attr_store(a, sbi, buf, len); } - return 0; } static void ext4_sb_release(struct kobject *kobj) -- cgit v1.2.3-59-g8ed1b From 57341fe3179c7694c92dcf99e7f836cee4c800dd Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:19 +0800 Subject: ext4: refactor out ext4_generic_attr_show() Refactor out the function ext4_generic_attr_show() to handle the reading of values of various common types, with no functional changes. Signed-off-by: Baokun Li Reviewed-by: Zhang Yi Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-4-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/sysfs.c | 74 +++++++++++++++++++++++++-------------------------------- 1 file changed, 32 insertions(+), 42 deletions(-) (limited to 'fs') diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 2b1c529b7fdf..7f455b5f22c0 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -366,13 +366,42 @@ static ssize_t __print_tstamp(char *buf, __le32 lo, __u8 hi) #define print_tstamp(buf, es, tstamp) \ __print_tstamp(buf, (es)->tstamp, (es)->tstamp ## _hi) +static ssize_t ext4_generic_attr_show(struct ext4_attr *a, + struct ext4_sb_info *sbi, char *buf) +{ + void *ptr = calc_ptr(a, sbi); + + if (!ptr) + return 0; + + switch (a->attr_id) { + case attr_inode_readahead: + case attr_pointer_ui: + if (a->attr_ptr == ptr_ext4_super_block_offset) + return sysfs_emit(buf, "%u\n", le32_to_cpup(ptr)); + return sysfs_emit(buf, "%u\n", *((unsigned int *) ptr)); + case attr_pointer_ul: + return sysfs_emit(buf, "%lu\n", *((unsigned long *) ptr)); + case attr_pointer_u8: + return sysfs_emit(buf, "%u\n", *((unsigned char *) ptr)); + case attr_pointer_u64: + if (a->attr_ptr == ptr_ext4_super_block_offset) + return sysfs_emit(buf, "%llu\n", le64_to_cpup(ptr)); + return sysfs_emit(buf, "%llu\n", *((unsigned long long *) ptr)); + case attr_pointer_string: + return sysfs_emit(buf, "%.*s\n", a->attr_size, (char *) ptr); + case attr_pointer_atomic: + return sysfs_emit(buf, "%d\n", atomic_read((atomic_t *) ptr)); + } + return 0; +} + static ssize_t ext4_attr_show(struct kobject *kobj, struct attribute *attr, char *buf) { struct ext4_sb_info *sbi = container_of(kobj, struct ext4_sb_info, s_kobj); struct ext4_attr *a = container_of(attr, struct ext4_attr, attr); - void *ptr = calc_ptr(a, sbi); switch (a->attr_id) { case attr_delayed_allocation_blocks: @@ -391,45 +420,6 @@ static ssize_t ext4_attr_show(struct kobject *kobj, return sysfs_emit(buf, "%llu\n", (unsigned long long) percpu_counter_sum(&sbi->s_sra_exceeded_retry_limit)); - case attr_inode_readahead: - case attr_pointer_ui: - if (!ptr) - return 0; - if (a->attr_ptr == ptr_ext4_super_block_offset) - return sysfs_emit(buf, "%u\n", - le32_to_cpup(ptr)); - else - return sysfs_emit(buf, "%u\n", - *((unsigned int *) ptr)); - case attr_pointer_ul: - if (!ptr) - return 0; - return sysfs_emit(buf, "%lu\n", - *((unsigned long *) ptr)); - case attr_pointer_u8: - if (!ptr) - return 0; - return sysfs_emit(buf, "%u\n", - *((unsigned char *) ptr)); - case attr_pointer_u64: - if (!ptr) - return 0; - if (a->attr_ptr == ptr_ext4_super_block_offset) - return sysfs_emit(buf, "%llu\n", - le64_to_cpup(ptr)); - else - return sysfs_emit(buf, "%llu\n", - *((unsigned long long *) ptr)); - case attr_pointer_string: - if (!ptr) - return 0; - return sysfs_emit(buf, "%.*s\n", a->attr_size, - (char *) ptr); - case attr_pointer_atomic: - if (!ptr) - return 0; - return sysfs_emit(buf, "%d\n", - atomic_read((atomic_t *) ptr)); case attr_feature: return sysfs_emit(buf, "supported\n"); case attr_first_error_time: @@ -438,9 +428,9 @@ static ssize_t ext4_attr_show(struct kobject *kobj, return print_tstamp(buf, sbi->s_es, s_last_error_time); case attr_journal_task: return journal_task_show(sbi, buf); + default: + return ext4_generic_attr_show(a, sbi, buf); } - - return 0; } static ssize_t ext4_generic_attr_store(struct ext4_attr *a, -- cgit v1.2.3-59-g8ed1b From 13df4d44a3aaabe61cd01d277b6ee23ead2a5206 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:20 +0800 Subject: ext4: fix slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists() We can trigger a slab-out-of-bounds with the following commands: mkfs.ext4 -F /dev/$disk 10G mount /dev/$disk /tmp/test echo 2147483647 > /sys/fs/ext4/$disk/mb_group_prealloc echo test > /tmp/test/file && sync ================================================================== BUG: KASAN: slab-out-of-bounds in ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4] Read of size 8 at addr ffff888121b9d0f0 by task kworker/u2:0/11 CPU: 0 PID: 11 Comm: kworker/u2:0 Tainted: GL 6.7.0-next-20240118 #521 Call Trace: dump_stack_lvl+0x2c/0x50 kasan_report+0xb6/0xf0 ext4_mb_find_good_group_avg_frag_lists+0x8a/0x200 [ext4] ext4_mb_regular_allocator+0x19e9/0x2370 [ext4] ext4_mb_new_blocks+0x88a/0x1370 [ext4] ext4_ext_map_blocks+0x14f7/0x2390 [ext4] ext4_map_blocks+0x569/0xea0 [ext4] ext4_do_writepages+0x10f6/0x1bc0 [ext4] [...] ================================================================== The flow of issue triggering is as follows: // Set s_mb_group_prealloc to 2147483647 via sysfs ext4_mb_new_blocks ext4_mb_normalize_request ext4_mb_normalize_group_request ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc ext4_mb_regular_allocator ext4_mb_choose_next_group ext4_mb_choose_next_group_best_avail mb_avg_fragment_size_order order = fls(len) - 2 = 29 ext4_mb_find_good_group_avg_frag_lists frag_list = &sbi->s_mb_avg_fragment_size[order] if (list_empty(frag_list)) // Trigger SOOB! At 4k block size, the length of the s_mb_avg_fragment_size list is 14, but an oversized s_mb_group_prealloc is set, causing slab-out-of-bounds to be triggered by an attempt to access an element at index 29. Add a new attr_id attr_clusters_in_group with values in the range [0, sbi->s_clusters_per_group] and declare mb_group_prealloc as that type to fix the issue. In addition avoid returning an order from mb_avg_fragment_size_order() greater than MB_NUM_ORDERS(sb) and reduce some useless loops. Fixes: 7e170922f06b ("ext4: Add allocation criteria 1.5 (CR1_5)") CC: stable@vger.kernel.org Signed-off-by: Baokun Li Reviewed-by: Jan Kara Reviewed-by: Ojaswin Mujoo Link: https://lore.kernel.org/r/20240319113325.3110393-5-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 4 ++++ fs/ext4/sysfs.c | 13 ++++++++++++- 2 files changed, 16 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 12b3f196010b..dbf04f91516c 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -831,6 +831,8 @@ static int mb_avg_fragment_size_order(struct super_block *sb, ext4_grpblk_t len) return 0; if (order == MB_NUM_ORDERS(sb)) order--; + if (WARN_ON_ONCE(order > MB_NUM_ORDERS(sb))) + order = MB_NUM_ORDERS(sb) - 1; return order; } @@ -1008,6 +1010,8 @@ static void ext4_mb_choose_next_group_best_avail(struct ext4_allocation_context * goal length. */ order = fls(ac->ac_g_ex.fe_len) - 1; + if (WARN_ON_ONCE(order - 1 > MB_NUM_ORDERS(ac->ac_sb))) + order = MB_NUM_ORDERS(ac->ac_sb); min_order = order - sbi->s_mb_best_avail_max_trim_order; if (min_order < 0) min_order = 0; diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 7f455b5f22c0..ddd71673176c 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -29,6 +29,7 @@ typedef enum { attr_trigger_test_error, attr_first_error_time, attr_last_error_time, + attr_clusters_in_group, attr_feature, attr_pointer_ui, attr_pointer_ul, @@ -207,13 +208,14 @@ EXT4_ATTR_FUNC(sra_exceeded_retry_limit, 0444); EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, inode_readahead, ext4_sb_info, s_inode_readahead_blks); +EXT4_ATTR_OFFSET(mb_group_prealloc, 0644, clusters_in_group, + ext4_sb_info, s_mb_group_prealloc); EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal); EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats); EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan); EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan); EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs); EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request); -EXT4_RW_ATTR_SBI_UI(mb_group_prealloc, s_mb_group_prealloc); EXT4_RW_ATTR_SBI_UI(mb_max_linear_groups, s_mb_max_linear_groups); EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb); EXT4_ATTR(trigger_fs_error, 0200, trigger_test_error); @@ -376,6 +378,7 @@ static ssize_t ext4_generic_attr_show(struct ext4_attr *a, switch (a->attr_id) { case attr_inode_readahead: + case attr_clusters_in_group: case attr_pointer_ui: if (a->attr_ptr == ptr_ext4_super_block_offset) return sysfs_emit(buf, "%u\n", le32_to_cpup(ptr)); @@ -455,6 +458,14 @@ static ssize_t ext4_generic_attr_store(struct ext4_attr *a, else *((unsigned int *) ptr) = t; return len; + case attr_clusters_in_group: + ret = kstrtouint(skip_spaces(buf), 0, &t); + if (ret) + return ret; + if (t > sbi->s_clusters_per_group) + return -EINVAL; + *((unsigned int *) ptr) = t; + return len; case attr_pointer_ul: ret = kstrtoul(skip_spaces(buf), 0, <); if (ret) -- cgit v1.2.3-59-g8ed1b From b7b2a5799b8fafe95fcd5455c32ba2c643c86f99 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:21 +0800 Subject: ext4: add new attr pointer attr_mb_order The s_mb_best_avail_max_trim_order is of type unsigned int, and has a range of values well beyond the normal use of the mb_order. Although the mballoc code is careful enough that large numbers don't matter there, but this can mislead the sysadmin into thinking that it's normal to set such values. Hence add a new attr_id attr_mb_order with values in the range [0, 64] to avoid storing garbage values and make us more resilient to surprises in the future. Suggested-by: Jan Kara Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-6-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/sysfs.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index ddd71673176c..8e0473169458 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -30,6 +30,7 @@ typedef enum { attr_first_error_time, attr_last_error_time, attr_clusters_in_group, + attr_mb_order, attr_feature, attr_pointer_ui, attr_pointer_ul, @@ -210,6 +211,8 @@ EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, inode_readahead, ext4_sb_info, s_inode_readahead_blks); EXT4_ATTR_OFFSET(mb_group_prealloc, 0644, clusters_in_group, ext4_sb_info, s_mb_group_prealloc); +EXT4_ATTR_OFFSET(mb_best_avail_max_trim_order, 0644, mb_order, + ext4_sb_info, s_mb_best_avail_max_trim_order); EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal); EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats); EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan); @@ -225,7 +228,6 @@ EXT4_RW_ATTR_SBI_UI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.int EXT4_RW_ATTR_SBI_UI(warning_ratelimit_burst, s_warning_ratelimit_state.burst); EXT4_RW_ATTR_SBI_UI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval); EXT4_RW_ATTR_SBI_UI(msg_ratelimit_burst, s_msg_ratelimit_state.burst); -EXT4_RW_ATTR_SBI_UI(mb_best_avail_max_trim_order, s_mb_best_avail_max_trim_order); #ifdef CONFIG_EXT4_DEBUG EXT4_RW_ATTR_SBI_UL(simulate_fail, s_simulate_fail); #endif @@ -379,6 +381,7 @@ static ssize_t ext4_generic_attr_show(struct ext4_attr *a, switch (a->attr_id) { case attr_inode_readahead: case attr_clusters_in_group: + case attr_mb_order: case attr_pointer_ui: if (a->attr_ptr == ptr_ext4_super_block_offset) return sysfs_emit(buf, "%u\n", le32_to_cpup(ptr)); @@ -458,6 +461,14 @@ static ssize_t ext4_generic_attr_store(struct ext4_attr *a, else *((unsigned int *) ptr) = t; return len; + case attr_mb_order: + ret = kstrtouint(skip_spaces(buf), 0, &t); + if (ret) + return ret; + if (t > 64) + return -EINVAL; + *((unsigned int *) ptr) = t; + return len; case attr_clusters_in_group: ret = kstrtouint(skip_spaces(buf), 0, &t); if (ret) -- cgit v1.2.3-59-g8ed1b From 63bfe841053f8dda09c9d059d543486d9dc16104 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:22 +0800 Subject: ext4: add positive int attr pointer to avoid sysfs variables overflow The following variables controlled by the sysfs interface are of type int and are normally used in the range [0, INT_MAX], but are declared as attr_pointer_ui, and thus may be set to values that exceed INT_MAX and result in overflows to get negative values. err_ratelimit_burst msg_ratelimit_burst warning_ratelimit_burst err_ratelimit_interval_ms msg_ratelimit_interval_ms warning_ratelimit_interval_ms Therefore, we add attr_pointer_pi (aka positive int attr pointer) with a value range of 0-INT_MAX to avoid overflow. Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-7-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/sysfs.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 8e0473169458..ddb54608ca2e 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -32,6 +32,7 @@ typedef enum { attr_clusters_in_group, attr_mb_order, attr_feature, + attr_pointer_pi, attr_pointer_ui, attr_pointer_ul, attr_pointer_u64, @@ -180,6 +181,9 @@ static struct ext4_attr ext4_attr_##_name = { \ #define EXT4_RO_ATTR_ES_STRING(_name,_elname,_size) \ EXT4_ATTR_STRING(_name, 0444, _size, ext4_super_block, _elname) +#define EXT4_RW_ATTR_SBI_PI(_name,_elname) \ + EXT4_ATTR_OFFSET(_name, 0644, pointer_pi, ext4_sb_info, _elname) + #define EXT4_RW_ATTR_SBI_UI(_name,_elname) \ EXT4_ATTR_OFFSET(_name, 0644, pointer_ui, ext4_sb_info, _elname) @@ -222,12 +226,12 @@ EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request); EXT4_RW_ATTR_SBI_UI(mb_max_linear_groups, s_mb_max_linear_groups); EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb); EXT4_ATTR(trigger_fs_error, 0200, trigger_test_error); -EXT4_RW_ATTR_SBI_UI(err_ratelimit_interval_ms, s_err_ratelimit_state.interval); -EXT4_RW_ATTR_SBI_UI(err_ratelimit_burst, s_err_ratelimit_state.burst); -EXT4_RW_ATTR_SBI_UI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.interval); -EXT4_RW_ATTR_SBI_UI(warning_ratelimit_burst, s_warning_ratelimit_state.burst); -EXT4_RW_ATTR_SBI_UI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval); -EXT4_RW_ATTR_SBI_UI(msg_ratelimit_burst, s_msg_ratelimit_state.burst); +EXT4_RW_ATTR_SBI_PI(err_ratelimit_interval_ms, s_err_ratelimit_state.interval); +EXT4_RW_ATTR_SBI_PI(err_ratelimit_burst, s_err_ratelimit_state.burst); +EXT4_RW_ATTR_SBI_PI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.interval); +EXT4_RW_ATTR_SBI_PI(warning_ratelimit_burst, s_warning_ratelimit_state.burst); +EXT4_RW_ATTR_SBI_PI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval); +EXT4_RW_ATTR_SBI_PI(msg_ratelimit_burst, s_msg_ratelimit_state.burst); #ifdef CONFIG_EXT4_DEBUG EXT4_RW_ATTR_SBI_UL(simulate_fail, s_simulate_fail); #endif @@ -382,6 +386,7 @@ static ssize_t ext4_generic_attr_show(struct ext4_attr *a, case attr_inode_readahead: case attr_clusters_in_group: case attr_mb_order: + case attr_pointer_pi: case attr_pointer_ui: if (a->attr_ptr == ptr_ext4_super_block_offset) return sysfs_emit(buf, "%u\n", le32_to_cpup(ptr)); @@ -452,6 +457,14 @@ static ssize_t ext4_generic_attr_store(struct ext4_attr *a, return 0; switch (a->attr_id) { + case attr_pointer_pi: + ret = kstrtouint(skip_spaces(buf), 0, &t); + if (ret) + return ret; + if ((int)t < 0) + return -EINVAL; + *((unsigned int *) ptr) = t; + return len; case attr_pointer_ui: ret = kstrtouint(skip_spaces(buf), 0, &t); if (ret) -- cgit v1.2.3-59-g8ed1b From 9a9f3a9842927e4af7ca10c19c94dad83bebd713 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:23 +0800 Subject: ext4: set type of ac_groups_linear_remaining to __u32 to avoid overflow Now ac_groups_linear_remaining is of type __u16 and s_mb_max_linear_groups is of type unsigned int, so an overflow occurs when setting a value above 65535 through the mb_max_linear_groups sysfs interface. Therefore, the type of ac_groups_linear_remaining is set to __u32 to avoid overflow. Fixes: 196e402adf2e ("ext4: improve cr 0 / cr 1 group scanning") CC: stable@kernel.org Signed-off-by: Baokun Li Reviewed-by: Zhang Yi Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-8-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 56938532b4ce..7bfc5fb5a128 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -193,8 +193,8 @@ struct ext4_allocation_context { ext4_grpblk_t ac_orig_goal_len; __u32 ac_flags; /* allocation hints */ + __u32 ac_groups_linear_remaining; __u16 ac_groups_scanned; - __u16 ac_groups_linear_remaining; __u16 ac_found; __u16 ac_cX_found[EXT4_MB_NUM_CRS]; __u16 ac_tail; -- cgit v1.2.3-59-g8ed1b From 261341a932d9244cbcd372a3659428c8723e5a49 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:24 +0800 Subject: ext4: set the type of max_zeroout to unsigned int to avoid overflow The max_zeroout is of type int and the s_extent_max_zeroout_kb is of type uint, and the s_extent_max_zeroout_kb can be freely modified via the sysfs interface. When the block size is 1024, max_zeroout may overflow, so declare it as unsigned int to avoid overflow. Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-9-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/extents.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index e57054bdc5fd..e067f2dd0335 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3402,9 +3402,10 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, struct ext4_extent *ex, *abut_ex; ext4_lblk_t ee_block, eof_block; unsigned int ee_len, depth, map_len = map->m_len; - int allocated = 0, max_zeroout = 0; int err = 0; int split_flag = EXT4_EXT_DATA_VALID2; + int allocated = 0; + unsigned int max_zeroout = 0; ext_debug(inode, "logical block %llu, max_blocks %u\n", (unsigned long long)map->m_lblk, map_len); -- cgit v1.2.3-59-g8ed1b From e19089dff547c9e1f09712acc3536d7b0aa9ce3d Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 19 Mar 2024 19:33:25 +0800 Subject: ext4: clean up s_mb_rb_lock to fix build warnings with C=1 Running sparse (make C=1) on mballoc.c we get the following warning: fs/ext4/mballoc.c:3194:13: warning: context imbalance in 'ext4_mb_seq_structs_summary_start' - wrong count at exit This is because __acquires(&EXT4_SB(sb)->s_mb_rb_lock) was called in ext4_mb_seq_structs_summary_start(), but s_mb_rb_lock was removed in commit 83e80a6e3543 ("ext4: use buckets for cr 1 block scan instead of rbtree"), so remove the __acquires to silence the warning. Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240319113325.3110393-10-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index dbf04f91516c..c65fac9b8c72 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3190,7 +3190,6 @@ int ext4_seq_mb_stats_show(struct seq_file *seq, void *offset) } static void *ext4_mb_seq_structs_summary_start(struct seq_file *seq, loff_t *pos) -__acquires(&EXT4_SB(sb)->s_mb_rb_lock) { struct super_block *sb = pde_data(file_inode(seq->file)); unsigned long position; -- cgit v1.2.3-59-g8ed1b From 744a56389f7398f286231e062c2e63f0de01bcc6 Mon Sep 17 00:00:00 2001 From: Justin Stitt Date: Thu, 21 Mar 2024 01:03:10 +0000 Subject: ext4: replace deprecated strncpy with alternatives strncpy() is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. in file.c: s_last_mounted is marked as __nonstring meaning it does not need to be NUL-terminated. Let's instead use strtomem_pad() to copy bytes from the string source to the byte array destination -- while also ensuring to pad with zeroes. in ioctl.c: We can drop the memset and size argument in favor of using the new 2-argument version of strscpy_pad() -- which was introduced with Commit e6584c3964f2f ("string: Allow 2-argument strscpy()"). This guarantees NUL-termination and NUL-padding on the destination buffer -- which seems to be a requirement judging from this comment: | static int ext4_ioctl_getlabel(struct ext4_sb_info *sbi, char __user *user_label) | { | char label[EXT4_LABEL_MAX + 1]; | | /* | * EXT4_LABEL_MAX must always be smaller than FSLABEL_MAX because | * FSLABEL_MAX must include terminating null byte, while s_volume_name | * does not have to. | */ in super.c: s_first_error_func is marked as __nonstring meaning we can take the same approach as in file.c; just use strtomem_pad() Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2] Link: https://github.com/KSPP/linux/issues/90 Cc: linux-hardening@vger.kernel.org Signed-off-by: Justin Stitt Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/20240321-strncpy-fs-ext4-file-c-v1-1-36a6a09fef0c@google.com Signed-off-by: Theodore Ts'o --- fs/ext4/file.c | 3 +-- fs/ext4/ioctl.c | 3 +-- fs/ext4/super.c | 7 +++---- 3 files changed, 5 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 965febab1d04..77529c655f95 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -844,8 +844,7 @@ static int ext4_sample_last_mounted(struct super_block *sb, if (err) goto out_journal; lock_buffer(sbi->s_sbh); - strncpy(sbi->s_es->s_last_mounted, cp, - sizeof(sbi->s_es->s_last_mounted)); + strtomem_pad(sbi->s_es->s_last_mounted, cp, 0); ext4_superblock_csum_set(sb); unlock_buffer(sbi->s_sbh); ext4_handle_dirty_metadata(handle, NULL, sbi->s_sbh); diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 7160a71044c8..dab7acd49709 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -1150,9 +1150,8 @@ static int ext4_ioctl_getlabel(struct ext4_sb_info *sbi, char __user *user_label */ BUILD_BUG_ON(EXT4_LABEL_MAX >= FSLABEL_MAX); - memset(label, 0, sizeof(label)); lock_buffer(sbi->s_sbh); - strncpy(label, sbi->s_es->s_volume_name, EXT4_LABEL_MAX); + strscpy_pad(label, sbi->s_es->s_volume_name); unlock_buffer(sbi->s_sbh); if (copy_to_user(user_label, label, sizeof(label))) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9bde300b1551..f9b2b18374f3 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -6129,8 +6129,8 @@ static void ext4_update_super(struct super_block *sb) __ext4_update_tstamp(&es->s_first_error_time, &es->s_first_error_time_hi, sbi->s_first_error_time); - strncpy(es->s_first_error_func, sbi->s_first_error_func, - sizeof(es->s_first_error_func)); + strtomem_pad(es->s_first_error_func, + sbi->s_first_error_func, 0); es->s_first_error_line = cpu_to_le32(sbi->s_first_error_line); es->s_first_error_ino = @@ -6143,8 +6143,7 @@ static void ext4_update_super(struct super_block *sb) __ext4_update_tstamp(&es->s_last_error_time, &es->s_last_error_time_hi, sbi->s_last_error_time); - strncpy(es->s_last_error_func, sbi->s_last_error_func, - sizeof(es->s_last_error_func)); + strtomem_pad(es->s_last_error_func, sbi->s_last_error_func, 0); es->s_last_error_line = cpu_to_le32(sbi->s_last_error_line); es->s_last_error_ino = cpu_to_le32(sbi->s_last_error_ino); es->s_last_error_block = cpu_to_le64(sbi->s_last_error_block); -- cgit v1.2.3-59-g8ed1b From 4f3e6db3c3719952cfef89340290e0b7b03f7cbc Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 21 Mar 2024 17:26:49 +0100 Subject: Revert "ext4: drop duplicate ea_inode handling in ext4_xattr_block_set()" This reverts commit 7f48212678e91a057259b3e281701f7feb1ee397. We will need the special cleanup handling once we move allocation of EA inode outside of the buffer lock in the following patch. Signed-off-by: Jan Kara Link: https://lore.kernel.org/r/20240321162657.27420-1-jack@suse.cz Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'fs') diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index b67a176bfcf9..146690c10c73 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -2158,6 +2158,17 @@ getblk_failed: ENTRY(header(s->base)+1)); if (error) goto getblk_failed; + if (ea_inode) { + /* Drop the extra ref on ea_inode. */ + error = ext4_xattr_inode_dec_ref(handle, + ea_inode); + if (error) + ext4_warning_inode(ea_inode, + "dec ref error=%d", + error); + iput(ea_inode); + ea_inode = NULL; + } lock_buffer(new_bh); error = ext4_journal_get_create_access(handle, sb, -- cgit v1.2.3-59-g8ed1b From 0a46ef234756dca04623b7591e8ebb3440622f0b Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 21 Mar 2024 17:26:50 +0100 Subject: ext4: do not create EA inode under buffer lock ext4_xattr_set_entry() creates new EA inodes while holding buffer lock on the external xattr block. This is problematic as it nests all the allocation locking (which acquires locks on other buffers) under the buffer lock. This can even deadlock when the filesystem is corrupted and e.g. quota file is setup to contain xattr block as data block. Move the allocation of EA inode out of ext4_xattr_set_entry() into the callers. Reported-by: syzbot+a43d4f48b8397d0e41a9@syzkaller.appspotmail.com Signed-off-by: Jan Kara Link: https://lore.kernel.org/r/20240321162657.27420-2-jack@suse.cz Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 113 ++++++++++++++++++++++++++------------------------------ 1 file changed, 53 insertions(+), 60 deletions(-) (limited to 'fs') diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 146690c10c73..04f90df8dbae 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1619,6 +1619,7 @@ out_err: static int ext4_xattr_set_entry(struct ext4_xattr_info *i, struct ext4_xattr_search *s, handle_t *handle, struct inode *inode, + struct inode *new_ea_inode, bool is_block) { struct ext4_xattr_entry *last, *next; @@ -1626,7 +1627,6 @@ static int ext4_xattr_set_entry(struct ext4_xattr_info *i, size_t min_offs = s->end - s->base, name_len = strlen(i->name); int in_inode = i->in_inode; struct inode *old_ea_inode = NULL; - struct inode *new_ea_inode = NULL; size_t old_size, new_size; int ret; @@ -1711,38 +1711,11 @@ static int ext4_xattr_set_entry(struct ext4_xattr_info *i, old_ea_inode = NULL; goto out; } - } - if (i->value && in_inode) { - WARN_ON_ONCE(!i->value_len); - - new_ea_inode = ext4_xattr_inode_lookup_create(handle, inode, - i->value, i->value_len); - if (IS_ERR(new_ea_inode)) { - ret = PTR_ERR(new_ea_inode); - new_ea_inode = NULL; - goto out; - } - } - if (old_ea_inode) { /* We are ready to release ref count on the old_ea_inode. */ ret = ext4_xattr_inode_dec_ref(handle, old_ea_inode); - if (ret) { - /* Release newly required ref count on new_ea_inode. */ - if (new_ea_inode) { - int err; - - err = ext4_xattr_inode_dec_ref(handle, - new_ea_inode); - if (err) - ext4_warning_inode(new_ea_inode, - "dec ref new_ea_inode err=%d", - err); - ext4_xattr_inode_free_quota(inode, new_ea_inode, - i->value_len); - } + if (ret) goto out; - } ext4_xattr_inode_free_quota(inode, old_ea_inode, le32_to_cpu(here->e_value_size)); @@ -1866,7 +1839,6 @@ update_hash: ret = 0; out: iput(old_ea_inode); - iput(new_ea_inode); return ret; } @@ -1929,9 +1901,21 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, size_t old_ea_inode_quota = 0; unsigned int ea_ino; - #define header(x) ((struct ext4_xattr_header *)(x)) + /* If we need EA inode, prepare it before locking the buffer */ + if (i->value && i->in_inode) { + WARN_ON_ONCE(!i->value_len); + + ea_inode = ext4_xattr_inode_lookup_create(handle, inode, + i->value, i->value_len); + if (IS_ERR(ea_inode)) { + error = PTR_ERR(ea_inode); + ea_inode = NULL; + goto cleanup; + } + } + if (s->base) { int offset = (char *)s->here - bs->bh->b_data; @@ -1940,6 +1924,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, EXT4_JTR_NONE); if (error) goto cleanup; + lock_buffer(bs->bh); if (header(s->base)->h_refcount == cpu_to_le32(1)) { @@ -1966,7 +1951,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, } ea_bdebug(bs->bh, "modifying in-place"); error = ext4_xattr_set_entry(i, s, handle, inode, - true /* is_block */); + ea_inode, true /* is_block */); ext4_xattr_block_csum_set(inode, bs->bh); unlock_buffer(bs->bh); if (error == -EFSCORRUPTED) @@ -2034,29 +2019,13 @@ clone_block: s->end = s->base + sb->s_blocksize; } - error = ext4_xattr_set_entry(i, s, handle, inode, true /* is_block */); + error = ext4_xattr_set_entry(i, s, handle, inode, ea_inode, + true /* is_block */); if (error == -EFSCORRUPTED) goto bad_block; if (error) goto cleanup; - if (i->value && s->here->e_value_inum) { - /* - * A ref count on ea_inode has been taken as part of the call to - * ext4_xattr_set_entry() above. We would like to drop this - * extra ref but we have to wait until the xattr block is - * initialized and has its own ref count on the ea_inode. - */ - ea_ino = le32_to_cpu(s->here->e_value_inum); - error = ext4_xattr_inode_iget(inode, ea_ino, - le32_to_cpu(s->here->e_hash), - &ea_inode); - if (error) { - ea_inode = NULL; - goto cleanup; - } - } - inserted: if (!IS_LAST_ENTRY(s->first)) { new_bh = ext4_xattr_block_cache_find(inode, header(s->base), @@ -2209,17 +2178,16 @@ getblk_failed: cleanup: if (ea_inode) { - int error2; - - error2 = ext4_xattr_inode_dec_ref(handle, ea_inode); - if (error2) - ext4_warning_inode(ea_inode, "dec ref error=%d", - error2); + if (error) { + int error2; - /* If there was an error, revert the quota charge. */ - if (error) + error2 = ext4_xattr_inode_dec_ref(handle, ea_inode); + if (error2) + ext4_warning_inode(ea_inode, "dec ref error=%d", + error2); ext4_xattr_inode_free_quota(inode, ea_inode, i_size_read(ea_inode)); + } iput(ea_inode); } if (ce) @@ -2277,14 +2245,38 @@ int ext4_xattr_ibody_set(handle_t *handle, struct inode *inode, { struct ext4_xattr_ibody_header *header; struct ext4_xattr_search *s = &is->s; + struct inode *ea_inode = NULL; int error; if (!EXT4_INODE_HAS_XATTR_SPACE(inode)) return -ENOSPC; - error = ext4_xattr_set_entry(i, s, handle, inode, false /* is_block */); - if (error) + /* If we need EA inode, prepare it before locking the buffer */ + if (i->value && i->in_inode) { + WARN_ON_ONCE(!i->value_len); + + ea_inode = ext4_xattr_inode_lookup_create(handle, inode, + i->value, i->value_len); + if (IS_ERR(ea_inode)) + return PTR_ERR(ea_inode); + } + error = ext4_xattr_set_entry(i, s, handle, inode, ea_inode, + false /* is_block */); + if (error) { + if (ea_inode) { + int error2; + + error2 = ext4_xattr_inode_dec_ref(handle, ea_inode); + if (error2) + ext4_warning_inode(ea_inode, "dec ref error=%d", + error2); + + ext4_xattr_inode_free_quota(inode, ea_inode, + i_size_read(ea_inode)); + iput(ea_inode); + } return error; + } header = IHDR(inode, ext4_raw_inode(&is->iloc)); if (!IS_LAST_ENTRY(s->first)) { header->h_magic = cpu_to_le32(EXT4_XATTR_MAGIC); @@ -2293,6 +2285,7 @@ int ext4_xattr_ibody_set(handle_t *handle, struct inode *inode, header->h_magic = cpu_to_le32(0); ext4_clear_inode_state(inode, EXT4_STATE_XATTR); } + iput(ea_inode); return 0; } -- cgit v1.2.3-59-g8ed1b From a11adf7be9d8baefe798eab49c356ab8e3924f0e Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Sat, 23 Mar 2024 00:55:18 +0800 Subject: ext4: implement filesystem specific alloc_inode in unit test We expect inode with ext4_info_info type as following: mbt_kunit_init mbt_mb_init ext4_mb_init ext4_mb_init_backend sbi->s_buddy_cache = new_inode(sb); EXT4_I(sbi->s_buddy_cache)->i_disksize = 0; Implement alloc_inode ionde with ext4_inode_info type to avoid out-of-bounds write. Signed-off-by: Kemeng Shi Reported-by: Guenter Roeck Tested-by: Guenter Roeck Link: https://lore.kernel.org/r/20240322165518.8147-1-shikemeng@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc-test.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) (limited to 'fs') diff --git a/fs/ext4/mballoc-test.c b/fs/ext4/mballoc-test.c index 044ca5238f41..49aabcfe6b46 100644 --- a/fs/ext4/mballoc-test.c +++ b/fs/ext4/mballoc-test.c @@ -30,7 +30,31 @@ struct mbt_ext4_super_block { #define MBT_CTX(_sb) (&MBT_SB(_sb)->mbt_ctx) #define MBT_GRP_CTX(_sb, _group) (&MBT_CTX(_sb)->grp_ctx[_group]) +static struct inode *mbt_alloc_inode(struct super_block *sb) +{ + struct ext4_inode_info *ei; + + ei = kmalloc(sizeof(struct ext4_inode_info), GFP_KERNEL); + if (!ei) + return NULL; + + INIT_LIST_HEAD(&ei->i_orphan); + init_rwsem(&ei->xattr_sem); + init_rwsem(&ei->i_data_sem); + inode_init_once(&ei->vfs_inode); + ext4_fc_init_inode(&ei->vfs_inode); + + return &ei->vfs_inode; +} + +static void mbt_free_inode(struct inode *inode) +{ + kfree(EXT4_I(inode)); +} + static const struct super_operations mbt_sops = { + .alloc_inode = mbt_alloc_inode, + .free_inode = mbt_free_inode, }; static void mbt_kill_sb(struct super_block *sb) -- cgit v1.2.3-59-g8ed1b From 9c97c34a998a01bc0cf970b1685456bc2ea16b64 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 24 Apr 2024 14:19:00 +0800 Subject: ext4: keep "prefetch_grp" and "nr" consistent Keep "prefetch_grp" and "nr" consistent to avoid to call ext4_mb_prefetch_fini with non-prefetched groups. When we step into next criteria, "prefetch_grp" is set to prefetch start of new criteria while "nr" is number of the prefetched group in previous criteria. If previous criteria and next criteria are both inexpensive (< CR_GOAL_LEN_SLOW) and prefetch_ios reachs sbi->s_mb_prefetch_limit in previous criteria, "prefetch_grp" and "nr" will be inconsistent and may introduce unexpected cost to do ext4_mb_init_group for non-prefetched groups. Reset "nr" to 0 when we reset "prefetch_grp" to goal group to keep them consistent. Signed-off-by: Kemeng Shi Reviewed-by: Ojaswin Mujoo Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240424061904.987525-2-shikemeng@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index c65fac9b8c72..33e6a3ebdb55 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2860,6 +2860,7 @@ repeat: group = ac->ac_g_ex.fe_group; ac->ac_groups_linear_remaining = sbi->s_mb_max_linear_groups; prefetch_grp = group; + nr = 0; for (i = 0, new_cr = cr; i < ngroups; i++, ext4_mb_choose_next_group(ac, &new_cr, &group, ngroups)) { -- cgit v1.2.3-59-g8ed1b From d0b88624f81f272626c767f4b715f4b690fcbed2 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 24 Apr 2024 14:19:01 +0800 Subject: ext4: add test_mb_mark_used_cost to estimate cost of mb_mark_used Add test_mb_mark_used_cost to estimate cost of mb_mark_used Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240424061904.987525-3-shikemeng@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc-test.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) (limited to 'fs') diff --git a/fs/ext4/mballoc-test.c b/fs/ext4/mballoc-test.c index 49aabcfe6b46..bb2a223b207c 100644 --- a/fs/ext4/mballoc-test.c +++ b/fs/ext4/mballoc-test.c @@ -883,6 +883,56 @@ static void test_mb_free_blocks(struct kunit *test) ext4_mb_unload_buddy(&e4b); } +#define COUNT_FOR_ESTIMATE 100000 +static void test_mb_mark_used_cost(struct kunit *test) +{ + struct ext4_buddy e4b; + struct super_block *sb = (struct super_block *)test->priv; + struct ext4_free_extent ex; + int ret; + struct test_range ranges[TEST_RANGE_COUNT]; + int i, j; + unsigned long start, end, all = 0; + + /* buddy cache assumes that each page contains at least one block */ + if (sb->s_blocksize > PAGE_SIZE) + kunit_skip(test, "blocksize exceeds pagesize"); + + ret = ext4_mb_load_buddy(sb, TEST_GOAL_GROUP, &e4b); + KUNIT_ASSERT_EQ(test, ret, 0); + + ex.fe_group = TEST_GOAL_GROUP; + for (j = 0; j < COUNT_FOR_ESTIMATE; j++) { + mbt_generate_test_ranges(sb, ranges, TEST_RANGE_COUNT); + start = jiffies; + for (i = 0; i < TEST_RANGE_COUNT; i++) { + if (ranges[i].len == 0) + continue; + + ex.fe_start = ranges[i].start; + ex.fe_len = ranges[i].len; + ext4_lock_group(sb, TEST_GOAL_GROUP); + mb_mark_used(&e4b, &ex); + ext4_unlock_group(sb, TEST_GOAL_GROUP); + } + end = jiffies; + all += (end - start); + + for (i = 0; i < TEST_RANGE_COUNT; i++) { + if (ranges[i].len == 0) + continue; + + ext4_lock_group(sb, TEST_GOAL_GROUP); + mb_free_blocks(NULL, &e4b, ranges[i].start, + ranges[i].len); + ext4_unlock_group(sb, TEST_GOAL_GROUP); + } + } + + kunit_info(test, "costed jiffies %lu\n", all); + ext4_mb_unload_buddy(&e4b); +} + static const struct mbt_ext4_block_layout mbt_test_layouts[] = { { .blocksize_bits = 10, @@ -925,6 +975,8 @@ static struct kunit_case mbt_test_cases[] = { KUNIT_CASE_PARAM(test_mb_mark_used, mbt_layouts_gen_params), KUNIT_CASE_PARAM(test_mb_free_blocks, mbt_layouts_gen_params), KUNIT_CASE_PARAM(test_mark_diskspace_used, mbt_layouts_gen_params), + KUNIT_CASE_PARAM_ATTR(test_mb_mark_used_cost, mbt_layouts_gen_params, + { .speed = KUNIT_SPEED_SLOW }), {} }; -- cgit v1.2.3-59-g8ed1b From d1a3924e43a35860ed7edaeec7f901a1ade2ac37 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 24 Apr 2024 14:19:02 +0800 Subject: ext4: call ext4_mb_mark_free_simple to free continuous bits in found chunk In mb_mark_used, we will find free chunk and mark it inuse. For chunk in mid of passed range, we could simply mark whole chunk inuse. For chunk at end of range, we may need to mark a continuous bits at end of part of chunk inuse and keep rest part of chunk free. To only mark a part of chunk inuse, we firstly mark whole chunk inuse and then mark a continuous range at end of chunk free. Function mb_mark_used does several times of "mb_find_buddy; mb_clear_bit; ..." to mark a continuous range free which can be done by simply calling ext4_mb_mark_free_simple which free continuous bits in a more effective way. Just call ext4_mb_mark_free_simple in mb_mark_used to use existing and effective code to free continuous blocks in chunk at end of passed range. Signed-off-by: Kemeng Shi Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240424061904.987525-4-shikemeng@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 38 +++++++++++++++++++++----------------- 1 file changed, 21 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 33e6a3ebdb55..9a2b0bd0dcbc 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2044,13 +2044,12 @@ static int mb_mark_used(struct ext4_buddy *e4b, struct ext4_free_extent *ex) int ord; int mlen = 0; int max = 0; - int cur; int start = ex->fe_start; int len = ex->fe_len; unsigned ret = 0; int len0 = len; void *buddy; - bool split = false; + int ord_start, ord_end; BUG_ON(start + len > (e4b->bd_sb->s_blocksize << 3)); BUG_ON(e4b->bd_group != ex->fe_group); @@ -2075,16 +2074,12 @@ static int mb_mark_used(struct ext4_buddy *e4b, struct ext4_free_extent *ex) /* let's maintain buddy itself */ while (len) { - if (!split) - ord = mb_find_order_for_block(e4b, start); + ord = mb_find_order_for_block(e4b, start); if (((start >> ord) << ord) == start && len >= (1 << ord)) { /* the whole chunk may be allocated at once! */ mlen = 1 << ord; - if (!split) - buddy = mb_find_buddy(e4b, ord, &max); - else - split = false; + buddy = mb_find_buddy(e4b, ord, &max); BUG_ON((start >> ord) >= max); mb_set_bit(start >> ord, buddy); e4b->bd_info->bb_counters[ord]--; @@ -2098,20 +2093,29 @@ static int mb_mark_used(struct ext4_buddy *e4b, struct ext4_free_extent *ex) if (ret == 0) ret = len | (ord << 16); - /* we have to split large buddy */ BUG_ON(ord <= 0); buddy = mb_find_buddy(e4b, ord, &max); mb_set_bit(start >> ord, buddy); e4b->bd_info->bb_counters[ord]--; - ord--; - cur = (start >> ord) & ~1U; - buddy = mb_find_buddy(e4b, ord, &max); - mb_clear_bit(cur, buddy); - mb_clear_bit(cur + 1, buddy); - e4b->bd_info->bb_counters[ord]++; - e4b->bd_info->bb_counters[ord]++; - split = true; + ord_start = (start >> ord) << ord; + ord_end = ord_start + (1 << ord); + /* first chunk */ + if (start > ord_start) + ext4_mb_mark_free_simple(e4b->bd_sb, e4b->bd_buddy, + ord_start, start - ord_start, + e4b->bd_info); + + /* last chunk */ + if (start + len < ord_end) { + ext4_mb_mark_free_simple(e4b->bd_sb, e4b->bd_buddy, + start + len, + ord_end - (start + len), + e4b->bd_info); + break; + } + len = start + len - ord_end; + start = ord_end; } mb_set_largest_free_order(e4b->bd_sb, e4b->bd_info); -- cgit v1.2.3-59-g8ed1b From 2caffb6a277bb0f2a482a2eb824d012d5f45f4d0 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 24 Apr 2024 14:19:03 +0800 Subject: ext4: use correct criteria name instead stale integer number in comment Use correct criteria name instead stale integer number in comment Signed-off-by: Kemeng Shi Reviewed-by: Ojaswin Mujoo Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240424061904.987525-5-shikemeng@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/ext4.h | 9 ++++++--- fs/ext4/mballoc.c | 16 +++++++++------- fs/ext4/mballoc.h | 4 ++-- 3 files changed, 17 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 8d126654019e..983dad8c07ec 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -213,11 +213,14 @@ enum criteria { #define EXT4_MB_USE_RESERVED 0x2000 /* Do strict check for free blocks while retrying block allocation */ #define EXT4_MB_STRICT_CHECK 0x4000 -/* Large fragment size list lookup succeeded at least once for cr = 0 */ +/* Large fragment size list lookup succeeded at least once for + * CR_POWER2_ALIGNED */ #define EXT4_MB_CR_POWER2_ALIGNED_OPTIMIZED 0x8000 -/* Avg fragment size rb tree lookup succeeded at least once for cr = 1 */ +/* Avg fragment size rb tree lookup succeeded at least once for + * CR_GOAL_LEN_FAST */ #define EXT4_MB_CR_GOAL_LEN_FAST_OPTIMIZED 0x00010000 -/* Avg fragment size rb tree lookup succeeded at least once for cr = 1.5 */ +/* Avg fragment size rb tree lookup succeeded at least once for + * CR_BEST_AVAIL_LEN */ #define EXT4_MB_CR_BEST_AVAIL_LEN_OPTIMIZED 0x00020000 struct ext4_allocation_request { diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 9a2b0bd0dcbc..65ce2dd0cefb 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1135,8 +1135,9 @@ static void ext4_mb_choose_next_group(struct ext4_allocation_context *ac, ext4_mb_choose_next_group_best_avail(ac, new_cr, group); } else { /* - * TODO: For CR=2, we can arrange groups in an rb tree sorted by - * bb_free. But until that happens, we should never come here. + * TODO: For CR_GOAL_LEN_SLOW, we can arrange groups in an + * rb tree sorted by bb_free. But until that happens, we should + * never come here. */ WARN_ON(1); } @@ -2683,7 +2684,7 @@ static int ext4_mb_good_group_nolock(struct ext4_allocation_context *ac, int ret; /* - * cr=CR_POWER2_ALIGNED/CR_GOAL_LEN_FAST is a very optimistic + * CR_POWER2_ALIGNED/CR_GOAL_LEN_FAST is a very optimistic * search to find large good chunks almost for free. If buddy * data is not ready, then this optimization makes no sense. But * we never skip the first block group in a flex_bg, since this @@ -3448,10 +3449,11 @@ static int ext4_mb_init_backend(struct super_block *sb) } if (sbi->s_mb_prefetch > ext4_get_groups_count(sb)) sbi->s_mb_prefetch = ext4_get_groups_count(sb); - /* now many real IOs to prefetch within a single allocation at cr=0 - * given cr=0 is an CPU-related optimization we shouldn't try to - * load too many groups, at some point we should start to use what - * we've got in memory. + /* + * now many real IOs to prefetch within a single allocation at + * CR_POWER2_ALIGNED. Given CR_POWER2_ALIGNED is an CPU-related + * optimization we shouldn't try to load too many groups, at some point + * we should start to use what we've got in memory. * with an average random access time 5ms, it'd take a second to get * 200 groups (* N with flex_bg), so let's make this limit 4 */ diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 7bfc5fb5a128..7ee13d8eecd7 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -187,8 +187,8 @@ struct ext4_allocation_context { struct ext4_free_extent ac_f_ex; /* - * goal len can change in CR1.5, so save the original len. This is - * used while adjusting the PA window and for accounting. + * goal len can change in CR_BEST_AVAIL_LEN, so save the original len. + * This is used while adjusting the PA window and for accounting. */ ext4_grpblk_t ac_orig_goal_len; -- cgit v1.2.3-59-g8ed1b From da5704eef7037a5bc84a56519729d93d10a0e0a0 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 24 Apr 2024 14:19:04 +0800 Subject: ext4: open coding repeated check in next_linear_group Open coding repeated check in next_linear_group. Signed-off-by: Kemeng Shi Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240424061904.987525-6-shikemeng@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 31 +++++++++++++++---------------- 1 file changed, 15 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 65ce2dd0cefb..e89c1bfb7ce3 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1080,23 +1080,11 @@ static inline int should_optimize_scan(struct ext4_allocation_context *ac) } /* - * Return next linear group for allocation. If linear traversal should not be - * performed, this function just returns the same group + * Return next linear group for allocation. */ static ext4_group_t -next_linear_group(struct ext4_allocation_context *ac, ext4_group_t group, - ext4_group_t ngroups) +next_linear_group(ext4_group_t group, ext4_group_t ngroups) { - if (!should_optimize_scan(ac)) - goto inc_and_return; - - if (ac->ac_groups_linear_remaining) { - ac->ac_groups_linear_remaining--; - goto inc_and_return; - } - - return group; -inc_and_return: /* * Artificially restricted ngroups for non-extent * files makes group > ngroups possible on first loop. @@ -1122,8 +1110,19 @@ static void ext4_mb_choose_next_group(struct ext4_allocation_context *ac, { *new_cr = ac->ac_criteria; - if (!should_optimize_scan(ac) || ac->ac_groups_linear_remaining) { - *group = next_linear_group(ac, *group, ngroups); + if (!should_optimize_scan(ac)) { + *group = next_linear_group(*group, ngroups); + return; + } + + /* + * Optimized scanning can return non adjacent groups which can cause + * seek overhead for rotational disks. So try few linear groups before + * trying optimized scan. + */ + if (ac->ac_groups_linear_remaining) { + *group = next_linear_group(*group, ngroups); + ac->ac_groups_linear_remaining--; return; } -- cgit v1.2.3-59-g8ed1b From 45cf976008ddef4a9c9a30310c9b4fb2a9a6602a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 30 Apr 2024 06:07:55 +0200 Subject: xfs: fix log recovery buffer allocation for the legacy h_size fixup Commit a70f9fe52daa ("xfs: detect and handle invalid iclog size set by mkfs") added a fixup for incorrect h_size values used for the initial umount record in old xfsprogs versions. Later commit 0c771b99d6c9 ("xfs: clean up calculation of LR header blocks") cleaned up the log reover buffer calculation, but stoped using the fixed up h_size value to size the log recovery buffer, which can lead to an out of bounds access when the incorrect h_size does not come from the old mkfs tool, but a fuzzer. Fix this by open coding xlog_logrec_hblks and taking the fixed h_size into account for this calculation. Fixes: 0c771b99d6c9 ("xfs: clean up calculation of LR header blocks") Reported-by: Sam Sun Signed-off-by: Christoph Hellwig Reviewed-by: Brian Foster Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_log_recover.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index b445e8ce4a7d..bb8957927c3c 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2999,7 +2999,7 @@ xlog_do_recovery_pass( int error = 0, h_size, h_len; int error2 = 0; int bblks, split_bblks; - int hblks, split_hblks, wrapped_hblks; + int hblks = 1, split_hblks, wrapped_hblks; int i; struct hlist_head rhash[XLOG_RHASH_SIZE]; LIST_HEAD (buffer_list); @@ -3055,14 +3055,22 @@ xlog_do_recovery_pass( if (error) goto bread_err1; - hblks = xlog_logrec_hblks(log, rhead); - if (hblks != 1) { - kvfree(hbp); - hbp = xlog_alloc_buffer(log, hblks); + /* + * This open codes xlog_logrec_hblks so that we can reuse the + * fixed up h_size value calculated above. Without that we'd + * still allocate the buffer based on the incorrect on-disk + * size. + */ + if (h_size > XLOG_HEADER_CYCLE_SIZE && + (rhead->h_version & cpu_to_be32(XLOG_VERSION_2))) { + hblks = DIV_ROUND_UP(h_size, XLOG_HEADER_CYCLE_SIZE); + if (hblks > 1) { + kvfree(hbp); + hbp = xlog_alloc_buffer(log, hblks); + } } } else { ASSERT(log->l_sectBBsize == 1); - hblks = 1; hbp = xlog_alloc_buffer(log, 1); h_size = XLOG_BIG_RECORD_BSIZE; } -- cgit v1.2.3-59-g8ed1b From 67a841f9d7246e45dd8953dc936776132d90a048 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 30 Apr 2024 06:07:56 +0200 Subject: xfs: clean up buffer allocation in xlog_do_recovery_pass Merge the initial xlog_alloc_buffer calls, and pass the variable designating the length that is initialized to 1 above instead of passing the open coded 1 directly. Signed-off-by: Christoph Hellwig Reviewed-by: Brian Foster Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_log_recover.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index bb8957927c3c..4fe627991e86 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -3010,6 +3010,10 @@ xlog_do_recovery_pass( for (i = 0; i < XLOG_RHASH_SIZE; i++) INIT_HLIST_HEAD(&rhash[i]); + hbp = xlog_alloc_buffer(log, hblks); + if (!hbp) + return -ENOMEM; + /* * Read the header of the tail block and get the iclog buffer size from * h_size. Use this to tell how many sectors make up the log header. @@ -3020,10 +3024,6 @@ xlog_do_recovery_pass( * iclog header and extract the header size from it. Get a * new hbp that is the correct size. */ - hbp = xlog_alloc_buffer(log, 1); - if (!hbp) - return -ENOMEM; - error = xlog_bread(log, tail_blk, 1, hbp, &offset); if (error) goto bread_err1; @@ -3067,16 +3067,15 @@ xlog_do_recovery_pass( if (hblks > 1) { kvfree(hbp); hbp = xlog_alloc_buffer(log, hblks); + if (!hbp) + return -ENOMEM; } } } else { ASSERT(log->l_sectBBsize == 1); - hbp = xlog_alloc_buffer(log, 1); h_size = XLOG_BIG_RECORD_BSIZE; } - if (!hbp) - return -ENOMEM; dbp = xlog_alloc_buffer(log, BTOBB(h_size)); if (!dbp) { kvfree(hbp); -- cgit v1.2.3-59-g8ed1b From f7b9ee784511920545558db85da24d022fb2805f Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 30 Apr 2024 14:58:21 +0200 Subject: xfs: consolidate the xfs_quota_reserve_blkres definitions xfs_trans_reserve_quota_nblks is already stubbed out if quota support is disabled, no need for an extra xfs_quota_reserve_blkres stub. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_quota.h | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index 85a4ae1a17f6..621ea9d7cf06 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -123,12 +123,6 @@ extern int xfs_qm_newmount(struct xfs_mount *, uint *, uint *); extern void xfs_qm_mount_quotas(struct xfs_mount *); extern void xfs_qm_unmount(struct xfs_mount *); extern void xfs_qm_unmount_quotas(struct xfs_mount *); - -static inline int -xfs_quota_reserve_blkres(struct xfs_inode *ip, int64_t blocks) -{ - return xfs_trans_reserve_quota_nblks(NULL, ip, blocks, 0, false); -} bool xfs_inode_near_dquot_enforcement(struct xfs_inode *ip, xfs_dqtype_t type); # ifdef CONFIG_XFS_LIVE_HOOKS @@ -187,12 +181,6 @@ static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp, return 0; } -static inline int -xfs_quota_reserve_blkres(struct xfs_inode *ip, int64_t blocks) -{ - return 0; -} - static inline int xfs_trans_reserve_quota_icreate(struct xfs_trans *tp, struct xfs_dquot *udqp, struct xfs_dquot *gdqp, struct xfs_dquot *pdqp, int64_t dblocks) @@ -221,6 +209,12 @@ xfs_trans_reserve_quota_icreate(struct xfs_trans *tp, struct xfs_dquot *udqp, #endif /* CONFIG_XFS_QUOTA */ +static inline int +xfs_quota_reserve_blkres(struct xfs_inode *ip, int64_t blocks) +{ + return xfs_trans_reserve_quota_nblks(NULL, ip, blocks, 0, false); +} + static inline int xfs_quota_unreserve_blkres(struct xfs_inode *ip, int64_t blocks) { -- cgit v1.2.3-59-g8ed1b From cc3c92e7e79eb5f7f3ec4d5790ade384b7d294f7 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 30 Apr 2024 14:58:22 +0200 Subject: xfs: xfs_quota_unreserve_blkres can't fail Unreserving quotas can't fail due to quota limits, and we'll notice a shut down file system a bit later in all the callers anyway. Return void and remove the error checking and propagation in the callers. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_bmap.c | 16 +++++----------- fs/xfs/libxfs/xfs_bmap.h | 2 +- fs/xfs/xfs_aops.c | 6 +----- fs/xfs/xfs_bmap_util.c | 9 +++------ fs/xfs/xfs_bmap_util.h | 2 +- fs/xfs/xfs_iomap.c | 4 ++-- fs/xfs/xfs_quota.h | 7 ++++--- fs/xfs/xfs_reflink.c | 11 +++-------- 8 files changed, 20 insertions(+), 37 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index be515bac8ea1..0633f285875d 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4928,7 +4928,7 @@ xfs_bmap_split_indlen( *indlen2 = len2; } -int +void xfs_bmap_del_extent_delay( struct xfs_inode *ip, int whichfork, @@ -4944,7 +4944,6 @@ xfs_bmap_del_extent_delay( xfs_filblks_t got_indlen, new_indlen, stolen = 0; uint32_t state = xfs_bmap_fork_to_state(whichfork); uint64_t fdblocks; - int error = 0; bool isrt; XFS_STATS_INC(mp, xs_del_exlist); @@ -4964,9 +4963,7 @@ xfs_bmap_del_extent_delay( * sb counters as we might have to borrow some blocks for the * indirect block accounting. */ - error = xfs_quota_unreserve_blkres(ip, del->br_blockcount); - if (error) - return error; + xfs_quota_unreserve_blkres(ip, del->br_blockcount); ip->i_delayed_blks -= del->br_blockcount; if (got->br_startoff == del->br_startoff) @@ -5064,7 +5061,6 @@ xfs_bmap_del_extent_delay( xfs_add_fdblocks(mp, fdblocks); xfs_mod_delalloc(ip, -(int64_t)del->br_blockcount, -da_diff); - return error; } void @@ -5622,18 +5618,16 @@ __xfs_bunmapi( delete: if (wasdel) { - error = xfs_bmap_del_extent_delay(ip, whichfork, &icur, - &got, &del); + xfs_bmap_del_extent_delay(ip, whichfork, &icur, &got, &del); } else { error = xfs_bmap_del_extent_real(ip, tp, &icur, cur, &del, &tmp_logflags, whichfork, flags); logflags |= tmp_logflags; + if (error) + goto error0; } - if (error) - goto error0; - end = del.br_startoff - 1; nodelete: /* diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index e98849eb9bba..667b0c2b33d1 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -202,7 +202,7 @@ int xfs_bmapi_write(struct xfs_trans *tp, struct xfs_inode *ip, int xfs_bunmapi(struct xfs_trans *tp, struct xfs_inode *ip, xfs_fileoff_t bno, xfs_filblks_t len, uint32_t flags, xfs_extnum_t nexts, int *done); -int xfs_bmap_del_extent_delay(struct xfs_inode *ip, int whichfork, +void xfs_bmap_del_extent_delay(struct xfs_inode *ip, int whichfork, struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *got, struct xfs_bmbt_irec *del); void xfs_bmap_del_extent_cow(struct xfs_inode *ip, diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index f7520f375cee..6dead20338e2 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -443,7 +443,6 @@ xfs_discard_folio( { struct xfs_inode *ip = XFS_I(folio->mapping->host); struct xfs_mount *mp = ip->i_mount; - int error; if (xfs_is_shutdown(mp)) return; @@ -457,11 +456,8 @@ xfs_discard_folio( * byte of the next folio. Hence the end offset is only dependent on the * folio itself and not the start offset that is passed in. */ - error = xfs_bmap_punch_delalloc_range(ip, pos, + xfs_bmap_punch_delalloc_range(ip, pos, folio_pos(folio) + folio_size(folio)); - - if (error && !xfs_is_shutdown(mp)) - xfs_alert(mp, "page discard unable to remove delalloc mapping."); } static const struct iomap_writeback_ops xfs_writeback_ops = { diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 2e6f08198c07..14e0eb9b305f 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -440,7 +440,7 @@ out_unlock_iolock: * if the ranges only partially overlap them, so it is up to the caller to * ensure that partial blocks are not passed in. */ -int +void xfs_bmap_punch_delalloc_range( struct xfs_inode *ip, xfs_off_t start_byte, @@ -452,7 +452,6 @@ xfs_bmap_punch_delalloc_range( xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, end_byte); struct xfs_bmbt_irec got, del; struct xfs_iext_cursor icur; - int error = 0; ASSERT(!xfs_need_iread_extents(ifp)); @@ -476,15 +475,13 @@ xfs_bmap_punch_delalloc_range( continue; } - error = xfs_bmap_del_extent_delay(ip, XFS_DATA_FORK, &icur, - &got, &del); - if (error || !xfs_iext_get_extent(ifp, &icur, &got)) + xfs_bmap_del_extent_delay(ip, XFS_DATA_FORK, &icur, &got, &del); + if (!xfs_iext_get_extent(ifp, &icur, &got)) break; } out_unlock: xfs_iunlock(ip, XFS_ILOCK_EXCL); - return error; } /* diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h index 77ecbb753ef2..51f84d8ff372 100644 --- a/fs/xfs/xfs_bmap_util.h +++ b/fs/xfs/xfs_bmap_util.h @@ -30,7 +30,7 @@ xfs_bmap_rtalloc(struct xfs_bmalloca *ap) } #endif /* CONFIG_XFS_RT */ -int xfs_bmap_punch_delalloc_range(struct xfs_inode *ip, +void xfs_bmap_punch_delalloc_range(struct xfs_inode *ip, xfs_off_t start_byte, xfs_off_t end_byte); struct kgetbmap { diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index f5d0ed45721b..3db93b876de3 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -1232,8 +1232,8 @@ xfs_buffered_write_delalloc_punch( loff_t offset, loff_t length) { - return xfs_bmap_punch_delalloc_range(XFS_I(inode), offset, - offset + length); + xfs_bmap_punch_delalloc_range(XFS_I(inode), offset, offset + length); + return 0; } static int diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index 621ea9d7cf06..23d71a55bbc0 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -215,10 +215,11 @@ xfs_quota_reserve_blkres(struct xfs_inode *ip, int64_t blocks) return xfs_trans_reserve_quota_nblks(NULL, ip, blocks, 0, false); } -static inline int -xfs_quota_unreserve_blkres(struct xfs_inode *ip, int64_t blocks) +static inline void +xfs_quota_unreserve_blkres(struct xfs_inode *ip, uint64_t blocks) { - return xfs_quota_reserve_blkres(ip, -blocks); + /* don't return an error as unreserving quotas can't fail */ + xfs_quota_reserve_blkres(ip, -(int64_t)blocks); } extern int xfs_mount_reset_sbqflags(struct xfs_mount *); diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 5ecb52a234be..24761e89ce9a 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -592,10 +592,8 @@ xfs_reflink_cancel_cow_blocks( trace_xfs_reflink_cancel_cow(ip, &del); if (isnullstartblock(del.br_startblock)) { - error = xfs_bmap_del_extent_delay(ip, XFS_COW_FORK, - &icur, &got, &del); - if (error) - break; + xfs_bmap_del_extent_delay(ip, XFS_COW_FORK, &icur, &got, + &del); } else if (del.br_state == XFS_EXT_UNWRITTEN || cancel_real) { ASSERT((*tpp)->t_highest_agno == NULLAGNUMBER); @@ -618,10 +616,7 @@ xfs_reflink_cancel_cow_blocks( xfs_bmap_del_extent_cow(ip, &icur, &got, &del); /* Remove the quota reservation */ - error = xfs_quota_unreserve_blkres(ip, - del.br_blockcount); - if (error) - break; + xfs_quota_unreserve_blkres(ip, del.br_blockcount); } else { /* Didn't do anything, push cursor back. */ xfs_iext_prev(ifp, &icur); -- cgit v1.2.3-59-g8ed1b From 99fb6b7ad1f2fb83d8df2c1382be63a1f50b1ae0 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 2 May 2024 09:33:53 +0200 Subject: xfs: upgrade the extent counters in xfs_reflink_end_cow_extent later Defer the extent counter size upgrade until we know we're going to modify the extent mapping. This also defers dirtying the transaction and will allow us safely back out later in the function in later changes. Fixes: 4f86bb4b66c9 ("xfs: Conditionally upgrade existing inodes to use large extent counters") Reviewed-by: "Darrick J. Wong" Signed-off-by: Christoph Hellwig Signed-off-by: Chandan Babu R --- fs/xfs/xfs_reflink.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 24761e89ce9a..93f775c13b66 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -732,14 +732,6 @@ xfs_reflink_end_cow_extent( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, 0); - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, - XFS_IEXT_REFLINK_END_COW_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_REFLINK_END_COW_CNT); - if (error) - goto out_cancel; - /* * In case of racing, overlapping AIO writes no COW extents might be * left by the time I/O completes for the loser of the race. In that @@ -768,6 +760,14 @@ xfs_reflink_end_cow_extent( del = got; xfs_trim_extent(&del, *offset_fsb, end_fsb - *offset_fsb); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_REFLINK_END_COW_CNT); + if (error == -EFBIG) + error = xfs_iext_count_upgrade(tp, ip, + XFS_IEXT_REFLINK_END_COW_CNT); + if (error) + goto out_cancel; + /* Grab the corresponding mapping in the data fork. */ nmaps = 1; error = xfs_bmapi_read(ip, del.br_startoff, del.br_blockcount, &data, -- cgit v1.2.3-59-g8ed1b From 86de848403abda05bf9c16dcdb6bef65a8d88c41 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 2 May 2024 09:33:54 +0200 Subject: xfs: remove a racy if_bytes check in xfs_reflink_end_cow_extent Accessing if_bytes without the ilock is racy. Remove the initial if_bytes == 0 check in xfs_reflink_end_cow_extent and let ext_iext_lookup_extent fail for this case after we've taken the ilock. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Chandan Babu R --- fs/xfs/xfs_reflink.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 93f775c13b66..520f5ef3743c 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -712,12 +712,6 @@ xfs_reflink_end_cow_extent( int nmaps; int error; - /* No COW extents? That's easy! */ - if (ifp->if_bytes == 0) { - *offset_fsb = end_fsb; - return 0; - } - resblks = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK); error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, XFS_TRANS_RESERVE, &tp); -- cgit v1.2.3-59-g8ed1b From 25576c5420e61dea4c2b52942460f2221b8e46e8 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 2 May 2024 09:33:55 +0200 Subject: xfs: simplify iext overflow checking and upgrade Currently the calls to xfs_iext_count_may_overflow and xfs_iext_count_upgrade are always paired. Merge them into a single function to simplify the callers and the actual check and upgrade logic itself. Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Reviewed-by: Dave Chinner Signed-off-by: Chandan Babu R --- fs/xfs/libxfs/xfs_attr.c | 5 +--- fs/xfs/libxfs/xfs_bmap.c | 5 +--- fs/xfs/libxfs/xfs_inode_fork.c | 57 ++++++++++++++++++------------------------ fs/xfs/libxfs/xfs_inode_fork.h | 6 ++--- fs/xfs/xfs_bmap_item.c | 4 +-- fs/xfs/xfs_bmap_util.c | 23 ++++------------- fs/xfs/xfs_dquot.c | 5 +--- fs/xfs/xfs_iomap.c | 9 ++----- fs/xfs/xfs_reflink.c | 9 ++----- fs/xfs/xfs_rtalloc.c | 5 +--- 10 files changed, 41 insertions(+), 87 deletions(-) (limited to 'fs') diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index eb274d4d636d..430cd3244c14 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -1050,11 +1050,8 @@ xfs_attr_set( return error; if (op != XFS_ATTRUPDATE_REMOVE || xfs_inode_hasattr(dp)) { - error = xfs_iext_count_may_overflow(dp, XFS_ATTR_FORK, + error = xfs_iext_count_extend(args->trans, dp, XFS_ATTR_FORK, XFS_IEXT_ATTR_MANIP_CNT(rmt_blks)); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(args->trans, dp, - XFS_IEXT_ATTR_MANIP_CNT(rmt_blks)); if (error) goto out_trans_cancel; } diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 0633f285875d..3b3206d312d6 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -4651,11 +4651,8 @@ xfs_bmapi_convert_one_delalloc( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, 0); - error = xfs_iext_count_may_overflow(ip, whichfork, + error = xfs_iext_count_extend(tp, ip, whichfork, XFS_IEXT_ADD_NOSPLIT_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_ADD_NOSPLIT_CNT); if (error) goto out_trans_cancel; diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c index 7d660a973909..9d11ae015909 100644 --- a/fs/xfs/libxfs/xfs_inode_fork.c +++ b/fs/xfs/libxfs/xfs_inode_fork.c @@ -765,53 +765,46 @@ xfs_ifork_verify_local_attr( return 0; } +/* + * Check if the inode fork supports adding nr_to_add more extents. + * + * If it doesn't but we can upgrade it to large extent counters, do the upgrade. + * If we can't upgrade or are already using big counters but still can't fit the + * additional extents, return -EFBIG. + */ int -xfs_iext_count_may_overflow( +xfs_iext_count_extend( + struct xfs_trans *tp, struct xfs_inode *ip, int whichfork, - int nr_to_add) + uint nr_to_add) { + struct xfs_mount *mp = ip->i_mount; + bool has_large = + xfs_inode_has_large_extent_counts(ip); struct xfs_ifork *ifp = xfs_ifork_ptr(ip, whichfork); - uint64_t max_exts; uint64_t nr_exts; + ASSERT(nr_to_add <= XFS_MAX_EXTCNT_UPGRADE_NR); + if (whichfork == XFS_COW_FORK) return 0; - max_exts = xfs_iext_max_nextents(xfs_inode_has_large_extent_counts(ip), - whichfork); - - if (XFS_TEST_ERROR(false, ip->i_mount, XFS_ERRTAG_REDUCE_MAX_IEXTENTS)) - max_exts = 10; - + /* no point in upgrading if if_nextents overflows */ nr_exts = ifp->if_nextents + nr_to_add; - if (nr_exts < ifp->if_nextents || nr_exts > max_exts) + if (nr_exts < ifp->if_nextents) return -EFBIG; - return 0; -} - -/* - * Upgrade this inode's extent counter fields to be able to handle a potential - * increase in the extent count by nr_to_add. Normally this is the same - * quantity that caused xfs_iext_count_may_overflow() to return -EFBIG. - */ -int -xfs_iext_count_upgrade( - struct xfs_trans *tp, - struct xfs_inode *ip, - uint nr_to_add) -{ - ASSERT(nr_to_add <= XFS_MAX_EXTCNT_UPGRADE_NR); - - if (!xfs_has_large_extent_counts(ip->i_mount) || - xfs_inode_has_large_extent_counts(ip) || - XFS_TEST_ERROR(false, ip->i_mount, XFS_ERRTAG_REDUCE_MAX_IEXTENTS)) + if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_REDUCE_MAX_IEXTENTS) && + nr_exts > 10) return -EFBIG; - ip->i_diflags2 |= XFS_DIFLAG2_NREXT64; - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - + if (nr_exts > xfs_iext_max_nextents(has_large, whichfork)) { + if (has_large || !xfs_has_large_extent_counts(mp)) + return -EFBIG; + ip->i_diflags2 |= XFS_DIFLAG2_NREXT64; + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + } return 0; } diff --git a/fs/xfs/libxfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h index bd53eb951b65..2373d12fd474 100644 --- a/fs/xfs/libxfs/xfs_inode_fork.h +++ b/fs/xfs/libxfs/xfs_inode_fork.h @@ -256,10 +256,8 @@ extern void xfs_ifork_init_cow(struct xfs_inode *ip); int xfs_ifork_verify_local_data(struct xfs_inode *ip); int xfs_ifork_verify_local_attr(struct xfs_inode *ip); -int xfs_iext_count_may_overflow(struct xfs_inode *ip, int whichfork, - int nr_to_add); -int xfs_iext_count_upgrade(struct xfs_trans *tp, struct xfs_inode *ip, - uint nr_to_add); +int xfs_iext_count_extend(struct xfs_trans *tp, struct xfs_inode *ip, + int whichfork, uint nr_to_add); bool xfs_ifork_is_realtime(struct xfs_inode *ip, int whichfork); /* returns true if the fork has extents but they are not read in yet. */ diff --git a/fs/xfs/xfs_bmap_item.c b/fs/xfs/xfs_bmap_item.c index d27859a684aa..a19d62e78aa1 100644 --- a/fs/xfs/xfs_bmap_item.c +++ b/fs/xfs/xfs_bmap_item.c @@ -524,9 +524,7 @@ xfs_bmap_recover_work( else iext_delta = XFS_IEXT_PUNCH_HOLE_CNT; - error = xfs_iext_count_may_overflow(ip, work->bi_whichfork, iext_delta); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, iext_delta); + error = xfs_iext_count_extend(tp, ip, work->bi_whichfork, iext_delta); if (error) goto err_cancel; diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 14e0eb9b305f..ac2e77ebb54c 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -710,11 +710,8 @@ xfs_alloc_file_space( if (error) break; - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, XFS_IEXT_ADD_NOSPLIT_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_ADD_NOSPLIT_CNT); if (error) goto error; @@ -771,10 +768,8 @@ xfs_unmap_extent( if (error) return error; - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, XFS_IEXT_PUNCH_HOLE_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, XFS_IEXT_PUNCH_HOLE_CNT); if (error) goto out_trans_cancel; @@ -1050,10 +1045,8 @@ xfs_insert_file_space( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, 0); - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, XFS_IEXT_PUNCH_HOLE_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, XFS_IEXT_PUNCH_HOLE_CNT); if (error) goto out_trans_cancel; @@ -1279,23 +1272,17 @@ xfs_swap_extent_rmap( trace_xfs_swap_extent_rmap_remap_piece(tip, &uirec); if (xfs_bmap_is_real_extent(&uirec)) { - error = xfs_iext_count_may_overflow(ip, + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, XFS_IEXT_SWAP_RMAP_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_SWAP_RMAP_CNT); if (error) goto out; } if (xfs_bmap_is_real_extent(&irec)) { - error = xfs_iext_count_may_overflow(tip, + error = xfs_iext_count_extend(tp, tip, XFS_DATA_FORK, XFS_IEXT_SWAP_RMAP_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_SWAP_RMAP_CNT); if (error) goto out; } diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 43acb4f0d174..c1b211c260a9 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -341,11 +341,8 @@ xfs_dquot_disk_alloc( goto err_cancel; } - error = xfs_iext_count_may_overflow(quotip, XFS_DATA_FORK, + error = xfs_iext_count_extend(tp, quotip, XFS_DATA_FORK, XFS_IEXT_ADD_NOSPLIT_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, quotip, - XFS_IEXT_ADD_NOSPLIT_CNT); if (error) goto err_cancel; diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 3db93b876de3..378342673925 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -299,9 +299,7 @@ xfs_iomap_write_direct( if (error) return error; - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, nr_exts); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, nr_exts); + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, nr_exts); if (error) goto out_trans_cancel; @@ -617,11 +615,8 @@ xfs_iomap_write_unwritten( if (error) return error; - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, XFS_IEXT_WRITE_UNWRITTEN_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_WRITE_UNWRITTEN_CNT); if (error) goto error_on_bmapi_transaction; diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 520f5ef3743c..063a2e00d169 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -754,11 +754,8 @@ xfs_reflink_end_cow_extent( del = got; xfs_trim_extent(&del, *offset_fsb, end_fsb - *offset_fsb); - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, XFS_IEXT_REFLINK_END_COW_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_REFLINK_END_COW_CNT); if (error) goto out_cancel; @@ -1258,9 +1255,7 @@ xfs_reflink_remap_extent( if (dmap_written) ++iext_delta; - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, iext_delta); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, iext_delta); + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, iext_delta); if (error) goto out_cancel; diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 150f544445ca..5a7ddfed1bb8 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -695,11 +695,8 @@ xfs_growfs_rt_alloc( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + error = xfs_iext_count_extend(tp, ip, XFS_DATA_FORK, XFS_IEXT_ADD_NOSPLIT_CNT); - if (error == -EFBIG) - error = xfs_iext_count_upgrade(tp, ip, - XFS_IEXT_ADD_NOSPLIT_CNT); if (error) goto out_trans_cancel; -- cgit v1.2.3-59-g8ed1b From 559428a915e8a06e9e61a8327d0b2216116d3b14 Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 11 Apr 2024 15:53:42 +0100 Subject: ext4: remove block_device_ejected() block_device_ejected() is added by commit bdfe0cbd746a ("Revert "ext4: remove block_device_ejected"") in 2015. At that time 'bdi->wb' is destroyed synchronized from del_gendisk(), hence if ext4 is still mounted, and then mark_buffer_dirty() will reference destroyed 'wb'. However, such problem doesn't exist anymore: - commit d03f6cdc1fc4 ("block: Dynamically allocate and refcount backing_dev_info") switch bdi to use refcounting; - commit 13eec2363ef0 ("fs: Get proper reference for s_bdi"), will grab additional reference of bdi while mounting, so that 'bdi->wb' will not be destroyed until generic_shutdown_super(). Hence remove this dead function block_device_ejected(). Signed-off-by: Yu Kuai Reviewed-by: Jan Kara Reviewed-by: Christoph Hellwig Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240411145346.2516848-7-viro@zeniv.linux.org.uk Signed-off-by: Christian Brauner --- fs/ext4/super.c | 18 ------------------ 1 file changed, 18 deletions(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9988b3a40b42..b255f798f449 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -492,22 +492,6 @@ static void ext4_maybe_update_superblock(struct super_block *sb) schedule_work(&EXT4_SB(sb)->s_sb_upd_work); } -/* - * The del_gendisk() function uninitializes the disk-specific data - * structures, including the bdi structure, without telling anyone - * else. Once this happens, any attempt to call mark_buffer_dirty() - * (for example, by ext4_commit_super), will cause a kernel OOPS. - * This is a kludge to prevent these oops until we can put in a proper - * hook in del_gendisk() to inform the VFS and file system layers. - */ -static int block_device_ejected(struct super_block *sb) -{ - struct inode *bd_inode = sb->s_bdev->bd_inode; - struct backing_dev_info *bdi = inode_to_bdi(bd_inode); - - return bdi->dev == NULL; -} - static void ext4_journal_commit_callback(journal_t *journal, transaction_t *txn) { struct super_block *sb = journal->j_private; @@ -6172,8 +6156,6 @@ static int ext4_commit_super(struct super_block *sb) if (!sbh) return -EINVAL; - if (block_device_ejected(sb)) - return -ENODEV; ext4_update_super(sb); -- cgit v1.2.3-59-g8ed1b From 9baf50dfff1c55e704cae74c5f1c65f0fe254f1e Mon Sep 17 00:00:00 2001 From: Yu Kuai Date: Thu, 11 Apr 2024 15:53:45 +0100 Subject: bcachefs: remove dead function bdev_sectors() bdev_sectors() is not used hence remove it. Signed-off-by: Yu Kuai Reviewed-by: Christoph Hellwig Reviewed-by: Jan Kara Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240411145346.2516848-10-viro@zeniv.linux.org.uk Acked-by: Kent Overstreet Signed-off-by: Christian Brauner --- fs/bcachefs/util.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index 5cf885b09986..5d2c470a49ac 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -445,11 +445,6 @@ static inline unsigned fract_exp_two(unsigned x, unsigned fract_bits) void bch2_bio_map(struct bio *bio, void *base, size_t); int bch2_bio_alloc_pages(struct bio *, size_t, gfp_t); -static inline sector_t bdev_sectors(struct block_device *bdev) -{ - return bdev->bd_inode->i_size >> 9; -} - #define closure_bio_submit(bio, cl) \ do { \ closure_get(cl); \ -- cgit v1.2.3-59-g8ed1b From 224941e8379a0de8652ffec768cc8394f0b1cb95 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 11 Apr 2024 15:53:37 +0100 Subject: use ->bd_mapping instead of ->bd_inode->i_mapping Just the low-hanging fruit... Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240411145346.2516848-2-viro@zeniv.linux.org.uk Signed-off-by: Christian Brauner --- block/bdev.c | 18 +++++++++--------- block/blk-zoned.c | 4 ++-- block/genhd.c | 2 +- block/ioctl.c | 4 ++-- block/partitions/core.c | 2 +- drivers/md/bcache/super.c | 2 +- drivers/scsi/scsicam.c | 2 +- fs/btrfs/disk-io.c | 6 +++--- fs/btrfs/volumes.c | 2 +- fs/btrfs/zoned.c | 2 +- fs/buffer.c | 2 +- fs/cramfs/inode.c | 2 +- fs/erofs/data.c | 2 +- fs/ext4/dir.c | 2 +- fs/ext4/ext4_jbd2.c | 2 +- fs/ext4/super.c | 6 +++--- fs/jbd2/journal.c | 2 +- include/linux/buffer_head.h | 4 ++-- include/linux/jbd2.h | 4 ++-- 19 files changed, 35 insertions(+), 35 deletions(-) (limited to 'fs') diff --git a/block/bdev.c b/block/bdev.c index 8e19101cbbb0..00017af92a51 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -76,7 +76,7 @@ static void bdev_write_inode(struct block_device *bdev) /* Kill _all_ buffers and pagecache , dirty or not.. */ static void kill_bdev(struct block_device *bdev) { - struct address_space *mapping = bdev->bd_inode->i_mapping; + struct address_space *mapping = bdev->bd_mapping; if (mapping_empty(mapping)) return; @@ -88,7 +88,7 @@ static void kill_bdev(struct block_device *bdev) /* Invalidate clean unused buffers and pagecache. */ void invalidate_bdev(struct block_device *bdev) { - struct address_space *mapping = bdev->bd_inode->i_mapping; + struct address_space *mapping = bdev->bd_mapping; if (mapping->nrpages) { invalidate_bh_lrus(); @@ -116,7 +116,7 @@ int truncate_bdev_range(struct block_device *bdev, blk_mode_t mode, goto invalidate; } - truncate_inode_pages_range(bdev->bd_inode->i_mapping, lstart, lend); + truncate_inode_pages_range(bdev->bd_mapping, lstart, lend); if (!(mode & BLK_OPEN_EXCL)) bd_abort_claiming(bdev, truncate_bdev_range); return 0; @@ -126,7 +126,7 @@ invalidate: * Someone else has handle exclusively open. Try invalidating instead. * The 'end' argument is inclusive so the rounding is safe. */ - return invalidate_inode_pages2_range(bdev->bd_inode->i_mapping, + return invalidate_inode_pages2_range(bdev->bd_mapping, lstart >> PAGE_SHIFT, lend >> PAGE_SHIFT); } @@ -198,7 +198,7 @@ int sync_blockdev_nowait(struct block_device *bdev) { if (!bdev) return 0; - return filemap_flush(bdev->bd_inode->i_mapping); + return filemap_flush(bdev->bd_mapping); } EXPORT_SYMBOL_GPL(sync_blockdev_nowait); @@ -210,13 +210,13 @@ int sync_blockdev(struct block_device *bdev) { if (!bdev) return 0; - return filemap_write_and_wait(bdev->bd_inode->i_mapping); + return filemap_write_and_wait(bdev->bd_mapping); } EXPORT_SYMBOL(sync_blockdev); int sync_blockdev_range(struct block_device *bdev, loff_t lstart, loff_t lend) { - return filemap_write_and_wait_range(bdev->bd_inode->i_mapping, + return filemap_write_and_wait_range(bdev->bd_mapping, lstart, lend); } EXPORT_SYMBOL(sync_blockdev_range); @@ -445,7 +445,7 @@ void bdev_set_nr_sectors(struct block_device *bdev, sector_t sectors) void bdev_add(struct block_device *bdev, dev_t dev) { if (bdev_stable_writes(bdev)) - mapping_set_stable_writes(bdev->bd_inode->i_mapping); + mapping_set_stable_writes(bdev->bd_mapping); bdev->bd_dev = dev; bdev->bd_inode->i_rdev = dev; bdev->bd_inode->i_ino = dev; @@ -925,7 +925,7 @@ int bdev_open(struct block_device *bdev, blk_mode_t mode, void *holder, bdev_file->f_mode |= FMODE_NOWAIT; if (mode & BLK_OPEN_RESTRICT_WRITES) bdev_file->f_mode |= FMODE_WRITE_RESTRICTED; - bdev_file->f_mapping = bdev->bd_inode->i_mapping; + bdev_file->f_mapping = bdev->bd_mapping; bdev_file->f_wb_err = filemap_sample_wb_err(bdev_file->f_mapping); bdev_file->private_data = holder; diff --git a/block/blk-zoned.c b/block/blk-zoned.c index da0f4b2a8fa0..b008bcd4889c 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -398,7 +398,7 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, blk_mode_t mode, op = REQ_OP_ZONE_RESET; /* Invalidate the page cache, including dirty pages. */ - filemap_invalidate_lock(bdev->bd_inode->i_mapping); + filemap_invalidate_lock(bdev->bd_mapping); ret = blkdev_truncate_zone_range(bdev, mode, &zrange); if (ret) goto fail; @@ -420,7 +420,7 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, blk_mode_t mode, fail: if (cmd == BLKRESETZONE) - filemap_invalidate_unlock(bdev->bd_inode->i_mapping); + filemap_invalidate_unlock(bdev->bd_mapping); return ret; } diff --git a/block/genhd.c b/block/genhd.c index 93f5118b7d41..2bf05bd0f071 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -745,7 +745,7 @@ void invalidate_disk(struct gendisk *disk) struct block_device *bdev = disk->part0; invalidate_bdev(bdev); - bdev->bd_inode->i_mapping->wb_err = 0; + bdev->bd_mapping->wb_err = 0; set_capacity(disk, 0); } EXPORT_SYMBOL(invalidate_disk); diff --git a/block/ioctl.c b/block/ioctl.c index 1c800364bc70..7c13d8bed453 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -152,12 +152,12 @@ static int blk_ioctl_secure_erase(struct block_device *bdev, blk_mode_t mode, if (start + len > bdev_nr_bytes(bdev)) return -EINVAL; - filemap_invalidate_lock(bdev->bd_inode->i_mapping); + filemap_invalidate_lock(bdev->bd_mapping); err = truncate_bdev_range(bdev, mode, start, start + len - 1); if (!err) err = blkdev_issue_secure_erase(bdev, start >> 9, len >> 9, GFP_KERNEL); - filemap_invalidate_unlock(bdev->bd_inode->i_mapping); + filemap_invalidate_unlock(bdev->bd_mapping); return err; } diff --git a/block/partitions/core.c b/block/partitions/core.c index 2b75e325c63b..63ee317dfff0 100644 --- a/block/partitions/core.c +++ b/block/partitions/core.c @@ -704,7 +704,7 @@ EXPORT_SYMBOL_GPL(bdev_disk_changed); void *read_part_sector(struct parsed_partitions *state, sector_t n, Sector *p) { - struct address_space *mapping = state->disk->part0->bd_inode->i_mapping; + struct address_space *mapping = state->disk->part0->bd_mapping; struct folio *folio; if (n >= get_capacity(state->disk)) { diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index 0ee5e17ae2dd..277ad958cf53 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -171,7 +171,7 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev, struct page *page; unsigned int i; - page = read_cache_page_gfp(bdev->bd_inode->i_mapping, + page = read_cache_page_gfp(bdev->bd_mapping, SB_OFFSET >> PAGE_SHIFT, GFP_KERNEL); if (IS_ERR(page)) return "IO error"; diff --git a/drivers/scsi/scsicam.c b/drivers/scsi/scsicam.c index e2c7d8ef205f..dd69342bbe78 100644 --- a/drivers/scsi/scsicam.c +++ b/drivers/scsi/scsicam.c @@ -32,7 +32,7 @@ */ unsigned char *scsi_bios_ptable(struct block_device *dev) { - struct address_space *mapping = bdev_whole(dev)->bd_inode->i_mapping; + struct address_space *mapping = bdev_whole(dev)->bd_mapping; unsigned char *res = NULL; struct folio *folio; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 3df5477d48a8..f10e894b0bf5 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3651,7 +3651,7 @@ struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev, struct btrfs_super_block *super; struct page *page; u64 bytenr, bytenr_orig; - struct address_space *mapping = bdev->bd_inode->i_mapping; + struct address_space *mapping = bdev->bd_mapping; int ret; bytenr_orig = btrfs_sb_offset(copy_num); @@ -3738,7 +3738,7 @@ static int write_dev_supers(struct btrfs_device *device, struct btrfs_super_block *sb, int max_mirrors) { struct btrfs_fs_info *fs_info = device->fs_info; - struct address_space *mapping = device->bdev->bd_inode->i_mapping; + struct address_space *mapping = device->bdev->bd_mapping; SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); int i; int errors = 0; @@ -3855,7 +3855,7 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) device->commit_total_bytes) break; - page = find_get_page(device->bdev->bd_inode->i_mapping, + page = find_get_page(device->bdev->bd_mapping, bytenr >> PAGE_SHIFT); if (!page) { errors++; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 65c03ddecc59..33d357e5f9c5 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1290,7 +1290,7 @@ static struct btrfs_super_block *btrfs_read_disk_super(struct block_device *bdev return ERR_PTR(-EINVAL); /* pull in the page with our super */ - page = read_cache_page_gfp(bdev->bd_inode->i_mapping, index, GFP_KERNEL); + page = read_cache_page_gfp(bdev->bd_mapping, index, GFP_KERNEL); if (IS_ERR(page)) return ERR_CAST(page); diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 4cba80b34387..9b43fa493219 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -118,7 +118,7 @@ static int sb_write_pointer(struct block_device *bdev, struct blk_zone *zones, return -ENOENT; } else if (full[0] && full[1]) { /* Compare two super blocks */ - struct address_space *mapping = bdev->bd_inode->i_mapping; + struct address_space *mapping = bdev->bd_mapping; struct page *page[BTRFS_NR_SB_LOG_ZONES]; struct btrfs_super_block *super[BTRFS_NR_SB_LOG_ZONES]; int i; diff --git a/fs/buffer.c b/fs/buffer.c index 4f73d23c2c46..d5a0932ae68d 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1463,7 +1463,7 @@ __bread_gfp(struct block_device *bdev, sector_t block, { struct buffer_head *bh; - gfp |= mapping_gfp_constraint(bdev->bd_inode->i_mapping, ~__GFP_FS); + gfp |= mapping_gfp_constraint(bdev->bd_mapping, ~__GFP_FS); /* * Prefer looping in the allocator rather than here, at least that diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 9901057a15ba..460690ca0174 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -183,7 +183,7 @@ static int next_buffer; static void *cramfs_blkdev_read(struct super_block *sb, unsigned int offset, unsigned int len) { - struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping; + struct address_space *mapping = sb->s_bdev->bd_mapping; struct file_ra_state ra = {}; struct page *pages[BLKS_PER_BUF]; unsigned i, blocknr, buffer; diff --git a/fs/erofs/data.c b/fs/erofs/data.c index e1a170e45c70..5fc03c1e2757 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -68,7 +68,7 @@ void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb) if (erofs_is_fscache_mode(sb)) buf->mapping = EROFS_SB(sb)->s_fscache->inode->i_mapping; else - buf->mapping = sb->s_bdev->bd_inode->i_mapping; + buf->mapping = sb->s_bdev->bd_mapping; } void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb, diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 3985f8c33f95..ff4514e4626b 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -192,7 +192,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx) (PAGE_SHIFT - inode->i_blkbits); if (!ra_has_index(&file->f_ra, index)) page_cache_sync_readahead( - sb->s_bdev->bd_inode->i_mapping, + sb->s_bdev->bd_mapping, &file->f_ra, file, index, 1); file->f_ra.prev_pos = (loff_t)index << PAGE_SHIFT; diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index 5d8055161acd..da4a82456383 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -206,7 +206,7 @@ static void ext4_journal_abort_handle(const char *caller, unsigned int line, static void ext4_check_bdev_write_error(struct super_block *sb) { - struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping; + struct address_space *mapping = sb->s_bdev->bd_mapping; struct ext4_sb_info *sbi = EXT4_SB(sb); int err; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index b255f798f449..02042eb91806 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -244,7 +244,7 @@ static struct buffer_head *__ext4_sb_bread_gfp(struct super_block *sb, struct buffer_head *ext4_sb_bread(struct super_block *sb, sector_t block, blk_opf_t op_flags) { - gfp_t gfp = mapping_gfp_constraint(sb->s_bdev->bd_inode->i_mapping, + gfp_t gfp = mapping_gfp_constraint(sb->s_bdev->bd_mapping, ~__GFP_FS) | __GFP_MOVABLE; return __ext4_sb_bread_gfp(sb, block, op_flags, gfp); @@ -253,7 +253,7 @@ struct buffer_head *ext4_sb_bread(struct super_block *sb, sector_t block, struct buffer_head *ext4_sb_bread_unmovable(struct super_block *sb, sector_t block) { - gfp_t gfp = mapping_gfp_constraint(sb->s_bdev->bd_inode->i_mapping, + gfp_t gfp = mapping_gfp_constraint(sb->s_bdev->bd_mapping, ~__GFP_FS); return __ext4_sb_bread_gfp(sb, block, 0, gfp); @@ -5556,7 +5556,7 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) * used to detect the metadata async write error. */ spin_lock_init(&sbi->s_bdev_wb_lock); - errseq_check_and_advance(&sb->s_bdev->bd_inode->i_mapping->wb_err, + errseq_check_and_advance(&sb->s_bdev->bd_mapping->wb_err, &sbi->s_bdev_wb_err); EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS; ext4_orphan_cleanup(sb, es); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index b6c114c11b97..03c4b9214f56 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -2009,7 +2009,7 @@ static int __jbd2_journal_erase(journal_t *journal, unsigned int flags) byte_count = (block_stop - block_start + 1) * journal->j_blocksize; - truncate_inode_pages_range(journal->j_dev->bd_inode->i_mapping, + truncate_inode_pages_range(journal->j_dev->bd_mapping, byte_start, byte_stop); if (flags & JBD2_JOURNAL_FLUSH_DISCARD) { diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index d78454a4dd1f..e58a0d63409a 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -338,7 +338,7 @@ static inline struct buffer_head *getblk_unmovable(struct block_device *bdev, { gfp_t gfp; - gfp = mapping_gfp_constraint(bdev->bd_inode->i_mapping, ~__GFP_FS); + gfp = mapping_gfp_constraint(bdev->bd_mapping, ~__GFP_FS); gfp |= __GFP_NOFAIL; return bdev_getblk(bdev, block, size, gfp); @@ -349,7 +349,7 @@ static inline struct buffer_head *__getblk(struct block_device *bdev, { gfp_t gfp; - gfp = mapping_gfp_constraint(bdev->bd_inode->i_mapping, ~__GFP_FS); + gfp = mapping_gfp_constraint(bdev->bd_mapping, ~__GFP_FS); gfp |= __GFP_MOVABLE | __GFP_NOFAIL; return bdev_getblk(bdev, block, size, gfp); diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 971f3e826e15..ac31c37816f7 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1696,7 +1696,7 @@ static inline void jbd2_journal_abort_handle(handle_t *handle) static inline void jbd2_init_fs_dev_write_error(journal_t *journal) { - struct address_space *mapping = journal->j_fs_dev->bd_inode->i_mapping; + struct address_space *mapping = journal->j_fs_dev->bd_mapping; /* * Save the original wb_err value of client fs's bdev mapping which @@ -1707,7 +1707,7 @@ static inline void jbd2_init_fs_dev_write_error(journal_t *journal) static inline int jbd2_check_fs_dev_write_error(journal_t *journal) { - struct address_space *mapping = journal->j_fs_dev->bd_inode->i_mapping; + struct address_space *mapping = journal->j_fs_dev->bd_mapping; return errseq_check(&mapping->wb_err, READ_ONCE(journal->j_fs_dev_wb_err)); -- cgit v1.2.3-59-g8ed1b From 22f89a4f8c049b884c320bc7477397916ee97b63 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 11 Apr 2024 15:53:38 +0100 Subject: grow_dev_folio(): we only want ->bd_inode->i_mapping there Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240411145346.2516848-3-viro@zeniv.linux.org.uk Reviewed-by: Matthew Wilcox (Oracle) Signed-off-by: Christian Brauner --- fs/buffer.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index d5a0932ae68d..78a4e95ba2f2 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1034,12 +1034,12 @@ static sector_t folio_init_buffers(struct folio *folio, static bool grow_dev_folio(struct block_device *bdev, sector_t block, pgoff_t index, unsigned size, gfp_t gfp) { - struct inode *inode = bdev->bd_inode; + struct address_space *mapping = bdev->bd_mapping; struct folio *folio; struct buffer_head *bh; sector_t end_block = 0; - folio = __filemap_get_folio(inode->i_mapping, index, + folio = __filemap_get_folio(mapping, index, FGP_LOCK | FGP_ACCESSED | FGP_CREAT, gfp); if (IS_ERR(folio)) return false; @@ -1073,10 +1073,10 @@ static bool grow_dev_folio(struct block_device *bdev, sector_t block, * lock to be atomic wrt __find_get_block(), which does not * run under the folio lock. */ - spin_lock(&inode->i_mapping->i_private_lock); + spin_lock(&mapping->i_private_lock); link_dev_buffers(folio, bh); end_block = folio_init_buffers(folio, bdev, size); - spin_unlock(&inode->i_mapping->i_private_lock); + spin_unlock(&mapping->i_private_lock); unlock: folio_unlock(folio); folio_put(folio); -- cgit v1.2.3-59-g8ed1b From 53cd4cd3b12d8a62719bef950f298a2bc5a6b415 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 28 Apr 2024 19:41:13 -0400 Subject: fs/buffer.c: massage the remaining users of ->bd_inode to ->bd_mapping both for ->i_blkbits and both want the address_space in question anyway. Signed-off-by: Al Viro --- fs/buffer.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 78a4e95ba2f2..ac29e0f221bc 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -189,8 +189,8 @@ EXPORT_SYMBOL(end_buffer_write_sync); static struct buffer_head * __find_get_block_slow(struct block_device *bdev, sector_t block) { - struct inode *bd_inode = bdev->bd_inode; - struct address_space *bd_mapping = bd_inode->i_mapping; + struct address_space *bd_mapping = bdev->bd_mapping; + const int blkbits = bd_mapping->host->i_blkbits; struct buffer_head *ret = NULL; pgoff_t index; struct buffer_head *bh; @@ -199,7 +199,7 @@ __find_get_block_slow(struct block_device *bdev, sector_t block) int all_mapped = 1; static DEFINE_RATELIMIT_STATE(last_warned, HZ, 1); - index = ((loff_t)block << bd_inode->i_blkbits) / PAGE_SIZE; + index = ((loff_t)block << blkbits) / PAGE_SIZE; folio = __filemap_get_folio(bd_mapping, index, FGP_ACCESSED, 0); if (IS_ERR(folio)) goto out; @@ -233,7 +233,7 @@ __find_get_block_slow(struct block_device *bdev, sector_t block) (unsigned long long)block, (unsigned long long)bh->b_blocknr, bh->b_state, bh->b_size, bdev, - 1 << bd_inode->i_blkbits); + 1 << blkbits); } out_unlock: spin_unlock(&bd_mapping->i_private_lock); @@ -1696,16 +1696,16 @@ EXPORT_SYMBOL(create_empty_buffers); */ void clean_bdev_aliases(struct block_device *bdev, sector_t block, sector_t len) { - struct inode *bd_inode = bdev->bd_inode; - struct address_space *bd_mapping = bd_inode->i_mapping; + struct address_space *bd_mapping = bdev->bd_mapping; + const int blkbits = bd_mapping->host->i_blkbits; struct folio_batch fbatch; - pgoff_t index = ((loff_t)block << bd_inode->i_blkbits) / PAGE_SIZE; + pgoff_t index = ((loff_t)block << blkbits) / PAGE_SIZE; pgoff_t end; int i, count; struct buffer_head *bh; struct buffer_head *head; - end = ((loff_t)(block + len - 1) << bd_inode->i_blkbits) / PAGE_SIZE; + end = ((loff_t)(block + len - 1) << blkbits) / PAGE_SIZE; folio_batch_init(&fbatch); while (filemap_get_folios(bd_mapping, &index, end, &fbatch)) { count = folio_batch_count(&fbatch); -- cgit v1.2.3-59-g8ed1b From 2d0026a4901be98000137af41bb42b6ead5a4c27 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 11 Apr 2024 15:53:39 +0100 Subject: gfs2: more obvious initializations of mapping->host what's going on is copying the ->host of bdev's address_space Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240411145346.2516848-4-viro@zeniv.linux.org.uk Signed-off-by: Christian Brauner --- fs/gfs2/glock.c | 2 +- fs/gfs2/ops_fstype.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 34540f9d011c..1ebcf6c90f2b 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1227,7 +1227,7 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, mapping = gfs2_glock2aspace(gl); if (mapping) { mapping->a_ops = &gfs2_meta_aops; - mapping->host = s->s_bdev->bd_inode; + mapping->host = s->s_bdev->bd_mapping->host; mapping->flags = 0; mapping_set_gfp_mask(mapping, GFP_NOFS); mapping->i_private_data = NULL; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 572d58e86296..fcf7dfd14f52 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -114,7 +114,7 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) address_space_init_once(mapping); mapping->a_ops = &gfs2_rgrp_aops; - mapping->host = sb->s_bdev->bd_inode; + mapping->host = sb->s_bdev->bd_mapping->host; mapping->flags = 0; mapping_set_gfp_mask(mapping, GFP_NOFS); mapping->i_private_data = NULL; -- cgit v1.2.3-59-g8ed1b From 338618338233a8567f48f70b228abca626ffb3ed Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 28 Apr 2024 20:39:55 -0400 Subject: nilfs_attach_log_writer(): use ->bd_mapping->host instead of ->bd_inode I suspect that inode_attach_wb() use is rather unidiomatic, but that's a separate story - in any case, its use is a few times per mount *and* the route by which we access that inode is "the host of address_space a page belongs to". Signed-off-by: Al Viro --- fs/nilfs2/segment.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index aa5290cb7467..15188e799580 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -2790,7 +2790,7 @@ int nilfs_attach_log_writer(struct super_block *sb, struct nilfs_root *root) if (!nilfs->ns_writer) return -ENOMEM; - inode_attach_wb(nilfs->ns_bdev->bd_inode, NULL); + inode_attach_wb(nilfs->ns_bdev->bd_mapping->host, NULL); err = nilfs_segctor_start_thread(nilfs->ns_writer); if (unlikely(err)) -- cgit v1.2.3-59-g8ed1b From 290fa9471ee040734bbe873c3d7b2a20240b375a Mon Sep 17 00:00:00 2001 From: "Ritesh Harjani (IBM)" Date: Fri, 3 May 2024 14:58:52 +0530 Subject: ext2: Remove LEGACY_DIRECT_IO dependency commit fb5de4358e1a ("ext2: Move direct-io to use iomap"), converted ext2 direct-io to iomap which killed the call to blockdev_direct_IO(). So let's remove LEGACY_DIRECT_IO config dependency from ext2 Kconfig. Signed-off-by: Ritesh Harjani (IBM) Signed-off-by: Jan Kara Message-Id: --- fs/ext2/Kconfig | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/ext2/Kconfig b/fs/ext2/Kconfig index d6cfb1849580..d5bce83ad905 100644 --- a/fs/ext2/Kconfig +++ b/fs/ext2/Kconfig @@ -3,7 +3,6 @@ config EXT2_FS tristate "Second extended fs support (DEPRECATED)" select BUFFER_HEAD select FS_IOMAP - select LEGACY_DIRECT_IO help Ext2 is a standard Linux file system for hard disks. -- cgit v1.2.3-59-g8ed1b From ee5814dddefbaa181cb247a75676dd5103775db1 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Tue, 30 Apr 2024 19:53:31 -0700 Subject: fsverity: use register_sysctl_init() to avoid kmemleak warning Since the fsverity sysctl registration runs as a builtin initcall, there is no corresponding sysctl deregistration and the resulting struct ctl_table_header is not used. This can cause a kmemleak warning just after the system boots up. (A pointer to the ctl_table_header is stored in the fsverity_sysctl_header static variable, which kmemleak should detect; however, the compiler can optimize out that variable.) Avoid the kmemleak warning by using register_sysctl_init() which is intended for use by builtin initcalls and uses kmemleak_not_leak(). Reported-by: Yi Zhang Closes: https://lore.kernel.org/r/CAHj4cs8DTSvR698UE040rs_pX1k-WVe7aR6N2OoXXuhXJPDC-w@mail.gmail.com Cc: stable@vger.kernel.org Reviewed-by: Joel Granados Link: https://lore.kernel.org/r/20240501025331.594183-1-ebiggers@kernel.org Signed-off-by: Eric Biggers --- fs/verity/init.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/verity/init.c b/fs/verity/init.c index cb2c9aac61ed..f440f0e61e3e 100644 --- a/fs/verity/init.c +++ b/fs/verity/init.c @@ -10,8 +10,6 @@ #include #ifdef CONFIG_SYSCTL -static struct ctl_table_header *fsverity_sysctl_header; - static struct ctl_table fsverity_sysctl_table[] = { #ifdef CONFIG_FS_VERITY_BUILTIN_SIGNATURES { @@ -28,10 +26,7 @@ static struct ctl_table fsverity_sysctl_table[] = { static void __init fsverity_init_sysctl(void) { - fsverity_sysctl_header = register_sysctl("fs/verity", - fsverity_sysctl_table); - if (!fsverity_sysctl_header) - panic("fsverity sysctl registration failed"); + register_sysctl_init("fs/verity", fsverity_sysctl_table); } #else /* CONFIG_SYSCTL */ static inline void fsverity_init_sysctl(void) -- cgit v1.2.3-59-g8ed1b From b844048011d3c771e7f8ae9354450f4e0c5c041a Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:23:49 +0100 Subject: gfs2: Add a migrate_folio operation for journalled files For journalled data, folio migration currently works by writing the folio back, freeing the folio and faulting the new folio back in. We can bypass that by telling the migration code to migrate the buffer_heads attached to our folios. Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Andreas Gruenbacher --- fs/gfs2/aops.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index 974aca9c8ea8..10d5acd3f742 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -116,8 +116,7 @@ static int gfs2_write_jdata_folio(struct folio *folio, * @folio: The folio to write * @wbc: The writeback control * - * This is shared between writepage and writepages and implements the - * core of the writepage operation. If a transaction is required then + * Implements the core of write back. If a transaction is required then * the checked flag will have been set and the transaction will have * already been started before this is called. */ @@ -755,6 +754,7 @@ static const struct address_space_operations gfs2_jdata_aops = { .readahead = gfs2_readahead, .dirty_folio = jdata_dirty_folio, .bmap = gfs2_bmap, + .migrate_folio = buffer_migrate_folio, .invalidate_folio = gfs2_invalidate_folio, .release_folio = gfs2_release_folio, .is_partially_uptodate = block_is_partially_uptodate, -- cgit v1.2.3-59-g8ed1b From 50fabd42cb2fa02b727e7786fb80f6e172334066 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Wed, 3 Apr 2024 18:23:51 +0100 Subject: gfs2: Convert gfs2_aspace_writepage() to use a folio Convert the incoming struct page to a folio and use it throughout. Saves six calls to compound_head(). Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Andreas Gruenbacher --- fs/gfs2/meta_io.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index f814054c8cd0..2b26e8d529aa 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -32,14 +32,14 @@ static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wbc) { + struct folio *folio = page_folio(page); struct buffer_head *bh, *head; int nr_underway = 0; blk_opf_t write_flags = REQ_META | REQ_PRIO | wbc_to_write_flags(wbc); - BUG_ON(!PageLocked(page)); - BUG_ON(!page_has_buffers(page)); + BUG_ON(!folio_test_locked(folio)); - head = page_buffers(page); + head = folio_buffers(folio); bh = head; do { @@ -55,7 +55,7 @@ static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wb if (wbc->sync_mode != WB_SYNC_NONE) { lock_buffer(bh); } else if (!trylock_buffer(bh)) { - redirty_page_for_writepage(wbc, page); + folio_redirty_for_writepage(wbc, folio); continue; } if (test_clear_buffer_dirty(bh)) { @@ -69,8 +69,8 @@ static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wb * The page and its buffers are protected by PageWriteback(), so we can * drop the bh refcounts early. */ - BUG_ON(PageWriteback(page)); - set_page_writeback(page); + BUG_ON(folio_test_writeback(folio)); + folio_start_writeback(folio); do { struct buffer_head *next = bh->b_this_page; @@ -80,10 +80,10 @@ static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wb } bh = next; } while (bh != head); - unlock_page(page); + folio_unlock(folio); if (nr_underway == 0) - end_page_writeback(page); + folio_end_writeback(folio); return 0; } -- cgit v1.2.3-59-g8ed1b From b63db58e2fa5d6963db9c45df88e60060f0ff35f Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 2 May 2024 09:03:15 -0400 Subject: eventfs/tracing: Add callback for release of an eventfs_inode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Synthetic events create and destroy tracefs files when they are created and removed. The tracing subsystem has its own file descriptor representing the state of the events attached to the tracefs files. There's a race between the eventfs files and this file descriptor of the tracing system where the following can cause an issue: With two scripts 'A' and 'B' doing: Script 'A': echo "hello int aaa" > /sys/kernel/tracing/synthetic_events while : do echo 0 > /sys/kernel/tracing/events/synthetic/hello/enable done Script 'B': echo > /sys/kernel/tracing/synthetic_events Script 'A' creates a synthetic event "hello" and then just writes zero into its enable file. Script 'B' removes all synthetic events (including the newly created "hello" event). What happens is that the opening of the "enable" file has: { struct trace_event_file *file = inode->i_private; int ret; ret = tracing_check_open_get_tr(file->tr); [..] But deleting the events frees the "file" descriptor, and a "use after free" happens with the dereference at "file->tr". The file descriptor does have a reference counter, but there needs to be a way to decrement it from the eventfs when the eventfs_inode is removed that represents this file descriptor. Add an optional "release" callback to the eventfs_entry array structure, that gets called when the eventfs file is about to be removed. This allows for the creating on the eventfs file to increment the tracing file descriptor ref counter. When the eventfs file is deleted, it can call the release function that will call the put function for the tracing file descriptor. This will protect the tracing file from being freed while a eventfs file that references it is being opened. Link: https://lore.kernel.org/linux-trace-kernel/20240426073410.17154-1-Tze-nan.Wu@mediatek.com/ Link: https://lore.kernel.org/linux-trace-kernel/20240502090315.448cba46@gandalf.local.home Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Fixes: 5790b1fb3d672 ("eventfs: Remove eventfs_file and just use eventfs_inode") Reported-by: Tze-nan wu Tested-by: Tze-nan Wu (吳澤南) Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 23 +++++++++++++++++++++-- include/linux/tracefs.h | 3 +++ kernel/trace/trace_events.c | 12 ++++++++++++ 3 files changed, 36 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 894c6ca1e500..f5510e26f0f6 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -84,10 +84,17 @@ enum { static void release_ei(struct kref *ref) { struct eventfs_inode *ei = container_of(ref, struct eventfs_inode, kref); + const struct eventfs_entry *entry; struct eventfs_root_inode *rei; WARN_ON_ONCE(!ei->is_freed); + for (int i = 0; i < ei->nr_entries; i++) { + entry = &ei->entries[i]; + if (entry->release) + entry->release(entry->name, ei->data); + } + kfree(ei->entry_attrs); kfree_const(ei->name); if (ei->is_events) { @@ -112,6 +119,18 @@ static inline void free_ei(struct eventfs_inode *ei) } } +/* + * Called when creation of an ei fails, do not call release() functions. + */ +static inline void cleanup_ei(struct eventfs_inode *ei) +{ + if (ei) { + /* Set nr_entries to 0 to prevent release() function being called */ + ei->nr_entries = 0; + free_ei(ei); + } +} + static inline struct eventfs_inode *get_ei(struct eventfs_inode *ei) { if (ei) @@ -734,7 +753,7 @@ struct eventfs_inode *eventfs_create_dir(const char *name, struct eventfs_inode /* Was the parent freed? */ if (list_empty(&ei->list)) { - free_ei(ei); + cleanup_ei(ei); ei = NULL; } return ei; @@ -835,7 +854,7 @@ struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry return ei; fail: - free_ei(ei); + cleanup_ei(ei); tracefs_failed_creating(dentry); return ERR_PTR(-ENOMEM); } diff --git a/include/linux/tracefs.h b/include/linux/tracefs.h index 7a5fe17b6bf9..d03f74658716 100644 --- a/include/linux/tracefs.h +++ b/include/linux/tracefs.h @@ -62,6 +62,8 @@ struct eventfs_file; typedef int (*eventfs_callback)(const char *name, umode_t *mode, void **data, const struct file_operations **fops); +typedef void (*eventfs_release)(const char *name, void *data); + /** * struct eventfs_entry - dynamically created eventfs file call back handler * @name: Then name of the dynamic file in an eventfs directory @@ -72,6 +74,7 @@ typedef int (*eventfs_callback)(const char *name, umode_t *mode, void **data, struct eventfs_entry { const char *name; eventfs_callback callback; + eventfs_release release; }; struct eventfs_inode; diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 52f75c36bbca..6ef29eba90ce 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -2552,6 +2552,14 @@ static int event_callback(const char *name, umode_t *mode, void **data, return 0; } +/* The file is incremented on creation and freeing the enable file decrements it */ +static void event_release(const char *name, void *data) +{ + struct trace_event_file *file = data; + + event_file_put(file); +} + static int event_create_dir(struct eventfs_inode *parent, struct trace_event_file *file) { @@ -2566,6 +2574,7 @@ event_create_dir(struct eventfs_inode *parent, struct trace_event_file *file) { .name = "enable", .callback = event_callback, + .release = event_release, }, { .name = "filter", @@ -2634,6 +2643,9 @@ event_create_dir(struct eventfs_inode *parent, struct trace_event_file *file) return ret; } + /* Gets decremented on freeing of the "enable" file */ + event_file_get(file); + return 0; } -- cgit v1.2.3-59-g8ed1b From ee4e0379475e4fe723986ae96293e465014fa8d9 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 2 May 2024 16:08:22 -0400 Subject: eventfs: Free all of the eventfs_inode after RCU The freeing of eventfs_inode via a kfree_rcu() callback. But the content of the eventfs_inode was being freed after the last kref. This is dangerous, as changes are being made that can access the content of an eventfs_inode from an RCU loop. Instead of using kfree_rcu() use call_rcu() that calls a function to do all the freeing of the eventfs_inode after a RCU grace period has expired. Link: https://lore.kernel.org/linux-trace-kernel/20240502200905.370261163@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 43aa6f97c2d03 ("eventfs: Get rid of dentry pointers without refcounts") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index f5510e26f0f6..cc8b838bbe62 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -73,6 +73,21 @@ enum { #define EVENTFS_MODE_MASK (EVENTFS_SAVE_MODE - 1) +static void free_ei_rcu(struct rcu_head *rcu) +{ + struct eventfs_inode *ei = container_of(rcu, struct eventfs_inode, rcu); + struct eventfs_root_inode *rei; + + kfree(ei->entry_attrs); + kfree_const(ei->name); + if (ei->is_events) { + rei = get_root_inode(ei); + kfree(rei); + } else { + kfree(ei); + } +} + /* * eventfs_inode reference count management. * @@ -85,7 +100,6 @@ static void release_ei(struct kref *ref) { struct eventfs_inode *ei = container_of(ref, struct eventfs_inode, kref); const struct eventfs_entry *entry; - struct eventfs_root_inode *rei; WARN_ON_ONCE(!ei->is_freed); @@ -95,14 +109,7 @@ static void release_ei(struct kref *ref) entry->release(entry->name, ei->data); } - kfree(ei->entry_attrs); - kfree_const(ei->name); - if (ei->is_events) { - rei = get_root_inode(ei); - kfree_rcu(rei, ei.rcu); - } else { - kfree_rcu(ei, rcu); - } + call_rcu(&ei->rcu, free_ei_rcu); } static inline void put_ei(struct eventfs_inode *ei) -- cgit v1.2.3-59-g8ed1b From baa23a8d4360d981a49913841a726edede5cdd54 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 2 May 2024 16:08:23 -0400 Subject: tracefs: Reset permissions on remount if permissions are options There's an inconsistency with the way permissions are handled in tracefs. Because the permissions are generated when accessed, they default to the root inode's permission if they were never set by the user. If the user sets the permissions, then a flag is set and the permissions are saved via the inode (for tracefs files) or an internal attribute field (for eventfs). But if a remount happens that specify the permissions, all the files that were not changed by the user gets updated, but the ones that were are not. If the user were to remount the file system with a given permission, then all files and directories within that file system should be updated. This can cause security issues if a file's permission was updated but the admin forgot about it. They could incorrectly think that remounting with permissions set would update all files, but miss some. For example: # cd /sys/kernel/tracing # chgrp 1002 current_tracer # ls -l [..] -rw-r----- 1 root root 0 May 1 21:25 buffer_size_kb -rw-r----- 1 root root 0 May 1 21:25 buffer_subbuf_size_kb -r--r----- 1 root root 0 May 1 21:25 buffer_total_size_kb -rw-r----- 1 root lkp 0 May 1 21:25 current_tracer -rw-r----- 1 root root 0 May 1 21:25 dynamic_events -r--r----- 1 root root 0 May 1 21:25 dyn_ftrace_total_info -r--r----- 1 root root 0 May 1 21:25 enabled_functions Where current_tracer now has group "lkp". # mount -o remount,gid=1001 . # ls -l -rw-r----- 1 root tracing 0 May 1 21:25 buffer_size_kb -rw-r----- 1 root tracing 0 May 1 21:25 buffer_subbuf_size_kb -r--r----- 1 root tracing 0 May 1 21:25 buffer_total_size_kb -rw-r----- 1 root lkp 0 May 1 21:25 current_tracer -rw-r----- 1 root tracing 0 May 1 21:25 dynamic_events -r--r----- 1 root tracing 0 May 1 21:25 dyn_ftrace_total_info -r--r----- 1 root tracing 0 May 1 21:25 enabled_functions Everything changed but the "current_tracer". Add a new link list that keeps track of all the tracefs_inodes which has the permission flags that tell if the file/dir should use the root inode's permission or not. Then on remount, clear all the flags so that the default behavior of using the root inode's permission is done for all files and directories. Link: https://lore.kernel.org/linux-trace-kernel/20240502200905.529542160@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 8186fff7ab649 ("tracefs/eventfs: Use root and instance inodes as default ownership") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 29 +++++++++++++++++++++ fs/tracefs/inode.c | 65 +++++++++++++++++++++++++++++++++++++++++++++++- fs/tracefs/internal.h | 7 +++++- 3 files changed, 99 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index cc8b838bbe62..15a2a9c3c62b 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -308,6 +308,35 @@ static const struct file_operations eventfs_file_operations = { .llseek = generic_file_llseek, }; +/* + * On a remount of tracefs, if UID or GID options are set, then + * the mount point inode permissions should be used. + * Reset the saved permission flags appropriately. + */ +void eventfs_remount(struct tracefs_inode *ti, bool update_uid, bool update_gid) +{ + struct eventfs_inode *ei = ti->private; + + if (!ei) + return; + + if (update_uid) + ei->attr.mode &= ~EVENTFS_SAVE_UID; + + if (update_gid) + ei->attr.mode &= ~EVENTFS_SAVE_GID; + + if (!ei->entry_attrs) + return; + + for (int i = 0; i < ei->nr_entries; i++) { + if (update_uid) + ei->entry_attrs[i].mode &= ~EVENTFS_SAVE_UID; + if (update_gid) + ei->entry_attrs[i].mode &= ~EVENTFS_SAVE_GID; + } +} + /* Return the evenfs_inode of the "events" directory */ static struct eventfs_inode *eventfs_find_events(struct dentry *dentry) { diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index 5545e6bf7d26..52aa14bd2994 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -30,20 +30,47 @@ static struct vfsmount *tracefs_mount; static int tracefs_mount_count; static bool tracefs_registered; +/* + * Keep track of all tracefs_inodes in order to update their + * flags if necessary on a remount. + */ +static DEFINE_SPINLOCK(tracefs_inode_lock); +static LIST_HEAD(tracefs_inodes); + static struct inode *tracefs_alloc_inode(struct super_block *sb) { struct tracefs_inode *ti; + unsigned long flags; ti = kmem_cache_alloc(tracefs_inode_cachep, GFP_KERNEL); if (!ti) return NULL; + spin_lock_irqsave(&tracefs_inode_lock, flags); + list_add_rcu(&ti->list, &tracefs_inodes); + spin_unlock_irqrestore(&tracefs_inode_lock, flags); + return &ti->vfs_inode; } +static void tracefs_free_inode_rcu(struct rcu_head *rcu) +{ + struct tracefs_inode *ti; + + ti = container_of(rcu, struct tracefs_inode, rcu); + kmem_cache_free(tracefs_inode_cachep, ti); +} + static void tracefs_free_inode(struct inode *inode) { - kmem_cache_free(tracefs_inode_cachep, get_tracefs(inode)); + struct tracefs_inode *ti = get_tracefs(inode); + unsigned long flags; + + spin_lock_irqsave(&tracefs_inode_lock, flags); + list_del_rcu(&ti->list); + spin_unlock_irqrestore(&tracefs_inode_lock, flags); + + call_rcu(&ti->rcu, tracefs_free_inode_rcu); } static ssize_t default_read_file(struct file *file, char __user *buf, @@ -313,6 +340,8 @@ static int tracefs_apply_options(struct super_block *sb, bool remount) struct tracefs_fs_info *fsi = sb->s_fs_info; struct inode *inode = d_inode(sb->s_root); struct tracefs_mount_opts *opts = &fsi->mount_opts; + struct tracefs_inode *ti; + bool update_uid, update_gid; umode_t tmp_mode; /* @@ -332,6 +361,25 @@ static int tracefs_apply_options(struct super_block *sb, bool remount) if (!remount || opts->opts & BIT(Opt_gid)) inode->i_gid = opts->gid; + if (remount && (opts->opts & BIT(Opt_uid) || opts->opts & BIT(Opt_gid))) { + + update_uid = opts->opts & BIT(Opt_uid); + update_gid = opts->opts & BIT(Opt_gid); + + rcu_read_lock(); + list_for_each_entry_rcu(ti, &tracefs_inodes, list) { + if (update_uid) + ti->flags &= ~TRACEFS_UID_PERM_SET; + + if (update_gid) + ti->flags &= ~TRACEFS_GID_PERM_SET; + + if (ti->flags & TRACEFS_EVENT_INODE) + eventfs_remount(ti, update_uid, update_gid); + } + rcu_read_unlock(); + } + return 0; } @@ -398,7 +446,22 @@ static int tracefs_d_revalidate(struct dentry *dentry, unsigned int flags) return !(ei && ei->is_freed); } +static void tracefs_d_iput(struct dentry *dentry, struct inode *inode) +{ + struct tracefs_inode *ti = get_tracefs(inode); + + /* + * This inode is being freed and cannot be used for + * eventfs. Clear the flag so that it doesn't call into + * eventfs during the remount flag updates. The eventfs_inode + * gets freed after an RCU cycle, so the content will still + * be safe if the iteration is going on now. + */ + ti->flags &= ~TRACEFS_EVENT_INODE; +} + static const struct dentry_operations tracefs_dentry_operations = { + .d_iput = tracefs_d_iput, .d_revalidate = tracefs_d_revalidate, .d_release = tracefs_d_release, }; diff --git a/fs/tracefs/internal.h b/fs/tracefs/internal.h index 15c26f9aaad4..29f0c999975b 100644 --- a/fs/tracefs/internal.h +++ b/fs/tracefs/internal.h @@ -11,8 +11,12 @@ enum { }; struct tracefs_inode { - struct inode vfs_inode; + union { + struct inode vfs_inode; + struct rcu_head rcu; + }; /* The below gets initialized with memset_after(ti, 0, vfs_inode) */ + struct list_head list; unsigned long flags; void *private; }; @@ -73,6 +77,7 @@ struct dentry *tracefs_end_creating(struct dentry *dentry); struct dentry *tracefs_failed_creating(struct dentry *dentry); struct inode *tracefs_get_inode(struct super_block *sb); +void eventfs_remount(struct tracefs_inode *ti, bool update_uid, bool update_gid); void eventfs_d_release(struct dentry *dentry); #endif /* _TRACEFS_INTERNAL_H */ -- cgit v1.2.3-59-g8ed1b From 6599bd5517be66c8344f869f3ca3a91bc10f2b9e Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 2 May 2024 16:08:24 -0400 Subject: tracefs: Still use mount point as default permissions for instances If the instances directory's permissions were never change, then have it and its children use the mount point permissions as the default. Currently, the permissions of instance directories are determined by the instance directory's permissions itself. But if the tracefs file system is remounted and changes the permissions, the instance directory and its children should use the new permission. But because both the instance directory and its children use the instance directory's inode for permissions, it misses the update. To demonstrate this: # cd /sys/kernel/tracing/ # mkdir instances/foo # ls -ld instances/foo drwxr-x--- 5 root root 0 May 1 19:07 instances/foo # ls -ld instances drwxr-x--- 3 root root 0 May 1 18:57 instances # ls -ld current_tracer -rw-r----- 1 root root 0 May 1 18:57 current_tracer # mount -o remount,gid=1002 . # ls -ld instances drwxr-x--- 3 root root 0 May 1 18:57 instances # ls -ld instances/foo/ drwxr-x--- 5 root root 0 May 1 19:07 instances/foo/ # ls -ld current_tracer -rw-r----- 1 root lkp 0 May 1 18:57 current_tracer Notice that changing the group id to that of "lkp" did not affect the instances directory nor its children. It should have been: # ls -ld current_tracer -rw-r----- 1 root root 0 May 1 19:19 current_tracer # ls -ld instances/foo/ drwxr-x--- 5 root root 0 May 1 19:25 instances/foo/ # ls -ld instances drwxr-x--- 3 root root 0 May 1 19:19 instances # mount -o remount,gid=1002 . # ls -ld current_tracer -rw-r----- 1 root lkp 0 May 1 19:19 current_tracer # ls -ld instances drwxr-x--- 3 root lkp 0 May 1 19:19 instances # ls -ld instances/foo/ drwxr-x--- 5 root lkp 0 May 1 19:25 instances/foo/ Where all files were updated by the remount gid update. Link: https://lore.kernel.org/linux-trace-kernel/20240502200905.686838327@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 8186fff7ab649 ("tracefs/eventfs: Use root and instance inodes as default ownership") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/inode.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index 52aa14bd2994..417c840e6403 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -180,16 +180,39 @@ static void set_tracefs_inode_owner(struct inode *inode) { struct tracefs_inode *ti = get_tracefs(inode); struct inode *root_inode = ti->private; + kuid_t uid; + kgid_t gid; + + uid = root_inode->i_uid; + gid = root_inode->i_gid; + + /* + * If the root is not the mount point, then check the root's + * permissions. If it was never set, then default to the + * mount point. + */ + if (root_inode != d_inode(root_inode->i_sb->s_root)) { + struct tracefs_inode *rti; + + rti = get_tracefs(root_inode); + root_inode = d_inode(root_inode->i_sb->s_root); + + if (!(rti->flags & TRACEFS_UID_PERM_SET)) + uid = root_inode->i_uid; + + if (!(rti->flags & TRACEFS_GID_PERM_SET)) + gid = root_inode->i_gid; + } /* * If this inode has never been referenced, then update * the permissions to the superblock. */ if (!(ti->flags & TRACEFS_UID_PERM_SET)) - inode->i_uid = root_inode->i_uid; + inode->i_uid = uid; if (!(ti->flags & TRACEFS_GID_PERM_SET)) - inode->i_gid = root_inode->i_gid; + inode->i_gid = gid; } static int tracefs_permission(struct mnt_idmap *idmap, -- cgit v1.2.3-59-g8ed1b From d53891d348ac3eceaf48f4732a1f4f5c0e0a55ce Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 2 May 2024 16:08:25 -0400 Subject: eventfs: Do not differentiate the toplevel events directory The toplevel events directory is really no different than the events directory of instances. Having the two be different caused inconsistencies and made it harder to fix the permissions bugs. Make all events directories act the same. Link: https://lore.kernel.org/linux-trace-kernel/20240502200905.846448710@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 8186fff7ab649 ("tracefs/eventfs: Use root and instance inodes as default ownership") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 29 ++++++++--------------------- fs/tracefs/internal.h | 7 +++---- 2 files changed, 11 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 15a2a9c3c62b..9dacf65c0b6e 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -68,7 +68,6 @@ enum { EVENTFS_SAVE_MODE = BIT(16), EVENTFS_SAVE_UID = BIT(17), EVENTFS_SAVE_GID = BIT(18), - EVENTFS_TOPLEVEL = BIT(19), }; #define EVENTFS_MODE_MASK (EVENTFS_SAVE_MODE - 1) @@ -239,14 +238,10 @@ static int eventfs_set_attr(struct mnt_idmap *idmap, struct dentry *dentry, return ret; } -static void update_top_events_attr(struct eventfs_inode *ei, struct super_block *sb) +static void update_events_attr(struct eventfs_inode *ei, struct super_block *sb) { struct inode *root; - /* Only update if the "events" was on the top level */ - if (!ei || !(ei->attr.mode & EVENTFS_TOPLEVEL)) - return; - /* Get the tracefs root inode. */ root = d_inode(sb->s_root); ei->attr.uid = root->i_uid; @@ -259,10 +254,10 @@ static void set_top_events_ownership(struct inode *inode) struct eventfs_inode *ei = ti->private; /* The top events directory doesn't get automatically updated */ - if (!ei || !ei->is_events || !(ei->attr.mode & EVENTFS_TOPLEVEL)) + if (!ei || !ei->is_events) return; - update_top_events_attr(ei, inode->i_sb); + update_events_attr(ei, inode->i_sb); if (!(ei->attr.mode & EVENTFS_SAVE_UID)) inode->i_uid = ei->attr.uid; @@ -291,7 +286,7 @@ static int eventfs_permission(struct mnt_idmap *idmap, return generic_permission(idmap, inode, mask); } -static const struct inode_operations eventfs_root_dir_inode_operations = { +static const struct inode_operations eventfs_dir_inode_operations = { .lookup = eventfs_root_lookup, .setattr = eventfs_set_attr, .getattr = eventfs_get_attr, @@ -359,7 +354,7 @@ static struct eventfs_inode *eventfs_find_events(struct dentry *dentry) // Walk upwards until you find the events inode } while (!ei->is_events); - update_top_events_attr(ei, dentry->d_sb); + update_events_attr(ei, dentry->d_sb); return ei; } @@ -465,7 +460,7 @@ static struct dentry *lookup_dir_entry(struct dentry *dentry, update_inode_attr(dentry, inode, &ei->attr, S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO); - inode->i_op = &eventfs_root_dir_inode_operations; + inode->i_op = &eventfs_dir_inode_operations; inode->i_fop = &eventfs_file_operations; /* All directories will have the same inode number */ @@ -845,14 +840,6 @@ struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry uid = d_inode(dentry->d_parent)->i_uid; gid = d_inode(dentry->d_parent)->i_gid; - /* - * If the events directory is of the top instance, then parent - * is NULL. Set the attr.mode to reflect this and its permissions will - * default to the tracefs root dentry. - */ - if (!parent) - ei->attr.mode = EVENTFS_TOPLEVEL; - /* This is used as the default ownership of the files and directories */ ei->attr.uid = uid; ei->attr.gid = gid; @@ -861,13 +848,13 @@ struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry INIT_LIST_HEAD(&ei->list); ti = get_tracefs(inode); - ti->flags |= TRACEFS_EVENT_INODE | TRACEFS_EVENT_TOP_INODE; + ti->flags |= TRACEFS_EVENT_INODE; ti->private = ei; inode->i_mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO; inode->i_uid = uid; inode->i_gid = gid; - inode->i_op = &eventfs_root_dir_inode_operations; + inode->i_op = &eventfs_dir_inode_operations; inode->i_fop = &eventfs_file_operations; dentry->d_fsdata = get_ei(ei); diff --git a/fs/tracefs/internal.h b/fs/tracefs/internal.h index 29f0c999975b..f704d8348357 100644 --- a/fs/tracefs/internal.h +++ b/fs/tracefs/internal.h @@ -4,10 +4,9 @@ enum { TRACEFS_EVENT_INODE = BIT(1), - TRACEFS_EVENT_TOP_INODE = BIT(2), - TRACEFS_GID_PERM_SET = BIT(3), - TRACEFS_UID_PERM_SET = BIT(4), - TRACEFS_INSTANCE_INODE = BIT(5), + TRACEFS_GID_PERM_SET = BIT(2), + TRACEFS_UID_PERM_SET = BIT(3), + TRACEFS_INSTANCE_INODE = BIT(4), }; struct tracefs_inode { -- cgit v1.2.3-59-g8ed1b From 22e61e15af731dbe46704c775d2335e56fcef4e9 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 2 May 2024 16:08:26 -0400 Subject: eventfs: Do not treat events directory different than other directories Treat the events directory the same as other directories when it comes to permissions. The events directory was considered different because it's dentry is persistent, whereas the other directory dentries are created when accessed. But the way tracefs now does its ownership by using the root dentry's permissions as the default permissions, the events directory can get out of sync when a remount is performed setting the group and user permissions. Remove the special case for the events directory on setting the attributes. This allows the updates caused by remount to work properly as well as simplifies the code. Link: https://lore.kernel.org/linux-trace-kernel/20240502200906.002923579@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 8186fff7ab649 ("tracefs/eventfs: Use root and instance inodes as default ownership") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 9dacf65c0b6e..6e08405892ae 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -206,21 +206,7 @@ static int eventfs_set_attr(struct mnt_idmap *idmap, struct dentry *dentry, * determined by the parent directory. */ if (dentry->d_inode->i_mode & S_IFDIR) { - /* - * The events directory dentry is never freed, unless its - * part of an instance that is deleted. It's attr is the - * default for its child files and directories. - * Do not update it. It's not used for its own mode or ownership. - */ - if (ei->is_events) { - /* But it still needs to know if it was modified */ - if (iattr->ia_valid & ATTR_UID) - ei->attr.mode |= EVENTFS_SAVE_UID; - if (iattr->ia_valid & ATTR_GID) - ei->attr.mode |= EVENTFS_SAVE_GID; - } else { - update_attr(&ei->attr, iattr); - } + update_attr(&ei->attr, iattr); } else { name = dentry->d_name.name; -- cgit v1.2.3-59-g8ed1b From d57cf30c4c07837799edec949102b0adf58bae79 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 2 May 2024 16:08:27 -0400 Subject: eventfs: Have "events" directory get permissions from its parent The events directory gets its permissions from the root inode. But this can cause an inconsistency if the instances directory changes its permissions, as the permissions of the created directories under it should inherit the permissions of the instances directory when directories under it are created. Currently the behavior is: # cd /sys/kernel/tracing # chgrp 1002 instances # mkdir instances/foo # ls -l instances/foo [..] -r--r----- 1 root lkp 0 May 1 18:55 buffer_total_size_kb -rw-r----- 1 root lkp 0 May 1 18:55 current_tracer -rw-r----- 1 root lkp 0 May 1 18:55 error_log drwxr-xr-x 1 root root 0 May 1 18:55 events --w------- 1 root lkp 0 May 1 18:55 free_buffer drwxr-x--- 2 root lkp 0 May 1 18:55 options drwxr-x--- 10 root lkp 0 May 1 18:55 per_cpu -rw-r----- 1 root lkp 0 May 1 18:55 set_event All the files and directories under "foo" has the "lkp" group except the "events" directory. That's because its getting its default value from the mount point instead of its parent. Have the "events" directory make its default value based on its parent's permissions. That now gives: # ls -l instances/foo [..] -rw-r----- 1 root lkp 0 May 1 21:16 buffer_subbuf_size_kb -r--r----- 1 root lkp 0 May 1 21:16 buffer_total_size_kb -rw-r----- 1 root lkp 0 May 1 21:16 current_tracer -rw-r----- 1 root lkp 0 May 1 21:16 error_log drwxr-xr-x 1 root lkp 0 May 1 21:16 events --w------- 1 root lkp 0 May 1 21:16 free_buffer drwxr-x--- 2 root lkp 0 May 1 21:16 options drwxr-x--- 10 root lkp 0 May 1 21:16 per_cpu -rw-r----- 1 root lkp 0 May 1 21:16 set_event Link: https://lore.kernel.org/linux-trace-kernel/20240502200906.161887248@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 8186fff7ab649 ("tracefs/eventfs: Use root and instance inodes as default ownership") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 30 ++++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 6e08405892ae..a878cea70f4c 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -37,6 +37,7 @@ static DEFINE_MUTEX(eventfs_mutex); struct eventfs_root_inode { struct eventfs_inode ei; + struct inode *parent_inode; struct dentry *events_dir; }; @@ -226,12 +227,23 @@ static int eventfs_set_attr(struct mnt_idmap *idmap, struct dentry *dentry, static void update_events_attr(struct eventfs_inode *ei, struct super_block *sb) { - struct inode *root; + struct eventfs_root_inode *rei; + struct inode *parent; + + rei = get_root_inode(ei); + + /* Use the parent inode permissions unless root set its permissions */ + parent = rei->parent_inode; - /* Get the tracefs root inode. */ - root = d_inode(sb->s_root); - ei->attr.uid = root->i_uid; - ei->attr.gid = root->i_gid; + if (rei->ei.attr.mode & EVENTFS_SAVE_UID) + ei->attr.uid = rei->ei.attr.uid; + else + ei->attr.uid = parent->i_uid; + + if (rei->ei.attr.mode & EVENTFS_SAVE_GID) + ei->attr.gid = rei->ei.attr.gid; + else + ei->attr.gid = parent->i_gid; } static void set_top_events_ownership(struct inode *inode) @@ -817,6 +829,7 @@ struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry // Note: we have a ref to the dentry from tracefs_start_creating() rei = get_root_inode(ei); rei->events_dir = dentry; + rei->parent_inode = d_inode(dentry->d_sb->s_root); ei->entries = entries; ei->nr_entries = size; @@ -826,10 +839,15 @@ struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry uid = d_inode(dentry->d_parent)->i_uid; gid = d_inode(dentry->d_parent)->i_gid; - /* This is used as the default ownership of the files and directories */ ei->attr.uid = uid; ei->attr.gid = gid; + /* + * When the "events" directory is created, it takes on the + * permissions of its parent. But can be reset on remount. + */ + ei->attr.mode |= EVENTFS_SAVE_UID | EVENTFS_SAVE_GID; + INIT_LIST_HEAD(&ei->children); INIT_LIST_HEAD(&ei->list); -- cgit v1.2.3-59-g8ed1b From 75cde4e37a14bdb8554c44cbc22354fc10fdf127 Mon Sep 17 00:00:00 2001 From: Li zeming Date: Mon, 15 Apr 2024 18:20:09 +0800 Subject: kernfs: mount: Remove unnecessary ‘NULL’ values from knparent MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit knparent is assigned first, so it does not need to initialize the assignment. Signed-off-by: Li zeming Acked-by: Tejun Heo Link: https://lore.kernel.org/r/20240415102009.9926-1-zeming@nfschina.com Signed-off-by: Greg Kroah-Hartman --- fs/kernfs/mount.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/kernfs/mount.c b/fs/kernfs/mount.c index e29f4edf9572..1358c21837f1 100644 --- a/fs/kernfs/mount.c +++ b/fs/kernfs/mount.c @@ -206,7 +206,7 @@ struct dentry *kernfs_node_dentry(struct kernfs_node *kn, struct super_block *sb) { struct dentry *dentry; - struct kernfs_node *knparent = NULL; + struct kernfs_node *knparent; BUG_ON(sb->s_op != &kernfs_sops); -- cgit v1.2.3-59-g8ed1b From cc00bc83f26eb8f2d8d9f56b949b62fd774d8432 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Wed, 1 May 2024 21:41:50 +0900 Subject: ksmbd: off ipv6only for both ipv4/ipv6 binding MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ΕΛΕΝΗ reported that ksmbd binds to the IPV6 wildcard (::) by default for ipv4 and ipv6 binding. So IPV4 connections are successful only when the Linux system parameter bindv6only is set to 0 [default value]. If this parameter is set to 1, then the ipv6 wildcard only represents any IPV6 address. Samba creates different sockets for ipv4 and ipv6 by default. This patch off sk_ipv6only to support IPV4/IPV6 connections without creating two sockets. Cc: stable@vger.kernel.org Reported-by: ΕΛΕΝΗ ΤΖΑΒΕΛΛΑ Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/transport_tcp.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/smb/server/transport_tcp.c b/fs/smb/server/transport_tcp.c index 002a3f0dc7c5..6633fa78e9b9 100644 --- a/fs/smb/server/transport_tcp.c +++ b/fs/smb/server/transport_tcp.c @@ -448,6 +448,10 @@ static int create_socket(struct interface *iface) sin6.sin6_family = PF_INET6; sin6.sin6_addr = in6addr_any; sin6.sin6_port = htons(server_conf.tcp_port); + + lock_sock(ksmbd_socket->sk); + ksmbd_socket->sk->sk_ipv6only = false; + release_sock(ksmbd_socket->sk); } ksmbd_tcp_nodelay(ksmbd_socket); -- cgit v1.2.3-59-g8ed1b From 97c2ec64667bacc49881d2b2dd9afd4d1c3fbaeb Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Wed, 1 May 2024 21:44:02 +0900 Subject: ksmbd: avoid to send duplicate lease break notifications This patch fixes generic/011 when enable smb2 leases. if ksmbd sends multiple notifications for a file, cifs increments the reference count of the file but it does not decrement the count by the failure of queue_work. So even if the file is closed, cifs does not send a SMB2_CLOSE request. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/oplock.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index 4978edfb15f9..6fd8cb7064dc 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -613,13 +613,23 @@ static int oplock_break_pending(struct oplock_info *opinfo, int req_op_level) if (opinfo->op_state == OPLOCK_CLOSING) return -ENOENT; - else if (!opinfo->is_lease && opinfo->level <= req_op_level) - return 1; + else if (opinfo->level <= req_op_level) { + if (opinfo->is_lease && + opinfo->o_lease->state != + (SMB2_LEASE_HANDLE_CACHING_LE | + SMB2_LEASE_READ_CACHING_LE)) + return 1; + } } - if (!opinfo->is_lease && opinfo->level <= req_op_level) { - wake_up_oplock_break(opinfo); - return 1; + if (opinfo->level <= req_op_level) { + if (opinfo->is_lease && + opinfo->o_lease->state != + (SMB2_LEASE_HANDLE_CACHING_LE | + SMB2_LEASE_READ_CACHING_LE)) { + wake_up_oplock_break(opinfo); + return 1; + } } return 0; } @@ -887,7 +897,6 @@ static int oplock_break(struct oplock_info *brk_opinfo, int req_op_level) struct lease *lease = brk_opinfo->o_lease; atomic_inc(&brk_opinfo->breaking_cnt); - err = oplock_break_pending(brk_opinfo, req_op_level); if (err) return err < 0 ? err : 0; -- cgit v1.2.3-59-g8ed1b From d1c189c6cb8b0fb7b5ee549237d27889c40c2f8b Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Thu, 2 May 2024 10:07:50 +0900 Subject: ksmbd: use rwsem instead of rwlock for lease break lease break wait for lease break acknowledgment. rwsem is more suitable than unlock while traversing the list for parent lease break in ->m_op_list. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/oplock.c | 30 +++++++++++------------------- fs/smb/server/smb2pdu.c | 4 ++-- fs/smb/server/smb_common.c | 4 ++-- fs/smb/server/vfs_cache.c | 28 ++++++++++++++-------------- fs/smb/server/vfs_cache.h | 2 +- 5 files changed, 30 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index 6fd8cb7064dc..c2abf109010d 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -207,9 +207,9 @@ static void opinfo_add(struct oplock_info *opinfo) { struct ksmbd_inode *ci = opinfo->o_fp->f_ci; - write_lock(&ci->m_lock); + down_write(&ci->m_lock); list_add_rcu(&opinfo->op_entry, &ci->m_op_list); - write_unlock(&ci->m_lock); + up_write(&ci->m_lock); } static void opinfo_del(struct oplock_info *opinfo) @@ -221,9 +221,9 @@ static void opinfo_del(struct oplock_info *opinfo) lease_del_list(opinfo); write_unlock(&lease_list_lock); } - write_lock(&ci->m_lock); + down_write(&ci->m_lock); list_del_rcu(&opinfo->op_entry); - write_unlock(&ci->m_lock); + up_write(&ci->m_lock); } static unsigned long opinfo_count(struct ksmbd_file *fp) @@ -526,21 +526,18 @@ static struct oplock_info *same_client_has_lease(struct ksmbd_inode *ci, * Compare lease key and client_guid to know request from same owner * of same client */ - read_lock(&ci->m_lock); + down_read(&ci->m_lock); list_for_each_entry(opinfo, &ci->m_op_list, op_entry) { if (!opinfo->is_lease || !opinfo->conn) continue; - read_unlock(&ci->m_lock); lease = opinfo->o_lease; ret = compare_guid_key(opinfo, client_guid, lctx->lease_key); if (ret) { m_opinfo = opinfo; /* skip upgrading lease about breaking lease */ - if (atomic_read(&opinfo->breaking_cnt)) { - read_lock(&ci->m_lock); + if (atomic_read(&opinfo->breaking_cnt)) continue; - } /* upgrading lease */ if ((atomic_read(&ci->op_count) + @@ -570,9 +567,8 @@ static struct oplock_info *same_client_has_lease(struct ksmbd_inode *ci, lease_none_upgrade(opinfo, lctx->req_state); } } - read_lock(&ci->m_lock); } - read_unlock(&ci->m_lock); + up_read(&ci->m_lock); return m_opinfo; } @@ -1114,7 +1110,7 @@ void smb_send_parent_lease_break_noti(struct ksmbd_file *fp, if (!p_ci) return; - read_lock(&p_ci->m_lock); + down_read(&p_ci->m_lock); list_for_each_entry(opinfo, &p_ci->m_op_list, op_entry) { if (opinfo->conn == NULL || !opinfo->is_lease) continue; @@ -1132,13 +1128,11 @@ void smb_send_parent_lease_break_noti(struct ksmbd_file *fp, continue; } - read_unlock(&p_ci->m_lock); oplock_break(opinfo, SMB2_OPLOCK_LEVEL_NONE); opinfo_conn_put(opinfo); - read_lock(&p_ci->m_lock); } } - read_unlock(&p_ci->m_lock); + up_read(&p_ci->m_lock); ksmbd_inode_put(p_ci); } @@ -1159,7 +1153,7 @@ void smb_lazy_parent_lease_break_close(struct ksmbd_file *fp) if (!p_ci) return; - read_lock(&p_ci->m_lock); + down_read(&p_ci->m_lock); list_for_each_entry(opinfo, &p_ci->m_op_list, op_entry) { if (opinfo->conn == NULL || !opinfo->is_lease) continue; @@ -1173,13 +1167,11 @@ void smb_lazy_parent_lease_break_close(struct ksmbd_file *fp) atomic_dec(&opinfo->conn->r_count); continue; } - read_unlock(&p_ci->m_lock); oplock_break(opinfo, SMB2_OPLOCK_LEVEL_NONE); opinfo_conn_put(opinfo); - read_lock(&p_ci->m_lock); } } - read_unlock(&p_ci->m_lock); + up_read(&p_ci->m_lock); ksmbd_inode_put(p_ci); } diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 30229161b346..b6c5a8ea3887 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -3376,9 +3376,9 @@ int smb2_open(struct ksmbd_work *work) * after daccess, saccess, attrib_only, and stream are * initialized. */ - write_lock(&fp->f_ci->m_lock); + down_write(&fp->f_ci->m_lock); list_add(&fp->node, &fp->f_ci->m_fp_list); - write_unlock(&fp->f_ci->m_lock); + up_write(&fp->f_ci->m_lock); /* Check delete pending among previous fp before oplock break */ if (ksmbd_inode_pending_delete(fp)) { diff --git a/fs/smb/server/smb_common.c b/fs/smb/server/smb_common.c index fcaf373cc008..474dadf6b7b8 100644 --- a/fs/smb/server/smb_common.c +++ b/fs/smb/server/smb_common.c @@ -646,7 +646,7 @@ int ksmbd_smb_check_shared_mode(struct file *filp, struct ksmbd_file *curr_fp) * Lookup fp in master fp list, and check desired access and * shared mode between previous open and current open. */ - read_lock(&curr_fp->f_ci->m_lock); + down_read(&curr_fp->f_ci->m_lock); list_for_each_entry(prev_fp, &curr_fp->f_ci->m_fp_list, node) { if (file_inode(filp) != file_inode(prev_fp->filp)) continue; @@ -722,7 +722,7 @@ int ksmbd_smb_check_shared_mode(struct file *filp, struct ksmbd_file *curr_fp) break; } } - read_unlock(&curr_fp->f_ci->m_lock); + up_read(&curr_fp->f_ci->m_lock); return rc; } diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c index 030f70700036..6cb599cd287e 100644 --- a/fs/smb/server/vfs_cache.c +++ b/fs/smb/server/vfs_cache.c @@ -165,7 +165,7 @@ static int ksmbd_inode_init(struct ksmbd_inode *ci, struct ksmbd_file *fp) ci->m_fattr = 0; INIT_LIST_HEAD(&ci->m_fp_list); INIT_LIST_HEAD(&ci->m_op_list); - rwlock_init(&ci->m_lock); + init_rwsem(&ci->m_lock); ci->m_de = fp->filp->f_path.dentry; return 0; } @@ -261,14 +261,14 @@ static void __ksmbd_inode_close(struct ksmbd_file *fp) } if (atomic_dec_and_test(&ci->m_count)) { - write_lock(&ci->m_lock); + down_write(&ci->m_lock); if (ci->m_flags & (S_DEL_ON_CLS | S_DEL_PENDING)) { ci->m_flags &= ~(S_DEL_ON_CLS | S_DEL_PENDING); - write_unlock(&ci->m_lock); + up_write(&ci->m_lock); ksmbd_vfs_unlink(filp); - write_lock(&ci->m_lock); + down_write(&ci->m_lock); } - write_unlock(&ci->m_lock); + up_write(&ci->m_lock); ksmbd_inode_free(ci); } @@ -289,9 +289,9 @@ static void __ksmbd_remove_fd(struct ksmbd_file_table *ft, struct ksmbd_file *fp if (!has_file_id(fp->volatile_id)) return; - write_lock(&fp->f_ci->m_lock); + down_write(&fp->f_ci->m_lock); list_del_init(&fp->node); - write_unlock(&fp->f_ci->m_lock); + up_write(&fp->f_ci->m_lock); write_lock(&ft->lock); idr_remove(ft->idr, fp->volatile_id); @@ -523,17 +523,17 @@ struct ksmbd_file *ksmbd_lookup_fd_inode(struct dentry *dentry) if (!ci) return NULL; - read_lock(&ci->m_lock); + down_read(&ci->m_lock); list_for_each_entry(lfp, &ci->m_fp_list, node) { if (inode == file_inode(lfp->filp)) { atomic_dec(&ci->m_count); lfp = ksmbd_fp_get(lfp); - read_unlock(&ci->m_lock); + up_read(&ci->m_lock); return lfp; } } atomic_dec(&ci->m_count); - read_unlock(&ci->m_lock); + up_read(&ci->m_lock); return NULL; } @@ -705,13 +705,13 @@ static bool session_fd_check(struct ksmbd_tree_connect *tcon, conn = fp->conn; ci = fp->f_ci; - write_lock(&ci->m_lock); + down_write(&ci->m_lock); list_for_each_entry_rcu(op, &ci->m_op_list, op_entry) { if (op->conn != conn) continue; op->conn = NULL; } - write_unlock(&ci->m_lock); + up_write(&ci->m_lock); fp->conn = NULL; fp->tcon = NULL; @@ -801,13 +801,13 @@ int ksmbd_reopen_durable_fd(struct ksmbd_work *work, struct ksmbd_file *fp) fp->tcon = work->tcon; ci = fp->f_ci; - write_lock(&ci->m_lock); + down_write(&ci->m_lock); list_for_each_entry_rcu(op, &ci->m_op_list, op_entry) { if (op->conn) continue; op->conn = fp->conn; } - write_unlock(&ci->m_lock); + up_write(&ci->m_lock); __open_id(&work->sess->file_table, fp, OPEN_ID_TYPE_VOLATILE_ID); if (!has_file_id(fp->volatile_id)) { diff --git a/fs/smb/server/vfs_cache.h b/fs/smb/server/vfs_cache.h index ed44fb4e18e7..5a225e7055f1 100644 --- a/fs/smb/server/vfs_cache.h +++ b/fs/smb/server/vfs_cache.h @@ -47,7 +47,7 @@ struct stream { }; struct ksmbd_inode { - rwlock_t m_lock; + struct rw_semaphore m_lock; atomic_t m_count; atomic_t op_count; /* opinfo count for streams */ -- cgit v1.2.3-59-g8ed1b From 691aae4f36f9825df6781da4399a1e718951085a Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Wed, 1 May 2024 21:58:15 +0900 Subject: ksmbd: do not grant v2 lease if parent lease key and epoch are not set This patch fix xfstests generic/070 test with smb2 leases = yes. cifs.ko doesn't set parent lease key and epoch in create context v2 lease. ksmbd suppose that parent lease and epoch are vaild if data length is v2 lease context size and handle directory lease using this values. ksmbd should hanle it as v1 lease not v2 lease if parent lease key and epoch are not set in create context v2 lease. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/oplock.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index c2abf109010d..b9d9116fc2b3 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -1201,7 +1201,9 @@ int smb_grant_oplock(struct ksmbd_work *work, int req_op_level, u64 pid, /* Only v2 leases handle the directory */ if (S_ISDIR(file_inode(fp->filp)->i_mode)) { - if (!lctx || lctx->version != 2) + if (!lctx || lctx->version != 2 || + (lctx->flags != SMB2_LEASE_FLAG_PARENT_LEASE_KEY_SET_LE && + !lctx->epoch)) return 0; } @@ -1466,8 +1468,9 @@ void create_lease_buf(u8 *rbuf, struct lease *lease) buf->lcontext.LeaseFlags = lease->flags; buf->lcontext.Epoch = cpu_to_le16(lease->epoch); buf->lcontext.LeaseState = lease->state; - memcpy(buf->lcontext.ParentLeaseKey, lease->parent_lease_key, - SMB2_LEASE_KEY_SIZE); + if (lease->flags == SMB2_LEASE_FLAG_PARENT_LEASE_KEY_SET_LE) + memcpy(buf->lcontext.ParentLeaseKey, lease->parent_lease_key, + SMB2_LEASE_KEY_SIZE); buf->ccontext.DataOffset = cpu_to_le16(offsetof (struct create_lease_v2, lcontext)); buf->ccontext.DataLength = cpu_to_le32(sizeof(struct lease_context_v2)); @@ -1526,8 +1529,9 @@ struct lease_ctx_info *parse_lease_state(void *open_req) lreq->flags = lc->lcontext.LeaseFlags; lreq->epoch = lc->lcontext.Epoch; lreq->duration = lc->lcontext.LeaseDuration; - memcpy(lreq->parent_lease_key, lc->lcontext.ParentLeaseKey, - SMB2_LEASE_KEY_SIZE); + if (lreq->flags == SMB2_LEASE_FLAG_PARENT_LEASE_KEY_SET_LE) + memcpy(lreq->parent_lease_key, lc->lcontext.ParentLeaseKey, + SMB2_LEASE_KEY_SIZE); lreq->version = 2; } else { struct create_lease *lc = (struct create_lease *)cc; -- cgit v1.2.3-59-g8ed1b From 4efaa5acf0a1d2b5947f98abb3acf8bfd966422b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 3 May 2024 13:36:09 -0700 Subject: epoll: be better about file lifetimes epoll can call out to vfs_poll() with a file pointer that may race with the last 'fput()'. That would make f_count go down to zero, and while the ep->mtx locking means that the resulting file pointer tear-down will be blocked until the poll returns, it means that f_count is already dead, and any use of it won't actually get a reference to the file any more: it's dead regardless. Make sure we have a valid ref on the file pointer before we call down to vfs_poll() from the epoll routines. Link: https://lore.kernel.org/lkml/0000000000002d631f0615918f1e@google.com/ Reported-by: syzbot+045b454ab35fd82a35fb@syzkaller.appspotmail.com Reviewed-by: Jens Axboe Signed-off-by: Linus Torvalds --- fs/eventpoll.c | 38 +++++++++++++++++++++++++++++++++++++- 1 file changed, 37 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 882b89edc52a..f53ca4f7fced 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -979,6 +979,34 @@ static __poll_t __ep_eventpoll_poll(struct file *file, poll_table *wait, int dep return res; } +/* + * The ffd.file pointer may be in the process of being torn down due to + * being closed, but we may not have finished eventpoll_release() yet. + * + * Normally, even with the atomic_long_inc_not_zero, the file may have + * been free'd and then gotten re-allocated to something else (since + * files are not RCU-delayed, they are SLAB_TYPESAFE_BY_RCU). + * + * But for epoll, users hold the ep->mtx mutex, and as such any file in + * the process of being free'd will block in eventpoll_release_file() + * and thus the underlying file allocation will not be free'd, and the + * file re-use cannot happen. + * + * For the same reason we can avoid a rcu_read_lock() around the + * operation - 'ffd.file' cannot go away even if the refcount has + * reached zero (but we must still not call out to ->poll() functions + * etc). + */ +static struct file *epi_fget(const struct epitem *epi) +{ + struct file *file; + + file = epi->ffd.file; + if (!atomic_long_inc_not_zero(&file->f_count)) + file = NULL; + return file; +} + /* * Differs from ep_eventpoll_poll() in that internal callers already have * the ep->mtx so we need to start from depth=1, such that mutex_lock_nested() @@ -987,14 +1015,22 @@ static __poll_t __ep_eventpoll_poll(struct file *file, poll_table *wait, int dep static __poll_t ep_item_poll(const struct epitem *epi, poll_table *pt, int depth) { - struct file *file = epi->ffd.file; + struct file *file = epi_fget(epi); __poll_t res; + /* + * We could return EPOLLERR | EPOLLHUP or something, but let's + * treat this more as "file doesn't exist, poll didn't happen". + */ + if (!file) + return 0; + pt->_key = epi->event.events; if (!is_file_epoll(file)) res = vfs_poll(file, pt); else res = __ep_eventpoll_poll(file, pt, depth); + fput(file); return res & epi->event.events; } -- cgit v1.2.3-59-g8ed1b From c88033efe9a391e72ba6b5df4b01d6e628f4e734 Mon Sep 17 00:00:00 2001 From: Peter Xu Date: Mon, 22 Apr 2024 09:33:11 -0400 Subject: mm/userfaultfd: reset ptes when close() for wr-protected ones Userfaultfd unregister includes a step to remove wr-protect bits from all the relevant pgtable entries, but that only covered an explicit UFFDIO_UNREGISTER ioctl, not a close() on the userfaultfd itself. Cover that too. This fixes a WARN trace. The only user visible side effect is the user can observe leftover wr-protect bits even if the user close()ed on an userfaultfd when releasing the last reference of it. However hopefully that should be harmless, and nothing bad should happen even if so. This change is now more important after the recent page-table-check patch we merged in mm-unstable (446dd9ad37d0 ("mm/page_table_check: support userfault wr-protect entries")), as we'll do sanity check on uffd-wp bits without vma context. So it's better if we can 100% guarantee no uffd-wp bit leftovers, to make sure each report will be valid. Link: https://lore.kernel.org/all/000000000000ca4df20616a0fe16@google.com/ Fixes: f369b07c8614 ("mm/uffd: reset write protection when unregister with wp-mode") Analyzed-by: David Hildenbrand Link: https://lkml.kernel.org/r/20240422133311.2987675-1-peterx@redhat.com Reported-by: syzbot+d8426b591c36b21c750e@syzkaller.appspotmail.com Signed-off-by: Peter Xu Reviewed-by: David Hildenbrand Cc: Nadav Amit Cc: Signed-off-by: Andrew Morton --- fs/userfaultfd.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 60dcfafdc11a..292f5fd50104 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -895,6 +895,10 @@ static int userfaultfd_release(struct inode *inode, struct file *file) prev = vma; continue; } + /* Reset ptes for the whole vma range if wr-protected */ + if (userfaultfd_wp(vma)) + uffd_wp_range(vma, vma->vm_start, + vma->vm_end - vma->vm_start, false); new_flags = vma->vm_flags & ~__VM_UFFD_FLAGS; vma = vma_modify_flags_uffd(&vmi, prev, vma, vma->vm_start, vma->vm_end, new_flags, -- cgit v1.2.3-59-g8ed1b From c70dce4982ce1718bf978a35f8e26160b82081f4 Mon Sep 17 00:00:00 2001 From: Ryan Roberts Date: Mon, 29 Apr 2024 12:40:17 +0100 Subject: fs/proc/task_mmu: fix loss of young/dirty bits during pagemap scan make_uffd_wp_pte() was previously doing: pte = ptep_get(ptep); ptep_modify_prot_start(ptep); pte = pte_mkuffd_wp(pte); ptep_modify_prot_commit(ptep, pte); But if another thread accessed or dirtied the pte between the first 2 calls, this could lead to loss of that information. Since ptep_modify_prot_start() gets and clears atomically, the following is the correct pattern and prevents any possible race. Any access after the first call would see an invalid pte and cause a fault: pte = ptep_modify_prot_start(ptep); pte = pte_mkuffd_wp(pte); ptep_modify_prot_commit(ptep, pte); Link: https://lkml.kernel.org/r/20240429114017.182570-1-ryan.roberts@arm.com Fixes: 52526ca7fdb9 ("fs/proc/task_mmu: implement IOCTL to get and optionally clear info about PTEs") Signed-off-by: Ryan Roberts Acked-by: David Hildenbrand Cc: Muhammad Usama Anjum Cc: Peter Xu Cc: Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 23fbab954c20..af4bc1da0c01 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1825,7 +1825,7 @@ static void make_uffd_wp_pte(struct vm_area_struct *vma, pte_t old_pte; old_pte = ptep_modify_prot_start(vma, addr, pte); - ptent = pte_mkuffd_wp(ptent); + ptent = pte_mkuffd_wp(old_pte); ptep_modify_prot_commit(vma, addr, pte, old_pte, ptent); } else if (is_swap_pte(ptent)) { ptent = pte_swp_mkuffd_wp(ptent); -- cgit v1.2.3-59-g8ed1b From 2c7ad9a590d1a99ec59c7d90cef41e2b296944c4 Mon Sep 17 00:00:00 2001 From: Ryan Roberts Date: Mon, 29 Apr 2024 12:41:04 +0100 Subject: fs/proc/task_mmu: fix uffd-wp confusion in pagemap_scan_pmd_entry() pagemap_scan_pmd_entry() checks if uffd-wp is set on each pte to avoid unnecessary if set. However it was previously checking with `pte_uffd_wp(ptep_get(pte))` without first confirming that the pte was present. It is only valid to call pte_uffd_wp() for present ptes. For swap ptes, pte_swp_uffd_wp() must be called because the uffd-wp bit may be kept in a different position, depending on the arch. This was leading to test failures in the pagemap_ioctl mm selftest, when bringing up uffd-wp support on arm64 due to incorrectly interpretting the uffd-wp status of migration entries. Let's fix this by using the correct check based on pte_present(). While we are at it, let's pass the pte to make_uffd_wp_pte() to avoid the pointless extra ptep_get() which can't be optimized out due to READ_ONCE() on many arches. Link: https://lkml.kernel.org/r/20240429114104.182890-1-ryan.roberts@arm.com Fixes: 12f6b01a0bcb ("fs/proc/task_mmu: add fast paths to get/clear PAGE_IS_WRITTEN flag") Closes: https://lore.kernel.org/linux-arm-kernel/ZiuyGXt0XWwRgFh9@x1n/ Signed-off-by: Ryan Roberts Acked-by: David Hildenbrand Reviewed-by: Muhammad Usama Anjum Tested-by: Muhammad Usama Anjum Cc: Peter Xu Cc: Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index af4bc1da0c01..102f48668c35 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1817,10 +1817,8 @@ static unsigned long pagemap_page_category(struct pagemap_scan_private *p, } static void make_uffd_wp_pte(struct vm_area_struct *vma, - unsigned long addr, pte_t *pte) + unsigned long addr, pte_t *pte, pte_t ptent) { - pte_t ptent = ptep_get(pte); - if (pte_present(ptent)) { pte_t old_pte; @@ -2175,9 +2173,12 @@ static int pagemap_scan_pmd_entry(pmd_t *pmd, unsigned long start, if ((p->arg.flags & PM_SCAN_WP_MATCHING) && !p->vec_out) { /* Fast path for performing exclusive WP */ for (addr = start; addr != end; pte++, addr += PAGE_SIZE) { - if (pte_uffd_wp(ptep_get(pte))) + pte_t ptent = ptep_get(pte); + + if ((pte_present(ptent) && pte_uffd_wp(ptent)) || + pte_swp_uffd_wp_any(ptent)) continue; - make_uffd_wp_pte(vma, addr, pte); + make_uffd_wp_pte(vma, addr, pte, ptent); if (!flush_end) start = addr; flush_end = addr + PAGE_SIZE; @@ -2190,8 +2191,10 @@ static int pagemap_scan_pmd_entry(pmd_t *pmd, unsigned long start, p->arg.return_mask == PAGE_IS_WRITTEN) { for (addr = start; addr < end; pte++, addr += PAGE_SIZE) { unsigned long next = addr + PAGE_SIZE; + pte_t ptent = ptep_get(pte); - if (pte_uffd_wp(ptep_get(pte))) + if ((pte_present(ptent) && pte_uffd_wp(ptent)) || + pte_swp_uffd_wp_any(ptent)) continue; ret = pagemap_scan_output(p->cur_vma_category | PAGE_IS_WRITTEN, p, addr, &next); @@ -2199,7 +2202,7 @@ static int pagemap_scan_pmd_entry(pmd_t *pmd, unsigned long start, break; if (~p->arg.flags & PM_SCAN_WP_MATCHING) continue; - make_uffd_wp_pte(vma, addr, pte); + make_uffd_wp_pte(vma, addr, pte, ptent); if (!flush_end) start = addr; flush_end = next; @@ -2208,8 +2211,9 @@ static int pagemap_scan_pmd_entry(pmd_t *pmd, unsigned long start, } for (addr = start; addr != end; pte++, addr += PAGE_SIZE) { + pte_t ptent = ptep_get(pte); unsigned long categories = p->cur_vma_category | - pagemap_page_category(p, vma, addr, ptep_get(pte)); + pagemap_page_category(p, vma, addr, ptent); unsigned long next = addr + PAGE_SIZE; if (!pagemap_scan_is_interesting_page(categories, p)) @@ -2224,7 +2228,7 @@ static int pagemap_scan_pmd_entry(pmd_t *pmd, unsigned long start, if (~categories & PAGE_IS_WRITTEN) continue; - make_uffd_wp_pte(vma, addr, pte); + make_uffd_wp_pte(vma, addr, pte, ptent); if (!flush_end) start = addr; flush_end = next; -- cgit v1.2.3-59-g8ed1b From 3814ec89540d9ce1a92cb4c9a6f9f7a0a343d73d Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 04:17:46 +0100 Subject: buffer: add kernel-doc for block_dirty_folio() Turn the excellent documentation for this function into kernel-doc. Replace 'page' with 'folio' and make a few other minor updates. Link: https://lkml.kernel.org/r/20240416031754.4076917-3-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Pankaj Raghav Tested-by: Randy Dunlap Signed-off-by: Andrew Morton --- fs/buffer.c | 55 +++++++++++++++++++++++++++++++------------------------ 1 file changed, 31 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 4f73d23c2c46..b08526bdcb54 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -687,30 +687,37 @@ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) } EXPORT_SYMBOL(mark_buffer_dirty_inode); -/* - * Add a page to the dirty page list. - * - * It is a sad fact of life that this function is called from several places - * deeply under spinlocking. It may not sleep. - * - * If the page has buffers, the uptodate buffers are set dirty, to preserve - * dirty-state coherency between the page and the buffers. It the page does - * not have buffers then when they are later attached they will all be set - * dirty. - * - * The buffers are dirtied before the page is dirtied. There's a small race - * window in which a writepage caller may see the page cleanness but not the - * buffer dirtiness. That's fine. If this code were to set the page dirty - * before the buffers, a concurrent writepage caller could clear the page dirty - * bit, see a bunch of clean buffers and we'd end up with dirty buffers/clean - * page on the dirty page list. - * - * We use i_private_lock to lock against try_to_free_buffers while using the - * page's buffer list. Also use this to protect against clean buffers being - * added to the page after it was set dirty. - * - * FIXME: may need to call ->reservepage here as well. That's rather up to the - * address_space though. +/** + * block_dirty_folio - Mark a folio as dirty. + * @mapping: The address space containing this folio. + * @folio: The folio to mark dirty. + * + * Filesystems which use buffer_heads can use this function as their + * ->dirty_folio implementation. Some filesystems need to do a little + * work before calling this function. Filesystems which do not use + * buffer_heads should call filemap_dirty_folio() instead. + * + * If the folio has buffers, the uptodate buffers are set dirty, to + * preserve dirty-state coherency between the folio and the buffers. + * Buffers added to a dirty folio are created dirty. + * + * The buffers are dirtied before the folio is dirtied. There's a small + * race window in which writeback may see the folio cleanness but not the + * buffer dirtiness. That's fine. If this code were to set the folio + * dirty before the buffers, writeback could clear the folio dirty flag, + * see a bunch of clean buffers and we'd end up with dirty buffers/clean + * folio on the dirty folio list. + * + * We use i_private_lock to lock against try_to_free_buffers() while + * using the folio's buffer list. This also prevents clean buffers + * being added to the folio after it was set dirty. + * + * Context: May only be called from process context. Does not sleep. + * Caller must ensure that @folio cannot be truncated during this call, + * typically by holding the folio lock or having a page in the folio + * mapped and holding the page table lock. + * + * Return: True if the folio was dirtied; false if it was already dirtied. */ bool block_dirty_folio(struct address_space *mapping, struct folio *folio) { -- cgit v1.2.3-59-g8ed1b From b1888d143203589b71ab31b39d1070737287bc79 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 04:17:47 +0100 Subject: buffer: add kernel-doc for try_to_free_buffers() The documentation for this function has become separated from it over time; move it to the right place and turn it into kernel-doc. Mild editing of the content to make it more about what the function does, and less about how it does it. Link: https://lkml.kernel.org/r/20240416031754.4076917-4-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Pankaj Raghav Tested-by: Randy Dunlap Signed-off-by: Andrew Morton --- fs/buffer.c | 44 ++++++++++++++++++++++++-------------------- 1 file changed, 24 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index b08526bdcb54..0466ed7ed95a 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2868,26 +2868,6 @@ int sync_dirty_buffer(struct buffer_head *bh) } EXPORT_SYMBOL(sync_dirty_buffer); -/* - * try_to_free_buffers() checks if all the buffers on this particular folio - * are unused, and releases them if so. - * - * Exclusion against try_to_free_buffers may be obtained by either - * locking the folio or by holding its mapping's i_private_lock. - * - * If the folio is dirty but all the buffers are clean then we need to - * be sure to mark the folio clean as well. This is because the folio - * may be against a block device, and a later reattachment of buffers - * to a dirty folio will set *all* buffers dirty. Which would corrupt - * filesystem data on the same device. - * - * The same applies to regular filesystem folios: if all the buffers are - * clean then we set the folio clean and proceed. To do that, we require - * total exclusion from block_dirty_folio(). That is obtained with - * i_private_lock. - * - * try_to_free_buffers() is non-blocking. - */ static inline int buffer_busy(struct buffer_head *bh) { return atomic_read(&bh->b_count) | @@ -2921,6 +2901,30 @@ failed: return false; } +/** + * try_to_free_buffers - Release buffers attached to this folio. + * @folio: The folio. + * + * If any buffers are in use (dirty, under writeback, elevated refcount), + * no buffers will be freed. + * + * If the folio is dirty but all the buffers are clean then we need to + * be sure to mark the folio clean as well. This is because the folio + * may be against a block device, and a later reattachment of buffers + * to a dirty folio will set *all* buffers dirty. Which would corrupt + * filesystem data on the same device. + * + * The same applies to regular filesystem folios: if all the buffers are + * clean then we set the folio clean and proceed. To do that, we require + * total exclusion from block_dirty_folio(). That is obtained with + * i_private_lock. + * + * Exclusion against try_to_free_buffers may be obtained by either + * locking the folio or by holding its mapping's i_private_lock. + * + * Context: Process context. @folio must be locked. Will not sleep. + * Return: true if all buffers attached to this folio were freed. + */ bool try_to_free_buffers(struct folio *folio) { struct address_space * const mapping = folio->mapping; -- cgit v1.2.3-59-g8ed1b From 324ecaee46f86c1eaf083fd82eaf997335e70163 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 04:17:48 +0100 Subject: buffer: fix __bread and __bread_gfp kernel-doc The extra indentation confused the kernel-doc parser, so remove it. Fix some other wording while I'm here, and advise the user they need to call brelse() on this buffer. __bread_gfp() isn't used directly by filesystems, but the other wrappers for it don't have documentation, so document it accordingly. Link: https://lkml.kernel.org/r/20240416031754.4076917-5-willy@infradead.org Co-developed-by: Pankaj Raghav Signed-off-by: Pankaj Raghav Signed-off-by: Matthew Wilcox (Oracle) Tested-by: Randy Dunlap Signed-off-by: Andrew Morton --- fs/buffer.c | 35 ++++++++++++++++++++++------------- include/linux/buffer_head.h | 22 +++++++++++++--------- 2 files changed, 35 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 0466ed7ed95a..32ab3eddc44f 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1453,20 +1453,29 @@ void __breadahead(struct block_device *bdev, sector_t block, unsigned size) EXPORT_SYMBOL(__breadahead); /** - * __bread_gfp() - reads a specified block and returns the bh - * @bdev: the block_device to read from - * @block: number of block - * @size: size (in bytes) to read - * @gfp: page allocation flag - * - * Reads a specified block, and returns buffer head that contains it. - * The page cache can be allocated from non-movable area - * not to prevent page migration if you set gfp to zero. - * It returns NULL if the block was unreadable. + * __bread_gfp() - Read a block. + * @bdev: The block device to read from. + * @block: Block number in units of block size. + * @size: The block size of this device in bytes. + * @gfp: Not page allocation flags; see below. + * + * You are not expected to call this function. You should use one of + * sb_bread(), sb_bread_unmovable() or __bread(). + * + * Read a specified block, and return the buffer head that refers to it. + * If @gfp is 0, the memory will be allocated using the block device's + * default GFP flags. If @gfp is __GFP_MOVABLE, the memory may be + * allocated from a movable area. Do not pass in a complete set of + * GFP flags. + * + * The returned buffer head has its refcount increased. The caller should + * call brelse() when it has finished with the buffer. + * + * Context: May sleep waiting for I/O. + * Return: NULL if the block was unreadable. */ -struct buffer_head * -__bread_gfp(struct block_device *bdev, sector_t block, - unsigned size, gfp_t gfp) +struct buffer_head *__bread_gfp(struct block_device *bdev, sector_t block, + unsigned size, gfp_t gfp) { struct buffer_head *bh; diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index d78454a4dd1f..56a1e9c1e71e 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -437,17 +437,21 @@ static inline void bh_readahead_batch(int nr, struct buffer_head *bhs[], } /** - * __bread() - reads a specified block and returns the bh - * @bdev: the block_device to read from - * @block: number of block - * @size: size (in bytes) to read + * __bread() - Read a block. + * @bdev: The block device to read from. + * @block: Block number in units of block size. + * @size: The block size of this device in bytes. * - * Reads a specified block, and returns buffer head that contains it. - * The page cache is allocated from movable area so that it can be migrated. - * It returns NULL if the block was unreadable. + * Read a specified block, and return the buffer head that refers + * to it. The memory is allocated from the movable area so that it can + * be migrated. The returned buffer head has its refcount increased. + * The caller should call brelse() when it has finished with the buffer. + * + * Context: May sleep waiting for I/O. + * Return: NULL if the block was unreadable. */ -static inline struct buffer_head * -__bread(struct block_device *bdev, sector_t block, unsigned size) +static inline struct buffer_head *__bread(struct block_device *bdev, + sector_t block, unsigned size) { return __bread_gfp(bdev, block, size, __GFP_MOVABLE); } -- cgit v1.2.3-59-g8ed1b From 66924fdaf835f5fc6fe78d92a79afcca7d4db7ec Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 04:17:49 +0100 Subject: buffer: add kernel-doc for brelse() and __brelse() Move the documentation for __brelse() to brelse(), format it as kernel-doc and update it from talking about pages to folios. Link: https://lkml.kernel.org/r/20240416031754.4076917-6-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Tested-by: Randy Dunlap Cc: Pankaj Raghav Signed-off-by: Andrew Morton --- fs/buffer.c | 17 ++++++++--------- include/linux/buffer_head.h | 16 ++++++++++++++++ 2 files changed, 24 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 32ab3eddc44f..e5beca3868a7 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1226,17 +1226,16 @@ void mark_buffer_write_io_error(struct buffer_head *bh) } EXPORT_SYMBOL(mark_buffer_write_io_error); -/* - * Decrement a buffer_head's reference count. If all buffers against a page - * have zero reference count, are clean and unlocked, and if the page is clean - * and unlocked then try_to_free_buffers() may strip the buffers from the page - * in preparation for freeing it (sometimes, rarely, buffers are removed from - * a page but it ends up not being freed, and buffers may later be reattached). +/** + * __brelse - Release a buffer. + * @bh: The buffer to release. + * + * This variant of brelse() can be called if @bh is guaranteed to not be NULL. */ -void __brelse(struct buffer_head * buf) +void __brelse(struct buffer_head *bh) { - if (atomic_read(&buf->b_count)) { - put_bh(buf); + if (atomic_read(&bh->b_count)) { + put_bh(bh); return; } WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n"); diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 56a1e9c1e71e..c145817c6ca0 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -303,6 +303,22 @@ static inline void put_bh(struct buffer_head *bh) atomic_dec(&bh->b_count); } +/** + * brelse - Release a buffer. + * @bh: The buffer to release. + * + * Decrement a buffer_head's reference count. If @bh is NULL, this + * function is a no-op. + * + * If all buffers on a folio have zero reference count, are clean + * and unlocked, and if the folio is unlocked and not under writeback + * then try_to_free_buffers() may strip the buffers from the folio in + * preparation for freeing it (sometimes, rarely, buffers are removed + * from a folio but it ends up not being freed, and buffers may later + * be reattached). + * + * Context: Any context. + */ static inline void brelse(struct buffer_head *bh) { if (bh) -- cgit v1.2.3-59-g8ed1b From b73a936f99914e23fbe236f75ecf257923cb06e7 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 04:17:50 +0100 Subject: buffer: add kernel-doc for bforget() and __bforget() Distinguish these functions from brelse() and __brelse(). Link: https://lkml.kernel.org/r/20240416031754.4076917-7-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Tested-by: Randy Dunlap Cc: Pankaj Raghav Signed-off-by: Andrew Morton --- fs/buffer.c | 9 ++++++--- include/linux/buffer_head.h | 10 ++++++++++ 2 files changed, 16 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index e5beca3868a7..60829312787a 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1242,9 +1242,12 @@ void __brelse(struct buffer_head *bh) } EXPORT_SYMBOL(__brelse); -/* - * bforget() is like brelse(), except it discards any - * potentially dirty data. +/** + * __bforget - Discard any dirty data in a buffer. + * @bh: The buffer to forget. + * + * This variant of bforget() can be called if @bh is guaranteed to not + * be NULL. */ void __bforget(struct buffer_head *bh) { diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index c145817c6ca0..a1c0bdd0cca6 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -325,6 +325,16 @@ static inline void brelse(struct buffer_head *bh) __brelse(bh); } +/** + * bforget - Discard any dirty data in a buffer. + * @bh: The buffer to forget. + * + * Call this function instead of brelse() if the data written to a buffer + * no longer needs to be written back. It will clear the buffer's dirty + * flag so writeback of this buffer will be skipped. + * + * Context: Any context. + */ static inline void bforget(struct buffer_head *bh) { if (bh) -- cgit v1.2.3-59-g8ed1b From 0b116ff4dc40ec84ce4bfd451436e66ab2bbc86d Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 04:17:51 +0100 Subject: buffer: improve bdev_getblk documentation Add some more information about the state of the buffer_head returned. Link: https://lkml.kernel.org/r/20240416031754.4076917-8-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Cc: Pankaj Raghav Signed-off-by: Andrew Morton --- fs/buffer.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 60829312787a..ed698caa8834 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1424,6 +1424,11 @@ EXPORT_SYMBOL(__find_get_block); * @size: The size of buffer_heads for this @bdev. * @gfp: The memory allocation flags to use. * + * The returned buffer head has its reference count incremented, but is + * not locked. The caller should call brelse() when it has finished + * with the buffer. The buffer may not be uptodate. If needed, the + * caller can bring it uptodate either by reading it or overwriting it. + * * Return: The buffer head, or NULL if memory could not be allocated. */ struct buffer_head *bdev_getblk(struct block_device *bdev, sector_t block, -- cgit v1.2.3-59-g8ed1b From 88e4e47c12836cf6875f73fca87baaf20c6ca1a2 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Wed, 17 Apr 2024 11:23:12 +0200 Subject: fs/proc/task_mmu: convert pagemap_hugetlb_range() to work on folios Patch series "fs/proc/task_mmu: convert hugetlb functions to work on folis". Let's convert two more functions, getting rid of two more page_mapcount() calls. This patch (of 2): Let's get rid of another page_mapcount() check and simply use folio_likely_mapped_shared(), which is precise for hugetlb folios. While at it, also check for PMD table sharing, like we do in smaps_hugetlb_range(). No functional change intended, except that we would now detect hugetlb folios shared via PMD table sharing correctly. Link: https://lkml.kernel.org/r/20240417092313.753919-1-david@redhat.com Link: https://lkml.kernel.org/r/20240417092313.753919-2-david@redhat.com Signed-off-by: David Hildenbrand Reviewed-by: Oscar Salvador Cc: Muchun Song Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 8ff79bd427ec..cd6e45e0cde8 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1578,12 +1578,13 @@ static int pagemap_hugetlb_range(pte_t *ptep, unsigned long hmask, pte = huge_ptep_get(ptep); if (pte_present(pte)) { - struct page *page = pte_page(pte); + struct folio *folio = page_folio(pte_page(pte)); - if (!PageAnon(page)) + if (!folio_test_anon(folio)) flags |= PM_FILE; - if (page_mapcount(page) == 1) + if (!folio_likely_mapped_shared(folio) && + !hugetlb_pmd_shared(ptep)) flags |= PM_MMAP_EXCLUSIVE; if (huge_pte_uffd_wp(pte)) -- cgit v1.2.3-59-g8ed1b From 6401a2e6900843a77a27873c0529dea68f61193d Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Wed, 17 Apr 2024 11:23:13 +0200 Subject: fs/proc/task_mmu: convert smaps_hugetlb_range() to work on folios Let's get rid of another page_mapcount() check and simply use folio_likely_mapped_shared(), which is precise for hugetlb folios. While at it, use huge_ptep_get() + pte_page() instead of ptep_get() + vm_normal_page(), just like we do in pagemap_hugetlb_range(). No functional change intended. Link: https://lkml.kernel.org/r/20240417092313.753919-3-david@redhat.com Signed-off-by: David Hildenbrand Reviewed-by: Oscar Salvador Cc: Muchun Song Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index cd6e45e0cde8..f4259b7edfde 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -730,19 +730,20 @@ static int smaps_hugetlb_range(pte_t *pte, unsigned long hmask, { struct mem_size_stats *mss = walk->private; struct vm_area_struct *vma = walk->vma; - struct page *page = NULL; - pte_t ptent = ptep_get(pte); + pte_t ptent = huge_ptep_get(pte); + struct folio *folio = NULL; if (pte_present(ptent)) { - page = vm_normal_page(vma, addr, ptent); + folio = page_folio(pte_page(ptent)); } else if (is_swap_pte(ptent)) { swp_entry_t swpent = pte_to_swp_entry(ptent); if (is_pfn_swap_entry(swpent)) - page = pfn_swap_entry_to_page(swpent); + folio = pfn_swap_entry_folio(swpent); } - if (page) { - if (page_mapcount(page) >= 2 || hugetlb_pmd_shared(pte)) + if (folio) { + if (folio_likely_mapped_shared(folio) || + hugetlb_pmd_shared(pte)) mss->shared_hugetlb += huge_page_size(hstate_vma(vma)); else mss->private_hugetlb += huge_page_size(hstate_vma(vma)); -- cgit v1.2.3-59-g8ed1b From 262f014dd7de3747f20ae7d1c3cb5cc68241e770 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 23 Apr 2024 23:55:32 +0100 Subject: fscrypt: convert bh_get_inode_and_lblk_num to use a folio Patch series "Remove page_mapping()". There are only a few users left. Convert them all to either call folio_mapping() or just use folio->mapping directly. This patch (of 6): Remove uses of page->index, page_mapping() and b_page. Saves a call to compound_head(). Link: https://lkml.kernel.org/r/20240423225552.4113447-1-willy@infradead.org Link: https://lkml.kernel.org/r/20240423225552.4113447-2-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Eric Biggers Reviewed-by: David Hildenbrand Cc: Sidhartha Kumar Signed-off-by: Andrew Morton --- fs/crypto/inline_crypt.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/crypto/inline_crypt.c b/fs/crypto/inline_crypt.c index b4002aea7cdb..40de69860dcf 100644 --- a/fs/crypto/inline_crypt.c +++ b/fs/crypto/inline_crypt.c @@ -284,7 +284,7 @@ static bool bh_get_inode_and_lblk_num(const struct buffer_head *bh, const struct inode **inode_ret, u64 *lblk_num_ret) { - struct page *page = bh->b_page; + struct folio *folio = bh->b_folio; const struct address_space *mapping; const struct inode *inode; @@ -292,13 +292,13 @@ static bool bh_get_inode_and_lblk_num(const struct buffer_head *bh, * The ext4 journal (jbd2) can submit a buffer_head it directly created * for a non-pagecache page. fscrypt doesn't care about these. */ - mapping = page_mapping(page); + mapping = folio_mapping(folio); if (!mapping) return false; inode = mapping->host; *inode_ret = inode; - *lblk_num_ret = ((u64)page->index << (PAGE_SHIFT - inode->i_blkbits)) + + *lblk_num_ret = ((u64)folio->index << (PAGE_SHIFT - inode->i_blkbits)) + (bh_offset(bh) >> inode->i_blkbits); return true; } -- cgit v1.2.3-59-g8ed1b From 196ad49cd62614979903c3a6033afd2299b29441 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 23 Apr 2024 23:55:33 +0100 Subject: f2fs: convert f2fs_clear_page_cache_dirty_tag to use a folio Removes uses of page_mapping() and page_index(). Link: https://lkml.kernel.org/r/20240423225552.4113447-3-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Hildenbrand Cc: Eric Biggers Cc: Sidhartha Kumar Signed-off-by: Andrew Morton --- fs/f2fs/data.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index d9494b5fc7c1..961e6ff77c72 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -4082,11 +4082,12 @@ const struct address_space_operations f2fs_dblock_aops = { void f2fs_clear_page_cache_dirty_tag(struct page *page) { - struct address_space *mapping = page_mapping(page); + struct folio *folio = page_folio(page); + struct address_space *mapping = folio->mapping; unsigned long flags; xa_lock_irqsave(&mapping->i_pages, flags); - __xa_clear_mark(&mapping->i_pages, page_index(page), + __xa_clear_mark(&mapping->i_pages, folio->index, PAGECACHE_TAG_DIRTY); xa_unlock_irqrestore(&mapping->i_pages, flags); } -- cgit v1.2.3-59-g8ed1b From e0ffb29bc54d86b9ab10ebafc66eb1b7229e0cd7 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Thu, 25 Apr 2024 05:00:55 +0100 Subject: mm: simplify thp_vma_allowable_order Combine the three boolean arguments into one flags argument for readability. Signed-off-by: Matthew Wilcox (Oracle) Cc: David Hildenbrand Cc: Kefeng Wang Cc: Ryan Roberts Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 4 ++-- include/linux/huge_mm.h | 29 +++++++++++++++-------------- mm/huge_memory.c | 7 +++++-- mm/khugepaged.c | 16 +++++++--------- mm/memory.c | 10 ++++++---- 5 files changed, 35 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index f4259b7edfde..81fbecfe5ff6 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -871,8 +871,8 @@ static int show_smap(struct seq_file *m, void *v) __show_smap(m, &mss, false); seq_printf(m, "THPeligible: %8u\n", - !!thp_vma_allowable_orders(vma, vma->vm_flags, true, false, - true, THP_ORDERS_ALL)); + !!thp_vma_allowable_orders(vma, vma->vm_flags, + TVA_SMAPS | TVA_ENFORCE_SYSFS, THP_ORDERS_ALL)); if (arch_pkeys_enabled()) seq_printf(m, "ProtectionKey: %8u\n", vma_pkey(vma)); diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 7cd07b83a3d0..c8d3ec116e29 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -81,8 +81,12 @@ extern struct kobj_attribute shmem_enabled_attr; */ #define THP_ORDERS_ALL (THP_ORDERS_ALL_ANON | THP_ORDERS_ALL_FILE) -#define thp_vma_allowable_order(vma, vm_flags, smaps, in_pf, enforce_sysfs, order) \ - (!!thp_vma_allowable_orders(vma, vm_flags, smaps, in_pf, enforce_sysfs, BIT(order))) +#define TVA_SMAPS (1 << 0) /* Will be used for procfs */ +#define TVA_IN_PF (1 << 1) /* Page fault handler */ +#define TVA_ENFORCE_SYSFS (1 << 2) /* Obey sysfs configuration */ + +#define thp_vma_allowable_order(vma, vm_flags, tva_flags, order) \ + (!!thp_vma_allowable_orders(vma, vm_flags, tva_flags, BIT(order))) #ifdef CONFIG_PGTABLE_HAS_HUGE_LEAVES #define HPAGE_PMD_SHIFT PMD_SHIFT @@ -218,17 +222,15 @@ static inline bool file_thp_enabled(struct vm_area_struct *vma) } unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma, - unsigned long vm_flags, bool smaps, - bool in_pf, bool enforce_sysfs, + unsigned long vm_flags, + unsigned long tva_flags, unsigned long orders); /** * thp_vma_allowable_orders - determine hugepage orders that are allowed for vma * @vma: the vm area to check * @vm_flags: use these vm_flags instead of vma->vm_flags - * @smaps: whether answer will be used for smaps file - * @in_pf: whether answer will be used by page fault handler - * @enforce_sysfs: whether sysfs config should be taken into account + * @tva_flags: Which TVA flags to honour * @orders: bitfield of all orders to consider * * Calculates the intersection of the requested hugepage orders and the allowed @@ -241,12 +243,12 @@ unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma, */ static inline unsigned long thp_vma_allowable_orders(struct vm_area_struct *vma, - unsigned long vm_flags, bool smaps, - bool in_pf, bool enforce_sysfs, + unsigned long vm_flags, + unsigned long tva_flags, unsigned long orders) { /* Optimization to check if required orders are enabled early. */ - if (enforce_sysfs && vma_is_anonymous(vma)) { + if ((tva_flags & TVA_ENFORCE_SYSFS) && vma_is_anonymous(vma)) { unsigned long mask = READ_ONCE(huge_anon_orders_always); if (vm_flags & VM_HUGEPAGE) @@ -260,8 +262,7 @@ unsigned long thp_vma_allowable_orders(struct vm_area_struct *vma, return 0; } - return __thp_vma_allowable_orders(vma, vm_flags, smaps, in_pf, - enforce_sysfs, orders); + return __thp_vma_allowable_orders(vma, vm_flags, tva_flags, orders); } enum mthp_stat_item { @@ -428,8 +429,8 @@ static inline unsigned long thp_vma_suitable_orders(struct vm_area_struct *vma, } static inline unsigned long thp_vma_allowable_orders(struct vm_area_struct *vma, - unsigned long vm_flags, bool smaps, - bool in_pf, bool enforce_sysfs, + unsigned long vm_flags, + unsigned long tva_flags, unsigned long orders) { return 0; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index eabf088cb444..aaa327bd5155 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -81,10 +81,13 @@ unsigned long huge_anon_orders_madvise __read_mostly; unsigned long huge_anon_orders_inherit __read_mostly; unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma, - unsigned long vm_flags, bool smaps, - bool in_pf, bool enforce_sysfs, + unsigned long vm_flags, + unsigned long tva_flags, unsigned long orders) { + bool smaps = tva_flags & TVA_SMAPS; + bool in_pf = tva_flags & TVA_IN_PF; + bool enforce_sysfs = tva_flags & TVA_ENFORCE_SYSFS; /* Check the intersection of requested and supported orders. */ orders &= vma_is_anonymous(vma) ? THP_ORDERS_ALL_ANON : THP_ORDERS_ALL_FILE; diff --git a/mm/khugepaged.c b/mm/khugepaged.c index cf518fc44098..774a97e6e2da 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -453,7 +453,7 @@ void khugepaged_enter_vma(struct vm_area_struct *vma, { if (!test_bit(MMF_VM_HUGEPAGE, &vma->vm_mm->flags) && hugepage_flags_enabled()) { - if (thp_vma_allowable_order(vma, vm_flags, false, false, true, + if (thp_vma_allowable_order(vma, vm_flags, TVA_ENFORCE_SYSFS, PMD_ORDER)) __khugepaged_enter(vma->vm_mm); } @@ -900,6 +900,7 @@ static int hugepage_vma_revalidate(struct mm_struct *mm, unsigned long address, struct collapse_control *cc) { struct vm_area_struct *vma; + unsigned long tva_flags = cc->is_khugepaged ? TVA_ENFORCE_SYSFS : 0; if (unlikely(hpage_collapse_test_exit_or_disable(mm))) return SCAN_ANY_PROCESS; @@ -910,8 +911,7 @@ static int hugepage_vma_revalidate(struct mm_struct *mm, unsigned long address, if (!thp_vma_suitable_order(vma, address, PMD_ORDER)) return SCAN_ADDRESS_RANGE; - if (!thp_vma_allowable_order(vma, vma->vm_flags, false, false, - cc->is_khugepaged, PMD_ORDER)) + if (!thp_vma_allowable_order(vma, vma->vm_flags, tva_flags, PMD_ORDER)) return SCAN_VMA_CHECK; /* * Anon VMA expected, the address may be unmapped then @@ -1501,8 +1501,7 @@ int collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr, * and map it by a PMD, regardless of sysfs THP settings. As such, let's * analogously elide sysfs THP settings here. */ - if (!thp_vma_allowable_order(vma, vma->vm_flags, false, false, false, - PMD_ORDER)) + if (!thp_vma_allowable_order(vma, vma->vm_flags, 0, PMD_ORDER)) return SCAN_VMA_CHECK; /* Keep pmd pgtable for uffd-wp; see comment in retract_page_tables() */ @@ -2363,8 +2362,8 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result, progress++; break; } - if (!thp_vma_allowable_order(vma, vma->vm_flags, false, false, - true, PMD_ORDER)) { + if (!thp_vma_allowable_order(vma, vma->vm_flags, + TVA_ENFORCE_SYSFS, PMD_ORDER)) { skip: progress++; continue; @@ -2701,8 +2700,7 @@ int madvise_collapse(struct vm_area_struct *vma, struct vm_area_struct **prev, *prev = vma; - if (!thp_vma_allowable_order(vma, vma->vm_flags, false, false, false, - PMD_ORDER)) + if (!thp_vma_allowable_order(vma, vma->vm_flags, 0, PMD_ORDER)) return -EINVAL; cc = kmalloc(sizeof(*cc), GFP_KERNEL); diff --git a/mm/memory.c b/mm/memory.c index c0bd4e0d5e7a..79d851be8ab2 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4334,8 +4334,8 @@ static struct folio *alloc_anon_folio(struct vm_fault *vmf) * for this vma. Then filter out the orders that can't be allocated over * the faulting address and still be fully contained in the vma. */ - orders = thp_vma_allowable_orders(vma, vma->vm_flags, false, true, true, - BIT(PMD_ORDER) - 1); + orders = thp_vma_allowable_orders(vma, vma->vm_flags, + TVA_IN_PF | TVA_ENFORCE_SYSFS, BIT(PMD_ORDER) - 1); orders = thp_vma_suitable_orders(vma, vmf->address, orders); if (!orders) @@ -5438,7 +5438,8 @@ static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma, return VM_FAULT_OOM; retry_pud: if (pud_none(*vmf.pud) && - thp_vma_allowable_order(vma, vm_flags, false, true, true, PUD_ORDER)) { + thp_vma_allowable_order(vma, vm_flags, + TVA_IN_PF | TVA_ENFORCE_SYSFS, PUD_ORDER)) { ret = create_huge_pud(&vmf); if (!(ret & VM_FAULT_FALLBACK)) return ret; @@ -5472,7 +5473,8 @@ retry_pud: goto retry_pud; if (pmd_none(*vmf.pmd) && - thp_vma_allowable_order(vma, vm_flags, false, true, true, PMD_ORDER)) { + thp_vma_allowable_order(vma, vm_flags, + TVA_IN_PF | TVA_ENFORCE_SYSFS, PMD_ORDER)) { ret = create_huge_pmd(&vmf); if (!(ret & VM_FAULT_FALLBACK)) return ret; -- cgit v1.2.3-59-g8ed1b From 23df17788c6212809848f836b10c4f85b16843a5 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 8 Apr 2024 12:09:15 +1000 Subject: nfsd: perform all find_openstateowner_str calls in the one place. Currently find_openstateowner_str look ups are done both in nfsd4_process_open1() and alloc_init_open_stateowner() - the latter possibly being a surprise based on its name. It would be easier to follow, and more conformant to common patterns, if the lookup was all in the one place. So replace alloc_init_open_stateowner() with find_or_alloc_open_stateowner() and use the latter in nfsd4_process_open1() without any calls to find_openstateowner_str(). This means all finds are find_openstateowner_str_locked() and find_openstateowner_str() is no longer needed. So discard find_openstateowner_str() and rename find_openstateowner_str_locked() to find_openstateowner_str(). Reviewed-by: Jeff Layton Signed-off-by: NeilBrown Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 93 +++++++++++++++++++++++------------------------------ 1 file changed, 40 insertions(+), 53 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 84d4093ca713..65225b3d1d93 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -541,7 +541,7 @@ same_owner_str(struct nfs4_stateowner *sop, struct xdr_netobj *owner) } static struct nfs4_openowner * -find_openstateowner_str_locked(unsigned int hashval, struct nfsd4_open *open, +find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open, struct nfs4_client *clp) { struct nfs4_stateowner *so; @@ -558,18 +558,6 @@ find_openstateowner_str_locked(unsigned int hashval, struct nfsd4_open *open, return NULL; } -static struct nfs4_openowner * -find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open, - struct nfs4_client *clp) -{ - struct nfs4_openowner *oo; - - spin_lock(&clp->cl_lock); - oo = find_openstateowner_str_locked(hashval, open, clp); - spin_unlock(&clp->cl_lock); - return oo; -} - static inline u32 opaque_hashval(const void *ptr, int nbytes) { @@ -4866,34 +4854,46 @@ nfsd4_find_and_lock_existing_open(struct nfs4_file *fp, struct nfsd4_open *open) } static struct nfs4_openowner * -alloc_init_open_stateowner(unsigned int strhashval, struct nfsd4_open *open, - struct nfsd4_compound_state *cstate) +find_or_alloc_open_stateowner(unsigned int strhashval, struct nfsd4_open *open, + struct nfsd4_compound_state *cstate) { struct nfs4_client *clp = cstate->clp; - struct nfs4_openowner *oo, *ret; + struct nfs4_openowner *oo, *new = NULL; - oo = alloc_stateowner(openowner_slab, &open->op_owner, clp); - if (!oo) - return NULL; - oo->oo_owner.so_ops = &openowner_ops; - oo->oo_owner.so_is_open_owner = 1; - oo->oo_owner.so_seqid = open->op_seqid; - oo->oo_flags = 0; - if (nfsd4_has_session(cstate)) - oo->oo_flags |= NFS4_OO_CONFIRMED; - oo->oo_time = 0; - oo->oo_last_closed_stid = NULL; - INIT_LIST_HEAD(&oo->oo_close_lru); +retry: spin_lock(&clp->cl_lock); - ret = find_openstateowner_str_locked(strhashval, open, clp); - if (ret == NULL) { - hash_openowner(oo, clp, strhashval); - ret = oo; - } else - nfs4_free_stateowner(&oo->oo_owner); - + oo = find_openstateowner_str(strhashval, open, clp); + if (!oo && new) { + hash_openowner(new, clp, strhashval); + spin_unlock(&clp->cl_lock); + return new; + } spin_unlock(&clp->cl_lock); - return ret; + + if (oo && !(oo->oo_flags & NFS4_OO_CONFIRMED)) { + /* Replace unconfirmed owners without checking for replay. */ + release_openowner(oo); + oo = NULL; + } + if (oo) { + if (new) + nfs4_free_stateowner(&new->oo_owner); + return oo; + } + + new = alloc_stateowner(openowner_slab, &open->op_owner, clp); + if (!new) + return NULL; + new->oo_owner.so_ops = &openowner_ops; + new->oo_owner.so_is_open_owner = 1; + new->oo_owner.so_seqid = open->op_seqid; + new->oo_flags = 0; + if (nfsd4_has_session(cstate)) + new->oo_flags |= NFS4_OO_CONFIRMED; + new->oo_time = 0; + new->oo_last_closed_stid = NULL; + INIT_LIST_HEAD(&new->oo_close_lru); + goto retry; } static struct nfs4_ol_stateid * @@ -5342,27 +5342,14 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate, clp = cstate->clp; strhashval = ownerstr_hashval(&open->op_owner); - oo = find_openstateowner_str(strhashval, open, clp); + oo = find_or_alloc_open_stateowner(strhashval, open, cstate); open->op_openowner = oo; - if (!oo) { - goto new_owner; - } - if (!(oo->oo_flags & NFS4_OO_CONFIRMED)) { - /* Replace unconfirmed owners without checking for replay. */ - release_openowner(oo); - open->op_openowner = NULL; - goto new_owner; - } + if (!oo) + return nfserr_jukebox; status = nfsd4_check_seqid(cstate, &oo->oo_owner, open->op_seqid); if (status) return status; - goto alloc_stateid; -new_owner: - oo = alloc_init_open_stateowner(strhashval, open, cstate); - if (oo == NULL) - return nfserr_jukebox; - open->op_openowner = oo; -alloc_stateid: + open->op_stp = nfs4_alloc_open_stateid(clp); if (!open->op_stp) return nfserr_jukebox; -- cgit v1.2.3-59-g8ed1b From b3f03739ca8cd6058a5d8754ea1354bc21fa0f2f Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 8 Apr 2024 12:09:16 +1000 Subject: nfsd: move nfsd4_cstate_assign_replay() earlier in open handling. Rather than taking the rp_mutex (via nfsd4_cstate_assign_replay) in nfsd4_cleanup_open_state() (which seems counter-intuitive), take it and assign rp_owner as soon as possible - in nfsd4_process_open1(). This will support a future change when nfsd4_cstate_assign_replay() might fail. Reviewed-by: Jeff Layton Signed-off-by: NeilBrown Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 65225b3d1d93..43bdff49aa03 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -5346,6 +5346,7 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate, open->op_openowner = oo; if (!oo) return nfserr_jukebox; + nfsd4_cstate_assign_replay(cstate, &oo->oo_owner); status = nfsd4_check_seqid(cstate, &oo->oo_owner, open->op_seqid); if (status) return status; @@ -6120,12 +6121,8 @@ out: void nfsd4_cleanup_open_state(struct nfsd4_compound_state *cstate, struct nfsd4_open *open) { - if (open->op_openowner) { - struct nfs4_stateowner *so = &open->op_openowner->oo_owner; - - nfsd4_cstate_assign_replay(cstate, so); - nfs4_put_stateowner(so); - } + if (open->op_openowner) + nfs4_put_stateowner(&open->op_openowner->oo_owner); if (open->op_file) kmem_cache_free(file_slab, open->op_file); if (open->op_stp) -- cgit v1.2.3-59-g8ed1b From eec7620800081e27dbf8019ac2e66259f0d5bf6f Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 8 Apr 2024 12:09:17 +1000 Subject: nfsd: replace rp_mutex to avoid deadlock in move_to_close_lru() move_to_close_lru() waits for sc_count to become zero while holding rp_mutex. This can deadlock if another thread holds a reference and is waiting for rp_mutex. By the time we get to move_to_close_lru() the openowner is unhashed and cannot be found any more. So code waiting for the mutex can safely retry the lookup if move_to_close_lru() has started. So change rp_mutex to an atomic_t with three states: RP_UNLOCK - state is still hashed, not locked for reply RP_LOCKED - state is still hashed, is locked for reply RP_UNHASHED - state is not hashed, no code can get a lock. Use wait_var_event() to wait for either a lock, or for the owner to be unhashed. In the latter case, retry the lookup. Reviewed-by: Jeff Layton Signed-off-by: NeilBrown Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 38 +++++++++++++++++++++++++++++++------- fs/nfsd/state.h | 2 +- 2 files changed, 32 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 43bdff49aa03..e91e741719a0 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -4650,21 +4650,32 @@ nfsd4_init_leases_net(struct nfsd_net *nn) atomic_set(&nn->nfsd_courtesy_clients, 0); } +enum rp_lock { + RP_UNLOCKED, + RP_LOCKED, + RP_UNHASHED, +}; + static void init_nfs4_replay(struct nfs4_replay *rp) { rp->rp_status = nfserr_serverfault; rp->rp_buflen = 0; rp->rp_buf = rp->rp_ibuf; - mutex_init(&rp->rp_mutex); + atomic_set(&rp->rp_locked, RP_UNLOCKED); } -static void nfsd4_cstate_assign_replay(struct nfsd4_compound_state *cstate, - struct nfs4_stateowner *so) +static int nfsd4_cstate_assign_replay(struct nfsd4_compound_state *cstate, + struct nfs4_stateowner *so) { if (!nfsd4_has_session(cstate)) { - mutex_lock(&so->so_replay.rp_mutex); + wait_var_event(&so->so_replay.rp_locked, + atomic_cmpxchg(&so->so_replay.rp_locked, + RP_UNLOCKED, RP_LOCKED) != RP_LOCKED); + if (atomic_read(&so->so_replay.rp_locked) == RP_UNHASHED) + return -EAGAIN; cstate->replay_owner = nfs4_get_stateowner(so); } + return 0; } void nfsd4_cstate_clear_replay(struct nfsd4_compound_state *cstate) @@ -4673,7 +4684,8 @@ void nfsd4_cstate_clear_replay(struct nfsd4_compound_state *cstate) if (so != NULL) { cstate->replay_owner = NULL; - mutex_unlock(&so->so_replay.rp_mutex); + atomic_set(&so->so_replay.rp_locked, RP_UNLOCKED); + wake_up_var(&so->so_replay.rp_locked); nfs4_put_stateowner(so); } } @@ -4969,7 +4981,11 @@ move_to_close_lru(struct nfs4_ol_stateid *s, struct net *net) * Wait for the refcount to drop to 2. Since it has been unhashed, * there should be no danger of the refcount going back up again at * this point. + * Some threads with a reference might be waiting for rp_locked, + * so tell them to stop waiting. */ + atomic_set(&oo->oo_owner.so_replay.rp_locked, RP_UNHASHED); + wake_up_var(&oo->oo_owner.so_replay.rp_locked); wait_event(close_wq, refcount_read(&s->st_stid.sc_count) == 2); release_all_access(s); @@ -5342,11 +5358,15 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate, clp = cstate->clp; strhashval = ownerstr_hashval(&open->op_owner); +retry: oo = find_or_alloc_open_stateowner(strhashval, open, cstate); open->op_openowner = oo; if (!oo) return nfserr_jukebox; - nfsd4_cstate_assign_replay(cstate, &oo->oo_owner); + if (nfsd4_cstate_assign_replay(cstate, &oo->oo_owner) == -EAGAIN) { + nfs4_put_stateowner(&oo->oo_owner); + goto retry; + } status = nfsd4_check_seqid(cstate, &oo->oo_owner, open->op_seqid); if (status) return status; @@ -7186,12 +7206,16 @@ nfs4_preprocess_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid, trace_nfsd_preprocess(seqid, stateid); *stpp = NULL; +retry: status = nfsd4_lookup_stateid(cstate, stateid, typemask, statusmask, &s, nn); if (status) return status; stp = openlockstateid(s); - nfsd4_cstate_assign_replay(cstate, stp->st_stateowner); + if (nfsd4_cstate_assign_replay(cstate, stp->st_stateowner) == -EAGAIN) { + nfs4_put_stateowner(stp->st_stateowner); + goto retry; + } status = nfs4_seqid_op_checks(cstate, stateid, seqid, stp); if (!status) diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index 2ed0fcf879fd..a6261754deed 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -486,7 +486,7 @@ struct nfs4_replay { unsigned int rp_buflen; char *rp_buf; struct knfsd_fh rp_openfh; - struct mutex rp_mutex; + atomic_t rp_locked; char rp_ibuf[NFSD4_REPLAY_ISIZE]; }; -- cgit v1.2.3-59-g8ed1b From 56c35f43eef013579c76c007ba1f386d8c2cac14 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 8 Apr 2024 12:09:18 +1000 Subject: nfsd: drop st_mutex before calling move_to_close_lru() move_to_close_lru() is currently called with ->st_mutex held. This can lead to a deadlock as move_to_close_lru() waits for sc_count to drop to 2, and some threads holding a reference might be waiting for the mutex. These references will never be dropped so sc_count will never reach 2. There can be no harm in dropping ->st_mutex before move_to_close_lru() because the only place that takes the mutex is nfsd4_lock_ol_stateid(), and it quickly aborts if sc_type is NFS4_CLOSED_STID, which it will be before move_to_close_lru() is called. See also https://lore.kernel.org/lkml/4dd1fe21e11344e5969bb112e954affb@jd.com/T/ where this problem was raised but not successfully resolved. Reviewed-by: Jeff Layton Signed-off-by: NeilBrown Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index e91e741719a0..df07202cc53e 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -7357,7 +7357,7 @@ out: return status; } -static void nfsd4_close_open_stateid(struct nfs4_ol_stateid *s) +static bool nfsd4_close_open_stateid(struct nfs4_ol_stateid *s) { struct nfs4_client *clp = s->st_stid.sc_client; bool unhashed; @@ -7374,11 +7374,11 @@ static void nfsd4_close_open_stateid(struct nfs4_ol_stateid *s) list_for_each_entry(stp, &reaplist, st_locks) nfs4_free_cpntf_statelist(clp->net, &stp->st_stid); free_ol_stateid_reaplist(&reaplist); + return false; } else { spin_unlock(&clp->cl_lock); free_ol_stateid_reaplist(&reaplist); - if (unhashed) - move_to_close_lru(s, clp->net); + return unhashed; } } @@ -7394,6 +7394,7 @@ nfsd4_close(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfs4_ol_stateid *stp; struct net *net = SVC_NET(rqstp); struct nfsd_net *nn = net_generic(net, nfsd_net_id); + bool need_move_to_close_list; dprintk("NFSD: nfsd4_close on file %pd\n", cstate->current_fh.fh_dentry); @@ -7418,8 +7419,10 @@ nfsd4_close(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, */ nfs4_inc_and_copy_stateid(&close->cl_stateid, &stp->st_stid); - nfsd4_close_open_stateid(stp); + need_move_to_close_list = nfsd4_close_open_stateid(stp); mutex_unlock(&stp->st_mutex); + if (need_move_to_close_list) + move_to_close_lru(stp, net); /* v4.1+ suggests that we send a special stateid in here, since the * clients should just ignore this anyway. Since this is not useful -- cgit v1.2.3-59-g8ed1b From 38f080f3cd19fbd87eb7f6c4f6c236c1c9df6fba Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 1 Apr 2024 16:05:39 -0400 Subject: NFSD: Move callback_wq into struct nfs4_client Commit 883820366747 ("nfsd: update workqueue creation") made the callback_wq single-threaded, presumably to protect modifications of cl_cb_client. See documenting comment for nfsd4_process_cb_update(). However, cl_cb_client is per-lease. There's no other reason that all callback operations need to be dispatched via a single thread. The single threading here means all client callbacks can be blocked by a problem with one client. Change the NFSv4 callback client so it serializes per-lease instead of serializing all NFSv4 callback operations on the server. Reported-by: Dai Ngo Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs4callback.c | 31 +++++++++---------------------- fs/nfsd/nfs4state.c | 14 +++++++------- fs/nfsd/state.h | 4 ++-- 3 files changed, 18 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index e88aca0c6e8e..d756f443fc44 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -978,12 +978,12 @@ static int max_cb_time(struct net *net) return max(((u32)nn->nfsd4_lease)/10, 1u) * HZ; } -static struct workqueue_struct *callback_wq; - static bool nfsd4_queue_cb(struct nfsd4_callback *cb) { - trace_nfsd_cb_queue(cb->cb_clp, cb); - return queue_work(callback_wq, &cb->cb_work); + struct nfs4_client *clp = cb->cb_clp; + + trace_nfsd_cb_queue(clp, cb); + return queue_work(clp->cl_callback_wq, &cb->cb_work); } static void nfsd41_cb_inflight_begin(struct nfs4_client *clp) @@ -1153,7 +1153,7 @@ void nfsd4_probe_callback(struct nfs4_client *clp) void nfsd4_probe_callback_sync(struct nfs4_client *clp) { nfsd4_probe_callback(clp); - flush_workqueue(callback_wq); + flush_workqueue(clp->cl_callback_wq); } void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn *conn) @@ -1372,19 +1372,6 @@ static const struct rpc_call_ops nfsd4_cb_ops = { .rpc_release = nfsd4_cb_release, }; -int nfsd4_create_callback_queue(void) -{ - callback_wq = alloc_ordered_workqueue("nfsd4_callbacks", 0); - if (!callback_wq) - return -ENOMEM; - return 0; -} - -void nfsd4_destroy_callback_queue(void) -{ - destroy_workqueue(callback_wq); -} - /* must be called under the state lock */ void nfsd4_shutdown_callback(struct nfs4_client *clp) { @@ -1398,7 +1385,7 @@ void nfsd4_shutdown_callback(struct nfs4_client *clp) * client, destroy the rpc client, and stop: */ nfsd4_run_cb(&clp->cl_cb_null); - flush_workqueue(callback_wq); + flush_workqueue(clp->cl_callback_wq); nfsd41_cb_inflight_wait_complete(clp); } @@ -1420,9 +1407,9 @@ static struct nfsd4_conn * __nfsd4_find_backchannel(struct nfs4_client *clp) /* * Note there isn't a lot of locking in this code; instead we depend on - * the fact that it is run from the callback_wq, which won't run two - * work items at once. So, for example, callback_wq handles all access - * of cl_cb_client and all calls to rpc_create or rpc_shutdown_client. + * the fact that it is run from clp->cl_callback_wq, which won't run two + * work items at once. So, for example, clp->cl_callback_wq handles all + * access of cl_cb_client and all calls to rpc_create or rpc_shutdown_client. */ static void nfsd4_process_cb_update(struct nfsd4_callback *cb) { diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index df07202cc53e..5ae70fb1d3a2 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2233,6 +2233,10 @@ static struct nfs4_client *alloc_client(struct xdr_netobj name, GFP_KERNEL); if (!clp->cl_ownerstr_hashtbl) goto err_no_hashtbl; + clp->cl_callback_wq = alloc_ordered_workqueue("nfsd4_callbacks", 0); + if (!clp->cl_callback_wq) + goto err_no_callback_wq; + for (i = 0; i < OWNER_HASH_SIZE; i++) INIT_LIST_HEAD(&clp->cl_ownerstr_hashtbl[i]); INIT_LIST_HEAD(&clp->cl_sessions); @@ -2255,6 +2259,8 @@ static struct nfs4_client *alloc_client(struct xdr_netobj name, spin_lock_init(&clp->cl_lock); rpc_init_wait_queue(&clp->cl_cb_waitq, "Backchannel slot table"); return clp; +err_no_callback_wq: + kfree(clp->cl_ownerstr_hashtbl); err_no_hashtbl: kfree(clp->cl_name.data); err_no_name: @@ -2268,6 +2274,7 @@ static void __free_client(struct kref *k) struct nfs4_client *clp = container_of(c, struct nfs4_client, cl_nfsdfs); free_svc_cred(&clp->cl_cred); + destroy_workqueue(clp->cl_callback_wq); kfree(clp->cl_ownerstr_hashtbl); kfree(clp->cl_name.data); kfree(clp->cl_nii_domain.data); @@ -8636,12 +8643,6 @@ nfs4_state_start(void) if (ret) return ret; - ret = nfsd4_create_callback_queue(); - if (ret) { - rhltable_destroy(&nfs4_file_rhltable); - return ret; - } - set_max_delegations(); return 0; } @@ -8682,7 +8683,6 @@ nfs4_state_shutdown_net(struct net *net) void nfs4_state_shutdown(void) { - nfsd4_destroy_callback_queue(); rhltable_destroy(&nfs4_file_rhltable); } diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index a6261754deed..ffc217099d19 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -408,6 +408,8 @@ struct nfs4_client { 1 << NFSD4_CLIENT_CB_KILL) #define NFSD4_CLIENT_CB_RECALL_ANY (6) unsigned long cl_flags; + + struct workqueue_struct *cl_callback_wq; const struct cred *cl_cb_cred; struct rpc_clnt *cl_cb_client; u32 cl_cb_ident; @@ -735,8 +737,6 @@ extern void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn * extern void nfsd4_init_cb(struct nfsd4_callback *cb, struct nfs4_client *clp, const struct nfsd4_callback_ops *ops, enum nfsd4_cb_op op); extern bool nfsd4_run_cb(struct nfsd4_callback *cb); -extern int nfsd4_create_callback_queue(void); -extern void nfsd4_destroy_callback_queue(void); extern void nfsd4_shutdown_callback(struct nfs4_client *); extern void nfsd4_shutdown_copy(struct nfs4_client *clp); extern struct nfs4_client_reclaim *nfs4_client_to_reclaim(struct xdr_netobj name, -- cgit v1.2.3-59-g8ed1b From 33a1e6ea73e5f1defe6706f006c0930a82ebdaaa Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Sat, 23 Mar 2024 09:25:54 -0400 Subject: nfsd: trivial GET_DIR_DELEGATION support This adds basic infrastructure for handing GET_DIR_DELEGATION calls from clients, including the decoders and encoders. For now, it always just returns NFS4_OK + GDD4_UNAVAIL. Eventually clients may start sending this operation, and it's better if we can return GDD4_UNAVAIL instead of having to abort the whole compound. Signed-off-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs4proc.c | 41 ++++++++++++++++++++++++++++ fs/nfsd/nfs4xdr.c | 76 ++++++++++++++++++++++++++++++++++++++++++++++++++-- fs/nfsd/xdr4.h | 19 +++++++++++++ include/linux/nfs4.h | 6 +++++ 4 files changed, 140 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 2927b1263f08..3dc173b29803 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -2154,6 +2154,29 @@ nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, return status == nfserr_same ? nfs_ok : status; } +static __be32 +nfsd4_get_dir_delegation(struct svc_rqst *rqstp, + struct nfsd4_compound_state *cstate, + union nfsd4_op_u *u) +{ + struct nfsd4_get_dir_delegation *gdd = &u->get_dir_delegation; + + /* + * RFC 8881, section 18.39.3 says: + * + * "The server may refuse to grant the delegation. In that case, the + * server will return NFS4ERR_DIRDELEG_UNAVAIL." + * + * This is sub-optimal, since it means that the server would need to + * abort compound processing just because the delegation wasn't + * available. RFC8881bis should change this to allow the server to + * return NFS4_OK with a non-fatal status of GDD4_UNAVAIL in this + * situation. + */ + gdd->gddrnf_status = GDD4_UNAVAIL; + return nfs_ok; +} + #ifdef CONFIG_NFSD_PNFS static const struct nfsd4_layout_ops * nfsd4_layout_verify(struct svc_export *exp, unsigned int layout_type) @@ -3082,6 +3105,18 @@ static u32 nfsd4_copy_notify_rsize(const struct svc_rqst *rqstp, * sizeof(__be32); } +static u32 nfsd4_get_dir_delegation_rsize(const struct svc_rqst *rqstp, + const struct nfsd4_op *op) +{ + return (op_encode_hdr_size + + 1 /* gddr_status */ + + op_encode_verifier_maxsz + + op_encode_stateid_maxsz + + 2 /* gddr_notification */ + + 2 /* gddr_child_attributes */ + + 2 /* gddr_dir_attributes */); +} + #ifdef CONFIG_NFSD_PNFS static u32 nfsd4_getdeviceinfo_rsize(const struct svc_rqst *rqstp, const struct nfsd4_op *op) @@ -3470,6 +3505,12 @@ static const struct nfsd4_operation nfsd4_ops[] = { .op_get_currentstateid = nfsd4_get_freestateid, .op_rsize_bop = nfsd4_only_status_rsize, }, + [OP_GET_DIR_DELEGATION] = { + .op_func = nfsd4_get_dir_delegation, + .op_flags = OP_MODIFIES_SOMETHING, + .op_name = "OP_GET_DIR_DELEGATION", + .op_rsize_bop = nfsd4_get_dir_delegation_rsize, + }, #ifdef CONFIG_NFSD_PNFS [OP_GETDEVICEINFO] = { .op_func = nfsd4_getdeviceinfo, diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index a644460f3a5e..ea76100e50ca 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1732,6 +1732,35 @@ nfsd4_decode_free_stateid(struct nfsd4_compoundargs *argp, return nfsd4_decode_stateid4(argp, &free_stateid->fr_stateid); } +static __be32 +nfsd4_decode_get_dir_delegation(struct nfsd4_compoundargs *argp, + union nfsd4_op_u *u) +{ + struct nfsd4_get_dir_delegation *gdd = &u->get_dir_delegation; + __be32 status; + + memset(gdd, 0, sizeof(*gdd)); + + if (xdr_stream_decode_bool(argp->xdr, &gdd->gdda_signal_deleg_avail) < 0) + return nfserr_bad_xdr; + status = nfsd4_decode_bitmap4(argp, gdd->gdda_notification_types, + ARRAY_SIZE(gdd->gdda_notification_types)); + if (status) + return status; + status = nfsd4_decode_nfstime4(argp, &gdd->gdda_child_attr_delay); + if (status) + return status; + status = nfsd4_decode_nfstime4(argp, &gdd->gdda_dir_attr_delay); + if (status) + return status; + status = nfsd4_decode_bitmap4(argp, gdd->gdda_child_attributes, + ARRAY_SIZE(gdd->gdda_child_attributes)); + if (status) + return status; + return nfsd4_decode_bitmap4(argp, gdd->gdda_dir_attributes, + ARRAY_SIZE(gdd->gdda_dir_attributes)); +} + #ifdef CONFIG_NFSD_PNFS static __be32 nfsd4_decode_getdeviceinfo(struct nfsd4_compoundargs *argp, @@ -2370,7 +2399,7 @@ static const nfsd4_dec nfsd4_dec_ops[] = { [OP_CREATE_SESSION] = nfsd4_decode_create_session, [OP_DESTROY_SESSION] = nfsd4_decode_destroy_session, [OP_FREE_STATEID] = nfsd4_decode_free_stateid, - [OP_GET_DIR_DELEGATION] = nfsd4_decode_notsupp, + [OP_GET_DIR_DELEGATION] = nfsd4_decode_get_dir_delegation, #ifdef CONFIG_NFSD_PNFS [OP_GETDEVICEINFO] = nfsd4_decode_getdeviceinfo, [OP_GETDEVICELIST] = nfsd4_decode_notsupp, @@ -4963,6 +4992,49 @@ nfsd4_encode_test_stateid(struct nfsd4_compoundres *resp, __be32 nfserr, return nfs_ok; } +static __be32 +nfsd4_encode_get_dir_delegation(struct nfsd4_compoundres *resp, __be32 nfserr, + union nfsd4_op_u *u) +{ + struct nfsd4_get_dir_delegation *gdd = &u->get_dir_delegation; + struct xdr_stream *xdr = resp->xdr; + __be32 status = nfserr_resource; + + switch(gdd->gddrnf_status) { + case GDD4_OK: + if (xdr_stream_encode_u32(xdr, GDD4_OK) != XDR_UNIT) + break; + status = nfsd4_encode_verifier4(xdr, &gdd->gddr_cookieverf); + if (status) + break; + status = nfsd4_encode_stateid4(xdr, &gdd->gddr_stateid); + if (status) + break; + status = nfsd4_encode_bitmap4(xdr, gdd->gddr_notification[0], 0, 0); + if (status) + break; + status = nfsd4_encode_bitmap4(xdr, gdd->gddr_child_attributes[0], + gdd->gddr_child_attributes[1], + gdd->gddr_child_attributes[2]); + if (status) + break; + status = nfsd4_encode_bitmap4(xdr, gdd->gddr_dir_attributes[0], + gdd->gddr_dir_attributes[1], + gdd->gddr_dir_attributes[2]); + break; + default: + pr_warn("nfsd: bad gddrnf_status (%u)\n", gdd->gddrnf_status); + gdd->gddrnf_will_signal_deleg_avail = 0; + fallthrough; + case GDD4_UNAVAIL: + if (xdr_stream_encode_u32(xdr, GDD4_UNAVAIL) != XDR_UNIT) + break; + status = nfsd4_encode_bool(xdr, gdd->gddrnf_will_signal_deleg_avail); + break; + } + return status; +} + #ifdef CONFIG_NFSD_PNFS static __be32 nfsd4_encode_device_addr4(struct xdr_stream *xdr, @@ -5579,7 +5651,7 @@ static const nfsd4_enc nfsd4_enc_ops[] = { [OP_CREATE_SESSION] = nfsd4_encode_create_session, [OP_DESTROY_SESSION] = nfsd4_encode_noop, [OP_FREE_STATEID] = nfsd4_encode_noop, - [OP_GET_DIR_DELEGATION] = nfsd4_encode_noop, + [OP_GET_DIR_DELEGATION] = nfsd4_encode_get_dir_delegation, #ifdef CONFIG_NFSD_PNFS [OP_GETDEVICEINFO] = nfsd4_encode_getdeviceinfo, [OP_GETDEVICELIST] = nfsd4_encode_noop, diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index 415516c1b27e..446e72b0385e 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -518,6 +518,24 @@ struct nfsd4_free_stateid { stateid_t fr_stateid; /* request */ }; +struct nfsd4_get_dir_delegation { + /* request */ + u32 gdda_signal_deleg_avail; + u32 gdda_notification_types[1]; + struct timespec64 gdda_child_attr_delay; + struct timespec64 gdda_dir_attr_delay; + u32 gdda_child_attributes[3]; + u32 gdda_dir_attributes[3]; + /* response */ + u32 gddrnf_status; + nfs4_verifier gddr_cookieverf; + stateid_t gddr_stateid; + u32 gddr_notification[1]; + u32 gddr_child_attributes[3]; + u32 gddr_dir_attributes[3]; + bool gddrnf_will_signal_deleg_avail; +}; + /* also used for NVERIFY */ struct nfsd4_verify { u32 ve_bmval[3]; /* request */ @@ -797,6 +815,7 @@ struct nfsd4_op { struct nfsd4_reclaim_complete reclaim_complete; struct nfsd4_test_stateid test_stateid; struct nfsd4_free_stateid free_stateid; + struct nfsd4_get_dir_delegation get_dir_delegation; struct nfsd4_getdeviceinfo getdeviceinfo; struct nfsd4_layoutget layoutget; struct nfsd4_layoutcommit layoutcommit; diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index ef8d2d618d5b..0d896ce296ce 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -701,6 +701,12 @@ enum state_protect_how4 { SP4_SSV = 2 }; +/* GET_DIR_DELEGATION non-fatal status codes */ +enum gddrnf4_status { + GDD4_OK = 0, + GDD4_UNAVAIL = 1 +}; + enum pnfs_layouttype { LAYOUT_NFSV4_1_FILES = 1, LAYOUT_OSD2_OBJECTS = 2, -- cgit v1.2.3-59-g8ed1b From 7d12cce8784c66ad9351e91743cd68a573850732 Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Mon, 25 Mar 2024 21:21:39 +0800 Subject: fs: nfsd: use group allocation/free of per-cpu counters API Use group allocation/free of per-cpu counters api to accelerate nfsd percpu_counters init/destroy(), and also squash the nfsd_percpu_counters_init/reset/destroy() and nfsd_counters_init/destroy() into callers to simplify code. Signed-off-by: Kefeng Wang Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/export.c | 16 ++++++++++------ fs/nfsd/nfsctl.c | 5 +++-- fs/nfsd/stats.c | 42 ------------------------------------------ fs/nfsd/stats.h | 5 ----- 4 files changed, 13 insertions(+), 55 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 7b641095a665..50b3135d07ac 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -334,21 +334,25 @@ static void nfsd4_fslocs_free(struct nfsd4_fs_locations *fsloc) static int export_stats_init(struct export_stats *stats) { stats->start_time = ktime_get_seconds(); - return nfsd_percpu_counters_init(stats->counter, EXP_STATS_COUNTERS_NUM); + return percpu_counter_init_many(stats->counter, 0, GFP_KERNEL, + EXP_STATS_COUNTERS_NUM); } static void export_stats_reset(struct export_stats *stats) { - if (stats) - nfsd_percpu_counters_reset(stats->counter, - EXP_STATS_COUNTERS_NUM); + if (stats) { + int i; + + for (i = 0; i < EXP_STATS_COUNTERS_NUM; i++) + percpu_counter_set(&stats->counter[i], 0); + } } static void export_stats_destroy(struct export_stats *stats) { if (stats) - nfsd_percpu_counters_destroy(stats->counter, - EXP_STATS_COUNTERS_NUM); + percpu_counter_destroy_many(stats->counter, + EXP_STATS_COUNTERS_NUM); } static void svc_export_put(struct kref *ref) diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index ecd18bffeebc..93c87587e646 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -1672,7 +1672,8 @@ static __net_init int nfsd_net_init(struct net *net) retval = nfsd_idmap_init(net); if (retval) goto out_idmap_error; - retval = nfsd_stat_counters_init(nn); + retval = percpu_counter_init_many(nn->counter, 0, GFP_KERNEL, + NFSD_STATS_COUNTERS_NUM); if (retval) goto out_repcache_error; memset(&nn->nfsd_svcstats, 0, sizeof(nn->nfsd_svcstats)); @@ -1704,7 +1705,7 @@ static __net_exit void nfsd_net_exit(struct net *net) struct nfsd_net *nn = net_generic(net, nfsd_net_id); nfsd_proc_stat_shutdown(net); - nfsd_stat_counters_destroy(nn); + percpu_counter_destroy_many(nn->counter, NFSD_STATS_COUNTERS_NUM); nfsd_idmap_shutdown(net); nfsd_export_shutdown(net); nfsd_netns_free_versions(nn); diff --git a/fs/nfsd/stats.c b/fs/nfsd/stats.c index be52fb1e928e..bb22893f1157 100644 --- a/fs/nfsd/stats.c +++ b/fs/nfsd/stats.c @@ -73,48 +73,6 @@ static int nfsd_show(struct seq_file *seq, void *v) DEFINE_PROC_SHOW_ATTRIBUTE(nfsd); -int nfsd_percpu_counters_init(struct percpu_counter *counters, int num) -{ - int i, err = 0; - - for (i = 0; !err && i < num; i++) - err = percpu_counter_init(&counters[i], 0, GFP_KERNEL); - - if (!err) - return 0; - - for (; i > 0; i--) - percpu_counter_destroy(&counters[i-1]); - - return err; -} - -void nfsd_percpu_counters_reset(struct percpu_counter counters[], int num) -{ - int i; - - for (i = 0; i < num; i++) - percpu_counter_set(&counters[i], 0); -} - -void nfsd_percpu_counters_destroy(struct percpu_counter counters[], int num) -{ - int i; - - for (i = 0; i < num; i++) - percpu_counter_destroy(&counters[i]); -} - -int nfsd_stat_counters_init(struct nfsd_net *nn) -{ - return nfsd_percpu_counters_init(nn->counter, NFSD_STATS_COUNTERS_NUM); -} - -void nfsd_stat_counters_destroy(struct nfsd_net *nn) -{ - nfsd_percpu_counters_destroy(nn->counter, NFSD_STATS_COUNTERS_NUM); -} - void nfsd_proc_stat_init(struct net *net) { struct nfsd_net *nn = net_generic(net, nfsd_net_id); diff --git a/fs/nfsd/stats.h b/fs/nfsd/stats.h index d2753e975dfd..04aacb6c36e2 100644 --- a/fs/nfsd/stats.h +++ b/fs/nfsd/stats.h @@ -10,11 +10,6 @@ #include #include -int nfsd_percpu_counters_init(struct percpu_counter *counters, int num); -void nfsd_percpu_counters_reset(struct percpu_counter *counters, int num); -void nfsd_percpu_counters_destroy(struct percpu_counter *counters, int num); -int nfsd_stat_counters_init(struct nfsd_net *nn); -void nfsd_stat_counters_destroy(struct nfsd_net *nn); void nfsd_proc_stat_init(struct net *net); void nfsd_proc_stat_shutdown(struct net *net); -- cgit v1.2.3-59-g8ed1b From db2acd6e8ab1606c47bcea3e269ac7cff897eb8d Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 5 Apr 2024 14:40:49 -0400 Subject: nfsd: drop extraneous newline from nfsd tracepoints We never want a newline in tracepoint output. Signed-off-by: Jeff Layton Tested-by: Vladimir Benes Signed-off-by: Chuck Lever --- fs/nfsd/trace.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index 1cd2076210b1..c0cc345a576e 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -1534,7 +1534,7 @@ TRACE_EVENT(nfsd_cb_seq_status, __entry->seq_status = cb->cb_seq_status; ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER - " sessionid=%08x:%08x:%08x:%08x tk_status=%d seq_status=%d\n", + " sessionid=%08x:%08x:%08x:%08x tk_status=%d seq_status=%d", __entry->task_id, __entry->client_id, __entry->cl_boot, __entry->cl_id, __entry->seqno, __entry->reserved, @@ -1573,7 +1573,7 @@ TRACE_EVENT(nfsd_cb_free_slot, __entry->slot_seqno = session->se_cb_seq_nr; ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER - " sessionid=%08x:%08x:%08x:%08x new slot seqno=%u\n", + " sessionid=%08x:%08x:%08x:%08x new slot seqno=%u", __entry->task_id, __entry->client_id, __entry->cl_boot, __entry->cl_id, __entry->seqno, __entry->reserved, @@ -1978,7 +1978,7 @@ TRACE_EVENT(nfsd_ctl_time, __entry->time = time; __assign_str(name, name); ), - TP_printk("file=%s time=%d\n", + TP_printk("file=%s time=%d", __get_str(name), __entry->time ) ); -- cgit v1.2.3-59-g8ed1b From 2d49901150ecf82f9a3d3d2dc30e38b17fef71a0 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 5 Apr 2024 14:40:50 -0400 Subject: nfsd: new tracepoint for check_slot_seqid Replace a dprintk in check_slot_seqid with tracepoints. These new tracepoints track slot sequence numbers during operation. Suggested-by: Jeffrey Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 14 +++++------ fs/nfsd/trace.h | 70 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 77 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 5ae70fb1d3a2..bd81217ada7f 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -3636,12 +3636,8 @@ out_nolock: return status; } -static __be32 -check_slot_seqid(u32 seqid, u32 slot_seqid, int slot_inuse) +static __be32 check_slot_seqid(u32 seqid, u32 slot_seqid, bool slot_inuse) { - dprintk("%s enter. seqid %d slot_seqid %d\n", __func__, seqid, - slot_seqid); - /* The slot is in use, and no response has been sent. */ if (slot_inuse) { if (seqid == slot_seqid) @@ -3818,10 +3814,13 @@ nfsd4_create_session(struct svc_rqst *rqstp, } /* RFC 8881 Section 18.36.4 Phase 2: Sequence ID processing. */ - if (conf) + if (conf) { cs_slot = &conf->cl_cs_slot; - else + trace_nfsd_slot_seqid_conf(conf, cr_ses); + } else { cs_slot = &unconf->cl_cs_slot; + trace_nfsd_slot_seqid_unconf(unconf, cr_ses); + } status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0); switch (status) { case nfs_ok: @@ -4216,6 +4215,7 @@ nfsd4_sequence(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, * sr_highest_slotid and the sr_target_slot id to maxslots */ seq->maxslots = session->se_fchannel.maxreqs; + trace_nfsd_slot_seqid_sequence(clp, seq, slot); status = check_slot_seqid(seq->seqid, slot->sl_seqid, slot->sl_flags & NFSD4_SLOT_INUSE); if (status == nfserr_replay_cache) { diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index c0cc345a576e..ac2124461a08 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -749,6 +749,76 @@ TRACE_EVENT_CONDITION(nfsd_seq4_status, ) ); +DECLARE_EVENT_CLASS(nfsd_cs_slot_class, + TP_PROTO( + const struct nfs4_client *clp, + const struct nfsd4_create_session *cs + ), + TP_ARGS(clp, cs), + TP_STRUCT__entry( + __field(u32, seqid) + __field(u32, slot_seqid) + __field(u32, cl_boot) + __field(u32, cl_id) + __sockaddr(addr, clp->cl_cb_conn.cb_addrlen) + ), + TP_fast_assign( + const struct nfsd4_clid_slot *slot = &clp->cl_cs_slot; + + __entry->cl_boot = clp->cl_clientid.cl_boot; + __entry->cl_id = clp->cl_clientid.cl_id; + __assign_sockaddr(addr, &clp->cl_cb_conn.cb_addr, + clp->cl_cb_conn.cb_addrlen); + __entry->seqid = cs->seqid; + __entry->slot_seqid = slot->sl_seqid; + ), + TP_printk("addr=%pISpc client %08x:%08x seqid=%u slot_seqid=%u", + __get_sockaddr(addr), __entry->cl_boot, __entry->cl_id, + __entry->seqid, __entry->slot_seqid + ) +); + +#define DEFINE_CS_SLOT_EVENT(name) \ +DEFINE_EVENT(nfsd_cs_slot_class, nfsd_##name, \ + TP_PROTO( \ + const struct nfs4_client *clp, \ + const struct nfsd4_create_session *cs \ + ), \ + TP_ARGS(clp, cs)) + +DEFINE_CS_SLOT_EVENT(slot_seqid_conf); +DEFINE_CS_SLOT_EVENT(slot_seqid_unconf); + +TRACE_EVENT(nfsd_slot_seqid_sequence, + TP_PROTO( + const struct nfs4_client *clp, + const struct nfsd4_sequence *seq, + const struct nfsd4_slot *slot + ), + TP_ARGS(clp, seq, slot), + TP_STRUCT__entry( + __field(u32, seqid) + __field(u32, slot_seqid) + __field(u32, cl_boot) + __field(u32, cl_id) + __sockaddr(addr, clp->cl_cb_conn.cb_addrlen) + __field(bool, in_use) + ), + TP_fast_assign( + __entry->cl_boot = clp->cl_clientid.cl_boot; + __entry->cl_id = clp->cl_clientid.cl_id; + __assign_sockaddr(addr, &clp->cl_cb_conn.cb_addr, + clp->cl_cb_conn.cb_addrlen); + __entry->seqid = seq->seqid; + __entry->slot_seqid = slot->sl_seqid; + ), + TP_printk("addr=%pISpc client %08x:%08x seqid=%u slot_seqid=%u (%sin use)", + __get_sockaddr(addr), __entry->cl_boot, __entry->cl_id, + __entry->seqid, __entry->slot_seqid, + __entry->in_use ? "" : "not " + ) +); + DECLARE_EVENT_CLASS(nfsd_clientid_class, TP_PROTO(const clientid_t *clid), TP_ARGS(clid), -- cgit v1.2.3-59-g8ed1b From 9320f27fda5498c80f9e0c7d8c80d50e8297d89e Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 5 Apr 2024 14:40:51 -0400 Subject: nfsd: add tracepoint in mark_client_expired_locked Show client info alongside the number of cl_rpc_users. If that's elevated, then we can infer that this function returned nfserr_jukebox. [ cel: For additional debugging of RPC user refcounting ] Signed-off-by: Jeff Layton Tested-by: Vladimir Benes Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 6 +++++- fs/nfsd/trace.h | 24 ++++++++++++++++++++++++ 2 files changed, 29 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index bd81217ada7f..66f9a4f031a4 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2347,7 +2347,11 @@ unhash_client(struct nfs4_client *clp) static __be32 mark_client_expired_locked(struct nfs4_client *clp) { - if (atomic_read(&clp->cl_rpc_users)) + int users = atomic_read(&clp->cl_rpc_users); + + trace_nfsd_mark_client_expired(clp, users); + + if (users) return nfserr_jukebox; unhash_client_locked(clp); return nfs_ok; diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index ac2124461a08..b5e48d504062 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -848,6 +848,30 @@ DEFINE_CLIENTID_EVENT(purged); DEFINE_CLIENTID_EVENT(renew); DEFINE_CLIENTID_EVENT(stale); +TRACE_EVENT(nfsd_mark_client_expired, + TP_PROTO( + const struct nfs4_client *clp, + int cl_rpc_users + ), + TP_ARGS(clp, cl_rpc_users), + TP_STRUCT__entry( + __field(int, cl_rpc_users) + __field(u32, cl_boot) + __field(u32, cl_id) + __sockaddr(addr, clp->cl_cb_conn.cb_addrlen) + ), + TP_fast_assign( + __entry->cl_rpc_users = cl_rpc_users; + __entry->cl_boot = clp->cl_clientid.cl_boot; + __entry->cl_id = clp->cl_clientid.cl_id; + __assign_sockaddr(addr, &clp->cl_cb_conn.cb_addr, + clp->cl_cb_conn.cb_addrlen) + ), + TP_printk("addr=%pISpc client %08x:%08x cl_rpc_users=%d", + __get_sockaddr(addr), __entry->cl_boot, __entry->cl_id, + __entry->cl_rpc_users) +); + DECLARE_EVENT_CLASS(nfsd_net_class, TP_PROTO(const struct nfsd_net *nn), TP_ARGS(nn), -- cgit v1.2.3-59-g8ed1b From d43113fbbf79c00fe3ecd637f8cadebc4df9a63c Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 8 Apr 2024 11:29:52 +1000 Subject: nfsd: optimise recalculate_deny_mode() for a common case recalculate_deny_mode() takes time that is linear in the number of stateids active on the file. When called from release_openowner -> free_ol_stateid_reaplist ->nfs4_free_ol_stateid -> release_all_access the number of times it is called is linear in the number of stateids. The net result is that time taken by release_openowner is quadratic in the number of stateids. When the nfsd server is shut down while there are many active stateids this can result in a soft lockup. ("CPU stuck for 302s" seen in one case). In many cases all the states have the same deny modes and there is no need to examine the entire list in recalculate_deny_mode(). In particular, recalculate_deny_mode() will only reduce the deny mode, never increase it. So if some prefix of the list causes the original deny mode to be required, there is no need to examine the remainder of the list. So we can improve recalculate_deny_mode() to usually run in constant time, so release_openowner will typically be only linear in the number of states. Signed-off-by: NeilBrown Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs4state.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 66f9a4f031a4..a20c2c9d7d45 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1397,11 +1397,16 @@ static void recalculate_deny_mode(struct nfs4_file *fp) { struct nfs4_ol_stateid *stp; + u32 old_deny; spin_lock(&fp->fi_lock); + old_deny = fp->fi_share_deny; fp->fi_share_deny = 0; - list_for_each_entry(stp, &fp->fi_stateids, st_perfile) + list_for_each_entry(stp, &fp->fi_stateids, st_perfile) { fp->fi_share_deny |= bmap_to_share_mode(stp->st_deny_bmap); + if (fp->fi_share_deny == old_deny) + break; + } spin_unlock(&fp->fi_lock); } -- cgit v1.2.3-59-g8ed1b From 0770249b90f9d9f69714b76adc36cf6c895bc1f9 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 17 Apr 2024 07:23:02 +1000 Subject: nfsd: don't create nfsv4recoverydir in nfsdfs when not used. When CONFIG_NFSD_LEGACY_CLIENT_TRACKING is not set, the virtual file /proc/fs/nfsd/nfsv4recoverydir is created but responds EINVAL to any access. This is not useful, is somewhat surprising, and it causes ltp to complain. The only known user of this file is in nfs-utils, which handles non-existence and read-failure equally well. So there is nothing to gain from leaving the file present but inaccessible. So this patch removes the file when its content is not available - i.e. when that config option is not selected. Also remove the #ifdef which hides some of the enum values when CONFIG_NFSD_V$ not selection. simple_fill_super() quietly ignores array entries that are not present, so having slots in the array that don't get used is perfectly acceptable. So there is no value in this #ifdef. Reported-by: Petr Vorel Reviewed-by: Jeff Layton Fixes: 74fd48739d04 ("nfsd: new Kconfig option for legacy client tracking") Signed-off-by: NeilBrown Reviewed-by: Petr Vorel Signed-off-by: Chuck Lever --- fs/nfsd/nfsctl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 93c87587e646..340c5d61f199 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -48,12 +48,10 @@ enum { NFSD_MaxBlkSize, NFSD_MaxConnections, NFSD_Filecache, -#ifdef CONFIG_NFSD_V4 NFSD_Leasetime, NFSD_Gracetime, NFSD_RecoveryDir, NFSD_V4EndGrace, -#endif NFSD_MaxReserved }; @@ -1360,7 +1358,9 @@ static int nfsd_fill_super(struct super_block *sb, struct fs_context *fc) #ifdef CONFIG_NFSD_V4 [NFSD_Leasetime] = {"nfsv4leasetime", &transaction_ops, S_IWUSR|S_IRUSR}, [NFSD_Gracetime] = {"nfsv4gracetime", &transaction_ops, S_IWUSR|S_IRUSR}, +#ifdef CONFIG_NFSD_LEGACY_CLIENT_TRACKING [NFSD_RecoveryDir] = {"nfsv4recoverydir", &transaction_ops, S_IWUSR|S_IRUSR}, +#endif [NFSD_V4EndGrace] = {"v4_end_grace", &transaction_ops, S_IWUSR|S_IRUGO}, #endif /* last one */ {""} -- cgit v1.2.3-59-g8ed1b From 03b0036f452d244b6bfeab3973e9e57e8eac9c04 Mon Sep 17 00:00:00 2001 From: Li kunyu Date: Wed, 17 Apr 2024 16:28:07 +0800 Subject: lockd: host: Remove unnecessary statements＇host = NULL;＇ MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In 'nlm_alloc_host', the host has already been assigned a value of NULL when defined, so 'host=NULL;' Can be deleted. Signed-off-by: Li kunyu Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/lockd/host.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/lockd/host.c b/fs/lockd/host.c index 127a728fcbc8..c11516801784 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -117,7 +117,6 @@ static struct nlm_host *nlm_alloc_host(struct nlm_lookup_host_info *ni, if (nsm != NULL) refcount_inc(&nsm->sm_count); else { - host = NULL; nsm = nsm_get_handle(ni->net, ni->sap, ni->salen, ni->hostname, ni->hostname_len); if (unlikely(nsm == NULL)) { -- cgit v1.2.3-59-g8ed1b From 0842b4c80bc7162a803b1558dc6d887e0e8feb39 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 23 Apr 2024 15:25:38 +0200 Subject: NFSD: move nfsd_mutex handling into nfsd_svc callers Currently nfsd_svc holds the nfsd_mutex over the whole function. For some of the later netlink patches though, we want to do some other things to the server before starting it. Move the mutex handling into the callers. Signed-off-by: Jeff Layton Signed-off-by: Lorenzo Bianconi Signed-off-by: Chuck Lever --- fs/nfsd/nfsctl.c | 2 ++ fs/nfsd/nfssvc.c | 4 ++-- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 340c5d61f199..2fe78b802f98 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -404,7 +404,9 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) if (newthreads < 0) return -EINVAL; trace_nfsd_ctl_threads(net, newthreads); + mutex_lock(&nfsd_mutex); rv = nfsd_svc(newthreads, net, file->f_cred); + mutex_unlock(&nfsd_mutex); if (rv < 0) return rv; } else diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index c0d17b92b249..ca193f7ff0e1 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -775,7 +775,8 @@ nfsd_svc(int nrservs, struct net *net, const struct cred *cred) struct nfsd_net *nn = net_generic(net, nfsd_net_id); struct svc_serv *serv; - mutex_lock(&nfsd_mutex); + lockdep_assert_held(&nfsd_mutex); + dprintk("nfsd: creating service\n"); nrservs = max(nrservs, 0); @@ -804,7 +805,6 @@ out_put: if (serv->sv_nrthreads == 0) nfsd_destroy_serv(net); out: - mutex_unlock(&nfsd_mutex); return error; } -- cgit v1.2.3-59-g8ed1b From 9077d59847896745712dff00839eac14f84b21f8 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 23 Apr 2024 15:25:39 +0200 Subject: NFSD: allow callers to pass in scope string to nfsd_svc Currently admins set this by using unshare to create a new uts namespace, and then resetting the hostname. With the new netlink interface we can just pass this in directly. Prepare nfsd_svc for this change. Signed-off-by: Jeff Layton Signed-off-by: Lorenzo Bianconi Signed-off-by: Chuck Lever --- fs/nfsd/nfsctl.c | 2 +- fs/nfsd/nfsd.h | 2 +- fs/nfsd/nfssvc.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 2fe78b802f98..34148d73b6c1 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -405,7 +405,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) return -EINVAL; trace_nfsd_ctl_threads(net, newthreads); mutex_lock(&nfsd_mutex); - rv = nfsd_svc(newthreads, net, file->f_cred); + rv = nfsd_svc(newthreads, net, file->f_cred, NULL); mutex_unlock(&nfsd_mutex); if (rv < 0) return rv; diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h index 16c5a05f340e..2f6c6f3815b4 100644 --- a/fs/nfsd/nfsd.h +++ b/fs/nfsd/nfsd.h @@ -103,7 +103,7 @@ bool nfssvc_encode_voidres(struct svc_rqst *rqstp, /* * Function prototypes. */ -int nfsd_svc(int nrservs, struct net *net, const struct cred *cred); +int nfsd_svc(int nrservs, struct net *net, const struct cred *cred, const char *scope); int nfsd_dispatch(struct svc_rqst *rqstp); int nfsd_nrthreads(struct net *); diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index ca193f7ff0e1..e25b9b829749 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -769,7 +769,7 @@ int nfsd_set_nrthreads(int n, int *nthreads, struct net *net) * this is the first time nrservs is nonzero. */ int -nfsd_svc(int nrservs, struct net *net, const struct cred *cred) +nfsd_svc(int nrservs, struct net *net, const struct cred *cred, const char *scope) { int error; struct nfsd_net *nn = net_generic(net, nfsd_net_id); @@ -786,7 +786,7 @@ nfsd_svc(int nrservs, struct net *net, const struct cred *cred) if (nrservs == 0 && nn->nfsd_serv == NULL) goto out; - strscpy(nn->nfsd_name, utsname()->nodename, + strscpy(nn->nfsd_name, scope ? scope : utsname()->nodename, sizeof(nn->nfsd_name)); error = nfsd_create_serv(net); -- cgit v1.2.3-59-g8ed1b From 924f4fb003ba114c60b3c07a011dcd86a8956cd1 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 23 Apr 2024 15:25:40 +0200 Subject: NFSD: convert write_threads to netlink command Introduce write_threads netlink command similar to the one available through the procfs. Tested-by: Jeff Layton Reviewed-by: Jeff Layton Co-developed-by: Jeff Layton Signed-off-by: Jeff Layton Signed-off-by: Lorenzo Bianconi Signed-off-by: Chuck Lever --- Documentation/netlink/specs/nfsd.yaml | 39 ++++++++++ fs/nfsd/netlink.c | 20 +++++ fs/nfsd/netlink.h | 2 + fs/nfsd/nfsctl.c | 143 ++++++++++++++++++++++++++++++++++ include/uapi/linux/nfsd_netlink.h | 12 +++ 5 files changed, 216 insertions(+) (limited to 'fs') diff --git a/Documentation/netlink/specs/nfsd.yaml b/Documentation/netlink/specs/nfsd.yaml index 05acc73e2e33..703c2d11d0a9 100644 --- a/Documentation/netlink/specs/nfsd.yaml +++ b/Documentation/netlink/specs/nfsd.yaml @@ -62,6 +62,22 @@ attribute-sets: name: compound-ops type: u32 multi-attr: true + - + name: server + attributes: + - + name: threads + type: u32 + multi-attr: true + - + name: gracetime + type: u32 + - + name: leasetime + type: u32 + - + name: scope + type: string operations: list: @@ -87,3 +103,26 @@ operations: - sport - dport - compound-ops + - + name: threads-set + doc: set the number of running threads + attribute-set: server + flags: [ admin-perm ] + do: + request: + attributes: + - threads + - gracetime + - leasetime + - scope + - + name: threads-get + doc: get the number of running threads + attribute-set: server + do: + reply: + attributes: + - threads + - gracetime + - leasetime + - scope diff --git a/fs/nfsd/netlink.c b/fs/nfsd/netlink.c index 0e1d635ec5f9..fe9eef5c7f27 100644 --- a/fs/nfsd/netlink.c +++ b/fs/nfsd/netlink.c @@ -10,6 +10,14 @@ #include +/* NFSD_CMD_THREADS_SET - do */ +static const struct nla_policy nfsd_threads_set_nl_policy[NFSD_A_SERVER_SCOPE + 1] = { + [NFSD_A_SERVER_THREADS] = { .type = NLA_U32, }, + [NFSD_A_SERVER_GRACETIME] = { .type = NLA_U32, }, + [NFSD_A_SERVER_LEASETIME] = { .type = NLA_U32, }, + [NFSD_A_SERVER_SCOPE] = { .type = NLA_NUL_STRING, }, +}; + /* Ops table for nfsd */ static const struct genl_split_ops nfsd_nl_ops[] = { { @@ -19,6 +27,18 @@ static const struct genl_split_ops nfsd_nl_ops[] = { .done = nfsd_nl_rpc_status_get_done, .flags = GENL_CMD_CAP_DUMP, }, + { + .cmd = NFSD_CMD_THREADS_SET, + .doit = nfsd_nl_threads_set_doit, + .policy = nfsd_threads_set_nl_policy, + .maxattr = NFSD_A_SERVER_SCOPE, + .flags = GENL_ADMIN_PERM | GENL_CMD_CAP_DO, + }, + { + .cmd = NFSD_CMD_THREADS_GET, + .doit = nfsd_nl_threads_get_doit, + .flags = GENL_CMD_CAP_DO, + }, }; struct genl_family nfsd_nl_family __ro_after_init = { diff --git a/fs/nfsd/netlink.h b/fs/nfsd/netlink.h index d83dd6bdee92..4137fac477e4 100644 --- a/fs/nfsd/netlink.h +++ b/fs/nfsd/netlink.h @@ -16,6 +16,8 @@ int nfsd_nl_rpc_status_get_done(struct netlink_callback *cb); int nfsd_nl_rpc_status_get_dumpit(struct sk_buff *skb, struct netlink_callback *cb); +int nfsd_nl_threads_set_doit(struct sk_buff *skb, struct genl_info *info); +int nfsd_nl_threads_get_doit(struct sk_buff *skb, struct genl_info *info); extern struct genl_family nfsd_nl_family; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 34148d73b6c1..5bfdebb2e7e9 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -1653,6 +1654,148 @@ int nfsd_nl_rpc_status_get_done(struct netlink_callback *cb) return 0; } +/** + * nfsd_nl_threads_set_doit - set the number of running threads + * @skb: reply buffer + * @info: netlink metadata and command arguments + * + * Return 0 on success or a negative errno. + */ +int nfsd_nl_threads_set_doit(struct sk_buff *skb, struct genl_info *info) +{ + int nthreads = 0, count = 0, nrpools, ret = -EOPNOTSUPP, rem; + struct net *net = genl_info_net(info); + struct nfsd_net *nn = net_generic(net, nfsd_net_id); + const struct nlattr *attr; + const char *scope = NULL; + + if (GENL_REQ_ATTR_CHECK(info, NFSD_A_SERVER_THREADS)) + return -EINVAL; + + /* count number of SERVER_THREADS values */ + nlmsg_for_each_attr(attr, info->nlhdr, GENL_HDRLEN, rem) { + if (nla_type(attr) == NFSD_A_SERVER_THREADS) + count++; + } + + mutex_lock(&nfsd_mutex); + + nrpools = nfsd_nrpools(net); + if (nrpools && count > nrpools) + count = nrpools; + + /* XXX: make this handle non-global pool-modes */ + if (count > 1) + goto out_unlock; + + nthreads = nla_get_u32(info->attrs[NFSD_A_SERVER_THREADS]); + if (info->attrs[NFSD_A_SERVER_GRACETIME] || + info->attrs[NFSD_A_SERVER_LEASETIME] || + info->attrs[NFSD_A_SERVER_SCOPE]) { + ret = -EBUSY; + if (nn->nfsd_serv && nn->nfsd_serv->sv_nrthreads) + goto out_unlock; + + ret = -EINVAL; + attr = info->attrs[NFSD_A_SERVER_GRACETIME]; + if (attr) { + u32 gracetime = nla_get_u32(attr); + + if (gracetime < 10 || gracetime > 3600) + goto out_unlock; + + nn->nfsd4_grace = gracetime; + } + + attr = info->attrs[NFSD_A_SERVER_LEASETIME]; + if (attr) { + u32 leasetime = nla_get_u32(attr); + + if (leasetime < 10 || leasetime > 3600) + goto out_unlock; + + nn->nfsd4_lease = leasetime; + } + + attr = info->attrs[NFSD_A_SERVER_SCOPE]; + if (attr) + scope = nla_data(attr); + } + + ret = nfsd_svc(nthreads, net, get_current_cred(), scope); + +out_unlock: + mutex_unlock(&nfsd_mutex); + + return ret == nthreads ? 0 : ret; +} + +/** + * nfsd_nl_threads_get_doit - get the number of running threads + * @skb: reply buffer + * @info: netlink metadata and command arguments + * + * Return 0 on success or a negative errno. + */ +int nfsd_nl_threads_get_doit(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct nfsd_net *nn = net_generic(net, nfsd_net_id); + void *hdr; + int err; + + skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + hdr = genlmsg_iput(skb, info); + if (!hdr) { + err = -EMSGSIZE; + goto err_free_msg; + } + + mutex_lock(&nfsd_mutex); + + err = nla_put_u32(skb, NFSD_A_SERVER_GRACETIME, + nn->nfsd4_grace) || + nla_put_u32(skb, NFSD_A_SERVER_LEASETIME, + nn->nfsd4_lease) || + nla_put_string(skb, NFSD_A_SERVER_SCOPE, + nn->nfsd_name); + if (err) + goto err_unlock; + + if (nn->nfsd_serv) { + int i; + + for (i = 0; i < nfsd_nrpools(net); ++i) { + struct svc_pool *sp = &nn->nfsd_serv->sv_pools[i]; + + err = nla_put_u32(skb, NFSD_A_SERVER_THREADS, + atomic_read(&sp->sp_nrthreads)); + if (err) + goto err_unlock; + } + } else { + err = nla_put_u32(skb, NFSD_A_SERVER_THREADS, 0); + if (err) + goto err_unlock; + } + + mutex_unlock(&nfsd_mutex); + + genlmsg_end(skb, hdr); + + return genlmsg_reply(skb, info); + +err_unlock: + mutex_unlock(&nfsd_mutex); +err_free_msg: + nlmsg_free(skb); + + return err; +} + /** * nfsd_net_init - Prepare the nfsd_net portion of a new net namespace * @net: a freshly-created network namespace diff --git a/include/uapi/linux/nfsd_netlink.h b/include/uapi/linux/nfsd_netlink.h index 3cd044edee5d..4bbccd5db7cd 100644 --- a/include/uapi/linux/nfsd_netlink.h +++ b/include/uapi/linux/nfsd_netlink.h @@ -29,8 +29,20 @@ enum { NFSD_A_RPC_STATUS_MAX = (__NFSD_A_RPC_STATUS_MAX - 1) }; +enum { + NFSD_A_SERVER_THREADS = 1, + NFSD_A_SERVER_GRACETIME, + NFSD_A_SERVER_LEASETIME, + NFSD_A_SERVER_SCOPE, + + __NFSD_A_SERVER_MAX, + NFSD_A_SERVER_MAX = (__NFSD_A_SERVER_MAX - 1) +}; + enum { NFSD_CMD_RPC_STATUS_GET = 1, + NFSD_CMD_THREADS_SET, + NFSD_CMD_THREADS_GET, __NFSD_CMD_MAX, NFSD_CMD_MAX = (__NFSD_CMD_MAX - 1) -- cgit v1.2.3-59-g8ed1b From 5a939bea25be9793d9aa5d8494df667dfe625e6b Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 23 Apr 2024 15:25:41 +0200 Subject: NFSD: add write_version to netlink command Introduce write_version netlink command through a "declarative" interface. This patch introduces a change in behavior since for version-set userspace is expected to provide a NFS major/minor version list it wants to enable while all the other ones will be disabled. (procfs write_version command implements imperative interface where the admin writes +3/-3 to enable/disable a single version. Reviewed-by: Jeff Layton Tested-by: Jeff Layton Signed-off-by: Lorenzo Bianconi Signed-off-by: Chuck Lever --- Documentation/netlink/specs/nfsd.yaml | 37 +++++++++ fs/nfsd/netlink.c | 24 ++++++ fs/nfsd/netlink.h | 5 ++ fs/nfsd/netns.h | 1 + fs/nfsd/nfsctl.c | 150 ++++++++++++++++++++++++++++++++++ fs/nfsd/nfssvc.c | 3 +- include/uapi/linux/nfsd_netlink.h | 18 ++++ 7 files changed, 236 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/Documentation/netlink/specs/nfsd.yaml b/Documentation/netlink/specs/nfsd.yaml index 703c2d11d0a9..77adda1425e2 100644 --- a/Documentation/netlink/specs/nfsd.yaml +++ b/Documentation/netlink/specs/nfsd.yaml @@ -78,6 +78,26 @@ attribute-sets: - name: scope type: string + - + name: version + attributes: + - + name: major + type: u32 + - + name: minor + type: u32 + - + name: enabled + type: flag + - + name: server-proto + attributes: + - + name: version + type: nest + nested-attributes: version + multi-attr: true operations: list: @@ -126,3 +146,20 @@ operations: - gracetime - leasetime - scope + - + name: version-set + doc: set nfs enabled versions + attribute-set: server-proto + flags: [ admin-perm ] + do: + request: + attributes: + - version + - + name: version-get + doc: get nfs enabled versions + attribute-set: server-proto + do: + reply: + attributes: + - version diff --git a/fs/nfsd/netlink.c b/fs/nfsd/netlink.c index fe9eef5c7f27..b23b0b84a59a 100644 --- a/fs/nfsd/netlink.c +++ b/fs/nfsd/netlink.c @@ -10,6 +10,13 @@ #include +/* Common nested types */ +const struct nla_policy nfsd_version_nl_policy[NFSD_A_VERSION_ENABLED + 1] = { + [NFSD_A_VERSION_MAJOR] = { .type = NLA_U32, }, + [NFSD_A_VERSION_MINOR] = { .type = NLA_U32, }, + [NFSD_A_VERSION_ENABLED] = { .type = NLA_FLAG, }, +}; + /* NFSD_CMD_THREADS_SET - do */ static const struct nla_policy nfsd_threads_set_nl_policy[NFSD_A_SERVER_SCOPE + 1] = { [NFSD_A_SERVER_THREADS] = { .type = NLA_U32, }, @@ -18,6 +25,11 @@ static const struct nla_policy nfsd_threads_set_nl_policy[NFSD_A_SERVER_SCOPE + [NFSD_A_SERVER_SCOPE] = { .type = NLA_NUL_STRING, }, }; +/* NFSD_CMD_VERSION_SET - do */ +static const struct nla_policy nfsd_version_set_nl_policy[NFSD_A_SERVER_PROTO_VERSION + 1] = { + [NFSD_A_SERVER_PROTO_VERSION] = NLA_POLICY_NESTED(nfsd_version_nl_policy), +}; + /* Ops table for nfsd */ static const struct genl_split_ops nfsd_nl_ops[] = { { @@ -39,6 +51,18 @@ static const struct genl_split_ops nfsd_nl_ops[] = { .doit = nfsd_nl_threads_get_doit, .flags = GENL_CMD_CAP_DO, }, + { + .cmd = NFSD_CMD_VERSION_SET, + .doit = nfsd_nl_version_set_doit, + .policy = nfsd_version_set_nl_policy, + .maxattr = NFSD_A_SERVER_PROTO_VERSION, + .flags = GENL_ADMIN_PERM | GENL_CMD_CAP_DO, + }, + { + .cmd = NFSD_CMD_VERSION_GET, + .doit = nfsd_nl_version_get_doit, + .flags = GENL_CMD_CAP_DO, + }, }; struct genl_family nfsd_nl_family __ro_after_init = { diff --git a/fs/nfsd/netlink.h b/fs/nfsd/netlink.h index 4137fac477e4..c7c0da275481 100644 --- a/fs/nfsd/netlink.h +++ b/fs/nfsd/netlink.h @@ -11,6 +11,9 @@ #include +/* Common nested types */ +extern const struct nla_policy nfsd_version_nl_policy[NFSD_A_VERSION_ENABLED + 1]; + int nfsd_nl_rpc_status_get_start(struct netlink_callback *cb); int nfsd_nl_rpc_status_get_done(struct netlink_callback *cb); @@ -18,6 +21,8 @@ int nfsd_nl_rpc_status_get_dumpit(struct sk_buff *skb, struct netlink_callback *cb); int nfsd_nl_threads_set_doit(struct sk_buff *skb, struct genl_info *info); int nfsd_nl_threads_get_doit(struct sk_buff *skb, struct genl_info *info); +int nfsd_nl_version_set_doit(struct sk_buff *skb, struct genl_info *info); +int nfsd_nl_version_get_doit(struct sk_buff *skb, struct genl_info *info); extern struct genl_family nfsd_nl_family; diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h index d4be519b5734..14ec15656320 100644 --- a/fs/nfsd/netns.h +++ b/fs/nfsd/netns.h @@ -218,6 +218,7 @@ struct nfsd_net { /* Simple check to find out if a given net was properly initialized */ #define nfsd_netns_ready(nn) ((nn)->sessionid_hashtbl) +extern bool nfsd_support_version(int vers); extern void nfsd_netns_free_versions(struct nfsd_net *nn); extern unsigned int nfsd_net_id; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 5bfdebb2e7e9..b6b405437aa0 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -1796,6 +1796,156 @@ err_free_msg: return err; } +/** + * nfsd_nl_version_set_doit - set the nfs enabled versions + * @skb: reply buffer + * @info: netlink metadata and command arguments + * + * Return 0 on success or a negative errno. + */ +int nfsd_nl_version_set_doit(struct sk_buff *skb, struct genl_info *info) +{ + const struct nlattr *attr; + struct nfsd_net *nn; + int i, rem; + + if (GENL_REQ_ATTR_CHECK(info, NFSD_A_SERVER_PROTO_VERSION)) + return -EINVAL; + + mutex_lock(&nfsd_mutex); + + nn = net_generic(genl_info_net(info), nfsd_net_id); + if (nn->nfsd_serv) { + mutex_unlock(&nfsd_mutex); + return -EBUSY; + } + + /* clear current supported versions. */ + nfsd_vers(nn, 2, NFSD_CLEAR); + nfsd_vers(nn, 3, NFSD_CLEAR); + for (i = 0; i <= NFSD_SUPPORTED_MINOR_VERSION; i++) + nfsd_minorversion(nn, i, NFSD_CLEAR); + + nlmsg_for_each_attr(attr, info->nlhdr, GENL_HDRLEN, rem) { + struct nlattr *tb[NFSD_A_VERSION_MAX + 1]; + u32 major, minor = 0; + bool enabled; + + if (nla_type(attr) != NFSD_A_SERVER_PROTO_VERSION) + continue; + + if (nla_parse_nested(tb, NFSD_A_VERSION_MAX, attr, + nfsd_version_nl_policy, info->extack) < 0) + continue; + + if (!tb[NFSD_A_VERSION_MAJOR]) + continue; + + major = nla_get_u32(tb[NFSD_A_VERSION_MAJOR]); + if (tb[NFSD_A_VERSION_MINOR]) + minor = nla_get_u32(tb[NFSD_A_VERSION_MINOR]); + + enabled = nla_get_flag(tb[NFSD_A_VERSION_ENABLED]); + + switch (major) { + case 4: + nfsd_minorversion(nn, minor, enabled ? NFSD_SET : NFSD_CLEAR); + break; + case 3: + case 2: + if (!minor) + nfsd_vers(nn, major, enabled ? NFSD_SET : NFSD_CLEAR); + break; + default: + break; + } + } + + mutex_unlock(&nfsd_mutex); + + return 0; +} + +/** + * nfsd_nl_version_get_doit - get the enabled status for all supported nfs versions + * @skb: reply buffer + * @info: netlink metadata and command arguments + * + * Return 0 on success or a negative errno. + */ +int nfsd_nl_version_get_doit(struct sk_buff *skb, struct genl_info *info) +{ + struct nfsd_net *nn; + int i, err; + void *hdr; + + skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + hdr = genlmsg_iput(skb, info); + if (!hdr) { + err = -EMSGSIZE; + goto err_free_msg; + } + + mutex_lock(&nfsd_mutex); + nn = net_generic(genl_info_net(info), nfsd_net_id); + + for (i = 2; i <= 4; i++) { + int j; + + for (j = 0; j <= NFSD_SUPPORTED_MINOR_VERSION; j++) { + struct nlattr *attr; + + /* Don't record any versions the kernel doesn't have + * compiled in + */ + if (!nfsd_support_version(i)) + continue; + + /* NFSv{2,3} does not support minor numbers */ + if (i < 4 && j) + continue; + + attr = nla_nest_start(skb, + NFSD_A_SERVER_PROTO_VERSION); + if (!attr) { + err = -EINVAL; + goto err_nfsd_unlock; + } + + if (nla_put_u32(skb, NFSD_A_VERSION_MAJOR, i) || + nla_put_u32(skb, NFSD_A_VERSION_MINOR, j)) { + err = -EINVAL; + goto err_nfsd_unlock; + } + + /* Set the enabled flag if the version is enabled */ + if (nfsd_vers(nn, i, NFSD_TEST) && + (i < 4 || nfsd_minorversion(nn, j, NFSD_TEST)) && + nla_put_flag(skb, NFSD_A_VERSION_ENABLED)) { + err = -EINVAL; + goto err_nfsd_unlock; + } + + nla_nest_end(skb, attr); + } + } + + mutex_unlock(&nfsd_mutex); + genlmsg_end(skb, hdr); + + return genlmsg_reply(skb, info); + +err_nfsd_unlock: + mutex_unlock(&nfsd_mutex); +err_free_msg: + nlmsg_free(skb); + + return err; +} + /** * nfsd_net_init - Prepare the nfsd_net portion of a new net namespace * @net: a freshly-created network namespace diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index e25b9b829749..cd9a6a1a9fc8 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -133,8 +133,7 @@ struct svc_program nfsd_program = { .pg_rpcbind_set = nfsd_rpcbind_set, }; -static bool -nfsd_support_version(int vers) +bool nfsd_support_version(int vers) { if (vers >= NFSD_MINVERS && vers < NFSD_NRVERS) return nfsd_version[vers] != NULL; diff --git a/include/uapi/linux/nfsd_netlink.h b/include/uapi/linux/nfsd_netlink.h index 4bbccd5db7cd..7176cd8e6f09 100644 --- a/include/uapi/linux/nfsd_netlink.h +++ b/include/uapi/linux/nfsd_netlink.h @@ -39,10 +39,28 @@ enum { NFSD_A_SERVER_MAX = (__NFSD_A_SERVER_MAX - 1) }; +enum { + NFSD_A_VERSION_MAJOR = 1, + NFSD_A_VERSION_MINOR, + NFSD_A_VERSION_ENABLED, + + __NFSD_A_VERSION_MAX, + NFSD_A_VERSION_MAX = (__NFSD_A_VERSION_MAX - 1) +}; + +enum { + NFSD_A_SERVER_PROTO_VERSION = 1, + + __NFSD_A_SERVER_PROTO_MAX, + NFSD_A_SERVER_PROTO_MAX = (__NFSD_A_SERVER_PROTO_MAX - 1) +}; + enum { NFSD_CMD_RPC_STATUS_GET = 1, NFSD_CMD_THREADS_SET, NFSD_CMD_THREADS_GET, + NFSD_CMD_VERSION_SET, + NFSD_CMD_VERSION_GET, __NFSD_CMD_MAX, NFSD_CMD_MAX = (__NFSD_CMD_MAX - 1) -- cgit v1.2.3-59-g8ed1b From 16a471177496c8e04a9793812c187a2c1a2192fa Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 23 Apr 2024 15:25:44 +0200 Subject: NFSD: add listener-{set,get} netlink command Introduce write_ports netlink command. For listener-set, userspace is expected to provide a NFS listeners list it wants enabled. All other sockets will be closed. Reviewed-by: Jeff Layton Co-developed-by: Jeff Layton Signed-off-by: Jeff Layton Signed-off-by: Lorenzo Bianconi Signed-off-by: Chuck Lever --- Documentation/netlink/specs/nfsd.yaml | 34 ++++++ fs/nfsd/netlink.c | 22 ++++ fs/nfsd/netlink.h | 3 + fs/nfsd/nfsctl.c | 220 ++++++++++++++++++++++++++++++++++ include/uapi/linux/nfsd_netlink.h | 17 +++ 5 files changed, 296 insertions(+) (limited to 'fs') diff --git a/Documentation/netlink/specs/nfsd.yaml b/Documentation/netlink/specs/nfsd.yaml index 77adda1425e2..d21234097167 100644 --- a/Documentation/netlink/specs/nfsd.yaml +++ b/Documentation/netlink/specs/nfsd.yaml @@ -98,6 +98,23 @@ attribute-sets: type: nest nested-attributes: version multi-attr: true + - + name: sock + attributes: + - + name: addr + type: binary + - + name: transport-name + type: string + - + name: server-sock + attributes: + - + name: addr + type: nest + nested-attributes: sock + multi-attr: true operations: list: @@ -163,3 +180,20 @@ operations: reply: attributes: - version + - + name: listener-set + doc: set nfs running sockets + attribute-set: server-sock + flags: [ admin-perm ] + do: + request: + attributes: + - addr + - + name: listener-get + doc: get nfs running listeners + attribute-set: server-sock + do: + reply: + attributes: + - addr diff --git a/fs/nfsd/netlink.c b/fs/nfsd/netlink.c index b23b0b84a59a..62d2586d9902 100644 --- a/fs/nfsd/netlink.c +++ b/fs/nfsd/netlink.c @@ -11,6 +11,11 @@ #include /* Common nested types */ +const struct nla_policy nfsd_sock_nl_policy[NFSD_A_SOCK_TRANSPORT_NAME + 1] = { + [NFSD_A_SOCK_ADDR] = { .type = NLA_BINARY, }, + [NFSD_A_SOCK_TRANSPORT_NAME] = { .type = NLA_NUL_STRING, }, +}; + const struct nla_policy nfsd_version_nl_policy[NFSD_A_VERSION_ENABLED + 1] = { [NFSD_A_VERSION_MAJOR] = { .type = NLA_U32, }, [NFSD_A_VERSION_MINOR] = { .type = NLA_U32, }, @@ -30,6 +35,11 @@ static const struct nla_policy nfsd_version_set_nl_policy[NFSD_A_SERVER_PROTO_VE [NFSD_A_SERVER_PROTO_VERSION] = NLA_POLICY_NESTED(nfsd_version_nl_policy), }; +/* NFSD_CMD_LISTENER_SET - do */ +static const struct nla_policy nfsd_listener_set_nl_policy[NFSD_A_SERVER_SOCK_ADDR + 1] = { + [NFSD_A_SERVER_SOCK_ADDR] = NLA_POLICY_NESTED(nfsd_sock_nl_policy), +}; + /* Ops table for nfsd */ static const struct genl_split_ops nfsd_nl_ops[] = { { @@ -63,6 +73,18 @@ static const struct genl_split_ops nfsd_nl_ops[] = { .doit = nfsd_nl_version_get_doit, .flags = GENL_CMD_CAP_DO, }, + { + .cmd = NFSD_CMD_LISTENER_SET, + .doit = nfsd_nl_listener_set_doit, + .policy = nfsd_listener_set_nl_policy, + .maxattr = NFSD_A_SERVER_SOCK_ADDR, + .flags = GENL_ADMIN_PERM | GENL_CMD_CAP_DO, + }, + { + .cmd = NFSD_CMD_LISTENER_GET, + .doit = nfsd_nl_listener_get_doit, + .flags = GENL_CMD_CAP_DO, + }, }; struct genl_family nfsd_nl_family __ro_after_init = { diff --git a/fs/nfsd/netlink.h b/fs/nfsd/netlink.h index c7c0da275481..e3724637d64d 100644 --- a/fs/nfsd/netlink.h +++ b/fs/nfsd/netlink.h @@ -12,6 +12,7 @@ #include /* Common nested types */ +extern const struct nla_policy nfsd_sock_nl_policy[NFSD_A_SOCK_TRANSPORT_NAME + 1]; extern const struct nla_policy nfsd_version_nl_policy[NFSD_A_VERSION_ENABLED + 1]; int nfsd_nl_rpc_status_get_start(struct netlink_callback *cb); @@ -23,6 +24,8 @@ int nfsd_nl_threads_set_doit(struct sk_buff *skb, struct genl_info *info); int nfsd_nl_threads_get_doit(struct sk_buff *skb, struct genl_info *info); int nfsd_nl_version_set_doit(struct sk_buff *skb, struct genl_info *info); int nfsd_nl_version_get_doit(struct sk_buff *skb, struct genl_info *info); +int nfsd_nl_listener_set_doit(struct sk_buff *skb, struct genl_info *info); +int nfsd_nl_listener_get_doit(struct sk_buff *skb, struct genl_info *info); extern struct genl_family nfsd_nl_family; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index b6b405437aa0..202140df8f82 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -1946,6 +1946,226 @@ err_free_msg: return err; } +/** + * nfsd_nl_listener_set_doit - set the nfs running sockets + * @skb: reply buffer + * @info: netlink metadata and command arguments + * + * Return 0 on success or a negative errno. + */ +int nfsd_nl_listener_set_doit(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct svc_xprt *xprt, *tmp; + const struct nlattr *attr; + struct svc_serv *serv; + LIST_HEAD(permsocks); + struct nfsd_net *nn; + int err, rem; + + mutex_lock(&nfsd_mutex); + + err = nfsd_create_serv(net); + if (err) { + mutex_unlock(&nfsd_mutex); + return err; + } + + nn = net_generic(net, nfsd_net_id); + serv = nn->nfsd_serv; + + spin_lock_bh(&serv->sv_lock); + + /* Move all of the old listener sockets to a temp list */ + list_splice_init(&serv->sv_permsocks, &permsocks); + + /* + * Walk the list of server_socks from userland and move any that match + * back to sv_permsocks + */ + nlmsg_for_each_attr(attr, info->nlhdr, GENL_HDRLEN, rem) { + struct nlattr *tb[NFSD_A_SOCK_MAX + 1]; + const char *xcl_name; + struct sockaddr *sa; + + if (nla_type(attr) != NFSD_A_SERVER_SOCK_ADDR) + continue; + + if (nla_parse_nested(tb, NFSD_A_SOCK_MAX, attr, + nfsd_sock_nl_policy, info->extack) < 0) + continue; + + if (!tb[NFSD_A_SOCK_ADDR] || !tb[NFSD_A_SOCK_TRANSPORT_NAME]) + continue; + + if (nla_len(tb[NFSD_A_SOCK_ADDR]) < sizeof(*sa)) + continue; + + xcl_name = nla_data(tb[NFSD_A_SOCK_TRANSPORT_NAME]); + sa = nla_data(tb[NFSD_A_SOCK_ADDR]); + + /* Put back any matching sockets */ + list_for_each_entry_safe(xprt, tmp, &permsocks, xpt_list) { + /* This shouldn't be possible */ + if (WARN_ON_ONCE(xprt->xpt_net != net)) { + list_move(&xprt->xpt_list, &serv->sv_permsocks); + continue; + } + + /* If everything matches, put it back */ + if (!strcmp(xprt->xpt_class->xcl_name, xcl_name) && + rpc_cmp_addr_port(sa, (struct sockaddr *)&xprt->xpt_local)) { + list_move(&xprt->xpt_list, &serv->sv_permsocks); + break; + } + } + } + + /* For now, no removing old sockets while server is running */ + if (serv->sv_nrthreads && !list_empty(&permsocks)) { + list_splice_init(&permsocks, &serv->sv_permsocks); + spin_unlock_bh(&serv->sv_lock); + err = -EBUSY; + goto out_unlock_mtx; + } + + /* Close the remaining sockets on the permsocks list */ + while (!list_empty(&permsocks)) { + xprt = list_first_entry(&permsocks, struct svc_xprt, xpt_list); + list_move(&xprt->xpt_list, &serv->sv_permsocks); + + /* + * Newly-created sockets are born with the BUSY bit set. Clear + * it if there are no threads, since nothing can pick it up + * in that case. + */ + if (!serv->sv_nrthreads) + clear_bit(XPT_BUSY, &xprt->xpt_flags); + + set_bit(XPT_CLOSE, &xprt->xpt_flags); + spin_unlock_bh(&serv->sv_lock); + svc_xprt_close(xprt); + spin_lock_bh(&serv->sv_lock); + } + + spin_unlock_bh(&serv->sv_lock); + + /* walk list of addrs again, open any that still don't exist */ + nlmsg_for_each_attr(attr, info->nlhdr, GENL_HDRLEN, rem) { + struct nlattr *tb[NFSD_A_SOCK_MAX + 1]; + const char *xcl_name; + struct sockaddr *sa; + int ret; + + if (nla_type(attr) != NFSD_A_SERVER_SOCK_ADDR) + continue; + + if (nla_parse_nested(tb, NFSD_A_SOCK_MAX, attr, + nfsd_sock_nl_policy, info->extack) < 0) + continue; + + if (!tb[NFSD_A_SOCK_ADDR] || !tb[NFSD_A_SOCK_TRANSPORT_NAME]) + continue; + + if (nla_len(tb[NFSD_A_SOCK_ADDR]) < sizeof(*sa)) + continue; + + xcl_name = nla_data(tb[NFSD_A_SOCK_TRANSPORT_NAME]); + sa = nla_data(tb[NFSD_A_SOCK_ADDR]); + + xprt = svc_find_listener(serv, xcl_name, net, sa); + if (xprt) { + svc_xprt_put(xprt); + continue; + } + + ret = svc_xprt_create_from_sa(serv, xcl_name, net, sa, + SVC_SOCK_ANONYMOUS, + get_current_cred()); + /* always save the latest error */ + if (ret < 0) + err = ret; + } + + if (!serv->sv_nrthreads && list_empty(&nn->nfsd_serv->sv_permsocks)) + nfsd_destroy_serv(net); + +out_unlock_mtx: + mutex_unlock(&nfsd_mutex); + + return err; +} + +/** + * nfsd_nl_listener_get_doit - get the nfs running listeners + * @skb: reply buffer + * @info: netlink metadata and command arguments + * + * Return 0 on success or a negative errno. + */ +int nfsd_nl_listener_get_doit(struct sk_buff *skb, struct genl_info *info) +{ + struct svc_xprt *xprt; + struct svc_serv *serv; + struct nfsd_net *nn; + void *hdr; + int err; + + skb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + hdr = genlmsg_iput(skb, info); + if (!hdr) { + err = -EMSGSIZE; + goto err_free_msg; + } + + mutex_lock(&nfsd_mutex); + nn = net_generic(genl_info_net(info), nfsd_net_id); + + /* no nfs server? Just send empty socket list */ + if (!nn->nfsd_serv) + goto out_unlock_mtx; + + serv = nn->nfsd_serv; + spin_lock_bh(&serv->sv_lock); + list_for_each_entry(xprt, &serv->sv_permsocks, xpt_list) { + struct nlattr *attr; + + attr = nla_nest_start(skb, NFSD_A_SERVER_SOCK_ADDR); + if (!attr) { + err = -EINVAL; + goto err_serv_unlock; + } + + if (nla_put_string(skb, NFSD_A_SOCK_TRANSPORT_NAME, + xprt->xpt_class->xcl_name) || + nla_put(skb, NFSD_A_SOCK_ADDR, + sizeof(struct sockaddr_storage), + &xprt->xpt_local)) { + err = -EINVAL; + goto err_serv_unlock; + } + + nla_nest_end(skb, attr); + } + spin_unlock_bh(&serv->sv_lock); +out_unlock_mtx: + mutex_unlock(&nfsd_mutex); + genlmsg_end(skb, hdr); + + return genlmsg_reply(skb, info); + +err_serv_unlock: + spin_unlock_bh(&serv->sv_lock); + mutex_unlock(&nfsd_mutex); +err_free_msg: + nlmsg_free(skb); + + return err; +} + /** * nfsd_net_init - Prepare the nfsd_net portion of a new net namespace * @net: a freshly-created network namespace diff --git a/include/uapi/linux/nfsd_netlink.h b/include/uapi/linux/nfsd_netlink.h index 7176cd8e6f09..24c86dbc7ed5 100644 --- a/include/uapi/linux/nfsd_netlink.h +++ b/include/uapi/linux/nfsd_netlink.h @@ -55,12 +55,29 @@ enum { NFSD_A_SERVER_PROTO_MAX = (__NFSD_A_SERVER_PROTO_MAX - 1) }; +enum { + NFSD_A_SOCK_ADDR = 1, + NFSD_A_SOCK_TRANSPORT_NAME, + + __NFSD_A_SOCK_MAX, + NFSD_A_SOCK_MAX = (__NFSD_A_SOCK_MAX - 1) +}; + +enum { + NFSD_A_SERVER_SOCK_ADDR = 1, + + __NFSD_A_SERVER_SOCK_MAX, + NFSD_A_SERVER_SOCK_MAX = (__NFSD_A_SERVER_SOCK_MAX - 1) +}; + enum { NFSD_CMD_RPC_STATUS_GET = 1, NFSD_CMD_THREADS_SET, NFSD_CMD_THREADS_GET, NFSD_CMD_VERSION_SET, NFSD_CMD_VERSION_GET, + NFSD_CMD_LISTENER_SET, + NFSD_CMD_LISTENER_GET, __NFSD_CMD_MAX, NFSD_CMD_MAX = (__NFSD_CMD_MAX - 1) -- cgit v1.2.3-59-g8ed1b From a8483b9ad92c9d07122efe8697f0f42f6c41d1b1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 30 Apr 2024 16:05:10 -0400 Subject: NFSD: Record status of async copy operation in struct nfsd4_copy After a client has started an asynchronous COPY operation, a subsequent OFFLOAD_STATUS operation will need to report the status code once that COPY operation has completed. The recorded status record will be used by a subsequent patch. Signed-off-by: Chuck Lever --- fs/nfsd/nfs4proc.c | 21 ++++++++++----------- fs/nfsd/xdr4.h | 1 + 2 files changed, 11 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 3dc173b29803..7e6580384cdb 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1737,7 +1737,7 @@ static void cleanup_async_copy(struct nfsd4_copy *copy) nfs4_put_copy(copy); } -static void nfsd4_send_cb_offload(struct nfsd4_copy *copy, __be32 nfserr) +static void nfsd4_send_cb_offload(struct nfsd4_copy *copy) { struct nfsd4_cb_offload *cbo; @@ -1747,12 +1747,12 @@ static void nfsd4_send_cb_offload(struct nfsd4_copy *copy, __be32 nfserr) memcpy(&cbo->co_res, ©->cp_res, sizeof(copy->cp_res)); memcpy(&cbo->co_fh, ©->fh, sizeof(copy->fh)); - cbo->co_nfserr = nfserr; + cbo->co_nfserr = copy->nfserr; nfsd4_init_cb(&cbo->co_cb, copy->cp_clp, &nfsd4_cb_offload_ops, NFSPROC4_CLNT_CB_OFFLOAD); trace_nfsd_cb_offload(copy->cp_clp, &cbo->co_res.cb_stateid, - &cbo->co_fh, copy->cp_count, nfserr); + &cbo->co_fh, copy->cp_count, copy->nfserr); nfsd4_run_cb(&cbo->co_cb); } @@ -1766,7 +1766,6 @@ static void nfsd4_send_cb_offload(struct nfsd4_copy *copy, __be32 nfserr) static int nfsd4_do_async_copy(void *data) { struct nfsd4_copy *copy = (struct nfsd4_copy *)data; - __be32 nfserr; trace_nfsd_copy_do_async(copy); if (nfsd4_ssc_is_inter(copy)) { @@ -1777,24 +1776,24 @@ static int nfsd4_do_async_copy(void *data) if (IS_ERR(filp)) { switch (PTR_ERR(filp)) { case -EBADF: - nfserr = nfserr_wrong_type; + copy->nfserr = nfserr_wrong_type; break; default: - nfserr = nfserr_offload_denied; + copy->nfserr = nfserr_offload_denied; } /* ss_mnt will be unmounted by the laundromat */ goto do_callback; } - nfserr = nfsd4_do_copy(copy, filp, copy->nf_dst->nf_file, - false); + copy->nfserr = nfsd4_do_copy(copy, filp, copy->nf_dst->nf_file, + false); nfsd4_cleanup_inter_ssc(copy->ss_nsui, filp, copy->nf_dst); } else { - nfserr = nfsd4_do_copy(copy, copy->nf_src->nf_file, - copy->nf_dst->nf_file, false); + copy->nfserr = nfsd4_do_copy(copy, copy->nf_src->nf_file, + copy->nf_dst->nf_file, false); } do_callback: - nfsd4_send_cb_offload(copy, nfserr); + nfsd4_send_cb_offload(copy); cleanup_async_copy(copy); return 0; } diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index 446e72b0385e..62805931e857 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -694,6 +694,7 @@ struct nfsd4_copy { #define NFSD4_COPY_F_COMMITTED (3) /* response */ + __be32 nfserr; struct nfsd42_write_res cp_res; struct knfsd_fh fh; -- cgit v1.2.3-59-g8ed1b From cc63c21682a51e24338baf30424b2987e05a556a Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 30 Apr 2024 16:05:11 -0400 Subject: NFSD: Add COPY status code to OFFLOAD_STATUS response Clients that send an OFFLOAD_STATUS might want to distinguish between an async COPY operation that is still running, has completed successfully, or that has failed. The intention of this patch is to make NFSD behave like this: * Copy still running: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied so far, and an empty osr_status array * Copy completed successfully: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status of NFS4_OK * Copy failed: OFFLOAD_STATUS returns NFS4_OK, the number of bytes copied, and an osr_status other than NFS4_OK * Copy operation lost, canceled, or otherwise unrecognized: OFFLOAD_STATUS returns NFS4ERR_BAD_STATEID NB: Though RFC 7862 Section 11.2 lists a small set of NFS status codes that are valid for OFFLOAD_STATUS, there do not seem to be any explicit spec limits on the status codes that may be returned in the osr_status field. At this time we have no unit tests for COPY and its brethren, as pynfs does not yet implement support for NFSv4.2. Signed-off-by: Chuck Lever --- fs/nfsd/nfs4proc.c | 10 ++++++++-- fs/nfsd/nfs4xdr.c | 7 ++++++- fs/nfsd/xdr4.h | 4 +++- 3 files changed, 17 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 7e6580384cdb..ea3cc3e870a7 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1793,6 +1793,7 @@ static int nfsd4_do_async_copy(void *data) } do_callback: + set_bit(NFSD4_COPY_F_COMPLETED, ©->cp_flags); nfsd4_send_cb_offload(copy); cleanup_async_copy(copy); return 0; @@ -2002,11 +2003,16 @@ nfsd4_offload_status(struct svc_rqst *rqstp, struct nfsd4_copy *copy; struct nfs4_client *clp = cstate->clp; + os->completed = false; spin_lock(&clp->async_lock); copy = find_async_copy_locked(clp, &os->stateid); - if (copy) + if (copy) { os->count = copy->cp_res.wr_bytes_written; - else + if (test_bit(NFSD4_COPY_F_COMPLETED, ©->cp_flags)) { + os->completed = true; + os->status = copy->nfserr; + } + } else status = nfserr_bad_stateid; spin_unlock(&clp->async_lock); diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index ea76100e50ca..c7bfd2180e3f 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -5271,7 +5271,12 @@ nfsd4_encode_offload_status(struct nfsd4_compoundres *resp, __be32 nfserr, if (nfserr != nfs_ok) return nfserr; /* osr_complete<1> */ - if (xdr_stream_encode_u32(xdr, 0) != XDR_UNIT) + if (os->completed) { + if (xdr_stream_encode_u32(xdr, 1) != XDR_UNIT) + return nfserr_resource; + if (xdr_stream_encode_be32(xdr, os->status) != XDR_UNIT) + return nfserr_resource; + } else if (xdr_stream_encode_u32(xdr, 0) != XDR_UNIT) return nfserr_resource; return nfs_ok; } diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index 62805931e857..fbdd42cde1fa 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -692,6 +692,7 @@ struct nfsd4_copy { #define NFSD4_COPY_F_INTRA (1) #define NFSD4_COPY_F_SYNCHRONOUS (2) #define NFSD4_COPY_F_COMMITTED (3) +#define NFSD4_COPY_F_COMPLETED (4) /* response */ __be32 nfserr; @@ -754,7 +755,8 @@ struct nfsd4_offload_status { /* response */ u64 count; - u32 status; + __be32 status; + bool completed; }; struct nfsd4_copy_notify { -- cgit v1.2.3-59-g8ed1b From 442d27ff09a218b61020ab56387dbc508ad6bfa6 Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Fri, 3 May 2024 09:09:06 -0400 Subject: nfsd: set security label during create operations When security labeling is enabled, the client can pass a file security label as part of a create operation for the new file, similar to mode and other attributes. At present, the security label is received by nfsd and passed down to nfsd_create_setattr(), but nfsd_setattr() is never called and therefore the label is never set on the new file. This bug may have been introduced on or around commit d6a97d3f589a ("NFSD: add security label to struct nfsd_attrs"). Looking at nfsd_setattr() I am uncertain as to whether the same issue presents for file ACLs and therefore requires a similar fix for those. An alternative approach would be to introduce a new LSM hook to set the "create SID" of the current task prior to the actual file creation, which would atomically label the new inode at creation time. This would be better for SELinux and a similar approach has been used previously (see security_dentry_create_files_as) but perhaps not usable by other LSMs. Reproducer: 1. Install a Linux distro with SELinux - Fedora is easiest 2. git clone https://github.com/SELinuxProject/selinux-testsuite 3. Install the requisite dependencies per selinux-testsuite/README.md 4. Run something like the following script: MOUNT=$HOME/selinux-testsuite sudo systemctl start nfs-server sudo exportfs -o rw,no_root_squash,security_label localhost:$MOUNT sudo mkdir -p /mnt/selinux-testsuite sudo mount -t nfs -o vers=4.2 localhost:$MOUNT /mnt/selinux-testsuite pushd /mnt/selinux-testsuite/ sudo make -C policy load pushd tests/filesystem sudo runcon -t test_filesystem_t ./create_file -f trans_test_file \ -e test_filesystem_filetranscon_t -v sudo rm -f trans_test_file popd sudo make -C policy unload popd sudo umount /mnt/selinux-testsuite sudo exportfs -u localhost:$MOUNT sudo rmdir /mnt/selinux-testsuite sudo systemctl stop nfs-server Expected output: Process context: unconfined_u:unconfined_r:test_filesystem_t:s0-s0:c0.c1023 Created file: trans_test_file File context: unconfined_u:object_r:test_filesystem_filetranscon_t:s0 File context is correct Actual output: Process context: unconfined_u:unconfined_r:test_filesystem_t:s0-s0:c0.c1023 Created file: trans_test_file File context: system_u:object_r:test_file_t:s0 File context error, expected: test_filesystem_filetranscon_t got: test_file_t Signed-off-by: Stephen Smalley Reviewed-by: Jeff Layton Reviewed-by: NeilBrown Signed-off-by: Chuck Lever --- fs/nfsd/vfs.c | 2 +- fs/nfsd/vfs.h | 8 ++++++++ 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 2e41eb4c3cec..29b1f3613800 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1422,7 +1422,7 @@ nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, * Callers expect new file metadata to be committed even * if the attributes have not changed. */ - if (iap->ia_valid) + if (nfsd_attrs_valid(attrs)) status = nfsd_setattr(rqstp, resfhp, attrs, NULL); else status = nfserrno(commit_metadata(resfhp)); diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h index c60fdb6200fd..57cd70062048 100644 --- a/fs/nfsd/vfs.h +++ b/fs/nfsd/vfs.h @@ -60,6 +60,14 @@ static inline void nfsd_attrs_free(struct nfsd_attrs *attrs) posix_acl_release(attrs->na_dpacl); } +static inline bool nfsd_attrs_valid(struct nfsd_attrs *attrs) +{ + struct iattr *iap = attrs->na_iattr; + + return (iap->ia_valid || (attrs->na_seclabel && + attrs->na_seclabel->len)); +} + __be32 nfserrno (int errno); int nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, struct svc_export **expp); -- cgit v1.2.3-59-g8ed1b From 53e4efa470d5fc6a96662d2d3322cfc925818517 Mon Sep 17 00:00:00 2001 From: Mike Marshall Date: Wed, 1 May 2024 16:20:36 -0400 Subject: orangefs: fix out-of-bounds fsid access Arnd Bergmann sent a patch to fsdevel, he says: "orangefs_statfs() copies two consecutive fields of the superblock into the statfs structure, which triggers a warning from the string fortification helpers" Jan Kara suggested an alternate way to do the patch to make it more readable. I ran both ideas through xfstests and both seem fine. This patch is based on Jan Kara's suggestion. Signed-off-by: Mike Marshall --- fs/orangefs/super.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/orangefs/super.c b/fs/orangefs/super.c index 34849b4a3243..907765673765 100644 --- a/fs/orangefs/super.c +++ b/fs/orangefs/super.c @@ -201,7 +201,8 @@ static int orangefs_statfs(struct dentry *dentry, struct kstatfs *buf) (long)new_op->downcall.resp.statfs.files_avail); buf->f_type = sb->s_magic; - memcpy(&buf->f_fsid, &ORANGEFS_SB(sb)->fs_id, sizeof(buf->f_fsid)); + buf->f_fsid.val[0] = ORANGEFS_SB(sb)->fs_id; + buf->f_fsid.val[1] = ORANGEFS_SB(sb)->id; buf->f_bsize = new_op->downcall.resp.statfs.block_size; buf->f_namelen = ORANGEFS_NAME_MAX; -- cgit v1.2.3-59-g8ed1b From 72e71bf0298c7ed985bcd0d3c7ff4ca19de60373 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 6 May 2024 10:14:13 -0400 Subject: bcachefs: Fix a scheduler splat in __bch2_next_write_buffer_flush_journal_buf() We're using mutex_lock() inside a wait_event() conditional - prepare_to_wait() has already flipped task state, so potentially blocking ops need annotation. Signed-off-by: Kent Overstreet --- fs/bcachefs/journal.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index 9c9a25dbd613..9c2af544251a 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -870,6 +870,8 @@ static struct journal_buf *__bch2_next_write_buffer_flush_journal_buf(struct jou { struct journal_buf *ret = NULL; + /* We're inside wait_event(), but using mutex_lock(: */ + sched_annotate_sleep(); mutex_lock(&j->buf_lock); spin_lock(&j->lock); max_seq = min(max_seq, journal_cur_seq(j)); -- cgit v1.2.3-59-g8ed1b From 7ffec9ccdc6ad8356792f9a7823b1fe9c8a10cbf Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 10:55:17 -0400 Subject: bcachefs: don't free error pointers Reported-by: syzbot+3333603f569fc2ef258c@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index be5b47619327..8091d0686029 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -902,7 +902,8 @@ out: bch2_journal_keys_put_initial(c); bch2_find_btree_nodes_exit(&c->found_btree_nodes); } - kfree(clean); + if (!IS_ERR(clean)) + kfree(clean); if (!ret && test_bit(BCH_FS_need_delete_dead_snapshots, &c->flags) && -- cgit v1.2.3-59-g8ed1b From a2ddaf965f6a15c316f483e7446fbe3d81fba27c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 11:06:54 -0400 Subject: bcachefs: bucket_pos_to_bp_noerror() We don't want the assert when we're checking if the backpointer is valid. Reported-by: syzbot+bf7215c0525098e7747a@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 2 +- fs/bcachefs/backpointers.h | 14 ++++++++++---- 2 files changed, 11 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index a20044201002..af7a71de1bdf 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -54,7 +54,7 @@ int bch2_backpointer_invalid(struct bch_fs *c, struct bkey_s_c k, int ret = 0; bkey_fsck_err_on((bp.v->bucket_offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT) >= ca->mi.bucket_size || - !bpos_eq(bp.k->p, bucket_pos_to_bp(c, bucket, bp.v->bucket_offset)), + !bpos_eq(bp.k->p, bucket_pos_to_bp_noerror(ca, bucket, bp.v->bucket_offset)), c, err, backpointer_bucket_offset_wrong, "backpointer bucket_offset wrong"); diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index 85949b9fd880..c1b274eadda1 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -45,6 +45,15 @@ static inline struct bpos bp_pos_to_bucket(const struct bch_fs *c, return POS(bp_pos.inode, sector_to_bucket(ca, bucket_sector)); } +static inline struct bpos bucket_pos_to_bp_noerror(const struct bch_dev *ca, + struct bpos bucket, + u64 bucket_offset) +{ + return POS(bucket.inode, + (bucket_to_sector(ca, bucket.offset) << + MAX_EXTENT_COMPRESS_RATIO_SHIFT) + bucket_offset); +} + /* * Convert from pos in alloc btree + bucket offset to pos in backpointer btree: */ @@ -53,10 +62,7 @@ static inline struct bpos bucket_pos_to_bp(const struct bch_fs *c, u64 bucket_offset) { struct bch_dev *ca = bch_dev_bkey_exists(c, bucket.inode); - struct bpos ret = POS(bucket.inode, - (bucket_to_sector(ca, bucket.offset) << - MAX_EXTENT_COMPRESS_RATIO_SHIFT) + bucket_offset); - + struct bpos ret = bucket_pos_to_bp_noerror(ca, bucket, bucket_offset); EBUG_ON(!bkey_eq(bucket, bp_pos_to_bucket(c, ret))); return ret; } -- cgit v1.2.3-59-g8ed1b From b30b70ad8bffcde513d34d525820ec411f48e3d7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 11:39:53 -0400 Subject: bcachefs: Fix early error path in bch2_fs_btree_key_cache_exit() Reported-by: syzbot+a35cdb62ec34d44fb062@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index e8c1c530cd95..7dafa1accec2 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -956,13 +956,15 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc) } #ifdef __KERNEL__ - for_each_possible_cpu(cpu) { - struct btree_key_cache_freelist *f = - per_cpu_ptr(bc->pcpu_freed, cpu); - - for (i = 0; i < f->nr; i++) { - ck = f->objs[i]; - list_add(&ck->list, &items); + if (bc->pcpu_freed) { + for_each_possible_cpu(cpu) { + struct btree_key_cache_freelist *f = + per_cpu_ptr(bc->pcpu_freed, cpu); + + for (i = 0; i < f->nr; i++) { + ck = f->objs[i]; + list_add(&ck->list, &items); + } } } #endif -- cgit v1.2.3-59-g8ed1b From 4a8521b6bb81abba9d80d60b80908c77c9236ced Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 11:31:22 -0400 Subject: bcachefs: Inodes need extra padding for varint_decode_fast() Reported-by: syzbot+66b9b74f6520068596a9@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/io_write.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index f137252bccc5..b72cf31f7274 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -199,9 +199,6 @@ static inline int bch2_extent_update_i_size_sectors(struct btree_trans *trans, u64 new_i_size, s64 i_sectors_delta) { - struct btree_iter iter; - struct bkey_i *k; - struct bkey_i_inode_v3 *inode; /* * Crazy performance optimization: * Every extent update needs to also update the inode: the inode trigger @@ -214,25 +211,36 @@ static inline int bch2_extent_update_i_size_sectors(struct btree_trans *trans, * lost, but that's fine. */ unsigned inode_update_flags = BTREE_UPDATE_NOJOURNAL; - int ret; - k = bch2_bkey_get_mut_noupdate(trans, &iter, BTREE_ID_inodes, + struct btree_iter iter; + struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, SPOS(0, extent_iter->pos.inode, extent_iter->snapshot), BTREE_ITER_CACHED); - ret = PTR_ERR_OR_ZERO(k); + int ret = bkey_err(k); if (unlikely(ret)) return ret; - if (unlikely(k->k.type != KEY_TYPE_inode_v3)) { - k = bch2_inode_to_v3(trans, k); - ret = PTR_ERR_OR_ZERO(k); + /* + * varint_decode_fast(), in the inode .invalid method, reads up to 7 + * bytes past the end of the buffer: + */ + struct bkey_i *k_mut = bch2_trans_kmalloc_nomemzero(trans, bkey_bytes(k.k) + 8); + ret = PTR_ERR_OR_ZERO(k_mut); + if (unlikely(ret)) + goto err; + + bkey_reassemble(k_mut, k); + + if (unlikely(k_mut->k.type != KEY_TYPE_inode_v3)) { + k_mut = bch2_inode_to_v3(trans, k_mut); + ret = PTR_ERR_OR_ZERO(k_mut); if (unlikely(ret)) goto err; } - inode = bkey_i_to_inode_v3(k); + struct bkey_i_inode_v3 *inode = bkey_i_to_inode_v3(k_mut); if (!(le64_to_cpu(inode->v.bi_flags) & BCH_INODE_i_size_dirty) && new_i_size > le64_to_cpu(inode->v.bi_size)) { -- cgit v1.2.3-59-g8ed1b From feb077c1774e559cddfc41ef26c864780d158fe2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 17:13:21 -0400 Subject: bcachefs: Fix refcount put in sb_field_resize error path Signed-off-by: Kent Overstreet --- fs/bcachefs/super-io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index 08ea3dbbbe97..e7527d551e3c 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -232,7 +232,7 @@ struct bch_sb_field *bch2_sb_field_resize_id(struct bch_sb_handle *sb, struct bch_sb_handle *dev_sb = &ca->disk_sb; if (bch2_sb_realloc(dev_sb, le32_to_cpu(dev_sb->sb->u64s) + d)) { - percpu_ref_put(&ca->ref); + percpu_ref_put(&ca->io_ref); return NULL; } } -- cgit v1.2.3-59-g8ed1b From 1267df40acb2da62b1641abae26132411d093fb3 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 4 May 2024 12:29:46 -0400 Subject: bcachefs: Initialize bch_write_op->failed in inline data path Normally this is initialized in __bch2_write(), which is executed in a loop, but the inline data path skips this. Reported-by: syzbot+fd3ccb331eb21f05d13b@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/io_write.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index b72cf31f7274..40d7df7607df 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1513,6 +1513,8 @@ static void bch2_write_data_inline(struct bch_write_op *op, unsigned data_len) unsigned sectors; int ret; + memset(&op->failed, 0, sizeof(op->failed)); + op->flags |= BCH_WRITE_WROTE_DATA_INLINE; op->flags |= BCH_WRITE_DONE; -- cgit v1.2.3-59-g8ed1b From 3a2d0259274202432e5119463dd4cf5f9fabed98 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 4 May 2024 12:51:49 -0400 Subject: bcachefs: Fix bch2_dev_lookup() refcounting bch2_dev_lookup() is supposed to take a ref on the device it returns, but for_each_member_device() takes refs as it iterates, for_each_member_device_rcu() does not. Signed-off-by: Kent Overstreet --- fs/bcachefs/super.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 88e214c609bb..c2c80e6890ae 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -2004,13 +2004,9 @@ err: /* return with ref on ca->ref: */ struct bch_dev *bch2_dev_lookup(struct bch_fs *c, const char *name) { - rcu_read_lock(); - for_each_member_device_rcu(c, ca, NULL) - if (!strcmp(name, ca->name)) { - rcu_read_unlock(); + for_each_member_device(c, ca) + if (!strcmp(name, ca->name)) return ca; - } - rcu_read_unlock(); return ERR_PTR(-BCH_ERR_ENOENT_dev_not_found); } -- cgit v1.2.3-59-g8ed1b From 18b4abcead744322feb90ba31450e7d770e928bd Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 4 May 2024 12:55:44 -0400 Subject: bcachefs: Fix lifetime issue in device iterator helpers bch2_get_next_dev() and bch2_get_next_online_dev() iterate over devices, dropping and taking refs as they go; we can't access the previous device (for ca->dev_idx) after we've dropped our ref to it, unless we take rcu_read_lock() first. Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-members.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index 5efa64eca5f8..5bf27d30ca29 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -107,10 +107,10 @@ static inline struct bch_dev *__bch2_next_dev(struct bch_fs *c, struct bch_dev * static inline struct bch_dev *bch2_get_next_dev(struct bch_fs *c, struct bch_dev *ca) { + rcu_read_lock(); if (ca) percpu_ref_put(&ca->ref); - rcu_read_lock(); if ((ca = __bch2_next_dev(c, ca, NULL))) percpu_ref_get(&ca->ref); rcu_read_unlock(); @@ -132,10 +132,10 @@ static inline struct bch_dev *bch2_get_next_online_dev(struct bch_fs *c, struct bch_dev *ca, unsigned state_mask) { + rcu_read_lock(); if (ca) percpu_ref_put(&ca->io_ref); - rcu_read_lock(); while ((ca = __bch2_next_dev(c, ca, NULL)) && (!((1 << ca->mi.state) & state_mask) || !percpu_ref_tryget(&ca->io_ref))) -- cgit v1.2.3-59-g8ed1b From db42549d402cb44fe67c95d08f1a9ea902d32e7e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 4 May 2024 13:26:37 -0400 Subject: bcachefs: Add a better limit for maximum number of buckets The bucket_gens array is a single array allocation (one byte per bucket), and kernel allocations are still limited to INT_MAX. Check this limit to avoid failing the bucket_gens array allocation. Reported-by: syzbot+b29f436493184ea42e2b@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 6 ++++++ fs/bcachefs/errcode.h | 1 + fs/bcachefs/sb-members.c | 6 +++--- fs/bcachefs/super.c | 7 +++++++ 4 files changed, 17 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index f7fbfccd2b1e..8345a2b2d05b 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -591,6 +591,12 @@ struct bch_member { __le64 btree_allocated_bitmap; }; +/* + * This limit comes from the bucket_gens array - it's a single allocation, and + * kernel allocation are limited to INT_MAX + */ +#define BCH_MEMBER_NBUCKETS_MAX (INT_MAX - 64) + #define BCH_MEMBER_V1_BYTES 56 LE64_BITMASK(BCH_MEMBER_STATE, struct bch_member, flags, 0, 4) diff --git a/fs/bcachefs/errcode.h b/fs/bcachefs/errcode.h index 01a79fa3eacb..dbe35b80bc0b 100644 --- a/fs/bcachefs/errcode.h +++ b/fs/bcachefs/errcode.h @@ -175,6 +175,7 @@ x(EINVAL, block_size_too_small) \ x(EINVAL, bucket_size_too_small) \ x(EINVAL, device_size_too_small) \ + x(EINVAL, device_size_too_big) \ x(EINVAL, device_not_a_member_of_filesystem) \ x(EINVAL, device_has_been_removed) \ x(EINVAL, device_splitbrain) \ diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 5b8e621ac5eb..44b3f0cb7b49 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -124,9 +124,9 @@ static int validate_member(struct printbuf *err, struct bch_sb *sb, int i) { - if (le64_to_cpu(m.nbuckets) > LONG_MAX) { - prt_printf(err, "device %u: too many buckets (got %llu, max %lu)", - i, le64_to_cpu(m.nbuckets), LONG_MAX); + if (le64_to_cpu(m.nbuckets) > BCH_MEMBER_NBUCKETS_MAX) { + prt_printf(err, "device %u: too many buckets (got %llu, max %u)", + i, le64_to_cpu(m.nbuckets), BCH_MEMBER_NBUCKETS_MAX); return -BCH_ERR_invalid_sb_members; } diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index c2c80e6890ae..dddf57ec4511 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -1959,6 +1959,13 @@ int bch2_dev_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) goto err; } + if (nbuckets > BCH_MEMBER_NBUCKETS_MAX) { + bch_err(ca, "New device size too big (%llu greater than max %u)", + nbuckets, BCH_MEMBER_NBUCKETS_MAX); + ret = -BCH_ERR_device_size_too_big; + goto err; + } + if (bch2_dev_is_online(ca) && get_capacity(ca->disk_sb.bdev->bd_disk) < ca->mi.bucket_size * nbuckets) { -- cgit v1.2.3-59-g8ed1b From 9a0ec045110dbaad4b8d609142b534f913354101 Mon Sep 17 00:00:00 2001 From: Reed Riley Date: Sat, 4 May 2024 22:12:23 +0000 Subject: bcachefs: fix overflow in fiemap filefrag (and potentially other utilities that call fiemap) sometimes pass ULONG_MAX as the length. fiemap_prep clamps excessively large lengths - but the calculation of end can overflow if it occurs before calling fiemap_prep. When this happens, filefrag assumes it has read to the end and exits. Signed-off-by: Reed Riley Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index fce690007edf..6f114803c6f2 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -964,7 +964,6 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, struct btree_iter iter; struct bkey_s_c k; struct bkey_buf cur, prev; - struct bpos end = POS(ei->v.i_ino, (start + len) >> 9); unsigned offset_into_extent, sectors; bool have_extent = false; u32 snapshot; @@ -974,6 +973,7 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, if (ret) return ret; + struct bpos end = POS(ei->v.i_ino, (start + len) >> 9); if (start + len < start) return -EINVAL; -- cgit v1.2.3-59-g8ed1b From 6b8cbfc3db7582d6f26c6b757d8e949174641709 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 5 May 2024 22:02:28 -0400 Subject: bcachefs: Fix assert in bch2_alloc_v4_invalid() Reported-by: syzbot+10827fa6b176e1acf1d0@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 4 ++-- fs/bcachefs/alloc_background.h | 8 ++++++-- 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 4ff56fa4d539..534ba2b02bd6 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -244,10 +244,10 @@ int bch2_alloc_v4_invalid(struct bch_fs *c, struct bkey_s_c k, struct bkey_s_c_alloc_v4 a = bkey_s_c_to_alloc_v4(k); int ret = 0; - bkey_fsck_err_on(alloc_v4_u64s(a.v) > bkey_val_u64s(k.k), c, err, + bkey_fsck_err_on(alloc_v4_u64s_noerror(a.v) > bkey_val_u64s(k.k), c, err, alloc_v4_val_size_bad, "bad val size (%u > %zu)", - alloc_v4_u64s(a.v), bkey_val_u64s(k.k)); + alloc_v4_u64s_noerror(a.v), bkey_val_u64s(k.k)); bkey_fsck_err_on(!BCH_ALLOC_V4_BACKPOINTERS_START(a.v) && BCH_ALLOC_V4_NR_BACKPOINTERS(a.v), c, err, diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 052b2fac25d6..2790e516383d 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -126,13 +126,17 @@ static inline struct bpos alloc_freespace_pos(struct bpos pos, struct bch_alloc_ return pos; } -static inline unsigned alloc_v4_u64s(const struct bch_alloc_v4 *a) +static inline unsigned alloc_v4_u64s_noerror(const struct bch_alloc_v4 *a) { - unsigned ret = (BCH_ALLOC_V4_BACKPOINTERS_START(a) ?: + return (BCH_ALLOC_V4_BACKPOINTERS_START(a) ?: BCH_ALLOC_V4_U64s_V0) + BCH_ALLOC_V4_NR_BACKPOINTERS(a) * (sizeof(struct bch_backpointer) / sizeof(u64)); +} +static inline unsigned alloc_v4_u64s(const struct bch_alloc_v4 *a) +{ + unsigned ret = alloc_v4_u64s_noerror(a); BUG_ON(ret > U8_MAX - BKEY_U64s); return ret; } -- cgit v1.2.3-59-g8ed1b From f39055220f6f98a180e3503fe05bbf9921c425c8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 5 May 2024 22:28:00 -0400 Subject: bcachefs: Add missing validation for superblock section clean We were forgetting to check for jset entries that overrun the end of the section - both in validate and to_text(); to_text() needs to be safe for types that fail to validate. Reported-by: syzbot+c48865e11e7e893ec4ab@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-clean.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index 35ca3f138de6..194e55b11137 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -278,6 +278,17 @@ static int bch2_sb_clean_validate(struct bch_sb *sb, return -BCH_ERR_invalid_sb_clean; } + for (struct jset_entry *entry = clean->start; + entry != vstruct_end(&clean->field); + entry = vstruct_next(entry)) { + if ((void *) vstruct_next(entry) > vstruct_end(&clean->field)) { + prt_str(err, "entry type "); + bch2_prt_jset_entry_type(err, le16_to_cpu(entry->type)); + prt_str(err, " overruns end of section"); + return -BCH_ERR_invalid_sb_clean; + } + } + return 0; } @@ -295,6 +306,9 @@ static void bch2_sb_clean_to_text(struct printbuf *out, struct bch_sb *sb, for (entry = clean->start; entry != vstruct_end(&clean->field); entry = vstruct_next(entry)) { + if ((void *) vstruct_next(entry) > vstruct_end(&clean->field)) + break; + if (entry->type == BCH_JSET_ENTRY_btree_keys && !entry->u64s) continue; -- cgit v1.2.3-59-g8ed1b From 2bb9600d5d4735953c47dd1ee99382c68dd04caa Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 5 May 2024 22:33:05 -0400 Subject: bcachefs: Guard against unknown k.k->type in __bkey_invalid() For forwards compatibility we have to allow unknown key types, and only run the checks that make sense against them. Fix a missing guard on k.k->type being known. Reported-by: syzbot+ae4dc916da3ce51f284f@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/bkey_methods.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c index db336a43fc08..a275a9e8e341 100644 --- a/fs/bcachefs/bkey_methods.c +++ b/fs/bcachefs/bkey_methods.c @@ -171,8 +171,8 @@ int __bch2_bkey_invalid(struct bch_fs *c, struct bkey_s_c k, if (type >= BKEY_TYPE_NR) return 0; - bkey_fsck_err_on((type == BKEY_TYPE_btree || - (flags & BKEY_INVALID_COMMIT)) && + bkey_fsck_err_on(k.k->type < KEY_TYPE_MAX && + (type == BKEY_TYPE_btree || (flags & BKEY_INVALID_COMMIT)) && !(bch2_key_types_allowed[type] & BIT_ULL(k.k->type)), c, err, bkey_invalid_type_for_btree, "invalid key type for btree %s (%s)", -- cgit v1.2.3-59-g8ed1b From 0ec5b3b7ccfcdca02ab322abf86455d0050ae98f Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 5 May 2024 22:44:27 -0400 Subject: bcachefs: Fix shift-by-64 in bformat_needs_redo() Ancient versions of bcachefs produced packed formats that could represent keys that our in memory format cannot represent; bformat_needs_redo() has some tricky shifts to check for this sort of overflow. Reported-by: syzbot+594427aebfefeebe91c6@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/move.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index bf68ea49447b..4d94b7742dbb 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -968,24 +968,30 @@ static bool migrate_btree_pred(struct bch_fs *c, void *arg, return migrate_pred(c, arg, bkey_i_to_s_c(&b->key), io_opts, data_opts); } +/* + * Ancient versions of bcachefs produced packed formats which could represent + * keys that the in memory format cannot represent; this checks for those + * formats so we can get rid of them. + */ static bool bformat_needs_redo(struct bkey_format *f) { - unsigned i; - - for (i = 0; i < f->nr_fields; i++) { + for (unsigned i = 0; i < f->nr_fields; i++) { + unsigned f_bits = f->bits_per_field[i]; unsigned unpacked_bits = bch2_bkey_format_current.bits_per_field[i]; u64 unpacked_mask = ~((~0ULL << 1) << (unpacked_bits - 1)); u64 field_offset = le64_to_cpu(f->field_offset[i]); - if (f->bits_per_field[i] > unpacked_bits) + if (f_bits > unpacked_bits) return true; - if ((f->bits_per_field[i] == unpacked_bits) && field_offset) + if ((f_bits == unpacked_bits) && field_offset) return true; - if (((field_offset + ((1ULL << f->bits_per_field[i]) - 1)) & - unpacked_mask) < - field_offset) + u64 f_mask = f_bits + ? ~((~0ULL << (f_bits - 1)) << 1) + : 0; + + if (((field_offset + f_mask) & unpacked_mask) < field_offset) return true; } -- cgit v1.2.3-59-g8ed1b From 8060bf1d83f7d404bacb0e5a38f2d4d8f4c9dfb7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 5 May 2024 22:56:54 -0400 Subject: bcachefs: Fix snapshot_t() usage in bch2_fs_quota_read_inode() bch2_fs_quota_read_inode() wasn't entirely updated to the bch2_snapshot_tree() helper, which takes rcu lock. Reported-by: syzbot+a3a9a61224ed3b7f0010@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/quota.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/quota.c b/fs/bcachefs/quota.c index e68b34eab90a..556da0738106 100644 --- a/fs/bcachefs/quota.c +++ b/fs/bcachefs/quota.c @@ -560,13 +560,11 @@ static int bch2_fs_quota_read_inode(struct btree_trans *trans, struct bch_fs *c = trans->c; struct bch_inode_unpacked u; struct bch_snapshot_tree s_t; - int ret; + u32 tree = bch2_snapshot_tree(c, k.k->p.snapshot); - ret = bch2_snapshot_tree_lookup(trans, - bch2_snapshot_tree(c, k.k->p.snapshot), &s_t); + int ret = bch2_snapshot_tree_lookup(trans, tree, &s_t); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), c, - "%s: snapshot tree %u not found", __func__, - snapshot_t(c, k.k->p.snapshot)->tree); + "%s: snapshot tree %u not found", __func__, tree); if (ret) return ret; -- cgit v1.2.3-59-g8ed1b From 88ab10186c44d0a8b90842beab8648b5fd14432d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 6 May 2024 08:40:46 -0400 Subject: bcachefs: Add missing skcipher_request_set_callback() call Signed-off-by: Kent Overstreet --- fs/bcachefs/checksum.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index 7ed779b411f6..088fd2e7bdf1 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -102,6 +102,7 @@ static inline int do_encrypt_sg(struct crypto_sync_skcipher *tfm, int ret; skcipher_request_set_sync_tfm(req, tfm); + skcipher_request_set_callback(req, 0, NULL, NULL); skcipher_request_set_crypt(req, sg, sg, len, nonce.d); ret = crypto_skcipher_encrypt(req); -- cgit v1.2.3-59-g8ed1b From 71dac2482ad3c8d4a8b8998a96751f009bad895f Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 6 May 2024 09:10:29 -0400 Subject: bcachefs: BCH_SB_LAYOUT_SIZE_BITS_MAX Define a constant for the max superblock size, to avoid a too-large shift. Reported-by: syzbot+a8b0fb419355c91dda7f@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 2 ++ fs/bcachefs/super-io.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 8345a2b2d05b..2e8b1a489c20 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -903,6 +903,8 @@ unsigned bcachefs_metadata_required_upgrade_below = bcachefs_metadata_version_re #define BCH_SB_SECTOR 8 #define BCH_SB_MEMBERS_MAX 64 /* XXX kill */ +#define BCH_SB_LAYOUT_SIZE_BITS_MAX 16 /* 32 MB */ + struct bch_sb_layout { __uuid_t magic; /* bcachefs superblock UUID */ __u8 layout_type; diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index e7527d551e3c..989d16bba8f0 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -649,7 +649,7 @@ reread: bytes = vstruct_bytes(sb->sb); - if (bytes > 512 << sb->sb->layout.sb_max_size_bits) { + if (bytes > 512ULL << min(BCH_SB_LAYOUT_SIZE_BITS_MAX, sb->sb->layout.sb_max_size_bits)) { prt_printf(err, "Invalid superblock: too big (got %zu bytes, layout max %lu)", bytes, 512UL << sb->sb->layout.sb_max_size_bits); return -BCH_ERR_invalid_sb_too_big; -- cgit v1.2.3-59-g8ed1b From e221c45da3770962418fb30c27d941bbc70d595a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 6 May 2024 12:30:04 -0400 Subject: knfsd: LOOKUP can return an illegal error value The 'NFS error' NFSERR_OPNOTSUPP is not described by any of the official NFS related RFCs, but appears to have snuck into some older .x files for NFSv2. Either way, it is not in RFC1094, RFC1813 or any of the NFSv4 RFCs, so should not be returned by the knfsd server, and particularly not by the "LOOKUP" operation. Instead, let's return NFSERR_STALE, which is more appropriate if the filesystem encodes the filehandle as FILEID_INVALID. Cc: stable@vger.kernel.org Signed-off-by: Trond Myklebust Signed-off-by: Chuck Lever --- fs/nfsd/nfsfh.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 40fecf7b224f..0b75305fb5f5 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -573,7 +573,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, _fh_update(fhp, exp, dentry); if (fhp->fh_handle.fh_fileid_type == FILEID_INVALID) { fh_put(fhp); - return nfserr_opnotsupp; + return nfserr_stale; } return 0; @@ -599,7 +599,7 @@ fh_update(struct svc_fh *fhp) _fh_update(fhp, fhp->fh_export, dentry); if (fhp->fh_handle.fh_fileid_type == FILEID_INVALID) - return nfserr_opnotsupp; + return nfserr_stale; return 0; out_bad: printk(KERN_ERR "fh_update: fh not verified!\n"); -- cgit v1.2.3-59-g8ed1b From 939cb14d51a150e3c12ef7a8ce0ba04ce6131bd2 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 6 May 2024 12:30:05 -0400 Subject: NFS/knfsd: Remove the invalid NFS error 'NFSERR_OPNOTSUPP' NFSERR_OPNOTSUPP is not described by any RFC, and should not be used. Signed-off-by: Trond Myklebust Signed-off-by: Chuck Lever --- fs/nfsd/nfsd.h | 1 - include/trace/misc/nfs.h | 2 -- include/uapi/linux/nfs.h | 1 - 3 files changed, 4 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h index 2f6c6f3815b4..8f4f239d9f8a 100644 --- a/fs/nfsd/nfsd.h +++ b/fs/nfsd/nfsd.h @@ -230,7 +230,6 @@ void nfsd_lockd_shutdown(void); #define nfserr_nospc cpu_to_be32(NFSERR_NOSPC) #define nfserr_rofs cpu_to_be32(NFSERR_ROFS) #define nfserr_mlink cpu_to_be32(NFSERR_MLINK) -#define nfserr_opnotsupp cpu_to_be32(NFSERR_OPNOTSUPP) #define nfserr_nametoolong cpu_to_be32(NFSERR_NAMETOOLONG) #define nfserr_notempty cpu_to_be32(NFSERR_NOTEMPTY) #define nfserr_dquot cpu_to_be32(NFSERR_DQUOT) diff --git a/include/trace/misc/nfs.h b/include/trace/misc/nfs.h index e43e74591561..7b221d51133a 100644 --- a/include/trace/misc/nfs.h +++ b/include/trace/misc/nfs.h @@ -28,7 +28,6 @@ TRACE_DEFINE_ENUM(NFSERR_FBIG); TRACE_DEFINE_ENUM(NFSERR_NOSPC); TRACE_DEFINE_ENUM(NFSERR_ROFS); TRACE_DEFINE_ENUM(NFSERR_MLINK); -TRACE_DEFINE_ENUM(NFSERR_OPNOTSUPP); TRACE_DEFINE_ENUM(NFSERR_NAMETOOLONG); TRACE_DEFINE_ENUM(NFSERR_NOTEMPTY); TRACE_DEFINE_ENUM(NFSERR_DQUOT); @@ -64,7 +63,6 @@ TRACE_DEFINE_ENUM(NFSERR_JUKEBOX); { NFSERR_NOSPC, "NOSPC" }, \ { NFSERR_ROFS, "ROFS" }, \ { NFSERR_MLINK, "MLINK" }, \ - { NFSERR_OPNOTSUPP, "OPNOTSUPP" }, \ { NFSERR_NAMETOOLONG, "NAMETOOLONG" }, \ { NFSERR_NOTEMPTY, "NOTEMPTY" }, \ { NFSERR_DQUOT, "DQUOT" }, \ diff --git a/include/uapi/linux/nfs.h b/include/uapi/linux/nfs.h index 946cb62d64b0..f356f2ba3814 100644 --- a/include/uapi/linux/nfs.h +++ b/include/uapi/linux/nfs.h @@ -61,7 +61,6 @@ NFSERR_NOSPC = 28, /* v2 v3 v4 */ NFSERR_ROFS = 30, /* v2 v3 v4 */ NFSERR_MLINK = 31, /* v3 v4 */ - NFSERR_OPNOTSUPP = 45, /* v2 v3 */ NFSERR_NAMETOOLONG = 63, /* v2 v3 v4 */ NFSERR_NOTEMPTY = 66, /* v2 v3 v4 */ NFSERR_DQUOT = 69, /* v2 v3 v4 */ -- cgit v1.2.3-59-g8ed1b From c1c53c26e3380a79b65e6b53dac6c3c797a7e8f1 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Tue, 7 May 2024 11:04:49 +0200 Subject: gfs2: make timeout values more explicit 'timeout' is a vague name for the return value of wait_event_*_timeout because it actually returns the time left. Because the variable is never used later, just drop the return value. Since variable 'timeout' is then only used to carry a fixed timeout value, drop this in favor of a fixed function argument as in the other call to wait_event_timeout() above. Signed-off-by: Wolfram Sang Signed-off-by: Andreas Gruenbacher --- fs/gfs2/super.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 713dd24082c9..7a5aedfcd52a 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -1262,7 +1262,6 @@ static bool gfs2_upgrade_iopen_glock(struct inode *inode) struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); struct gfs2_holder *gh = &ip->i_iopen_gh; - long timeout = 5 * HZ; int error; gh->gh_flags |= GL_NOCACHE; @@ -1293,10 +1292,10 @@ static bool gfs2_upgrade_iopen_glock(struct inode *inode) if (error) return false; - timeout = wait_event_interruptible_timeout(sdp->sd_async_glock_wait, + wait_event_interruptible_timeout(sdp->sd_async_glock_wait, !test_bit(HIF_WAIT, &gh->gh_iflags) || test_bit(GLF_DEMOTE, &ip->i_gl->gl_flags), - timeout); + 5 * HZ); if (!test_bit(HIF_HOLDER, &gh->gh_iflags)) { gfs2_glock_dq(gh); return false; -- cgit v1.2.3-59-g8ed1b From 54541c1f78e12a78487ae63e2e199a7d4f6dbd26 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 6 May 2024 20:49:24 -0400 Subject: bcachefs: Fix race in bch2_write_super() bch2_write_super() was looping over online devices multiple times - dropping and retaking io_ref each time. This meant it could race with device removal; it could increment the sequence number on a device but fail to write it - and then if the device was re-added, it would get confused the next time around thinking a superblock write was silently dropped. Fix this by taking io_ref once, and stashing pointers to online devices in a darray. Signed-off-by: Kent Overstreet --- fs/bcachefs/super-io.c | 47 ++++++++++++++++++++++++++++++++--------------- 1 file changed, 32 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index 989d16bba8f0..bfdb15e7d778 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -923,6 +923,7 @@ int bch2_write_super(struct bch_fs *c) struct bch_devs_mask sb_written; bool wrote, can_mount_without_written, can_mount_with_written; unsigned degraded_flags = BCH_FORCE_IF_DEGRADED; + DARRAY(struct bch_dev *) online_devices = {}; int ret = 0; trace_and_count(c, write_super, c, _RET_IP_); @@ -935,6 +936,15 @@ int bch2_write_super(struct bch_fs *c) closure_init_stack(cl); memset(&sb_written, 0, sizeof(sb_written)); + for_each_online_member(c, ca) { + ret = darray_push(&online_devices, ca); + if (bch2_fs_fatal_err_on(ret, c, "%s: error allocating online devices", __func__)) { + percpu_ref_put(&ca->io_ref); + goto out; + } + percpu_ref_get(&ca->io_ref); + } + /* Make sure we're using the new magic numbers: */ c->disk_sb.sb->magic = BCHFS_MAGIC; c->disk_sb.sb->layout.magic = BCHFS_MAGIC; @@ -942,8 +952,8 @@ int bch2_write_super(struct bch_fs *c) le64_add_cpu(&c->disk_sb.sb->seq, 1); struct bch_sb_field_members_v2 *mi = bch2_sb_field_get(c->disk_sb.sb, members_v2); - for_each_online_member(c, ca) - __bch2_members_v2_get_mut(mi, ca->dev_idx)->seq = c->disk_sb.sb->seq; + darray_for_each(online_devices, ca) + __bch2_members_v2_get_mut(mi, (*ca)->dev_idx)->seq = c->disk_sb.sb->seq; c->disk_sb.sb->write_time = cpu_to_le64(ktime_get_real_seconds()); if (test_bit(BCH_FS_error, &c->flags)) @@ -959,16 +969,15 @@ int bch2_write_super(struct bch_fs *c) bch2_sb_errors_from_cpu(c); bch2_sb_downgrade_update(c); - for_each_online_member(c, ca) - bch2_sb_from_fs(c, ca); + darray_for_each(online_devices, ca) + bch2_sb_from_fs(c, (*ca)); - for_each_online_member(c, ca) { + darray_for_each(online_devices, ca) { printbuf_reset(&err); - ret = bch2_sb_validate(&ca->disk_sb, &err, WRITE); + ret = bch2_sb_validate(&(*ca)->disk_sb, &err, WRITE); if (ret) { bch2_fs_inconsistent(c, "sb invalid before write: %s", err.buf); - percpu_ref_put(&ca->io_ref); goto out; } } @@ -995,16 +1004,18 @@ int bch2_write_super(struct bch_fs *c) return -BCH_ERR_sb_not_downgraded; } - for_each_online_member(c, ca) { - __set_bit(ca->dev_idx, sb_written.d); - ca->sb_write_error = 0; + darray_for_each(online_devices, ca) { + __set_bit((*ca)->dev_idx, sb_written.d); + (*ca)->sb_write_error = 0; } - for_each_online_member(c, ca) - read_back_super(c, ca); + darray_for_each(online_devices, ca) + read_back_super(c, *ca); closure_sync(cl); - for_each_online_member(c, ca) { + darray_for_each(online_devices, cap) { + struct bch_dev *ca = *cap; + if (ca->sb_write_error) continue; @@ -1031,17 +1042,20 @@ int bch2_write_super(struct bch_fs *c) do { wrote = false; - for_each_online_member(c, ca) + darray_for_each(online_devices, cap) { + struct bch_dev *ca = *cap; if (!ca->sb_write_error && sb < ca->disk_sb.sb->layout.nr_superblocks) { write_one_super(c, ca, sb); wrote = true; } + } closure_sync(cl); sb++; } while (wrote); - for_each_online_member(c, ca) { + darray_for_each(online_devices, cap) { + struct bch_dev *ca = *cap; if (ca->sb_write_error) __clear_bit(ca->dev_idx, sb_written.d); else @@ -1077,6 +1091,9 @@ int bch2_write_super(struct bch_fs *c) out: /* Make new options visible after they're persistent: */ bch2_sb_update(c); + darray_for_each(online_devices, ca) + percpu_ref_put(&(*ca)->io_ref); + darray_exit(&online_devices); printbuf_exit(&err); return ret; } -- cgit v1.2.3-59-g8ed1b From 6e297a73bccf852e7716207caa8eb868737c7155 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 6 May 2024 23:11:43 -0400 Subject: bcachefs: Add missing sched_annotate_sleep() in bch2_journal_flush_seq_async() Signed-off-by: Kent Overstreet --- fs/bcachefs/journal.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index 9c2af544251a..a8b08e76d0d0 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -706,6 +706,12 @@ recheck_need_open: spin_unlock(&j->lock); + /* + * We're called from bch2_journal_flush_seq() -> wait_event(); + * but this might block. We won't usually block, so we won't + * livelock: + */ + sched_annotate_sleep(); ret = bch2_journal_res_get(j, &res, jset_u64s(0), 0); if (ret) return ret; -- cgit v1.2.3-59-g8ed1b From 05aa02438223e19dbc2ee14ee5260d6c559cba52 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 7 Mar 2024 12:11:17 +0000 Subject: btrfs: remove pointless BUG_ON() when creating snapshot When creating a snapshot we first check with btrfs_lookup_dir_item() if there is a name collision in the parent directory and then return an error if there's a collision. Then later on when trying to insert a dir item for the snapshot we BUG_ON() if the return value is -EEXIST or -EOVERFLOW: static noinline int create_pending_snapshot(...) { (...) /* check if there is a file/dir which has the same name. */ dir_item = btrfs_lookup_dir_item(...); (...) ret = btrfs_insert_dir_item(...); /* We have check then name at the beginning, so it is impossible. */ BUG_ON(ret == -EEXIST || ret == -EOVERFLOW); if (ret) { btrfs_abort_transaction(trans, ret); goto fail; } (...) } It's impossible to get the -EEXIST because we previously checked for a potential collision with btrfs_lookup_dir_item() and we know that after that no one could have added a colliding name because at this point the transaction is in its critical section, state TRANS_STATE_COMMIT_DOING, so no one can join this transaction to add a colliding name and neither can anyone start a new transaction to do that. As for the -EOVERFLOW, that can't happen as long as we have the extended references feature enabled, which is a mkfs default for many years now. In either case, the BUG_ON() is excessive as we can properly deal with any error and can abort the transaction and jump to the 'fail' label, in which case we'll also get the useful stack trace (just like a BUG_ON()) from the abort if the error is either -EEXIST or -EOVERFLOW. So remove the BUG_ON(). Reviewed-by: Qu Wenruo Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 85f359e0e0a7..9c5663b43b1c 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1863,8 +1863,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, ret = btrfs_insert_dir_item(trans, &fname.disk_name, BTRFS_I(parent_inode), &key, BTRFS_FT_DIR, index); - /* We have check then name at the beginning, so it is impossible. */ - BUG_ON(ret == -EEXIST || ret == -EOVERFLOW); if (ret) { btrfs_abort_transaction(trans, ret); goto fail; -- cgit v1.2.3-59-g8ed1b From f40ca9cb581009f92f35351acf0129165c3729ee Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 15 Mar 2024 12:15:53 +0000 Subject: btrfs: locking: inline btrfs_tree_lock() and btrfs_tree_read_lock() The functions btrfs_tree_lock() and btrfs_tree_read_lock() are very trivial so that can be made inline and avoid call overhead, as they are very often called inside critical sections (when searching a btree for example, attempting to lock a child node/leaf while holding a lock on the parent). So make them static inline, which even reduces the size of the btrfs module a little bit. Before this change: $ size fs/btrfs/btrfs.ko text data bss dec hex filename 1718786 156276 16920 1891982 1cde8e fs/btrfs/btrfs.ko After this change: $ size fs/btrfs/btrfs.ko text data bss dec hex filename 1718650 156260 16920 1891830 1cddf6 fs/btrfs/btrfs.ko Running fs_mark also showed a tiny improvement with this script: $ cat test.sh #!/bin/bash DEV=/dev/nullb0 MNT=/mnt/nullb0 FILES=100000 THREADS=$(nproc --all) echo "performance" | \ tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor umount $DEV &> /dev/null mkfs.btrfs -f $DEV mount $DEV $MNT OPTS="-S 0 -L 5 -n $FILES -s 0 -t $THREADS -k" for ((i = 1; i <= $THREADS; i++)); do OPTS="$OPTS -d $MNT/d$i" done fs_mark $OPTS umount $MNT Before this change: FSUse% Count Size Files/sec App Overhead 10 1200000 0 180894.0 10705410 16 2400000 0 228211.4 10765738 23 3600000 0 215969.6 11011072 30 4800000 0 199077.1 11145587 46 6000000 0 176624.1 11658470 After this change: FSUse% Count Size Files/sec App Overhead 10 1200000 0 185312.3 10708377 16 2400000 0 229320.4 10858013 23 3600000 0 217958.7 11006167 30 4800000 0 205122.9 11112899 46 6000000 0 178039.1 11438852 Reviewed-by: Anand Jain Reviewed-by: Johannes Thumshirn Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/locking.c | 10 ---------- fs/btrfs/locking.h | 14 ++++++++++++-- 2 files changed, 12 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 99ccab86bb86..1f355ca65910 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -147,11 +147,6 @@ void __btrfs_tree_read_lock(struct extent_buffer *eb, enum btrfs_lock_nesting ne trace_btrfs_tree_read_lock(eb, start_ns); } -void btrfs_tree_read_lock(struct extent_buffer *eb) -{ - __btrfs_tree_read_lock(eb, BTRFS_NESTING_NORMAL); -} - /* * Try-lock for read. * @@ -211,11 +206,6 @@ void __btrfs_tree_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest) trace_btrfs_tree_lock(eb, start_ns); } -void btrfs_tree_lock(struct extent_buffer *eb) -{ - __btrfs_tree_lock(eb, BTRFS_NESTING_NORMAL); -} - /* * Release the write lock. */ diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h index 9576f485a300..c30aff66e86f 100644 --- a/fs/btrfs/locking.h +++ b/fs/btrfs/locking.h @@ -164,11 +164,21 @@ static_assert(BTRFS_NESTING_MAX <= MAX_LOCKDEP_SUBCLASSES, "too many lock subclasses defined"); void __btrfs_tree_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest); -void btrfs_tree_lock(struct extent_buffer *eb); + +static inline void btrfs_tree_lock(struct extent_buffer *eb) +{ + __btrfs_tree_lock(eb, BTRFS_NESTING_NORMAL); +} + void btrfs_tree_unlock(struct extent_buffer *eb); void __btrfs_tree_read_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest); -void btrfs_tree_read_lock(struct extent_buffer *eb); + +static inline void btrfs_tree_read_lock(struct extent_buffer *eb) +{ + __btrfs_tree_read_lock(eb, BTRFS_NESTING_NORMAL); +} + void btrfs_tree_read_unlock(struct extent_buffer *eb); int btrfs_try_tree_read_lock(struct extent_buffer *eb); int btrfs_try_tree_write_lock(struct extent_buffer *eb); -- cgit v1.2.3-59-g8ed1b From 2066bbfccf4a7bb40e61475c1a2aa3e59fad9619 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 15 Mar 2024 12:41:45 +0000 Subject: btrfs: locking: rename __btrfs_tree_lock() and __btrfs_tree_read_lock() The __btrfs_tree_lock() and __btrfs_tree_read_lock() are using a naming with a double underscore prefix, which is specially not proper for exported functions. Remove the double underscore prefix from their name and add the "_nested" suffix. Reviewed-by: Anand Jain Reviewed-by: Johannes Thumshirn Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ctree.c | 12 ++++++------ fs/btrfs/extent-tree.c | 2 +- fs/btrfs/locking.c | 6 +++--- fs/btrfs/locking.h | 8 ++++---- 4 files changed, 14 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index aaf53fd84358..f6a98e7cf006 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1003,7 +1003,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, goto out; } - __btrfs_tree_lock(left, BTRFS_NESTING_LEFT); + btrfs_tree_lock_nested(left, BTRFS_NESTING_LEFT); wret = btrfs_cow_block(trans, root, left, parent, pslot - 1, &left, BTRFS_NESTING_LEFT_COW); @@ -1021,7 +1021,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, goto out; } - __btrfs_tree_lock(right, BTRFS_NESTING_RIGHT); + btrfs_tree_lock_nested(right, BTRFS_NESTING_RIGHT); wret = btrfs_cow_block(trans, root, right, parent, pslot + 1, &right, BTRFS_NESTING_RIGHT_COW); @@ -1205,7 +1205,7 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans, if (IS_ERR(left)) return PTR_ERR(left); - __btrfs_tree_lock(left, BTRFS_NESTING_LEFT); + btrfs_tree_lock_nested(left, BTRFS_NESTING_LEFT); left_nr = btrfs_header_nritems(left); if (left_nr >= BTRFS_NODEPTRS_PER_BLOCK(fs_info) - 1) { @@ -1265,7 +1265,7 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans, if (IS_ERR(right)) return PTR_ERR(right); - __btrfs_tree_lock(right, BTRFS_NESTING_RIGHT); + btrfs_tree_lock_nested(right, BTRFS_NESTING_RIGHT); right_nr = btrfs_header_nritems(right); if (right_nr >= BTRFS_NODEPTRS_PER_BLOCK(fs_info) - 1) { @@ -3267,7 +3267,7 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root if (IS_ERR(right)) return PTR_ERR(right); - __btrfs_tree_lock(right, BTRFS_NESTING_RIGHT); + btrfs_tree_lock_nested(right, BTRFS_NESTING_RIGHT); free_space = btrfs_leaf_free_space(right); if (free_space < data_size) @@ -3483,7 +3483,7 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root if (IS_ERR(left)) return PTR_ERR(left); - __btrfs_tree_lock(left, BTRFS_NESTING_LEFT); + btrfs_tree_lock_nested(left, BTRFS_NESTING_LEFT); free_space = btrfs_leaf_free_space(left); if (free_space < data_size) { diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 257d044bca91..42525dc8a551 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -5101,7 +5101,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root, */ btrfs_set_buffer_lockdep_class(lockdep_owner, buf, level); - __btrfs_tree_lock(buf, nest); + btrfs_tree_lock_nested(buf, nest); btrfs_clear_buffer_dirty(trans, buf); clear_bit(EXTENT_BUFFER_STALE, &buf->bflags); clear_bit(EXTENT_BUFFER_ZONED_ZEROOUT, &buf->bflags); diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 1f355ca65910..508a3fdfcd58 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -129,14 +129,14 @@ static void btrfs_set_eb_lock_owner(struct extent_buffer *eb, pid_t owner) { } */ /* - * __btrfs_tree_read_lock - lock extent buffer for read + * btrfs_tree_read_lock_nested - lock extent buffer for read * @eb: the eb to be locked * @nest: the nesting level to be used for lockdep * * This takes the read lock on the extent buffer, using the specified nesting * level for lockdep purposes. */ -void __btrfs_tree_read_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest) +void btrfs_tree_read_lock_nested(struct extent_buffer *eb, enum btrfs_lock_nesting nest) { u64 start_ns = 0; @@ -193,7 +193,7 @@ void btrfs_tree_read_unlock(struct extent_buffer *eb) * * Returns with the eb->lock write locked. */ -void __btrfs_tree_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest) +void btrfs_tree_lock_nested(struct extent_buffer *eb, enum btrfs_lock_nesting nest) __acquires(&eb->lock) { u64 start_ns = 0; diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h index c30aff66e86f..1bc8e6738879 100644 --- a/fs/btrfs/locking.h +++ b/fs/btrfs/locking.h @@ -163,20 +163,20 @@ enum btrfs_lockdep_trans_states { static_assert(BTRFS_NESTING_MAX <= MAX_LOCKDEP_SUBCLASSES, "too many lock subclasses defined"); -void __btrfs_tree_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest); +void btrfs_tree_lock_nested(struct extent_buffer *eb, enum btrfs_lock_nesting nest); static inline void btrfs_tree_lock(struct extent_buffer *eb) { - __btrfs_tree_lock(eb, BTRFS_NESTING_NORMAL); + btrfs_tree_lock_nested(eb, BTRFS_NESTING_NORMAL); } void btrfs_tree_unlock(struct extent_buffer *eb); -void __btrfs_tree_read_lock(struct extent_buffer *eb, enum btrfs_lock_nesting nest); +void btrfs_tree_read_lock_nested(struct extent_buffer *eb, enum btrfs_lock_nesting nest); static inline void btrfs_tree_read_lock(struct extent_buffer *eb) { - __btrfs_tree_read_lock(eb, BTRFS_NESTING_NORMAL); + btrfs_tree_read_lock_nested(eb, BTRFS_NESTING_NORMAL); } void btrfs_tree_read_unlock(struct extent_buffer *eb); -- cgit v1.2.3-59-g8ed1b From 7938d38b94c98e7a48ddc0a43ddf54482b940b90 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 18 Mar 2024 11:52:00 +0000 Subject: btrfs: remove pointless readahead callback wrapper There's no point in having a static readahead callback in inode.c that does nothing besides calling extent_readahead() from extent_io.c. So just remove the callback at inode.c and rename extent_readahead() to btrfs_readahead(). Reviewed-by: Johannes Thumshirn Reviewed-by: Anand Jain Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 2 +- fs/btrfs/extent_io.h | 2 +- fs/btrfs/inode.c | 5 ----- 3 files changed, 2 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 2776112dbdf8..f863eefe0f1c 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2267,7 +2267,7 @@ int extent_writepages(struct address_space *mapping, return ret; } -void extent_readahead(struct readahead_control *rac) +void btrfs_readahead(struct readahead_control *rac) { struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ | REQ_RAHEAD }; struct page *pagepool[16]; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index e3530d427e1f..eb123b0499e1 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -241,7 +241,7 @@ int extent_writepages(struct address_space *mapping, struct writeback_control *wbc); int btree_write_cache_pages(struct address_space *mapping, struct writeback_control *wbc); -void extent_readahead(struct readahead_control *rac); +void btrfs_readahead(struct readahead_control *rac); int extent_fiemap(struct btrfs_inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len); int set_folio_extent_mapped(struct folio *folio); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7fed887e700c..ce923f207e2d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7929,11 +7929,6 @@ static int btrfs_writepages(struct address_space *mapping, return extent_writepages(mapping, wbc); } -static void btrfs_readahead(struct readahead_control *rac) -{ - extent_readahead(rac); -} - /* * For release_folio() and invalidate_folio() we have a race window where * folio_end_writeback() is called but the subpage spinlock is not yet released. -- cgit v1.2.3-59-g8ed1b From c66f2afc714867cf7e685680d848748e0d636bef Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 18 Mar 2024 11:58:28 +0000 Subject: btrfs: remove pointless writepages callback wrapper There's no point in having a static writepages callback in inode.c that does nothing besides calling extent_writepages from extent_io.c. So just remove the callback at inode.c and rename extent_writepages() to btrfs_writepages(). Reviewed-by: Johannes Thumshirn Reviewed-by: Anand Jain Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 3 +-- fs/btrfs/extent_io.h | 3 +-- fs/btrfs/inode.c | 6 ------ 3 files changed, 2 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index f863eefe0f1c..7bc23e26a530 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2246,8 +2246,7 @@ next_page: submit_write_bio(&bio_ctrl, found_error ? ret : 0); } -int extent_writepages(struct address_space *mapping, - struct writeback_control *wbc) +int btrfs_writepages(struct address_space *mapping, struct writeback_control *wbc) { struct inode *inode = mapping->host; int ret = 0; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index eb123b0499e1..818431b37124 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -237,8 +237,7 @@ int btrfs_read_folio(struct file *file, struct folio *folio); void extent_write_locked_range(struct inode *inode, struct page *locked_page, u64 start, u64 end, struct writeback_control *wbc, bool pages_dirty); -int extent_writepages(struct address_space *mapping, - struct writeback_control *wbc); +int btrfs_writepages(struct address_space *mapping, struct writeback_control *wbc); int btree_write_cache_pages(struct address_space *mapping, struct writeback_control *wbc); void btrfs_readahead(struct readahead_control *rac); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index ce923f207e2d..a6ebaa5438be 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7923,12 +7923,6 @@ static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, return ret; } -static int btrfs_writepages(struct address_space *mapping, - struct writeback_control *wbc) -{ - return extent_writepages(mapping, wbc); -} - /* * For release_folio() and invalidate_folio() we have a race window where * folio_end_writeback() is called but the subpage spinlock is not yet released. -- cgit v1.2.3-59-g8ed1b From c79f57eafc5e01190fa05f2de8cbd5f579d68acd Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 18 Mar 2024 11:08:30 +0000 Subject: btrfs: avoid pointless wake ups of drew lock readers When unlocking a write lock on a drew lock, at btrfs_drew_write_unlock(), it's pointless to wake up tasks waiting to acquire a read lock if we didn't decrement the 'writers' counter down to 0, since a read lock can only be acquired when the counter reaches a value of 0. Doing so is harmless from a functional point of view, but it's not efficient due to unnecessarily waking up tasks just for them to sleep again on the waitqueue. So change this to wake up readers only if we decremented the 'writers' counter to 0. Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/locking.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 508a3fdfcd58..72992e74c479 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -364,8 +364,12 @@ void btrfs_drew_write_lock(struct btrfs_drew_lock *lock) void btrfs_drew_write_unlock(struct btrfs_drew_lock *lock) { - atomic_dec(&lock->writers); - cond_wake_up(&lock->pending_readers); + /* + * atomic_dec_and_test() implies a full barrier, so woken up readers are + * guaranteed to see the decrement. + */ + if (atomic_dec_and_test(&lock->writers)) + wake_up(&lock->pending_readers); } void btrfs_drew_read_lock(struct btrfs_drew_lock *lock) -- cgit v1.2.3-59-g8ed1b From 1e2d1837091bf70f204802bcac48495358e75673 Mon Sep 17 00:00:00 2001 From: Tavian Barnes Date: Mon, 18 Mar 2024 09:56:53 -0400 Subject: btrfs: add helper to clear EXTENT_BUFFER_READING We are clearing the bit and waking up any waiters in two different places. Factor that code out into a static helper function. Reviewed-by: Qu Wenruo Signed-off-by: Tavian Barnes Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 7bc23e26a530..23bdd05b5cec 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4260,6 +4260,13 @@ void set_extent_buffer_uptodate(struct extent_buffer *eb) } } +static void clear_extent_buffer_reading(struct extent_buffer *eb) +{ + clear_bit(EXTENT_BUFFER_READING, &eb->bflags); + smp_mb__after_atomic(); + wake_up_bit(&eb->bflags, EXTENT_BUFFER_READING); +} + static void end_bbio_meta_read(struct btrfs_bio *bbio) { struct extent_buffer *eb = bbio->private; @@ -4294,9 +4301,7 @@ static void end_bbio_meta_read(struct btrfs_bio *bbio) bio_offset += len; } - clear_bit(EXTENT_BUFFER_READING, &eb->bflags); - smp_mb__after_atomic(); - wake_up_bit(&eb->bflags, EXTENT_BUFFER_READING); + clear_extent_buffer_reading(eb); free_extent_buffer(eb); bio_put(&bbio->bio); @@ -4330,9 +4335,7 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num, * will now be set, and we shouldn't read it in again. */ if (unlikely(test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags))) { - clear_bit(EXTENT_BUFFER_READING, &eb->bflags); - smp_mb__after_atomic(); - wake_up_bit(&eb->bflags, EXTENT_BUFFER_READING); + clear_extent_buffer_reading(eb); return 0; } -- cgit v1.2.3-59-g8ed1b From f32f20e2bd1f3b83925f703704840eebb56faedb Mon Sep 17 00:00:00 2001 From: Tavian Barnes Date: Mon, 18 Mar 2024 09:56:54 -0400 Subject: btrfs: warn if EXTENT_BUFFER_UPTODATE is set while reading We recently tracked down a race condition that triggered a read for an extent buffer with EXTENT_BUFFER_UPTODATE already set. While this read was in progress, other concurrent readers would see the UPTODATE bit and return early as if the read was already complete, making accesses to the extent buffer conflict with the read operation that was overwriting it. Add a WARN_ON() to end_bbio_meta_read() for this situation to make similar races easier to spot in the future. Reviewed-by: Qu Wenruo Signed-off-by: Tavian Barnes Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 23bdd05b5cec..ecb18a8db373 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4275,6 +4275,13 @@ static void end_bbio_meta_read(struct btrfs_bio *bbio) struct folio_iter fi; u32 bio_offset = 0; + /* + * If the extent buffer is marked UPTODATE before the read operation + * completes, other calls to read_extent_buffer_pages() will return + * early without waiting for the read to finish, causing data races. + */ + WARN_ON(test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)); + eb->read_mirror = bbio->mirror_num; if (uptodate && -- cgit v1.2.3-59-g8ed1b From c87b979d9f27f4368d01a53d7dd17062d3c13fda Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 11:34:32 +0800 Subject: btrfs: rename err to ret in btrfs_initxattrs() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/xattr.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index 6287763fdccc..15d0999e340e 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -504,7 +504,7 @@ static int btrfs_initxattrs(struct inode *inode, const struct xattr *xattr; unsigned int nofs_flag; char *name; - int err = 0; + int ret = 0; /* * We're holding a transaction handle, so use a NOFS memory allocation @@ -515,7 +515,7 @@ static int btrfs_initxattrs(struct inode *inode, name = kmalloc(XATTR_SECURITY_PREFIX_LEN + strlen(xattr->name) + 1, GFP_KERNEL); if (!name) { - err = -ENOMEM; + ret = -ENOMEM; break; } strcpy(name, XATTR_SECURITY_PREFIX); @@ -524,14 +524,14 @@ static int btrfs_initxattrs(struct inode *inode, if (strcmp(name, XATTR_NAME_CAPS) == 0) clear_bit(BTRFS_INODE_NO_CAP_XATTR, &BTRFS_I(inode)->runtime_flags); - err = btrfs_setxattr(trans, inode, name, xattr->value, + ret = btrfs_setxattr(trans, inode, name, xattr->value, xattr->value_len, 0); kfree(name); - if (err < 0) + if (ret < 0) break; } memalloc_nofs_restore(nofs_flag); - return err; + return ret; } int btrfs_xattr_security_init(struct btrfs_trans_handle *trans, -- cgit v1.2.3-59-g8ed1b From c3a1cc8ff48875b050fda5285ac8a9889eb7898d Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 11:40:23 +0800 Subject: btrfs: rename err to ret in btrfs_rmdir() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a6ebaa5438be..a8964ae2d5d6 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4645,7 +4645,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) { struct inode *inode = d_inode(dentry); struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; - int err = 0; + int ret = 0; struct btrfs_trans_handle *trans; u64 last_unlink_trans; struct fscrypt_name fname; @@ -4661,33 +4661,33 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) return btrfs_delete_subvolume(BTRFS_I(dir), dentry); } - err = fscrypt_setup_filename(dir, &dentry->d_name, 1, &fname); - if (err) - return err; + ret = fscrypt_setup_filename(dir, &dentry->d_name, 1, &fname); + if (ret) + return ret; /* This needs to handle no-key deletions later on */ trans = __unlink_start_trans(BTRFS_I(dir)); if (IS_ERR(trans)) { - err = PTR_ERR(trans); + ret = PTR_ERR(trans); goto out_notrans; } if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { - err = btrfs_unlink_subvol(trans, BTRFS_I(dir), dentry); + ret = btrfs_unlink_subvol(trans, BTRFS_I(dir), dentry); goto out; } - err = btrfs_orphan_add(trans, BTRFS_I(inode)); - if (err) + ret = btrfs_orphan_add(trans, BTRFS_I(inode)); + if (ret) goto out; last_unlink_trans = BTRFS_I(inode)->last_unlink_trans; /* now the directory is empty */ - err = btrfs_unlink_inode(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)), + ret = btrfs_unlink_inode(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)), &fname.disk_name); - if (!err) { + if (!ret) { btrfs_i_size_write(BTRFS_I(inode), 0); /* * Propagate the last_unlink_trans value of the deleted dir to @@ -4709,7 +4709,7 @@ out_notrans: btrfs_btree_balance_dirty(fs_info); fscrypt_free_filename(&fname); - return err; + return ret; } /* -- cgit v1.2.3-59-g8ed1b From 5e45b044b791dea9abdefa0f95ca0c4c5f8e818d Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 11:43:16 +0800 Subject: btrfs: rename err to ret in btrfs_cont_expand() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a8964ae2d5d6..2d75e04a5a9b 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4933,16 +4933,16 @@ int btrfs_cont_expand(struct btrfs_inode *inode, loff_t oldsize, loff_t size) u64 last_byte; u64 cur_offset; u64 hole_size; - int err = 0; + int ret = 0; /* * If our size started in the middle of a block we need to zero out the * rest of the block before we expand the i_size, otherwise we could * expose stale data. */ - err = btrfs_truncate_block(inode, oldsize, 0, 0); - if (err) - return err; + ret = btrfs_truncate_block(inode, oldsize, 0, 0); + if (ret) + return ret; if (size <= hole_start) return 0; @@ -4953,7 +4953,7 @@ int btrfs_cont_expand(struct btrfs_inode *inode, loff_t oldsize, loff_t size) while (1) { em = btrfs_get_extent(inode, NULL, cur_offset, block_end - cur_offset); if (IS_ERR(em)) { - err = PTR_ERR(em); + ret = PTR_ERR(em); em = NULL; break; } @@ -4964,13 +4964,13 @@ int btrfs_cont_expand(struct btrfs_inode *inode, loff_t oldsize, loff_t size) if (!(em->flags & EXTENT_FLAG_PREALLOC)) { struct extent_map *hole_em; - err = maybe_insert_hole(inode, cur_offset, hole_size); - if (err) + ret = maybe_insert_hole(inode, cur_offset, hole_size); + if (ret) break; - err = btrfs_inode_set_file_extent_range(inode, + ret = btrfs_inode_set_file_extent_range(inode, cur_offset, hole_size); - if (err) + if (ret) break; hole_em = alloc_extent_map(); @@ -4991,12 +4991,12 @@ int btrfs_cont_expand(struct btrfs_inode *inode, loff_t oldsize, loff_t size) hole_em->ram_bytes = hole_size; hole_em->generation = btrfs_get_fs_generation(fs_info); - err = btrfs_replace_extent_map_range(inode, hole_em, true); + ret = btrfs_replace_extent_map_range(inode, hole_em, true); free_extent_map(hole_em); } else { - err = btrfs_inode_set_file_extent_range(inode, + ret = btrfs_inode_set_file_extent_range(inode, cur_offset, hole_size); - if (err) + if (ret) break; } next: @@ -5008,7 +5008,7 @@ next: } free_extent_map(em); unlock_extent(io_tree, hole_start, block_end - 1, &cached_state); - return err; + return ret; } static int btrfs_setsize(struct inode *inode, struct iattr *attr) -- cgit v1.2.3-59-g8ed1b From 93bc66f4b6f14203c3f0c490a532e94d6ec92416 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 12:12:04 +0800 Subject: btrfs: rename err to ret in btrfs_ioctl_snap_destroy() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ioctl.c | 66 ++++++++++++++++++++++++++++---------------------------- 1 file changed, 33 insertions(+), 33 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 0493272a7668..16aa667c4778 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2386,7 +2386,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, struct mnt_idmap *idmap = file_mnt_idmap(file); char *subvol_name, *subvol_name_ptr = NULL; int subvol_namelen; - int err = 0; + int ret = 0; bool destroy_parent = false; /* We don't support snapshots with extent tree v2 yet. */ @@ -2402,7 +2402,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, return PTR_ERR(vol_args2); if (vol_args2->flags & ~BTRFS_SUBVOL_DELETE_ARGS_MASK) { - err = -EOPNOTSUPP; + ret = -EOPNOTSUPP; goto out; } @@ -2411,31 +2411,31 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, * name, same as v1 currently does. */ if (!(vol_args2->flags & BTRFS_SUBVOL_SPEC_BY_ID)) { - err = btrfs_check_ioctl_vol_args2_subvol_name(vol_args2); - if (err < 0) + ret = btrfs_check_ioctl_vol_args2_subvol_name(vol_args2); + if (ret < 0) goto out; subvol_name = vol_args2->name; - err = mnt_want_write_file(file); - if (err) + ret = mnt_want_write_file(file); + if (ret) goto out; } else { struct inode *old_dir; if (vol_args2->subvolid < BTRFS_FIRST_FREE_OBJECTID) { - err = -EINVAL; + ret = -EINVAL; goto out; } - err = mnt_want_write_file(file); - if (err) + ret = mnt_want_write_file(file); + if (ret) goto out; dentry = btrfs_get_dentry(fs_info->sb, BTRFS_FIRST_FREE_OBJECTID, vol_args2->subvolid, 0); if (IS_ERR(dentry)) { - err = PTR_ERR(dentry); + ret = PTR_ERR(dentry); goto out_drop_write; } @@ -2455,7 +2455,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, */ dput(dentry); if (IS_ERR(parent)) { - err = PTR_ERR(parent); + ret = PTR_ERR(parent); goto out_drop_write; } old_dir = dir; @@ -2479,14 +2479,14 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, * to delete without an idmapped mount. */ if (old_dir != dir && idmap != &nop_mnt_idmap) { - err = -EOPNOTSUPP; + ret = -EOPNOTSUPP; goto free_parent; } subvol_name_ptr = btrfs_get_subvol_name_from_objectid( fs_info, vol_args2->subvolid); if (IS_ERR(subvol_name_ptr)) { - err = PTR_ERR(subvol_name_ptr); + ret = PTR_ERR(subvol_name_ptr); goto free_parent; } /* subvol_name_ptr is already nul terminated */ @@ -2497,14 +2497,14 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, if (IS_ERR(vol_args)) return PTR_ERR(vol_args); - err = btrfs_check_ioctl_vol_args_path(vol_args); - if (err < 0) + ret = btrfs_check_ioctl_vol_args_path(vol_args); + if (ret < 0) goto out; subvol_name = vol_args->name; - err = mnt_want_write_file(file); - if (err) + ret = mnt_want_write_file(file); + if (ret) goto out; } @@ -2512,26 +2512,26 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, if (strchr(subvol_name, '/') || strncmp(subvol_name, "..", subvol_namelen) == 0) { - err = -EINVAL; + ret = -EINVAL; goto free_subvol_name; } if (!S_ISDIR(dir->i_mode)) { - err = -ENOTDIR; + ret = -ENOTDIR; goto free_subvol_name; } - err = down_write_killable_nested(&dir->i_rwsem, I_MUTEX_PARENT); - if (err == -EINTR) + ret = down_write_killable_nested(&dir->i_rwsem, I_MUTEX_PARENT); + if (ret == -EINTR) goto free_subvol_name; dentry = lookup_one(idmap, subvol_name, parent, subvol_namelen); if (IS_ERR(dentry)) { - err = PTR_ERR(dentry); + ret = PTR_ERR(dentry); goto out_unlock_dir; } if (d_really_is_negative(dentry)) { - err = -ENOENT; + ret = -ENOENT; goto out_dput; } @@ -2551,7 +2551,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, * Users who want to delete empty subvols should try * rmdir(2). */ - err = -EPERM; + ret = -EPERM; if (!btrfs_test_opt(fs_info, USER_SUBVOL_RM_ALLOWED)) goto out_dput; @@ -2562,29 +2562,29 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, * of the subvol, not a random directory contained * within it. */ - err = -EINVAL; + ret = -EINVAL; if (root == dest) goto out_dput; - err = inode_permission(idmap, inode, MAY_WRITE | MAY_EXEC); - if (err) + ret = inode_permission(idmap, inode, MAY_WRITE | MAY_EXEC); + if (ret) goto out_dput; } /* check if subvolume may be deleted by a user */ - err = btrfs_may_delete(idmap, dir, dentry, 1); - if (err) + ret = btrfs_may_delete(idmap, dir, dentry, 1); + if (ret) goto out_dput; if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) { - err = -EINVAL; + ret = -EINVAL; goto out_dput; } btrfs_inode_lock(BTRFS_I(inode), 0); - err = btrfs_delete_subvolume(BTRFS_I(dir), dentry); + ret = btrfs_delete_subvolume(BTRFS_I(dir), dentry); btrfs_inode_unlock(BTRFS_I(inode), 0); - if (!err) + if (!ret) d_delete_notify(dir, dentry); out_dput: @@ -2601,7 +2601,7 @@ out_drop_write: out: kfree(vol_args2); kfree(vol_args); - return err; + return ret; } static int btrfs_ioctl_defrag(struct file *file, void __user *argp) -- cgit v1.2.3-59-g8ed1b From cbb6b5d208855b381f9b0ce075a93766c963f039 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 12:15:16 +0800 Subject: btrfs: rename err to ret in __set_extent_bit() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-io-tree.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent-io-tree.c b/fs/btrfs/extent-io-tree.c index c09b428823d7..0d564860464d 100644 --- a/fs/btrfs/extent-io-tree.c +++ b/fs/btrfs/extent-io-tree.c @@ -1059,7 +1059,7 @@ static int __set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, struct extent_state *prealloc = NULL; struct rb_node **p = NULL; struct rb_node *parent = NULL; - int err = 0; + int ret = 0; u64 last_start; u64 last_end; u32 exclusive_bits = (bits & EXTENT_LOCKED); @@ -1122,7 +1122,7 @@ hit_next: if (state->state & exclusive_bits) { *failed_start = state->start; cache_state(state, failed_state); - err = -EEXIST; + ret = -EEXIST; goto out; } @@ -1158,7 +1158,7 @@ hit_next: if (state->state & exclusive_bits) { *failed_start = start; cache_state(state, failed_state); - err = -EEXIST; + ret = -EEXIST; goto out; } @@ -1175,12 +1175,12 @@ hit_next: prealloc = alloc_extent_state_atomic(prealloc); if (!prealloc) goto search_again; - err = split_state(tree, state, prealloc, start); - if (err) - extent_io_tree_panic(tree, state, "split", err); + ret = split_state(tree, state, prealloc, start); + if (ret) + extent_io_tree_panic(tree, state, "split", ret); prealloc = NULL; - if (err) + if (ret) goto out; if (state->end <= end) { set_state_bits(tree, state, bits, changeset); @@ -1224,8 +1224,8 @@ hit_next: prealloc->end = this_end; inserted_state = insert_state(tree, prealloc, bits, changeset); if (IS_ERR(inserted_state)) { - err = PTR_ERR(inserted_state); - extent_io_tree_panic(tree, prealloc, "insert", err); + ret = PTR_ERR(inserted_state); + extent_io_tree_panic(tree, prealloc, "insert", ret); } cache_state(inserted_state, cached_state); @@ -1244,16 +1244,16 @@ hit_next: if (state->state & exclusive_bits) { *failed_start = start; cache_state(state, failed_state); - err = -EEXIST; + ret = -EEXIST; goto out; } prealloc = alloc_extent_state_atomic(prealloc); if (!prealloc) goto search_again; - err = split_state(tree, state, prealloc, end + 1); - if (err) - extent_io_tree_panic(tree, state, "split", err); + ret = split_state(tree, state, prealloc, end + 1); + if (ret) + extent_io_tree_panic(tree, state, "split", ret); set_state_bits(tree, prealloc, bits, changeset); cache_state(prealloc, cached_state); @@ -1275,7 +1275,7 @@ out: if (prealloc) free_extent_state(prealloc); - return err; + return ret; } -- cgit v1.2.3-59-g8ed1b From d5b634ae1fa428d111bdda05b075d65ffe1549dd Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 12:16:42 +0800 Subject: btrfs: rename err to ret in convert_extent_bit() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-io-tree.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent-io-tree.c b/fs/btrfs/extent-io-tree.c index 0d564860464d..ed2cfc3d5d8a 100644 --- a/fs/btrfs/extent-io-tree.c +++ b/fs/btrfs/extent-io-tree.c @@ -1312,7 +1312,7 @@ int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, struct extent_state *prealloc = NULL; struct rb_node **p = NULL; struct rb_node *parent = NULL; - int err = 0; + int ret = 0; u64 last_start; u64 last_end; bool first_iteration = true; @@ -1351,7 +1351,7 @@ again: if (!state) { prealloc = alloc_extent_state_atomic(prealloc); if (!prealloc) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } prealloc->start = start; @@ -1402,14 +1402,14 @@ hit_next: if (state->start < start) { prealloc = alloc_extent_state_atomic(prealloc); if (!prealloc) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } - err = split_state(tree, state, prealloc, start); - if (err) - extent_io_tree_panic(tree, state, "split", err); + ret = split_state(tree, state, prealloc, start); + if (ret) + extent_io_tree_panic(tree, state, "split", ret); prealloc = NULL; - if (err) + if (ret) goto out; if (state->end <= end) { set_state_bits(tree, state, bits, NULL); @@ -1442,7 +1442,7 @@ hit_next: prealloc = alloc_extent_state_atomic(prealloc); if (!prealloc) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } @@ -1454,8 +1454,8 @@ hit_next: prealloc->end = this_end; inserted_state = insert_state(tree, prealloc, bits, NULL); if (IS_ERR(inserted_state)) { - err = PTR_ERR(inserted_state); - extent_io_tree_panic(tree, prealloc, "insert", err); + ret = PTR_ERR(inserted_state); + extent_io_tree_panic(tree, prealloc, "insert", ret); } cache_state(inserted_state, cached_state); if (inserted_state == prealloc) @@ -1472,13 +1472,13 @@ hit_next: if (state->start <= end && state->end > end) { prealloc = alloc_extent_state_atomic(prealloc); if (!prealloc) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } - err = split_state(tree, state, prealloc, end + 1); - if (err) - extent_io_tree_panic(tree, state, "split", err); + ret = split_state(tree, state, prealloc, end + 1); + if (ret) + extent_io_tree_panic(tree, state, "split", ret); set_state_bits(tree, prealloc, bits, NULL); cache_state(prealloc, cached_state); @@ -1500,7 +1500,7 @@ out: if (prealloc) free_extent_state(prealloc); - return err; + return ret; } /* -- cgit v1.2.3-59-g8ed1b From fdee5e557f44be6badecf795868cfd5567470b03 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 18:08:06 +0800 Subject: btrfs: rename err to ret in __btrfs_end_transaction() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 9c5663b43b1c..df2e58aa824a 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1052,7 +1052,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *info = trans->fs_info; struct btrfs_transaction *cur_trans = trans->transaction; - int err = 0; + int ret = 0; if (refcount_read(&trans->use_count) > 1) { refcount_dec(&trans->use_count); @@ -1091,13 +1091,13 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, if (TRANS_ABORTED(trans) || BTRFS_FS_ERROR(info)) { wake_up_process(info->transaction_kthread); if (TRANS_ABORTED(trans)) - err = trans->aborted; + ret = trans->aborted; else - err = -EROFS; + ret = -EROFS; } kmem_cache_free(btrfs_trans_handle_cachep, trans); - return err; + return ret; } int btrfs_end_transaction(struct btrfs_trans_handle *trans) -- cgit v1.2.3-59-g8ed1b From 04e4e189dddc463a7e51b6b63f0a21b7816465ca Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 19:01:07 +0800 Subject: btrfs: rename err to ret in create_reloc_inode() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/relocation.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index f96f267fb4aa..07f1166fec0f 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -3928,7 +3928,7 @@ static noinline_for_stack struct inode *create_reloc_inode( struct btrfs_trans_handle *trans; struct btrfs_root *root; u64 objectid; - int err = 0; + int ret = 0; root = btrfs_grab_root(fs_info->data_reloc_root); trans = btrfs_start_transaction(root, 6); @@ -3937,31 +3937,31 @@ static noinline_for_stack struct inode *create_reloc_inode( return ERR_CAST(trans); } - err = btrfs_get_free_objectid(root, &objectid); - if (err) + ret = btrfs_get_free_objectid(root, &objectid); + if (ret) goto out; - err = __insert_orphan_inode(trans, root, objectid); - if (err) + ret = __insert_orphan_inode(trans, root, objectid); + if (ret) goto out; inode = btrfs_iget(fs_info->sb, objectid, root); if (IS_ERR(inode)) { delete_orphan_inode(trans, root, objectid); - err = PTR_ERR(inode); + ret = PTR_ERR(inode); inode = NULL; goto out; } BTRFS_I(inode)->index_cnt = group->start; - err = btrfs_orphan_add(trans, BTRFS_I(inode)); + ret = btrfs_orphan_add(trans, BTRFS_I(inode)); out: btrfs_put_root(root); btrfs_end_transaction(trans); btrfs_btree_balance_dirty(fs_info); - if (err) { + if (ret) { iput(inode); - inode = ERR_PTR(err); + inode = ERR_PTR(ret); } return inode; } -- cgit v1.2.3-59-g8ed1b From 35cb2e90f4bafd301813a12c9229b511896b54da Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:41:19 +0800 Subject: btrfs: rename err to ret in btrfs_dirty_pages() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index f9d76072398d..f55ac15d727a 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -128,7 +128,7 @@ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages, struct extent_state **cached, bool noreserve) { struct btrfs_fs_info *fs_info = inode->root->fs_info; - int err = 0; + int ret = 0; int i; u64 num_bytes; u64 start_pos; @@ -158,10 +158,10 @@ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages, EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, cached); - err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block, + ret = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block, extra_bits, cached); - if (err) - return err; + if (ret) + return ret; for (i = 0; i < num_pages; i++) { struct page *p = pages[i]; -- cgit v1.2.3-59-g8ed1b From aefee7f1d892dd6609054bbfc1384a4f0939bbdb Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:44:20 +0800 Subject: btrfs: rename err to ret in prepare_pages() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index f55ac15d727a..c22264c9cc45 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -915,7 +915,7 @@ static noinline int prepare_pages(struct inode *inode, struct page **pages, unsigned long index = pos >> PAGE_SHIFT; gfp_t mask = get_prepare_gfp_flags(inode, nowait); fgf_t fgp_flags = get_prepare_fgp_flags(nowait); - int err = 0; + int ret = 0; int faili; for (i = 0; i < num_pages; i++) { @@ -925,28 +925,28 @@ again: if (!pages[i]) { faili = i - 1; if (nowait) - err = -EAGAIN; + ret = -EAGAIN; else - err = -ENOMEM; + ret = -ENOMEM; goto fail; } - err = set_page_extent_mapped(pages[i]); - if (err < 0) { + ret = set_page_extent_mapped(pages[i]); + if (ret < 0) { faili = i; goto fail; } if (i == 0) - err = prepare_uptodate_page(inode, pages[i], pos, + ret = prepare_uptodate_page(inode, pages[i], pos, force_uptodate); - if (!err && i == num_pages - 1) - err = prepare_uptodate_page(inode, pages[i], + if (!ret && i == num_pages - 1) + ret = prepare_uptodate_page(inode, pages[i], pos + write_bytes, false); - if (err) { + if (ret) { put_page(pages[i]); - if (!nowait && err == -EAGAIN) { - err = 0; + if (!nowait && ret == -EAGAIN) { + ret = 0; goto again; } faili = i - 1; @@ -962,7 +962,7 @@ fail: put_page(pages[faili]); faili--; } - return err; + return ret; } -- cgit v1.2.3-59-g8ed1b From 70f1e5b6db56ae99ede369d25d5996fa50d7bb74 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:47:32 +0800 Subject: btrfs: rename err to ret in btrfs_direct_write() Unify naming of return value to the preferred way. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file.c | 48 ++++++++++++++++++++++++------------------------ 1 file changed, 24 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index c22264c9cc45..0c23053951be 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1465,7 +1465,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from) ssize_t written_buffered; size_t prev_left = 0; loff_t endbyte; - ssize_t err; + ssize_t ret; unsigned int ilock_flags = 0; struct iomap_dio *dio; @@ -1482,9 +1482,9 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from) ilock_flags |= BTRFS_ILOCK_SHARED; relock: - err = btrfs_inode_lock(BTRFS_I(inode), ilock_flags); - if (err < 0) - return err; + ret = btrfs_inode_lock(BTRFS_I(inode), ilock_flags); + if (ret < 0) + return ret; /* Shared lock cannot be used with security bits set. */ if ((ilock_flags & BTRFS_ILOCK_SHARED) && !IS_NOSEC(inode)) { @@ -1493,14 +1493,14 @@ relock: goto relock; } - err = generic_write_checks(iocb, from); - if (err <= 0) { + ret = generic_write_checks(iocb, from); + if (ret <= 0) { btrfs_inode_unlock(BTRFS_I(inode), ilock_flags); - return err; + return ret; } - err = btrfs_write_check(iocb, from, err); - if (err < 0) { + ret = btrfs_write_check(iocb, from, ret); + if (ret < 0) { btrfs_inode_unlock(BTRFS_I(inode), ilock_flags); goto out; } @@ -1552,15 +1552,15 @@ relock: btrfs_inode_unlock(BTRFS_I(inode), ilock_flags); if (IS_ERR_OR_NULL(dio)) - err = PTR_ERR_OR_ZERO(dio); + ret = PTR_ERR_OR_ZERO(dio); else - err = iomap_dio_complete(dio); + ret = iomap_dio_complete(dio); /* No increment (+=) because iomap returns a cumulative value. */ - if (err > 0) - written = err; + if (ret > 0) + written = ret; - if (iov_iter_count(from) > 0 && (err == -EFAULT || err > 0)) { + if (iov_iter_count(from) > 0 && (ret == -EFAULT || ret > 0)) { const size_t left = iov_iter_count(from); /* * We have more data left to write. Try to fault in as many as @@ -1577,7 +1577,7 @@ relock: * to buffered IO in case we haven't made any progress. */ if (left == prev_left) { - err = -ENOTBLK; + ret = -ENOTBLK; } else { fault_in_iov_iter_readable(from, left); prev_left = left; @@ -1586,10 +1586,10 @@ relock: } /* - * If 'err' is -ENOTBLK or we have not written all data, then it means + * If 'ret' is -ENOTBLK or we have not written all data, then it means * we must fallback to buffered IO. */ - if ((err < 0 && err != -ENOTBLK) || !iov_iter_count(from)) + if ((ret < 0 && ret != -ENOTBLK) || !iov_iter_count(from)) goto out; buffered: @@ -1600,14 +1600,14 @@ buffered: * below, we will block when flushing and waiting for the IO. */ if (iocb->ki_flags & IOCB_NOWAIT) { - err = -EAGAIN; + ret = -EAGAIN; goto out; } pos = iocb->ki_pos; written_buffered = btrfs_buffered_write(iocb, from); if (written_buffered < 0) { - err = written_buffered; + ret = written_buffered; goto out; } /* @@ -1615,18 +1615,18 @@ buffered: * able to read what was just written. */ endbyte = pos + written_buffered - 1; - err = btrfs_fdatawrite_range(inode, pos, endbyte); - if (err) + ret = btrfs_fdatawrite_range(inode, pos, endbyte); + if (ret) goto out; - err = filemap_fdatawait_range(inode->i_mapping, pos, endbyte); - if (err) + ret = filemap_fdatawait_range(inode->i_mapping, pos, endbyte); + if (ret) goto out; written += written_buffered; iocb->ki_pos = pos + written_buffered; invalidate_mapping_pages(file->f_mapping, pos >> PAGE_SHIFT, endbyte >> PAGE_SHIFT); out: - return err < 0 ? err : written; + return ret < 0 ? ret : written; } static ssize_t btrfs_encoded_write(struct kiocb *iocb, struct iov_iter *from, -- cgit v1.2.3-59-g8ed1b From 8d6e5f9a0ad7932461645c47fd696e05e446fed2 Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Tue, 23 Jan 2024 13:28:05 -0600 Subject: btrfs: page to folio conversion: prealloc_file_extent_cluster() Convert usage of page to folio in prealloc_file_extent_cluster() Reviewed-by: Boris Burkov Signed-off-by: Goldwyn Rodrigues Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/relocation.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 07f1166fec0f..8da09288aedf 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2859,7 +2859,7 @@ static noinline_for_stack int prealloc_file_extent_cluster( struct address_space *mapping = inode->vfs_inode.i_mapping; struct btrfs_fs_info *fs_info = inode->root->fs_info; const u32 sectorsize = fs_info->sectorsize; - struct page *page; + struct folio *folio; ASSERT(sectorsize < PAGE_SIZE); ASSERT(IS_ALIGNED(i_size, sectorsize)); @@ -2890,16 +2890,16 @@ static noinline_for_stack int prealloc_file_extent_cluster( clear_extent_bits(&inode->io_tree, i_size, round_up(i_size, PAGE_SIZE) - 1, EXTENT_UPTODATE); - page = find_lock_page(mapping, i_size >> PAGE_SHIFT); + folio = filemap_lock_folio(mapping, i_size >> PAGE_SHIFT); /* * If page is freed we don't need to do anything then, as we * will re-read the whole page anyway. */ - if (page) { - btrfs_subpage_clear_uptodate(fs_info, page_folio(page), i_size, + if (!IS_ERR(folio)) { + btrfs_subpage_clear_uptodate(fs_info, folio, i_size, round_up(i_size, PAGE_SIZE) - i_size); - unlock_page(page); - put_page(page); + folio_unlock(folio); + folio_put(folio); } } -- cgit v1.2.3-59-g8ed1b From a16c2c48f438e9edd3330f0754fa129108f2e9b3 Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Tue, 23 Jan 2024 13:28:06 -0600 Subject: btrfs: convert relocate_one_page() to folios and rename Convert page references to folios and call the respective folio functions. Since find_or_create_page() takes a mask argument, call __filemap_get_folio() instead of filemap_grab_folio(). The patch assumes folio size is PAGE_SIZE, add a warning in case it's a higher order that will be implemented in the future. Signed-off-by: Goldwyn Rodrigues Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/relocation.c | 93 ++++++++++++++++++++++++++------------------------- 1 file changed, 47 insertions(+), 46 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 8da09288aedf..21ba5f0bcc44 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2850,7 +2850,7 @@ static noinline_for_stack int prealloc_file_extent_cluster( * btrfs_do_readpage() call of previously relocated file cluster. * * If the current cluster starts in the above range, btrfs_do_readpage() - * will skip the read, and relocate_one_page() will later writeback + * will skip the read, and relocate_one_folio() will later writeback * the padding zeros as new data, causing data corruption. * * Here we have to manually invalidate the range (i_size, PAGE_END + 1). @@ -2984,68 +2984,71 @@ static u64 get_cluster_boundary_end(const struct file_extent_cluster *cluster, return cluster->boundary[cluster_nr + 1] - 1; } -static int relocate_one_page(struct inode *inode, struct file_ra_state *ra, - const struct file_extent_cluster *cluster, - int *cluster_nr, unsigned long page_index) +static int relocate_one_folio(struct inode *inode, struct file_ra_state *ra, + const struct file_extent_cluster *cluster, + int *cluster_nr, unsigned long index) { struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); u64 offset = BTRFS_I(inode)->index_cnt; const unsigned long last_index = (cluster->end - offset) >> PAGE_SHIFT; gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); - struct page *page; - u64 page_start; - u64 page_end; + struct folio *folio; + u64 folio_start; + u64 folio_end; u64 cur; int ret; - ASSERT(page_index <= last_index); - page = find_lock_page(inode->i_mapping, page_index); - if (!page) { + ASSERT(index <= last_index); + folio = filemap_lock_folio(inode->i_mapping, index); + if (IS_ERR(folio)) { page_cache_sync_readahead(inode->i_mapping, ra, NULL, - page_index, last_index + 1 - page_index); - page = find_or_create_page(inode->i_mapping, page_index, mask); - if (!page) - return -ENOMEM; + index, last_index + 1 - index); + folio = __filemap_get_folio(inode->i_mapping, index, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, mask); + if (IS_ERR(folio)) + return PTR_ERR(folio); } - if (PageReadahead(page)) + WARN_ON(folio_order(folio)); + + if (folio_test_readahead(folio)) page_cache_async_readahead(inode->i_mapping, ra, NULL, - page_folio(page), page_index, - last_index + 1 - page_index); + folio, index, + last_index + 1 - index); - if (!PageUptodate(page)) { - btrfs_read_folio(NULL, page_folio(page)); - lock_page(page); - if (!PageUptodate(page)) { + if (!folio_test_uptodate(folio)) { + btrfs_read_folio(NULL, folio); + folio_lock(folio); + if (!folio_test_uptodate(folio)) { ret = -EIO; - goto release_page; + goto release_folio; } } /* - * We could have lost page private when we dropped the lock to read the - * page above, make sure we set_page_extent_mapped here so we have any + * We could have lost folio private when we dropped the lock to read the + * folio above, make sure we set_page_extent_mapped here so we have any * of the subpage blocksize stuff we need in place. */ - ret = set_page_extent_mapped(page); + ret = set_folio_extent_mapped(folio); if (ret < 0) - goto release_page; + goto release_folio; - page_start = page_offset(page); - page_end = page_start + PAGE_SIZE - 1; + folio_start = folio_pos(folio); + folio_end = folio_start + PAGE_SIZE - 1; /* * Start from the cluster, as for subpage case, the cluster can start - * inside the page. + * inside the folio. */ - cur = max(page_start, cluster->boundary[*cluster_nr] - offset); - while (cur <= page_end) { + cur = max(folio_start, cluster->boundary[*cluster_nr] - offset); + while (cur <= folio_end) { struct extent_state *cached_state = NULL; u64 extent_start = cluster->boundary[*cluster_nr] - offset; u64 extent_end = get_cluster_boundary_end(cluster, *cluster_nr) - offset; - u64 clamped_start = max(page_start, extent_start); - u64 clamped_end = min(page_end, extent_end); + u64 clamped_start = max(folio_start, extent_start); + u64 clamped_end = min(folio_end, extent_end); u32 clamped_len = clamped_end + 1 - clamped_start; /* Reserve metadata for this range */ @@ -3053,7 +3056,7 @@ static int relocate_one_page(struct inode *inode, struct file_ra_state *ra, clamped_len, clamped_len, false); if (ret) - goto release_page; + goto release_folio; /* Mark the range delalloc and dirty for later writeback */ lock_extent(&BTRFS_I(inode)->io_tree, clamped_start, clamped_end, @@ -3069,20 +3072,18 @@ static int relocate_one_page(struct inode *inode, struct file_ra_state *ra, clamped_len, true); btrfs_delalloc_release_extents(BTRFS_I(inode), clamped_len); - goto release_page; + goto release_folio; } - btrfs_folio_set_dirty(fs_info, page_folio(page), - clamped_start, clamped_len); + btrfs_folio_set_dirty(fs_info, folio, clamped_start, clamped_len); /* - * Set the boundary if it's inside the page. + * Set the boundary if it's inside the folio. * Data relocation requires the destination extents to have the * same size as the source. * EXTENT_BOUNDARY bit prevents current extent from being merged * with previous extent. */ - if (in_range(cluster->boundary[*cluster_nr] - offset, - page_start, PAGE_SIZE)) { + if (in_range(cluster->boundary[*cluster_nr] - offset, folio_start, PAGE_SIZE)) { u64 boundary_start = cluster->boundary[*cluster_nr] - offset; u64 boundary_end = boundary_start + @@ -3105,8 +3106,8 @@ static int relocate_one_page(struct inode *inode, struct file_ra_state *ra, break; } } - unlock_page(page); - put_page(page); + folio_unlock(folio); + folio_put(folio); balance_dirty_pages_ratelimited(inode->i_mapping); btrfs_throttle(fs_info); @@ -3114,9 +3115,9 @@ static int relocate_one_page(struct inode *inode, struct file_ra_state *ra, ret = -ECANCELED; return ret; -release_page: - unlock_page(page); - put_page(page); +release_folio: + folio_unlock(folio); + folio_put(folio); return ret; } @@ -3151,7 +3152,7 @@ static int relocate_file_extent_cluster(struct inode *inode, last_index = (cluster->end - offset) >> PAGE_SHIFT; for (index = (cluster->start - offset) >> PAGE_SHIFT; index <= last_index && !ret; index++) - ret = relocate_one_page(inode, ra, cluster, &cluster_nr, index); + ret = relocate_one_folio(inode, ra, cluster, &cluster_nr, index); if (ret == 0) WARN_ON(cluster_nr != cluster->nr); out: -- cgit v1.2.3-59-g8ed1b From 01b69bf9906bf5bbea79c5427e318a836d5d14e3 Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Tue, 23 Jan 2024 13:28:07 -0600 Subject: btrfs: convert put_file_data() to folios Use folio instead of page in put_file_data(). Add a warning in case higher order folio is found, this will be implemented in the future. Signed-off-by: Goldwyn Rodrigues Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/send.c | 45 +++++++++++++++++++++++---------------------- 1 file changed, 23 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 50b4a76ac88e..342fe5ff0f05 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -5274,10 +5274,11 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len) { struct btrfs_root *root = sctx->send_root; struct btrfs_fs_info *fs_info = root->fs_info; - struct page *page; + struct folio *folio; pgoff_t index = offset >> PAGE_SHIFT; pgoff_t last_index; unsigned pg_offset = offset_in_page(offset); + struct address_space *mapping = sctx->cur_inode->i_mapping; int ret; ret = put_data_header(sctx, len); @@ -5290,44 +5291,44 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len) unsigned cur_len = min_t(unsigned, len, PAGE_SIZE - pg_offset); - page = find_lock_page(sctx->cur_inode->i_mapping, index); - if (!page) { - page_cache_sync_readahead(sctx->cur_inode->i_mapping, + folio = filemap_lock_folio(mapping, index); + if (IS_ERR(folio)) { + page_cache_sync_readahead(mapping, &sctx->ra, NULL, index, last_index + 1 - index); - page = find_or_create_page(sctx->cur_inode->i_mapping, - index, GFP_KERNEL); - if (!page) { - ret = -ENOMEM; + folio = filemap_grab_folio(mapping, index); + if (IS_ERR(folio)) { + ret = PTR_ERR(folio); break; } } - if (PageReadahead(page)) - page_cache_async_readahead(sctx->cur_inode->i_mapping, - &sctx->ra, NULL, page_folio(page), + WARN_ON(folio_order(folio)); + + if (folio_test_readahead(folio)) + page_cache_async_readahead(mapping, &sctx->ra, NULL, folio, index, last_index + 1 - index); - if (!PageUptodate(page)) { - btrfs_read_folio(NULL, page_folio(page)); - lock_page(page); - if (!PageUptodate(page)) { - unlock_page(page); + if (!folio_test_uptodate(folio)) { + btrfs_read_folio(NULL, folio); + folio_lock(folio); + if (!folio_test_uptodate(folio)) { + folio_unlock(folio); btrfs_err(fs_info, "send: IO error at offset %llu for inode %llu root %llu", - page_offset(page), sctx->cur_ino, + folio_pos(folio), sctx->cur_ino, sctx->send_root->root_key.objectid); - put_page(page); + folio_put(folio); ret = -EIO; break; } } - memcpy_from_page(sctx->send_buf + sctx->send_size, page, - pg_offset, cur_len); - unlock_page(page); - put_page(page); + memcpy_from_folio(sctx->send_buf + sctx->send_size, folio, + pg_offset, cur_len); + folio_unlock(folio); + folio_put(folio); index++; pg_offset = 0; len -= cur_len; -- cgit v1.2.3-59-g8ed1b From 4a43d735a6028a9852e72a3ce02343942fd73c22 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 7 Mar 2024 17:53:47 +0300 Subject: btrfs: qgroup: delete unnecessary check in btrfs_qgroup_check_inherit() This check "if (inherit->num_qgroups > PAGE_SIZE)" is confusing and unnecessary. The problem with the check is that static checkers flag it as a potential mixup of between units of bytes vs number of elements. Fortunately, the check can safely be deleted because the next check is correct and applies an even stricter limit: if (size != struct_size(inherit, qgroups, inherit->num_qgroups)) return -EINVAL; The "inherit" struct ends in a variable array of __u64 and "inherit->num_qgroups" is the number of elements in the array. At the start of the function we check that: if (size < sizeof(*inherit) || size > PAGE_SIZE) return -EINVAL; Thus, since we verify that the whole struct fits within one page, that means that the number of elements in the inherit->qgroups[] array must be less than PAGE_SIZE. Reviewed-by: Qu Wenruo Signed-off-by: Dan Carpenter Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 40e5f7f2fcb7..dbb114be8e79 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3064,9 +3064,6 @@ int btrfs_qgroup_check_inherit(struct btrfs_fs_info *fs_info, if (inherit->num_ref_copies > 0 || inherit->num_excl_copies > 0) return -EINVAL; - if (inherit->num_qgroups > PAGE_SIZE) - return -EINVAL; - if (size != struct_size(inherit, qgroups, inherit->num_qgroups)) return -EINVAL; -- cgit v1.2.3-59-g8ed1b From 5d6f0e9890ed857a0bafb7fa73c85bf49bbe1e14 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 22 Mar 2024 18:23:03 +0000 Subject: btrfs: stop locking the source extent range during reflink Nowadays before starting a reflink operation we do this: 1) Take the VFS lock of the inodes in exclusive mode (a rw semaphore); 2) Take the mmap lock of the inodes (struct btrfs_inode::i_mmap_lock); 3) Flush all delalloc in the source and target ranges; 4) Wait for all ordered extents in the source and target ranges to complete; 5) Lock the source and destination ranges in the inodes' io trees. In step 5 we lock the source range because: 1) We needed to serialize against mmap writes, but that is not needed anymore because nowadays we do that through the inode's i_mmap_lock (step 2). This happens since commit 8c99516a8cdd ("btrfs: exclude mmaps while doing remap"); 2) To serialize against a concurrent relocation and avoid generating a delayed ref for an extent that was just dropped by relocation, see commit d8b552424210 ("Btrfs: fix race between reflink/dedupe and relocation"). Locking the source range however blocks any concurrent reads for that range and makes test case generic/733 fail. So instead of locking the source range during reflinks, make relocation read lock the inode's i_mmap_lock, so that it serializes with a concurrent reflink while still able to run concurrently with mmap writes and allow concurrent reads too. Reviewed-by: Boris Burkov Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/reflink.c | 54 ++++++++++++++++----------------------------------- fs/btrfs/relocation.c | 8 +++++++- 2 files changed, 24 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c index 08d0fb46ceec..f12ba2b75141 100644 --- a/fs/btrfs/reflink.c +++ b/fs/btrfs/reflink.c @@ -616,35 +616,6 @@ out: return ret; } -static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1, - struct inode *inode2, u64 loff2, u64 len) -{ - unlock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1, NULL); - unlock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1, NULL); -} - -static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1, - struct inode *inode2, u64 loff2, u64 len) -{ - u64 range1_end = loff1 + len - 1; - u64 range2_end = loff2 + len - 1; - - if (inode1 < inode2) { - swap(inode1, inode2); - swap(loff1, loff2); - swap(range1_end, range2_end); - } else if (inode1 == inode2 && loff2 < loff1) { - swap(loff1, loff2); - swap(range1_end, range2_end); - } - - lock_extent(&BTRFS_I(inode1)->io_tree, loff1, range1_end, NULL); - lock_extent(&BTRFS_I(inode2)->io_tree, loff2, range2_end, NULL); - - btrfs_assert_inode_range_clean(BTRFS_I(inode1), loff1, range1_end); - btrfs_assert_inode_range_clean(BTRFS_I(inode2), loff2, range2_end); -} - static void btrfs_double_mmap_lock(struct inode *inode1, struct inode *inode2) { if (inode1 < inode2) @@ -662,17 +633,21 @@ static void btrfs_double_mmap_unlock(struct inode *inode1, struct inode *inode2) static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 len, struct inode *dst, u64 dst_loff) { + const u64 end = dst_loff + len - 1; + struct extent_state *cached_state = NULL; struct btrfs_fs_info *fs_info = BTRFS_I(src)->root->fs_info; const u64 bs = fs_info->sectorsize; int ret; /* - * Lock destination range to serialize with concurrent readahead() and - * source range to serialize with relocation. + * Lock destination range to serialize with concurrent readahead(), and + * we are safe from concurrency with relocation of source extents + * because we have already locked the inode's i_mmap_lock in exclusive + * mode. */ - btrfs_double_extent_lock(src, loff, dst, dst_loff, len); + lock_extent(&BTRFS_I(dst)->io_tree, dst_loff, end, &cached_state); ret = btrfs_clone(src, dst, loff, len, ALIGN(len, bs), dst_loff, 1); - btrfs_double_extent_unlock(src, loff, dst, dst_loff, len); + unlock_extent(&BTRFS_I(dst)->io_tree, dst_loff, end, &cached_state); btrfs_btree_balance_dirty(fs_info); @@ -724,6 +699,7 @@ out: static noinline int btrfs_clone_files(struct file *file, struct file *file_src, u64 off, u64 olen, u64 destoff) { + struct extent_state *cached_state = NULL; struct inode *inode = file_inode(file); struct inode *src = file_inode(file_src); struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); @@ -731,6 +707,7 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src, int wb_ret; u64 len = olen; u64 bs = fs_info->sectorsize; + u64 end; /* * VFS's generic_remap_file_range_prep() protects us from cloning the @@ -763,12 +740,15 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src, } /* - * Lock destination range to serialize with concurrent readahead() and - * source range to serialize with relocation. + * Lock destination range to serialize with concurrent readahead(), and + * we are safe from concurrency with relocation of source extents + * because we have already locked the inode's i_mmap_lock in exclusive + * mode. */ - btrfs_double_extent_lock(src, off, inode, destoff, len); + end = destoff + len - 1; + lock_extent(&BTRFS_I(inode)->io_tree, destoff, end, &cached_state); ret = btrfs_clone(src, inode, off, olen, len, destoff, 0); - btrfs_double_extent_unlock(src, off, inode, destoff, len); + unlock_extent(&BTRFS_I(inode)->io_tree, destoff, end, &cached_state); /* * We may have copied an inline extent into a page of the destination diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 21ba5f0bcc44..5c9ef6717f84 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1127,16 +1127,22 @@ int replace_file_extents(struct btrfs_trans_handle *trans, fs_info->sectorsize)); WARN_ON(!IS_ALIGNED(end, fs_info->sectorsize)); end--; + /* Take mmap lock to serialize with reflinks. */ + if (!down_read_trylock(&BTRFS_I(inode)->i_mmap_lock)) + continue; ret = try_lock_extent(&BTRFS_I(inode)->io_tree, key.offset, end, &cached_state); - if (!ret) + if (!ret) { + up_read(&BTRFS_I(inode)->i_mmap_lock); continue; + } btrfs_drop_extent_map_range(BTRFS_I(inode), key.offset, end, true); unlock_extent(&BTRFS_I(inode)->io_tree, key.offset, end, &cached_state); + up_read(&BTRFS_I(inode)->i_mmap_lock); } } -- cgit v1.2.3-59-g8ed1b From 6de3595473b0bae11102ef6db40e6f2334f13ed2 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 29 Jan 2024 20:16:06 +1030 Subject: btrfs: compression: add error handling for missed page cache For all the supported compression algorithms, the compression path would always need to grab the page cache, then do the compression. Normally we would get a page reference without any problem, since the write path should have already locked the pages in the write range. For the sake of error handling, we should handle the page cache miss case. Adds a common wrapper, btrfs_compress_find_get_page(), which calls find_get_page(), and do the error handling along with an error message. Callers inside compression path would only need to call btrfs_compress_find_get_page(), and error out if it returned any error. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/compression.c | 23 +++++++++++++++++++++++ fs/btrfs/compression.h | 3 +++ fs/btrfs/lzo.c | 5 +++-- fs/btrfs/zlib.c | 14 ++++++++++---- fs/btrfs/zstd.c | 9 +++++++-- 5 files changed, 46 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index b2b94009959d..3c9e22b5481f 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -974,6 +974,29 @@ static unsigned int btrfs_compress_set_level(int type, unsigned level) return level; } +/* Wrapper around find_get_page(), with extra error message. */ +int btrfs_compress_find_get_page(struct address_space *mapping, u64 start, + struct page **in_page_ret) +{ + struct page *in_page; + + /* + * The compressed write path should have the page locked already, thus + * we only need to grab one reference of the page cache. + */ + in_page = find_get_page(mapping, start >> PAGE_SHIFT); + if (unlikely(!in_page)) { + struct btrfs_inode *inode = BTRFS_I(mapping->host); + + btrfs_crit(inode->root->fs_info, + "failed to get page cache, root %lld ino %llu file offset %llu", + inode->root->root_key.objectid, btrfs_ino(inode), start); + return -ENOENT; + } + *in_page_ret = in_page; + return 0; +} + /* * Given an address space and start and length, compress the bytes into @pages * that are allocated on demand. diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h index 4691a84ca838..7590dc86d040 100644 --- a/fs/btrfs/compression.h +++ b/fs/btrfs/compression.h @@ -149,6 +149,9 @@ bool btrfs_compress_is_valid_type(const char *str, size_t len); int btrfs_compress_heuristic(struct inode *inode, u64 start, u64 end); +int btrfs_compress_find_get_page(struct address_space *mapping, u64 start, + struct page **in_page_ret); + int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, u64 start, struct page **pages, unsigned long *out_pages, unsigned long *total_in, unsigned long *total_out); diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c index 3e5d3b7028e8..6ac2cd177d44 100644 --- a/fs/btrfs/lzo.c +++ b/fs/btrfs/lzo.c @@ -244,8 +244,9 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, /* Get the input page first */ if (!page_in) { - page_in = find_get_page(mapping, cur_in >> PAGE_SHIFT); - ASSERT(page_in); + ret = btrfs_compress_find_get_page(mapping, cur_in, &page_in); + if (ret < 0) + goto out; } /* Compress at most one sector of data each time */ diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index e5b3f2003896..ad6f011eab69 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -151,9 +151,12 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, if (data_in) { kunmap_local(data_in); put_page(in_page); + data_in = NULL; } - in_page = find_get_page(mapping, - start >> PAGE_SHIFT); + ret = btrfs_compress_find_get_page(mapping, + start, &in_page); + if (ret < 0) + goto out; data_in = kmap_local_page(in_page); copy_page(workspace->buf + i * PAGE_SIZE, data_in); @@ -164,9 +167,12 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, if (data_in) { kunmap_local(data_in); put_page(in_page); + data_in = NULL; } - in_page = find_get_page(mapping, - start >> PAGE_SHIFT); + ret = btrfs_compress_find_get_page(mapping, + start, &in_page); + if (ret < 0) + goto out; data_in = kmap_local_page(in_page); start += PAGE_SIZE; workspace->strm.next_in = data_in; diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c index 92b3744b819b..b647ad036af3 100644 --- a/fs/btrfs/zstd.c +++ b/fs/btrfs/zstd.c @@ -406,7 +406,9 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, } /* map in the first page of input data */ - in_page = find_get_page(mapping, start >> PAGE_SHIFT); + ret = btrfs_compress_find_get_page(mapping, start, &in_page); + if (ret < 0) + goto out; workspace->in_buf.src = kmap_local_page(in_page); workspace->in_buf.pos = 0; workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE); @@ -479,10 +481,13 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, if (workspace->in_buf.pos == workspace->in_buf.size) { tot_in += PAGE_SIZE; kunmap_local(workspace->in_buf.src); + workspace->in_buf.src = NULL; put_page(in_page); start += PAGE_SIZE; len -= PAGE_SIZE; - in_page = find_get_page(mapping, start >> PAGE_SHIFT); + ret = btrfs_compress_find_get_page(mapping, start, &in_page); + if (ret < 0) + goto out; workspace->in_buf.src = kmap_local_page(in_page); workspace->in_buf.pos = 0; workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE); -- cgit v1.2.3-59-g8ed1b From 98fe01af7ebe44bcc11afe4b4d681e27b959adb4 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 29 Jan 2024 20:16:07 +1030 Subject: btrfs: compression: convert page allocation to folio interfaces Currently we have two wrappers to allocate and free a page for compression usage: - btrfs_alloc_compr_page() - btrfs_free_compr_page() The allocator would try to grab a page from the pool, and only allocate a new page if the pool is empty. The reclaimer would check if the pool is full, and if not full it would put the page into the pool. This patch converts both helpers to use folio interfaces, and allowing further conversion of compression path to folios. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/compression.c | 24 ++++++++++++------------ fs/btrfs/compression.h | 4 ++-- fs/btrfs/inode.c | 4 ++-- fs/btrfs/lzo.c | 4 ++-- fs/btrfs/zlib.c | 6 +++--- fs/btrfs/zstd.c | 6 +++--- 6 files changed, 24 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 3c9e22b5481f..03648ec1bf73 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -161,7 +161,7 @@ static int compression_decompress(int type, struct list_head *ws, static void btrfs_free_compressed_pages(struct compressed_bio *cb) { for (unsigned int i = 0; i < cb->nr_pages; i++) - btrfs_free_compr_page(cb->compressed_pages[i]); + btrfs_free_compr_folio(page_folio(cb->compressed_pages[i])); kfree(cb->compressed_pages); } @@ -223,25 +223,25 @@ static unsigned long btrfs_compr_pool_scan(struct shrinker *sh, struct shrink_co /* * Common wrappers for page allocation from compression wrappers */ -struct page *btrfs_alloc_compr_page(void) +struct folio *btrfs_alloc_compr_folio(void) { - struct page *page = NULL; + struct folio *folio = NULL; spin_lock(&compr_pool.lock); if (compr_pool.count > 0) { - page = list_first_entry(&compr_pool.list, struct page, lru); - list_del_init(&page->lru); + folio = list_first_entry(&compr_pool.list, struct folio, lru); + list_del_init(&folio->lru); compr_pool.count--; } spin_unlock(&compr_pool.lock); - if (page) - return page; + if (folio) + return folio; - return alloc_page(GFP_NOFS); + return folio_alloc(GFP_NOFS, 0); } -void btrfs_free_compr_page(struct page *page) +void btrfs_free_compr_folio(struct folio *folio) { bool do_free = false; @@ -249,7 +249,7 @@ void btrfs_free_compr_page(struct page *page) if (compr_pool.count > compr_pool.thresh) { do_free = true; } else { - list_add(&page->lru, &compr_pool.list); + list_add(&folio->lru, &compr_pool.list); compr_pool.count++; } spin_unlock(&compr_pool.lock); @@ -257,8 +257,8 @@ void btrfs_free_compr_page(struct page *page) if (!do_free) return; - ASSERT(page_ref_count(page) == 1); - put_page(page); + ASSERT(folio_ref_count(folio) == 1); + folio_put(folio); } static void end_bbio_comprssed_read(struct btrfs_bio *bbio) diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h index 7590dc86d040..2673c25415e5 100644 --- a/fs/btrfs/compression.h +++ b/fs/btrfs/compression.h @@ -104,8 +104,8 @@ void btrfs_submit_compressed_read(struct btrfs_bio *bbio); unsigned int btrfs_compress_str2level(unsigned int type, const char *str); -struct page *btrfs_alloc_compr_page(void); -void btrfs_free_compr_page(struct page *page); +struct folio *btrfs_alloc_compr_folio(void); +void btrfs_free_compr_folio(struct folio *folio); enum btrfs_compression_type { BTRFS_COMPRESS_NONE = 0, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 2d75e04a5a9b..2deb662f171c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1047,7 +1047,7 @@ free_pages: if (pages) { for (i = 0; i < nr_pages; i++) { WARN_ON(pages[i]->mapping); - btrfs_free_compr_page(pages[i]); + btrfs_free_compr_folio(page_folio(pages[i])); } kfree(pages); } @@ -1062,7 +1062,7 @@ static void free_async_extent_pages(struct async_extent *async_extent) for (i = 0; i < async_extent->nr_pages; i++) { WARN_ON(async_extent->pages[i]->mapping); - btrfs_free_compr_page(async_extent->pages[i]); + btrfs_free_compr_folio(page_folio(async_extent->pages[i])); } kfree(async_extent->pages); async_extent->nr_pages = 0; diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c index 6ac2cd177d44..c0e8894c9727 100644 --- a/fs/btrfs/lzo.c +++ b/fs/btrfs/lzo.c @@ -152,7 +152,7 @@ static int copy_compressed_data_to_page(char *compressed_data, cur_page = out_pages[*cur_out / PAGE_SIZE]; /* Allocate a new page */ if (!cur_page) { - cur_page = btrfs_alloc_compr_page(); + cur_page = folio_page(btrfs_alloc_compr_folio(), 0); if (!cur_page) return -ENOMEM; out_pages[*cur_out / PAGE_SIZE] = cur_page; @@ -178,7 +178,7 @@ static int copy_compressed_data_to_page(char *compressed_data, cur_page = out_pages[*cur_out / PAGE_SIZE]; /* Allocate a new page */ if (!cur_page) { - cur_page = btrfs_alloc_compr_page(); + cur_page = folio_page(btrfs_alloc_compr_folio(), 0); if (!cur_page) return -ENOMEM; out_pages[*cur_out / PAGE_SIZE] = cur_page; diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index ad6f011eab69..c260e42b0a33 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -121,7 +121,7 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, workspace->strm.total_in = 0; workspace->strm.total_out = 0; - out_page = btrfs_alloc_compr_page(); + out_page = folio_page(btrfs_alloc_compr_folio(), 0); if (out_page == NULL) { ret = -ENOMEM; goto out; @@ -206,7 +206,7 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, ret = -E2BIG; goto out; } - out_page = btrfs_alloc_compr_page(); + out_page = folio_page(btrfs_alloc_compr_folio(), 0); if (out_page == NULL) { ret = -ENOMEM; goto out; @@ -242,7 +242,7 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, ret = -E2BIG; goto out; } - out_page = btrfs_alloc_compr_page(); + out_page = folio_page(btrfs_alloc_compr_folio(), 0); if (out_page == NULL) { ret = -ENOMEM; goto out; diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c index b647ad036af3..4ec5dd84b93e 100644 --- a/fs/btrfs/zstd.c +++ b/fs/btrfs/zstd.c @@ -414,7 +414,7 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE); /* Allocate and map in the output buffer */ - out_page = btrfs_alloc_compr_page(); + out_page = folio_page(btrfs_alloc_compr_folio(), 0); if (out_page == NULL) { ret = -ENOMEM; goto out; @@ -459,7 +459,7 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, ret = -E2BIG; goto out; } - out_page = btrfs_alloc_compr_page(); + out_page = folio_page(btrfs_alloc_compr_folio(), 0); if (out_page == NULL) { ret = -ENOMEM; goto out; @@ -519,7 +519,7 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, ret = -E2BIG; goto out; } - out_page = btrfs_alloc_compr_page(); + out_page = folio_page(btrfs_alloc_compr_folio(), 0); if (out_page == NULL) { ret = -ENOMEM; goto out; -- cgit v1.2.3-59-g8ed1b From eb1fa9ab470285d39bee268bc2a90d1786610711 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 29 Jan 2024 20:16:08 +1030 Subject: btrfs: make insert_inline_extent() accept one page directly Since our inline extent cannot accept anything larger than a sector, there is really no need to pass all the compressed pages to insert_inline_extent(). And just in case, expand the ASSERT()s to make sure we only try inline with compressed size no larger than sectorsize. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 47 +++++++++++++++++++++++++---------------------- 1 file changed, 25 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 2deb662f171c..9096f8a34be9 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -512,12 +512,13 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, struct btrfs_inode *inode, bool extent_inserted, size_t size, size_t compressed_size, int compress_type, - struct page **compressed_pages, + struct page *compressed_page, bool update_i_size) { struct btrfs_root *root = inode->root; struct extent_buffer *leaf; struct page *page = NULL; + const u32 sectorsize = trans->fs_info->sectorsize; char *kaddr; unsigned long ptr; struct btrfs_file_extent_item *ei; @@ -525,10 +526,23 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, size_t cur_size = size; u64 i_size; - ASSERT((compressed_size > 0 && compressed_pages) || - (compressed_size == 0 && !compressed_pages)); + /* + * The decompressed size must still be no larger than a sector. Under + * heavy race, we can have size == 0 passed in, but that shouldn't be a + * big deal and we can continue the insertion. + */ + ASSERT(size <= sectorsize); - if (compressed_size && compressed_pages) + /* + * The compressed size also needs to be no larger than a sector. + * That's also why we only need one page as the parameter. + */ + if (compressed_page) + ASSERT(compressed_size <= sectorsize); + else + ASSERT(compressed_size == 0); + + if (compressed_size && compressed_page) cur_size = compressed_size; if (!extent_inserted) { @@ -556,21 +570,10 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, ptr = btrfs_file_extent_inline_start(ei); if (compress_type != BTRFS_COMPRESS_NONE) { - struct page *cpage; - int i = 0; - while (compressed_size > 0) { - cpage = compressed_pages[i]; - cur_size = min_t(unsigned long, compressed_size, - PAGE_SIZE); - - kaddr = kmap_local_page(cpage); - write_extent_buffer(leaf, kaddr, ptr, cur_size); - kunmap_local(kaddr); + kaddr = kmap_local_page(compressed_page); + write_extent_buffer(leaf, kaddr, ptr, compressed_size); + kunmap_local(kaddr); - i++; - ptr += cur_size; - compressed_size -= cur_size; - } btrfs_set_file_extent_compression(leaf, ei, compress_type); } else { @@ -620,7 +623,7 @@ fail: static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 size, size_t compressed_size, int compress_type, - struct page **compressed_pages, + struct page *compressed_page, bool update_i_size) { struct btrfs_drop_extents_args drop_args = { 0 }; @@ -668,7 +671,7 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 size, ret = insert_inline_extent(trans, path, inode, drop_args.extent_inserted, size, compressed_size, compress_type, - compressed_pages, update_i_size); + compressed_page, update_i_size); if (ret && ret != -ENOSPC) { btrfs_abort_transaction(trans, ret); goto out; @@ -976,7 +979,7 @@ again: } else { ret = cow_file_range_inline(inode, actual_end, total_compressed, - compress_type, pages, + compress_type, pages[0], false); } if (ret <= 0) { @@ -10460,7 +10463,7 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, if (start == 0 && encoded->unencoded_len == encoded->len && encoded->unencoded_offset == 0) { ret = cow_file_range_inline(inode, encoded->len, orig_count, - compression, pages, true); + compression, pages[0], true); if (ret <= 0) { if (ret == 0) ret = orig_count; -- cgit v1.2.3-59-g8ed1b From ae0d22a7fcb63863742c58be11b9e75ce2cbc9f5 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 29 Jan 2024 20:16:09 +1030 Subject: btrfs: migrate insert_inline_extent() to folio interfaces Since insert_inline_extent() now only accepts a single page, it's much easier to convert it to use folio interfaces. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 9096f8a34be9..979c6f8608e7 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -512,7 +512,7 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, struct btrfs_inode *inode, bool extent_inserted, size_t size, size_t compressed_size, int compress_type, - struct page *compressed_page, + struct folio *compressed_folio, bool update_i_size) { struct btrfs_root *root = inode->root; @@ -537,12 +537,12 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, * The compressed size also needs to be no larger than a sector. * That's also why we only need one page as the parameter. */ - if (compressed_page) + if (compressed_folio) ASSERT(compressed_size <= sectorsize); else ASSERT(compressed_size == 0); - if (compressed_size && compressed_page) + if (compressed_size && compressed_folio) cur_size = compressed_size; if (!extent_inserted) { @@ -570,7 +570,7 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, ptr = btrfs_file_extent_inline_start(ei); if (compress_type != BTRFS_COMPRESS_NONE) { - kaddr = kmap_local_page(compressed_page); + kaddr = kmap_local_folio(compressed_folio, 0); write_extent_buffer(leaf, kaddr, ptr, compressed_size); kunmap_local(kaddr); @@ -623,7 +623,7 @@ fail: static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 size, size_t compressed_size, int compress_type, - struct page *compressed_page, + struct folio *compressed_folio, bool update_i_size) { struct btrfs_drop_extents_args drop_args = { 0 }; @@ -671,7 +671,7 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 size, ret = insert_inline_extent(trans, path, inode, drop_args.extent_inserted, size, compressed_size, compress_type, - compressed_page, update_i_size); + compressed_folio, update_i_size); if (ret && ret != -ENOSPC) { btrfs_abort_transaction(trans, ret); goto out; @@ -979,7 +979,8 @@ again: } else { ret = cow_file_range_inline(inode, actual_end, total_compressed, - compress_type, pages[0], + compress_type, + page_folio(pages[0]), false); } if (ret <= 0) { @@ -10463,7 +10464,8 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, if (start == 0 && encoded->unencoded_len == encoded->len && encoded->unencoded_offset == 0) { ret = cow_file_range_inline(inode, encoded->len, orig_count, - compression, pages[0], true); + compression, page_folio(pages[0]), + true); if (ret <= 0) { if (ret == 0) ret = orig_count; -- cgit v1.2.3-59-g8ed1b From 11e03f2f4b79eac2176d8ae5120bc9857e7fbb29 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 29 Jan 2024 20:16:10 +1030 Subject: btrfs: introduce btrfs_alloc_folio_array() The new helper will do the same thing as btrfs_alloc_page_array(), but with folios. One extra difference is, there is no extra helper for bulk allocation, thus it may not be as efficient as the page version. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 31 +++++++++++++++++++++++++++++++ fs/btrfs/extent_io.h | 2 ++ 2 files changed, 33 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index ecb18a8db373..d90330f26827 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -666,6 +666,37 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) bio_put(bio); } +/* + * Populate every free slot in a provided array with folios. + * + * @nr_folios: number of folios to allocate + * @folio_array: the array to fill with folios; any existing non-NULL entries in + * the array will be skipped + * @extra_gfp: the extra GFP flags for the allocation + * + * Return: 0 if all folios were able to be allocated; + * -ENOMEM otherwise, the partially allocated folios would be freed and + * the array slots zeroed + */ +int btrfs_alloc_folio_array(unsigned int nr_folios, struct folio **folio_array, + gfp_t extra_gfp) +{ + for (int i = 0; i < nr_folios; i++) { + if (folio_array[i]) + continue; + folio_array[i] = folio_alloc(GFP_NOFS | extra_gfp, 0); + if (!folio_array[i]) + goto error; + } + return 0; +error: + for (int i = 0; i < nr_folios; i++) { + if (folio_array[i]) + folio_put(folio_array[i]); + } + return -ENOMEM; +} + /* * Populate every free slot in a provided array with pages. * diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 818431b37124..c81a9b546c9f 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -360,6 +360,8 @@ void btrfs_clear_buffer_dirty(struct btrfs_trans_handle *trans, int btrfs_alloc_page_array(unsigned int nr_pages, struct page **page_array, gfp_t extra_gfp); +int btrfs_alloc_folio_array(unsigned int nr_folios, struct folio **folio_array, + gfp_t extra_gfp); #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS bool find_lock_delalloc_range(struct inode *inode, -- cgit v1.2.3-59-g8ed1b From 400b172b8cdcff87038697169fe9e4222120dd9a Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 29 Jan 2024 20:16:11 +1030 Subject: btrfs: compression: migrate compression/decompression paths to folios For both compression and decompression paths, we always require a "struct page **pages" and "unsigned long nr_pages", this involves quite some part of the btrfs compression paths: - All the compression entry points - compressed_bio structure This affects both compression and decompression. - async_extent structure Unfortunately with all those involved parts, there is no good way to split the conversion into smaller patches while still passing compiling. So do this in one big conversion in one go. Please note this is direct page->folio conversion, no change on the page sized folio requirement yet. Signed-off-by: Qu Wenruo Reviewed-by: David Sterba [ minor style fixups ] Signed-off-by: David Sterba --- fs/btrfs/compression.c | 90 ++++++++++++++++++++-------------------- fs/btrfs/compression.h | 39 ++++++++---------- fs/btrfs/inode.c | 110 ++++++++++++++++++++++++------------------------- fs/btrfs/lzo.c | 86 +++++++++++++++++++------------------- fs/btrfs/zlib.c | 106 +++++++++++++++++++++++------------------------ fs/btrfs/zstd.c | 75 ++++++++++++++++----------------- 6 files changed, 251 insertions(+), 255 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 03648ec1bf73..c981903c8cd7 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -90,20 +90,20 @@ bool btrfs_compress_is_valid_type(const char *str, size_t len) } static int compression_compress_pages(int type, struct list_head *ws, - struct address_space *mapping, u64 start, struct page **pages, - unsigned long *out_pages, unsigned long *total_in, - unsigned long *total_out) + struct address_space *mapping, u64 start, + struct folio **folios, unsigned long *out_folios, + unsigned long *total_in, unsigned long *total_out) { switch (type) { case BTRFS_COMPRESS_ZLIB: - return zlib_compress_pages(ws, mapping, start, pages, - out_pages, total_in, total_out); + return zlib_compress_folios(ws, mapping, start, folios, + out_folios, total_in, total_out); case BTRFS_COMPRESS_LZO: - return lzo_compress_pages(ws, mapping, start, pages, - out_pages, total_in, total_out); + return lzo_compress_folios(ws, mapping, start, folios, + out_folios, total_in, total_out); case BTRFS_COMPRESS_ZSTD: - return zstd_compress_pages(ws, mapping, start, pages, - out_pages, total_in, total_out); + return zstd_compress_folios(ws, mapping, start, folios, + out_folios, total_in, total_out); case BTRFS_COMPRESS_NONE: default: /* @@ -115,7 +115,7 @@ static int compression_compress_pages(int type, struct list_head *ws, * Not a big deal, just need to inform caller that we * haven't allocated any pages yet. */ - *out_pages = 0; + *out_folios = 0; return -E2BIG; } } @@ -158,11 +158,11 @@ static int compression_decompress(int type, struct list_head *ws, } } -static void btrfs_free_compressed_pages(struct compressed_bio *cb) +static void btrfs_free_compressed_folios(struct compressed_bio *cb) { - for (unsigned int i = 0; i < cb->nr_pages; i++) - btrfs_free_compr_folio(page_folio(cb->compressed_pages[i])); - kfree(cb->compressed_pages); + for (unsigned int i = 0; i < cb->nr_folios; i++) + btrfs_free_compr_folio(cb->compressed_folios[i]); + kfree(cb->compressed_folios); } static int btrfs_decompress_bio(struct compressed_bio *cb); @@ -269,7 +269,7 @@ static void end_bbio_comprssed_read(struct btrfs_bio *bbio) if (!status) status = errno_to_blk_status(btrfs_decompress_bio(cb)); - btrfs_free_compressed_pages(cb); + btrfs_free_compressed_folios(cb); btrfs_bio_end_io(cb->orig_bbio, status); bio_put(&bbio->bio); } @@ -323,7 +323,7 @@ static void btrfs_finish_compressed_write_work(struct work_struct *work) end_compressed_writeback(cb); /* Note, our inode could be gone now */ - btrfs_free_compressed_pages(cb); + btrfs_free_compressed_folios(cb); bio_put(&cb->bbio.bio); } @@ -342,17 +342,19 @@ static void end_bbio_comprssed_write(struct btrfs_bio *bbio) queue_work(fs_info->compressed_write_workers, &cb->write_end_work); } -static void btrfs_add_compressed_bio_pages(struct compressed_bio *cb) +static void btrfs_add_compressed_bio_folios(struct compressed_bio *cb) { struct bio *bio = &cb->bbio.bio; u32 offset = 0; while (offset < cb->compressed_len) { + int ret; u32 len = min_t(u32, cb->compressed_len - offset, PAGE_SIZE); /* Maximum compressed extent is smaller than bio size limit. */ - __bio_add_page(bio, cb->compressed_pages[offset >> PAGE_SHIFT], - len, 0); + ret = bio_add_folio(bio, cb->compressed_folios[offset >> PAGE_SHIFT], + len, 0); + ASSERT(ret); offset += len; } } @@ -367,8 +369,8 @@ static void btrfs_add_compressed_bio_pages(struct compressed_bio *cb) * the end io hooks. */ void btrfs_submit_compressed_write(struct btrfs_ordered_extent *ordered, - struct page **compressed_pages, - unsigned int nr_pages, + struct folio **compressed_folios, + unsigned int nr_folios, blk_opf_t write_flags, bool writeback) { @@ -384,14 +386,14 @@ void btrfs_submit_compressed_write(struct btrfs_ordered_extent *ordered, end_bbio_comprssed_write); cb->start = ordered->file_offset; cb->len = ordered->num_bytes; - cb->compressed_pages = compressed_pages; + cb->compressed_folios = compressed_folios; cb->compressed_len = ordered->disk_num_bytes; cb->writeback = writeback; INIT_WORK(&cb->write_end_work, btrfs_finish_compressed_write_work); - cb->nr_pages = nr_pages; + cb->nr_folios = nr_folios; cb->bbio.bio.bi_iter.bi_sector = ordered->disk_bytenr >> SECTOR_SHIFT; cb->bbio.ordered = ordered; - btrfs_add_compressed_bio_pages(cb); + btrfs_add_compressed_bio_folios(cb); btrfs_submit_bio(&cb->bbio, 0); } @@ -599,14 +601,14 @@ void btrfs_submit_compressed_read(struct btrfs_bio *bbio) free_extent_map(em); - cb->nr_pages = DIV_ROUND_UP(compressed_len, PAGE_SIZE); - cb->compressed_pages = kcalloc(cb->nr_pages, sizeof(struct page *), GFP_NOFS); - if (!cb->compressed_pages) { + cb->nr_folios = DIV_ROUND_UP(compressed_len, PAGE_SIZE); + cb->compressed_folios = kcalloc(cb->nr_folios, sizeof(struct page *), GFP_NOFS); + if (!cb->compressed_folios) { ret = BLK_STS_RESOURCE; goto out_free_bio; } - ret2 = btrfs_alloc_page_array(cb->nr_pages, cb->compressed_pages, 0); + ret2 = btrfs_alloc_folio_array(cb->nr_folios, cb->compressed_folios, 0); if (ret2) { ret = BLK_STS_RESOURCE; goto out_free_compressed_pages; @@ -618,7 +620,7 @@ void btrfs_submit_compressed_read(struct btrfs_bio *bbio) /* include any pages we added in add_ra-bio_pages */ cb->len = bbio->bio.bi_iter.bi_size; cb->bbio.bio.bi_iter.bi_sector = bbio->bio.bi_iter.bi_sector; - btrfs_add_compressed_bio_pages(cb); + btrfs_add_compressed_bio_folios(cb); if (memstall) psi_memstall_leave(&pflags); @@ -627,7 +629,7 @@ void btrfs_submit_compressed_read(struct btrfs_bio *bbio) return; out_free_compressed_pages: - kfree(cb->compressed_pages); + kfree(cb->compressed_folios); out_free_bio: bio_put(&cb->bbio.bio); out: @@ -975,17 +977,17 @@ static unsigned int btrfs_compress_set_level(int type, unsigned level) } /* Wrapper around find_get_page(), with extra error message. */ -int btrfs_compress_find_get_page(struct address_space *mapping, u64 start, - struct page **in_page_ret) +int btrfs_compress_filemap_get_folio(struct address_space *mapping, u64 start, + struct folio **in_folio_ret) { - struct page *in_page; + struct folio *in_folio; /* - * The compressed write path should have the page locked already, thus - * we only need to grab one reference of the page cache. + * The compressed write path should have the folio locked already, thus + * we only need to grab one reference. */ - in_page = find_get_page(mapping, start >> PAGE_SHIFT); - if (unlikely(!in_page)) { + in_folio = filemap_get_folio(mapping, start >> PAGE_SHIFT); + if (IS_ERR(in_folio)) { struct btrfs_inode *inode = BTRFS_I(mapping->host); btrfs_crit(inode->root->fs_info, @@ -993,7 +995,7 @@ int btrfs_compress_find_get_page(struct address_space *mapping, u64 start, inode->root->root_key.objectid, btrfs_ino(inode), start); return -ENOENT; } - *in_page_ret = in_page; + *in_folio_ret = in_folio; return 0; } @@ -1017,11 +1019,9 @@ int btrfs_compress_find_get_page(struct address_space *mapping, u64 start, * @total_out is an in/out parameter, must be set to the input length and will * be also used to return the total number of compressed bytes */ -int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping, - u64 start, struct page **pages, - unsigned long *out_pages, - unsigned long *total_in, - unsigned long *total_out) +int btrfs_compress_folios(unsigned int type_level, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, + unsigned long *total_in, unsigned long *total_out) { int type = btrfs_compress_type(type_level); int level = btrfs_compress_level(type_level); @@ -1030,8 +1030,8 @@ int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping, level = btrfs_compress_set_level(type, level); workspace = get_workspace(type, level); - ret = compression_compress_pages(type, workspace, mapping, start, pages, - out_pages, total_in, total_out); + ret = compression_compress_pages(type, workspace, mapping, start, folios, + out_folios, total_in, total_out); put_workspace(type, workspace); return ret; } diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h index 2673c25415e5..c20c1a1b09d5 100644 --- a/fs/btrfs/compression.h +++ b/fs/btrfs/compression.h @@ -41,11 +41,11 @@ static_assert((BTRFS_MAX_COMPRESSED % PAGE_SIZE) == 0); #define BTRFS_ZLIB_DEFAULT_LEVEL 3 struct compressed_bio { - /* Number of compressed pages in the array */ - unsigned int nr_pages; + /* Number of compressed folios in the array. */ + unsigned int nr_folios; - /* the pages with the compressed data on them */ - struct page **compressed_pages; + /* The folios with the compressed data on them. */ + struct folio **compressed_folios; /* starting offset in the inode for our pages */ u64 start; @@ -85,21 +85,18 @@ static inline unsigned int btrfs_compress_level(unsigned int type_level) int __init btrfs_init_compress(void); void __cold btrfs_exit_compress(void); -int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping, - u64 start, struct page **pages, - unsigned long *out_pages, - unsigned long *total_in, - unsigned long *total_out); +int btrfs_compress_folios(unsigned int type_level, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, + unsigned long *total_in, unsigned long *total_out); int btrfs_decompress(int type, const u8 *data_in, struct page *dest_page, unsigned long start_byte, size_t srclen, size_t destlen); int btrfs_decompress_buf2page(const char *buf, u32 buf_len, struct compressed_bio *cb, u32 decompressed); void btrfs_submit_compressed_write(struct btrfs_ordered_extent *ordered, - struct page **compressed_pages, - unsigned int nr_pages, - blk_opf_t write_flags, - bool writeback); + struct folio **compressed_folios, + unsigned int nr_folios, blk_opf_t write_flags, + bool writeback); void btrfs_submit_compressed_read(struct btrfs_bio *bbio); unsigned int btrfs_compress_str2level(unsigned int type, const char *str); @@ -149,11 +146,11 @@ bool btrfs_compress_is_valid_type(const char *str, size_t len); int btrfs_compress_heuristic(struct inode *inode, u64 start, u64 end); -int btrfs_compress_find_get_page(struct address_space *mapping, u64 start, - struct page **in_page_ret); +int btrfs_compress_filemap_get_folio(struct address_space *mapping, u64 start, + struct folio **in_folio_ret); -int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, - u64 start, struct page **pages, unsigned long *out_pages, +int zlib_compress_folios(struct list_head *ws, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, unsigned long *total_in, unsigned long *total_out); int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb); int zlib_decompress(struct list_head *ws, const u8 *data_in, @@ -163,8 +160,8 @@ struct list_head *zlib_alloc_workspace(unsigned int level); void zlib_free_workspace(struct list_head *ws); struct list_head *zlib_get_workspace(unsigned int level); -int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, - u64 start, struct page **pages, unsigned long *out_pages, +int lzo_compress_folios(struct list_head *ws, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, unsigned long *total_in, unsigned long *total_out); int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb); int lzo_decompress(struct list_head *ws, const u8 *data_in, @@ -173,8 +170,8 @@ int lzo_decompress(struct list_head *ws, const u8 *data_in, struct list_head *lzo_alloc_workspace(unsigned int level); void lzo_free_workspace(struct list_head *ws); -int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, - u64 start, struct page **pages, unsigned long *out_pages, +int zstd_compress_folios(struct list_head *ws, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, unsigned long *total_in, unsigned long *total_out); int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb); int zstd_decompress(struct list_head *ws, const u8 *data_in, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 979c6f8608e7..15a13e191ee7 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -708,8 +708,8 @@ struct async_extent { u64 start; u64 ram_size; u64 compressed_size; - struct page **pages; - unsigned long nr_pages; + struct folio **folios; + unsigned long nr_folios; int compress_type; struct list_head list; }; @@ -734,8 +734,8 @@ struct async_cow { static noinline int add_async_extent(struct async_chunk *cow, u64 start, u64 ram_size, u64 compressed_size, - struct page **pages, - unsigned long nr_pages, + struct folio **folios, + unsigned long nr_folios, int compress_type) { struct async_extent *async_extent; @@ -746,8 +746,8 @@ static noinline int add_async_extent(struct async_chunk *cow, async_extent->start = start; async_extent->ram_size = ram_size; async_extent->compressed_size = compressed_size; - async_extent->pages = pages; - async_extent->nr_pages = nr_pages; + async_extent->folios = folios; + async_extent->nr_folios = nr_folios; async_extent->compress_type = compress_type; list_add_tail(&async_extent->list, &cow->extents); return 0; @@ -851,8 +851,8 @@ static void compress_file_range(struct btrfs_work *work) u64 actual_end; u64 i_size; int ret = 0; - struct page **pages; - unsigned long nr_pages; + struct folio **folios; + unsigned long nr_folios; unsigned long total_compressed = 0; unsigned long total_in = 0; unsigned int poff; @@ -882,9 +882,9 @@ static void compress_file_range(struct btrfs_work *work) barrier(); actual_end = min_t(u64, i_size, end + 1); again: - pages = NULL; - nr_pages = (end >> PAGE_SHIFT) - (start >> PAGE_SHIFT) + 1; - nr_pages = min_t(unsigned long, nr_pages, BTRFS_MAX_COMPRESSED_PAGES); + folios = NULL; + nr_folios = (end >> PAGE_SHIFT) - (start >> PAGE_SHIFT) + 1; + nr_folios = min_t(unsigned long, nr_folios, BTRFS_MAX_COMPRESSED_PAGES); /* * we don't want to send crud past the end of i_size through @@ -933,8 +933,8 @@ again: if (!inode_need_compress(inode, start, end)) goto cleanup_and_bail_uncompressed; - pages = kcalloc(nr_pages, sizeof(struct page *), GFP_NOFS); - if (!pages) { + folios = kcalloc(nr_folios, sizeof(struct folio *), GFP_NOFS); + if (!folios) { /* * Memory allocation failure is not a fatal error, we can fall * back to uncompressed code. @@ -948,9 +948,9 @@ again: compress_type = inode->prop_compress; /* Compression level is applied here. */ - ret = btrfs_compress_pages(compress_type | (fs_info->compress_level << 4), - mapping, start, pages, &nr_pages, &total_in, - &total_compressed); + ret = btrfs_compress_folios(compress_type | (fs_info->compress_level << 4), + mapping, start, folios, &nr_folios, &total_in, + &total_compressed); if (ret) goto mark_incompressible; @@ -960,7 +960,7 @@ again: */ poff = offset_in_page(total_compressed); if (poff) - memzero_page(pages[nr_pages - 1], poff, PAGE_SIZE - poff); + folio_zero_range(folios[nr_folios - 1], poff, PAGE_SIZE - poff); /* * Try to create an inline extent. @@ -979,8 +979,7 @@ again: } else { ret = cow_file_range_inline(inode, actual_end, total_compressed, - compress_type, - page_folio(pages[0]), + compress_type, folios[0], false); } if (ret <= 0) { @@ -1030,8 +1029,8 @@ again: * The async work queues will take care of doing actual allocation on * disk for these compressed pages, and will submit the bios. */ - ret = add_async_extent(async_chunk, start, total_in, total_compressed, pages, - nr_pages, compress_type); + ret = add_async_extent(async_chunk, start, total_in, total_compressed, folios, + nr_folios, compress_type); BUG_ON(ret); if (start + total_in < end) { start += total_in; @@ -1048,12 +1047,12 @@ cleanup_and_bail_uncompressed: BTRFS_COMPRESS_NONE); BUG_ON(ret); free_pages: - if (pages) { - for (i = 0; i < nr_pages; i++) { - WARN_ON(pages[i]->mapping); - btrfs_free_compr_folio(page_folio(pages[i])); + if (folios) { + for (i = 0; i < nr_folios; i++) { + WARN_ON(folios[i]->mapping); + btrfs_free_compr_folio(folios[i]); } - kfree(pages); + kfree(folios); } } @@ -1061,16 +1060,16 @@ static void free_async_extent_pages(struct async_extent *async_extent) { int i; - if (!async_extent->pages) + if (!async_extent->folios) return; - for (i = 0; i < async_extent->nr_pages; i++) { - WARN_ON(async_extent->pages[i]->mapping); - btrfs_free_compr_folio(page_folio(async_extent->pages[i])); + for (i = 0; i < async_extent->nr_folios; i++) { + WARN_ON(async_extent->folios[i]->mapping); + btrfs_free_compr_folio(async_extent->folios[i]); } - kfree(async_extent->pages); - async_extent->nr_pages = 0; - async_extent->pages = NULL; + kfree(async_extent->folios); + async_extent->nr_folios = 0; + async_extent->folios = NULL; } static void submit_uncompressed_range(struct btrfs_inode *inode, @@ -1194,8 +1193,8 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, NULL, EXTENT_LOCKED | EXTENT_DELALLOC, PAGE_UNLOCK | PAGE_START_WRITEBACK); btrfs_submit_compressed_write(ordered, - async_extent->pages, /* compressed_pages */ - async_extent->nr_pages, + async_extent->folios, /* compressed_folios */ + async_extent->nr_folios, async_chunk->write_flags, true); *alloc_hint = ins.objectid + ins.offset; done: @@ -10309,8 +10308,8 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, size_t orig_count; u64 start, end; u64 num_bytes, ram_bytes, disk_num_bytes; - unsigned long nr_pages, i; - struct page **pages; + unsigned long nr_folios, i; + struct folio **folios; struct btrfs_key ins; bool extent_reserved = false; struct extent_map *em; @@ -10399,24 +10398,24 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, * isn't. */ disk_num_bytes = ALIGN(orig_count, fs_info->sectorsize); - nr_pages = DIV_ROUND_UP(disk_num_bytes, PAGE_SIZE); - pages = kvcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL_ACCOUNT); - if (!pages) + nr_folios = DIV_ROUND_UP(disk_num_bytes, PAGE_SIZE); + folios = kvcalloc(nr_folios, sizeof(struct page *), GFP_KERNEL_ACCOUNT); + if (!folios) return -ENOMEM; - for (i = 0; i < nr_pages; i++) { + for (i = 0; i < nr_folios; i++) { size_t bytes = min_t(size_t, PAGE_SIZE, iov_iter_count(from)); char *kaddr; - pages[i] = alloc_page(GFP_KERNEL_ACCOUNT); - if (!pages[i]) { + folios[i] = folio_alloc(GFP_KERNEL_ACCOUNT, 0); + if (!folios[i]) { ret = -ENOMEM; - goto out_pages; + goto out_folios; } - kaddr = kmap_local_page(pages[i]); + kaddr = kmap_local_folio(folios[i], 0); if (copy_from_iter(kaddr, bytes, from) != bytes) { kunmap_local(kaddr); ret = -EFAULT; - goto out_pages; + goto out_folios; } if (bytes < PAGE_SIZE) memset(kaddr + bytes, 0, PAGE_SIZE - bytes); @@ -10428,12 +10427,12 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, ret = btrfs_wait_ordered_range(&inode->vfs_inode, start, num_bytes); if (ret) - goto out_pages; + goto out_folios; ret = invalidate_inode_pages2_range(inode->vfs_inode.i_mapping, start >> PAGE_SHIFT, end >> PAGE_SHIFT); if (ret) - goto out_pages; + goto out_folios; lock_extent(io_tree, start, end, &cached_state); ordered = btrfs_lookup_ordered_range(inode, start, num_bytes); if (!ordered && @@ -10464,8 +10463,7 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, if (start == 0 && encoded->unencoded_len == encoded->len && encoded->unencoded_offset == 0) { ret = cow_file_range_inline(inode, encoded->len, orig_count, - compression, page_folio(pages[0]), - true); + compression, folios[0], true); if (ret <= 0) { if (ret == 0) ret = orig_count; @@ -10509,7 +10507,7 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, btrfs_delalloc_release_extents(inode, num_bytes); - btrfs_submit_compressed_write(ordered, pages, nr_pages, 0, false); + btrfs_submit_compressed_write(ordered, folios, nr_folios, 0, false); ret = orig_count; goto out; @@ -10531,12 +10529,12 @@ out_free_data_space: btrfs_free_reserved_data_space_noquota(fs_info, disk_num_bytes); out_unlock: unlock_extent(io_tree, start, end, &cached_state); -out_pages: - for (i = 0; i < nr_pages; i++) { - if (pages[i]) - __free_page(pages[i]); +out_folios: + for (i = 0; i < nr_folios; i++) { + if (folios[i]) + __folio_put(folios[i]); } - kvfree(pages); + kvfree(folios); out: if (ret >= 0) iocb->ki_pos += encoded->len; diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c index c0e8894c9727..1c396ac167aa 100644 --- a/fs/btrfs/lzo.c +++ b/fs/btrfs/lzo.c @@ -130,17 +130,17 @@ static inline size_t read_compress_length(const char *buf) */ static int copy_compressed_data_to_page(char *compressed_data, size_t compressed_size, - struct page **out_pages, - unsigned long max_nr_page, + struct folio **out_folios, + unsigned long max_nr_folio, u32 *cur_out, const u32 sectorsize) { u32 sector_bytes_left; u32 orig_out; - struct page *cur_page; + struct folio *cur_folio; char *kaddr; - if ((*cur_out / PAGE_SIZE) >= max_nr_page) + if ((*cur_out / PAGE_SIZE) >= max_nr_folio) return -E2BIG; /* @@ -149,16 +149,16 @@ static int copy_compressed_data_to_page(char *compressed_data, */ ASSERT((*cur_out / sectorsize) == (*cur_out + LZO_LEN - 1) / sectorsize); - cur_page = out_pages[*cur_out / PAGE_SIZE]; + cur_folio = out_folios[*cur_out / PAGE_SIZE]; /* Allocate a new page */ - if (!cur_page) { - cur_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (!cur_page) + if (!cur_folio) { + cur_folio = btrfs_alloc_compr_folio(); + if (!cur_folio) return -ENOMEM; - out_pages[*cur_out / PAGE_SIZE] = cur_page; + out_folios[*cur_out / PAGE_SIZE] = cur_folio; } - kaddr = kmap_local_page(cur_page); + kaddr = kmap_local_folio(cur_folio, 0); write_compress_length(kaddr + offset_in_page(*cur_out), compressed_size); *cur_out += LZO_LEN; @@ -172,18 +172,18 @@ static int copy_compressed_data_to_page(char *compressed_data, kunmap_local(kaddr); - if ((*cur_out / PAGE_SIZE) >= max_nr_page) + if ((*cur_out / PAGE_SIZE) >= max_nr_folio) return -E2BIG; - cur_page = out_pages[*cur_out / PAGE_SIZE]; + cur_folio = out_folios[*cur_out / PAGE_SIZE]; /* Allocate a new page */ - if (!cur_page) { - cur_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (!cur_page) + if (!cur_folio) { + cur_folio = btrfs_alloc_compr_folio(); + if (!cur_folio) return -ENOMEM; - out_pages[*cur_out / PAGE_SIZE] = cur_page; + out_folios[*cur_out / PAGE_SIZE] = cur_folio; } - kaddr = kmap_local_page(cur_page); + kaddr = kmap_local_folio(cur_folio, 0); memcpy(kaddr + offset_in_page(*cur_out), compressed_data + *cur_out - orig_out, copy_len); @@ -209,15 +209,15 @@ out: return 0; } -int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, - u64 start, struct page **pages, unsigned long *out_pages, - unsigned long *total_in, unsigned long *total_out) +int lzo_compress_folios(struct list_head *ws, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, + unsigned long *total_in, unsigned long *total_out) { struct workspace *workspace = list_entry(ws, struct workspace, list); const u32 sectorsize = inode_to_fs_info(mapping->host)->sectorsize; - struct page *page_in = NULL; + struct folio *folio_in = NULL; char *sizes_ptr; - const unsigned long max_nr_page = *out_pages; + const unsigned long max_nr_folio = *out_folios; int ret = 0; /* Points to the file offset of input data */ u64 cur_in = start; @@ -225,8 +225,8 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, u32 cur_out = 0; u32 len = *total_out; - ASSERT(max_nr_page > 0); - *out_pages = 0; + ASSERT(max_nr_folio > 0); + *out_folios = 0; *total_out = 0; *total_in = 0; @@ -243,8 +243,8 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, size_t out_len; /* Get the input page first */ - if (!page_in) { - ret = btrfs_compress_find_get_page(mapping, cur_in, &page_in); + if (!folio_in) { + ret = btrfs_compress_filemap_get_folio(mapping, cur_in, &folio_in); if (ret < 0) goto out; } @@ -252,7 +252,7 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, /* Compress at most one sector of data each time */ in_len = min_t(u32, start + len - cur_in, sectorsize - sector_off); ASSERT(in_len); - data_in = kmap_local_page(page_in); + data_in = kmap_local_folio(folio_in, 0); ret = lzo1x_1_compress(data_in + offset_in_page(cur_in), in_len, workspace->cbuf, &out_len, @@ -265,7 +265,7 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, } ret = copy_compressed_data_to_page(workspace->cbuf, out_len, - pages, max_nr_page, + folios, max_nr_folio, &cur_out, sectorsize); if (ret < 0) goto out; @@ -283,13 +283,13 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, /* Check if we have reached page boundary */ if (PAGE_ALIGNED(cur_in)) { - put_page(page_in); - page_in = NULL; + folio_put(folio_in); + folio_in = NULL; } } /* Store the size of all chunks of compressed data */ - sizes_ptr = kmap_local_page(pages[0]); + sizes_ptr = kmap_local_folio(folios[0], 0); write_compress_length(sizes_ptr, cur_out); kunmap_local(sizes_ptr); @@ -297,9 +297,9 @@ int lzo_compress_pages(struct list_head *ws, struct address_space *mapping, *total_out = cur_out; *total_in = cur_in - start; out: - if (page_in) - put_page(page_in); - *out_pages = DIV_ROUND_UP(cur_out, PAGE_SIZE); + if (folio_in) + folio_put(folio_in); + *out_folios = DIV_ROUND_UP(cur_out, PAGE_SIZE); return ret; } @@ -314,15 +314,15 @@ static void copy_compressed_segment(struct compressed_bio *cb, u32 orig_in = *cur_in; while (*cur_in < orig_in + len) { - struct page *cur_page; + struct folio *cur_folio; u32 copy_len = min_t(u32, PAGE_SIZE - offset_in_page(*cur_in), orig_in + len - *cur_in); ASSERT(copy_len); - cur_page = cb->compressed_pages[*cur_in / PAGE_SIZE]; + cur_folio = cb->compressed_folios[*cur_in / PAGE_SIZE]; - memcpy_from_page(dest + *cur_in - orig_in, cur_page, - offset_in_page(*cur_in), copy_len); + memcpy_from_folio(dest + *cur_in - orig_in, cur_folio, + offset_in_folio(cur_folio, *cur_in), copy_len); *cur_in += copy_len; } @@ -342,7 +342,7 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) /* Bytes decompressed so far */ u32 cur_out = 0; - kaddr = kmap_local_page(cb->compressed_pages[0]); + kaddr = kmap_local_folio(cb->compressed_folios[0], 0); len_in = read_compress_length(kaddr); kunmap_local(kaddr); cur_in += LZO_LEN; @@ -364,7 +364,7 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) /* Go through each lzo segment */ while (cur_in < len_in) { - struct page *cur_page; + struct folio *cur_folio; /* Length of the compressed segment */ u32 seg_len; u32 sector_bytes_left; @@ -376,9 +376,9 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) */ ASSERT(cur_in / sectorsize == (cur_in + LZO_LEN - 1) / sectorsize); - cur_page = cb->compressed_pages[cur_in / PAGE_SIZE]; - ASSERT(cur_page); - kaddr = kmap_local_page(cur_page); + cur_folio = cb->compressed_folios[cur_in / PAGE_SIZE]; + ASSERT(cur_folio); + kaddr = kmap_local_folio(cur_folio, 0); seg_len = read_compress_length(kaddr + offset_in_page(cur_in)); kunmap_local(kaddr); cur_in += LZO_LEN; diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index c260e42b0a33..d9e5c88a0f85 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -91,24 +91,24 @@ fail: return ERR_PTR(-ENOMEM); } -int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, - u64 start, struct page **pages, unsigned long *out_pages, - unsigned long *total_in, unsigned long *total_out) +int zlib_compress_folios(struct list_head *ws, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, + unsigned long *total_in, unsigned long *total_out) { struct workspace *workspace = list_entry(ws, struct workspace, list); int ret; char *data_in = NULL; - char *cpage_out; - int nr_pages = 0; - struct page *in_page = NULL; - struct page *out_page = NULL; + char *cfolio_out; + int nr_folios = 0; + struct folio *in_folio = NULL; + struct folio *out_folio = NULL; unsigned long bytes_left; - unsigned int in_buf_pages; + unsigned int in_buf_folios; unsigned long len = *total_out; - unsigned long nr_dest_pages = *out_pages; - const unsigned long max_out = nr_dest_pages * PAGE_SIZE; + unsigned long nr_dest_folios = *out_folios; + const unsigned long max_out = nr_dest_folios * PAGE_SIZE; - *out_pages = 0; + *out_folios = 0; *total_out = 0; *total_in = 0; @@ -121,18 +121,18 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, workspace->strm.total_in = 0; workspace->strm.total_out = 0; - out_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (out_page == NULL) { + out_folio = btrfs_alloc_compr_folio(); + if (out_folio == NULL) { ret = -ENOMEM; goto out; } - cpage_out = page_address(out_page); - pages[0] = out_page; - nr_pages = 1; + cfolio_out = folio_address(out_folio); + folios[0] = out_folio; + nr_folios = 1; workspace->strm.next_in = workspace->buf; workspace->strm.avail_in = 0; - workspace->strm.next_out = cpage_out; + workspace->strm.next_out = cfolio_out; workspace->strm.avail_out = PAGE_SIZE; while (workspace->strm.total_in < len) { @@ -142,22 +142,22 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, */ if (workspace->strm.avail_in == 0) { bytes_left = len - workspace->strm.total_in; - in_buf_pages = min(DIV_ROUND_UP(bytes_left, PAGE_SIZE), - workspace->buf_size / PAGE_SIZE); - if (in_buf_pages > 1) { + in_buf_folios = min(DIV_ROUND_UP(bytes_left, PAGE_SIZE), + workspace->buf_size / PAGE_SIZE); + if (in_buf_folios > 1) { int i; - for (i = 0; i < in_buf_pages; i++) { + for (i = 0; i < in_buf_folios; i++) { if (data_in) { kunmap_local(data_in); - put_page(in_page); + folio_put(in_folio); data_in = NULL; } - ret = btrfs_compress_find_get_page(mapping, - start, &in_page); + ret = btrfs_compress_filemap_get_folio(mapping, + start, &in_folio); if (ret < 0) goto out; - data_in = kmap_local_page(in_page); + data_in = kmap_local_folio(in_folio, 0); copy_page(workspace->buf + i * PAGE_SIZE, data_in); start += PAGE_SIZE; @@ -166,14 +166,14 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, } else { if (data_in) { kunmap_local(data_in); - put_page(in_page); + folio_put(in_folio); data_in = NULL; } - ret = btrfs_compress_find_get_page(mapping, - start, &in_page); + ret = btrfs_compress_filemap_get_folio(mapping, + start, &in_folio); if (ret < 0) goto out; - data_in = kmap_local_page(in_page); + data_in = kmap_local_folio(in_folio, 0); start += PAGE_SIZE; workspace->strm.next_in = data_in; } @@ -202,20 +202,20 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, * the stream end if required */ if (workspace->strm.avail_out == 0) { - if (nr_pages == nr_dest_pages) { + if (nr_folios == nr_dest_folios) { ret = -E2BIG; goto out; } - out_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (out_page == NULL) { + out_folio = btrfs_alloc_compr_folio(); + if (out_folio == NULL) { ret = -ENOMEM; goto out; } - cpage_out = page_address(out_page); - pages[nr_pages] = out_page; - nr_pages++; + cfolio_out = folio_address(out_folio); + folios[nr_folios] = out_folio; + nr_folios++; workspace->strm.avail_out = PAGE_SIZE; - workspace->strm.next_out = cpage_out; + workspace->strm.next_out = cfolio_out; } /* we're all done */ if (workspace->strm.total_in >= len) @@ -237,21 +237,21 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, ret = -EIO; goto out; } else if (workspace->strm.avail_out == 0) { - /* get another page for the stream end */ - if (nr_pages == nr_dest_pages) { + /* Get another folio for the stream end. */ + if (nr_folios == nr_dest_folios) { ret = -E2BIG; goto out; } - out_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (out_page == NULL) { + out_folio = btrfs_alloc_compr_folio(); + if (out_folio == NULL) { ret = -ENOMEM; goto out; } - cpage_out = page_address(out_page); - pages[nr_pages] = out_page; - nr_pages++; + cfolio_out = folio_address(out_folio); + folios[nr_folios] = out_folio; + nr_folios++; workspace->strm.avail_out = PAGE_SIZE; - workspace->strm.next_out = cpage_out; + workspace->strm.next_out = cfolio_out; } } zlib_deflateEnd(&workspace->strm); @@ -265,10 +265,10 @@ int zlib_compress_pages(struct list_head *ws, struct address_space *mapping, *total_out = workspace->strm.total_out; *total_in = workspace->strm.total_in; out: - *out_pages = nr_pages; + *out_folios = nr_folios; if (data_in) { kunmap_local(data_in); - put_page(in_page); + folio_put(in_folio); } return ret; @@ -281,13 +281,13 @@ int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb) int wbits = MAX_WBITS; char *data_in; size_t total_out = 0; - unsigned long page_in_index = 0; + unsigned long folio_in_index = 0; size_t srclen = cb->compressed_len; - unsigned long total_pages_in = DIV_ROUND_UP(srclen, PAGE_SIZE); + unsigned long total_folios_in = DIV_ROUND_UP(srclen, PAGE_SIZE); unsigned long buf_start; - struct page **pages_in = cb->compressed_pages; + struct folio **folios_in = cb->compressed_folios; - data_in = kmap_local_page(pages_in[page_in_index]); + data_in = kmap_local_folio(folios_in[folio_in_index], 0); workspace->strm.next_in = data_in; workspace->strm.avail_in = min_t(size_t, srclen, PAGE_SIZE); workspace->strm.total_in = 0; @@ -337,12 +337,12 @@ int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb) if (workspace->strm.avail_in == 0) { unsigned long tmp; kunmap_local(data_in); - page_in_index++; - if (page_in_index >= total_pages_in) { + folio_in_index++; + if (folio_in_index >= total_folios_in) { data_in = NULL; break; } - data_in = kmap_local_page(pages_in[page_in_index]); + data_in = kmap_local_folio(folios_in[folio_in_index], 0); workspace->strm.next_in = data_in; tmp = srclen - workspace->strm.total_in; workspace->strm.avail_in = min(tmp, PAGE_SIZE); diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c index 4ec5dd84b93e..2b232b82c3a8 100644 --- a/fs/btrfs/zstd.c +++ b/fs/btrfs/zstd.c @@ -374,25 +374,25 @@ fail: return ERR_PTR(-ENOMEM); } -int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, - u64 start, struct page **pages, unsigned long *out_pages, - unsigned long *total_in, unsigned long *total_out) +int zstd_compress_folios(struct list_head *ws, struct address_space *mapping, + u64 start, struct folio **folios, unsigned long *out_folios, + unsigned long *total_in, unsigned long *total_out) { struct workspace *workspace = list_entry(ws, struct workspace, list); zstd_cstream *stream; int ret = 0; - int nr_pages = 0; - struct page *in_page = NULL; /* The current page to read */ - struct page *out_page = NULL; /* The current page to write to */ + int nr_folios = 0; + struct folio *in_folio = NULL; /* The current folio to read. */ + struct folio *out_folio = NULL; /* The current folio to write to. */ unsigned long tot_in = 0; unsigned long tot_out = 0; unsigned long len = *total_out; - const unsigned long nr_dest_pages = *out_pages; - unsigned long max_out = nr_dest_pages * PAGE_SIZE; + const unsigned long nr_dest_folios = *out_folios; + unsigned long max_out = nr_dest_folios * PAGE_SIZE; zstd_parameters params = zstd_get_btrfs_parameters(workspace->req_level, len); - *out_pages = 0; + *out_folios = 0; *total_out = 0; *total_in = 0; @@ -406,21 +406,21 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, } /* map in the first page of input data */ - ret = btrfs_compress_find_get_page(mapping, start, &in_page); + ret = btrfs_compress_filemap_get_folio(mapping, start, &in_folio); if (ret < 0) goto out; - workspace->in_buf.src = kmap_local_page(in_page); + workspace->in_buf.src = kmap_local_folio(in_folio, 0); workspace->in_buf.pos = 0; workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE); /* Allocate and map in the output buffer */ - out_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (out_page == NULL) { + out_folio = btrfs_alloc_compr_folio(); + if (out_folio == NULL) { ret = -ENOMEM; goto out; } - pages[nr_pages++] = out_page; - workspace->out_buf.dst = page_address(out_page); + folios[nr_folios++] = out_folio; + workspace->out_buf.dst = folio_address(out_folio); workspace->out_buf.pos = 0; workspace->out_buf.size = min_t(size_t, max_out, PAGE_SIZE); @@ -455,17 +455,17 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, if (workspace->out_buf.pos == workspace->out_buf.size) { tot_out += PAGE_SIZE; max_out -= PAGE_SIZE; - if (nr_pages == nr_dest_pages) { + if (nr_folios == nr_dest_folios) { ret = -E2BIG; goto out; } - out_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (out_page == NULL) { + out_folio = btrfs_alloc_compr_folio(); + if (out_folio == NULL) { ret = -ENOMEM; goto out; } - pages[nr_pages++] = out_page; - workspace->out_buf.dst = page_address(out_page); + folios[nr_folios++] = out_folio; + workspace->out_buf.dst = folio_address(out_folio); workspace->out_buf.pos = 0; workspace->out_buf.size = min_t(size_t, max_out, PAGE_SIZE); @@ -482,13 +482,13 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, tot_in += PAGE_SIZE; kunmap_local(workspace->in_buf.src); workspace->in_buf.src = NULL; - put_page(in_page); + folio_put(in_folio); start += PAGE_SIZE; len -= PAGE_SIZE; - ret = btrfs_compress_find_get_page(mapping, start, &in_page); + ret = btrfs_compress_filemap_get_folio(mapping, start, &in_folio); if (ret < 0) goto out; - workspace->in_buf.src = kmap_local_page(in_page); + workspace->in_buf.src = kmap_local_folio(in_folio, 0); workspace->in_buf.pos = 0; workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE); } @@ -515,17 +515,17 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, tot_out += PAGE_SIZE; max_out -= PAGE_SIZE; - if (nr_pages == nr_dest_pages) { + if (nr_folios == nr_dest_folios) { ret = -E2BIG; goto out; } - out_page = folio_page(btrfs_alloc_compr_folio(), 0); - if (out_page == NULL) { + out_folio = btrfs_alloc_compr_folio(); + if (out_folio == NULL) { ret = -ENOMEM; goto out; } - pages[nr_pages++] = out_page; - workspace->out_buf.dst = page_address(out_page); + folios[nr_folios++] = out_folio; + workspace->out_buf.dst = folio_address(out_folio); workspace->out_buf.pos = 0; workspace->out_buf.size = min_t(size_t, max_out, PAGE_SIZE); } @@ -539,10 +539,10 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping, *total_in = tot_in; *total_out = tot_out; out: - *out_pages = nr_pages; + *out_folios = nr_folios; if (workspace->in_buf.src) { kunmap_local(workspace->in_buf.src); - put_page(in_page); + folio_put(in_folio); } return ret; } @@ -550,12 +550,12 @@ out: int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb) { struct workspace *workspace = list_entry(ws, struct workspace, list); - struct page **pages_in = cb->compressed_pages; + struct folio **folios_in = cb->compressed_folios; size_t srclen = cb->compressed_len; zstd_dstream *stream; int ret = 0; - unsigned long page_in_index = 0; - unsigned long total_pages_in = DIV_ROUND_UP(srclen, PAGE_SIZE); + unsigned long folio_in_index = 0; + unsigned long total_folios_in = DIV_ROUND_UP(srclen, PAGE_SIZE); unsigned long buf_start; unsigned long total_out = 0; @@ -567,7 +567,7 @@ int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb) goto done; } - workspace->in_buf.src = kmap_local_page(pages_in[page_in_index]); + workspace->in_buf.src = kmap_local_folio(folios_in[folio_in_index], 0); workspace->in_buf.pos = 0; workspace->in_buf.size = min_t(size_t, srclen, PAGE_SIZE); @@ -604,14 +604,15 @@ int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb) if (workspace->in_buf.pos == workspace->in_buf.size) { kunmap_local(workspace->in_buf.src); - page_in_index++; - if (page_in_index >= total_pages_in) { + folio_in_index++; + if (folio_in_index >= total_folios_in) { workspace->in_buf.src = NULL; ret = -EIO; goto done; } srclen -= PAGE_SIZE; - workspace->in_buf.src = kmap_local_page(pages_in[page_in_index]); + workspace->in_buf.src = + kmap_local_folio(folios_in[folio_in_index], 0); workspace->in_buf.pos = 0; workspace->in_buf.size = min_t(size_t, srclen, PAGE_SIZE); } -- cgit v1.2.3-59-g8ed1b From 5f2fb819f6bfda870c4bef81a0ae7830e34044db Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Tue, 26 Mar 2024 11:17:12 -0700 Subject: btrfs: free PERTRANS at the end of cleanup_transaction() Some of the operations after the free might convert more PERTRANS metadata. Do the freeing as late as possible to eliminate a source of leaked PERTRANS metadata. This helps with the pass rate of generic/269 and generic/475. Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 3df5477d48a8..5a35c2c0bbc9 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -4844,14 +4844,10 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans, cur_trans->state = TRANS_STATE_UNBLOCKED; wake_up(&fs_info->transaction_wait); - btrfs_destroy_delayed_inodes(fs_info); - btrfs_destroy_marked_extents(fs_info, &cur_trans->dirty_pages, EXTENT_DIRTY); btrfs_destroy_pinned_extent(fs_info, &cur_trans->pinned_extents); - btrfs_free_all_qgroup_pertrans(fs_info); - cur_trans->state =TRANS_STATE_COMPLETED; wake_up(&cur_trans->commit_wait); } @@ -4904,6 +4900,7 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info) btrfs_assert_delayed_root_empty(fs_info); btrfs_destroy_all_delalloc_inodes(fs_info); btrfs_drop_all_logs(fs_info); + btrfs_free_all_qgroup_pertrans(fs_info); mutex_unlock(&fs_info->transaction_kthread_mutex); return 0; -- cgit v1.2.3-59-g8ed1b From 2e438442ba4d864c474ff38158054146010bd29b Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 2 Apr 2024 14:24:00 +0100 Subject: btrfs: remove not needed mod_start and mod_len from struct extent_map The mod_start and mod_len fields of struct extent_map were introduced by commit 4e2f84e63dc1 ("Btrfs: improve fsync by filtering extents that we want") in order to avoid too low performance when fsyncing a file that keeps getting extent maps merge, because it resulted in each fsync logging again csum ranges that were already merged before. We don't need this anymore as extent maps in the list of modified extents are never merged with other extent maps and once we log an extent map we remove it from the list of modified extent maps, so it's never logged twice. So remove the mod_start and mod_len fields from struct extent_map and use instead the start and len fields when logging checksums in the fast fsync path. This also makes EXTENT_FLAG_FILLING unused so remove it as well. Running the reproducer from the commit mentioned before, with a larger number of extents and against a null block device, so that IO is fast and we can better see any impact from searching checksums items and logging them, gave the following results from dd: Before this change: 409600000 bytes (410 MB, 391 MiB) copied, 22.948 s, 17.8 MB/s After this change: 409600000 bytes (410 MB, 391 MiB) copied, 22.9997 s, 17.8 MB/s So no changes in throughput. The test was done in a release kernel (non-debug, Debian's default kernel config) and its steps are the following: $ mkfs.btrfs -f /dev/nullb0 $ mount /dev/sdb /mnt $ dd if=/dev/zero of=/mnt/foobar bs=4k count=100000 oflag=sync $ umount /mnt This also reduces the size of struct extent_map from 128 bytes down to 112 bytes, so now we can have 36 extents maps per 4K page instead of 32. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 18 ------------------ fs/btrfs/extent_map.h | 4 ---- fs/btrfs/inode.c | 4 +--- fs/btrfs/tree-log.c | 4 ++-- include/trace/events/btrfs.h | 3 +-- 5 files changed, 4 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 24a048210b15..955ce300e5a1 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -252,8 +252,6 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em) em->len += merge->len; em->block_len += merge->block_len; em->block_start = merge->block_start; - em->mod_len = (em->mod_len + em->mod_start) - merge->mod_start; - em->mod_start = merge->mod_start; em->generation = max(em->generation, merge->generation); em->flags |= EXTENT_FLAG_MERGED; @@ -271,7 +269,6 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em) em->block_len += merge->block_len; rb_erase_cached(&merge->rb_node, &tree->map); RB_CLEAR_NODE(&merge->rb_node); - em->mod_len = (merge->mod_start + merge->mod_len) - em->mod_start; em->generation = max(em->generation, merge->generation); em->flags |= EXTENT_FLAG_MERGED; free_extent_map(merge); @@ -300,7 +297,6 @@ int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen) struct extent_map_tree *tree = &inode->extent_tree; int ret = 0; struct extent_map *em; - bool prealloc = false; write_lock(&tree->lock); em = lookup_extent_mapping(tree, start, len); @@ -325,21 +321,9 @@ int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen) em->generation = gen; em->flags &= ~EXTENT_FLAG_PINNED; - em->mod_start = em->start; - em->mod_len = em->len; - - if (em->flags & EXTENT_FLAG_FILLING) { - prealloc = true; - em->flags &= ~EXTENT_FLAG_FILLING; - } try_merge_map(tree, em); - if (prealloc) { - em->mod_start = em->start; - em->mod_len = em->len; - } - out: write_unlock(&tree->lock); free_extent_map(em); @@ -361,8 +345,6 @@ static inline void setup_extent_mapping(struct extent_map_tree *tree, int modified) { refcount_inc(&em->refs); - em->mod_start = em->start; - em->mod_len = em->len; ASSERT(list_empty(&em->list)); diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index c5a098c99cc6..10e9491865c9 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -30,8 +30,6 @@ enum { ENUM_BIT(EXTENT_FLAG_PREALLOC), /* Logging this extent */ ENUM_BIT(EXTENT_FLAG_LOGGING), - /* Filling in a preallocated extent */ - ENUM_BIT(EXTENT_FLAG_FILLING), /* This em is merged from two or more physically adjacent ems */ ENUM_BIT(EXTENT_FLAG_MERGED), }; @@ -46,8 +44,6 @@ struct extent_map { /* all of these are in bytes */ u64 start; u64 len; - u64 mod_start; - u64 mod_len; u64 orig_start; u64 orig_block_len; u64 ram_bytes; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 15a13e191ee7..12b1a83a4303 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7337,9 +7337,7 @@ static struct extent_map *create_io_em(struct btrfs_inode *inode, u64 start, em->ram_bytes = ram_bytes; em->generation = -1; em->flags |= EXTENT_FLAG_PINNED; - if (type == BTRFS_ORDERED_PREALLOC) - em->flags |= EXTENT_FLAG_FILLING; - else if (type == BTRFS_ORDERED_COMPRESSED) + if (type == BTRFS_ORDERED_COMPRESSED) extent_map_set_compression(em, compress_type); ret = btrfs_replace_extent_map_range(inode, em, true); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 472918a5bc73..d9777649e170 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -4574,8 +4574,8 @@ static int log_extent_csums(struct btrfs_trans_handle *trans, struct btrfs_root *csum_root; u64 csum_offset; u64 csum_len; - u64 mod_start = em->mod_start; - u64 mod_len = em->mod_len; + u64 mod_start = em->start; + u64 mod_len = em->len; LIST_HEAD(ordered_sums); int ret = 0; diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index 90b0222390e5..766cfd48386c 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -277,8 +277,7 @@ DEFINE_EVENT(btrfs__inode, btrfs_inode_evict, { EXTENT_FLAG_COMPRESS_LZO, "COMPRESS_LZO" },\ { EXTENT_FLAG_COMPRESS_ZSTD, "COMPRESS_ZSTD" },\ { EXTENT_FLAG_PREALLOC, "PREALLOC" },\ - { EXTENT_FLAG_LOGGING, "LOGGING" },\ - { EXTENT_FLAG_FILLING, "FILLING" }) + { EXTENT_FLAG_LOGGING, "LOGGING" }) TRACE_EVENT_CONDITION(btrfs_get_extent, -- cgit v1.2.3-59-g8ed1b From 47f694487738af319d0e0fae26c61d77fd6d0c5e Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 16 Jan 2024 15:05:00 +0000 Subject: btrfs: remove pointless return value assignment at btrfs_finish_one_ordered() At btrfs_finish_one_ordered() it's pointless to assign 0 to the 'ret' variable because if it has a non-zero value (error), we have already jumped to the 'out' label. So remove that redundant assignment. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 12b1a83a4303..1e19cb34b59a 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3184,7 +3184,6 @@ int btrfs_finish_one_ordered(struct btrfs_ordered_extent *ordered_extent) btrfs_abort_transaction(trans, ret); goto out; } - ret = 0; out: clear_extent_bit(&inode->io_tree, start, end, clear_bits, &cached_state); -- cgit v1.2.3-59-g8ed1b From 606a1c5de13e8cb017dfcf485c1a5c3565b1d2a5 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 3 Apr 2024 12:38:21 +0100 Subject: btrfs: remove list_empty() check at warn_about_uncommitted_trans() At warn_about_uncommitted_trans(), there's no need to check if the list is empty and return, because list_for_each_entry_safe() is safe to call for an empty list, it simply does nothing. So remove the check. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 5a35c2c0bbc9..0474e9b6d302 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -4182,9 +4182,6 @@ static void warn_about_uncommitted_trans(struct btrfs_fs_info *fs_info) struct btrfs_transaction *tmp; bool found = false; - if (list_empty(&fs_info->trans_list)) - return; - /* * This function is only called at the very end of close_ctree(), * thus no other running transaction, no need to take trans_lock. -- cgit v1.2.3-59-g8ed1b From 590e2c4a1e82cd9b878bdaa70ab54026d0165bbc Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 3 Apr 2024 12:47:51 +0100 Subject: btrfs: remove no longer used btrfs_clone_chunk_map() There are no more users of btrfs_clone_chunk_map(), the last one (and only one ever) was removed in commit 1ec17ef59168 ("btrfs: zoned: fix use-after-free in do_zone_finish()"). So remove btrfs_clone_chunk_map(). Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 15 --------------- fs/btrfs/volumes.h | 1 - 2 files changed, 16 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index ef6bd2f4251b..b6a701011fb0 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -5615,21 +5615,6 @@ struct btrfs_chunk_map *btrfs_alloc_chunk_map(int num_stripes, gfp_t gfp) return map; } -struct btrfs_chunk_map *btrfs_clone_chunk_map(struct btrfs_chunk_map *map, gfp_t gfp) -{ - const int size = btrfs_chunk_map_size(map->num_stripes); - struct btrfs_chunk_map *clone; - - clone = kmemdup(map, size, gfp); - if (!clone) - return NULL; - - refcount_set(&clone->refs, 1); - RB_CLEAR_NODE(&clone->rb_node); - - return clone; -} - static struct btrfs_block_group *create_chunk(struct btrfs_trans_handle *trans, struct alloc_chunk_ctl *ctl, struct btrfs_device_info *devices_info) diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 93854609a4d5..cf555f5b47ce 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -743,7 +743,6 @@ struct btrfs_chunk_map *btrfs_alloc_chunk_map(int num_stripes, gfp_t gfp); int btrfs_add_chunk_map(struct btrfs_fs_info *fs_info, struct btrfs_chunk_map *map); #endif -struct btrfs_chunk_map *btrfs_clone_chunk_map(struct btrfs_chunk_map *map, gfp_t gfp); struct btrfs_chunk_map *btrfs_find_chunk_map(struct btrfs_fs_info *fs_info, u64 logical, u64 length); struct btrfs_chunk_map *btrfs_find_chunk_map_nolock(struct btrfs_fs_info *fs_info, -- cgit v1.2.3-59-g8ed1b From 0ddefc2a7ce4be261a4ac3dffdca4a2ee9b716af Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 10 Apr 2024 12:36:51 +0100 Subject: btrfs: move btrfs_page_mkwrite() from inode.c into file.c btrfs_page_mkwrite() is a struct vm_operations_struct callback and we define that structure in file.c. Currently the function is in inode.c and has to be exported to be used in file.c, which makes no sense because it's not used anywhere else. So move btrfs_page_mkwrite() from inode.c and into file.c. While at it do a few minor style changes: 1) Capitalize the first word of every comment and end each sentence with punctuation; 2) Avoid splitting some statements into two lines when everything fits in 85 characters or less. Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/btrfs_inode.h | 1 - fs/btrfs/file.c | 166 ++++++++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/inode.c | 167 ------------------------------------------------- 3 files changed, 166 insertions(+), 168 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 100020ca4658..ed8bd15aa3e2 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -496,7 +496,6 @@ void btrfs_merge_delalloc_extent(struct btrfs_inode *inode, struct extent_state void btrfs_split_delalloc_extent(struct btrfs_inode *inode, struct extent_state *orig, u64 split); void btrfs_set_range_writeback(struct btrfs_inode *inode, u64 start, u64 end); -vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf); void btrfs_evict_inode(struct inode *inode); struct inode *btrfs_alloc_inode(struct super_block *sb); void btrfs_destroy_inode(struct inode *inode); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 0c23053951be..1b972bd05af1 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -2029,6 +2029,172 @@ out_release_extents: goto out; } +/* + * btrfs_page_mkwrite() is not allowed to change the file size as it gets + * called from a page fault handler when a page is first dirtied. Hence we must + * be careful to check for EOF conditions here. We set the page up correctly + * for a written page which means we get ENOSPC checking when writing into + * holes and correct delalloc and unwritten extent mapping on filesystems that + * support these features. + * + * We are not allowed to take the i_mutex here so we have to play games to + * protect against truncate races as the page could now be beyond EOF. Because + * truncate_setsize() writes the inode size before removing pages, once we have + * the page lock we can determine safely if the page is beyond EOF. If it is not + * beyond EOF, then the page is guaranteed safe against truncation until we + * unlock the page. + */ +static vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf) +{ + struct page *page = vmf->page; + struct folio *folio = page_folio(page); + struct inode *inode = file_inode(vmf->vma->vm_file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); + struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; + struct btrfs_ordered_extent *ordered; + struct extent_state *cached_state = NULL; + struct extent_changeset *data_reserved = NULL; + unsigned long zero_start; + loff_t size; + vm_fault_t ret; + int ret2; + int reserved = 0; + u64 reserved_space; + u64 page_start; + u64 page_end; + u64 end; + + ASSERT(folio_order(folio) == 0); + + reserved_space = PAGE_SIZE; + + sb_start_pagefault(inode->i_sb); + page_start = page_offset(page); + page_end = page_start + PAGE_SIZE - 1; + end = page_end; + + /* + * Reserving delalloc space after obtaining the page lock can lead to + * deadlock. For example, if a dirty page is locked by this function + * and the call to btrfs_delalloc_reserve_space() ends up triggering + * dirty page write out, then the btrfs_writepages() function could + * end up waiting indefinitely to get a lock on the page currently + * being processed by btrfs_page_mkwrite() function. + */ + ret2 = btrfs_delalloc_reserve_space(BTRFS_I(inode), &data_reserved, + page_start, reserved_space); + if (!ret2) { + ret2 = file_update_time(vmf->vma->vm_file); + reserved = 1; + } + if (ret2) { + ret = vmf_error(ret2); + if (reserved) + goto out; + goto out_noreserve; + } + + /* Make the VM retry the fault. */ + ret = VM_FAULT_NOPAGE; +again: + down_read(&BTRFS_I(inode)->i_mmap_lock); + lock_page(page); + size = i_size_read(inode); + + if ((page->mapping != inode->i_mapping) || + (page_start >= size)) { + /* Page got truncated out from underneath us. */ + goto out_unlock; + } + wait_on_page_writeback(page); + + lock_extent(io_tree, page_start, page_end, &cached_state); + ret2 = set_page_extent_mapped(page); + if (ret2 < 0) { + ret = vmf_error(ret2); + unlock_extent(io_tree, page_start, page_end, &cached_state); + goto out_unlock; + } + + /* + * We can't set the delalloc bits if there are pending ordered + * extents. Drop our locks and wait for them to finish. + */ + ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start, PAGE_SIZE); + if (ordered) { + unlock_extent(io_tree, page_start, page_end, &cached_state); + unlock_page(page); + up_read(&BTRFS_I(inode)->i_mmap_lock); + btrfs_start_ordered_extent(ordered); + btrfs_put_ordered_extent(ordered); + goto again; + } + + if (page->index == ((size - 1) >> PAGE_SHIFT)) { + reserved_space = round_up(size - page_start, fs_info->sectorsize); + if (reserved_space < PAGE_SIZE) { + end = page_start + reserved_space - 1; + btrfs_delalloc_release_space(BTRFS_I(inode), + data_reserved, page_start, + PAGE_SIZE - reserved_space, true); + } + } + + /* + * page_mkwrite gets called when the page is firstly dirtied after it's + * faulted in, but write(2) could also dirty a page and set delalloc + * bits, thus in this case for space account reason, we still need to + * clear any delalloc bits within this page range since we have to + * reserve data&meta space before lock_page() (see above comments). + */ + clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, end, + EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | + EXTENT_DEFRAG, &cached_state); + + ret2 = btrfs_set_extent_delalloc(BTRFS_I(inode), page_start, end, 0, + &cached_state); + if (ret2) { + unlock_extent(io_tree, page_start, page_end, &cached_state); + ret = VM_FAULT_SIGBUS; + goto out_unlock; + } + + /* Page is wholly or partially inside EOF. */ + if (page_start + PAGE_SIZE > size) + zero_start = offset_in_page(size); + else + zero_start = PAGE_SIZE; + + if (zero_start != PAGE_SIZE) + memzero_page(page, zero_start, PAGE_SIZE - zero_start); + + btrfs_folio_clear_checked(fs_info, folio, page_start, PAGE_SIZE); + btrfs_folio_set_dirty(fs_info, folio, page_start, end + 1 - page_start); + btrfs_folio_set_uptodate(fs_info, folio, page_start, end + 1 - page_start); + + btrfs_set_inode_last_sub_trans(BTRFS_I(inode)); + + unlock_extent(io_tree, page_start, page_end, &cached_state); + up_read(&BTRFS_I(inode)->i_mmap_lock); + + btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); + sb_end_pagefault(inode->i_sb); + extent_changeset_free(data_reserved); + return VM_FAULT_LOCKED; + +out_unlock: + unlock_page(page); + up_read(&BTRFS_I(inode)->i_mmap_lock); +out: + btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); + btrfs_delalloc_release_space(BTRFS_I(inode), data_reserved, page_start, + reserved_space, (ret != 0)); +out_noreserve: + sb_end_pagefault(inode->i_sb); + extent_changeset_free(data_reserved); + return ret; +} + static const struct vm_operations_struct btrfs_file_vm_ops = { .fault = filemap_fault, .map_pages = filemap_map_pages, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1e19cb34b59a..7c009bbc1a5c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8163,173 +8163,6 @@ next: clear_page_extent_mapped(&folio->page); } -/* - * btrfs_page_mkwrite() is not allowed to change the file size as it gets - * called from a page fault handler when a page is first dirtied. Hence we must - * be careful to check for EOF conditions here. We set the page up correctly - * for a written page which means we get ENOSPC checking when writing into - * holes and correct delalloc and unwritten extent mapping on filesystems that - * support these features. - * - * We are not allowed to take the i_mutex here so we have to play games to - * protect against truncate races as the page could now be beyond EOF. Because - * truncate_setsize() writes the inode size before removing pages, once we have - * the page lock we can determine safely if the page is beyond EOF. If it is not - * beyond EOF, then the page is guaranteed safe against truncation until we - * unlock the page. - */ -vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf) -{ - struct page *page = vmf->page; - struct folio *folio = page_folio(page); - struct inode *inode = file_inode(vmf->vma->vm_file); - struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); - struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; - struct btrfs_ordered_extent *ordered; - struct extent_state *cached_state = NULL; - struct extent_changeset *data_reserved = NULL; - unsigned long zero_start; - loff_t size; - vm_fault_t ret; - int ret2; - int reserved = 0; - u64 reserved_space; - u64 page_start; - u64 page_end; - u64 end; - - ASSERT(folio_order(folio) == 0); - - reserved_space = PAGE_SIZE; - - sb_start_pagefault(inode->i_sb); - page_start = page_offset(page); - page_end = page_start + PAGE_SIZE - 1; - end = page_end; - - /* - * Reserving delalloc space after obtaining the page lock can lead to - * deadlock. For example, if a dirty page is locked by this function - * and the call to btrfs_delalloc_reserve_space() ends up triggering - * dirty page write out, then the btrfs_writepages() function could - * end up waiting indefinitely to get a lock on the page currently - * being processed by btrfs_page_mkwrite() function. - */ - ret2 = btrfs_delalloc_reserve_space(BTRFS_I(inode), &data_reserved, - page_start, reserved_space); - if (!ret2) { - ret2 = file_update_time(vmf->vma->vm_file); - reserved = 1; - } - if (ret2) { - ret = vmf_error(ret2); - if (reserved) - goto out; - goto out_noreserve; - } - - ret = VM_FAULT_NOPAGE; /* make the VM retry the fault */ -again: - down_read(&BTRFS_I(inode)->i_mmap_lock); - lock_page(page); - size = i_size_read(inode); - - if ((page->mapping != inode->i_mapping) || - (page_start >= size)) { - /* page got truncated out from underneath us */ - goto out_unlock; - } - wait_on_page_writeback(page); - - lock_extent(io_tree, page_start, page_end, &cached_state); - ret2 = set_page_extent_mapped(page); - if (ret2 < 0) { - ret = vmf_error(ret2); - unlock_extent(io_tree, page_start, page_end, &cached_state); - goto out_unlock; - } - - /* - * we can't set the delalloc bits if there are pending ordered - * extents. Drop our locks and wait for them to finish - */ - ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start, - PAGE_SIZE); - if (ordered) { - unlock_extent(io_tree, page_start, page_end, &cached_state); - unlock_page(page); - up_read(&BTRFS_I(inode)->i_mmap_lock); - btrfs_start_ordered_extent(ordered); - btrfs_put_ordered_extent(ordered); - goto again; - } - - if (page->index == ((size - 1) >> PAGE_SHIFT)) { - reserved_space = round_up(size - page_start, - fs_info->sectorsize); - if (reserved_space < PAGE_SIZE) { - end = page_start + reserved_space - 1; - btrfs_delalloc_release_space(BTRFS_I(inode), - data_reserved, page_start, - PAGE_SIZE - reserved_space, true); - } - } - - /* - * page_mkwrite gets called when the page is firstly dirtied after it's - * faulted in, but write(2) could also dirty a page and set delalloc - * bits, thus in this case for space account reason, we still need to - * clear any delalloc bits within this page range since we have to - * reserve data&meta space before lock_page() (see above comments). - */ - clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, end, - EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | - EXTENT_DEFRAG, &cached_state); - - ret2 = btrfs_set_extent_delalloc(BTRFS_I(inode), page_start, end, 0, - &cached_state); - if (ret2) { - unlock_extent(io_tree, page_start, page_end, &cached_state); - ret = VM_FAULT_SIGBUS; - goto out_unlock; - } - - /* page is wholly or partially inside EOF */ - if (page_start + PAGE_SIZE > size) - zero_start = offset_in_page(size); - else - zero_start = PAGE_SIZE; - - if (zero_start != PAGE_SIZE) - memzero_page(page, zero_start, PAGE_SIZE - zero_start); - - btrfs_folio_clear_checked(fs_info, folio, page_start, PAGE_SIZE); - btrfs_folio_set_dirty(fs_info, folio, page_start, end + 1 - page_start); - btrfs_folio_set_uptodate(fs_info, folio, page_start, end + 1 - page_start); - - btrfs_set_inode_last_sub_trans(BTRFS_I(inode)); - - unlock_extent(io_tree, page_start, page_end, &cached_state); - up_read(&BTRFS_I(inode)->i_mmap_lock); - - btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); - sb_end_pagefault(inode->i_sb); - extent_changeset_free(data_reserved); - return VM_FAULT_LOCKED; - -out_unlock: - unlock_page(page); - up_read(&BTRFS_I(inode)->i_mmap_lock); -out: - btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); - btrfs_delalloc_release_space(BTRFS_I(inode), data_reserved, page_start, - reserved_space, (ret != 0)); -out_noreserve: - sb_end_pagefault(inode->i_sb); - extent_changeset_free(data_reserved); - return ret; -} - static int btrfs_truncate(struct btrfs_inode *inode, bool skip_writeback) { struct btrfs_truncate_control control = { -- cgit v1.2.3-59-g8ed1b From d800a9065b1e44e1cceb05990e7ebe73284b5f6d Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 18:17:05 +0100 Subject: btrfs: add function comment to btrfs_lookup_csums_list() Add a function comment to btrfs_lookup_csums_list() to document it. With another upcoming change its parameter list and return value will be less obvious. So add the documentation now so that it can be updated where needed later. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file-item.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index e58fb5347e65..909438540119 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -450,6 +450,19 @@ blk_status_t btrfs_lookup_bio_sums(struct btrfs_bio *bbio) return ret; } +/* + * Search for checksums for a given logical range. + * + * @root: The root where to look for checksums. + * @start: Logical address of target checksum range. + * @end: End offset (inclusive) of the target checksum range. + * @list: List for adding each checksum that was found. + * @search_commit: Indicate if the commit root of the @root should be used + * for the search. + * @nowait: Indicate if the search must be non-blocking or not. + * + * Return < 0 on error and 0 on success. + */ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, struct list_head *list, int search_commit, bool nowait) -- cgit v1.2.3-59-g8ed1b From afcb80624f5bc2fd2f40dc8a7f4fc2198a68e40f Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 18:26:56 +0100 Subject: btrfs: remove search_commit parameter from btrfs_lookup_csums_list() All the callers of btrfs_lookup_csums_list() pass a value of 0 as the "search_commit" parameter. So remove it and make the function behave as to always search from the regular root. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file-item.c | 10 +--------- fs/btrfs/file-item.h | 3 +-- fs/btrfs/inode.c | 2 +- fs/btrfs/relocation.c | 2 +- fs/btrfs/tree-log.c | 6 +++--- 5 files changed, 7 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 909438540119..0712a0aa2dd0 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -457,15 +457,12 @@ blk_status_t btrfs_lookup_bio_sums(struct btrfs_bio *bbio) * @start: Logical address of target checksum range. * @end: End offset (inclusive) of the target checksum range. * @list: List for adding each checksum that was found. - * @search_commit: Indicate if the commit root of the @root should be used - * for the search. * @nowait: Indicate if the search must be non-blocking or not. * * Return < 0 on error and 0 on success. */ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, - struct list_head *list, int search_commit, - bool nowait) + struct list_head *list, bool nowait) { struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_key key; @@ -484,11 +481,6 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, return -ENOMEM; path->nowait = nowait; - if (search_commit) { - path->skip_locking = 1; - path->reada = READA_FORWARD; - path->search_commit_root = 1; - } key.objectid = BTRFS_EXTENT_CSUM_OBJECTID; key.offset = start; diff --git a/fs/btrfs/file-item.h b/fs/btrfs/file-item.h index 15c05cc0fce6..557dc43d7142 100644 --- a/fs/btrfs/file-item.h +++ b/fs/btrfs/file-item.h @@ -68,8 +68,7 @@ int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end, struct list_head *list, int search_commit, bool nowait); int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, - struct list_head *list, int search_commit, - bool nowait); + struct list_head *list, bool nowait); int btrfs_lookup_csums_bitmap(struct btrfs_root *root, struct btrfs_path *path, u64 start, u64 end, u8 *csum_buf, unsigned long *csum_bitmap); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7c009bbc1a5c..f4c54596c5c4 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1745,7 +1745,7 @@ static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, LIST_HEAD(list); ret = btrfs_lookup_csums_list(csum_root, bytenr, bytenr + num_bytes - 1, - &list, 0, nowait); + &list, nowait); if (ret == 0 && list_empty(&list)) return 0; diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 5c9ef6717f84..516c6e4610d7 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4446,7 +4446,7 @@ int btrfs_reloc_clone_csums(struct btrfs_ordered_extent *ordered) ret = btrfs_lookup_csums_list(csum_root, disk_bytenr, disk_bytenr + ordered->num_bytes - 1, - &list, 0, false); + &list, false); if (ret) return ret; diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index d9777649e170..079fd0288d9c 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -797,7 +797,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, ret = btrfs_lookup_csums_list(root->log_root, csum_start, csum_end - 1, - &ordered_sums, 0, false); + &ordered_sums, false); if (ret) goto out; /* @@ -4460,7 +4460,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, disk_bytenr += extent_offset; ret = btrfs_lookup_csums_list(csum_root, disk_bytenr, disk_bytenr + extent_num_bytes - 1, - &ordered_sums, 0, false); + &ordered_sums, false); if (ret) goto out; @@ -4655,7 +4655,7 @@ static int log_extent_csums(struct btrfs_trans_handle *trans, csum_root = btrfs_csum_root(trans->fs_info, em->block_start); ret = btrfs_lookup_csums_list(csum_root, em->block_start + csum_offset, em->block_start + csum_offset + - csum_len - 1, &ordered_sums, 0, false); + csum_len - 1, &ordered_sums, false); if (ret) return ret; -- cgit v1.2.3-59-g8ed1b From c0dce8b6a3b257b8492260f2ea41b7056b795295 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 18:33:43 +0100 Subject: btrfs: remove use of a temporary list at btrfs_lookup_csums_list() There's no need to use a temporary list to add the checksums, we can just add them to input list and then on error delete and free any checksums that were added. So simplify and remove the temporary list. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file-item.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 0712a0aa2dd0..c2799325706f 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -470,7 +470,6 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, struct extent_buffer *leaf; struct btrfs_ordered_sum *sums; struct btrfs_csum_item *item; - LIST_HEAD(tmplist); int ret; ASSERT(IS_ALIGNED(start, fs_info->sectorsize) && @@ -572,18 +571,17 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, bytes_to_csum_size(fs_info, size)); start += size; - list_add_tail(&sums->list, &tmplist); + list_add_tail(&sums->list, list); } path->slots[0]++; } ret = 0; fail: - while (ret < 0 && !list_empty(&tmplist)) { - sums = list_entry(tmplist.next, struct btrfs_ordered_sum, list); + while (ret < 0 && !list_empty(list)) { + sums = list_entry(list->next, struct btrfs_ordered_sum, list); list_del(&sums->list); kfree(sums); } - list_splice_tail(&tmplist, list); btrfs_free_path(path); return ret; -- cgit v1.2.3-59-g8ed1b From fb90e1caf00d0797d355c8f15d004e41edf22e96 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 18:39:51 +0100 Subject: btrfs: simplify error path for btrfs_lookup_csums_list() In the error path we have this while loop that keeps iterating over the csums of the list and then delete them from the list and free them, testing for an error (ret < 0) and list emptyness as the conditions of the while loop. Simplify this by using list_for_each_entry_safe() so there's no need to delete elements from the list and need to test the error condition on each iteration. Also rename the 'fail' label to 'out' since the label is not exclusive to a failure path, as we also end up there when the function succeeds, and it's also a more common label name. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file-item.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index c2799325706f..231abcc87f63 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -487,7 +487,7 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) - goto fail; + goto out; if (ret > 0 && path->slots[0] > 0) { leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &key, path->slots[0] - 1); @@ -522,7 +522,7 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, if (path->slots[0] >= btrfs_header_nritems(leaf)) { ret = btrfs_next_leaf(root, path); if (ret < 0) - goto fail; + goto out; if (ret > 0) break; leaf = path->nodes[0]; @@ -557,7 +557,7 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, GFP_NOFS); if (!sums) { ret = -ENOMEM; - goto fail; + goto out; } sums->logical = start; @@ -576,11 +576,12 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, path->slots[0]++; } ret = 0; -fail: - while (ret < 0 && !list_empty(list)) { - sums = list_entry(list->next, struct btrfs_ordered_sum, list); - list_del(&sums->list); - kfree(sums); +out: + if (ret < 0) { + struct btrfs_ordered_sum *tmp_sums; + + list_for_each_entry_safe(sums, tmp_sums, list, list) + kfree(sums); } btrfs_free_path(path); -- cgit v1.2.3-59-g8ed1b From 8d2a83a97f1c86d23161b4d2e37e1b2c5220710a Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 12 Apr 2024 11:48:31 +0100 Subject: btrfs: make NOCOW checks for existence of checksums in a range more efficient Before deciding if we can do a NOCOW write into a range, one of the things we have to do is check if there are checksum items for that range. We do that through the btrfs_lookup_csums_list() function, which searches for checksums and adds them to a list supplied by the caller. But all we need is to check if there is any checksum, we don't need to look for all of them and collect them into a list, which requires more search time in the checksums tree, allocating memory for checksums items to add to the list, copy checksums from a leaf into those list items, then free that memory, etc. This is all unnecessary overhead, wasting mostly CPU time, and perhaps some occasional IO if we need to read from disk any extent buffers. So change btrfs_lookup_csums_list() to allow to return immediately in case it finds any checksum, without the need to add it to a list and read it from a leaf. This is accomplished by allowing a NULL list parameter and making the function return 1 if it found any checksum, 0 if it didn't found any, and a negative value in case of an error. The following test with fio was used to measure performance: $ cat test.sh #!/bin/bash DEV=/dev/nullb0 MNT=/mnt/nullb0 cat < /tmp/fio-job.ini [global] name=fio-rand-write filename=$MNT/fio-rand-write rw=randwrite bssplit=4k/20:8k/20:16k/20:32k/20:64k/20 direct=1 numjobs=16 fallocate=posix time_based runtime=300 [file1] size=8G ioengine=io_uring iodepth=16 EOF umount $MNT &> /dev/null mkfs.btrfs -f $DEV mount -o ssd $DEV $MNT fio /tmp/fio-job.ini umount $MNT The test was run on a release kernel (Debian's default kernel config). The results before this patch: WRITE: bw=139MiB/s (146MB/s), 8204KiB/s-9504KiB/s (8401kB/s-9732kB/s), io=17.0GiB (18.3GB), run=125317-125344msec The results after this patch: WRITE: bw=153MiB/s (160MB/s), 9241KiB/s-10.0MiB/s (9463kB/s-10.5MB/s), io=17.0GiB (18.3GB), run=114054-114071msec Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file-item.c | 25 ++++++++++++++++++------- fs/btrfs/inode.c | 18 ++---------------- fs/btrfs/relocation.c | 2 +- fs/btrfs/tree-log.c | 9 ++++++--- 4 files changed, 27 insertions(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 231abcc87f63..1ea1ed44fe42 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -457,9 +457,12 @@ blk_status_t btrfs_lookup_bio_sums(struct btrfs_bio *bbio) * @start: Logical address of target checksum range. * @end: End offset (inclusive) of the target checksum range. * @list: List for adding each checksum that was found. + * Can be NULL in case the caller only wants to check if + * there any checksums for the range. * @nowait: Indicate if the search must be non-blocking or not. * - * Return < 0 on error and 0 on success. + * Return < 0 on error, 0 if no checksums were found, or 1 if checksums were + * found. */ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, struct list_head *list, bool nowait) @@ -471,6 +474,7 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, struct btrfs_ordered_sum *sums; struct btrfs_csum_item *item; int ret; + bool found_csums = false; ASSERT(IS_ALIGNED(start, fs_info->sectorsize) && IS_ALIGNED(end + 1, fs_info->sectorsize)); @@ -544,6 +548,10 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, continue; } + found_csums = true; + if (!list) + goto out; + csum_end = min(csum_end, end + 1); item = btrfs_item_ptr(path->nodes[0], path->slots[0], struct btrfs_csum_item); @@ -575,17 +583,20 @@ int btrfs_lookup_csums_list(struct btrfs_root *root, u64 start, u64 end, } path->slots[0]++; } - ret = 0; out: + btrfs_free_path(path); if (ret < 0) { - struct btrfs_ordered_sum *tmp_sums; + if (list) { + struct btrfs_ordered_sum *tmp_sums; - list_for_each_entry_safe(sums, tmp_sums, list, list) - kfree(sums); + list_for_each_entry_safe(sums, tmp_sums, list, list) + kfree(sums); + } + + return ret; } - btrfs_free_path(path); - return ret; + return found_csums ? 1 : 0; } /* diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index f4c54596c5c4..99816d5d605b 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1740,23 +1740,9 @@ static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes, bool nowait) { struct btrfs_root *csum_root = btrfs_csum_root(fs_info, bytenr); - struct btrfs_ordered_sum *sums; - int ret; - LIST_HEAD(list); - - ret = btrfs_lookup_csums_list(csum_root, bytenr, bytenr + num_bytes - 1, - &list, nowait); - if (ret == 0 && list_empty(&list)) - return 0; - while (!list_empty(&list)) { - sums = list_entry(list.next, struct btrfs_ordered_sum, list); - list_del(&sums->list); - kfree(sums); - } - if (ret < 0) - return ret; - return 1; + return btrfs_lookup_csums_list(csum_root, bytenr, bytenr + num_bytes - 1, + NULL, nowait); } static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page, diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 516c6e4610d7..ca70b62163d8 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4447,7 +4447,7 @@ int btrfs_reloc_clone_csums(struct btrfs_ordered_extent *ordered) ret = btrfs_lookup_csums_list(csum_root, disk_bytenr, disk_bytenr + ordered->num_bytes - 1, &list, false); - if (ret) + if (ret < 0) return ret; while (!list_empty(&list)) { diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 079fd0288d9c..201141eb2a8b 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -798,8 +798,9 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, ret = btrfs_lookup_csums_list(root->log_root, csum_start, csum_end - 1, &ordered_sums, false); - if (ret) + if (ret < 0) goto out; + ret = 0; /* * Now delete all existing cums in the csum root that * cover our range. We do this because we can have an @@ -4461,8 +4462,9 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, ret = btrfs_lookup_csums_list(csum_root, disk_bytenr, disk_bytenr + extent_num_bytes - 1, &ordered_sums, false); - if (ret) + if (ret < 0) goto out; + ret = 0; list_for_each_entry_safe(sums, sums_next, &ordered_sums, list) { if (!ret) @@ -4656,8 +4658,9 @@ static int log_extent_csums(struct btrfs_trans_handle *trans, ret = btrfs_lookup_csums_list(csum_root, em->block_start + csum_offset, em->block_start + csum_offset + csum_len - 1, &ordered_sums, false); - if (ret) + if (ret < 0) return ret; + ret = 0; while (!list_empty(&ordered_sums)) { struct btrfs_ordered_sum *sums = list_entry(ordered_sums.next, -- cgit v1.2.3-59-g8ed1b From 236e3107fc807d38a62691b3e276d7a190a67953 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 12 Apr 2024 12:40:25 +0100 Subject: btrfs: open code csum_exist_in_range() The csum_exist_in_range() function is now too trivial and is only used in one place, so open code it in its single caller. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 99816d5d605b..9f3ed58f7fde 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1736,15 +1736,6 @@ static noinline int run_delalloc_cow(struct btrfs_inode *inode, return 1; } -static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, - u64 bytenr, u64 num_bytes, bool nowait) -{ - struct btrfs_root *csum_root = btrfs_csum_root(fs_info, bytenr); - - return btrfs_lookup_csums_list(csum_root, bytenr, bytenr + num_bytes - 1, - NULL, nowait); -} - static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page, const u64 start, const u64 end) { @@ -1859,6 +1850,7 @@ static int can_nocow_file_extent(struct btrfs_path *path, struct extent_buffer *leaf = path->nodes[0]; struct btrfs_root *root = inode->root; struct btrfs_file_extent_item *fi; + struct btrfs_root *csum_root; u64 extent_end; u8 extent_type; int can_nocow = 0; @@ -1919,7 +1911,7 @@ static int can_nocow_file_extent(struct btrfs_path *path, if (args->free_path) { /* * We don't need the path anymore, plus through the - * csum_exist_in_range() call below we will end up allocating + * btrfs_lookup_csums_list() call below we will end up allocating * another path. So free the path to avoid unnecessary extra * memory usage. */ @@ -1940,8 +1932,11 @@ static int can_nocow_file_extent(struct btrfs_path *path, * Force COW if csums exist in the range. This ensures that csums for a * given extent are either valid or do not exist. */ - ret = csum_exist_in_range(root->fs_info, args->disk_bytenr, args->num_bytes, - nowait); + + csum_root = btrfs_csum_root(root->fs_info, args->disk_bytenr); + ret = btrfs_lookup_csums_list(csum_root, args->disk_bytenr, + args->disk_bytenr + args->num_bytes - 1, + NULL, nowait); WARN_ON_ONCE(ret > 0 && is_freespace_inode); if (ret != 0) goto out; -- cgit v1.2.3-59-g8ed1b From 0a308f8095f74c732e055c3d9f897d56e8ed6e82 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Jan 2024 15:13:35 +0000 Subject: btrfs: pass an inode to btrfs_add_extent_mapping() Instead of passing fs_info and extent map tree arguments to btrfs_add_extent_mapping(), we can pass an inode instead, as extent maps are always inserted in the extent map tree of an inode, and the fs_info can be extracted from the inode (inode->root->fs_info). The only exception is in the self tests where we allocate an extent map tree and then use it to insert/update/remove extent maps. However the tests can be changed to use a test inode and then use the inode's extent map tree. So change btrfs_add_extent_mapping() to have an inode as an argument instead of a fs_info and an extent map tree. This reduces the number of parameters and will also be needed for an upcoming change. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 14 +-- fs/btrfs/extent_map.h | 3 +- fs/btrfs/inode.c | 2 +- fs/btrfs/tests/extent-map-tests.c | 174 +++++++++++++++++++------------------- 4 files changed, 95 insertions(+), 98 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 955ce300e5a1..ac4361792ffc 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -546,10 +546,9 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, } /* - * Add extent mapping into em_tree. + * Add extent mapping into an inode's extent map tree. * - * @fs_info: the filesystem - * @em_tree: extent tree into which we want to insert the extent mapping + * @inode: target inode * @em_in: extent we are inserting * @start: start of the logical range btrfs_get_extent() is requesting * @len: length of the logical range btrfs_get_extent() is requesting @@ -557,8 +556,8 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, * Note that @em_in's range may be different from [start, start+len), * but they must be overlapped. * - * Insert @em_in into @em_tree. In case there is an overlapping range, handle - * the -EEXIST by either: + * Insert @em_in into the inode's extent map tree. In case there is an + * overlapping range, handle the -EEXIST by either: * a) Returning the existing extent in @em_in if @start is within the * existing em. * b) Merge the existing extent with @em_in passed in. @@ -566,12 +565,13 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, * Return 0 on success, otherwise -EEXIST. * */ -int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree, +int btrfs_add_extent_mapping(struct btrfs_inode *inode, struct extent_map **em_in, u64 start, u64 len) { int ret; struct extent_map *em = *em_in; + struct extent_map_tree *em_tree = &inode->extent_tree; + struct btrfs_fs_info *fs_info = inode->root->fs_info; /* * Tree-checker should have rejected any inline extent with non-zero diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index 10e9491865c9..f287ab46e368 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -132,8 +132,7 @@ int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen); void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em); struct extent_map *search_extent_mapping(struct extent_map_tree *tree, u64 start, u64 len); -int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree, +int btrfs_add_extent_mapping(struct btrfs_inode *inode, struct extent_map **em_in, u64 start, u64 len); void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 9f3ed58f7fde..7ce436262e4a 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6972,7 +6972,7 @@ insert: } write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); + ret = btrfs_add_extent_mapping(inode, &em, start, len); write_unlock(&em_tree->lock); out: btrfs_free_path(path); diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c index 47b5d301038e..de2e8483220e 100644 --- a/fs/btrfs/tests/extent-map-tests.c +++ b/fs/btrfs/tests/extent-map-tests.c @@ -53,9 +53,9 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree) * ->add_extent_mapping(0, 16K) * -> #handle -EEXIST */ -static int test_case_1(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree) +static int test_case_1(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; u64 start = 0; u64 len = SZ_8K; @@ -73,7 +73,7 @@ static int test_case_1(struct btrfs_fs_info *fs_info, em->block_start = 0; em->block_len = SZ_16K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("cannot add extent range [0, 16K)"); @@ -94,7 +94,7 @@ static int test_case_1(struct btrfs_fs_info *fs_info, em->block_start = SZ_32K; /* avoid merging */ em->block_len = SZ_4K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("cannot add extent range [16K, 20K)"); @@ -115,7 +115,7 @@ static int test_case_1(struct btrfs_fs_info *fs_info, em->block_start = start; em->block_len = len; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret) { test_err("case1 [%llu %llu]: ret %d", start, start + len, ret); @@ -148,9 +148,9 @@ out: * Reading the inline ending up with EEXIST, ie. read an inline * extent and discard page cache and read it again. */ -static int test_case_2(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree) +static int test_case_2(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; int ret; @@ -166,7 +166,7 @@ static int test_case_2(struct btrfs_fs_info *fs_info, em->block_start = EXTENT_MAP_INLINE; em->block_len = (u64)-1; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("cannot add extent range [0, 1K)"); @@ -187,7 +187,7 @@ static int test_case_2(struct btrfs_fs_info *fs_info, em->block_start = SZ_4K; em->block_len = SZ_4K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("cannot add extent range [4K, 8K)"); @@ -208,7 +208,7 @@ static int test_case_2(struct btrfs_fs_info *fs_info, em->block_start = EXTENT_MAP_INLINE; em->block_len = (u64)-1; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret) { test_err("case2 [0 1K]: ret %d", ret); @@ -235,8 +235,9 @@ out: } static int __test_case_3(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree, u64 start) + struct btrfs_inode *inode, u64 start) { + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; u64 len = SZ_4K; int ret; @@ -253,7 +254,7 @@ static int __test_case_3(struct btrfs_fs_info *fs_info, em->block_start = SZ_4K; em->block_len = SZ_4K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("cannot add extent range [4K, 8K)"); @@ -274,7 +275,7 @@ static int __test_case_3(struct btrfs_fs_info *fs_info, em->block_start = 0; em->block_len = SZ_16K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); + ret = btrfs_add_extent_mapping(inode, &em, start, len); write_unlock(&em_tree->lock); if (ret) { test_err("case3 [%llu %llu): ret %d", @@ -322,25 +323,25 @@ out: * -> add_extent_mapping() * -> add_extent_mapping() */ -static int test_case_3(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree) +static int test_case_3(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { int ret; - ret = __test_case_3(fs_info, em_tree, 0); + ret = __test_case_3(fs_info, inode, 0); if (ret) return ret; - ret = __test_case_3(fs_info, em_tree, SZ_8K); + ret = __test_case_3(fs_info, inode, SZ_8K); if (ret) return ret; - ret = __test_case_3(fs_info, em_tree, (12 * SZ_1K)); + ret = __test_case_3(fs_info, inode, (12 * SZ_1K)); return ret; } static int __test_case_4(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree, u64 start) + struct btrfs_inode *inode, u64 start) { + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; u64 len = SZ_4K; int ret; @@ -357,7 +358,7 @@ static int __test_case_4(struct btrfs_fs_info *fs_info, em->block_start = 0; em->block_len = SZ_8K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("cannot add extent range [0, 8K)"); @@ -378,7 +379,7 @@ static int __test_case_4(struct btrfs_fs_info *fs_info, em->block_start = SZ_16K; /* avoid merging */ em->block_len = 24 * SZ_1K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("cannot add extent range [8K, 32K)"); @@ -398,7 +399,7 @@ static int __test_case_4(struct btrfs_fs_info *fs_info, em->block_start = 0; em->block_len = SZ_32K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); + ret = btrfs_add_extent_mapping(inode, &em, start, len); write_unlock(&em_tree->lock); if (ret) { test_err("case4 [%llu %llu): ret %d", @@ -450,23 +451,22 @@ out: * # handle -EEXIST when adding * # [0, 32K) */ -static int test_case_4(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree) +static int test_case_4(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { int ret; - ret = __test_case_4(fs_info, em_tree, 0); + ret = __test_case_4(fs_info, inode, 0); if (ret) return ret; - ret = __test_case_4(fs_info, em_tree, SZ_4K); + ret = __test_case_4(fs_info, inode, SZ_4K); return ret; } -static int add_compressed_extent(struct btrfs_fs_info *fs_info, - struct extent_map_tree *em_tree, +static int add_compressed_extent(struct btrfs_inode *inode, u64 start, u64 len, u64 block_start) { + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; int ret; @@ -482,7 +482,7 @@ static int add_compressed_extent(struct btrfs_fs_info *fs_info, em->block_len = SZ_4K; em->flags |= EXTENT_FLAG_COMPRESS_ZLIB; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); free_extent_map(em); if (ret < 0) { @@ -588,53 +588,43 @@ static int validate_range(struct extent_map_tree *em_tree, int index) * They'll have the EXTENT_FLAG_COMPRESSED flag set to keep the em tree from * merging the em's. */ -static int test_case_5(struct btrfs_fs_info *fs_info) +static int test_case_5(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { - struct extent_map_tree *em_tree; - struct inode *inode; u64 start, end; int ret; test_msg("Running btrfs_drop_extent_map_range tests"); - inode = btrfs_new_test_inode(); - if (!inode) { - test_std_err(TEST_ALLOC_INODE); - return -ENOMEM; - } - - em_tree = &BTRFS_I(inode)->extent_tree; - /* [0, 12k) */ - ret = add_compressed_extent(fs_info, em_tree, 0, SZ_4K * 3, 0); + ret = add_compressed_extent(inode, 0, SZ_4K * 3, 0); if (ret) { test_err("cannot add extent range [0, 12K)"); goto out; } /* [12k, 24k) */ - ret = add_compressed_extent(fs_info, em_tree, SZ_4K * 3, SZ_4K * 3, SZ_4K); + ret = add_compressed_extent(inode, SZ_4K * 3, SZ_4K * 3, SZ_4K); if (ret) { test_err("cannot add extent range [12k, 24k)"); goto out; } /* [24k, 36k) */ - ret = add_compressed_extent(fs_info, em_tree, SZ_4K * 6, SZ_4K * 3, SZ_8K); + ret = add_compressed_extent(inode, SZ_4K * 6, SZ_4K * 3, SZ_8K); if (ret) { test_err("cannot add extent range [12k, 24k)"); goto out; } /* [36k, 40k) */ - ret = add_compressed_extent(fs_info, em_tree, SZ_32K + SZ_4K, SZ_4K, SZ_4K * 3); + ret = add_compressed_extent(inode, SZ_32K + SZ_4K, SZ_4K, SZ_4K * 3); if (ret) { test_err("cannot add extent range [12k, 24k)"); goto out; } /* [40k, 64k) */ - ret = add_compressed_extent(fs_info, em_tree, SZ_4K * 10, SZ_4K * 6, SZ_16K); + ret = add_compressed_extent(inode, SZ_4K * 10, SZ_4K * 6, SZ_16K); if (ret) { test_err("cannot add extent range [12k, 24k)"); goto out; @@ -643,36 +633,36 @@ static int test_case_5(struct btrfs_fs_info *fs_info) /* Drop [8k, 12k) */ start = SZ_8K; end = (3 * SZ_4K) - 1; - btrfs_drop_extent_map_range(BTRFS_I(inode), start, end, false); - ret = validate_range(&BTRFS_I(inode)->extent_tree, 0); + btrfs_drop_extent_map_range(inode, start, end, false); + ret = validate_range(&inode->extent_tree, 0); if (ret) goto out; /* Drop [12k, 20k) */ start = SZ_4K * 3; end = SZ_16K + SZ_4K - 1; - btrfs_drop_extent_map_range(BTRFS_I(inode), start, end, false); - ret = validate_range(&BTRFS_I(inode)->extent_tree, 1); + btrfs_drop_extent_map_range(inode, start, end, false); + ret = validate_range(&inode->extent_tree, 1); if (ret) goto out; /* Drop [28k, 32k) */ start = SZ_32K - SZ_4K; end = SZ_32K - 1; - btrfs_drop_extent_map_range(BTRFS_I(inode), start, end, false); - ret = validate_range(&BTRFS_I(inode)->extent_tree, 2); + btrfs_drop_extent_map_range(inode, start, end, false); + ret = validate_range(&inode->extent_tree, 2); if (ret) goto out; /* Drop [32k, 64k) */ start = SZ_32K; end = SZ_64K - 1; - btrfs_drop_extent_map_range(BTRFS_I(inode), start, end, false); - ret = validate_range(&BTRFS_I(inode)->extent_tree, 3); + btrfs_drop_extent_map_range(inode, start, end, false); + ret = validate_range(&inode->extent_tree, 3); if (ret) goto out; out: - iput(inode); + free_extent_map_tree(&inode->extent_tree); return ret; } @@ -681,23 +671,25 @@ out: * for areas between two existing ems. Validate it doesn't do this when there * are two unmerged em's side by side. */ -static int test_case_6(struct btrfs_fs_info *fs_info, struct extent_map_tree *em_tree) +static int test_case_6(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em = NULL; int ret; - ret = add_compressed_extent(fs_info, em_tree, 0, SZ_4K, 0); + ret = add_compressed_extent(inode, 0, SZ_4K, 0); if (ret) goto out; - ret = add_compressed_extent(fs_info, em_tree, SZ_4K, SZ_4K, 0); + ret = add_compressed_extent(inode, SZ_4K, SZ_4K, 0); if (ret) goto out; em = alloc_extent_map(); if (!em) { test_std_err(TEST_ALLOC_EXTENT_MAP); - return -ENOMEM; + ret = -ENOMEM; + goto out; } em->start = SZ_4K; @@ -705,7 +697,7 @@ static int test_case_6(struct btrfs_fs_info *fs_info, struct extent_map_tree *em em->block_start = SZ_16K; em->block_len = SZ_16K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, 0, SZ_8K); + ret = btrfs_add_extent_mapping(inode, &em, 0, SZ_8K); write_unlock(&em_tree->lock); if (ret != 0) { @@ -734,28 +726,19 @@ out: * true would mess up the start/end calculations and subsequent splits would be * incorrect. */ -static int test_case_7(struct btrfs_fs_info *fs_info) +static int test_case_7(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { - struct extent_map_tree *em_tree; + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; - struct inode *inode; int ret; + int ret2; test_msg("Running btrfs_drop_extent_cache with pinned"); - inode = btrfs_new_test_inode(); - if (!inode) { - test_std_err(TEST_ALLOC_INODE); - return -ENOMEM; - } - - em_tree = &BTRFS_I(inode)->extent_tree; - em = alloc_extent_map(); if (!em) { test_std_err(TEST_ALLOC_EXTENT_MAP); - ret = -ENOMEM; - goto out; + return -ENOMEM; } /* [0, 16K), pinned */ @@ -765,7 +748,7 @@ static int test_case_7(struct btrfs_fs_info *fs_info) em->block_len = SZ_4K; em->flags |= EXTENT_FLAG_PINNED; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("couldn't add extent map"); @@ -786,7 +769,7 @@ static int test_case_7(struct btrfs_fs_info *fs_info) em->block_start = SZ_32K; em->block_len = SZ_16K; write_lock(&em_tree->lock); - ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(inode, &em, em->start, em->len); write_unlock(&em_tree->lock); if (ret < 0) { test_err("couldn't add extent map"); @@ -798,7 +781,7 @@ static int test_case_7(struct btrfs_fs_info *fs_info) * Drop [0, 36K) This should skip the [0, 4K) extent and then split the * [32K, 48K) extent. */ - btrfs_drop_extent_map_range(BTRFS_I(inode), 0, (36 * SZ_1K) - 1, true); + btrfs_drop_extent_map_range(inode, 0, (36 * SZ_1K) - 1, true); /* Make sure our extent maps look sane. */ ret = -EINVAL; @@ -865,7 +848,11 @@ static int test_case_7(struct btrfs_fs_info *fs_info) ret = 0; out: free_extent_map(em); - iput(inode); + /* Unpin our extent to prevent warning when removing it below. */ + ret2 = unpin_extent_cache(inode, 0, SZ_16K, 0); + if (ret == 0) + ret = ret2; + free_extent_map_tree(em_tree); return ret; } @@ -959,7 +946,8 @@ out_free: int btrfs_test_extent_map(void) { struct btrfs_fs_info *fs_info = NULL; - struct extent_map_tree *em_tree; + struct inode *inode; + struct btrfs_root *root = NULL; int ret = 0, i; struct rmap_test_vector rmap_tests[] = { { @@ -1008,33 +996,42 @@ int btrfs_test_extent_map(void) return -ENOMEM; } - em_tree = kzalloc(sizeof(*em_tree), GFP_KERNEL); - if (!em_tree) { + inode = btrfs_new_test_inode(); + if (!inode) { + test_std_err(TEST_ALLOC_INODE); ret = -ENOMEM; goto out; } - extent_map_tree_init(em_tree); + root = btrfs_alloc_dummy_root(fs_info); + if (IS_ERR(root)) { + test_std_err(TEST_ALLOC_ROOT); + ret = PTR_ERR(root); + root = NULL; + goto out; + } - ret = test_case_1(fs_info, em_tree); + BTRFS_I(inode)->root = root; + + ret = test_case_1(fs_info, BTRFS_I(inode)); if (ret) goto out; - ret = test_case_2(fs_info, em_tree); + ret = test_case_2(fs_info, BTRFS_I(inode)); if (ret) goto out; - ret = test_case_3(fs_info, em_tree); + ret = test_case_3(fs_info, BTRFS_I(inode)); if (ret) goto out; - ret = test_case_4(fs_info, em_tree); + ret = test_case_4(fs_info, BTRFS_I(inode)); if (ret) goto out; - ret = test_case_5(fs_info); + ret = test_case_5(fs_info, BTRFS_I(inode)); if (ret) goto out; - ret = test_case_6(fs_info, em_tree); + ret = test_case_6(fs_info, BTRFS_I(inode)); if (ret) goto out; - ret = test_case_7(fs_info); + ret = test_case_7(fs_info, BTRFS_I(inode)); if (ret) goto out; @@ -1046,7 +1043,8 @@ int btrfs_test_extent_map(void) } out: - kfree(em_tree); + iput(inode); + btrfs_free_dummy_root(root); btrfs_free_dummy_fs_info(fs_info); return ret; -- cgit v1.2.3-59-g8ed1b From 071533da5f2e7623089e69252bd949610d6c8194 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Jan 2024 16:04:26 +0000 Subject: btrfs: tests: error out on unexpected extent map reference count In the extent map self tests, when freeing all extent maps from a test extent map tree we are not expecting to find any extent map with a reference count different from 1 (the tree reference). If we find any, we just log a message but we don't fail the test, which makes it very easy to miss any bug/regression - no one reads the test messages unless a test fails. So change the behaviour to make a test fail if we find an extent map in the tree with a reference count different from 1. Make the failure happen only after removing all extent maps, so that we don't leak memory. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/tests/extent-map-tests.c | 43 +++++++++++++++++++++++++++++++-------- 1 file changed, 35 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c index de2e8483220e..0f5c9c9304d9 100644 --- a/fs/btrfs/tests/extent-map-tests.c +++ b/fs/btrfs/tests/extent-map-tests.c @@ -11,10 +11,11 @@ #include "../disk-io.h" #include "../block-group.h" -static void free_extent_map_tree(struct extent_map_tree *em_tree) +static int free_extent_map_tree(struct extent_map_tree *em_tree) { struct extent_map *em; struct rb_node *node; + int ret = 0; write_lock(&em_tree->lock); while (!RB_EMPTY_ROOT(&em_tree->map.rb_root)) { @@ -24,6 +25,7 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree) #ifdef CONFIG_BTRFS_DEBUG if (refcount_read(&em->refs) != 1) { + ret = -EINVAL; test_err( "em leak: em (start %llu len %llu block_start %llu block_len %llu) refs %d", em->start, em->len, em->block_start, @@ -35,6 +37,8 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree) free_extent_map(em); } write_unlock(&em_tree->lock); + + return ret; } /* @@ -60,6 +64,7 @@ static int test_case_1(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) u64 start = 0; u64 len = SZ_8K; int ret; + int ret2; em = alloc_extent_map(); if (!em) { @@ -137,7 +142,9 @@ static int test_case_1(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) } free_extent_map(em); out: - free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(em_tree); + if (ret == 0) + ret = ret2; return ret; } @@ -153,6 +160,7 @@ static int test_case_2(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; int ret; + int ret2; em = alloc_extent_map(); if (!em) { @@ -229,7 +237,9 @@ static int test_case_2(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) } free_extent_map(em); out: - free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(em_tree); + if (ret == 0) + ret = ret2; return ret; } @@ -241,6 +251,7 @@ static int __test_case_3(struct btrfs_fs_info *fs_info, struct extent_map *em; u64 len = SZ_4K; int ret; + int ret2; em = alloc_extent_map(); if (!em) { @@ -302,7 +313,9 @@ static int __test_case_3(struct btrfs_fs_info *fs_info, } free_extent_map(em); out: - free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(em_tree); + if (ret == 0) + ret = ret2; return ret; } @@ -345,6 +358,7 @@ static int __test_case_4(struct btrfs_fs_info *fs_info, struct extent_map *em; u64 len = SZ_4K; int ret; + int ret2; em = alloc_extent_map(); if (!em) { @@ -421,7 +435,9 @@ static int __test_case_4(struct btrfs_fs_info *fs_info, } free_extent_map(em); out: - free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(em_tree); + if (ret == 0) + ret = ret2; return ret; } @@ -592,6 +608,7 @@ static int test_case_5(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) { u64 start, end; int ret; + int ret2; test_msg("Running btrfs_drop_extent_map_range tests"); @@ -662,7 +679,10 @@ static int test_case_5(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) if (ret) goto out; out: - free_extent_map_tree(&inode->extent_tree); + ret2 = free_extent_map_tree(&inode->extent_tree); + if (ret == 0) + ret = ret2; + return ret; } @@ -676,6 +696,7 @@ static int test_case_6(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em = NULL; int ret; + int ret2; ret = add_compressed_extent(inode, 0, SZ_4K, 0); if (ret) @@ -717,7 +738,10 @@ static int test_case_6(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) ret = 0; out: free_extent_map(em); - free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(em_tree); + if (ret == 0) + ret = ret2; + return ret; } @@ -852,7 +876,10 @@ out: ret2 = unpin_extent_cache(inode, 0, SZ_16K, 0); if (ret == 0) ret = ret2; - free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(em_tree); + if (ret == 0) + ret = ret2; + return ret; } -- cgit v1.2.3-59-g8ed1b From ed48adf83e09041d7ec3ec9ef256f1b3f660c52d Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 16 Jan 2024 16:00:16 +0000 Subject: btrfs: simplify add_extent_mapping() by removing pointless label The add_extent_mapping() function is short and trivial, there's no need to have a label for a quick exit in case of an error, even because there's no error handling needed, we just need to return the error. So remove that label and return directly. Also while at it remove the redundant initialization of 'ret', as that may help avoid some warnings with clang tools such as the one reported/fixed by commit 966de47ff0c9 ("btrfs: remove redundant initialization of variables in log_new_ancestors"). Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index ac4361792ffc..4a9fbcb24d9b 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -370,17 +370,17 @@ static inline void setup_extent_mapping(struct extent_map_tree *tree, static int add_extent_mapping(struct extent_map_tree *tree, struct extent_map *em, int modified) { - int ret = 0; + int ret; lockdep_assert_held_write(&tree->lock); ret = tree_insert(&tree->map, em); if (ret) - goto out; + return ret; setup_extent_mapping(tree, em, modified); -out: - return ret; + + return 0; } static struct extent_map * -- cgit v1.2.3-59-g8ed1b From 5e485ac6f0f34126ba82267c09cc5ab38cee8908 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 12:30:59 +0100 Subject: btrfs: export find_next_inode() as btrfs_find_first_inode() Export the relocation private helper find_next_inode() to inode.c, as this same logic is also used at btrfs_prune_dentries() and will be used by an upcoming change that adds an extent map shrinker. The next patch will change btrfs_prune_dentries() to use this helper. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/btrfs_inode.h | 1 + fs/btrfs/inode.c | 59 +++++++++++++++++++++++++++ fs/btrfs/relocation.c | 105 ++++++++++++------------------------------------- 3 files changed, 85 insertions(+), 80 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index ed8bd15aa3e2..9a87ada7fe52 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -543,6 +543,7 @@ ssize_t btrfs_dio_read(struct kiocb *iocb, struct iov_iter *iter, size_t done_before); struct iomap_dio *btrfs_dio_write(struct kiocb *iocb, struct iov_iter *iter, size_t done_before); +struct btrfs_inode *btrfs_find_first_inode(struct btrfs_root *root, u64 min_ino); extern const struct dentry_operations btrfs_dentry_operations; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7ce436262e4a..adf7522a1f56 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -10807,6 +10807,65 @@ void btrfs_assert_inode_range_clean(struct btrfs_inode *inode, u64 start, u64 en ASSERT(ordered == NULL); } +/* + * Find the first inode with a minimum number. + * + * @root: The root to search for. + * @min_ino: The minimum inode number. + * + * Find the first inode in the @root with a number >= @min_ino and return it. + * Returns NULL if no such inode found. + */ +struct btrfs_inode *btrfs_find_first_inode(struct btrfs_root *root, u64 min_ino) +{ + struct rb_node *node; + struct rb_node *prev; + struct btrfs_inode *inode; + + spin_lock(&root->inode_lock); +again: + node = root->inode_tree.rb_node; + prev = NULL; + while (node) { + prev = node; + inode = rb_entry(node, struct btrfs_inode, rb_node); + if (min_ino < btrfs_ino(inode)) + node = node->rb_left; + else if (min_ino > btrfs_ino(inode)) + node = node->rb_right; + else + break; + } + + if (!node) { + while (prev) { + inode = rb_entry(prev, struct btrfs_inode, rb_node); + if (min_ino <= btrfs_ino(inode)) { + node = prev; + break; + } + prev = rb_next(prev); + } + } + + while (node) { + inode = rb_entry(prev, struct btrfs_inode, rb_node); + if (igrab(&inode->vfs_inode)) { + spin_unlock(&root->inode_lock); + return inode; + } + + min_ino = btrfs_ino(inode) + 1; + if (cond_resched_lock(&root->inode_lock)) + goto again; + + node = rb_next(node); + } + spin_unlock(&root->inode_lock); + + return NULL; +} + static const struct inode_operations btrfs_dir_inode_operations = { .getattr = btrfs_getattr, .lookup = btrfs_lookup, diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index ca70b62163d8..5a01aaa164de 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -951,60 +951,6 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans, return ret; } -/* - * helper to find first cached inode with inode number >= objectid - * in a subvolume - */ -static struct inode *find_next_inode(struct btrfs_root *root, u64 objectid) -{ - struct rb_node *node; - struct rb_node *prev; - struct btrfs_inode *entry; - struct inode *inode; - - spin_lock(&root->inode_lock); -again: - node = root->inode_tree.rb_node; - prev = NULL; - while (node) { - prev = node; - entry = rb_entry(node, struct btrfs_inode, rb_node); - - if (objectid < btrfs_ino(entry)) - node = node->rb_left; - else if (objectid > btrfs_ino(entry)) - node = node->rb_right; - else - break; - } - if (!node) { - while (prev) { - entry = rb_entry(prev, struct btrfs_inode, rb_node); - if (objectid <= btrfs_ino(entry)) { - node = prev; - break; - } - prev = rb_next(prev); - } - } - while (node) { - entry = rb_entry(node, struct btrfs_inode, rb_node); - inode = igrab(&entry->vfs_inode); - if (inode) { - spin_unlock(&root->inode_lock); - return inode; - } - - objectid = btrfs_ino(entry) + 1; - if (cond_resched_lock(&root->inode_lock)) - goto again; - - node = rb_next(node); - } - spin_unlock(&root->inode_lock); - return NULL; -} - /* * get new location of data */ @@ -1065,7 +1011,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_key key; struct btrfs_file_extent_item *fi; - struct inode *inode = NULL; + struct btrfs_inode *inode = NULL; u64 parent; u64 bytenr; u64 new_bytenr = 0; @@ -1112,13 +1058,13 @@ int replace_file_extents(struct btrfs_trans_handle *trans, */ if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) { if (first) { - inode = find_next_inode(root, key.objectid); + inode = btrfs_find_first_inode(root, key.objectid); first = 0; - } else if (inode && btrfs_ino(BTRFS_I(inode)) < key.objectid) { - btrfs_add_delayed_iput(BTRFS_I(inode)); - inode = find_next_inode(root, key.objectid); + } else if (inode && btrfs_ino(inode) < key.objectid) { + btrfs_add_delayed_iput(inode); + inode = btrfs_find_first_inode(root, key.objectid); } - if (inode && btrfs_ino(BTRFS_I(inode)) == key.objectid) { + if (inode && btrfs_ino(inode) == key.objectid) { struct extent_state *cached_state = NULL; end = key.offset + @@ -1128,21 +1074,19 @@ int replace_file_extents(struct btrfs_trans_handle *trans, WARN_ON(!IS_ALIGNED(end, fs_info->sectorsize)); end--; /* Take mmap lock to serialize with reflinks. */ - if (!down_read_trylock(&BTRFS_I(inode)->i_mmap_lock)) + if (!down_read_trylock(&inode->i_mmap_lock)) continue; - ret = try_lock_extent(&BTRFS_I(inode)->io_tree, - key.offset, end, - &cached_state); + ret = try_lock_extent(&inode->io_tree, key.offset, + end, &cached_state); if (!ret) { - up_read(&BTRFS_I(inode)->i_mmap_lock); + up_read(&inode->i_mmap_lock); continue; } - btrfs_drop_extent_map_range(BTRFS_I(inode), - key.offset, end, true); - unlock_extent(&BTRFS_I(inode)->io_tree, - key.offset, end, &cached_state); - up_read(&BTRFS_I(inode)->i_mmap_lock); + btrfs_drop_extent_map_range(inode, key.offset, end, true); + unlock_extent(&inode->io_tree, key.offset, end, + &cached_state); + up_read(&inode->i_mmap_lock); } } @@ -1185,7 +1129,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans, if (dirty) btrfs_mark_buffer_dirty(trans, leaf); if (inode) - btrfs_add_delayed_iput(BTRFS_I(inode)); + btrfs_add_delayed_iput(inode); return ret; } @@ -1527,7 +1471,7 @@ static int invalidate_extent_cache(struct btrfs_root *root, const struct btrfs_key *max_key) { struct btrfs_fs_info *fs_info = root->fs_info; - struct inode *inode = NULL; + struct btrfs_inode *inode = NULL; u64 objectid; u64 start, end; u64 ino; @@ -1537,23 +1481,24 @@ static int invalidate_extent_cache(struct btrfs_root *root, struct extent_state *cached_state = NULL; cond_resched(); - iput(inode); + if (inode) + iput(&inode->vfs_inode); if (objectid > max_key->objectid) break; - inode = find_next_inode(root, objectid); + inode = btrfs_find_first_inode(root, objectid); if (!inode) break; - ino = btrfs_ino(BTRFS_I(inode)); + ino = btrfs_ino(inode); if (ino > max_key->objectid) { - iput(inode); + iput(&inode->vfs_inode); break; } objectid = ino + 1; - if (!S_ISREG(inode->i_mode)) + if (!S_ISREG(inode->vfs_inode.i_mode)) continue; if (unlikely(min_key->objectid == ino)) { @@ -1586,9 +1531,9 @@ static int invalidate_extent_cache(struct btrfs_root *root, } /* the lock_extent waits for read_folio to complete */ - lock_extent(&BTRFS_I(inode)->io_tree, start, end, &cached_state); - btrfs_drop_extent_map_range(BTRFS_I(inode), start, end, true); - unlock_extent(&BTRFS_I(inode)->io_tree, start, end, &cached_state); + lock_extent(&inode->io_tree, start, end, &cached_state); + btrfs_drop_extent_map_range(inode, start, end, true); + unlock_extent(&inode->io_tree, start, end, &cached_state); } return 0; } -- cgit v1.2.3-59-g8ed1b From 26c0fae3e72ff1cf269891aae75c762dc04ee1dd Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 12:45:34 +0100 Subject: btrfs: use btrfs_find_first_inode() at btrfs_prune_dentries() Currently btrfs_prune_dentries() has open code to find the first inode in a root with a minimum inode number. Remove that code and make it use the helper btrfs_find_first_inode() for that task. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/inode.c | 66 ++++++++++++-------------------------------------------- 1 file changed, 14 insertions(+), 52 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index adf7522a1f56..629706b98e31 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4416,64 +4416,26 @@ out: static void btrfs_prune_dentries(struct btrfs_root *root) { struct btrfs_fs_info *fs_info = root->fs_info; - struct rb_node *node; - struct rb_node *prev; - struct btrfs_inode *entry; - struct inode *inode; - u64 objectid = 0; + struct btrfs_inode *inode; + u64 min_ino = 0; if (!BTRFS_FS_ERROR(fs_info)) WARN_ON(btrfs_root_refs(&root->root_item) != 0); - spin_lock(&root->inode_lock); -again: - node = root->inode_tree.rb_node; - prev = NULL; - while (node) { - prev = node; - entry = rb_entry(node, struct btrfs_inode, rb_node); - - if (objectid < btrfs_ino(entry)) - node = node->rb_left; - else if (objectid > btrfs_ino(entry)) - node = node->rb_right; - else - break; - } - if (!node) { - while (prev) { - entry = rb_entry(prev, struct btrfs_inode, rb_node); - if (objectid <= btrfs_ino(entry)) { - node = prev; - break; - } - prev = rb_next(prev); - } - } - while (node) { - entry = rb_entry(node, struct btrfs_inode, rb_node); - objectid = btrfs_ino(entry) + 1; - inode = igrab(&entry->vfs_inode); - if (inode) { - spin_unlock(&root->inode_lock); - if (atomic_read(&inode->i_count) > 1) - d_prune_aliases(inode); - /* - * btrfs_drop_inode will have it removed from the inode - * cache when its usage count hits zero. - */ - iput(inode); - cond_resched(); - spin_lock(&root->inode_lock); - goto again; - } - - if (cond_resched_lock(&root->inode_lock)) - goto again; + inode = btrfs_find_first_inode(root, min_ino); + while (inode) { + if (atomic_read(&inode->vfs_inode.i_count) > 1) + d_prune_aliases(&inode->vfs_inode); - node = rb_next(node); + min_ino = btrfs_ino(inode) + 1; + /* + * btrfs_drop_inode() will have it removed from the inode + * cache when its usage count hits zero. + */ + iput(&inode->vfs_inode); + cond_resched(); + inode = btrfs_find_first_inode(root, min_ino); } - spin_unlock(&root->inode_lock); } int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) -- cgit v1.2.3-59-g8ed1b From 0eea355fc0f48916f29d1f066d06df49f075a593 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 16:42:28 -0400 Subject: btrfs: add a helper to get the delayed ref node from the data/tree ref We have several different ways we refer to references throughout the code and it's not consistent and there's a bit of duplication. In order to clean this up I want to have one structure we use to define reference information, and one structure we use for the delayed reference information. Start this process by adding a helper to get from the btrfs_delayed_data_ref/btrfs_delayed_tree_ref to the btrfs_delayed_ref_node so that it'll make moving these structures around simpler. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 28 +++++++++++++++++++--------- fs/btrfs/delayed-ref.h | 12 ++++++++++++ 2 files changed, 31 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index e44e62cf76bc..d920663a18fd 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -310,7 +310,9 @@ int btrfs_delayed_refs_rsv_refill(struct btrfs_fs_info *fs_info, static int comp_tree_refs(struct btrfs_delayed_tree_ref *ref1, struct btrfs_delayed_tree_ref *ref2) { - if (ref1->node.type == BTRFS_TREE_BLOCK_REF_KEY) { + struct btrfs_delayed_ref_node *node = btrfs_delayed_tree_ref_to_node(ref1); + + if (node->type == BTRFS_TREE_BLOCK_REF_KEY) { if (ref1->root < ref2->root) return -1; if (ref1->root > ref2->root) @@ -330,7 +332,9 @@ static int comp_tree_refs(struct btrfs_delayed_tree_ref *ref1, static int comp_data_refs(struct btrfs_delayed_data_ref *ref1, struct btrfs_delayed_data_ref *ref2) { - if (ref1->node.type == BTRFS_EXTENT_DATA_REF_KEY) { + struct btrfs_delayed_ref_node *node = btrfs_delayed_data_ref_to_node(ref1); + + if (node->type == BTRFS_EXTENT_DATA_REF_KEY) { if (ref1->root < ref2->root) return -1; if (ref1->root > ref2->root) @@ -1061,6 +1065,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_delayed_tree_ref *ref; + struct btrfs_delayed_ref_node *node; struct btrfs_delayed_ref_head *head_ref; struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_qgroup_extent_record *record = NULL; @@ -1096,12 +1101,14 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, } } + node = btrfs_delayed_tree_ref_to_node(ref); + if (parent) ref_type = BTRFS_SHARED_BLOCK_REF_KEY; else ref_type = BTRFS_TREE_BLOCK_REF_KEY; - init_delayed_ref_common(fs_info, &ref->node, bytenr, num_bytes, + init_delayed_ref_common(fs_info, node, bytenr, num_bytes, generic_ref->tree_ref.ref_root, action, ref_type); ref->root = generic_ref->tree_ref.ref_root; @@ -1123,7 +1130,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, head_ref = add_delayed_ref_head(trans, head_ref, record, action, &qrecord_inserted); - merged = insert_delayed_ref(trans, head_ref, &ref->node); + merged = insert_delayed_ref(trans, head_ref, node); spin_unlock(&delayed_refs->lock); /* @@ -1132,7 +1139,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, */ btrfs_update_delayed_refs_rsv(trans); - trace_add_delayed_tree_ref(fs_info, &ref->node, ref, + trace_add_delayed_tree_ref(fs_info, node, ref, action == BTRFS_ADD_DELAYED_EXTENT ? BTRFS_ADD_DELAYED_REF : action); if (merged) @@ -1153,6 +1160,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_delayed_data_ref *ref; + struct btrfs_delayed_ref_node *node; struct btrfs_delayed_ref_head *head_ref; struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_qgroup_extent_record *record = NULL; @@ -1172,12 +1180,14 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, if (!ref) return -ENOMEM; + node = btrfs_delayed_data_ref_to_node(ref); + if (parent) ref_type = BTRFS_SHARED_DATA_REF_KEY; else ref_type = BTRFS_EXTENT_DATA_REF_KEY; - init_delayed_ref_common(fs_info, &ref->node, bytenr, num_bytes, - ref_root, action, ref_type); + init_delayed_ref_common(fs_info, node, bytenr, num_bytes, ref_root, + action, ref_type); ref->root = ref_root; ref->parent = parent; ref->objectid = owner; @@ -1214,7 +1224,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, head_ref = add_delayed_ref_head(trans, head_ref, record, action, &qrecord_inserted); - merged = insert_delayed_ref(trans, head_ref, &ref->node); + merged = insert_delayed_ref(trans, head_ref, node); spin_unlock(&delayed_refs->lock); /* @@ -1223,7 +1233,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, */ btrfs_update_delayed_refs_rsv(trans); - trace_add_delayed_data_ref(trans->fs_info, &ref->node, ref, + trace_add_delayed_data_ref(trans->fs_info, node, ref, action == BTRFS_ADD_DELAYED_EXTENT ? BTRFS_ADD_DELAYED_REF : action); if (merged) diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index b291147cb8ab..b3a78bf7b072 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -413,4 +413,16 @@ btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node) return container_of(node, struct btrfs_delayed_data_ref, node); } +static inline struct btrfs_delayed_ref_node * +btrfs_delayed_tree_ref_to_node(struct btrfs_delayed_tree_ref *ref) +{ + return &ref->node; +} + +static inline struct btrfs_delayed_ref_node * +btrfs_delayed_data_ref_to_node(struct btrfs_delayed_data_ref *ref) +{ + return &ref->node; +} + #endif -- cgit v1.2.3-59-g8ed1b From d3fbb00f5e21c6dfaa6e820a21df0c9a3455a028 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 17:16:40 -0400 Subject: btrfs: embed data_ref and tree_ref in btrfs_delayed_ref_node We have been embedding btrfs_delayed_ref_node in the btrfs_delayed_data_ref and btrfs_delayed_tree_ref, and then we have two sets of cachep's and a variety of handling that is awkward because of this separation. Instead union these two members inside of btrfs_delayed_ref_node and make that the first class object. This allows us to go down to one cachep for our delayed ref nodes instead of two. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 51 +++++++++++++++++--------------------------------- fs/btrfs/delayed-ref.h | 44 ++++++++++++++++++++++--------------------- 2 files changed, 40 insertions(+), 55 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index d920663a18fd..9382f7c81c25 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -16,8 +16,7 @@ #include "fs.h" struct kmem_cache *btrfs_delayed_ref_head_cachep; -struct kmem_cache *btrfs_delayed_tree_ref_cachep; -struct kmem_cache *btrfs_delayed_data_ref_cachep; +struct kmem_cache *btrfs_delayed_ref_node_cachep; struct kmem_cache *btrfs_delayed_extent_op_cachep; /* * delayed back reference update tracking. For subvolume trees @@ -1082,26 +1081,26 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, is_system = (generic_ref->tree_ref.ref_root == BTRFS_CHUNK_TREE_OBJECTID); ASSERT(generic_ref->type == BTRFS_REF_METADATA && generic_ref->action); - ref = kmem_cache_alloc(btrfs_delayed_tree_ref_cachep, GFP_NOFS); - if (!ref) + node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); + if (!node) return -ENOMEM; head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); if (!head_ref) { - kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref); + kmem_cache_free(btrfs_delayed_ref_node_cachep, node); return -ENOMEM; } if (btrfs_qgroup_full_accounting(fs_info) && !generic_ref->skip_qgroup) { record = kzalloc(sizeof(*record), GFP_NOFS); if (!record) { - kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref); + kmem_cache_free(btrfs_delayed_ref_node_cachep, node); kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref); return -ENOMEM; } } - node = btrfs_delayed_tree_ref_to_node(ref); + ref = btrfs_delayed_node_to_tree_ref(node); if (parent) ref_type = BTRFS_SHARED_BLOCK_REF_KEY; @@ -1143,7 +1142,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, action == BTRFS_ADD_DELAYED_EXTENT ? BTRFS_ADD_DELAYED_REF : action); if (merged) - kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref); + kmem_cache_free(btrfs_delayed_ref_node_cachep, node); if (qrecord_inserted) btrfs_qgroup_trace_extent_post(trans, record); @@ -1176,11 +1175,11 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, u8 ref_type; ASSERT(generic_ref->type == BTRFS_REF_DATA && action); - ref = kmem_cache_alloc(btrfs_delayed_data_ref_cachep, GFP_NOFS); - if (!ref) + node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); + if (!node) return -ENOMEM; - node = btrfs_delayed_data_ref_to_node(ref); + ref = btrfs_delayed_node_to_data_ref(node); if (parent) ref_type = BTRFS_SHARED_DATA_REF_KEY; @@ -1196,14 +1195,14 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); if (!head_ref) { - kmem_cache_free(btrfs_delayed_data_ref_cachep, ref); + kmem_cache_free(btrfs_delayed_ref_node_cachep, node); return -ENOMEM; } if (btrfs_qgroup_full_accounting(fs_info) && !generic_ref->skip_qgroup) { record = kzalloc(sizeof(*record), GFP_NOFS); if (!record) { - kmem_cache_free(btrfs_delayed_data_ref_cachep, ref); + kmem_cache_free(btrfs_delayed_ref_node_cachep, node); kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref); return -ENOMEM; @@ -1237,7 +1236,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, action == BTRFS_ADD_DELAYED_EXTENT ? BTRFS_ADD_DELAYED_REF : action); if (merged) - kmem_cache_free(btrfs_delayed_data_ref_cachep, ref); + kmem_cache_free(btrfs_delayed_ref_node_cachep, node); if (qrecord_inserted) @@ -1280,18 +1279,7 @@ void btrfs_put_delayed_ref(struct btrfs_delayed_ref_node *ref) { if (refcount_dec_and_test(&ref->refs)) { WARN_ON(!RB_EMPTY_NODE(&ref->ref_node)); - switch (ref->type) { - case BTRFS_TREE_BLOCK_REF_KEY: - case BTRFS_SHARED_BLOCK_REF_KEY: - kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref); - break; - case BTRFS_EXTENT_DATA_REF_KEY: - case BTRFS_SHARED_DATA_REF_KEY: - kmem_cache_free(btrfs_delayed_data_ref_cachep, ref); - break; - default: - BUG(); - } + kmem_cache_free(btrfs_delayed_ref_node_cachep, ref); } } @@ -1310,8 +1298,7 @@ btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs, u64 byt void __cold btrfs_delayed_ref_exit(void) { kmem_cache_destroy(btrfs_delayed_ref_head_cachep); - kmem_cache_destroy(btrfs_delayed_tree_ref_cachep); - kmem_cache_destroy(btrfs_delayed_data_ref_cachep); + kmem_cache_destroy(btrfs_delayed_ref_node_cachep); kmem_cache_destroy(btrfs_delayed_extent_op_cachep); } @@ -1321,12 +1308,8 @@ int __init btrfs_delayed_ref_init(void) if (!btrfs_delayed_ref_head_cachep) goto fail; - btrfs_delayed_tree_ref_cachep = KMEM_CACHE(btrfs_delayed_tree_ref, 0); - if (!btrfs_delayed_tree_ref_cachep) - goto fail; - - btrfs_delayed_data_ref_cachep = KMEM_CACHE(btrfs_delayed_data_ref, 0); - if (!btrfs_delayed_data_ref_cachep) + btrfs_delayed_ref_node_cachep = KMEM_CACHE(btrfs_delayed_ref_node, 0); + if (!btrfs_delayed_ref_node_cachep) goto fail; btrfs_delayed_extent_op_cachep = KMEM_CACHE(btrfs_delayed_extent_op, 0); diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index b3a78bf7b072..2de447d9aaba 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -30,6 +30,19 @@ enum btrfs_delayed_ref_action { BTRFS_UPDATE_DELAYED_HEAD, } __packed; +struct btrfs_delayed_tree_ref { + u64 root; + u64 parent; + int level; +}; + +struct btrfs_delayed_data_ref { + u64 root; + u64 parent; + u64 objectid; + u64 offset; +}; + struct btrfs_delayed_ref_node { struct rb_node ref_node; /* @@ -64,6 +77,11 @@ struct btrfs_delayed_ref_node { unsigned int action:8; unsigned int type:8; + + union { + struct btrfs_delayed_tree_ref tree_ref; + struct btrfs_delayed_data_ref data_ref; + }; }; struct btrfs_delayed_extent_op { @@ -151,21 +169,6 @@ struct btrfs_delayed_ref_head { bool processing; }; -struct btrfs_delayed_tree_ref { - struct btrfs_delayed_ref_node node; - u64 root; - u64 parent; - int level; -}; - -struct btrfs_delayed_data_ref { - struct btrfs_delayed_ref_node node; - u64 root; - u64 parent; - u64 objectid; - u64 offset; -}; - enum btrfs_delayed_ref_flags { /* Indicate that we are flushing delayed refs for the commit */ BTRFS_DELAYED_REFS_FLUSHING, @@ -279,8 +282,7 @@ struct btrfs_ref { }; extern struct kmem_cache *btrfs_delayed_ref_head_cachep; -extern struct kmem_cache *btrfs_delayed_tree_ref_cachep; -extern struct kmem_cache *btrfs_delayed_data_ref_cachep; +extern struct kmem_cache *btrfs_delayed_ref_node_cachep; extern struct kmem_cache *btrfs_delayed_extent_op_cachep; int __init btrfs_delayed_ref_init(void); @@ -404,25 +406,25 @@ bool btrfs_check_space_for_delayed_refs(struct btrfs_fs_info *fs_info); static inline struct btrfs_delayed_tree_ref * btrfs_delayed_node_to_tree_ref(struct btrfs_delayed_ref_node *node) { - return container_of(node, struct btrfs_delayed_tree_ref, node); + return &node->tree_ref; } static inline struct btrfs_delayed_data_ref * btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node) { - return container_of(node, struct btrfs_delayed_data_ref, node); + return &node->data_ref; } static inline struct btrfs_delayed_ref_node * btrfs_delayed_tree_ref_to_node(struct btrfs_delayed_tree_ref *ref) { - return &ref->node; + return container_of(ref, struct btrfs_delayed_ref_node, tree_ref); } static inline struct btrfs_delayed_ref_node * btrfs_delayed_data_ref_to_node(struct btrfs_delayed_data_ref *ref) { - return &ref->node; + return container_of(ref, struct btrfs_delayed_ref_node, data_ref); } #endif -- cgit v1.2.3-59-g8ed1b From 4d09b4e942bce27c6fa66f3582dc639417107e2d Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 19:17:40 -0400 Subject: btrfs: do not use a function to initialize btrfs_ref btrfs_ref currently has ->owning_root, and ->ref_root is shared between the tree ref and data ref, so in order to move that into btrfs_ref proper I would need to add another root parameter to the initialization function. This function has too many arguments, and adding another root will make it easy to make mistakes about which root goes where. Drop the generic ref init function and statically initialize the btrfs_ref in every usage. This makes the code easier to read because we can see what elements we're assigning, and will make the upcoming change moving the ref_root into the btrfs_ref more clear and less error prone than adding a new element to the initialization function. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 10 ------ fs/btrfs/delayed-ref.h | 2 -- fs/btrfs/extent-tree.c | 87 ++++++++++++++++++++++++++++---------------------- fs/btrfs/file.c | 49 +++++++++++++++++----------- fs/btrfs/inode-item.c | 10 +++--- fs/btrfs/relocation.c | 58 ++++++++++++++++++++++----------- fs/btrfs/tree-log.c | 11 ++++--- 7 files changed, 131 insertions(+), 96 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 9382f7c81c25..1d0795aeba12 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -1007,16 +1007,6 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, INIT_LIST_HEAD(&ref->add_list); } -void btrfs_init_generic_ref(struct btrfs_ref *generic_ref, int action, u64 bytenr, - u64 len, u64 parent, u64 owning_root) -{ - generic_ref->action = action; - generic_ref->bytenr = bytenr; - generic_ref->len = len; - generic_ref->parent = parent; - generic_ref->owning_root = owning_root; -} - void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 root, u64 mod_root, bool skip_qgroup) { diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 2de447d9aaba..b0b2d0e93996 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -320,8 +320,6 @@ static inline u64 btrfs_calc_delayed_ref_csum_bytes(const struct btrfs_fs_info * return btrfs_calc_metadata_size(fs_info, num_csum_items); } -void btrfs_init_generic_ref(struct btrfs_ref *generic_ref, int action, u64 bytenr, - u64 len, u64 parent, u64 owning_root); void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 root, u64 mod_root, bool skip_qgroup); void btrfs_init_data_ref(struct btrfs_ref *generic_ref, u64 ref_root, u64 ino, diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 42525dc8a551..f5e322ac340e 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2492,14 +2492,11 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, int full_backref, int inc) { struct btrfs_fs_info *fs_info = root->fs_info; - u64 bytenr; - u64 num_bytes; u64 parent; u64 ref_root; u32 nritems; struct btrfs_key key; struct btrfs_file_extent_item *fi; - struct btrfs_ref generic_ref = { 0 }; bool for_reloc = btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC); int i; int action; @@ -2526,6 +2523,11 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, action = BTRFS_DROP_DELAYED_REF; for (i = 0; i < nritems; i++) { + struct btrfs_ref ref = { + .action = action, + .parent = parent, + }; + if (level == 0) { btrfs_item_key_to_cpu(buf, &key, i); if (key.type != BTRFS_EXTENT_DATA_KEY) @@ -2535,35 +2537,34 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, if (btrfs_file_extent_type(buf, fi) == BTRFS_FILE_EXTENT_INLINE) continue; - bytenr = btrfs_file_extent_disk_bytenr(buf, fi); - if (bytenr == 0) + ref.bytenr = btrfs_file_extent_disk_bytenr(buf, fi); + if (ref.bytenr == 0) continue; - num_bytes = btrfs_file_extent_disk_num_bytes(buf, fi); + ref.len = btrfs_file_extent_disk_num_bytes(buf, fi); + ref.owning_root = ref_root; + key.offset -= btrfs_file_extent_offset(buf, fi); - btrfs_init_generic_ref(&generic_ref, action, bytenr, - num_bytes, parent, ref_root); - btrfs_init_data_ref(&generic_ref, ref_root, key.objectid, + btrfs_init_data_ref(&ref, ref_root, key.objectid, key.offset, root->root_key.objectid, for_reloc); if (inc) - ret = btrfs_inc_extent_ref(trans, &generic_ref); + ret = btrfs_inc_extent_ref(trans, &ref); else - ret = btrfs_free_extent(trans, &generic_ref); + ret = btrfs_free_extent(trans, &ref); if (ret) goto fail; } else { - bytenr = btrfs_node_blockptr(buf, i); - num_bytes = fs_info->nodesize; - /* We don't know the owning_root, use 0. */ - btrfs_init_generic_ref(&generic_ref, action, bytenr, - num_bytes, parent, 0); - btrfs_init_tree_ref(&generic_ref, level - 1, ref_root, + /* We don't know the owning_root, leave as 0. */ + ref.bytenr = btrfs_node_blockptr(buf, i); + ref.len = fs_info->nodesize; + + btrfs_init_tree_ref(&ref, level - 1, ref_root, root->root_key.objectid, for_reloc); if (inc) - ret = btrfs_inc_extent_ref(trans, &generic_ref); + ret = btrfs_inc_extent_ref(trans, &ref); else - ret = btrfs_free_extent(trans, &generic_ref); + ret = btrfs_free_extent(trans, &ref); if (ret) goto fail; } @@ -3462,7 +3463,13 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, int ret; if (root_id != BTRFS_TREE_LOG_OBJECTID) { - struct btrfs_ref generic_ref = { 0 }; + struct btrfs_ref generic_ref = { + .action = BTRFS_DROP_DELAYED_REF, + .bytenr = buf->start, + .len = buf->len, + .parent = parent, + .owning_root = btrfs_header_owner(buf), + }; /* * Assert that the extent buffer is not cleared due to @@ -3472,9 +3479,6 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, */ ASSERT(btrfs_header_bytenr(buf) != 0); - btrfs_init_generic_ref(&generic_ref, BTRFS_DROP_DELAYED_REF, - buf->start, buf->len, parent, - btrfs_header_owner(buf)); btrfs_init_tree_ref(&generic_ref, btrfs_header_level(buf), root_id, 0, false); btrfs_ref_tree_mod(fs_info, &generic_ref); @@ -4966,17 +4970,19 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans, u64 offset, u64 ram_bytes, struct btrfs_key *ins) { - struct btrfs_ref generic_ref = { 0 }; + struct btrfs_ref generic_ref = { + .action = BTRFS_ADD_DELAYED_EXTENT, + .bytenr = ins->objectid, + .len = ins->offset, + .owning_root = root->root_key.objectid, + }; u64 root_objectid = root->root_key.objectid; - u64 owning_root = root_objectid; ASSERT(root_objectid != BTRFS_TREE_LOG_OBJECTID); if (btrfs_is_data_reloc_root(root) && is_fstree(root->relocation_src_root)) - owning_root = root->relocation_src_root; + generic_ref.owning_root = root->relocation_src_root; - btrfs_init_generic_ref(&generic_ref, BTRFS_ADD_DELAYED_EXTENT, - ins->objectid, ins->offset, 0, owning_root); btrfs_init_data_ref(&generic_ref, root_objectid, owner, offset, 0, false); btrfs_ref_tree_mod(root->fs_info, &generic_ref); @@ -5157,7 +5163,6 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, struct btrfs_block_rsv *block_rsv; struct extent_buffer *buf; struct btrfs_delayed_extent_op *extent_op; - struct btrfs_ref generic_ref = { 0 }; u64 flags = 0; int ret; u32 blocksize = fs_info->nodesize; @@ -5200,6 +5205,13 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, BUG_ON(parent > 0); if (root_objectid != BTRFS_TREE_LOG_OBJECTID) { + struct btrfs_ref generic_ref = { + .action = BTRFS_ADD_DELAYED_EXTENT, + .bytenr = ins.objectid, + .len = ins.offset, + .parent = parent, + .owning_root = owning_root, + }; extent_op = btrfs_alloc_delayed_extent_op(); if (!extent_op) { ret = -ENOMEM; @@ -5214,8 +5226,6 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, extent_op->update_flags = true; extent_op->level = level; - btrfs_init_generic_ref(&generic_ref, BTRFS_ADD_DELAYED_EXTENT, - ins.objectid, ins.offset, parent, owning_root); btrfs_init_tree_ref(&generic_ref, level, root_objectid, root->root_key.objectid, false); btrfs_ref_tree_mod(fs_info, &generic_ref); @@ -5460,11 +5470,9 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info = root->fs_info; u64 bytenr; u64 generation; - u64 parent; u64 owner_root = 0; struct btrfs_tree_parent_check check = { 0 }; struct btrfs_key key; - struct btrfs_ref ref = { 0 }; struct extent_buffer *next; int level = wc->level; int reada = 0; @@ -5581,8 +5589,14 @@ skip: wc->refs[level - 1] = 0; wc->flags[level - 1] = 0; if (wc->stage == DROP_REFERENCE) { + struct btrfs_ref ref = { + .action = BTRFS_DROP_DELAYED_REF, + .bytenr = bytenr, + .len = fs_info->nodesize, + .owning_root = owner_root, + }; if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) { - parent = path->nodes[level]->start; + ref.parent = path->nodes[level]->start; } else { ASSERT(root->root_key.objectid == btrfs_header_owner(path->nodes[level])); @@ -5593,7 +5607,6 @@ skip: ret = -EIO; goto out_unlock; } - parent = 0; } /* @@ -5603,7 +5616,7 @@ skip: * ->restarted flag. */ if (wc->restarted) { - ret = check_ref_exists(trans, root, bytenr, parent, + ret = check_ref_exists(trans, root, bytenr, ref.parent, level - 1); if (ret < 0) goto out_unlock; @@ -5638,8 +5651,6 @@ skip: wc->drop_level = level; find_next_key(path, level, &wc->drop_progress); - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, bytenr, - fs_info->nodesize, parent, owner_root); btrfs_init_tree_ref(&ref, level - 1, root->root_key.objectid, 0, false); ret = btrfs_free_extent(trans, &ref); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 1b972bd05af1..d8ee17df1b02 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -206,7 +206,6 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info = root->fs_info; struct extent_buffer *leaf; struct btrfs_file_extent_item *fi; - struct btrfs_ref ref = { 0 }; struct btrfs_key key; struct btrfs_key new_key; u64 ino = btrfs_ino(inode); @@ -373,10 +372,13 @@ next_slot: btrfs_mark_buffer_dirty(trans, leaf); if (update_refs && disk_bytenr > 0) { - btrfs_init_generic_ref(&ref, - BTRFS_ADD_DELAYED_REF, - disk_bytenr, num_bytes, 0, - root->root_key.objectid); + struct btrfs_ref ref = { + .action = BTRFS_ADD_DELAYED_REF, + .bytenr = disk_bytenr, + .len = num_bytes, + .parent = 0, + .owning_root = root->root_key.objectid, + }; btrfs_init_data_ref(&ref, root->root_key.objectid, new_key.objectid, @@ -464,10 +466,13 @@ delete_extent_item: extent_end = ALIGN(extent_end, fs_info->sectorsize); } else if (update_refs && disk_bytenr > 0) { - btrfs_init_generic_ref(&ref, - BTRFS_DROP_DELAYED_REF, - disk_bytenr, num_bytes, 0, - root->root_key.objectid); + struct btrfs_ref ref = { + .action = BTRFS_DROP_DELAYED_REF, + .bytenr = disk_bytenr, + .len = num_bytes, + .parent = 0, + .owning_root = root->root_key.objectid, + }; btrfs_init_data_ref(&ref, root->root_key.objectid, key.objectid, @@ -748,8 +753,11 @@ again: extent_end - split); btrfs_mark_buffer_dirty(trans, leaf); - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, bytenr, - num_bytes, 0, root->root_key.objectid); + ref.action = BTRFS_ADD_DELAYED_REF; + ref.bytenr = bytenr; + ref.len = num_bytes; + ref.parent = 0; + ref.owning_root = root->root_key.objectid; btrfs_init_data_ref(&ref, root->root_key.objectid, ino, orig_offset, 0, false); ret = btrfs_inc_extent_ref(trans, &ref); @@ -774,8 +782,12 @@ again: other_start = end; other_end = 0; - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, bytenr, - num_bytes, 0, root->root_key.objectid); + + ref.action = BTRFS_DROP_DELAYED_REF; + ref.bytenr = bytenr; + ref.len = num_bytes; + ref.parent = 0; + ref.owning_root = root->root_key.objectid; btrfs_init_data_ref(&ref, root->root_key.objectid, ino, orig_offset, 0, false); if (extent_mergeable(leaf, path->slots[0] + 1, @@ -2424,7 +2436,6 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, struct extent_buffer *leaf; struct btrfs_key key; int slot; - struct btrfs_ref ref = { 0 }; int ret; if (replace_len == 0) @@ -2480,12 +2491,14 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, extent_info->qgroup_reserved, &key); } else { + struct btrfs_ref ref = { + .action = BTRFS_ADD_DELAYED_REF, + .bytenr = extent_info->disk_offset, + .len = extent_info->disk_len, + .owning_root = root->root_key.objectid, + }; u64 ref_offset; - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, - extent_info->disk_offset, - extent_info->disk_len, 0, - root->root_key.objectid); ref_offset = extent_info->file_offset - extent_info->data_offset; btrfs_init_data_ref(&ref, root->root_key.objectid, btrfs_ino(inode), ref_offset, 0, false); diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c index 9c1394c0a6d7..d61bb65859a5 100644 --- a/fs/btrfs/inode-item.c +++ b/fs/btrfs/inode-item.c @@ -670,13 +670,15 @@ delete: } if (del_item && extent_start != 0 && !control->skip_ref_updates) { - struct btrfs_ref ref = { 0 }; + struct btrfs_ref ref = { + .action = BTRFS_DROP_DELAYED_REF, + .bytenr = extent_start, + .len = extent_num_bytes, + .owning_root = root->root_key.objectid, + }; bytes_deleted += extent_num_bytes; - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, - extent_start, extent_num_bytes, 0, - root->root_key.objectid); btrfs_init_data_ref(&ref, btrfs_header_owner(leaf), control->ino, extent_offset, root->root_key.objectid, false); diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 5a01aaa164de..fdcecce85f91 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1104,8 +1104,11 @@ int replace_file_extents(struct btrfs_trans_handle *trans, dirty = 1; key.offset -= btrfs_file_extent_offset(leaf, fi); - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, new_bytenr, - num_bytes, parent, root->root_key.objectid); + ref.action = BTRFS_ADD_DELAYED_REF; + ref.bytenr = new_bytenr; + ref.len = num_bytes; + ref.parent = parent; + ref.owning_root = root->root_key.objectid; btrfs_init_data_ref(&ref, btrfs_header_owner(leaf), key.objectid, key.offset, root->root_key.objectid, false); @@ -1115,8 +1118,11 @@ int replace_file_extents(struct btrfs_trans_handle *trans, break; } - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, bytenr, - num_bytes, parent, root->root_key.objectid); + ref.action = BTRFS_DROP_DELAYED_REF; + ref.bytenr = bytenr; + ref.len = num_bytes; + ref.parent = parent; + ref.owning_root = root->root_key.objectid; btrfs_init_data_ref(&ref, btrfs_header_owner(leaf), key.objectid, key.offset, root->root_key.objectid, false); @@ -1328,9 +1334,11 @@ again: path->slots[level], old_ptr_gen); btrfs_mark_buffer_dirty(trans, path->nodes[level]); - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, old_bytenr, - blocksize, path->nodes[level]->start, - src->root_key.objectid); + ref.action = BTRFS_ADD_DELAYED_REF; + ref.bytenr = old_bytenr; + ref.len = blocksize; + ref.parent = path->nodes[level]->start; + ref.owning_root = src->root_key.objectid; btrfs_init_tree_ref(&ref, level - 1, src->root_key.objectid, 0, true); ret = btrfs_inc_extent_ref(trans, &ref); @@ -1338,8 +1346,12 @@ again: btrfs_abort_transaction(trans, ret); break; } - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, new_bytenr, - blocksize, 0, dest->root_key.objectid); + + ref.action = BTRFS_ADD_DELAYED_REF; + ref.bytenr = new_bytenr; + ref.len = blocksize; + ref.parent = 0; + ref.owning_root = dest->root_key.objectid; btrfs_init_tree_ref(&ref, level - 1, dest->root_key.objectid, 0, true); ret = btrfs_inc_extent_ref(trans, &ref); @@ -1349,8 +1361,11 @@ again: } /* We don't know the real owning_root, use 0. */ - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, new_bytenr, - blocksize, path->nodes[level]->start, 0); + ref.action = BTRFS_DROP_DELAYED_REF; + ref.bytenr = new_bytenr; + ref.len = blocksize; + ref.parent = path->nodes[level]->start; + ref.owning_root = 0; btrfs_init_tree_ref(&ref, level - 1, src->root_key.objectid, 0, true); ret = btrfs_free_extent(trans, &ref); @@ -1360,8 +1375,11 @@ again: } /* We don't know the real owning_root, use 0. */ - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, old_bytenr, - blocksize, 0, 0); + ref.action = BTRFS_DROP_DELAYED_REF; + ref.bytenr = old_bytenr; + ref.len = blocksize; + ref.parent = 0; + ref.owning_root = 0; btrfs_init_tree_ref(&ref, level - 1, dest->root_key.objectid, 0, true); ret = btrfs_free_extent(trans, &ref); @@ -2374,8 +2392,6 @@ static int do_relocation(struct btrfs_trans_handle *trans, path->lowest_level = node->level + 1; rc->backref_cache.path[node->level] = node; list_for_each_entry(edge, &node->upper, list[LOWER]) { - struct btrfs_ref ref = { 0 }; - cond_resched(); upper = edge->node[UPPER]; @@ -2463,16 +2479,20 @@ static int do_relocation(struct btrfs_trans_handle *trans, */ ASSERT(node->eb == eb); } else { + struct btrfs_ref ref = { + .action = BTRFS_ADD_DELAYED_REF, + .bytenr = node->eb->start, + .len = blocksize, + .parent = upper->eb->start, + .owning_root = btrfs_header_owner(upper->eb), + }; + btrfs_set_node_blockptr(upper->eb, slot, node->eb->start); btrfs_set_node_ptr_generation(upper->eb, slot, trans->transid); btrfs_mark_buffer_dirty(trans, upper->eb); - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, - node->eb->start, blocksize, - upper->eb->start, - btrfs_header_owner(upper->eb)); btrfs_init_tree_ref(&ref, node->level, btrfs_header_owner(upper->eb), root->root_key.objectid, false); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 201141eb2a8b..f85cdabf69c7 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -748,7 +748,6 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, goto out; if (ins.objectid > 0) { - struct btrfs_ref ref = { 0 }; u64 csum_start; u64 csum_end; LIST_HEAD(ordered_sums); @@ -762,10 +761,12 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, if (ret < 0) { goto out; } else if (ret == 0) { - btrfs_init_generic_ref(&ref, - BTRFS_ADD_DELAYED_REF, - ins.objectid, ins.offset, 0, - root->root_key.objectid); + struct btrfs_ref ref = { + .action = BTRFS_ADD_DELAYED_REF, + .bytenr = ins.objectid, + .len = ins.offset, + .owning_root = root->root_key.objectid, + }; btrfs_init_data_ref(&ref, root->root_key.objectid, key->objectid, offset, 0, false); -- cgit v1.2.3-59-g8ed1b From f2e69a77aa5150d8a8ea13e45aad6dc42e581591 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 19:37:53 -0400 Subject: btrfs: move ref_root into btrfs_ref We have this in both btrfs_tree_ref and btrfs_data_ref, which is just wasting space and making the code more complicated. Move this into btrfs_ref proper and update all the call sites to do the assignment in btrfs_ref. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 29 +++++++++++++---------------- fs/btrfs/delayed-ref.h | 22 +++++++++------------- fs/btrfs/extent-tree.c | 38 ++++++++++++++++---------------------- fs/btrfs/file.c | 30 ++++++++++++++---------------- fs/btrfs/inode-item.c | 6 +++--- fs/btrfs/ref-verify.c | 4 ++-- fs/btrfs/relocation.c | 26 +++++++++++++------------- fs/btrfs/tree-log.c | 6 +++--- 8 files changed, 73 insertions(+), 88 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 1d0795aeba12..c6a1b6938654 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -1007,17 +1007,16 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, INIT_LIST_HEAD(&ref->add_list); } -void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 root, - u64 mod_root, bool skip_qgroup) +void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 mod_root, + bool skip_qgroup) { #ifdef CONFIG_BTRFS_FS_REF_VERIFY /* If @real_root not set, use @root as fallback */ - generic_ref->real_root = mod_root ?: root; + generic_ref->real_root = mod_root ?: generic_ref->ref_root; #endif generic_ref->tree_ref.level = level; - generic_ref->tree_ref.ref_root = root; generic_ref->type = BTRFS_REF_METADATA; - if (skip_qgroup || !(is_fstree(root) && + if (skip_qgroup || !(is_fstree(generic_ref->ref_root) && (!mod_root || is_fstree(mod_root)))) generic_ref->skip_qgroup = true; else @@ -1025,18 +1024,17 @@ void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 root, } -void btrfs_init_data_ref(struct btrfs_ref *generic_ref, u64 ref_root, u64 ino, - u64 offset, u64 mod_root, bool skip_qgroup) +void btrfs_init_data_ref(struct btrfs_ref *generic_ref, u64 ino, u64 offset, + u64 mod_root, bool skip_qgroup) { #ifdef CONFIG_BTRFS_FS_REF_VERIFY /* If @real_root not set, use @root as fallback */ - generic_ref->real_root = mod_root ?: ref_root; + generic_ref->real_root = mod_root ?: generic_ref->ref_root; #endif - generic_ref->data_ref.ref_root = ref_root; generic_ref->data_ref.ino = ino; generic_ref->data_ref.offset = offset; generic_ref->type = BTRFS_REF_DATA; - if (skip_qgroup || !(is_fstree(ref_root) && + if (skip_qgroup || !(is_fstree(generic_ref->ref_root) && (!mod_root || is_fstree(mod_root)))) generic_ref->skip_qgroup = true; else @@ -1068,7 +1066,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, u64 parent = generic_ref->parent; u8 ref_type; - is_system = (generic_ref->tree_ref.ref_root == BTRFS_CHUNK_TREE_OBJECTID); + is_system = (generic_ref->ref_root == BTRFS_CHUNK_TREE_OBJECTID); ASSERT(generic_ref->type == BTRFS_REF_METADATA && generic_ref->action); node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); @@ -1098,14 +1096,13 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, ref_type = BTRFS_TREE_BLOCK_REF_KEY; init_delayed_ref_common(fs_info, node, bytenr, num_bytes, - generic_ref->tree_ref.ref_root, action, - ref_type); - ref->root = generic_ref->tree_ref.ref_root; + generic_ref->ref_root, action, ref_type); + ref->root = generic_ref->ref_root; ref->parent = parent; ref->level = level; init_delayed_ref_head(head_ref, record, bytenr, num_bytes, - generic_ref->tree_ref.ref_root, 0, action, + generic_ref->ref_root, 0, action, false, is_system, generic_ref->owning_root); head_ref->extent_op = extent_op; @@ -1159,7 +1156,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, u64 bytenr = generic_ref->bytenr; u64 num_bytes = generic_ref->len; u64 parent = generic_ref->parent; - u64 ref_root = generic_ref->data_ref.ref_root; + u64 ref_root = generic_ref->ref_root; u64 owner = generic_ref->data_ref.ino; u64 offset = generic_ref->data_ref.offset; u8 ref_type; diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index b0b2d0e93996..bf2916906bb8 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -220,9 +220,6 @@ enum btrfs_ref_type { struct btrfs_data_ref { /* For EXTENT_DATA_REF */ - /* Root which owns this data reference. */ - u64 ref_root; - /* Inode which refers to this data extent */ u64 ino; @@ -243,13 +240,6 @@ struct btrfs_tree_ref { */ int level; - /* - * Root which owns this tree block reference. - * - * For TREE_BLOCK_REF (skinny metadata, either inline or keyed) - */ - u64 ref_root; - /* For non-skinny metadata, no special member needed */ }; @@ -273,6 +263,12 @@ struct btrfs_ref { u64 len; u64 owning_root; + /* + * The root that owns the reference for this reference, this will be set + * or ->parent will be set, depending on what type of reference this is. + */ + u64 ref_root; + /* Bytenr of the parent tree block */ u64 parent; union { @@ -320,10 +316,10 @@ static inline u64 btrfs_calc_delayed_ref_csum_bytes(const struct btrfs_fs_info * return btrfs_calc_metadata_size(fs_info, num_csum_items); } -void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 root, +void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 mod_root, + bool skip_qgroup); +void btrfs_init_data_ref(struct btrfs_ref *generic_ref, u64 ino, u64 offset, u64 mod_root, bool skip_qgroup); -void btrfs_init_data_ref(struct btrfs_ref *generic_ref, u64 ref_root, u64 ino, - u64 offset, u64 mod_root, bool skip_qgroup); static inline struct btrfs_delayed_extent_op * btrfs_alloc_delayed_extent_op(void) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index f5e322ac340e..9a5e553707b7 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1439,7 +1439,7 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, ASSERT(generic_ref->type != BTRFS_REF_NOT_SET && generic_ref->action); BUG_ON(generic_ref->type == BTRFS_REF_METADATA && - generic_ref->tree_ref.ref_root == BTRFS_TREE_LOG_OBJECTID); + generic_ref->ref_root == BTRFS_TREE_LOG_OBJECTID); if (generic_ref->type == BTRFS_REF_METADATA) ret = btrfs_add_delayed_tree_ref(trans, generic_ref, NULL); @@ -2526,6 +2526,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, struct btrfs_ref ref = { .action = action, .parent = parent, + .ref_root = ref_root, }; if (level == 0) { @@ -2545,9 +2546,8 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, ref.owning_root = ref_root; key.offset -= btrfs_file_extent_offset(buf, fi); - btrfs_init_data_ref(&ref, ref_root, key.objectid, - key.offset, root->root_key.objectid, - for_reloc); + btrfs_init_data_ref(&ref, key.objectid, key.offset, + root->root_key.objectid, for_reloc); if (inc) ret = btrfs_inc_extent_ref(trans, &ref); else @@ -2559,7 +2559,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, ref.bytenr = btrfs_node_blockptr(buf, i); ref.len = fs_info->nodesize; - btrfs_init_tree_ref(&ref, level - 1, ref_root, + btrfs_init_tree_ref(&ref, level - 1, root->root_key.objectid, for_reloc); if (inc) ret = btrfs_inc_extent_ref(trans, &ref); @@ -3469,6 +3469,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, .len = buf->len, .parent = parent, .owning_root = btrfs_header_owner(buf), + .ref_root = root_id, }; /* @@ -3479,8 +3480,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, */ ASSERT(btrfs_header_bytenr(buf) != 0); - btrfs_init_tree_ref(&generic_ref, btrfs_header_level(buf), - root_id, 0, false); + btrfs_init_tree_ref(&generic_ref, btrfs_header_level(buf), 0, false); btrfs_ref_tree_mod(fs_info, &generic_ref); ret = btrfs_add_delayed_tree_ref(trans, &generic_ref, NULL); BUG_ON(ret); /* -ENOMEM */ @@ -3559,10 +3559,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref) * tree log blocks never actually go into the extent allocation * tree, just update pinning info and exit early. */ - if ((ref->type == BTRFS_REF_METADATA && - ref->tree_ref.ref_root == BTRFS_TREE_LOG_OBJECTID) || - (ref->type == BTRFS_REF_DATA && - ref->data_ref.ref_root == BTRFS_TREE_LOG_OBJECTID)) { + if (ref->ref_root == BTRFS_TREE_LOG_OBJECTID) { btrfs_pin_extent(trans, ref->bytenr, ref->len, 1); ret = 0; } else if (ref->type == BTRFS_REF_METADATA) { @@ -3571,10 +3568,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref) ret = btrfs_add_delayed_data_ref(trans, ref, 0); } - if (!((ref->type == BTRFS_REF_METADATA && - ref->tree_ref.ref_root == BTRFS_TREE_LOG_OBJECTID) || - (ref->type == BTRFS_REF_DATA && - ref->data_ref.ref_root == BTRFS_TREE_LOG_OBJECTID))) + if (ref->ref_root != BTRFS_TREE_LOG_OBJECTID) btrfs_ref_tree_mod(fs_info, ref); return ret; @@ -4975,16 +4969,15 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans, .bytenr = ins->objectid, .len = ins->offset, .owning_root = root->root_key.objectid, + .ref_root = root->root_key.objectid, }; - u64 root_objectid = root->root_key.objectid; - ASSERT(root_objectid != BTRFS_TREE_LOG_OBJECTID); + ASSERT(generic_ref.ref_root != BTRFS_TREE_LOG_OBJECTID); if (btrfs_is_data_reloc_root(root) && is_fstree(root->relocation_src_root)) generic_ref.owning_root = root->relocation_src_root; - btrfs_init_data_ref(&generic_ref, root_objectid, owner, - offset, 0, false); + btrfs_init_data_ref(&generic_ref, owner, offset, 0, false); btrfs_ref_tree_mod(root->fs_info, &generic_ref); return btrfs_add_delayed_data_ref(trans, &generic_ref, ram_bytes); @@ -5211,6 +5204,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, .len = ins.offset, .parent = parent, .owning_root = owning_root, + .ref_root = root_objectid, }; extent_op = btrfs_alloc_delayed_extent_op(); if (!extent_op) { @@ -5226,7 +5220,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, extent_op->update_flags = true; extent_op->level = level; - btrfs_init_tree_ref(&generic_ref, level, root_objectid, + btrfs_init_tree_ref(&generic_ref, level, root->root_key.objectid, false); btrfs_ref_tree_mod(fs_info, &generic_ref); ret = btrfs_add_delayed_tree_ref(trans, &generic_ref, extent_op); @@ -5594,6 +5588,7 @@ skip: .bytenr = bytenr, .len = fs_info->nodesize, .owning_root = owner_root, + .ref_root = root->root_key.objectid, }; if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) { ref.parent = path->nodes[level]->start; @@ -5651,8 +5646,7 @@ skip: wc->drop_level = level; find_next_key(path, level, &wc->drop_progress); - btrfs_init_tree_ref(&ref, level - 1, root->root_key.objectid, - 0, false); + btrfs_init_tree_ref(&ref, level - 1, 0, false); ret = btrfs_free_extent(trans, &ref); if (ret) goto out_unlock; diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index d8ee17df1b02..f17170eb0700 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -378,12 +378,11 @@ next_slot: .len = num_bytes, .parent = 0, .owning_root = root->root_key.objectid, + .ref_root = root->root_key.objectid, }; - btrfs_init_data_ref(&ref, - root->root_key.objectid, - new_key.objectid, - args->start - extent_offset, - 0, false); + btrfs_init_data_ref(&ref, new_key.objectid, + args->start - extent_offset, + 0, false); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -472,12 +471,11 @@ delete_extent_item: .len = num_bytes, .parent = 0, .owning_root = root->root_key.objectid, + .ref_root = root->root_key.objectid, }; - btrfs_init_data_ref(&ref, - root->root_key.objectid, - key.objectid, - key.offset - extent_offset, 0, - false); + btrfs_init_data_ref(&ref, key.objectid, + key.offset - extent_offset, + 0, false); ret = btrfs_free_extent(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -758,8 +756,8 @@ again: ref.len = num_bytes; ref.parent = 0; ref.owning_root = root->root_key.objectid; - btrfs_init_data_ref(&ref, root->root_key.objectid, ino, - orig_offset, 0, false); + ref.ref_root = root->root_key.objectid; + btrfs_init_data_ref(&ref, ino, orig_offset, 0, false); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -788,8 +786,8 @@ again: ref.len = num_bytes; ref.parent = 0; ref.owning_root = root->root_key.objectid; - btrfs_init_data_ref(&ref, root->root_key.objectid, ino, orig_offset, - 0, false); + ref.ref_root = root->root_key.objectid; + btrfs_init_data_ref(&ref, ino, orig_offset, 0, false); if (extent_mergeable(leaf, path->slots[0] + 1, ino, bytenr, orig_offset, &other_start, &other_end)) { @@ -2496,12 +2494,12 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, .bytenr = extent_info->disk_offset, .len = extent_info->disk_len, .owning_root = root->root_key.objectid, + .ref_root = root->root_key.objectid, }; u64 ref_offset; ref_offset = extent_info->file_offset - extent_info->data_offset; - btrfs_init_data_ref(&ref, root->root_key.objectid, - btrfs_ino(inode), ref_offset, 0, false); + btrfs_init_data_ref(&ref, btrfs_ino(inode), ref_offset, 0, false); ret = btrfs_inc_extent_ref(trans, &ref); } diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c index d61bb65859a5..e24605df35bb 100644 --- a/fs/btrfs/inode-item.c +++ b/fs/btrfs/inode-item.c @@ -675,13 +675,13 @@ delete: .bytenr = extent_start, .len = extent_num_bytes, .owning_root = root->root_key.objectid, + .ref_root = btrfs_header_owner(leaf), }; bytes_deleted += extent_num_bytes; - btrfs_init_data_ref(&ref, btrfs_header_owner(leaf), - control->ino, extent_offset, - root->root_key.objectid, false); + btrfs_init_data_ref(&ref, control->ino, extent_offset, + root->root_key.objectid, false); ret = btrfs_free_extent(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c index 8c4fc98ca9ce..1108be7a050c 100644 --- a/fs/btrfs/ref-verify.c +++ b/fs/btrfs/ref-verify.c @@ -684,10 +684,10 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info, if (generic_ref->type == BTRFS_REF_METADATA) { if (!parent) - ref_root = generic_ref->tree_ref.ref_root; + ref_root = generic_ref->ref_root; owner = generic_ref->tree_ref.level; } else if (!parent) { - ref_root = generic_ref->data_ref.ref_root; + ref_root = generic_ref->ref_root; owner = generic_ref->data_ref.ino; offset = generic_ref->data_ref.offset; } diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index fdcecce85f91..085b2b616569 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1109,8 +1109,8 @@ int replace_file_extents(struct btrfs_trans_handle *trans, ref.len = num_bytes; ref.parent = parent; ref.owning_root = root->root_key.objectid; - btrfs_init_data_ref(&ref, btrfs_header_owner(leaf), - key.objectid, key.offset, + ref.ref_root = btrfs_header_owner(leaf); + btrfs_init_data_ref(&ref, key.objectid, key.offset, root->root_key.objectid, false); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { @@ -1123,8 +1123,8 @@ int replace_file_extents(struct btrfs_trans_handle *trans, ref.len = num_bytes; ref.parent = parent; ref.owning_root = root->root_key.objectid; - btrfs_init_data_ref(&ref, btrfs_header_owner(leaf), - key.objectid, key.offset, + ref.ref_root = btrfs_header_owner(leaf); + btrfs_init_data_ref(&ref, key.objectid, key.offset, root->root_key.objectid, false); ret = btrfs_free_extent(trans, &ref); if (ret) { @@ -1339,8 +1339,8 @@ again: ref.len = blocksize; ref.parent = path->nodes[level]->start; ref.owning_root = src->root_key.objectid; - btrfs_init_tree_ref(&ref, level - 1, src->root_key.objectid, - 0, true); + ref.ref_root = src->root_key.objectid; + btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -1352,8 +1352,8 @@ again: ref.len = blocksize; ref.parent = 0; ref.owning_root = dest->root_key.objectid; - btrfs_init_tree_ref(&ref, level - 1, dest->root_key.objectid, 0, - true); + ref.ref_root = dest->root_key.objectid; + btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -1366,8 +1366,8 @@ again: ref.len = blocksize; ref.parent = path->nodes[level]->start; ref.owning_root = 0; - btrfs_init_tree_ref(&ref, level - 1, src->root_key.objectid, - 0, true); + ref.ref_root = src->root_key.objectid; + btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_free_extent(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -1380,8 +1380,8 @@ again: ref.len = blocksize; ref.parent = 0; ref.owning_root = 0; - btrfs_init_tree_ref(&ref, level - 1, dest->root_key.objectid, - 0, true); + ref.ref_root = dest->root_key.objectid; + btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_free_extent(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -2485,6 +2485,7 @@ static int do_relocation(struct btrfs_trans_handle *trans, .len = blocksize, .parent = upper->eb->start, .owning_root = btrfs_header_owner(upper->eb), + .ref_root = btrfs_header_owner(upper->eb), }; btrfs_set_node_blockptr(upper->eb, slot, @@ -2494,7 +2495,6 @@ static int do_relocation(struct btrfs_trans_handle *trans, btrfs_mark_buffer_dirty(trans, upper->eb); btrfs_init_tree_ref(&ref, node->level, - btrfs_header_owner(upper->eb), root->root_key.objectid, false); ret = btrfs_inc_extent_ref(trans, &ref); if (!ret) diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index f85cdabf69c7..88f2bbd12170 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -766,10 +766,10 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, .bytenr = ins.objectid, .len = ins.offset, .owning_root = root->root_key.objectid, + .ref_root = root->root_key.objectid, }; - btrfs_init_data_ref(&ref, - root->root_key.objectid, - key->objectid, offset, 0, false); + btrfs_init_data_ref(&ref, key->objectid, offset, + 0, false); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) goto out; -- cgit v1.2.3-59-g8ed1b From da3c5485419767d0943480b7e30066691966ca9a Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 19:44:55 -0400 Subject: btrfs: pass btrfs_ref to init_delayed_ref_common We're extracting all of these values from the btrfs_ref we passed in already, just pass the btrfs_ref through to init_delayed_ref_common and get the values directly from the struct. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 29 ++++++++--------------------- fs/btrfs/delayed-ref.h | 19 +++++++++++++++++++ 2 files changed, 27 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index c6a1b6938654..f5e4a64283e4 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -985,24 +985,24 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans, */ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, struct btrfs_delayed_ref_node *ref, - u64 bytenr, u64 num_bytes, u64 ref_root, - int action, u8 ref_type) + struct btrfs_ref *generic_ref) { + int action = generic_ref->action; u64 seq = 0; if (action == BTRFS_ADD_DELAYED_EXTENT) action = BTRFS_ADD_DELAYED_REF; - if (is_fstree(ref_root)) + if (is_fstree(generic_ref->ref_root)) seq = atomic64_read(&fs_info->tree_mod_seq); refcount_set(&ref->refs, 1); - ref->bytenr = bytenr; - ref->num_bytes = num_bytes; + ref->bytenr = generic_ref->bytenr; + ref->num_bytes = generic_ref->len; ref->ref_mod = 1; ref->action = action; ref->seq = seq; - ref->type = ref_type; + ref->type = btrfs_ref_type(generic_ref); RB_CLEAR_NODE(&ref->ref_node); INIT_LIST_HEAD(&ref->add_list); } @@ -1064,7 +1064,6 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, u64 bytenr = generic_ref->bytenr; u64 num_bytes = generic_ref->len; u64 parent = generic_ref->parent; - u8 ref_type; is_system = (generic_ref->ref_root == BTRFS_CHUNK_TREE_OBJECTID); @@ -1090,13 +1089,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, ref = btrfs_delayed_node_to_tree_ref(node); - if (parent) - ref_type = BTRFS_SHARED_BLOCK_REF_KEY; - else - ref_type = BTRFS_TREE_BLOCK_REF_KEY; - - init_delayed_ref_common(fs_info, node, bytenr, num_bytes, - generic_ref->ref_root, action, ref_type); + init_delayed_ref_common(fs_info, node, generic_ref); ref->root = generic_ref->ref_root; ref->parent = parent; ref->level = level; @@ -1159,7 +1152,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, u64 ref_root = generic_ref->ref_root; u64 owner = generic_ref->data_ref.ino; u64 offset = generic_ref->data_ref.offset; - u8 ref_type; ASSERT(generic_ref->type == BTRFS_REF_DATA && action); node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); @@ -1168,12 +1160,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, ref = btrfs_delayed_node_to_data_ref(node); - if (parent) - ref_type = BTRFS_SHARED_DATA_REF_KEY; - else - ref_type = BTRFS_EXTENT_DATA_REF_KEY; - init_delayed_ref_common(fs_info, node, bytenr, num_bytes, ref_root, - action, ref_type); + init_delayed_ref_common(fs_info, node, generic_ref); ref->root = ref_root; ref->parent = parent; ref->objectid = owner; diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index bf2916906bb8..167c4407c3c4 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -421,4 +421,23 @@ btrfs_delayed_data_ref_to_node(struct btrfs_delayed_data_ref *ref) return container_of(ref, struct btrfs_delayed_ref_node, data_ref); } +static inline u8 btrfs_ref_type(struct btrfs_ref *ref) +{ + ASSERT(ref->type == BTRFS_REF_DATA || ref->type == BTRFS_REF_METADATA); + + if (ref->type == BTRFS_REF_DATA) { + if (ref->parent) + return BTRFS_SHARED_DATA_REF_KEY; + else + return BTRFS_EXTENT_DATA_REF_KEY; + } else { + if (ref->parent) + return BTRFS_SHARED_BLOCK_REF_KEY; + else + return BTRFS_TREE_BLOCK_REF_KEY; + } + + return 0; +} + #endif -- cgit v1.2.3-59-g8ed1b From 0509cc56619d63e946f1212d0f40767728e18fbc Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 20:03:09 -0400 Subject: btrfs: initialize btrfs_delayed_ref_head with btrfs_ref We are calling init_delayed_ref_head with all of the elements from btrfs_ref, clean this up to simply pass in the btrfs_ref and initialize the btrfs_delayed_ref_head with the values from the btrfs_ref directly. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 53 ++++++++++++++++++++++++-------------------------- 1 file changed, 25 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index f5e4a64283e4..5ff6c109e5bf 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -831,18 +831,20 @@ static noinline void update_existing_head_ref(struct btrfs_trans_handle *trans, } static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref, + struct btrfs_ref *generic_ref, struct btrfs_qgroup_extent_record *qrecord, - u64 bytenr, u64 num_bytes, u64 ref_root, - u64 reserved, int action, bool is_data, - bool is_system, u64 owning_root) + u64 reserved) { int count_mod = 1; bool must_insert_reserved = false; /* If reserved is provided, it must be a data extent. */ - BUG_ON(!is_data && reserved); + BUG_ON(generic_ref->type != BTRFS_REF_DATA && reserved); - switch (action) { + switch (generic_ref->action) { + case BTRFS_ADD_DELAYED_REF: + /* count_mod is already set to 1. */ + break; case BTRFS_UPDATE_DELAYED_HEAD: count_mod = 0; break; @@ -871,14 +873,14 @@ static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref, } refcount_set(&head_ref->refs, 1); - head_ref->bytenr = bytenr; - head_ref->num_bytes = num_bytes; + head_ref->bytenr = generic_ref->bytenr; + head_ref->num_bytes = generic_ref->len; head_ref->ref_mod = count_mod; head_ref->reserved_bytes = reserved; head_ref->must_insert_reserved = must_insert_reserved; - head_ref->owning_root = owning_root; - head_ref->is_data = is_data; - head_ref->is_system = is_system; + head_ref->owning_root = generic_ref->owning_root; + head_ref->is_data = (generic_ref->type == BTRFS_REF_DATA); + head_ref->is_system = (generic_ref->ref_root == BTRFS_CHUNK_TREE_OBJECTID); head_ref->ref_tree = RB_ROOT_CACHED; INIT_LIST_HEAD(&head_ref->ref_add_list); RB_CLEAR_NODE(&head_ref->href_node); @@ -888,12 +890,12 @@ static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref, mutex_init(&head_ref->mutex); if (qrecord) { - if (ref_root && reserved) { + if (generic_ref->ref_root && reserved) { qrecord->data_rsv = reserved; - qrecord->data_rsv_refroot = ref_root; + qrecord->data_rsv_refroot = generic_ref->ref_root; } - qrecord->bytenr = bytenr; - qrecord->num_bytes = num_bytes; + qrecord->bytenr = generic_ref->bytenr; + qrecord->num_bytes = generic_ref->len; qrecord->old_roots = NULL; } } @@ -1057,16 +1059,11 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_qgroup_extent_record *record = NULL; bool qrecord_inserted; - bool is_system; bool merged; int action = generic_ref->action; int level = generic_ref->tree_ref.level; - u64 bytenr = generic_ref->bytenr; - u64 num_bytes = generic_ref->len; u64 parent = generic_ref->parent; - is_system = (generic_ref->ref_root == BTRFS_CHUNK_TREE_OBJECTID); - ASSERT(generic_ref->type == BTRFS_REF_METADATA && generic_ref->action); node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); if (!node) @@ -1094,9 +1091,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, ref->parent = parent; ref->level = level; - init_delayed_ref_head(head_ref, record, bytenr, num_bytes, - generic_ref->ref_root, 0, action, - false, is_system, generic_ref->owning_root); + init_delayed_ref_head(head_ref, generic_ref, record, 0); head_ref->extent_op = extent_op; delayed_refs = &trans->transaction->delayed_refs; @@ -1146,8 +1141,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, bool qrecord_inserted; int action = generic_ref->action; bool merged; - u64 bytenr = generic_ref->bytenr; - u64 num_bytes = generic_ref->len; u64 parent = generic_ref->parent; u64 ref_root = generic_ref->ref_root; u64 owner = generic_ref->data_ref.ino; @@ -1183,8 +1176,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, } } - init_delayed_ref_head(head_ref, record, bytenr, num_bytes, ref_root, - reserved, action, true, false, generic_ref->owning_root); + init_delayed_ref_head(head_ref, generic_ref, record, reserved); head_ref->extent_op = NULL; delayed_refs = &trans->transaction->delayed_refs; @@ -1224,13 +1216,18 @@ int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans, { struct btrfs_delayed_ref_head *head_ref; struct btrfs_delayed_ref_root *delayed_refs; + struct btrfs_ref generic_ref = { + .type = BTRFS_REF_METADATA, + .action = BTRFS_UPDATE_DELAYED_HEAD, + .bytenr = bytenr, + .len = num_bytes, + }; head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); if (!head_ref) return -ENOMEM; - init_delayed_ref_head(head_ref, NULL, bytenr, num_bytes, 0, 0, - BTRFS_UPDATE_DELAYED_HEAD, false, false, 0); + init_delayed_ref_head(head_ref, &generic_ref, NULL, 0); head_ref->extent_op = extent_op; delayed_refs = &trans->transaction->delayed_refs; -- cgit v1.2.3-59-g8ed1b From 0ea4703cc27e4e0f4fd8a9403ecae21a8879e7e0 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 20:09:45 -0400 Subject: btrfs: move ref specific initialization into init_delayed_ref_common Now that the btrfs_delayed_ref_node contains a union of the data and metadata specific information we can move the initialization into init_delayed_ref_common and just use the btrfs_ref to initialize the correct fields of the reference. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 5ff6c109e5bf..743cc52c30af 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -1007,6 +1007,17 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, ref->type = btrfs_ref_type(generic_ref); RB_CLEAR_NODE(&ref->ref_node); INIT_LIST_HEAD(&ref->add_list); + + if (generic_ref->type == BTRFS_REF_DATA) { + ref->data_ref.root = generic_ref->ref_root; + ref->data_ref.parent = generic_ref->parent; + ref->data_ref.objectid = generic_ref->data_ref.ino; + ref->data_ref.offset = generic_ref->data_ref.offset; + } else { + ref->tree_ref.root = generic_ref->ref_root; + ref->tree_ref.parent = generic_ref->parent; + ref->tree_ref.level = generic_ref->tree_ref.level; + } } void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 mod_root, @@ -1061,8 +1072,6 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, bool qrecord_inserted; bool merged; int action = generic_ref->action; - int level = generic_ref->tree_ref.level; - u64 parent = generic_ref->parent; ASSERT(generic_ref->type == BTRFS_REF_METADATA && generic_ref->action); node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); @@ -1087,9 +1096,6 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, ref = btrfs_delayed_node_to_tree_ref(node); init_delayed_ref_common(fs_info, node, generic_ref); - ref->root = generic_ref->ref_root; - ref->parent = parent; - ref->level = level; init_delayed_ref_head(head_ref, generic_ref, record, 0); head_ref->extent_op = extent_op; @@ -1141,10 +1147,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, bool qrecord_inserted; int action = generic_ref->action; bool merged; - u64 parent = generic_ref->parent; - u64 ref_root = generic_ref->ref_root; - u64 owner = generic_ref->data_ref.ino; - u64 offset = generic_ref->data_ref.offset; ASSERT(generic_ref->type == BTRFS_REF_DATA && action); node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); @@ -1154,11 +1156,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, ref = btrfs_delayed_node_to_data_ref(node); init_delayed_ref_common(fs_info, node, generic_ref); - ref->root = ref_root; - ref->parent = parent; - ref->objectid = owner; - ref->offset = offset; - head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); if (!head_ref) { -- cgit v1.2.3-59-g8ed1b From 1bff6d4f873790cfc675afce9860208576508c5a Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 20:27:00 -0400 Subject: btrfs: simplify delayed ref tracepoints Now that all of the delayed ref information is in the delayed ref node, drastically simplify the delayed ref tracepoints by simply passing in the btrfs_delayed_ref_node and populating the tracepoints with the values from the structure itself. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 14 ++---------- fs/btrfs/extent-tree.c | 4 ++-- include/trace/events/btrfs.h | 54 +++++++++++++++++--------------------------- 3 files changed, 25 insertions(+), 47 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 743cc52c30af..cc1510d7eee8 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -1064,7 +1064,6 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, struct btrfs_delayed_extent_op *extent_op) { struct btrfs_fs_info *fs_info = trans->fs_info; - struct btrfs_delayed_tree_ref *ref; struct btrfs_delayed_ref_node *node; struct btrfs_delayed_ref_head *head_ref; struct btrfs_delayed_ref_root *delayed_refs; @@ -1093,8 +1092,6 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, } } - ref = btrfs_delayed_node_to_tree_ref(node); - init_delayed_ref_common(fs_info, node, generic_ref); init_delayed_ref_head(head_ref, generic_ref, record, 0); @@ -1119,9 +1116,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, */ btrfs_update_delayed_refs_rsv(trans); - trace_add_delayed_tree_ref(fs_info, node, ref, - action == BTRFS_ADD_DELAYED_EXTENT ? - BTRFS_ADD_DELAYED_REF : action); + trace_add_delayed_tree_ref(fs_info, node); if (merged) kmem_cache_free(btrfs_delayed_ref_node_cachep, node); @@ -1139,7 +1134,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, u64 reserved) { struct btrfs_fs_info *fs_info = trans->fs_info; - struct btrfs_delayed_data_ref *ref; struct btrfs_delayed_ref_node *node; struct btrfs_delayed_ref_head *head_ref; struct btrfs_delayed_ref_root *delayed_refs; @@ -1153,8 +1147,6 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, if (!node) return -ENOMEM; - ref = btrfs_delayed_node_to_data_ref(node); - init_delayed_ref_common(fs_info, node, generic_ref); head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); @@ -1195,9 +1187,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, */ btrfs_update_delayed_refs_rsv(trans); - trace_add_delayed_data_ref(trans->fs_info, node, ref, - action == BTRFS_ADD_DELAYED_EXTENT ? - BTRFS_ADD_DELAYED_REF : action); + trace_add_delayed_data_ref(trans->fs_info, node); if (merged) kmem_cache_free(btrfs_delayed_ref_node_cachep, node); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 9a5e553707b7..e633ca2dc498 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1574,7 +1574,7 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, u64 flags = 0; ref = btrfs_delayed_node_to_data_ref(node); - trace_run_delayed_data_ref(trans->fs_info, node, ref, node->action); + trace_run_delayed_data_ref(trans->fs_info, node); if (node->type == BTRFS_SHARED_DATA_REF_KEY) parent = ref->parent; @@ -1737,7 +1737,7 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, u64 ref_root = 0; ref = btrfs_delayed_node_to_tree_ref(node); - trace_run_delayed_tree_ref(trans->fs_info, node, ref, node->action); + trace_run_delayed_tree_ref(trans->fs_info, node); if (node->type == BTRFS_SHARED_BLOCK_REF_KEY) parent = ref->parent; diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index 766cfd48386c..dae29f6d6b4c 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -868,11 +868,9 @@ TRACE_EVENT(btrfs_add_block_group, DECLARE_EVENT_CLASS(btrfs_delayed_tree_ref, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, - const struct btrfs_delayed_tree_ref *full_ref, - int action), + const struct btrfs_delayed_ref_node *ref), - TP_ARGS(fs_info, ref, full_ref, action), + TP_ARGS(fs_info, ref), TP_STRUCT__entry_btrfs( __field( u64, bytenr ) @@ -888,10 +886,10 @@ DECLARE_EVENT_CLASS(btrfs_delayed_tree_ref, TP_fast_assign_btrfs(fs_info, __entry->bytenr = ref->bytenr; __entry->num_bytes = ref->num_bytes; - __entry->action = action; - __entry->parent = full_ref->parent; - __entry->ref_root = full_ref->root; - __entry->level = full_ref->level; + __entry->action = ref->action; + __entry->parent = ref->tree_ref.parent; + __entry->ref_root = ref->tree_ref.root; + __entry->level = ref->tree_ref.level; __entry->type = ref->type; __entry->seq = ref->seq; ), @@ -911,31 +909,25 @@ DECLARE_EVENT_CLASS(btrfs_delayed_tree_ref, DEFINE_EVENT(btrfs_delayed_tree_ref, add_delayed_tree_ref, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, - const struct btrfs_delayed_tree_ref *full_ref, - int action), + const struct btrfs_delayed_ref_node *ref), - TP_ARGS(fs_info, ref, full_ref, action) + TP_ARGS(fs_info, ref) ); DEFINE_EVENT(btrfs_delayed_tree_ref, run_delayed_tree_ref, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, - const struct btrfs_delayed_tree_ref *full_ref, - int action), + const struct btrfs_delayed_ref_node *ref), - TP_ARGS(fs_info, ref, full_ref, action) + TP_ARGS(fs_info, ref) ); DECLARE_EVENT_CLASS(btrfs_delayed_data_ref, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, - const struct btrfs_delayed_data_ref *full_ref, - int action), + const struct btrfs_delayed_ref_node *ref), - TP_ARGS(fs_info, ref, full_ref, action), + TP_ARGS(fs_info, ref), TP_STRUCT__entry_btrfs( __field( u64, bytenr ) @@ -952,11 +944,11 @@ DECLARE_EVENT_CLASS(btrfs_delayed_data_ref, TP_fast_assign_btrfs(fs_info, __entry->bytenr = ref->bytenr; __entry->num_bytes = ref->num_bytes; - __entry->action = action; - __entry->parent = full_ref->parent; - __entry->ref_root = full_ref->root; - __entry->owner = full_ref->objectid; - __entry->offset = full_ref->offset; + __entry->action = ref->action; + __entry->parent = ref->data_ref.parent; + __entry->ref_root = ref->data_ref.root; + __entry->owner = ref->data_ref.objectid; + __entry->offset = ref->data_ref.offset; __entry->type = ref->type; __entry->seq = ref->seq; ), @@ -978,21 +970,17 @@ DECLARE_EVENT_CLASS(btrfs_delayed_data_ref, DEFINE_EVENT(btrfs_delayed_data_ref, add_delayed_data_ref, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, - const struct btrfs_delayed_data_ref *full_ref, - int action), + const struct btrfs_delayed_ref_node *ref), - TP_ARGS(fs_info, ref, full_ref, action) + TP_ARGS(fs_info, ref) ); DEFINE_EVENT(btrfs_delayed_data_ref, run_delayed_data_ref, TP_PROTO(const struct btrfs_fs_info *fs_info, - const struct btrfs_delayed_ref_node *ref, - const struct btrfs_delayed_data_ref *full_ref, - int action), + const struct btrfs_delayed_ref_node *ref), - TP_ARGS(fs_info, ref, full_ref, action) + TP_ARGS(fs_info, ref) ); DECLARE_EVENT_CLASS(btrfs_delayed_ref_head, -- cgit v1.2.3-59-g8ed1b From f75464f7bbbdd3bd63b91a1f023e26ef85e348c3 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 20:43:09 -0400 Subject: btrfs: unify the btrfs_add_delayed_*_ref helpers into one helper Now that these helpers are identical, create a helper function that handles everything properly and strip the individual helpers down to use just the common helper. This cleans up a significant amount of duplicated code. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 102 ++++++++++++------------------------------------- 1 file changed, 25 insertions(+), 77 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index cc1510d7eee8..b31aa9aa26f1 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -1054,14 +1054,10 @@ void btrfs_init_data_ref(struct btrfs_ref *generic_ref, u64 ino, u64 offset, generic_ref->skip_qgroup = false; } -/* - * add a delayed tree ref. This does all of the accounting required - * to make sure the delayed ref is eventually processed before this - * transaction commits. - */ -int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, - struct btrfs_ref *generic_ref, - struct btrfs_delayed_extent_op *extent_op) +static int add_delayed_ref(struct btrfs_trans_handle *trans, + struct btrfs_ref *generic_ref, + struct btrfs_delayed_extent_op *extent_op, + u64 reserved) { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_delayed_ref_node *node; @@ -1069,10 +1065,9 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_qgroup_extent_record *record = NULL; bool qrecord_inserted; - bool merged; int action = generic_ref->action; + bool merged; - ASSERT(generic_ref->type == BTRFS_REF_METADATA && generic_ref->action); node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); if (!node) return -ENOMEM; @@ -1093,8 +1088,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, } init_delayed_ref_common(fs_info, node, generic_ref); - - init_delayed_ref_head(head_ref, generic_ref, record, 0); + init_delayed_ref_head(head_ref, generic_ref, record, reserved); head_ref->extent_op = extent_op; delayed_refs = &trans->transaction->delayed_refs; @@ -1116,16 +1110,30 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, */ btrfs_update_delayed_refs_rsv(trans); - trace_add_delayed_tree_ref(fs_info, node); + if (generic_ref->type == BTRFS_REF_DATA) + trace_add_delayed_data_ref(trans->fs_info, node); + else + trace_add_delayed_tree_ref(trans->fs_info, node); if (merged) kmem_cache_free(btrfs_delayed_ref_node_cachep, node); if (qrecord_inserted) - btrfs_qgroup_trace_extent_post(trans, record); - + return btrfs_qgroup_trace_extent_post(trans, record); return 0; } +/* + * Add a delayed tree ref. This does all of the accounting required to make sure + * the delayed ref is eventually processed before this transaction commits. + */ +int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, + struct btrfs_ref *generic_ref, + struct btrfs_delayed_extent_op *extent_op) +{ + ASSERT(generic_ref->type == BTRFS_REF_METADATA && generic_ref->action); + return add_delayed_ref(trans, generic_ref, extent_op, 0); +} + /* * add a delayed data ref. it's similar to btrfs_add_delayed_tree_ref. */ @@ -1133,68 +1141,8 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, struct btrfs_ref *generic_ref, u64 reserved) { - struct btrfs_fs_info *fs_info = trans->fs_info; - struct btrfs_delayed_ref_node *node; - struct btrfs_delayed_ref_head *head_ref; - struct btrfs_delayed_ref_root *delayed_refs; - struct btrfs_qgroup_extent_record *record = NULL; - bool qrecord_inserted; - int action = generic_ref->action; - bool merged; - - ASSERT(generic_ref->type == BTRFS_REF_DATA && action); - node = kmem_cache_alloc(btrfs_delayed_ref_node_cachep, GFP_NOFS); - if (!node) - return -ENOMEM; - - init_delayed_ref_common(fs_info, node, generic_ref); - - head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); - if (!head_ref) { - kmem_cache_free(btrfs_delayed_ref_node_cachep, node); - return -ENOMEM; - } - - if (btrfs_qgroup_full_accounting(fs_info) && !generic_ref->skip_qgroup) { - record = kzalloc(sizeof(*record), GFP_NOFS); - if (!record) { - kmem_cache_free(btrfs_delayed_ref_node_cachep, node); - kmem_cache_free(btrfs_delayed_ref_head_cachep, - head_ref); - return -ENOMEM; - } - } - - init_delayed_ref_head(head_ref, generic_ref, record, reserved); - head_ref->extent_op = NULL; - - delayed_refs = &trans->transaction->delayed_refs; - spin_lock(&delayed_refs->lock); - - /* - * insert both the head node and the new ref without dropping - * the spin lock - */ - head_ref = add_delayed_ref_head(trans, head_ref, record, - action, &qrecord_inserted); - - merged = insert_delayed_ref(trans, head_ref, node); - spin_unlock(&delayed_refs->lock); - - /* - * Need to update the delayed_refs_rsv with any changes we may have - * made. - */ - btrfs_update_delayed_refs_rsv(trans); - - trace_add_delayed_data_ref(trans->fs_info, node); - if (merged) - kmem_cache_free(btrfs_delayed_ref_node_cachep, node); - - - if (qrecord_inserted) - return btrfs_qgroup_trace_extent_post(trans, record); - return 0; + ASSERT(generic_ref->type == BTRFS_REF_DATA && generic_ref->action); + return add_delayed_ref(trans, generic_ref, NULL, reserved); } int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans, -- cgit v1.2.3-59-g8ed1b From 12390e42b69d7e439cc2a266b6c7b4e1116de7c9 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 20:52:26 -0400 Subject: btrfs: rename ->len to ->num_bytes in btrfs_ref We consistently use ->num_bytes everywhere through the delayed ref code, except in btrfs_ref. Rename btrfs_ref to match all the other code. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 8 ++++---- fs/btrfs/delayed-ref.h | 2 +- fs/btrfs/extent-tree.c | 14 +++++++------- fs/btrfs/file.c | 10 +++++----- fs/btrfs/inode-item.c | 2 +- fs/btrfs/ref-verify.c | 2 +- fs/btrfs/relocation.c | 14 +++++++------- fs/btrfs/tree-log.c | 2 +- 8 files changed, 27 insertions(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index b31aa9aa26f1..0f5e1f1f3efb 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -874,7 +874,7 @@ static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref, refcount_set(&head_ref->refs, 1); head_ref->bytenr = generic_ref->bytenr; - head_ref->num_bytes = generic_ref->len; + head_ref->num_bytes = generic_ref->num_bytes; head_ref->ref_mod = count_mod; head_ref->reserved_bytes = reserved; head_ref->must_insert_reserved = must_insert_reserved; @@ -895,7 +895,7 @@ static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref, qrecord->data_rsv_refroot = generic_ref->ref_root; } qrecord->bytenr = generic_ref->bytenr; - qrecord->num_bytes = generic_ref->len; + qrecord->num_bytes = generic_ref->num_bytes; qrecord->old_roots = NULL; } } @@ -1000,7 +1000,7 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, refcount_set(&ref->refs, 1); ref->bytenr = generic_ref->bytenr; - ref->num_bytes = generic_ref->len; + ref->num_bytes = generic_ref->num_bytes; ref->ref_mod = 1; ref->action = action; ref->seq = seq; @@ -1155,7 +1155,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans, .type = BTRFS_REF_METADATA, .action = BTRFS_UPDATE_DELAYED_HEAD, .bytenr = bytenr, - .len = num_bytes, + .num_bytes = num_bytes, }; head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 167c4407c3c4..05cf6eb2e604 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -260,7 +260,7 @@ struct btrfs_ref { u64 real_root; #endif u64 bytenr; - u64 len; + u64 num_bytes; u64 owning_root; /* diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index e633ca2dc498..96fb67a88c1e 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2542,7 +2542,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, if (ref.bytenr == 0) continue; - ref.len = btrfs_file_extent_disk_num_bytes(buf, fi); + ref.num_bytes = btrfs_file_extent_disk_num_bytes(buf, fi); ref.owning_root = ref_root; key.offset -= btrfs_file_extent_offset(buf, fi); @@ -2557,7 +2557,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, } else { /* We don't know the owning_root, leave as 0. */ ref.bytenr = btrfs_node_blockptr(buf, i); - ref.len = fs_info->nodesize; + ref.num_bytes = fs_info->nodesize; btrfs_init_tree_ref(&ref, level - 1, root->root_key.objectid, for_reloc); @@ -3466,7 +3466,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, struct btrfs_ref generic_ref = { .action = BTRFS_DROP_DELAYED_REF, .bytenr = buf->start, - .len = buf->len, + .num_bytes = buf->len, .parent = parent, .owning_root = btrfs_header_owner(buf), .ref_root = root_id, @@ -3560,7 +3560,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref) * tree, just update pinning info and exit early. */ if (ref->ref_root == BTRFS_TREE_LOG_OBJECTID) { - btrfs_pin_extent(trans, ref->bytenr, ref->len, 1); + btrfs_pin_extent(trans, ref->bytenr, ref->num_bytes, 1); ret = 0; } else if (ref->type == BTRFS_REF_METADATA) { ret = btrfs_add_delayed_tree_ref(trans, ref, NULL); @@ -4967,7 +4967,7 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans, struct btrfs_ref generic_ref = { .action = BTRFS_ADD_DELAYED_EXTENT, .bytenr = ins->objectid, - .len = ins->offset, + .num_bytes = ins->offset, .owning_root = root->root_key.objectid, .ref_root = root->root_key.objectid, }; @@ -5201,7 +5201,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, struct btrfs_ref generic_ref = { .action = BTRFS_ADD_DELAYED_EXTENT, .bytenr = ins.objectid, - .len = ins.offset, + .num_bytes = ins.offset, .parent = parent, .owning_root = owning_root, .ref_root = root_objectid, @@ -5586,7 +5586,7 @@ skip: struct btrfs_ref ref = { .action = BTRFS_DROP_DELAYED_REF, .bytenr = bytenr, - .len = fs_info->nodesize, + .num_bytes = fs_info->nodesize, .owning_root = owner_root, .ref_root = root->root_key.objectid, }; diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index f17170eb0700..065658f2c061 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -375,7 +375,7 @@ next_slot: struct btrfs_ref ref = { .action = BTRFS_ADD_DELAYED_REF, .bytenr = disk_bytenr, - .len = num_bytes, + .num_bytes = num_bytes, .parent = 0, .owning_root = root->root_key.objectid, .ref_root = root->root_key.objectid, @@ -468,7 +468,7 @@ delete_extent_item: struct btrfs_ref ref = { .action = BTRFS_DROP_DELAYED_REF, .bytenr = disk_bytenr, - .len = num_bytes, + .num_bytes = num_bytes, .parent = 0, .owning_root = root->root_key.objectid, .ref_root = root->root_key.objectid, @@ -753,7 +753,7 @@ again: ref.action = BTRFS_ADD_DELAYED_REF; ref.bytenr = bytenr; - ref.len = num_bytes; + ref.num_bytes = num_bytes; ref.parent = 0; ref.owning_root = root->root_key.objectid; ref.ref_root = root->root_key.objectid; @@ -783,7 +783,7 @@ again: ref.action = BTRFS_DROP_DELAYED_REF; ref.bytenr = bytenr; - ref.len = num_bytes; + ref.num_bytes = num_bytes; ref.parent = 0; ref.owning_root = root->root_key.objectid; ref.ref_root = root->root_key.objectid; @@ -2492,7 +2492,7 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, struct btrfs_ref ref = { .action = BTRFS_ADD_DELAYED_REF, .bytenr = extent_info->disk_offset, - .len = extent_info->disk_len, + .num_bytes = extent_info->disk_len, .owning_root = root->root_key.objectid, .ref_root = root->root_key.objectid, }; diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c index e24605df35bb..7565ff15a69a 100644 --- a/fs/btrfs/inode-item.c +++ b/fs/btrfs/inode-item.c @@ -673,7 +673,7 @@ delete: struct btrfs_ref ref = { .action = BTRFS_DROP_DELAYED_REF, .bytenr = extent_start, - .len = extent_num_bytes, + .num_bytes = extent_num_bytes, .owning_root = root->root_key.objectid, .ref_root = btrfs_header_owner(leaf), }; diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c index 1108be7a050c..94bbb7ef9a13 100644 --- a/fs/btrfs/ref-verify.c +++ b/fs/btrfs/ref-verify.c @@ -673,7 +673,7 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info, int ret = 0; bool metadata; u64 bytenr = generic_ref->bytenr; - u64 num_bytes = generic_ref->len; + u64 num_bytes = generic_ref->num_bytes; u64 parent = generic_ref->parent; u64 ref_root = 0; u64 owner = 0; diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 085b2b616569..00f33d8c4bf9 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1106,7 +1106,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans, key.offset -= btrfs_file_extent_offset(leaf, fi); ref.action = BTRFS_ADD_DELAYED_REF; ref.bytenr = new_bytenr; - ref.len = num_bytes; + ref.num_bytes = num_bytes; ref.parent = parent; ref.owning_root = root->root_key.objectid; ref.ref_root = btrfs_header_owner(leaf); @@ -1120,7 +1120,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans, ref.action = BTRFS_DROP_DELAYED_REF; ref.bytenr = bytenr; - ref.len = num_bytes; + ref.num_bytes = num_bytes; ref.parent = parent; ref.owning_root = root->root_key.objectid; ref.ref_root = btrfs_header_owner(leaf); @@ -1336,7 +1336,7 @@ again: ref.action = BTRFS_ADD_DELAYED_REF; ref.bytenr = old_bytenr; - ref.len = blocksize; + ref.num_bytes = blocksize; ref.parent = path->nodes[level]->start; ref.owning_root = src->root_key.objectid; ref.ref_root = src->root_key.objectid; @@ -1349,7 +1349,7 @@ again: ref.action = BTRFS_ADD_DELAYED_REF; ref.bytenr = new_bytenr; - ref.len = blocksize; + ref.num_bytes = blocksize; ref.parent = 0; ref.owning_root = dest->root_key.objectid; ref.ref_root = dest->root_key.objectid; @@ -1363,7 +1363,7 @@ again: /* We don't know the real owning_root, use 0. */ ref.action = BTRFS_DROP_DELAYED_REF; ref.bytenr = new_bytenr; - ref.len = blocksize; + ref.num_bytes = blocksize; ref.parent = path->nodes[level]->start; ref.owning_root = 0; ref.ref_root = src->root_key.objectid; @@ -1377,7 +1377,7 @@ again: /* We don't know the real owning_root, use 0. */ ref.action = BTRFS_DROP_DELAYED_REF; ref.bytenr = old_bytenr; - ref.len = blocksize; + ref.num_bytes = blocksize; ref.parent = 0; ref.owning_root = 0; ref.ref_root = dest->root_key.objectid; @@ -2482,7 +2482,7 @@ static int do_relocation(struct btrfs_trans_handle *trans, struct btrfs_ref ref = { .action = BTRFS_ADD_DELAYED_REF, .bytenr = node->eb->start, - .len = blocksize, + .num_bytes = blocksize, .parent = upper->eb->start, .owning_root = btrfs_header_owner(upper->eb), .ref_root = btrfs_header_owner(upper->eb), diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 88f2bbd12170..1c7efb7c2160 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -764,7 +764,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, struct btrfs_ref ref = { .action = BTRFS_ADD_DELAYED_REF, .bytenr = ins.objectid, - .len = ins.offset, + .num_bytes = ins.offset, .owning_root = root->root_key.objectid, .ref_root = root->root_key.objectid, }; -- cgit v1.2.3-59-g8ed1b From cf4f04325b2b27efa5697ba0ea4c1abdee0035b4 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 22:57:13 -0400 Subject: btrfs: move ->parent and ->ref_root into btrfs_delayed_ref_node These two members are shared by both the tree refs and data refs, so move them into btrfs_delayed_ref_node proper. This allows us to greatly simplify the comparison code, as the shared refs always only sort on parent, and the non shared refs always sort first on ref_root, and then only data refs sort on their specific fields. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/backref.c | 12 +++---- fs/btrfs/delayed-ref.c | 82 ++++++++++++++------------------------------ fs/btrfs/delayed-ref.h | 13 ++++--- fs/btrfs/extent-tree.c | 18 +++++----- include/trace/events/btrfs.h | 8 ++--- 5 files changed, 52 insertions(+), 81 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 58110c968667..abf2f4bd9f4a 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -928,7 +928,7 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, } ref = btrfs_delayed_node_to_tree_ref(node); - ret = add_indirect_ref(fs_info, preftrees, ref->root, + ret = add_indirect_ref(fs_info, preftrees, node->ref_root, key_ptr, ref->level + 1, node->bytenr, count, sc, GFP_ATOMIC); @@ -941,7 +941,7 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, ref = btrfs_delayed_node_to_tree_ref(node); ret = add_direct_ref(fs_info, preftrees, ref->level + 1, - ref->parent, node->bytenr, count, + node->parent, node->bytenr, count, sc, GFP_ATOMIC); break; } @@ -972,18 +972,14 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, if (sc && count < 0) sc->have_delayed_delete_refs = true; - ret = add_indirect_ref(fs_info, preftrees, ref->root, + ret = add_indirect_ref(fs_info, preftrees, node->ref_root, &key, 0, node->bytenr, count, sc, GFP_ATOMIC); break; } case BTRFS_SHARED_DATA_REF_KEY: { /* SHARED DIRECT FULL backref */ - struct btrfs_delayed_data_ref *ref; - - ref = btrfs_delayed_node_to_data_ref(node); - - ret = add_direct_ref(fs_info, preftrees, 0, ref->parent, + ret = add_direct_ref(fs_info, preftrees, 0, node->parent, node->bytenr, count, sc, GFP_ATOMIC); break; diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 0f5e1f1f3efb..7342e8778166 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -303,55 +303,20 @@ int btrfs_delayed_refs_rsv_refill(struct btrfs_fs_info *fs_info, return 0; } -/* - * compare two delayed tree backrefs with same bytenr and type - */ -static int comp_tree_refs(struct btrfs_delayed_tree_ref *ref1, - struct btrfs_delayed_tree_ref *ref2) -{ - struct btrfs_delayed_ref_node *node = btrfs_delayed_tree_ref_to_node(ref1); - - if (node->type == BTRFS_TREE_BLOCK_REF_KEY) { - if (ref1->root < ref2->root) - return -1; - if (ref1->root > ref2->root) - return 1; - } else { - if (ref1->parent < ref2->parent) - return -1; - if (ref1->parent > ref2->parent) - return 1; - } - return 0; -} - /* * compare two delayed data backrefs with same bytenr and type */ -static int comp_data_refs(struct btrfs_delayed_data_ref *ref1, - struct btrfs_delayed_data_ref *ref2) +static int comp_data_refs(struct btrfs_delayed_ref_node *ref1, + struct btrfs_delayed_ref_node *ref2) { - struct btrfs_delayed_ref_node *node = btrfs_delayed_data_ref_to_node(ref1); - - if (node->type == BTRFS_EXTENT_DATA_REF_KEY) { - if (ref1->root < ref2->root) - return -1; - if (ref1->root > ref2->root) - return 1; - if (ref1->objectid < ref2->objectid) - return -1; - if (ref1->objectid > ref2->objectid) - return 1; - if (ref1->offset < ref2->offset) - return -1; - if (ref1->offset > ref2->offset) - return 1; - } else { - if (ref1->parent < ref2->parent) - return -1; - if (ref1->parent > ref2->parent) - return 1; - } + if (ref1->data_ref.objectid < ref2->data_ref.objectid) + return -1; + if (ref1->data_ref.objectid > ref2->data_ref.objectid) + return 1; + if (ref1->data_ref.offset < ref2->data_ref.offset) + return -1; + if (ref1->data_ref.offset > ref2->data_ref.offset) + return 1; return 0; } @@ -365,13 +330,20 @@ static int comp_refs(struct btrfs_delayed_ref_node *ref1, return -1; if (ref1->type > ref2->type) return 1; - if (ref1->type == BTRFS_TREE_BLOCK_REF_KEY || - ref1->type == BTRFS_SHARED_BLOCK_REF_KEY) - ret = comp_tree_refs(btrfs_delayed_node_to_tree_ref(ref1), - btrfs_delayed_node_to_tree_ref(ref2)); - else - ret = comp_data_refs(btrfs_delayed_node_to_data_ref(ref1), - btrfs_delayed_node_to_data_ref(ref2)); + if (ref1->type == BTRFS_SHARED_BLOCK_REF_KEY || + ref1->type == BTRFS_SHARED_DATA_REF_KEY) { + if (ref1->parent < ref2->parent) + return -1; + if (ref1->parent > ref2->parent) + return 1; + } else { + if (ref1->ref_root < ref2->ref_root) + return -1; + if (ref1->ref_root > ref2->ref_root) + return -1; + if (ref1->type == BTRFS_EXTENT_DATA_REF_KEY) + ret = comp_data_refs(ref1, ref2); + } if (ret) return ret; if (check_seq) { @@ -1005,17 +977,15 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, ref->action = action; ref->seq = seq; ref->type = btrfs_ref_type(generic_ref); + ref->ref_root = generic_ref->ref_root; + ref->parent = generic_ref->parent; RB_CLEAR_NODE(&ref->ref_node); INIT_LIST_HEAD(&ref->add_list); if (generic_ref->type == BTRFS_REF_DATA) { - ref->data_ref.root = generic_ref->ref_root; - ref->data_ref.parent = generic_ref->parent; ref->data_ref.objectid = generic_ref->data_ref.ino; ref->data_ref.offset = generic_ref->data_ref.offset; } else { - ref->tree_ref.root = generic_ref->ref_root; - ref->tree_ref.parent = generic_ref->parent; ref->tree_ref.level = generic_ref->tree_ref.level; } } diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 05cf6eb2e604..ec53404cec0c 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -31,14 +31,10 @@ enum btrfs_delayed_ref_action { } __packed; struct btrfs_delayed_tree_ref { - u64 root; - u64 parent; int level; }; struct btrfs_delayed_data_ref { - u64 root; - u64 parent; u64 objectid; u64 offset; }; @@ -61,6 +57,15 @@ struct btrfs_delayed_ref_node { /* seq number to keep track of insertion order */ u64 seq; + /* The ref_root for this ref */ + u64 ref_root; + + /* + * The parent for this ref, if this isn't set the ref_root is the + * reference owner. + */ + u64 parent; + /* ref count on this data structure */ refcount_t refs; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 96fb67a88c1e..60fb481e7411 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1577,7 +1577,7 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, trace_run_delayed_data_ref(trans->fs_info, node); if (node->type == BTRFS_SHARED_DATA_REF_KEY) - parent = ref->parent; + parent = node->parent; if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) { struct btrfs_key key; @@ -1596,7 +1596,7 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, key.type = BTRFS_EXTENT_ITEM_KEY; key.offset = node->num_bytes; - ret = alloc_reserved_file_extent(trans, parent, ref->root, + ret = alloc_reserved_file_extent(trans, parent, node->ref_root, flags, ref->objectid, ref->offset, &key, node->ref_mod, href->owning_root); @@ -1604,12 +1604,12 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, if (!ret) ret = btrfs_record_squota_delta(trans->fs_info, &delta); } else if (node->action == BTRFS_ADD_DELAYED_REF) { - ret = __btrfs_inc_extent_ref(trans, node, parent, ref->root, + ret = __btrfs_inc_extent_ref(trans, node, parent, node->ref_root, ref->objectid, ref->offset, extent_op); } else if (node->action == BTRFS_DROP_DELAYED_REF) { ret = __btrfs_free_extent(trans, href, node, parent, - ref->root, ref->objectid, + node->ref_root, ref->objectid, ref->offset, extent_op); } else { BUG(); @@ -1740,8 +1740,8 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, trace_run_delayed_tree_ref(trans->fs_info, node); if (node->type == BTRFS_SHARED_BLOCK_REF_KEY) - parent = ref->parent; - ref_root = ref->root; + parent = node->parent; + ref_root = node->ref_root; if (unlikely(node->ref_mod != 1)) { btrfs_err(trans->fs_info, @@ -2359,7 +2359,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, * If our ref doesn't match the one we're currently looking at * then we have a cross reference. */ - if (data_ref->root != root->root_key.objectid || + if (ref->ref_root != root->root_key.objectid || data_ref->objectid != objectid || data_ref->offset != offset) { ret = 1; @@ -4946,11 +4946,11 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, if (node->type == BTRFS_SHARED_BLOCK_REF_KEY) { btrfs_set_extent_inline_ref_type(leaf, iref, BTRFS_SHARED_BLOCK_REF_KEY); - btrfs_set_extent_inline_ref_offset(leaf, iref, ref->parent); + btrfs_set_extent_inline_ref_offset(leaf, iref, node->parent); } else { btrfs_set_extent_inline_ref_type(leaf, iref, BTRFS_TREE_BLOCK_REF_KEY); - btrfs_set_extent_inline_ref_offset(leaf, iref, ref->root); + btrfs_set_extent_inline_ref_offset(leaf, iref, node->ref_root); } btrfs_mark_buffer_dirty(trans, leaf); diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index dae29f6d6b4c..e6cee75c384c 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -887,8 +887,8 @@ DECLARE_EVENT_CLASS(btrfs_delayed_tree_ref, __entry->bytenr = ref->bytenr; __entry->num_bytes = ref->num_bytes; __entry->action = ref->action; - __entry->parent = ref->tree_ref.parent; - __entry->ref_root = ref->tree_ref.root; + __entry->parent = ref->parent; + __entry->ref_root = ref->ref_root; __entry->level = ref->tree_ref.level; __entry->type = ref->type; __entry->seq = ref->seq; @@ -945,8 +945,8 @@ DECLARE_EVENT_CLASS(btrfs_delayed_data_ref, __entry->bytenr = ref->bytenr; __entry->num_bytes = ref->num_bytes; __entry->action = ref->action; - __entry->parent = ref->data_ref.parent; - __entry->ref_root = ref->data_ref.root; + __entry->parent = ref->parent; + __entry->ref_root = ref->ref_root; __entry->owner = ref->data_ref.objectid; __entry->offset = ref->data_ref.offset; __entry->type = ref->type; -- cgit v1.2.3-59-g8ed1b From 53667634467178f0049d9d5300fabd017dfb5477 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 23:01:38 -0400 Subject: btrfs: rename btrfs_data_ref->ino to ->objectid This is how we refer to it in the rest of the extent reference related code, make it consistent. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 4 ++-- fs/btrfs/delayed-ref.h | 2 +- fs/btrfs/ref-verify.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 7342e8778166..8d553347e11b 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -983,7 +983,7 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, INIT_LIST_HEAD(&ref->add_list); if (generic_ref->type == BTRFS_REF_DATA) { - ref->data_ref.objectid = generic_ref->data_ref.ino; + ref->data_ref.objectid = generic_ref->data_ref.objectid; ref->data_ref.offset = generic_ref->data_ref.offset; } else { ref->tree_ref.level = generic_ref->tree_ref.level; @@ -1014,7 +1014,7 @@ void btrfs_init_data_ref(struct btrfs_ref *generic_ref, u64 ino, u64 offset, /* If @real_root not set, use @root as fallback */ generic_ref->real_root = mod_root ?: generic_ref->ref_root; #endif - generic_ref->data_ref.ino = ino; + generic_ref->data_ref.objectid = ino; generic_ref->data_ref.offset = offset; generic_ref->type = BTRFS_REF_DATA; if (skip_qgroup || !(is_fstree(generic_ref->ref_root) && diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index ec53404cec0c..abfb565ec310 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -226,7 +226,7 @@ struct btrfs_data_ref { /* For EXTENT_DATA_REF */ /* Inode which refers to this data extent */ - u64 ino; + u64 objectid; /* * file_offset - extent_offset diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c index 94bbb7ef9a13..cf531255ab76 100644 --- a/fs/btrfs/ref-verify.c +++ b/fs/btrfs/ref-verify.c @@ -688,7 +688,7 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info, owner = generic_ref->tree_ref.level; } else if (!parent) { ref_root = generic_ref->ref_root; - owner = generic_ref->data_ref.ino; + owner = generic_ref->data_ref.objectid; offset = generic_ref->data_ref.offset; } metadata = owner < BTRFS_FIRST_FREE_OBJECTID; -- cgit v1.2.3-59-g8ed1b From a502f112adfcae499df43e3e49d7fe849d5c5df3 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 23:27:49 -0400 Subject: btrfs: make __btrfs_inc_extent_ref take a btrfs_delayed_ref_node We're just extracting the values from btrfs_delayed_ref_node and passing them through, simply pass the btrfs_delayed_ref_node into __btrfs_inc_extent_ref and shrink the function arguments. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.h | 16 ++++++++++++++++ fs/btrfs/extent-tree.c | 41 +++++++++-------------------------------- 2 files changed, 25 insertions(+), 32 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index abfb565ec310..83c5e010f619 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -426,6 +426,22 @@ btrfs_delayed_data_ref_to_node(struct btrfs_delayed_data_ref *ref) return container_of(ref, struct btrfs_delayed_ref_node, data_ref); } +static inline u64 btrfs_delayed_ref_owner(struct btrfs_delayed_ref_node *node) +{ + if (node->type == BTRFS_EXTENT_DATA_REF_KEY || + node->type == BTRFS_SHARED_DATA_REF_KEY) + return node->data_ref.objectid; + return node->tree_ref.level; +} + +static inline u64 btrfs_delayed_ref_offset(struct btrfs_delayed_ref_node *node) +{ + if (node->type == BTRFS_EXTENT_DATA_REF_KEY || + node->type == BTRFS_SHARED_DATA_REF_KEY) + return node->data_ref.offset; + return 0; +} + static inline u8 btrfs_ref_type(struct btrfs_ref *ref) { ASSERT(ref->type == BTRFS_REF_DATA || ref->type == BTRFS_REF_METADATA); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 60fb481e7411..6b5fa4f8cccc 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1462,34 +1462,12 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, * @node: The delayed ref node used to get the bytenr/length for * extent whose references are incremented. * - * @parent: If this is a shared extent (BTRFS_SHARED_DATA_REF_KEY/ - * BTRFS_SHARED_BLOCK_REF_KEY) then it holds the logical - * bytenr of the parent block. Since new extents are always - * created with indirect references, this will only be the case - * when relocating a shared extent. In that case, root_objectid - * will be BTRFS_TREE_RELOC_OBJECTID. Otherwise, parent must - * be 0 - * - * @root_objectid: The id of the root where this modification has originated, - * this can be either one of the well-known metadata trees or - * the subvolume id which references this extent. - * - * @owner: For data extents it is the inode number of the owning file. - * For metadata extents this parameter holds the level in the - * tree of the extent. - * - * @offset: For metadata extents the offset is ignored and is currently - * always passed as 0. For data extents it is the fileoffset - * this extent belongs to. - * * @extent_op Pointer to a structure, holding information necessary when * updating a tree block's flags * */ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_node *node, - u64 parent, u64 root_objectid, - u64 owner, u64 offset, struct btrfs_delayed_extent_op *extent_op) { struct btrfs_path *path; @@ -1498,6 +1476,8 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, struct btrfs_key key; u64 bytenr = node->bytenr; u64 num_bytes = node->num_bytes; + u64 owner = btrfs_delayed_ref_owner(node); + u64 offset = btrfs_delayed_ref_offset(node); u64 refs; int refs_to_add = node->ref_mod; int ret; @@ -1508,7 +1488,7 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, /* this will setup the path even if it fails to insert the back ref */ ret = insert_inline_extent_backref(trans, path, bytenr, num_bytes, - parent, root_objectid, owner, + node->parent, node->ref_root, owner, offset, refs_to_add, extent_op); if ((ret < 0 && ret != -EAGAIN) || !ret) goto out; @@ -1531,11 +1511,11 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, /* now insert the actual backref */ if (owner < BTRFS_FIRST_FREE_OBJECTID) - ret = insert_tree_block_ref(trans, path, bytenr, parent, - root_objectid); + ret = insert_tree_block_ref(trans, path, bytenr, node->parent, + node->ref_root); else - ret = insert_extent_data_ref(trans, path, bytenr, parent, - root_objectid, owner, offset, + ret = insert_extent_data_ref(trans, path, bytenr, node->parent, + node->ref_root, owner, offset, refs_to_add); if (ret) @@ -1604,9 +1584,7 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, if (!ret) ret = btrfs_record_squota_delta(trans->fs_info, &delta); } else if (node->action == BTRFS_ADD_DELAYED_REF) { - ret = __btrfs_inc_extent_ref(trans, node, parent, node->ref_root, - ref->objectid, ref->offset, - extent_op); + ret = __btrfs_inc_extent_ref(trans, node, extent_op); } else if (node->action == BTRFS_DROP_DELAYED_REF) { ret = __btrfs_free_extent(trans, href, node, parent, node->ref_root, ref->objectid, @@ -1764,8 +1742,7 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, if (!ret) btrfs_record_squota_delta(fs_info, &delta); } else if (node->action == BTRFS_ADD_DELAYED_REF) { - ret = __btrfs_inc_extent_ref(trans, node, parent, ref_root, - ref->level, 0, extent_op); + ret = __btrfs_inc_extent_ref(trans, node, extent_op); } else if (node->action == BTRFS_DROP_DELAYED_REF) { ret = __btrfs_free_extent(trans, href, node, parent, ref_root, ref->level, 0, extent_op); -- cgit v1.2.3-59-g8ed1b From 85bb9f544e8ef247143fb7b3f4ca2974d71d407e Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 23:34:48 -0400 Subject: btrfs: drop unnecessary arguments from __btrfs_free_extent We have all the information we need in our btrfs_delayed_ref_node, which we already pass into __btrfs_free_extent. Drop the extra arguments and just extract the values from btrfs_delayed_ref_node. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 23 ++++++++--------------- 1 file changed, 8 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 6b5fa4f8cccc..36c1be9adcb4 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -46,9 +46,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_head *href, - struct btrfs_delayed_ref_node *node, u64 parent, - u64 root_objectid, u64 owner_objectid, - u64 owner_offset, + struct btrfs_delayed_ref_node *node, struct btrfs_delayed_extent_op *extra_op); static void __run_delayed_extent_op(struct btrfs_delayed_extent_op *extent_op, struct extent_buffer *leaf, @@ -1586,9 +1584,7 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, } else if (node->action == BTRFS_ADD_DELAYED_REF) { ret = __btrfs_inc_extent_ref(trans, node, extent_op); } else if (node->action == BTRFS_DROP_DELAYED_REF) { - ret = __btrfs_free_extent(trans, href, node, parent, - node->ref_root, ref->objectid, - ref->offset, extent_op); + ret = __btrfs_free_extent(trans, href, node, extent_op); } else { BUG(); } @@ -1710,11 +1706,9 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, { int ret = 0; struct btrfs_fs_info *fs_info = trans->fs_info; - struct btrfs_delayed_tree_ref *ref; u64 parent = 0; u64 ref_root = 0; - ref = btrfs_delayed_node_to_tree_ref(node); trace_run_delayed_tree_ref(trans->fs_info, node); if (node->type == BTRFS_SHARED_BLOCK_REF_KEY) @@ -1744,8 +1738,7 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, } else if (node->action == BTRFS_ADD_DELAYED_REF) { ret = __btrfs_inc_extent_ref(trans, node, extent_op); } else if (node->action == BTRFS_DROP_DELAYED_REF) { - ret = __btrfs_free_extent(trans, href, node, parent, ref_root, - ref->level, 0, extent_op); + ret = __btrfs_free_extent(trans, href, node, extent_op); } else { BUG(); } @@ -3077,9 +3070,7 @@ static int do_free_extent_accounting(struct btrfs_trans_handle *trans, */ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_head *href, - struct btrfs_delayed_ref_node *node, u64 parent, - u64 root_objectid, u64 owner_objectid, - u64 owner_offset, + struct btrfs_delayed_ref_node *node, struct btrfs_delayed_extent_op *extent_op) { struct btrfs_fs_info *info = trans->fs_info; @@ -3099,6 +3090,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, u64 refs; u64 bytenr = node->bytenr; u64 num_bytes = node->num_bytes; + u64 owner_objectid = btrfs_delayed_ref_owner(node); + u64 owner_offset = btrfs_delayed_ref_offset(node); bool skinny_metadata = btrfs_fs_incompat(info, SKINNY_METADATA); u64 delayed_ref_root = href->owning_root; @@ -3124,7 +3117,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, skinny_metadata = false; ret = lookup_extent_backref(trans, path, &iref, bytenr, num_bytes, - parent, root_objectid, owner_objectid, + node->parent, node->ref_root, owner_objectid, owner_offset); if (ret == 0) { /* @@ -3226,7 +3219,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, } else if (WARN_ON(ret == -ENOENT)) { abort_and_dump(trans, path, "unable to find ref byte nr %llu parent %llu root %llu owner %llu offset %llu slot %d", - bytenr, parent, root_objectid, owner_objectid, + bytenr, node->parent, node->ref_root, owner_objectid, owner_offset, path->slots[0]); goto out; } else { -- cgit v1.2.3-59-g8ed1b From b4b5934ac1c3a6a29648f01486541ddf5b981234 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 23:43:46 -0400 Subject: btrfs: make the insert backref helpers take a btrfs_delayed_ref_node We don't need to pass in all the elements for the backrefs as function arguments, simply pass through the btrfs_delayed_ref_node and then extract the values we need from that. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 46 +++++++++++++++++++++------------------------- 1 file changed, 21 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 36c1be9adcb4..618dd1c50ecf 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -513,26 +513,26 @@ fail: static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans, struct btrfs_path *path, - u64 bytenr, u64 parent, - u64 root_objectid, u64 owner, - u64 offset, int refs_to_add) + struct btrfs_delayed_ref_node *node, + u64 bytenr) { struct btrfs_root *root = btrfs_extent_root(trans->fs_info, bytenr); struct btrfs_key key; struct extent_buffer *leaf; + u64 owner = btrfs_delayed_ref_owner(node); + u64 offset = btrfs_delayed_ref_offset(node); u32 size; u32 num_refs; int ret; key.objectid = bytenr; - if (parent) { + if (node->parent) { key.type = BTRFS_SHARED_DATA_REF_KEY; - key.offset = parent; + key.offset = node->parent; size = sizeof(struct btrfs_shared_data_ref); } else { key.type = BTRFS_EXTENT_DATA_REF_KEY; - key.offset = hash_extent_data_ref(root_objectid, - owner, offset); + key.offset = hash_extent_data_ref(node->ref_root, owner, offset); size = sizeof(struct btrfs_extent_data_ref); } @@ -541,15 +541,15 @@ static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans, goto fail; leaf = path->nodes[0]; - if (parent) { + if (node->parent) { struct btrfs_shared_data_ref *ref; ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_shared_data_ref); if (ret == 0) { - btrfs_set_shared_data_ref_count(leaf, ref, refs_to_add); + btrfs_set_shared_data_ref_count(leaf, ref, node->ref_mod); } else { num_refs = btrfs_shared_data_ref_count(leaf, ref); - num_refs += refs_to_add; + num_refs += node->ref_mod; btrfs_set_shared_data_ref_count(leaf, ref, num_refs); } } else { @@ -557,7 +557,7 @@ static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans, while (ret == -EEXIST) { ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_data_ref); - if (match_extent_data_ref(leaf, ref, root_objectid, + if (match_extent_data_ref(leaf, ref, node->ref_root, owner, offset)) break; btrfs_release_path(path); @@ -572,14 +572,13 @@ static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans, ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_data_ref); if (ret == 0) { - btrfs_set_extent_data_ref_root(leaf, ref, - root_objectid); + btrfs_set_extent_data_ref_root(leaf, ref, node->ref_root); btrfs_set_extent_data_ref_objectid(leaf, ref, owner); btrfs_set_extent_data_ref_offset(leaf, ref, offset); - btrfs_set_extent_data_ref_count(leaf, ref, refs_to_add); + btrfs_set_extent_data_ref_count(leaf, ref, node->ref_mod); } else { num_refs = btrfs_extent_data_ref_count(leaf, ref); - num_refs += refs_to_add; + num_refs += node->ref_mod; btrfs_set_extent_data_ref_count(leaf, ref, num_refs); } } @@ -703,20 +702,20 @@ static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans, static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans, struct btrfs_path *path, - u64 bytenr, u64 parent, - u64 root_objectid) + struct btrfs_delayed_ref_node *node, + u64 bytenr) { struct btrfs_root *root = btrfs_extent_root(trans->fs_info, bytenr); struct btrfs_key key; int ret; key.objectid = bytenr; - if (parent) { + if (node->parent) { key.type = BTRFS_SHARED_BLOCK_REF_KEY; - key.offset = parent; + key.offset = node->parent; } else { key.type = BTRFS_TREE_BLOCK_REF_KEY; - key.offset = root_objectid; + key.offset = node->ref_root; } ret = btrfs_insert_empty_item(trans, root, path, &key, 0); @@ -1509,12 +1508,9 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, /* now insert the actual backref */ if (owner < BTRFS_FIRST_FREE_OBJECTID) - ret = insert_tree_block_ref(trans, path, bytenr, node->parent, - node->ref_root); + ret = insert_tree_block_ref(trans, path, node, bytenr); else - ret = insert_extent_data_ref(trans, path, bytenr, node->parent, - node->ref_root, owner, offset, - refs_to_add); + ret = insert_extent_data_ref(trans, path, node, bytenr); if (ret) btrfs_abort_transaction(trans, ret); -- cgit v1.2.3-59-g8ed1b From 44cc2e38e67b7b7551e923571ef6ea51f3ae1e05 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 12 Apr 2024 23:53:49 -0400 Subject: btrfs: stop referencing btrfs_delayed_data_ref directly Now that most of our elements are inside of btrfs_delayed_ref_node directly and we have helpers for the delayed_data_ref bits, go ahead and remove all direct usage of btrfs_delayed_data_ref and use the helpers where needed. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/backref.c | 7 ++----- fs/btrfs/extent-tree.c | 20 +++++++++++--------- include/trace/events/btrfs.h | 1 - 3 files changed, 13 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index abf2f4bd9f4a..06849122a022 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -947,12 +947,9 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, } case BTRFS_EXTENT_DATA_REF_KEY: { /* NORMAL INDIRECT DATA backref */ - struct btrfs_delayed_data_ref *ref; - ref = btrfs_delayed_node_to_data_ref(node); - - key.objectid = ref->objectid; + key.objectid = btrfs_delayed_ref_owner(node); key.type = BTRFS_EXTENT_DATA_KEY; - key.offset = ref->offset; + key.offset = btrfs_delayed_ref_offset(node); /* * If we have a share check context and a reference for diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 618dd1c50ecf..24997c83dc14 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1543,11 +1543,9 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, bool insert_reserved) { int ret = 0; - struct btrfs_delayed_data_ref *ref; u64 parent = 0; u64 flags = 0; - ref = btrfs_delayed_node_to_data_ref(node); trace_run_delayed_data_ref(trans->fs_info, node); if (node->type == BTRFS_SHARED_DATA_REF_KEY) @@ -1562,6 +1560,8 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, .is_inc = true, .generation = trans->transid, }; + u64 owner = btrfs_delayed_ref_owner(node); + u64 offset = btrfs_delayed_ref_offset(node); if (extent_op) flags |= extent_op->flags_to_set; @@ -1571,9 +1571,9 @@ static int run_delayed_data_ref(struct btrfs_trans_handle *trans, key.offset = node->num_bytes; ret = alloc_reserved_file_extent(trans, parent, node->ref_root, - flags, ref->objectid, - ref->offset, &key, - node->ref_mod, href->owning_root); + flags, owner, offset, &key, + node->ref_mod, + href->owning_root); free_head_ref_squota_rsv(trans->fs_info, href); if (!ret) ret = btrfs_record_squota_delta(trans->fs_info, &delta); @@ -2258,7 +2258,6 @@ static noinline int check_delayed_ref(struct btrfs_root *root, { struct btrfs_delayed_ref_head *head; struct btrfs_delayed_ref_node *ref; - struct btrfs_delayed_data_ref *data_ref; struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_transaction *cur_trans; struct rb_node *node; @@ -2312,6 +2311,9 @@ static noinline int check_delayed_ref(struct btrfs_root *root, */ for (node = rb_first_cached(&head->ref_tree); node; node = rb_next(node)) { + u64 ref_owner; + u64 ref_offset; + ref = rb_entry(node, struct btrfs_delayed_ref_node, ref_node); /* If it's a shared ref we know a cross reference exists */ if (ref->type != BTRFS_EXTENT_DATA_REF_KEY) { @@ -2319,15 +2321,15 @@ static noinline int check_delayed_ref(struct btrfs_root *root, break; } - data_ref = btrfs_delayed_node_to_data_ref(ref); + ref_owner = btrfs_delayed_ref_owner(ref); + ref_offset = btrfs_delayed_ref_offset(ref); /* * If our ref doesn't match the one we're currently looking at * then we have a cross reference. */ if (ref->ref_root != root->root_key.objectid || - data_ref->objectid != objectid || - data_ref->offset != offset) { + ref_owner != objectid || ref_offset != offset) { ret = 1; break; } diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index e6cee75c384c..89fa96fd95b4 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -17,7 +17,6 @@ struct btrfs_file_extent_item; struct btrfs_ordered_extent; struct btrfs_delayed_ref_node; struct btrfs_delayed_tree_ref; -struct btrfs_delayed_data_ref; struct btrfs_delayed_ref_head; struct btrfs_block_group; struct btrfs_free_cluster; -- cgit v1.2.3-59-g8ed1b From efc7d5dbf8b42eb53db31443520e979b05cd1908 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Sat, 13 Apr 2024 00:07:06 -0400 Subject: btrfs: stop referencing btrfs_delayed_tree_ref directly We only ever need to use this to get the level of the tree block ref, so use the btrfs_delayed_ref_owner() helper, which returns the level for the given reference. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/backref.c | 21 +++++++++++---------- fs/btrfs/extent-tree.c | 10 +++++----- include/trace/events/btrfs.h | 1 - 3 files changed, 16 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 06849122a022..5bb71dd1ab2a 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -919,28 +919,29 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, switch (node->type) { case BTRFS_TREE_BLOCK_REF_KEY: { /* NORMAL INDIRECT METADATA backref */ - struct btrfs_delayed_tree_ref *ref; struct btrfs_key *key_ptr = NULL; + /* The owner of a tree block ref is the level. */ + int level = btrfs_delayed_ref_owner(node); if (head->extent_op && head->extent_op->update_key) { btrfs_disk_key_to_cpu(&key, &head->extent_op->key); key_ptr = &key; } - ref = btrfs_delayed_node_to_tree_ref(node); ret = add_indirect_ref(fs_info, preftrees, node->ref_root, - key_ptr, ref->level + 1, - node->bytenr, count, sc, - GFP_ATOMIC); + key_ptr, level + 1, node->bytenr, + count, sc, GFP_ATOMIC); break; } case BTRFS_SHARED_BLOCK_REF_KEY: { - /* SHARED DIRECT METADATA backref */ - struct btrfs_delayed_tree_ref *ref; - - ref = btrfs_delayed_node_to_tree_ref(node); + /* + * SHARED DIRECT METADATA backref + * + * The owner of a tree block ref is the level. + */ + int level = btrfs_delayed_ref_owner(node); - ret = add_direct_ref(fs_info, preftrees, ref->level + 1, + ret = add_direct_ref(fs_info, preftrees, level + 1, node->parent, node->bytenr, count, sc, GFP_ATOMIC); break; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 24997c83dc14..87a921b84bab 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4865,16 +4865,16 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, struct btrfs_extent_inline_ref *iref; struct btrfs_path *path; struct extent_buffer *leaf; - struct btrfs_delayed_tree_ref *ref; u32 size = sizeof(*extent_item) + sizeof(*iref); u64 flags = extent_op->flags_to_set; + /* The owner of a tree block is the level. */ + int level = btrfs_delayed_ref_owner(node); bool skinny_metadata = btrfs_fs_incompat(fs_info, SKINNY_METADATA); - ref = btrfs_delayed_node_to_tree_ref(node); - extent_key.objectid = node->bytenr; if (skinny_metadata) { - extent_key.offset = ref->level; + /* The owner of a tree block is the level. */ + extent_key.offset = level; extent_key.type = BTRFS_METADATA_ITEM_KEY; } else { extent_key.offset = node->num_bytes; @@ -4907,7 +4907,7 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, } else { block_info = (struct btrfs_tree_block_info *)(extent_item + 1); btrfs_set_tree_block_key(leaf, block_info, &extent_op->key); - btrfs_set_tree_block_level(leaf, block_info, ref->level); + btrfs_set_tree_block_level(leaf, block_info, level); iref = (struct btrfs_extent_inline_ref *)(block_info + 1); } diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index 89fa96fd95b4..8f2497603cb5 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -16,7 +16,6 @@ struct extent_map; struct btrfs_file_extent_item; struct btrfs_ordered_extent; struct btrfs_delayed_ref_node; -struct btrfs_delayed_tree_ref; struct btrfs_delayed_ref_head; struct btrfs_block_group; struct btrfs_free_cluster; -- cgit v1.2.3-59-g8ed1b From 7f6af7c434d9db0f937774e9f0e31bf5fc73a86e Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Sat, 13 Apr 2024 00:09:03 -0400 Subject: btrfs: remove the btrfs_delayed_ref_node container helpers Now that we don't use these helpers anywhere, remove them. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.h | 27 --------------------------- 1 file changed, 27 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 83c5e010f619..09f2c4cfea7f 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -399,33 +399,6 @@ void btrfs_migrate_to_delayed_refs_rsv(struct btrfs_fs_info *fs_info, u64 num_bytes); bool btrfs_check_space_for_delayed_refs(struct btrfs_fs_info *fs_info); -/* - * helper functions to cast a node into its container - */ -static inline struct btrfs_delayed_tree_ref * -btrfs_delayed_node_to_tree_ref(struct btrfs_delayed_ref_node *node) -{ - return &node->tree_ref; -} - -static inline struct btrfs_delayed_data_ref * -btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node) -{ - return &node->data_ref; -} - -static inline struct btrfs_delayed_ref_node * -btrfs_delayed_tree_ref_to_node(struct btrfs_delayed_tree_ref *ref) -{ - return container_of(ref, struct btrfs_delayed_ref_node, tree_ref); -} - -static inline struct btrfs_delayed_ref_node * -btrfs_delayed_data_ref_to_node(struct btrfs_delayed_data_ref *ref) -{ - return container_of(ref, struct btrfs_delayed_ref_node, data_ref); -} - static inline u64 btrfs_delayed_ref_owner(struct btrfs_delayed_ref_node *node) { if (node->type == BTRFS_EXTENT_DATA_REF_KEY || -- cgit v1.2.3-59-g8ed1b From 99f2be15229ae269afa9af0eeaca2deefa87fdb3 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Sat, 13 Apr 2024 00:11:22 -0400 Subject: btrfs: replace btrfs_delayed_*_ref with btrfs_*_ref Now that these two structs are the same, move the btrfs_data_ref and btrfs_tree_ref up and use these in the btrfs_delayed_ref_node. Then remove the btrfs_delayed_*_ref structs. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/delayed-ref.c | 10 ++++----- fs/btrfs/delayed-ref.h | 55 +++++++++++++++++++++----------------------------- 2 files changed, 27 insertions(+), 38 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 8d553347e11b..6cc80fb10da2 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -982,12 +982,10 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, RB_CLEAR_NODE(&ref->ref_node); INIT_LIST_HEAD(&ref->add_list); - if (generic_ref->type == BTRFS_REF_DATA) { - ref->data_ref.objectid = generic_ref->data_ref.objectid; - ref->data_ref.offset = generic_ref->data_ref.offset; - } else { - ref->tree_ref.level = generic_ref->tree_ref.level; - } + if (generic_ref->type == BTRFS_REF_DATA) + ref->data_ref = generic_ref->data_ref; + else + ref->tree_ref = generic_ref->tree_ref; } void btrfs_init_tree_ref(struct btrfs_ref *generic_ref, int level, u64 mod_root, diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 09f2c4cfea7f..04b180ebe1fe 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -30,15 +30,32 @@ enum btrfs_delayed_ref_action { BTRFS_UPDATE_DELAYED_HEAD, } __packed; -struct btrfs_delayed_tree_ref { - int level; -}; +struct btrfs_data_ref { + /* For EXTENT_DATA_REF */ -struct btrfs_delayed_data_ref { + /* Inode which refers to this data extent */ u64 objectid; + + /* + * file_offset - extent_offset + * + * file_offset is the key.offset of the EXTENT_DATA key. + * extent_offset is btrfs_file_extent_offset() of the EXTENT_DATA data. + */ u64 offset; }; +struct btrfs_tree_ref { + /* + * Level of this tree block. + * + * Shared for skinny (TREE_BLOCK_REF) and normal tree ref. + */ + int level; + + /* For non-skinny metadata, no special member needed */ +}; + struct btrfs_delayed_ref_node { struct rb_node ref_node; /* @@ -84,8 +101,8 @@ struct btrfs_delayed_ref_node { unsigned int type:8; union { - struct btrfs_delayed_tree_ref tree_ref; - struct btrfs_delayed_data_ref data_ref; + struct btrfs_tree_ref tree_ref; + struct btrfs_data_ref data_ref; }; }; @@ -222,32 +239,6 @@ enum btrfs_ref_type { BTRFS_REF_LAST, } __packed; -struct btrfs_data_ref { - /* For EXTENT_DATA_REF */ - - /* Inode which refers to this data extent */ - u64 objectid; - - /* - * file_offset - extent_offset - * - * file_offset is the key.offset of the EXTENT_DATA key. - * extent_offset is btrfs_file_extent_offset() of the EXTENT_DATA data. - */ - u64 offset; -}; - -struct btrfs_tree_ref { - /* - * Level of this tree block - * - * Shared for skinny (TREE_BLOCK_REF) and normal tree ref. - */ - int level; - - /* For non-skinny metadata, no special member needed */ -}; - struct btrfs_ref { enum btrfs_ref_type type; enum btrfs_delayed_ref_action action; -- cgit v1.2.3-59-g8ed1b From 53e24158684b527d013b5b2204ccb34d1f94c248 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Sun, 14 Apr 2024 05:42:43 +0000 Subject: btrfs: set start on clone before calling copy_extent_buffer_full Our subpage testing started hanging on generic/560 and I bisected it down to 1cab1375ba6d ("btrfs: reuse cloned extent buffer during fiemap to avoid re-allocations"). This is subtle because we use eb->start to figure out where in the folio we're copying to when we're subpage, as our ->start may refer to an area inside of the folio. For example, assume a 16K page size machine with a 4K node size, and assume that we already have a cloned extent buffer when we cloned the previous search. copy_extent_buffer_full() will do the following when copying the extent buffer path->nodes[0] (src) into cloned (dest): src->start = 8k; // this is the new leaf we're cloning cloned->start = 4k; // this is left over from the previous clone src_addr = folio_address(src->folios[0]); dest_addr = folio_address(dest->folios[0]); memcpy(dest_addr + get_eb_offset_in_folio(dst, 0), src_addr + get_eb_offset_in_folio(src, 0), src->len); Now get_eb_offset_in_folio() is where the problems occur, because for sub-pagesize blocksize we can have multiple eb's per folio, the code for this is as follows size_t get_eb_offset_in_folio(eb, offset) { return (eb->start + offset & (folio_size(eb->folio[0]) - 1)); } So in the above example we are copying into offset 4K inside the folio. However once we update cloned->start to 8K to match the src the math for get_eb_offset_in_folio() changes, and any subsequent reads (i.e. btrfs_item_key_to_cpu()) will start reading from the offset 8K instead of 4K where we copied to, giving us garbage. Fix this by setting start before we co copy_extent_buffer_full() to make sure that we're copying into the same offset inside of the folio that we will read from later. All other sites of copy_extent_buffer_full() are correct because we either set ->start beforehand or we simply don't change it in the case of the tree-log usage. With this fix we now pass generic/560 on our subpage tests. Fixes: 1cab1375ba6d ("btrfs: reuse cloned extent buffer during fiemap to avoid re-allocations") Reviewed-by: Filipe Manana Reviewed-by: Qu Wenruo Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index d90330f26827..91122817f137 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2803,13 +2803,19 @@ static int fiemap_next_leaf_item(struct btrfs_inode *inode, struct btrfs_path *p goto out; } - /* See the comment at fiemap_search_slot() about why we clone. */ - copy_extent_buffer_full(clone, path->nodes[0]); /* * Important to preserve the start field, for the optimizations when * checking if extents are shared (see extent_fiemap()). + * + * We must set ->start before calling copy_extent_buffer_full(). If we + * are on sub-pagesize blocksize, we use ->start to determine the offset + * into the folio where our eb exists, and if we update ->start after + * the fact then any subsequent reads of the eb may read from a + * different offset in the folio than where we originally copied into. */ clone->start = path->nodes[0]->start; + /* See the comment at fiemap_search_slot() about why we clone. */ + copy_extent_buffer_full(clone, path->nodes[0]); slot = path->slots[0]; btrfs_release_path(path); -- cgit v1.2.3-59-g8ed1b From e094f48040cda67622060857ad43dad9e418fbd7 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 15 Apr 2024 16:16:23 -0400 Subject: btrfs: change root->root_key.objectid to btrfs_root_id() A comment from Filipe on one of my previous cleanups brought my attention to a new helper we have for getting the root id of a root, which makes it easier to read in the code. The changes where made with the following Coccinelle semantic patch: // @@ expression E,E1; @@ ( E->root_key.objectid = E1 | - E->root_key.objectid + btrfs_root_id(E) ) // Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba [ minor style fixups ] Signed-off-by: David Sterba --- fs/btrfs/backref.c | 8 +++--- fs/btrfs/block-rsv.c | 11 ++++----- fs/btrfs/compression.c | 2 +- fs/btrfs/ctree.c | 39 ++++++++++++++---------------- fs/btrfs/defrag.c | 2 +- fs/btrfs/delayed-inode.c | 2 +- fs/btrfs/disk-io.c | 36 +++++++++++++-------------- fs/btrfs/export.c | 8 +++--- fs/btrfs/extent-tree.c | 56 +++++++++++++++++++----------------------- fs/btrfs/file-item.c | 11 ++++----- fs/btrfs/file.c | 22 ++++++++--------- fs/btrfs/inode-item.c | 4 +-- fs/btrfs/inode.c | 61 ++++++++++++++++++++++------------------------ fs/btrfs/ioctl.c | 20 +++++++-------- fs/btrfs/locking.c | 2 +- fs/btrfs/ordered-data.c | 2 +- fs/btrfs/props.c | 2 +- fs/btrfs/qgroup.c | 27 ++++++++++----------- fs/btrfs/reflink.c | 2 +- fs/btrfs/relocation.c | 63 +++++++++++++++++++++++------------------------- fs/btrfs/root-tree.c | 3 +-- fs/btrfs/send.c | 29 +++++++++++----------- fs/btrfs/super.c | 13 ++++------ fs/btrfs/transaction.c | 22 ++++++++--------- fs/btrfs/tree-log.c | 10 ++++---- fs/btrfs/tree-mod-log.c | 2 +- 26 files changed, 218 insertions(+), 241 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 5bb71dd1ab2a..a2de5c05f97c 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -261,7 +261,7 @@ static void update_share_count(struct share_check *sc, int oldcount, else if (oldcount < 1 && newcount > 0) sc->share_count++; - if (newref->root_id == sc->root->root_key.objectid && + if (newref->root_id == btrfs_root_id(sc->root) && newref->wanted_disk_byte == sc->data_bytenr && newref->key_for_search.objectid == sc->inum) sc->self_ref_count += newref->count; @@ -769,7 +769,7 @@ static int resolve_indirect_refs(struct btrfs_backref_walk_ctx *ctx, continue; } - if (sc && ref->root_id != sc->root->root_key.objectid) { + if (sc && ref->root_id != btrfs_root_id(sc->root)) { free_pref(ref); ret = BACKREF_FOUND_SHARED; goto out; @@ -2623,7 +2623,7 @@ static int iterate_inode_refs(u64 inum, struct inode_fs_paths *ipath) btrfs_debug(fs_root->fs_info, "following ref at offset %u for inode %llu in tree %llu", cur, found_key.objectid, - fs_root->root_key.objectid); + btrfs_root_id(fs_root)); ret = inode_to_path(parent, name_len, (unsigned long)(iref + 1), eb, ipath); if (ret) @@ -3355,7 +3355,7 @@ static int handle_indirect_tree_backref(struct btrfs_trans_handle *trans, if (btrfs_node_blockptr(eb, path->slots[level]) != cur->bytenr) { btrfs_err(fs_info, "couldn't find block (%llu) (level %d) in tree (%llu) with key (%llu %u %llu)", - cur->bytenr, level - 1, root->root_key.objectid, + cur->bytenr, level - 1, btrfs_root_id(root), tree_key->objectid, tree_key->type, tree_key->offset); btrfs_put_root(root); ret = -ENOENT; diff --git a/fs/btrfs/block-rsv.c b/fs/btrfs/block-rsv.c index 95c174f9fd4f..b299b82d676e 100644 --- a/fs/btrfs/block-rsv.c +++ b/fs/btrfs/block-rsv.c @@ -341,9 +341,9 @@ void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info) read_lock(&fs_info->global_root_lock); rbtree_postorder_for_each_entry_safe(root, tmp, &fs_info->global_root_tree, rb_node) { - if (root->root_key.objectid == BTRFS_EXTENT_TREE_OBJECTID || - root->root_key.objectid == BTRFS_CSUM_TREE_OBJECTID || - root->root_key.objectid == BTRFS_FREE_SPACE_TREE_OBJECTID) { + if (btrfs_root_id(root) == BTRFS_EXTENT_TREE_OBJECTID || + btrfs_root_id(root) == BTRFS_CSUM_TREE_OBJECTID || + btrfs_root_id(root) == BTRFS_FREE_SPACE_TREE_OBJECTID) { num_bytes += btrfs_root_used(&root->root_item); min_items++; } @@ -406,7 +406,7 @@ void btrfs_init_root_block_rsv(struct btrfs_root *root) { struct btrfs_fs_info *fs_info = root->fs_info; - switch (root->root_key.objectid) { + switch (btrfs_root_id(root)) { case BTRFS_CSUM_TREE_OBJECTID: case BTRFS_EXTENT_TREE_OBJECTID: case BTRFS_FREE_SPACE_TREE_OBJECTID: @@ -468,8 +468,7 @@ static struct btrfs_block_rsv *get_block_rsv( if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state) || (root == fs_info->uuid_root) || - (trans->adding_csums && - root->root_key.objectid == BTRFS_CSUM_TREE_OBJECTID)) + (trans->adding_csums && btrfs_root_id(root) == BTRFS_CSUM_TREE_OBJECTID)) block_rsv = trans->block_rsv; if (!block_rsv) diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index c981903c8cd7..6441e47d8a5e 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -992,7 +992,7 @@ int btrfs_compress_filemap_get_folio(struct address_space *mapping, u64 start, btrfs_crit(inode->root->fs_info, "failed to get page cache, root %lld ino %llu file offset %llu", - inode->root->root_key.objectid, btrfs_ino(inode), start); + btrfs_root_id(inode->root), btrfs_ino(inode), start); return -ENOENT; } *in_folio_ret = in_folio; diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index f6a98e7cf006..1a49b9232990 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -291,7 +291,7 @@ static void add_root_to_dirty_list(struct btrfs_root *root) spin_lock(&fs_info->trans_lock); if (!test_and_set_bit(BTRFS_ROOT_DIRTY, &root->state)) { /* Want the extent tree to be the last on the list */ - if (root->root_key.objectid == BTRFS_EXTENT_TREE_OBJECTID) + if (btrfs_root_id(root) == BTRFS_EXTENT_TREE_OBJECTID) list_move_tail(&root->dirty_list, &fs_info->dirty_cowonly_roots); else @@ -454,7 +454,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, } } else { refs = 1; - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID || + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID || btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV) flags = BTRFS_BLOCK_FLAG_FULL_BACKREF; else @@ -466,15 +466,14 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, !(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)); if (refs > 1) { - if ((owner == root->root_key.objectid || - root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) && + if ((owner == btrfs_root_id(root) || + btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) && !(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)) { ret = btrfs_inc_ref(trans, root, buf, 1); if (ret) return ret; - if (root->root_key.objectid == - BTRFS_TREE_RELOC_OBJECTID) { + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) { ret = btrfs_dec_ref(trans, root, buf, 0); if (ret) return ret; @@ -485,8 +484,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, new_flags |= BTRFS_BLOCK_FLAG_FULL_BACKREF; } else { - if (root->root_key.objectid == - BTRFS_TREE_RELOC_OBJECTID) + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) ret = btrfs_inc_ref(trans, root, cow, 1); else ret = btrfs_inc_ref(trans, root, cow, 0); @@ -500,8 +498,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, } } else { if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF) { - if (root->root_key.objectid == - BTRFS_TREE_RELOC_OBJECTID) + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) ret = btrfs_inc_ref(trans, root, cow, 1); else ret = btrfs_inc_ref(trans, root, cow, 0); @@ -563,13 +560,13 @@ int btrfs_force_cow_block(struct btrfs_trans_handle *trans, else btrfs_node_key(buf, &disk_key, 0); - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) { + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) { if (parent) parent_start = parent->start; reloc_src_root = btrfs_header_owner(buf); } cow = btrfs_alloc_tree_block(trans, root, parent_start, - root->root_key.objectid, &disk_key, level, + btrfs_root_id(root), &disk_key, level, search_start, empty_size, reloc_src_root, nest); if (IS_ERR(cow)) return PTR_ERR(cow); @@ -582,10 +579,10 @@ int btrfs_force_cow_block(struct btrfs_trans_handle *trans, btrfs_set_header_backref_rev(cow, BTRFS_MIXED_BACKREF_REV); btrfs_clear_header_flag(cow, BTRFS_HEADER_FLAG_WRITTEN | BTRFS_HEADER_FLAG_RELOC); - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) btrfs_set_header_flag(cow, BTRFS_HEADER_FLAG_RELOC); else - btrfs_set_header_owner(cow, root->root_key.objectid); + btrfs_set_header_owner(cow, btrfs_root_id(root)); write_extent_buffer_fsid(cow, fs_info->fs_devices->metadata_uuid); @@ -609,7 +606,7 @@ int btrfs_force_cow_block(struct btrfs_trans_handle *trans, if (buf == root->node) { WARN_ON(parent && parent != buf); - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID || + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID || btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV) parent_start = buf->start; @@ -685,7 +682,7 @@ static inline int should_cow_block(struct btrfs_trans_handle *trans, */ if (btrfs_header_generation(buf) == trans->transid && !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN) && - !(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID && + !(btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID && btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)) && !test_bit(BTRFS_ROOT_FORCE_COW, &root->state)) return 0; @@ -1511,7 +1508,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, check.has_first_key = true; check.level = parent_level - 1; check.transid = gen; - check.owner_root = root->root_key.objectid; + check.owner_root = btrfs_root_id(root); /* * If we need to read an extent buffer from disk and we are holding locks @@ -1556,7 +1553,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, btrfs_release_path(p); return -EIO; } - if (btrfs_check_eb_owner(tmp, root->root_key.objectid)) { + if (btrfs_check_eb_owner(tmp, btrfs_root_id(root))) { free_extent_buffer(tmp); btrfs_release_path(p); return -EUCLEAN; @@ -2865,7 +2862,7 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans, else btrfs_node_key(lower, &lower_key, 0); - c = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid, + c = btrfs_alloc_tree_block(trans, root, 0, btrfs_root_id(root), &lower_key, level, root->node->start, 0, 0, BTRFS_NESTING_NEW_ROOT); if (IS_ERR(c)) @@ -3009,7 +3006,7 @@ static noinline int split_node(struct btrfs_trans_handle *trans, mid = (c_nritems + 1) / 2; btrfs_node_key(c, &disk_key, mid); - split = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid, + split = btrfs_alloc_tree_block(trans, root, 0, btrfs_root_id(root), &disk_key, level, c->start, 0, 0, BTRFS_NESTING_SPLIT); if (IS_ERR(split)) @@ -3761,7 +3758,7 @@ again: * BTRFS_NESTING_SPLIT_THE_SPLITTENING if we need to, but for now just * use BTRFS_NESTING_NEW_ROOT. */ - right = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid, + right = btrfs_alloc_tree_block(trans, root, 0, btrfs_root_id(root), &disk_key, 0, l->start, 0, 0, num_doubles ? BTRFS_NESTING_NEW_ROOT : BTRFS_NESTING_SPLIT); diff --git a/fs/btrfs/defrag.c b/fs/btrfs/defrag.c index f015fa1b6301..407ccec3e57e 100644 --- a/fs/btrfs/defrag.c +++ b/fs/btrfs/defrag.c @@ -147,7 +147,7 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans, defrag->ino = btrfs_ino(inode); defrag->transid = transid; - defrag->root = root->root_key.objectid; + defrag->root = btrfs_root_id(root); defrag->extent_thresh = extent_thresh; spin_lock(&fs_info->defrag_inodes_lock); diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 121ab890bd05..95a0497fa866 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1651,7 +1651,7 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans, if (unlikely(ret)) { btrfs_err(trans->fs_info, "err add delayed dir index item(index: %llu) into the deletion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)", - index, node->root->root_key.objectid, + index, btrfs_root_id(node->root), node->inode_id, ret); btrfs_delayed_item_release_metadata(dir->root, item); btrfs_release_delayed_item(item); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 0474e9b6d302..cee2e4f43dc6 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -776,7 +776,7 @@ int btrfs_global_root_insert(struct btrfs_root *root) if (tmp) { ret = -EEXIST; btrfs_warn(fs_info, "global root %llu %llu already exists", - root->root_key.objectid, root->root_key.offset); + btrfs_root_id(root), root->root_key.offset); } return ret; } @@ -1012,7 +1012,7 @@ int btrfs_add_log_tree(struct btrfs_trans_handle *trans, } log_root->last_trans = trans->transid; - log_root->root_key.offset = root->root_key.objectid; + log_root->root_key.offset = btrfs_root_id(root); inode_item = &log_root->root_item.inode; btrfs_set_stack_inode_generation(inode_item, 1); @@ -1077,14 +1077,14 @@ static struct btrfs_root *read_tree_root_path(struct btrfs_root *tree_root, * match its root node owner */ if (!test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state) && - root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID && - root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID && - root->root_key.objectid != btrfs_header_owner(root->node)) { + btrfs_root_id(root) != BTRFS_TREE_LOG_OBJECTID && + btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID && + btrfs_root_id(root) != btrfs_header_owner(root->node)) { btrfs_crit(fs_info, "root=%llu block=%llu, tree root owner mismatch, have %llu expect %llu", - root->root_key.objectid, root->node->start, + btrfs_root_id(root), root->node->start, btrfs_header_owner(root->node), - root->root_key.objectid); + btrfs_root_id(root)); ret = -EUCLEAN; goto fail; } @@ -1121,9 +1121,9 @@ static int btrfs_init_fs_root(struct btrfs_root *root, dev_t anon_dev) btrfs_drew_lock_init(&root->snapshot_lock); - if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID && + if (btrfs_root_id(root) != BTRFS_TREE_LOG_OBJECTID && !btrfs_is_data_reloc_root(root) && - is_fstree(root->root_key.objectid)) { + is_fstree(btrfs_root_id(root))) { set_bit(BTRFS_ROOT_SHAREABLE, &root->state); btrfs_check_and_init_root_item(&root->root_item); } @@ -1132,7 +1132,7 @@ static int btrfs_init_fs_root(struct btrfs_root *root, dev_t anon_dev) * Don't assign anonymous block device to roots that are not exposed to * userspace, the id pool is limited to 1M */ - if (is_fstree(root->root_key.objectid) && + if (is_fstree(btrfs_root_id(root)) && btrfs_root_refs(&root->root_item) > 0) { if (!anon_dev) { ret = get_anon_bdev(&root->anon_dev); @@ -1219,7 +1219,7 @@ int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info, spin_lock(&fs_info->fs_roots_radix_lock); ret = radix_tree_insert(&fs_info->fs_roots_radix, - (unsigned long)root->root_key.objectid, + (unsigned long)btrfs_root_id(root), root); if (ret == 0) { btrfs_grab_root(root); @@ -2584,7 +2584,7 @@ static int load_super_root(struct btrfs_root *root, u64 bytenr, u64 gen, int lev struct btrfs_tree_parent_check check = { .level = level, .transid = gen, - .owner_root = root->root_key.objectid + .owner_root = btrfs_root_id(root) }; int ret = 0; @@ -2930,7 +2930,7 @@ static int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info) spin_unlock(&fs_info->fs_roots_radix_lock); break; } - root_objectid = gang[ret - 1]->root_key.objectid + 1; + root_objectid = btrfs_root_id(gang[ret - 1]) + 1; for (i = 0; i < ret; i++) { /* Avoid to grab roots in dead_roots. */ @@ -2946,7 +2946,7 @@ static int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info) for (i = 0; i < ret; i++) { if (!gang[i]) continue; - root_objectid = gang[i]->root_key.objectid; + root_objectid = btrfs_root_id(gang[i]); err = btrfs_orphan_cleanup(gang[i]); if (err) goto out; @@ -4139,7 +4139,7 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info, spin_lock(&fs_info->fs_roots_radix_lock); radix_tree_delete(&fs_info->fs_roots_radix, - (unsigned long)root->root_key.objectid); + (unsigned long)btrfs_root_id(root)); if (test_and_clear_bit(BTRFS_ROOT_IN_RADIX, &root->state)) drop_ref = true; spin_unlock(&fs_info->fs_roots_radix_lock); @@ -4481,7 +4481,7 @@ static void btrfs_drop_all_logs(struct btrfs_fs_info *fs_info) for (i = 0; i < ret; i++) { if (!gang[i]) continue; - root_objectid = gang[i]->root_key.objectid; + root_objectid = btrfs_root_id(gang[i]); btrfs_free_log(NULL, gang[i]); btrfs_put_root(gang[i]); } @@ -4812,7 +4812,7 @@ static void btrfs_free_all_qgroup_pertrans(struct btrfs_fs_info *fs_info) btrfs_qgroup_free_meta_all_pertrans(root); radix_tree_tag_clear(&fs_info->fs_roots_radix, - (unsigned long)root->root_key.objectid, + (unsigned long)btrfs_root_id(root), BTRFS_ROOT_TRANS_TAG); } } @@ -4953,7 +4953,7 @@ int btrfs_get_free_objectid(struct btrfs_root *root, u64 *objectid) if (unlikely(root->free_objectid >= BTRFS_LAST_FREE_OBJECTID)) { btrfs_warn(root->fs_info, "the objectid of root %llu reaches its highest value", - root->root_key.objectid); + btrfs_root_id(root)); ret = -ENOSPC; goto out; } diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c index 8398d345ec5b..9e81f89e76d8 100644 --- a/fs/btrfs/export.c +++ b/fs/btrfs/export.c @@ -34,7 +34,7 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len, type = FILEID_BTRFS_WITHOUT_PARENT; fid->objectid = btrfs_ino(BTRFS_I(inode)); - fid->root_objectid = BTRFS_I(inode)->root->root_key.objectid; + fid->root_objectid = btrfs_root_id(BTRFS_I(inode)->root); fid->gen = inode->i_generation; if (parent) { @@ -42,7 +42,7 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len, fid->parent_objectid = BTRFS_I(parent)->location.objectid; fid->parent_gen = parent->i_generation; - parent_root_id = BTRFS_I(parent)->root->root_key.objectid; + parent_root_id = btrfs_root_id(BTRFS_I(parent)->root); if (parent_root_id != fid->root_objectid) { fid->parent_root_objectid = parent_root_id; @@ -160,7 +160,7 @@ struct dentry *btrfs_get_parent(struct dentry *child) return ERR_PTR(-ENOMEM); if (btrfs_ino(BTRFS_I(dir)) == BTRFS_FIRST_FREE_OBJECTID) { - key.objectid = root->root_key.objectid; + key.objectid = btrfs_root_id(root); key.type = BTRFS_ROOT_BACKREF_KEY; key.offset = (u64)-1; root = fs_info->tree_root; @@ -243,7 +243,7 @@ static int btrfs_get_name(struct dentry *parent, char *name, return -ENOMEM; if (ino == BTRFS_FIRST_FREE_OBJECTID) { - key.objectid = BTRFS_I(inode)->root->root_key.objectid; + key.objectid = btrfs_root_id(BTRFS_I(inode)->root); key.type = BTRFS_ROOT_BACKREF_KEY; key.offset = (u64)-1; root = fs_info->tree_root; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 87a921b84bab..023920d0d971 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2328,7 +2328,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, * If our ref doesn't match the one we're currently looking at * then we have a cross reference. */ - if (ref->ref_root != root->root_key.objectid || + if (ref->ref_root != btrfs_root_id(root) || ref_owner != objectid || ref_offset != offset) { ret = 1; break; @@ -2422,8 +2422,7 @@ static noinline int check_committed_ref(struct btrfs_root *root, ref = (struct btrfs_extent_data_ref *)(&iref->offset); if (btrfs_extent_refs(leaf, ei) != btrfs_extent_data_ref_count(leaf, ref) || - btrfs_extent_data_ref_root(leaf, ref) != - root->root_key.objectid || + btrfs_extent_data_ref_root(leaf, ref) != btrfs_root_id(root) || btrfs_extent_data_ref_objectid(leaf, ref) != objectid || btrfs_extent_data_ref_offset(leaf, ref) != offset) goto out; @@ -2515,7 +2514,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, key.offset -= btrfs_file_extent_offset(buf, fi); btrfs_init_data_ref(&ref, key.objectid, key.offset, - root->root_key.objectid, for_reloc); + btrfs_root_id(root), for_reloc); if (inc) ret = btrfs_inc_extent_ref(trans, &ref); else @@ -2528,7 +2527,7 @@ static int __btrfs_mod_ref(struct btrfs_trans_handle *trans, ref.num_bytes = fs_info->nodesize; btrfs_init_tree_ref(&ref, level - 1, - root->root_key.objectid, for_reloc); + btrfs_root_id(root), for_reloc); if (inc) ret = btrfs_inc_extent_ref(trans, &ref); else @@ -4671,7 +4670,7 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes, bool final_tried = num_bytes == min_alloc_size; u64 flags; int ret; - bool for_treelog = (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID); + bool for_treelog = (btrfs_root_id(root) == BTRFS_TREE_LOG_OBJECTID); bool for_data_reloc = (btrfs_is_data_reloc_root(root) && is_data); flags = get_alloc_profile_by_root(root, is_data); @@ -4936,8 +4935,8 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans, .action = BTRFS_ADD_DELAYED_EXTENT, .bytenr = ins->objectid, .num_bytes = ins->offset, - .owning_root = root->root_key.objectid, - .ref_root = root->root_key.objectid, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), }; ASSERT(generic_ref.ref_root != BTRFS_TREE_LOG_OBJECTID); @@ -5083,7 +5082,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root, btrfs_set_header_owner(buf, owner); write_extent_buffer_fsid(buf, fs_info->fs_devices->metadata_uuid); write_extent_buffer_chunk_tree_uuid(buf, fs_info->chunk_tree_uuid); - if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) { + if (btrfs_root_id(root) == BTRFS_TREE_LOG_OBJECTID) { buf->log_index = root->log_transid % 2; /* * we allow two log transactions at a time, use different @@ -5188,8 +5187,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, extent_op->update_flags = true; extent_op->level = level; - btrfs_init_tree_ref(&generic_ref, level, - root->root_key.objectid, false); + btrfs_init_tree_ref(&generic_ref, level, btrfs_root_id(root), false); btrfs_ref_tree_mod(fs_info, &generic_ref); ret = btrfs_add_delayed_tree_ref(trans, &generic_ref, extent_op); if (ret) @@ -5327,8 +5325,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans, u64 flag = BTRFS_BLOCK_FLAG_FULL_BACKREF; int ret; - if (wc->stage == UPDATE_BACKREF && - btrfs_header_owner(eb) != root->root_key.objectid) + if (wc->stage == UPDATE_BACKREF && btrfs_header_owner(eb) != btrfs_root_id(root)) return 1; /* @@ -5402,7 +5399,7 @@ static int check_ref_exists(struct btrfs_trans_handle *trans, ret = lookup_extent_backref(trans, path, &iref, bytenr, root->fs_info->nodesize, parent, - root->root_key.objectid, level, 0); + btrfs_root_id(root), level, 0); btrfs_free_path(path); if (ret == -ENOENT) return 0; @@ -5458,7 +5455,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, check.level = level - 1; check.transid = generation; - check.owner_root = root->root_key.objectid; + check.owner_root = btrfs_root_id(root); check.has_first_key = true; btrfs_node_key_to_cpu(path->nodes[level], &check.first_key, path->slots[level]); @@ -5466,7 +5463,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, next = find_extent_buffer(fs_info, bytenr); if (!next) { next = btrfs_find_create_tree_block(fs_info, bytenr, - root->root_key.objectid, level - 1); + btrfs_root_id(root), level - 1); if (IS_ERR(next)) return PTR_ERR(next); reada = 1; @@ -5556,14 +5553,14 @@ skip: .bytenr = bytenr, .num_bytes = fs_info->nodesize, .owning_root = owner_root, - .ref_root = root->root_key.objectid, + .ref_root = btrfs_root_id(root), }; if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) { ref.parent = path->nodes[level]->start; } else { - ASSERT(root->root_key.objectid == + ASSERT(btrfs_root_id(root) == btrfs_header_owner(path->nodes[level])); - if (root->root_key.objectid != + if (btrfs_root_id(root) != btrfs_header_owner(path->nodes[level])) { btrfs_err(root->fs_info, "mismatched block owner"); @@ -5594,8 +5591,7 @@ skip: * already accounted them at merge time (replace_path), * thus we could skip expensive subtree trace here. */ - if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID && - need_account) { + if (btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID && need_account) { ret = btrfs_qgroup_trace_subtree(trans, next, generation, level - 1); if (ret) { @@ -5705,7 +5701,7 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans, else ret = btrfs_dec_ref(trans, root, eb, 0); BUG_ON(ret); /* -ENOMEM */ - if (is_fstree(root->root_key.objectid)) { + if (is_fstree(btrfs_root_id(root))) { ret = btrfs_qgroup_trace_leaf_items(trans, eb); if (ret) { btrfs_err_rl(fs_info, @@ -5725,12 +5721,12 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans, if (eb == root->node) { if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) parent = eb->start; - else if (root->root_key.objectid != btrfs_header_owner(eb)) + else if (btrfs_root_id(root) != btrfs_header_owner(eb)) goto owner_mismatch; } else { if (wc->flags[level + 1] & BTRFS_BLOCK_FLAG_FULL_BACKREF) parent = path->nodes[level + 1]->start; - else if (root->root_key.objectid != + else if (btrfs_root_id(root) != btrfs_header_owner(path->nodes[level + 1])) goto owner_mismatch; } @@ -5744,7 +5740,7 @@ out: owner_mismatch: btrfs_err_rl(fs_info, "unexpected tree owner, have %llu expect %llu", - btrfs_header_owner(eb), root->root_key.objectid); + btrfs_header_owner(eb), btrfs_root_id(root)); return -EUCLEAN; } @@ -5830,8 +5826,7 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans, */ int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref, int for_reloc) { - const bool is_reloc_root = (root->root_key.objectid == - BTRFS_TREE_RELOC_OBJECTID); + const bool is_reloc_root = (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID); struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; struct btrfs_trans_handle *trans; @@ -5845,7 +5840,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref, int for_reloc) bool root_dropped = false; bool unfinished_drop = false; - btrfs_debug(fs_info, "Drop subvolume %llu", root->root_key.objectid); + btrfs_debug(fs_info, "Drop subvolume %llu", btrfs_root_id(root)); path = btrfs_alloc_path(); if (!path) { @@ -6043,8 +6038,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref, int for_reloc) * * The most common failure here is just -ENOENT. */ - btrfs_del_orphan_item(trans, tree_root, - root->root_key.objectid); + btrfs_del_orphan_item(trans, tree_root, btrfs_root_id(root)); } } @@ -6108,7 +6102,7 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans, int ret = 0; int wret; - BUG_ON(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID); + BUG_ON(btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID); path = btrfs_alloc_path(); if (!path) diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 1ea1ed44fe42..5fead7a253df 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -430,8 +430,7 @@ blk_status_t btrfs_lookup_bio_sums(struct btrfs_bio *bbio) memset(csum_dst, 0, csum_size); count = 1; - if (inode->root->root_key.objectid == - BTRFS_DATA_RELOC_TREE_OBJECTID) { + if (btrfs_root_id(inode->root) == BTRFS_DATA_RELOC_TREE_OBJECTID) { u64 file_offset = bbio->file_offset + bio_offset; set_extent_bit(&inode->io_tree, file_offset, @@ -885,8 +884,8 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans, const u32 csum_size = fs_info->csum_size; u32 blocksize_bits = fs_info->sectorsize_bits; - ASSERT(root->root_key.objectid == BTRFS_CSUM_TREE_OBJECTID || - root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID); + ASSERT(btrfs_root_id(root) == BTRFS_CSUM_TREE_OBJECTID || + btrfs_root_id(root) == BTRFS_TREE_LOG_OBJECTID); path = btrfs_alloc_path(); if (!path) @@ -1186,7 +1185,7 @@ extend_csum: * search, etc, because log trees are temporary anyway and it * would only save a few bytes of leaf space. */ - if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) { + if (btrfs_root_id(root) == BTRFS_TREE_LOG_OBJECTID) { if (path->slots[0] + 1 >= btrfs_header_nritems(path->nodes[0])) { ret = find_next_csum_offset(root, path, &next_offset); @@ -1328,7 +1327,7 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode, btrfs_err(fs_info, "unknown file extent item type %d, inode %llu, offset %llu, " "root %llu", type, btrfs_ino(inode), extent_start, - root->root_key.objectid); + btrfs_root_id(root)); } } diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 065658f2c061..0c7c1b42028e 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -245,7 +245,7 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, if (args->start >= inode->disk_i_size && !args->replace_extent) modify_tree = 0; - update_refs = (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID); + update_refs = (btrfs_root_id(root) != BTRFS_TREE_LOG_OBJECTID); while (1) { recow = 0; ret = btrfs_lookup_file_extent(trans, root, path, ino, @@ -377,8 +377,8 @@ next_slot: .bytenr = disk_bytenr, .num_bytes = num_bytes, .parent = 0, - .owning_root = root->root_key.objectid, - .ref_root = root->root_key.objectid, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), }; btrfs_init_data_ref(&ref, new_key.objectid, args->start - extent_offset, @@ -470,8 +470,8 @@ delete_extent_item: .bytenr = disk_bytenr, .num_bytes = num_bytes, .parent = 0, - .owning_root = root->root_key.objectid, - .ref_root = root->root_key.objectid, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), }; btrfs_init_data_ref(&ref, key.objectid, key.offset - extent_offset, @@ -755,8 +755,8 @@ again: ref.bytenr = bytenr; ref.num_bytes = num_bytes; ref.parent = 0; - ref.owning_root = root->root_key.objectid; - ref.ref_root = root->root_key.objectid; + ref.owning_root = btrfs_root_id(root); + ref.ref_root = btrfs_root_id(root); btrfs_init_data_ref(&ref, ino, orig_offset, 0, false); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { @@ -785,8 +785,8 @@ again: ref.bytenr = bytenr; ref.num_bytes = num_bytes; ref.parent = 0; - ref.owning_root = root->root_key.objectid; - ref.ref_root = root->root_key.objectid; + ref.owning_root = btrfs_root_id(root); + ref.ref_root = btrfs_root_id(root); btrfs_init_data_ref(&ref, ino, orig_offset, 0, false); if (extent_mergeable(leaf, path->slots[0] + 1, ino, bytenr, orig_offset, @@ -2493,8 +2493,8 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, .action = BTRFS_ADD_DELAYED_REF, .bytenr = extent_info->disk_offset, .num_bytes = extent_info->disk_len, - .owning_root = root->root_key.objectid, - .ref_root = root->root_key.objectid, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), }; u64 ref_offset; diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c index 7565ff15a69a..84a94d19b22c 100644 --- a/fs/btrfs/inode-item.c +++ b/fs/btrfs/inode-item.c @@ -674,14 +674,14 @@ delete: .action = BTRFS_DROP_DELAYED_REF, .bytenr = extent_start, .num_bytes = extent_num_bytes, - .owning_root = root->root_key.objectid, + .owning_root = btrfs_root_id(root), .ref_root = btrfs_header_owner(leaf), }; bytes_deleted += extent_num_bytes; btrfs_init_data_ref(&ref, control->ino, extent_offset, - root->root_key.objectid, false); + btrfs_root_id(root), false); ret = btrfs_free_extent(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 629706b98e31..1dde8085271e 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -254,7 +254,7 @@ static void print_data_reloc_error(const struct btrfs_inode *inode, u64 file_off btrfs_warn_rl(fs_info, "has data reloc tree but no running relocation"); btrfs_warn_rl(fs_info, "csum failed root %lld ino %llu off %llu csum " CSUM_FMT " expected csum " CSUM_FMT " mirror %d", - inode->root->root_key.objectid, btrfs_ino(inode), file_off, + btrfs_root_id(inode->root), btrfs_ino(inode), file_off, CSUM_FMT_VALUE(csum_size, csum), CSUM_FMT_VALUE(csum_size, csum_expected), mirror_num); @@ -264,7 +264,7 @@ static void print_data_reloc_error(const struct btrfs_inode *inode, u64 file_off logical += file_off; btrfs_warn_rl(fs_info, "csum failed root %lld ino %llu off %llu logical %llu csum " CSUM_FMT " expected csum " CSUM_FMT " mirror %d", - inode->root->root_key.objectid, + btrfs_root_id(inode->root), btrfs_ino(inode), file_off, logical, CSUM_FMT_VALUE(csum_size, csum), CSUM_FMT_VALUE(csum_size, csum_expected), @@ -331,15 +331,15 @@ static void __cold btrfs_print_data_csum_error(struct btrfs_inode *inode, const u32 csum_size = root->fs_info->csum_size; /* For data reloc tree, it's better to do a backref lookup instead. */ - if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID) + if (btrfs_root_id(root) == BTRFS_DATA_RELOC_TREE_OBJECTID) return print_data_reloc_error(inode, logical_start, csum, csum_expected, mirror_num); /* Output without objectid, which is more meaningful */ - if (root->root_key.objectid >= BTRFS_LAST_FREE_OBJECTID) { + if (btrfs_root_id(root) >= BTRFS_LAST_FREE_OBJECTID) { btrfs_warn_rl(root->fs_info, "csum failed root %lld ino %lld off %llu csum " CSUM_FMT " expected csum " CSUM_FMT " mirror %d", - root->root_key.objectid, btrfs_ino(inode), + btrfs_root_id(root), btrfs_ino(inode), logical_start, CSUM_FMT_VALUE(csum_size, csum), CSUM_FMT_VALUE(csum_size, csum_expected), @@ -347,7 +347,7 @@ static void __cold btrfs_print_data_csum_error(struct btrfs_inode *inode, } else { btrfs_warn_rl(root->fs_info, "csum failed root %llu ino %llu off %llu csum " CSUM_FMT " expected csum " CSUM_FMT " mirror %d", - root->root_key.objectid, btrfs_ino(inode), + btrfs_root_id(root), btrfs_ino(inode), logical_start, CSUM_FMT_VALUE(csum_size, csum), CSUM_FMT_VALUE(csum_size, csum_expected), @@ -1218,7 +1218,7 @@ out_free_reserve: kthread_associate_blkcg(NULL); btrfs_debug(fs_info, "async extent submission failed root=%lld inode=%llu start=%llu len=%llu ret=%d", - root->root_key.objectid, btrfs_ino(inode), start, + btrfs_root_id(root), btrfs_ino(inode), start, async_extent->ram_size, ret); kfree(async_extent); } @@ -3239,7 +3239,7 @@ out: * Actually free the qgroup rsv which was released when * the ordered extent was created. */ - btrfs_qgroup_free_refroot(fs_info, inode->root->root_key.objectid, + btrfs_qgroup_free_refroot(fs_info, btrfs_root_id(inode->root), ordered_extent->qgroup_rsv, BTRFS_QGROUP_RSV_DATA); } @@ -3906,7 +3906,7 @@ cache_acl: btrfs_err(fs_info, "error loading props for ino %llu (root %llu): %d", btrfs_ino(BTRFS_I(inode)), - root->root_key.objectid, ret); + btrfs_root_id(root), ret); } if (path != in_path) btrfs_free_path(path); @@ -4265,7 +4265,7 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, /* This needs to handle no-key deletions later on */ if (btrfs_ino(inode) == BTRFS_FIRST_FREE_OBJECTID) { - objectid = inode->root->root_key.objectid; + objectid = btrfs_root_id(inode->root); } else if (btrfs_ino(inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) { objectid = inode->location.objectid; } else { @@ -4323,7 +4323,7 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, btrfs_release_path(path); } else { ret = btrfs_del_root_ref(trans, objectid, - root->root_key.objectid, dir_ino, + btrfs_root_id(root), dir_ino, &index, &fname.disk_name); if (ret) { btrfs_abort_transaction(trans, ret); @@ -4373,7 +4373,7 @@ static noinline int may_destroy_subvol(struct btrfs_root *root) dir_id, &name, 0); if (di && !IS_ERR(di)) { btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key); - if (key.objectid == root->root_key.objectid) { + if (key.objectid == btrfs_root_id(root)) { ret = -EPERM; btrfs_err(fs_info, "deleting default subvolume %llu is not allowed", @@ -4383,7 +4383,7 @@ static noinline int may_destroy_subvol(struct btrfs_root *root) btrfs_release_path(path); } - key.objectid = root->root_key.objectid; + key.objectid = btrfs_root_id(root); key.type = BTRFS_ROOT_REF_KEY; key.offset = (u64)-1; @@ -4403,8 +4403,7 @@ static noinline int may_destroy_subvol(struct btrfs_root *root) if (path->slots[0] > 0) { path->slots[0]--; btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); - if (key.objectid == root->root_key.objectid && - key.type == BTRFS_ROOT_REF_KEY) + if (key.objectid == btrfs_root_id(root) && key.type == BTRFS_ROOT_REF_KEY) ret = -ENOTEMPTY; } out: @@ -4462,7 +4461,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) spin_unlock(&dest->root_item_lock); btrfs_warn(fs_info, "attempt to delete subvolume %llu during send", - dest->root_key.objectid); + btrfs_root_id(dest)); ret = -EPERM; goto out_up_write; } @@ -4470,7 +4469,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) spin_unlock(&dest->root_item_lock); btrfs_warn(fs_info, "attempt to delete subvolume %llu with active swapfile", - root->root_key.objectid); + btrfs_root_id(root)); ret = -EPERM; goto out_up_write; } @@ -4531,7 +4530,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) if (!test_and_set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &dest->state)) { ret = btrfs_insert_orphan_item(trans, fs_info->tree_root, - dest->root_key.objectid); + btrfs_root_id(dest)); if (ret) { btrfs_abort_transaction(trans, ret); goto out_end_trans; @@ -4539,8 +4538,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) } ret = btrfs_uuid_tree_remove(trans, dest->root_item.uuid, - BTRFS_UUID_KEY_SUBVOL, - dest->root_key.objectid); + BTRFS_UUID_KEY_SUBVOL, btrfs_root_id(dest)); if (ret && ret != -ENOENT) { btrfs_abort_transaction(trans, ret); goto out_end_trans; @@ -4549,7 +4547,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) ret = btrfs_uuid_tree_remove(trans, dest->root_item.received_uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, - dest->root_key.objectid); + btrfs_root_id(dest)); if (ret && ret != -ENOENT) { btrfs_abort_transaction(trans, ret); goto out_end_trans; @@ -5229,7 +5227,7 @@ void btrfs_evict_inode(struct inode *inode) if (inode->i_nlink && ((btrfs_root_refs(&root->root_item) != 0 && - root->root_key.objectid != BTRFS_ROOT_TREE_OBJECTID) || + btrfs_root_id(root) != BTRFS_ROOT_TREE_OBJECTID) || btrfs_is_free_space_inode(BTRFS_I(inode)))) goto out; @@ -5241,7 +5239,7 @@ void btrfs_evict_inode(struct inode *inode) if (inode->i_nlink > 0) { BUG_ON(btrfs_root_refs(&root->root_item) != 0 && - root->root_key.objectid != BTRFS_ROOT_TREE_OBJECTID); + btrfs_root_id(root) != BTRFS_ROOT_TREE_OBJECTID); goto out; } @@ -5413,7 +5411,7 @@ static int fixup_tree_root_location(struct btrfs_fs_info *fs_info, } err = -ENOENT; - key.objectid = dir->root->root_key.objectid; + key.objectid = btrfs_root_id(dir->root); key.type = BTRFS_ROOT_REF_KEY; key.offset = location->objectid; @@ -6372,8 +6370,7 @@ int btrfs_create_new_inode(struct btrfs_trans_handle *trans, if (ret) { btrfs_err(fs_info, "error inheriting props for ino %llu (root %llu): %d", - btrfs_ino(BTRFS_I(inode)), root->root_key.objectid, - ret); + btrfs_ino(BTRFS_I(inode)), btrfs_root_id(root), ret); } /* @@ -6446,7 +6443,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) { ret = btrfs_add_root_ref(trans, key.objectid, - root->root_key.objectid, parent_ino, + btrfs_root_id(root), parent_ino, index, name); } else if (add_backref) { ret = btrfs_insert_inode_ref(trans, root, name, @@ -6489,7 +6486,7 @@ fail_dir_item: u64 local_index; int err; err = btrfs_del_root_ref(trans, key.objectid, - root->root_key.objectid, parent_ino, + btrfs_root_id(root), parent_ino, &local_index, name); if (err) btrfs_abort_transaction(trans, err); @@ -6587,7 +6584,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, int drop_inode = 0; /* do not allow sys_link's with other subvols of the same device */ - if (root->root_key.objectid != BTRFS_I(inode)->root->root_key.objectid) + if (btrfs_root_id(root) != btrfs_root_id(BTRFS_I(inode)->root)) return -EXDEV; if (inode->i_nlink >= BTRFS_LINK_MAX) @@ -9433,7 +9430,7 @@ free_qgroup: * or we leak qgroup data reservation. */ btrfs_qgroup_free_refroot(inode->root->fs_info, - inode->root->root_key.objectid, qgroup_released, + btrfs_root_id(inode->root), qgroup_released, BTRFS_QGROUP_RSV_DATA); return ERR_PTR(ret); } @@ -10534,7 +10531,7 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, btrfs_exclop_finish(fs_info); btrfs_warn(fs_info, "cannot activate swapfile because subvolume %llu is being deleted", - root->root_key.objectid); + btrfs_root_id(root)); return -EPERM; } atomic_inc(&root->nr_swapfiles); @@ -10760,7 +10757,7 @@ void btrfs_assert_inode_range_clean(struct btrfs_inode *inode, u64 start, u64 en if (ordered) { btrfs_err(root->fs_info, "found unexpected ordered extent in file range [%llu, %llu] for inode %llu root %llu (ordered range [%llu, %llu])", - start, end, btrfs_ino(inode), root->root_key.objectid, + start, end, btrfs_ino(inode), btrfs_root_id(root), ordered->file_offset, ordered->file_offset + ordered->num_bytes - 1); btrfs_put_ordered_extent(ordered); diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 16aa667c4778..efd5d6e9589e 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -668,7 +668,7 @@ static noinline int create_subvol(struct mnt_idmap *idmap, /* Tree log can't currently deal with an inode which is a new root. */ btrfs_set_log_full_commit(trans); - ret = btrfs_qgroup_inherit(trans, 0, objectid, root->root_key.objectid, inherit); + ret = btrfs_qgroup_inherit(trans, 0, objectid, btrfs_root_id(root), inherit); if (ret) goto out; @@ -1510,7 +1510,7 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file, spin_unlock(&root->root_item_lock); btrfs_warn(fs_info, "Attempt to set subvolume %llu read-write during send", - root->root_key.objectid); + btrfs_root_id(root)); ret = -EPERM; goto out_drop_sem; } @@ -1919,7 +1919,7 @@ static int btrfs_search_path_in_tree_user(struct mnt_idmap *idmap, struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; struct super_block *sb = inode->i_sb; struct btrfs_key upper_limit = BTRFS_I(inode)->location; - u64 treeid = BTRFS_I(inode)->root->root_key.objectid; + u64 treeid = btrfs_root_id(BTRFS_I(inode)->root); u64 dirid = args->dirid; unsigned long item_off; unsigned long item_len; @@ -2091,7 +2091,7 @@ static noinline int btrfs_ioctl_ino_lookup(struct btrfs_root *root, * path is reset so it's consistent with btrfs_search_path_in_tree. */ if (args->treeid == 0) - args->treeid = root->root_key.objectid; + args->treeid = btrfs_root_id(root); if (args->objectid == BTRFS_FIRST_FREE_OBJECTID) { args->name[0] = 0; @@ -2187,7 +2187,7 @@ static int btrfs_ioctl_get_subvol_info(struct inode *inode, void __user *argp) fs_info = BTRFS_I(inode)->root->fs_info; /* Get root_item of inode's subvolume */ - key.objectid = BTRFS_I(inode)->root->root_key.objectid; + key.objectid = btrfs_root_id(BTRFS_I(inode)->root); root = btrfs_get_fs_root(fs_info, key.objectid, true); if (IS_ERR(root)) { ret = PTR_ERR(root); @@ -2302,7 +2302,7 @@ static int btrfs_ioctl_get_subvol_rootref(struct btrfs_root *root, return PTR_ERR(rootrefs); } - objectid = root->root_key.objectid; + objectid = btrfs_root_id(root); key.objectid = objectid; key.type = BTRFS_ROOT_REF_KEY; key.offset = rootrefs->min_treeid; @@ -2981,7 +2981,7 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) ret = PTR_ERR(new_root); goto out; } - if (!is_fstree(new_root->root_key.objectid)) { + if (!is_fstree(btrfs_root_id(new_root))) { ret = -ENOENT; goto out_free; } @@ -3947,7 +3947,7 @@ static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg) qgroupid = sa->qgroupid; if (!qgroupid) { /* take the current subvol as qgroup */ - qgroupid = root->root_key.objectid; + qgroupid = btrfs_root_id(root); } ret = btrfs_limit_qgroup(trans, qgroupid, &sa->lim); @@ -4078,7 +4078,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, !btrfs_is_empty_uuid(root_item->received_uuid)) { ret = btrfs_uuid_tree_remove(trans, root_item->received_uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, - root->root_key.objectid); + btrfs_root_id(root)); if (ret && ret != -ENOENT) { btrfs_abort_transaction(trans, ret); btrfs_end_transaction(trans); @@ -4102,7 +4102,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, if (received_uuid_changed && !btrfs_is_empty_uuid(sa->uuid)) { ret = btrfs_uuid_tree_add(trans, sa->uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, - root->root_key.objectid); + btrfs_root_id(root)); if (ret < 0 && ret != -EEXIST) { btrfs_abort_transaction(trans, ret); btrfs_end_transaction(trans); diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 72992e74c479..6a0b7abb5bd9 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -97,7 +97,7 @@ void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb, int void btrfs_maybe_reset_lockdep_class(struct btrfs_root *root, struct extent_buffer *eb) { if (test_bit(BTRFS_ROOT_RESET_LOCKDEP_CLASS, &root->state)) - btrfs_set_buffer_lockdep_class(root->root_key.objectid, + btrfs_set_buffer_lockdep_class(btrfs_root_id(root), eb, btrfs_header_level(eb)); } diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index c2a42bcde98e..3c2921e32843 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -332,7 +332,7 @@ static bool can_finish_ordered_extent(struct btrfs_ordered_extent *ordered, if (WARN_ON_ONCE(len > ordered->bytes_left)) { btrfs_crit(fs_info, "bad ordered extent accounting, root=%llu ino=%llu OE offset=%llu OE len=%llu to_dec=%llu left=%llu", - inode->root->root_key.objectid, btrfs_ino(inode), + btrfs_root_id(inode->root), btrfs_ino(inode), ordered->file_offset, ordered->num_bytes, len, ordered->bytes_left); ordered->bytes_left = 0; diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c index 2a9b7b029eeb..155570e20f45 100644 --- a/fs/btrfs/props.c +++ b/fs/btrfs/props.c @@ -268,7 +268,7 @@ static void inode_prop_iterator(void *ctx, btrfs_warn(root->fs_info, "error applying prop %s to ino %llu (root %llu): %d", handler->xattr_name, btrfs_ino(BTRFS_I(inode)), - root->root_key.objectid, ret); + btrfs_root_id(root), ret); else set_bit(BTRFS_INODE_HAS_PROPS, &BTRFS_I(inode)->runtime_flags); } diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index dbb114be8e79..6488d9b888e4 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3468,7 +3468,7 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes, bool enforce, { struct btrfs_qgroup *qgroup; struct btrfs_fs_info *fs_info = root->fs_info; - u64 ref_root = root->root_key.objectid; + u64 ref_root = btrfs_root_id(root); int ret = 0; LIST_HEAD(qgroup_list); @@ -4109,7 +4109,7 @@ static int qgroup_reserve_data(struct btrfs_inode *inode, int ret; if (btrfs_qgroup_mode(root->fs_info) == BTRFS_QGROUP_MODE_DISABLED || - !is_fstree(root->root_key.objectid) || len == 0) + !is_fstree(btrfs_root_id(root)) || len == 0) return 0; /* @reserved parameter is mandatory for qgroup */ @@ -4225,7 +4225,7 @@ static int qgroup_free_reserved_data(struct btrfs_inode *inode, goto out; freed += changeset.bytes_changed; } - btrfs_qgroup_free_refroot(root->fs_info, root->root_key.objectid, freed, + btrfs_qgroup_free_refroot(root->fs_info, btrfs_root_id(root), freed, BTRFS_QGROUP_RSV_DATA); if (freed_ret) *freed_ret = freed; @@ -4266,7 +4266,7 @@ static int __btrfs_qgroup_release_data(struct btrfs_inode *inode, changeset.bytes_changed, trace_op); if (free) btrfs_qgroup_free_refroot(inode->root->fs_info, - inode->root->root_key.objectid, + btrfs_root_id(inode->root), changeset.bytes_changed, BTRFS_QGROUP_RSV_DATA); if (released) *released = changeset.bytes_changed; @@ -4361,7 +4361,7 @@ int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, int ret; if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_DISABLED || - !is_fstree(root->root_key.objectid) || num_bytes == 0) + !is_fstree(btrfs_root_id(root)) || num_bytes == 0) return 0; BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize)); @@ -4406,13 +4406,13 @@ void btrfs_qgroup_free_meta_all_pertrans(struct btrfs_root *root) struct btrfs_fs_info *fs_info = root->fs_info; if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_DISABLED || - !is_fstree(root->root_key.objectid)) + !is_fstree(btrfs_root_id(root))) return; /* TODO: Update trace point to handle such free */ trace_qgroup_meta_free_all_pertrans(root); /* Special value -1 means to free all reserved space */ - btrfs_qgroup_free_refroot(fs_info, root->root_key.objectid, (u64)-1, + btrfs_qgroup_free_refroot(fs_info, btrfs_root_id(root), (u64)-1, BTRFS_QGROUP_RSV_META_PERTRANS); } @@ -4422,7 +4422,7 @@ void __btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes, struct btrfs_fs_info *fs_info = root->fs_info; if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_DISABLED || - !is_fstree(root->root_key.objectid)) + !is_fstree(btrfs_root_id(root))) return; /* @@ -4433,8 +4433,7 @@ void __btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes, num_bytes = sub_root_meta_rsv(root, num_bytes, type); BUG_ON(num_bytes != round_down(num_bytes, fs_info->nodesize)); trace_qgroup_meta_reserve(root, -(s64)num_bytes, type); - btrfs_qgroup_free_refroot(fs_info, root->root_key.objectid, - num_bytes, type); + btrfs_qgroup_free_refroot(fs_info, btrfs_root_id(root), num_bytes, type); } static void qgroup_convert_meta(struct btrfs_fs_info *fs_info, u64 ref_root, @@ -4482,13 +4481,13 @@ void btrfs_qgroup_convert_reserved_meta(struct btrfs_root *root, int num_bytes) struct btrfs_fs_info *fs_info = root->fs_info; if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_DISABLED || - !is_fstree(root->root_key.objectid)) + !is_fstree(btrfs_root_id(root))) return; /* Same as btrfs_qgroup_free_meta_prealloc() */ num_bytes = sub_root_meta_rsv(root, num_bytes, BTRFS_QGROUP_RSV_META_PREALLOC); trace_qgroup_meta_convert(root, num_bytes); - qgroup_convert_meta(fs_info, root->root_key.objectid, num_bytes); + qgroup_convert_meta(fs_info, btrfs_root_id(root), num_bytes); if (!sb_rdonly(fs_info->sb)) add_root_meta_rsv(root, num_bytes, BTRFS_QGROUP_RSV_META_PERTRANS); } @@ -4517,7 +4516,7 @@ void btrfs_qgroup_check_reserved_leak(struct btrfs_inode *inode) btrfs_ino(inode), unode->val, unode->aux); } btrfs_qgroup_free_refroot(inode->root->fs_info, - inode->root->root_key.objectid, + btrfs_root_id(inode->root), changeset.bytes_changed, BTRFS_QGROUP_RSV_DATA); } @@ -4703,7 +4702,7 @@ int btrfs_qgroup_trace_subtree_after_cow(struct btrfs_trans_handle *trans, if (!btrfs_qgroup_full_accounting(fs_info)) return 0; - if (!is_fstree(root->root_key.objectid) || !root->reloc_root) + if (!is_fstree(btrfs_root_id(root)) || !root->reloc_root) return 0; spin_lock(&blocks->lock); diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c index f12ba2b75141..d0a3fcecc46a 100644 --- a/fs/btrfs/reflink.c +++ b/fs/btrfs/reflink.c @@ -665,7 +665,7 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen, if (root_dst->send_in_progress) { btrfs_warn_rl(root_dst->fs_info, "cannot deduplicate to root %llu while send operations are using it (%d in progress)", - root_dst->root_key.objectid, + btrfs_root_id(root_dst), root_dst->send_in_progress); spin_unlock(&root_dst->root_item_lock); return -EAGAIN; diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 00f33d8c4bf9..7e7799b4560b 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -754,7 +754,7 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans, root_key.type = BTRFS_ROOT_ITEM_KEY; root_key.offset = objectid; - if (root->root_key.objectid == objectid) { + if (btrfs_root_id(root) == objectid) { u64 commit_root_gen; /* called by btrfs_init_reloc_root */ @@ -798,7 +798,7 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans, btrfs_set_root_level(root_item, btrfs_header_level(eb)); btrfs_set_root_generation(root_item, trans->transid); - if (root->root_key.objectid == objectid) { + if (btrfs_root_id(root) == objectid) { btrfs_set_root_refs(root_item, 0); memset(&root_item->drop_progress, 0, sizeof(struct btrfs_disk_key)); @@ -876,8 +876,7 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans, * We are merging reloc roots, we do not need new reloc trees. Also * reloc trees never need their own reloc tree. */ - if (!rc->create_reloc_tree || - root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) + if (!rc->create_reloc_tree || btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) return 0; if (!trans->reloc_reserved) { @@ -885,7 +884,7 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans, trans->block_rsv = rc->block_rsv; clear_rsv = 1; } - reloc_root = create_reloc_root(trans, root, root->root_key.objectid); + reloc_root = create_reloc_root(trans, root, btrfs_root_id(root)); if (clear_rsv) trans->block_rsv = rsv; if (IS_ERR(reloc_root)) @@ -1027,7 +1026,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans, return 0; /* reloc trees always use full backref */ - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) parent = leaf->start; else parent = 0; @@ -1056,7 +1055,7 @@ int replace_file_extents(struct btrfs_trans_handle *trans, * if we are modifying block in fs tree, wait for read_folio * to complete and drop the extent cache */ - if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) { + if (btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID) { if (first) { inode = btrfs_find_first_inode(root, key.objectid); first = 0; @@ -1108,10 +1107,10 @@ int replace_file_extents(struct btrfs_trans_handle *trans, ref.bytenr = new_bytenr; ref.num_bytes = num_bytes; ref.parent = parent; - ref.owning_root = root->root_key.objectid; + ref.owning_root = btrfs_root_id(root); ref.ref_root = btrfs_header_owner(leaf); btrfs_init_data_ref(&ref, key.objectid, key.offset, - root->root_key.objectid, false); + btrfs_root_id(root), false); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -1122,10 +1121,10 @@ int replace_file_extents(struct btrfs_trans_handle *trans, ref.bytenr = bytenr; ref.num_bytes = num_bytes; ref.parent = parent; - ref.owning_root = root->root_key.objectid; + ref.owning_root = btrfs_root_id(root); ref.ref_root = btrfs_header_owner(leaf); btrfs_init_data_ref(&ref, key.objectid, key.offset, - root->root_key.objectid, false); + btrfs_root_id(root), false); ret = btrfs_free_extent(trans, &ref); if (ret) { btrfs_abort_transaction(trans, ret); @@ -1181,8 +1180,8 @@ int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc, int ret; int slot; - ASSERT(src->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID); - ASSERT(dest->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID); + ASSERT(btrfs_root_id(src) == BTRFS_TREE_RELOC_OBJECTID); + ASSERT(btrfs_root_id(dest) != BTRFS_TREE_RELOC_OBJECTID); last_snapshot = btrfs_root_last_snapshot(&src->root_item); again: @@ -1338,8 +1337,8 @@ again: ref.bytenr = old_bytenr; ref.num_bytes = blocksize; ref.parent = path->nodes[level]->start; - ref.owning_root = src->root_key.objectid; - ref.ref_root = src->root_key.objectid; + ref.owning_root = btrfs_root_id(src); + ref.ref_root = btrfs_root_id(src); btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { @@ -1351,8 +1350,8 @@ again: ref.bytenr = new_bytenr; ref.num_bytes = blocksize; ref.parent = 0; - ref.owning_root = dest->root_key.objectid; - ref.ref_root = dest->root_key.objectid; + ref.owning_root = btrfs_root_id(dest); + ref.ref_root = btrfs_root_id(dest); btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_inc_extent_ref(trans, &ref); if (ret) { @@ -1366,7 +1365,7 @@ again: ref.num_bytes = blocksize; ref.parent = path->nodes[level]->start; ref.owning_root = 0; - ref.ref_root = src->root_key.objectid; + ref.ref_root = btrfs_root_id(src); btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_free_extent(trans, &ref); if (ret) { @@ -1380,7 +1379,7 @@ again: ref.num_bytes = blocksize; ref.parent = 0; ref.owning_root = 0; - ref.ref_root = dest->root_key.objectid; + ref.ref_root = btrfs_root_id(dest); btrfs_init_tree_ref(&ref, level - 1, 0, true); ret = btrfs_free_extent(trans, &ref); if (ret) { @@ -1586,7 +1585,7 @@ static int insert_dirty_subvol(struct btrfs_trans_handle *trans, int ret; /* @root must be a subvolume tree root with a valid reloc tree */ - ASSERT(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID); + ASSERT(btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID); ASSERT(reloc_root); reloc_root_item = &reloc_root->root_item; @@ -1615,7 +1614,7 @@ static int clean_dirty_subvols(struct reloc_control *rc) list_for_each_entry_safe(root, next, &rc->dirty_subvol_roots, reloc_dirty_list) { - if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) { + if (btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID) { /* Merged subvolume, cleanup its reloc root */ struct btrfs_root *reloc_root = root->reloc_root; @@ -1890,13 +1889,13 @@ again: if (root->reloc_root) { btrfs_err(fs_info, "reloc tree mismatch, root %lld has reloc root key (%lld %u %llu) gen %llu, expect reloc root key (%lld %u %llu) gen %llu", - root->root_key.objectid, - root->reloc_root->root_key.objectid, + btrfs_root_id(root), + btrfs_root_id(root->reloc_root), root->reloc_root->root_key.type, root->reloc_root->root_key.offset, btrfs_root_generation( &root->reloc_root->root_item), - reloc_root->root_key.objectid, + btrfs_root_id(reloc_root), reloc_root->root_key.type, reloc_root->root_key.offset, btrfs_root_generation( @@ -1904,8 +1903,8 @@ again: } else { btrfs_err(fs_info, "reloc tree mismatch, root %lld has no reloc root, expect reloc root key (%lld %u %llu) gen %llu", - root->root_key.objectid, - reloc_root->root_key.objectid, + btrfs_root_id(root), + btrfs_root_id(reloc_root), reloc_root->root_key.type, reloc_root->root_key.offset, btrfs_root_generation( @@ -2162,7 +2161,7 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans, return ERR_PTR(-EUCLEAN); } - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) { + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) { ret = record_reloc_root_in_trans(trans, root); if (ret) return ERR_PTR(ret); @@ -2269,7 +2268,7 @@ struct btrfs_root *select_one_root(struct btrfs_backref_node *node) if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) return root; - if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) + if (btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID) fs_root = root; if (next != node) @@ -2495,7 +2494,7 @@ static int do_relocation(struct btrfs_trans_handle *trans, btrfs_mark_buffer_dirty(trans, upper->eb); btrfs_init_tree_ref(&ref, node->level, - root->root_key.objectid, false); + btrfs_root_id(root), false); ret = btrfs_inc_extent_ref(trans, &ref); if (!ret) ret = btrfs_drop_subtree(trans, root, eb, @@ -4463,8 +4462,7 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans, btrfs_root_last_snapshot(&root->root_item)) first_cow = 1; - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID && - rc->create_reloc_tree) { + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID && rc->create_reloc_tree) { WARN_ON(!first_cow && level == 0); node = rc->backref_cache.path[level]; @@ -4557,8 +4555,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans, } new_root = pending->snap; - reloc_root = create_reloc_root(trans, root->reloc_root, - new_root->root_key.objectid); + reloc_root = create_reloc_root(trans, root->reloc_root, btrfs_root_id(new_root)); if (IS_ERR(reloc_root)) return PTR_ERR(reloc_root); diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 7007f9e0c972..33962671a96c 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -148,8 +148,7 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root if (ret > 0) { btrfs_crit(fs_info, "unable to find root key (%llu %u %llu) in tree %llu", - key->objectid, key->type, key->offset, - root->root_key.objectid); + key->objectid, key->type, key->offset, btrfs_root_id(root)); ret = -EUCLEAN; btrfs_abort_transaction(trans, ret); goto out; diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 342fe5ff0f05..3dd4a48479a9 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -392,9 +392,8 @@ static void inconsistent_snapshot_error(struct send_ctx *sctx, btrfs_err(sctx->send_root->fs_info, "Send: inconsistent snapshot, found %s %s for inode %llu without updated inode item, send root is %llu, parent root is %llu", result_string, what, sctx->cmp_key->objectid, - sctx->send_root->root_key.objectid, - (sctx->parent_root ? - sctx->parent_root->root_key.objectid : 0)); + btrfs_root_id(sctx->send_root), + (sctx->parent_root ? btrfs_root_id(sctx->parent_root) : 0)); } __maybe_unused @@ -1316,9 +1315,9 @@ static int __clone_root_cmp_bsearch(const void *key, const void *elt) u64 root = (u64)(uintptr_t)key; const struct clone_root *cr = elt; - if (root < cr->root->root_key.objectid) + if (root < btrfs_root_id(cr->root)) return -1; - if (root > cr->root->root_key.objectid) + if (root > btrfs_root_id(cr->root)) return 1; return 0; } @@ -1328,9 +1327,9 @@ static int __clone_root_cmp_sort(const void *e1, const void *e2) const struct clone_root *cr1 = e1; const struct clone_root *cr2 = e2; - if (cr1->root->root_key.objectid < cr2->root->root_key.objectid) + if (btrfs_root_id(cr1->root) < btrfs_root_id(cr2->root)) return -1; - if (cr1->root->root_key.objectid > cr2->root->root_key.objectid) + if (btrfs_root_id(cr1->root) > btrfs_root_id(cr2->root)) return 1; return 0; } @@ -1778,7 +1777,7 @@ static int read_symlink(struct btrfs_root *root, */ btrfs_err(root->fs_info, "Found empty symlink inode %llu at root %llu", - ino, root->root_key.objectid); + ino, btrfs_root_id(root)); ret = -EIO; goto out; } @@ -2532,7 +2531,7 @@ static int send_subvol_begin(struct send_ctx *sctx) return -ENOMEM; } - key.objectid = send_root->root_key.objectid; + key.objectid = btrfs_root_id(send_root); key.type = BTRFS_ROOT_BACKREF_KEY; key.offset = 0; @@ -2548,7 +2547,7 @@ static int send_subvol_begin(struct send_ctx *sctx) leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); if (key.type != BTRFS_ROOT_BACKREF_KEY || - key.objectid != send_root->root_key.objectid) { + key.objectid != btrfs_root_id(send_root)) { ret = -ENOENT; goto out; } @@ -5318,7 +5317,7 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len) btrfs_err(fs_info, "send: IO error at offset %llu for inode %llu root %llu", folio_pos(folio), sctx->cur_ino, - sctx->send_root->root_key.objectid); + btrfs_root_id(sctx->send_root)); folio_put(folio); ret = -EIO; break; @@ -5389,7 +5388,7 @@ static int send_clone(struct send_ctx *sctx, btrfs_debug(sctx->send_root->fs_info, "send_clone offset=%llu, len=%d, clone_root=%llu, clone_inode=%llu, clone_offset=%llu", - offset, len, clone_root->root->root_key.objectid, + offset, len, btrfs_root_id(clone_root->root), clone_root->ino, clone_root->offset); p = fs_path_alloc(); @@ -7338,7 +7337,7 @@ static int search_key_again(const struct send_ctx *sctx, "send: key (%llu %u %llu) not found in %s root %llu, lowest_level %d, slot %d", key->objectid, key->type, key->offset, (root == sctx->parent_root ? "parent" : "send"), - root->root_key.objectid, path->lowest_level, + btrfs_root_id(root), path->lowest_level, path->slots[path->lowest_level]); return -EUCLEAN; } @@ -8072,7 +8071,7 @@ static void btrfs_root_dec_send_in_progress(struct btrfs_root* root) if (root->send_in_progress < 0) btrfs_err(root->fs_info, "send_in_progress unbalanced %d root %llu", - root->send_in_progress, root->root_key.objectid); + root->send_in_progress, btrfs_root_id(root)); spin_unlock(&root->root_item_lock); } @@ -8080,7 +8079,7 @@ static void dedupe_in_progress_warn(const struct btrfs_root *root) { btrfs_warn_rl(root->fs_info, "cannot use root %llu for send while deduplications on it are in progress (%d in progress)", - root->root_key.objectid, root->dedupe_in_progress); + btrfs_root_id(root), root->dedupe_in_progress); } long btrfs_ioctl_send(struct inode *inode, struct btrfs_ioctl_send_args *arg) diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 7e44ccaf348f..bb2e5294bb0c 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1097,10 +1097,9 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry) #endif if (btrfs_test_opt(info, REF_VERIFY)) seq_puts(seq, ",ref_verify"); - seq_printf(seq, ",subvolid=%llu", - BTRFS_I(d_inode(dentry))->root->root_key.objectid); + seq_printf(seq, ",subvolid=%llu", btrfs_root_id(BTRFS_I(d_inode(dentry))->root)); subvol_name = btrfs_get_subvol_name_from_objectid(info, - BTRFS_I(d_inode(dentry))->root->root_key.objectid); + btrfs_root_id(BTRFS_I(d_inode(dentry))->root)); if (!IS_ERR(subvol_name)) { seq_puts(seq, ",subvol="); seq_escape(seq, subvol_name, " \t\n\\"); @@ -1152,7 +1151,7 @@ static struct dentry *mount_subvol(const char *subvol_name, u64 subvol_objectid, struct super_block *s = root->d_sb; struct btrfs_fs_info *fs_info = btrfs_sb(s); struct inode *root_inode = d_inode(root); - u64 root_objectid = BTRFS_I(root_inode)->root->root_key.objectid; + u64 root_objectid = btrfs_root_id(BTRFS_I(root_inode)->root); ret = 0; if (!is_subvolume_inode(root_inode)) { @@ -1774,10 +1773,8 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_fsid.val[0] = be32_to_cpu(fsid[0]) ^ be32_to_cpu(fsid[2]); buf->f_fsid.val[1] = be32_to_cpu(fsid[1]) ^ be32_to_cpu(fsid[3]); /* Mask in the root object ID too, to disambiguate subvols */ - buf->f_fsid.val[0] ^= - BTRFS_I(d_inode(dentry))->root->root_key.objectid >> 32; - buf->f_fsid.val[1] ^= - BTRFS_I(d_inode(dentry))->root->root_key.objectid; + buf->f_fsid.val[0] ^= btrfs_root_id(BTRFS_I(d_inode(dentry))->root) >> 32; + buf->f_fsid.val[1] ^= btrfs_root_id(BTRFS_I(d_inode(dentry))->root); return 0; } diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index df2e58aa824a..3e3bcc5f64c6 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -426,7 +426,7 @@ static int record_root_in_trans(struct btrfs_trans_handle *trans, return 0; } radix_tree_tag_set(&fs_info->fs_roots_radix, - (unsigned long)root->root_key.objectid, + (unsigned long)btrfs_root_id(root), BTRFS_ROOT_TRANS_TAG); spin_unlock(&fs_info->fs_roots_radix_lock); root->last_trans = trans->transid; @@ -472,7 +472,7 @@ void btrfs_add_dropped_root(struct btrfs_trans_handle *trans, /* Make sure we don't try to update the root at commit time */ spin_lock(&fs_info->fs_roots_radix_lock); radix_tree_tag_clear(&fs_info->fs_roots_radix, - (unsigned long)root->root_key.objectid, + (unsigned long)btrfs_root_id(root), BTRFS_ROOT_TRANS_TAG); spin_unlock(&fs_info->fs_roots_radix_lock); } @@ -550,7 +550,7 @@ static inline bool need_reserve_reloc_root(struct btrfs_root *root) if (!fs_info->reloc_ctl || !test_bit(BTRFS_ROOT_SHAREABLE, &root->state) || - root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID || + btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID || root->reloc_root) return false; @@ -1229,7 +1229,7 @@ int btrfs_wait_tree_log_extents(struct btrfs_root *log_root, int mark) bool errors = false; int err; - ASSERT(log_root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID); + ASSERT(btrfs_root_id(log_root) == BTRFS_TREE_LOG_OBJECTID); err = __btrfs_wait_marked_extents(fs_info, dirty_pages); if ((mark & EXTENT_DIRTY) && @@ -1492,7 +1492,7 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans) ASSERT(atomic_read(&root->log_commit[1]) == 0); radix_tree_tag_clear(&fs_info->fs_roots_radix, - (unsigned long)root->root_key.objectid, + (unsigned long)btrfs_root_id(root), BTRFS_ROOT_TRANS_TAG); btrfs_qgroup_free_meta_all_pertrans(root); spin_unlock(&fs_info->fs_roots_radix_lock); @@ -1583,8 +1583,8 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans, goto out; /* Now qgroup are all updated, we can inherit it to new qgroups */ - ret = btrfs_qgroup_inherit(trans, src->root_key.objectid, dst_objectid, - parent->root_key.objectid, inherit); + ret = btrfs_qgroup_inherit(trans, btrfs_root_id(src), dst_objectid, + btrfs_root_id(parent), inherit); if (ret < 0) goto out; @@ -1822,7 +1822,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, * insert root back/forward references */ ret = btrfs_add_root_ref(trans, objectid, - parent_root->root_key.objectid, + btrfs_root_id(parent_root), btrfs_ino(BTRFS_I(parent_inode)), index, &fname.disk_name); if (ret) { @@ -1855,8 +1855,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, ret = qgroup_account_snapshot(trans, root, parent_root, pending->inherit, objectid); else if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_SIMPLE) - ret = btrfs_qgroup_inherit(trans, root->root_key.objectid, objectid, - parent_root->root_key.objectid, pending->inherit); + ret = btrfs_qgroup_inherit(trans, btrfs_root_id(root), objectid, + btrfs_root_id(parent_root), pending->inherit); if (ret < 0) goto fail; @@ -2623,7 +2623,7 @@ int btrfs_clean_one_deleted_snapshot(struct btrfs_fs_info *fs_info) list_del_init(&root->root_list); spin_unlock(&fs_info->trans_lock); - btrfs_debug(fs_info, "cleaner removing %llu", root->root_key.objectid); + btrfs_debug(fs_info, "cleaner removing %llu", btrfs_root_id(root)); btrfs_kill_all_delayed_nodes(root); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 1c7efb7c2160..a4e73081d477 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -391,7 +391,7 @@ static int overwrite_item(struct btrfs_trans_handle *trans, * the leaf before writing into the log tree. See the comments at * copy_items() for more details. */ - ASSERT(root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID); + ASSERT(btrfs_root_id(root) != BTRFS_TREE_LOG_OBJECTID); item_size = btrfs_item_size(eb, slot); src_ptr = btrfs_item_ptr_offset(eb, slot); @@ -765,8 +765,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, .action = BTRFS_ADD_DELAYED_REF, .bytenr = ins.objectid, .num_bytes = ins.offset, - .owning_root = root->root_key.objectid, - .ref_root = root->root_key.objectid, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), }; btrfs_init_data_ref(&ref, key->objectid, offset, 0, false); @@ -779,7 +779,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, * allocation tree */ ret = btrfs_alloc_logged_file_extent(trans, - root->root_key.objectid, + btrfs_root_id(root), key->objectid, offset, &ins); if (ret) goto out; @@ -3047,7 +3047,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, if (ret != -ENOSPC) btrfs_err(fs_info, "failed to update log for root %llu ret %d", - root->root_key.objectid, ret); + btrfs_root_id(root), ret); btrfs_wait_tree_log_extents(log, mark); mutex_unlock(&log_root_tree->log_mutex); goto out; diff --git a/fs/btrfs/tree-mod-log.c b/fs/btrfs/tree-mod-log.c index 43b3accbed7a..fa45b5fb9683 100644 --- a/fs/btrfs/tree-mod-log.c +++ b/fs/btrfs/tree-mod-log.c @@ -1004,7 +1004,7 @@ struct extent_buffer *btrfs_get_old_root(struct btrfs_root *root, u64 time_seq) free_extent_buffer(eb_root); check.level = level; - check.owner_root = root->root_key.objectid; + check.owner_root = btrfs_root_id(root); old = read_tree_block(fs_info, logical, &check); if (WARN_ON(IS_ERR(old) || !extent_buffer_uptodate(old))) { -- cgit v1.2.3-59-g8ed1b From 6c566def955cab6ae420d1c901674b8254d9d480 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 17 Jan 2024 11:54:39 +0000 Subject: btrfs: pass the extent map tree's inode to add_extent_mapping() Extent maps are always added to an inode's extent map tree, so there's no need to pass the extent map tree explicitly to add_extent_mapping(). In order to facilitate an upcoming change that adds a shrinker for extent maps, change add_extent_mapping() to receive the inode instead of its extent map tree. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 34 ++++++++++++++++------------------ 1 file changed, 16 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 4a9fbcb24d9b..c0846eed2fb2 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -355,21 +355,22 @@ static inline void setup_extent_mapping(struct extent_map_tree *tree, } /* - * Add new extent map to the extent tree + * Add a new extent map to an inode's extent map tree. * - * @tree: tree to insert new map in + * @inode: the target inode * @em: map to insert * @modified: indicate whether the given @em should be added to the * modified list, which indicates the extent needs to be logged * - * Insert @em into @tree or perform a simple forward/backward merge with - * existing mappings. The extent_map struct passed in will be inserted - * into the tree directly, with an additional reference taken, or a - * reference dropped if the merge attempt was successful. + * Insert @em into the @inode's extent map tree or perform a simple + * forward/backward merge with existing mappings. The extent_map struct passed + * in will be inserted into the tree directly, with an additional reference + * taken, or a reference dropped if the merge attempt was successful. */ -static int add_extent_mapping(struct extent_map_tree *tree, +static int add_extent_mapping(struct btrfs_inode *inode, struct extent_map *em, int modified) { + struct extent_map_tree *tree = &inode->extent_tree; int ret; lockdep_assert_held_write(&tree->lock); @@ -508,7 +509,7 @@ static struct extent_map *prev_extent_map(struct extent_map *em) * and an extent that you want to insert, deal with overlap and insert * the best fitted new extent into the tree. */ -static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, +static noinline int merge_extent_mapping(struct btrfs_inode *inode, struct extent_map *existing, struct extent_map *em, u64 map_start) @@ -542,7 +543,7 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, em->block_start += start_diff; em->block_len = em->len; } - return add_extent_mapping(em_tree, em, 0); + return add_extent_mapping(inode, em, 0); } /* @@ -570,7 +571,6 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode, { int ret; struct extent_map *em = *em_in; - struct extent_map_tree *em_tree = &inode->extent_tree; struct btrfs_fs_info *fs_info = inode->root->fs_info; /* @@ -580,7 +580,7 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode, if (em->block_start == EXTENT_MAP_INLINE) ASSERT(em->start == 0); - ret = add_extent_mapping(em_tree, em, 0); + ret = add_extent_mapping(inode, em, 0); /* it is possible that someone inserted the extent into the tree * while we had the lock dropped. It is also possible that * an overlapping map exists in the tree @@ -588,7 +588,7 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode, if (ret == -EEXIST) { struct extent_map *existing; - existing = search_extent_mapping(em_tree, start, len); + existing = search_extent_mapping(&inode->extent_tree, start, len); trace_btrfs_handle_em_exist(fs_info, existing, em, start, len); @@ -609,8 +609,7 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode, * The existing extent map is the one nearest to * the [start, start + len) range which overlaps */ - ret = merge_extent_mapping(em_tree, existing, - em, start); + ret = merge_extent_mapping(inode, existing, em, start); if (WARN_ON(ret)) { free_extent_map(em); *em_in = NULL; @@ -818,8 +817,7 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, } else { int ret; - ret = add_extent_mapping(em_tree, split, - modified); + ret = add_extent_mapping(inode, split, modified); /* Logic error, shouldn't happen. */ ASSERT(ret == 0); if (WARN_ON(ret != 0) && modified) @@ -909,7 +907,7 @@ int btrfs_replace_extent_map_range(struct btrfs_inode *inode, do { btrfs_drop_extent_map_range(inode, new_em->start, end, false); write_lock(&tree->lock); - ret = add_extent_mapping(tree, new_em, modified); + ret = add_extent_mapping(inode, new_em, modified); write_unlock(&tree->lock); } while (ret == -EEXIST); @@ -990,7 +988,7 @@ int split_extent_map(struct btrfs_inode *inode, u64 start, u64 len, u64 pre, split_mid->ram_bytes = split_mid->len; split_mid->flags = flags; split_mid->generation = em->generation; - add_extent_mapping(em_tree, split_mid, 1); + add_extent_mapping(inode, split_mid, 1); /* Once for us */ free_extent_map(em); -- cgit v1.2.3-59-g8ed1b From 002f3a2ce8b07d03aa46ed977ca86cdb30f3d668 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 21 Mar 2024 11:34:55 +0000 Subject: btrfs: pass the extent map tree's inode to clear_em_logging() Extent maps are always associated to an inode's extent map tree, so there's no need to pass the extent map tree explicitly to clear_em_logging(). In order to facilitate an upcoming change that adds a shrinker for extent maps, change clear_em_logging() to receive the inode instead of its extent map tree. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 4 +++- fs/btrfs/extent_map.h | 2 +- fs/btrfs/tree-log.c | 4 ++-- 3 files changed, 6 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index c0846eed2fb2..40f5a99ab382 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -331,8 +331,10 @@ out: } -void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em) +void clear_em_logging(struct btrfs_inode *inode, struct extent_map *em) { + struct extent_map_tree *tree = &inode->extent_tree; + lockdep_assert_held_write(&tree->lock); em->flags &= ~EXTENT_FLAG_LOGGING; diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index f287ab46e368..732fc8d7e534 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -129,7 +129,7 @@ void free_extent_map(struct extent_map *em); int __init extent_map_init(void); void __cold extent_map_exit(void); int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen); -void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em); +void clear_em_logging(struct btrfs_inode *inode, struct extent_map *em); struct extent_map *search_extent_mapping(struct extent_map_tree *tree, u64 start, u64 len); int btrfs_add_extent_mapping(struct btrfs_inode *inode, diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index a4e73081d477..5146387b416b 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -4949,7 +4949,7 @@ process: * private list. */ if (ret) { - clear_em_logging(tree, em); + clear_em_logging(inode, em); free_extent_map(em); continue; } @@ -4958,7 +4958,7 @@ process: ret = log_one_extent(trans, inode, em, path, ctx); write_lock(&tree->lock); - clear_em_logging(tree, em); + clear_em_logging(inode, em); free_extent_map(em); } WARN_ON(!list_empty(&extents)); -- cgit v1.2.3-59-g8ed1b From c2fbd812d749757c5abc6f995a7741da0653a4f4 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 21 Mar 2024 15:08:38 +0000 Subject: btrfs: pass the extent map tree's inode to remove_extent_mapping() Extent maps are always associated to an inode's extent map tree, so there's no need to pass the extent map tree explicitly to remove_extent_mapping(). In order to facilitate an upcoming change that adds a shrinker for extent maps, change remove_extent_mapping() to receive the inode instead of its extent map tree. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 2 +- fs/btrfs/extent_map.c | 22 +++++++++++++--------- fs/btrfs/extent_map.h | 2 +- fs/btrfs/tests/extent-map-tests.c | 19 ++++++++++--------- 4 files changed, 25 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 91122817f137..7b10f47d8f83 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2457,7 +2457,7 @@ remove_em: * hurts the fsync performance for workloads with a data * size that exceeds or is close to the system's memory). */ - remove_extent_mapping(map, em); + remove_extent_mapping(btrfs_inode, em); /* once for the rb tree */ free_extent_map(em); next: diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 40f5a99ab382..2b7e3666ebd3 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -449,16 +449,18 @@ struct extent_map *search_extent_mapping(struct extent_map_tree *tree, } /* - * Remove an extent_map from the extent tree. + * Remove an extent_map from its inode's extent tree. * - * @tree: extent tree to remove from + * @inode: the inode the extent map belongs to * @em: extent map being removed * - * Remove @em from @tree. No reference counts are dropped, and no checks - * are done to see if the range is in use. + * Remove @em from the extent tree of @inode. No reference counts are dropped, + * and no checks are done to see if the range is in use. */ -void remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em) +void remove_extent_mapping(struct btrfs_inode *inode, struct extent_map *em) { + struct extent_map_tree *tree = &inode->extent_tree; + lockdep_assert_held_write(&tree->lock); WARN_ON(em->flags & EXTENT_FLAG_PINNED); @@ -633,8 +635,10 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode, * if needed. This avoids searching the tree, from the root down to the first * extent map, before each deletion. */ -static void drop_all_extent_maps_fast(struct extent_map_tree *tree) +static void drop_all_extent_maps_fast(struct btrfs_inode *inode) { + struct extent_map_tree *tree = &inode->extent_tree; + write_lock(&tree->lock); while (!RB_EMPTY_ROOT(&tree->map.rb_root)) { struct extent_map *em; @@ -643,7 +647,7 @@ static void drop_all_extent_maps_fast(struct extent_map_tree *tree) node = rb_first_cached(&tree->map); em = rb_entry(node, struct extent_map, rb_node); em->flags &= ~(EXTENT_FLAG_PINNED | EXTENT_FLAG_LOGGING); - remove_extent_mapping(tree, em); + remove_extent_mapping(inode, em); free_extent_map(em); cond_resched_rwlock_write(&tree->lock); } @@ -676,7 +680,7 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, WARN_ON(end < start); if (end == (u64)-1) { if (start == 0 && !skip_pinned) { - drop_all_extent_maps_fast(em_tree); + drop_all_extent_maps_fast(inode); return; } len = (u64)-1; @@ -854,7 +858,7 @@ remove_em: ASSERT(!split); btrfs_set_inode_full_sync(inode); } - remove_extent_mapping(em_tree, em); + remove_extent_mapping(inode, em); } /* diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index 732fc8d7e534..c3707461ff62 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -120,7 +120,7 @@ static inline u64 extent_map_end(const struct extent_map *em) void extent_map_tree_init(struct extent_map_tree *tree); struct extent_map *lookup_extent_mapping(struct extent_map_tree *tree, u64 start, u64 len); -void remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em); +void remove_extent_mapping(struct btrfs_inode *inode, struct extent_map *em); int split_extent_map(struct btrfs_inode *inode, u64 start, u64 len, u64 pre, u64 new_logical); diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c index 0f5c9c9304d9..ba36794ba2d5 100644 --- a/fs/btrfs/tests/extent-map-tests.c +++ b/fs/btrfs/tests/extent-map-tests.c @@ -11,8 +11,9 @@ #include "../disk-io.h" #include "../block-group.h" -static int free_extent_map_tree(struct extent_map_tree *em_tree) +static int free_extent_map_tree(struct btrfs_inode *inode) { + struct extent_map_tree *em_tree = &inode->extent_tree; struct extent_map *em; struct rb_node *node; int ret = 0; @@ -21,7 +22,7 @@ static int free_extent_map_tree(struct extent_map_tree *em_tree) while (!RB_EMPTY_ROOT(&em_tree->map.rb_root)) { node = rb_first_cached(&em_tree->map); em = rb_entry(node, struct extent_map, rb_node); - remove_extent_mapping(em_tree, em); + remove_extent_mapping(inode, em); #ifdef CONFIG_BTRFS_DEBUG if (refcount_read(&em->refs) != 1) { @@ -142,7 +143,7 @@ static int test_case_1(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) } free_extent_map(em); out: - ret2 = free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(inode); if (ret == 0) ret = ret2; @@ -237,7 +238,7 @@ static int test_case_2(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) } free_extent_map(em); out: - ret2 = free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(inode); if (ret == 0) ret = ret2; @@ -313,7 +314,7 @@ static int __test_case_3(struct btrfs_fs_info *fs_info, } free_extent_map(em); out: - ret2 = free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(inode); if (ret == 0) ret = ret2; @@ -435,7 +436,7 @@ static int __test_case_4(struct btrfs_fs_info *fs_info, } free_extent_map(em); out: - ret2 = free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(inode); if (ret == 0) ret = ret2; @@ -679,7 +680,7 @@ static int test_case_5(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) if (ret) goto out; out: - ret2 = free_extent_map_tree(&inode->extent_tree); + ret2 = free_extent_map_tree(inode); if (ret == 0) ret = ret2; @@ -738,7 +739,7 @@ static int test_case_6(struct btrfs_fs_info *fs_info, struct btrfs_inode *inode) ret = 0; out: free_extent_map(em); - ret2 = free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(inode); if (ret == 0) ret = ret2; @@ -876,7 +877,7 @@ out: ret2 = unpin_extent_cache(inode, 0, SZ_16K, 0); if (ret == 0) ret = ret2; - ret2 = free_extent_map_tree(em_tree); + ret2 = free_extent_map_tree(inode); if (ret == 0) ret = ret2; -- cgit v1.2.3-59-g8ed1b From 6a3a9113ae1dbed188f0481815dcc94846a7b2e7 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 21 Mar 2024 16:07:26 +0000 Subject: btrfs: pass the extent map tree's inode to replace_extent_mapping() Extent maps are always associated to an inode's extent map tree, so there's no need to pass the extent map tree explicitly to replace_extent_mapping(). In order to facilitate an upcoming change that adds a shrinker for extent maps, change replace_extent_mapping() to receive the inode instead of its extent map tree. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 2b7e3666ebd3..66f50ebcb98a 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -470,11 +470,13 @@ void remove_extent_mapping(struct btrfs_inode *inode, struct extent_map *em) RB_CLEAR_NODE(&em->rb_node); } -static void replace_extent_mapping(struct extent_map_tree *tree, +static void replace_extent_mapping(struct btrfs_inode *inode, struct extent_map *cur, struct extent_map *new, int modified) { + struct extent_map_tree *tree = &inode->extent_tree; + lockdep_assert_held_write(&tree->lock); WARN_ON(cur->flags & EXTENT_FLAG_PINNED); @@ -777,7 +779,7 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, split->generation = gen; split->flags = flags; - replace_extent_mapping(em_tree, em, split, modified); + replace_extent_mapping(inode, em, split, modified); free_extent_map(split); split = split2; split2 = NULL; @@ -818,8 +820,7 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, } if (extent_map_in_tree(em)) { - replace_extent_mapping(em_tree, em, split, - modified); + replace_extent_mapping(inode, em, split, modified); } else { int ret; @@ -977,7 +978,7 @@ int split_extent_map(struct btrfs_inode *inode, u64 start, u64 len, u64 pre, split_pre->flags = flags; split_pre->generation = em->generation; - replace_extent_mapping(em_tree, em, split_pre, 1); + replace_extent_mapping(inode, em, split_pre, 1); /* * Now we only have an extent_map at: -- cgit v1.2.3-59-g8ed1b From e778724a5ef04e7144cb766a6904bdbe842695ee Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 11:29:17 +0100 Subject: btrfs: pass the extent map tree's inode to setup_extent_mapping() Extent maps are always associated to an inode's extent map tree, so there's no need to pass the extent map tree explicitly to setup_extent_mapping(). In order to facilitate an upcoming change that adds a shrinker for extent maps, change setup_extent_mapping() to receive the inode instead of its extent map tree. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 66f50ebcb98a..0d9066865dfb 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -342,7 +342,7 @@ void clear_em_logging(struct btrfs_inode *inode, struct extent_map *em) try_merge_map(tree, em); } -static inline void setup_extent_mapping(struct extent_map_tree *tree, +static inline void setup_extent_mapping(struct btrfs_inode *inode, struct extent_map *em, int modified) { @@ -351,9 +351,9 @@ static inline void setup_extent_mapping(struct extent_map_tree *tree, ASSERT(list_empty(&em->list)); if (modified) - list_add(&em->list, &tree->modified_extents); + list_add(&em->list, &inode->extent_tree.modified_extents); else - try_merge_map(tree, em); + try_merge_map(&inode->extent_tree, em); } /* @@ -381,7 +381,7 @@ static int add_extent_mapping(struct btrfs_inode *inode, if (ret) return ret; - setup_extent_mapping(tree, em, modified); + setup_extent_mapping(inode, em, modified); return 0; } @@ -486,7 +486,7 @@ static void replace_extent_mapping(struct btrfs_inode *inode, rb_replace_node_cached(&cur->rb_node, &new->rb_node, &tree->map); RB_CLEAR_NODE(&cur->rb_node); - setup_extent_mapping(tree, new, modified); + setup_extent_mapping(inode, new, modified); } static struct extent_map *next_extent_map(const struct extent_map *em) -- cgit v1.2.3-59-g8ed1b From 5fa8a6baff817c1b427aa7a8bfc1482043be6d58 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 11 Apr 2024 11:32:14 +0100 Subject: btrfs: pass the extent map tree's inode to try_merge_map() Extent maps are always associated to an inode's extent map tree, so there's no need to pass the extent map tree explicitly to try_merge_map(). In order to facilitate an upcoming change that adds a shrinker for extent maps, change try_merge_map() to receive the inode instead of its extent map tree. Reviewed-by: Qu Wenruo Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 0d9066865dfb..fca08ac95a4a 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -223,8 +223,9 @@ static bool mergeable_maps(const struct extent_map *prev, const struct extent_ma return next->block_start == prev->block_start; } -static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em) +static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em) { + struct extent_map_tree *tree = &inode->extent_tree; struct extent_map *merge = NULL; struct rb_node *rb; @@ -322,7 +323,7 @@ int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen) em->generation = gen; em->flags &= ~EXTENT_FLAG_PINNED; - try_merge_map(tree, em); + try_merge_map(inode, em); out: write_unlock(&tree->lock); @@ -333,13 +334,11 @@ out: void clear_em_logging(struct btrfs_inode *inode, struct extent_map *em) { - struct extent_map_tree *tree = &inode->extent_tree; - - lockdep_assert_held_write(&tree->lock); + lockdep_assert_held_write(&inode->extent_tree.lock); em->flags &= ~EXTENT_FLAG_LOGGING; if (extent_map_in_tree(em)) - try_merge_map(tree, em); + try_merge_map(inode, em); } static inline void setup_extent_mapping(struct btrfs_inode *inode, @@ -353,7 +352,7 @@ static inline void setup_extent_mapping(struct btrfs_inode *inode, if (modified) list_add(&em->list, &inode->extent_tree.modified_extents); else - try_merge_map(&inode->extent_tree, em); + try_merge_map(inode, em); } /* -- cgit v1.2.3-59-g8ed1b From f1d97e76915285013037c487d9513ab763005286 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 22 Mar 2024 18:02:59 +0000 Subject: btrfs: add a global per cpu counter to track number of used extent maps Add a per cpu counter that tracks the total number of extent maps that are in extent trees of inodes that belong to fs trees. This is going to be used in an upcoming change that adds a shrinker for extent maps. Only extent maps for fs trees are considered, because for special trees such as the data relocation tree we don't want to evict their extent maps which are critical for the relocation to work, and since those are limited, it's not a concern to have them in memory during the relocation of a block group. Another case are extent maps for free space cache inodes, which must always remain in memory, but those are limited (there's only one per free space cache inode, which means one per block group). Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 9 +++++++++ fs/btrfs/extent_map.c | 17 +++++++++++++++++ fs/btrfs/fs.h | 2 ++ 3 files changed, 28 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index cee2e4f43dc6..c2dc88f909b0 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1266,9 +1266,14 @@ static void free_global_roots(struct btrfs_fs_info *fs_info) void btrfs_free_fs_info(struct btrfs_fs_info *fs_info) { + struct percpu_counter *em_counter = &fs_info->evictable_extent_maps; + percpu_counter_destroy(&fs_info->dirty_metadata_bytes); percpu_counter_destroy(&fs_info->delalloc_bytes); percpu_counter_destroy(&fs_info->ordered_bytes); + if (percpu_counter_initialized(em_counter)) + ASSERT(percpu_counter_sum_positive(em_counter) == 0); + percpu_counter_destroy(em_counter); percpu_counter_destroy(&fs_info->dev_replace.bio_counter); btrfs_free_csum_hash(fs_info); btrfs_free_stripe_hash_table(fs_info); @@ -2848,6 +2853,10 @@ static int init_mount_fs_info(struct btrfs_fs_info *fs_info, struct super_block if (ret) return ret; + ret = percpu_counter_init(&fs_info->evictable_extent_maps, 0, GFP_KERNEL); + if (ret) + return ret; + ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0, GFP_KERNEL); if (ret) return ret; diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index fca08ac95a4a..031c124c514c 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -76,6 +76,14 @@ static u64 range_end(u64 start, u64 len) return start + len; } +static void dec_evictable_extent_maps(struct btrfs_inode *inode) +{ + struct btrfs_fs_info *fs_info = inode->root->fs_info; + + if (!btrfs_is_testing(fs_info) && is_fstree(btrfs_root_id(inode->root))) + percpu_counter_dec(&fs_info->evictable_extent_maps); +} + static int tree_insert(struct rb_root_cached *root, struct extent_map *em) { struct rb_node **p = &root->rb_root.rb_node; @@ -259,6 +267,7 @@ static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em) rb_erase_cached(&merge->rb_node, &tree->map); RB_CLEAR_NODE(&merge->rb_node); free_extent_map(merge); + dec_evictable_extent_maps(inode); } } @@ -273,6 +282,7 @@ static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em) em->generation = max(em->generation, merge->generation); em->flags |= EXTENT_FLAG_MERGED; free_extent_map(merge); + dec_evictable_extent_maps(inode); } } @@ -372,6 +382,8 @@ static int add_extent_mapping(struct btrfs_inode *inode, struct extent_map *em, int modified) { struct extent_map_tree *tree = &inode->extent_tree; + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; int ret; lockdep_assert_held_write(&tree->lock); @@ -382,6 +394,9 @@ static int add_extent_mapping(struct btrfs_inode *inode, setup_extent_mapping(inode, em, modified); + if (!btrfs_is_testing(fs_info) && is_fstree(btrfs_root_id(root))) + percpu_counter_inc(&fs_info->evictable_extent_maps); + return 0; } @@ -467,6 +482,8 @@ void remove_extent_mapping(struct btrfs_inode *inode, struct extent_map *em) if (!(em->flags & EXTENT_FLAG_LOGGING)) list_del_init(&em->list); RB_CLEAR_NODE(&em->rb_node); + + dec_evictable_extent_maps(inode); } static void replace_extent_mapping(struct btrfs_inode *inode, diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h index 93f5c57ea4e3..534d30dafe32 100644 --- a/fs/btrfs/fs.h +++ b/fs/btrfs/fs.h @@ -630,6 +630,8 @@ struct btrfs_fs_info { s32 dirty_metadata_batch; s32 delalloc_batch; + struct percpu_counter evictable_extent_maps; + /* Protected by 'trans_lock'. */ struct list_head dirty_cowonly_roots; -- cgit v1.2.3-59-g8ed1b From 956a17d9d050761e34ae6f2624e9c1ce456de204 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 15 Apr 2024 17:09:26 +0100 Subject: btrfs: add a shrinker for extent maps Extent maps are used either to represent existing file extent items, or to represent new extents that are going to be written and the respective file extent items are created when the ordered extent completes. We currently don't have any limit for how many extent maps we can have, neither per inode nor globally. Most of the time this not too noticeable because extent maps are removed in the following situations: 1) When evicting an inode; 2) When releasing folios (pages) through the btrfs_release_folio() address space operation callback. However we won't release extent maps in the folio range if the folio is either dirty or under writeback or if the inode's i_size is less than or equals to 16M (see try_release_extent_mapping(). This 16M i_size constraint was added back in 2008 with commit 70dec8079d78 ("Btrfs: extent_io and extent_state optimizations"), but there's no explanation about why we have it or why the 16M value. This means that for buffered IO we can reach an OOM situation due to too many extent maps if either of the following happens: 1) There's a set of tasks constantly doing IO on many files with a size not larger than 16M, specially if they keep the files open for very long periods, therefore preventing inode eviction. This requires a really high number of such files, and having many non mergeable extent maps (due to random 4K writes for example) and a machine with very little memory; 2) There's a set tasks constantly doing random write IO (therefore creating many non mergeable extent maps) on files and keeping them open for long periods of time, so inode eviction doesn't happen and there's always a lot of dirty pages or pages under writeback, preventing btrfs_release_folio() from releasing the respective extent maps. This second case was actually reported in the thread pointed by the Link tag below, and it requires a very large file under heavy IO and a machine with very little amount of RAM, which is probably hard to happen in practice in a real world use case. However when using direct IO this is not so hard to happen, because the page cache is not used, and therefore btrfs_release_folio() is never called. Which means extent maps are dropped only when evicting the inode, and that means that if we have tasks that keep a file descriptor open and keep doing IO on a very large file (or files), we can exhaust memory due to an unbounded amount of extent maps. This is especially easy to happen if we have a huge file with millions of small extents and their extent maps are not mergeable (non contiguous offsets and disk locations). This was reported in that thread with the following fio test: $ cat test.sh #!/bin/bash DEV=/dev/sdj MNT=/mnt/sdj MOUNT_OPTIONS="-o ssd" MKFS_OPTIONS="" cat < /tmp/fio-job.ini [global] name=fio-rand-write filename=$MNT/fio-rand-write rw=randwrite bs=4K direct=1 numjobs=16 fallocate=none time_based runtime=90000 [file1] size=300G ioengine=libaio iodepth=16 EOF umount $MNT &> /dev/null mkfs.btrfs -f $MKFS_OPTIONS $DEV mount $MOUNT_OPTIONS $DEV $MNT fio /tmp/fio-job.ini umount $MNT Monitoring the btrfs_extent_map slab while running the test with: $ watch -d -n 1 'cat /sys/kernel/slab/btrfs_extent_map/objects \ /sys/kernel/slab/btrfs_extent_map/total_objects' Shows the number of active and total extent maps skyrocketing to tens of millions, and on systems with a short amount of memory it's easy and quick to get into an OOM situation, as reported in that thread. So to avoid this issue add a shrinker that will remove extents maps, as long as they are not pinned, and takes proper care with any concurrent fsync to avoid missing extents (setting the full sync flag while in the middle of a fast fsync). This shrinker is triggered through the callbacks nr_cached_objects and free_cached_objects of struct super_operations. The shrinker will iterate over all roots and over all inodes of each root, and keeps track of the last scanned root and inode, so that the next time it runs, it starts from that root and from the next inode. This is similar to what xfs does for its inode reclaim (implements those callbacks, and cycles through inodes by starting from where it ended last time). Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 160 ++++++++++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/extent_map.h | 1 + fs/btrfs/fs.h | 2 + fs/btrfs/super.c | 17 ++++++ 4 files changed, 180 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 031c124c514c..62a21491c044 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -8,6 +8,7 @@ #include "extent_map.h" #include "compression.h" #include "btrfs_inode.h" +#include "disk-io.h" static struct kmem_cache *extent_map_cache; @@ -1026,3 +1027,162 @@ out_free_pre: free_extent_map(split_pre); return ret; } + +static long btrfs_scan_inode(struct btrfs_inode *inode, long *scanned, long nr_to_scan) +{ + const u64 cur_fs_gen = btrfs_get_fs_generation(inode->root->fs_info); + struct extent_map_tree *tree = &inode->extent_tree; + long nr_dropped = 0; + struct rb_node *node; + + /* + * Take the mmap lock so that we serialize with the inode logging phase + * of fsync because we may need to set the full sync flag on the inode, + * in case we have to remove extent maps in the tree's list of modified + * extents. If we set the full sync flag in the inode while an fsync is + * in progress, we may risk missing new extents because before the flag + * is set, fsync decides to only wait for writeback to complete and then + * during inode logging it sees the flag set and uses the subvolume tree + * to find new extents, which may not be there yet because ordered + * extents haven't completed yet. + * + * We also do a try lock because otherwise we could deadlock. This is + * because the shrinker for this filesystem may be invoked while we are + * in a path that is holding the mmap lock in write mode. For example in + * a reflink operation while COWing an extent buffer, when allocating + * pages for a new extent buffer and under memory pressure, the shrinker + * may be invoked, and therefore we would deadlock by attempting to read + * lock the mmap lock while we are holding already a write lock on it. + */ + if (!down_read_trylock(&inode->i_mmap_lock)) + return 0; + + write_lock(&tree->lock); + node = rb_first_cached(&tree->map); + while (node) { + struct extent_map *em; + + em = rb_entry(node, struct extent_map, rb_node); + node = rb_next(node); + (*scanned)++; + + if (em->flags & EXTENT_FLAG_PINNED) + goto next; + + /* + * If the inode is in the list of modified extents (new) and its + * generation is the same (or is greater than) the current fs + * generation, it means it was not yet persisted so we have to + * set the full sync flag so that the next fsync will not miss + * it. + */ + if (!list_empty(&em->list) && em->generation >= cur_fs_gen) + btrfs_set_inode_full_sync(inode); + + remove_extent_mapping(inode, em); + /* Drop the reference for the tree. */ + free_extent_map(em); + nr_dropped++; +next: + if (*scanned >= nr_to_scan) + break; + + /* + * Restart if we had to reschedule, and any extent maps that were + * pinned before may have become unpinned after we released the + * lock and took it again. + */ + if (cond_resched_rwlock_write(&tree->lock)) + node = rb_first_cached(&tree->map); + } + write_unlock(&tree->lock); + up_read(&inode->i_mmap_lock); + + return nr_dropped; +} + +static long btrfs_scan_root(struct btrfs_root *root, long *scanned, long nr_to_scan) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct btrfs_inode *inode; + long nr_dropped = 0; + u64 min_ino = fs_info->extent_map_shrinker_last_ino + 1; + + inode = btrfs_find_first_inode(root, min_ino); + while (inode) { + nr_dropped += btrfs_scan_inode(inode, scanned, nr_to_scan); + + min_ino = btrfs_ino(inode) + 1; + fs_info->extent_map_shrinker_last_ino = btrfs_ino(inode); + iput(&inode->vfs_inode); + + if (*scanned >= nr_to_scan) + break; + + cond_resched(); + inode = btrfs_find_first_inode(root, min_ino); + } + + if (inode) { + /* + * There are still inodes in this root or we happened to process + * the last one and reached the scan limit. In either case set + * the current root to this one, so we'll resume from the next + * inode if there is one or we will find out this was the last + * one and move to the next root. + */ + fs_info->extent_map_shrinker_last_root = btrfs_root_id(root); + } else { + /* + * No more inodes in this root, set extent_map_shrinker_last_ino to 0 so + * that when processing the next root we start from its first inode. + */ + fs_info->extent_map_shrinker_last_ino = 0; + fs_info->extent_map_shrinker_last_root = btrfs_root_id(root) + 1; + } + + return nr_dropped; +} + +long btrfs_free_extent_maps(struct btrfs_fs_info *fs_info, long nr_to_scan) +{ + const u64 start_root_id = fs_info->extent_map_shrinker_last_root; + u64 next_root_id = start_root_id; + bool cycled = false; + long nr_dropped = 0; + long scanned = 0; + + while (scanned < nr_to_scan) { + struct btrfs_root *root; + unsigned long count; + + spin_lock(&fs_info->fs_roots_radix_lock); + count = radix_tree_gang_lookup(&fs_info->fs_roots_radix, + (void **)&root, + (unsigned long)next_root_id, 1); + if (count == 0) { + spin_unlock(&fs_info->fs_roots_radix_lock); + if (start_root_id > 0 && !cycled) { + next_root_id = 0; + fs_info->extent_map_shrinker_last_root = 0; + fs_info->extent_map_shrinker_last_ino = 0; + cycled = true; + continue; + } + break; + } + next_root_id = btrfs_root_id(root) + 1; + root = btrfs_grab_root(root); + spin_unlock(&fs_info->fs_roots_radix_lock); + + if (!root) + continue; + + if (is_fstree(btrfs_root_id(root))) + nr_dropped += btrfs_scan_root(root, &scanned, nr_to_scan); + + btrfs_put_root(root); + } + + return nr_dropped; +} diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index c3707461ff62..8306a8f294d0 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -140,5 +140,6 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, int btrfs_replace_extent_map_range(struct btrfs_inode *inode, struct extent_map *new_em, bool modified); +long btrfs_free_extent_maps(struct btrfs_fs_info *fs_info, long nr_to_scan); #endif diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h index 534d30dafe32..9711c9c0e78f 100644 --- a/fs/btrfs/fs.h +++ b/fs/btrfs/fs.h @@ -631,6 +631,8 @@ struct btrfs_fs_info { s32 delalloc_batch; struct percpu_counter evictable_extent_maps; + u64 extent_map_shrinker_last_root; + u64 extent_map_shrinker_last_ino; /* Protected by 'trans_lock'. */ struct list_head dirty_cowonly_roots; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index bb2e5294bb0c..571f90d18c7f 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -2371,6 +2371,21 @@ static int btrfs_show_devname(struct seq_file *m, struct dentry *root) return 0; } +static long btrfs_nr_cached_objects(struct super_block *sb, struct shrink_control *sc) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(sb); + + return percpu_counter_sum_positive(&fs_info->evictable_extent_maps); +} + +static long btrfs_free_cached_objects(struct super_block *sb, struct shrink_control *sc) +{ + const long nr_to_scan = min_t(unsigned long, LONG_MAX, sc->nr_to_scan); + struct btrfs_fs_info *fs_info = btrfs_sb(sb); + + return btrfs_free_extent_maps(fs_info, nr_to_scan); +} + static const struct super_operations btrfs_super_ops = { .drop_inode = btrfs_drop_inode, .evict_inode = btrfs_evict_inode, @@ -2384,6 +2399,8 @@ static const struct super_operations btrfs_super_ops = { .statfs = btrfs_statfs, .freeze_fs = btrfs_freeze, .unfreeze_fs = btrfs_unfreeze, + .nr_cached_objects = btrfs_nr_cached_objects, + .free_cached_objects = btrfs_free_cached_objects, }; static const struct file_operations btrfs_ctl_fops = { -- cgit v1.2.3-59-g8ed1b From 65bb9fb00b7012a78b2f5d1cd042bf098900c5d3 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 9 Apr 2024 16:41:05 +0100 Subject: btrfs: update comment for btrfs_set_inode_full_sync() about locking Nowadays we have a lock used to synchronize mmap writes with reflink and fsync operations (struct btrfs_inode::i_mmap_lock), so update the comment for btrfs_set_inode_full_sync() to mention that it can also be called while holding that mmap lock. Besides being a valid alternative to the inode's VFS lock, we already have the extent map shrinker using that mmap lock instead. Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/btrfs_inode.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 9a87ada7fe52..91c994b569f3 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -381,9 +381,11 @@ static inline void btrfs_set_inode_last_sub_trans(struct btrfs_inode *inode) } /* - * Should be called while holding the inode's VFS lock in exclusive mode or in a - * context where no one else can access the inode concurrently (during inode - * creation or when loading an inode from disk). + * Should be called while holding the inode's VFS lock in exclusive mode, or + * while holding the inode's mmap lock (struct btrfs_inode::i_mmap_lock) in + * either shared or exclusive mode, or in a context where no one else can access + * the inode concurrently (during inode creation or when loading an inode from + * disk). */ static inline void btrfs_set_inode_full_sync(struct btrfs_inode *inode) { -- cgit v1.2.3-59-g8ed1b From 0d89a15e1a0dcc861481ee696ec332c998140510 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 9 Apr 2024 17:07:32 +0100 Subject: btrfs: add tracepoints for extent map shrinker events Add some tracepoints for the extent map shrinker to help debug and analyse main events. These have proved useful during development of the shrinker. Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_map.c | 13 ++++++ fs/btrfs/super.c | 5 ++- include/trace/events/btrfs.h | 99 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 116 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 62a21491c044..744e8952abb0 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -1080,6 +1080,7 @@ static long btrfs_scan_inode(struct btrfs_inode *inode, long *scanned, long nr_t btrfs_set_inode_full_sync(inode); remove_extent_mapping(inode, em); + trace_btrfs_extent_map_shrinker_remove_em(inode, em); /* Drop the reference for the tree. */ free_extent_map(em); nr_dropped++; @@ -1152,6 +1153,12 @@ long btrfs_free_extent_maps(struct btrfs_fs_info *fs_info, long nr_to_scan) long nr_dropped = 0; long scanned = 0; + if (trace_btrfs_extent_map_shrinker_scan_enter_enabled()) { + s64 nr = percpu_counter_sum_positive(&fs_info->evictable_extent_maps); + + trace_btrfs_extent_map_shrinker_scan_enter(fs_info, nr_to_scan, nr); + } + while (scanned < nr_to_scan) { struct btrfs_root *root; unsigned long count; @@ -1184,5 +1191,11 @@ long btrfs_free_extent_maps(struct btrfs_fs_info *fs_info, long nr_to_scan) btrfs_put_root(root); } + if (trace_btrfs_extent_map_shrinker_scan_exit_enabled()) { + s64 nr = percpu_counter_sum_positive(&fs_info->evictable_extent_maps); + + trace_btrfs_extent_map_shrinker_scan_exit(fs_info, nr_dropped, nr); + } + return nr_dropped; } diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 571f90d18c7f..2dbc930a20f7 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -2374,8 +2374,11 @@ static int btrfs_show_devname(struct seq_file *m, struct dentry *root) static long btrfs_nr_cached_objects(struct super_block *sb, struct shrink_control *sc) { struct btrfs_fs_info *fs_info = btrfs_sb(sb); + const s64 nr = percpu_counter_sum_positive(&fs_info->evictable_extent_maps); - return percpu_counter_sum_positive(&fs_info->evictable_extent_maps); + trace_btrfs_extent_map_shrinker_count(fs_info, nr); + + return nr; } static long btrfs_free_cached_objects(struct super_block *sb, struct shrink_control *sc) diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index 8f2497603cb5..d2d94d7c3fb5 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -2537,6 +2537,105 @@ TRACE_EVENT(btrfs_get_raid_extent_offset, __entry->devid) ); +TRACE_EVENT(btrfs_extent_map_shrinker_count, + + TP_PROTO(const struct btrfs_fs_info *fs_info, long nr), + + TP_ARGS(fs_info, nr), + + TP_STRUCT__entry_btrfs( + __field( long, nr ) + ), + + TP_fast_assign_btrfs(fs_info, + __entry->nr = nr; + ), + + TP_printk_btrfs("nr=%ld", __entry->nr) +); + +TRACE_EVENT(btrfs_extent_map_shrinker_scan_enter, + + TP_PROTO(const struct btrfs_fs_info *fs_info, long nr_to_scan, long nr), + + TP_ARGS(fs_info, nr_to_scan, nr), + + TP_STRUCT__entry_btrfs( + __field( long, nr_to_scan ) + __field( long, nr ) + __field( u64, last_root_id ) + __field( u64, last_ino ) + ), + + TP_fast_assign_btrfs(fs_info, + __entry->nr_to_scan = nr_to_scan; + __entry->nr = nr; + __entry->last_root_id = fs_info->extent_map_shrinker_last_root; + __entry->last_ino = fs_info->extent_map_shrinker_last_ino; + ), + + TP_printk_btrfs("nr_to_scan=%ld nr=%ld last_root=%llu(%s) last_ino=%llu", + __entry->nr_to_scan, __entry->nr, + show_root_type(__entry->last_root_id), __entry->last_ino) +); + +TRACE_EVENT(btrfs_extent_map_shrinker_scan_exit, + + TP_PROTO(const struct btrfs_fs_info *fs_info, long nr_dropped, long nr), + + TP_ARGS(fs_info, nr_dropped, nr), + + TP_STRUCT__entry_btrfs( + __field( long, nr_dropped ) + __field( long, nr ) + __field( u64, last_root_id ) + __field( u64, last_ino ) + ), + + TP_fast_assign_btrfs(fs_info, + __entry->nr_dropped = nr_dropped; + __entry->nr = nr; + __entry->last_root_id = fs_info->extent_map_shrinker_last_root; + __entry->last_ino = fs_info->extent_map_shrinker_last_ino; + ), + + TP_printk_btrfs("nr_dropped=%ld nr=%ld last_root=%llu(%s) last_ino=%llu", + __entry->nr_dropped, __entry->nr, + show_root_type(__entry->last_root_id), __entry->last_ino) +); + +TRACE_EVENT(btrfs_extent_map_shrinker_remove_em, + + TP_PROTO(const struct btrfs_inode *inode, const struct extent_map *em), + + TP_ARGS(inode, em), + + TP_STRUCT__entry_btrfs( + __field( u64, ino ) + __field( u64, root_id ) + __field( u64, start ) + __field( u64, len ) + __field( u64, block_start ) + __field( u32, flags ) + ), + + TP_fast_assign_btrfs(inode->root->fs_info, + __entry->ino = btrfs_ino(inode); + __entry->root_id = inode->root->root_key.objectid; + __entry->start = em->start; + __entry->len = em->len; + __entry->block_start = em->block_start; + __entry->flags = em->flags; + ), + + TP_printk_btrfs( +"ino=%llu root=%llu(%s) start=%llu len=%llu block_start=%llu(%s) flags=%s", + __entry->ino, show_root_type(__entry->root_id), + __entry->start, __entry->len, + show_map_type(__entry->block_start), + show_map_flags(__entry->flags)) +); + #endif /* _TRACE_BTRFS_H */ /* This part must be outside protection */ -- cgit v1.2.3-59-g8ed1b From 078b981aaa565040348cd3ca75b0ec9e138464a9 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 16 Apr 2024 15:07:13 +0100 Subject: btrfs: rename some variables at try_release_extent_mapping() Rename the following variables: 1) "btrfs_inode" to "inode", because it's shorter to type and clear, and we don't have a VFS inode here as well, so there's no confusion; 2) "tree" to "io_tree", to be clear which tree we are dealing with, since we use 2 different trees in the function; 3) "map" to "extent_tree" since "map" gives the idea we are dealing with an extent map for example, but we are dealing with the inode's extent tree (the tree which stores extent maps). These also make the next patches simpler. Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 7b10f47d8f83..6438c3e74756 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2398,9 +2398,9 @@ int try_release_extent_mapping(struct page *page, gfp_t mask) struct extent_map *em; u64 start = page_offset(page); u64 end = start + PAGE_SIZE - 1; - struct btrfs_inode *btrfs_inode = page_to_inode(page); - struct extent_io_tree *tree = &btrfs_inode->io_tree; - struct extent_map_tree *map = &btrfs_inode->extent_tree; + struct btrfs_inode *inode = page_to_inode(page); + struct extent_io_tree *io_tree = &inode->io_tree; + struct extent_map_tree *extent_tree = &inode->extent_tree; if (gfpflags_allow_blocking(mask) && page->mapping->host->i_size > SZ_16M) { @@ -2410,19 +2410,19 @@ int try_release_extent_mapping(struct page *page, gfp_t mask) u64 cur_gen; len = end - start + 1; - write_lock(&map->lock); - em = lookup_extent_mapping(map, start, len); + write_lock(&extent_tree->lock); + em = lookup_extent_mapping(extent_tree, start, len); if (!em) { - write_unlock(&map->lock); + write_unlock(&extent_tree->lock); break; } if ((em->flags & EXTENT_FLAG_PINNED) || em->start != start) { - write_unlock(&map->lock); + write_unlock(&extent_tree->lock); free_extent_map(em); break; } - if (test_range_bit_exists(tree, em->start, + if (test_range_bit_exists(io_tree, em->start, extent_map_end(em) - 1, EXTENT_LOCKED)) goto next; @@ -2442,7 +2442,7 @@ int try_release_extent_mapping(struct page *page, gfp_t mask) * Otherwise don't remove it, we could be racing with an * ongoing fast fsync that could miss the new extent. */ - fs_info = btrfs_inode->root->fs_info; + fs_info = inode->root->fs_info; spin_lock(&fs_info->trans_lock); cur_gen = fs_info->generation; spin_unlock(&fs_info->trans_lock); @@ -2457,12 +2457,12 @@ remove_em: * hurts the fsync performance for workloads with a data * size that exceeds or is close to the system's memory). */ - remove_extent_mapping(btrfs_inode, em); + remove_extent_mapping(inode, em); /* once for the rb tree */ free_extent_map(em); next: start = extent_map_end(em); - write_unlock(&map->lock); + write_unlock(&extent_tree->lock); /* once for us */ free_extent_map(em); @@ -2470,7 +2470,7 @@ next: cond_resched(); /* Allow large-extent preemption. */ } } - return try_release_extent_state(tree, page, mask); + return try_release_extent_state(io_tree, page, mask); } struct btrfs_fiemap_entry { -- cgit v1.2.3-59-g8ed1b From 85d288309ab5463140a2d00b3827262fb14e7db4 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 16 Apr 2024 15:13:03 +0100 Subject: btrfs: use btrfs_get_fs_generation() at try_release_extent_mapping() Nowadays we have the btrfs_get_fs_generation() to get the current generation of the filesystem, so there's no need anymore to lock the transaction spinlock to read it. Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 6438c3e74756..f689c53553e3 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2406,8 +2406,7 @@ int try_release_extent_mapping(struct page *page, gfp_t mask) page->mapping->host->i_size > SZ_16M) { u64 len; while (start <= end) { - struct btrfs_fs_info *fs_info; - u64 cur_gen; + const u64 cur_gen = btrfs_get_fs_generation(inode->root->fs_info); len = end - start + 1; write_lock(&extent_tree->lock); @@ -2442,10 +2441,6 @@ int try_release_extent_mapping(struct page *page, gfp_t mask) * Otherwise don't remove it, we could be racing with an * ongoing fast fsync that could miss the new extent. */ - fs_info = inode->root->fs_info; - spin_lock(&fs_info->trans_lock); - cur_gen = fs_info->generation; - spin_unlock(&fs_info->trans_lock); if (em->generation >= cur_gen) goto next; remove_em: -- cgit v1.2.3-59-g8ed1b From 433a3e01dda1d463159a9620b40ba027514f0ea5 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 16 Apr 2024 15:19:03 +0100 Subject: btrfs: remove i_size restriction at try_release_extent_mapping() Currently we don't attempt to release extent maps if the inode has an i_size that is not greater than 16M. This condition was added way back in 2008 by commit 70dec8079d78 ("Btrfs: extent_io and extent_state optimizations"), without any explanation about it. A quick chat with Chris on slack revealed that the goal was probably to release the extent maps for small files only when closing the inode. This however can be harmful in case we have tons of such files being kept open for very long periods of time, since we will consume more and more pages for extent maps. So remove the condition. Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index f689c53553e3..ff9132b897e3 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2402,8 +2402,7 @@ int try_release_extent_mapping(struct page *page, gfp_t mask) struct extent_io_tree *io_tree = &inode->io_tree; struct extent_map_tree *extent_tree = &inode->extent_tree; - if (gfpflags_allow_blocking(mask) && - page->mapping->host->i_size > SZ_16M) { + if (gfpflags_allow_blocking(mask)) { u64 len; while (start <= end) { const u64 cur_gen = btrfs_get_fs_generation(inode->root->fs_info); -- cgit v1.2.3-59-g8ed1b From 2e504418e4645302c40982a64de6a6979ec5489d Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 16 Apr 2024 15:34:51 +0100 Subject: btrfs: be better releasing extent maps at try_release_extent_mapping() At try_release_extent_mapping(), called during the release folio callback (btrfs_release_folio() callchain), we don't release any extent maps in the range if the GFP flags don't allow blocking. This behaviour is exaggerated because: 1) Both searching for extent maps and removing them are not blocking operations. The only thing that it is the cond_resched() call at the end of the loop that searches for and removes extent maps; 2) We currently only operate on a single page, so for the case where block size matches the page size, we can only have one extent map, and for the case where the block size is smaller than the page size, we can have at most 16 extent maps. So it's very unlikely the cond_resched() call will ever block even in the block size smaller than page size scenario. So instead of not removing any extent maps at all in case the GFP glags don't allow blocking, keep removing extent maps while we don't need to reschedule. This makes it safe for the subpage case and for a future where we can process folios with a size larger than a page. Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 120 ++++++++++++++++++++++++++------------------------- 1 file changed, 61 insertions(+), 59 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index ff9132b897e3..2230e6b6ba95 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2395,73 +2395,75 @@ static int try_release_extent_state(struct extent_io_tree *tree, */ int try_release_extent_mapping(struct page *page, gfp_t mask) { - struct extent_map *em; u64 start = page_offset(page); u64 end = start + PAGE_SIZE - 1; struct btrfs_inode *inode = page_to_inode(page); struct extent_io_tree *io_tree = &inode->io_tree; - struct extent_map_tree *extent_tree = &inode->extent_tree; - - if (gfpflags_allow_blocking(mask)) { - u64 len; - while (start <= end) { - const u64 cur_gen = btrfs_get_fs_generation(inode->root->fs_info); - - len = end - start + 1; - write_lock(&extent_tree->lock); - em = lookup_extent_mapping(extent_tree, start, len); - if (!em) { - write_unlock(&extent_tree->lock); - break; - } - if ((em->flags & EXTENT_FLAG_PINNED) || - em->start != start) { - write_unlock(&extent_tree->lock); - free_extent_map(em); - break; - } - if (test_range_bit_exists(io_tree, em->start, - extent_map_end(em) - 1, - EXTENT_LOCKED)) - goto next; - /* - * If it's not in the list of modified extents, used - * by a fast fsync, we can remove it. If it's being - * logged we can safely remove it since fsync took an - * extra reference on the em. - */ - if (list_empty(&em->list) || - (em->flags & EXTENT_FLAG_LOGGING)) - goto remove_em; - /* - * If it's in the list of modified extents, remove it - * only if its generation is older then the current one, - * in which case we don't need it for a fast fsync. - * Otherwise don't remove it, we could be racing with an - * ongoing fast fsync that could miss the new extent. - */ - if (em->generation >= cur_gen) - goto next; -remove_em: - /* - * We only remove extent maps that are not in the list of - * modified extents or that are in the list but with a - * generation lower then the current generation, so there - * is no need to set the full fsync flag on the inode (it - * hurts the fsync performance for workloads with a data - * size that exceeds or is close to the system's memory). - */ - remove_extent_mapping(inode, em); - /* once for the rb tree */ + + while (start <= end) { + const u64 cur_gen = btrfs_get_fs_generation(inode->root->fs_info); + const u64 len = end - start + 1; + struct extent_map_tree *extent_tree = &inode->extent_tree; + struct extent_map *em; + + write_lock(&extent_tree->lock); + em = lookup_extent_mapping(extent_tree, start, len); + if (!em) { + write_unlock(&extent_tree->lock); + break; + } + if ((em->flags & EXTENT_FLAG_PINNED) || em->start != start) { + write_unlock(&extent_tree->lock); free_extent_map(em); + break; + } + if (test_range_bit_exists(io_tree, em->start, + extent_map_end(em) - 1, EXTENT_LOCKED)) + goto next; + /* + * If it's not in the list of modified extents, used by a fast + * fsync, we can remove it. If it's being logged we can safely + * remove it since fsync took an extra reference on the em. + */ + if (list_empty(&em->list) || (em->flags & EXTENT_FLAG_LOGGING)) + goto remove_em; + /* + * If it's in the list of modified extents, remove it only if + * its generation is older then the current one, in which case + * we don't need it for a fast fsync. Otherwise don't remove it, + * we could be racing with an ongoing fast fsync that could miss + * the new extent. + */ + if (em->generation >= cur_gen) + goto next; +remove_em: + /* + * We only remove extent maps that are not in the list of + * modified extents or that are in the list but with a + * generation lower then the current generation, so there is no + * need to set the full fsync flag on the inode (it hurts the + * fsync performance for workloads with a data size that exceeds + * or is close to the system's memory). + */ + remove_extent_mapping(inode, em); + /* Once for the inode's extent map tree. */ + free_extent_map(em); next: - start = extent_map_end(em); - write_unlock(&extent_tree->lock); + start = extent_map_end(em); + write_unlock(&extent_tree->lock); - /* once for us */ - free_extent_map(em); + /* Once for us, for the lookup_extent_mapping() reference. */ + free_extent_map(em); + + if (need_resched()) { + /* + * If we need to resched but we can't block just exit + * and leave any remaining extent maps. + */ + if (!gfpflags_allow_blocking(mask)) + break; - cond_resched(); /* Allow large-extent preemption. */ + cond_resched(); } } return try_release_extent_state(io_tree, page, mask); -- cgit v1.2.3-59-g8ed1b From de6f14e83e6221e3ef7e949deabe041240bc1829 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 16 Apr 2024 20:52:30 +0100 Subject: btrfs: make try_release_extent_mapping() return a bool Currently try_release_extent_mapping() as an int return type, but we use it as a boolean. Its only caller, the release folio callback, also returns a boolean which corresponds to try_release_extent_mapping()'s return value. So change its return value type to bool as well as its helper try_release_extent_state(). Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 17 +++++++++-------- fs/btrfs/extent_io.h | 2 +- fs/btrfs/inode.c | 7 +++---- 3 files changed, 13 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 2230e6b6ba95..a9f9f5abdf53 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2355,19 +2355,20 @@ int extent_invalidate_folio(struct extent_io_tree *tree, * are locked or under IO and drops the related state bits if it is safe * to drop the page. */ -static int try_release_extent_state(struct extent_io_tree *tree, +static bool try_release_extent_state(struct extent_io_tree *tree, struct page *page, gfp_t mask) { u64 start = page_offset(page); u64 end = start + PAGE_SIZE - 1; - int ret = 1; + bool ret; if (test_range_bit_exists(tree, start, end, EXTENT_LOCKED)) { - ret = 0; + ret = false; } else { u32 clear_bits = ~(EXTENT_LOCKED | EXTENT_NODATASUM | EXTENT_DELALLOC_NEW | EXTENT_CTLBITS | EXTENT_QGROUP_RESERVED); + int ret2; /* * At this point we can safely clear everything except the @@ -2375,15 +2376,15 @@ static int try_release_extent_state(struct extent_io_tree *tree, * The delalloc new bit will be cleared by ordered extent * completion. */ - ret = __clear_extent_bit(tree, start, end, clear_bits, NULL, NULL); + ret2 = __clear_extent_bit(tree, start, end, clear_bits, NULL, NULL); /* if clear_extent_bit failed for enomem reasons, * we can't allow the release to continue. */ - if (ret < 0) - ret = 0; + if (ret2 < 0) + ret = false; else - ret = 1; + ret = true; } return ret; } @@ -2393,7 +2394,7 @@ static int try_release_extent_state(struct extent_io_tree *tree, * in the range corresponding to the page, both state records and extent * map records are removed */ -int try_release_extent_mapping(struct page *page, gfp_t mask) +bool try_release_extent_mapping(struct page *page, gfp_t mask) { u64 start = page_offset(page); u64 end = start + PAGE_SIZE - 1; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index c81a9b546c9f..f38397765e90 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -230,7 +230,7 @@ static inline void extent_changeset_free(struct extent_changeset *changeset) kfree(changeset); } -int try_release_extent_mapping(struct page *page, gfp_t mask); +bool try_release_extent_mapping(struct page *page, gfp_t mask); int try_release_extent_buffer(struct page *page); int btrfs_read_folio(struct file *file, struct folio *folio); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1dde8085271e..eb0dc913c33b 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7899,13 +7899,12 @@ static void wait_subpage_spinlock(struct page *page) static bool __btrfs_release_folio(struct folio *folio, gfp_t gfp_flags) { - int ret = try_release_extent_mapping(&folio->page, gfp_flags); - - if (ret == 1) { + if (try_release_extent_mapping(&folio->page, gfp_flags)) { wait_subpage_spinlock(&folio->page); clear_page_extent_mapped(&folio->page); + return true; } - return ret; + return false; } static bool btrfs_release_folio(struct folio *folio, gfp_t gfp_flags) -- cgit v1.2.3-59-g8ed1b From 905a95f3dd25abba64cad3e7a96947fcb9bf7006 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 17 Apr 2024 16:06:13 +0100 Subject: btrfs: initialize delayed inodes xarray without GFP_ATOMIC There's no need to initialize the delayed inodes xarray with a GFP_ATOMIC flag because that actually does nothing on the xarray operations. That was needed for radix trees, but for xarrays the allocation flags are passed as the last argument to xa_store() (which we are using correctly). So initialize the delayed inodes xarray with a simple xa_init(). Reviewed-by: Josef Bacik Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index c2dc88f909b0..3606fb18531e 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -663,8 +663,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, root->nr_delalloc_inodes = 0; root->nr_ordered_extents = 0; root->inode_tree = RB_ROOT; - /* GFP flags are compatible with XA_FLAGS_*. */ - xa_init_flags(&root->delayed_nodes, GFP_ATOMIC); + xa_init(&root->delayed_nodes); btrfs_init_root_block_rsv(root); -- cgit v1.2.3-59-g8ed1b From fef998d1a0b00ab35658484060354213a2c06c81 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Thu, 18 Apr 2024 00:47:13 +0200 Subject: btrfs: use btrfs_is_testing() everywhere There are open coded tests of BTRFS_FS_STATE_DUMMY_FS_INFO and we have a wrapper for that that's a compile-time constant when self-tests are not built in. As this is only for development we can save some bytes and conditions on release configs by using the helper in the remaining cases. Reviewed-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 4 ++-- fs/btrfs/sysfs.c | 8 ++++---- fs/btrfs/tests/btrfs-tests.c | 3 +-- fs/btrfs/tree-checker.c | 2 +- 4 files changed, 8 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 3606fb18531e..e0bdf3ed2449 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -646,7 +646,7 @@ struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr, static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, u64 objectid) { - bool dummy = test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state); + bool dummy = btrfs_is_testing(fs_info); memset(&root->root_key, 0, sizeof(root->root_key)); memset(&root->root_item, 0, sizeof(root->root_item)); @@ -1075,7 +1075,7 @@ static struct btrfs_root *read_tree_root_path(struct btrfs_root *tree_root, * For real fs, and not log/reloc trees, root owner must * match its root node owner */ - if (!test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state) && + if (!btrfs_is_testing(fs_info) && btrfs_root_id(root) != BTRFS_TREE_LOG_OBJECTID && btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID && btrfs_root_id(root) != btrfs_header_owner(root->node)) { diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index c6387a8ddb94..af545b6b1190 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -2339,7 +2339,7 @@ int btrfs_sysfs_add_one_qgroup(struct btrfs_fs_info *fs_info, struct kobject *qgroups_kobj = fs_info->qgroups_kobj; int ret; - if (test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state)) + if (btrfs_is_testing(fs_info)) return 0; if (qgroup->kobj.state_initialized) return 0; @@ -2360,7 +2360,7 @@ void btrfs_sysfs_del_qgroups(struct btrfs_fs_info *fs_info) struct btrfs_qgroup *qgroup; struct btrfs_qgroup *next; - if (test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state)) + if (btrfs_is_testing(fs_info)) return; rbtree_postorder_for_each_entry_safe(qgroup, next, @@ -2381,7 +2381,7 @@ int btrfs_sysfs_add_qgroups(struct btrfs_fs_info *fs_info) struct btrfs_qgroup *next; int ret = 0; - if (test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state)) + if (btrfs_is_testing(fs_info)) return 0; ASSERT(fsid_kobj); @@ -2413,7 +2413,7 @@ out: void btrfs_sysfs_del_one_qgroup(struct btrfs_fs_info *fs_info, struct btrfs_qgroup *qgroup) { - if (test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state)) + if (btrfs_is_testing(fs_info)) return; if (qgroup->kobj.state_initialized) { diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c index 709c6cc9706a..dce0387ef155 100644 --- a/fs/btrfs/tests/btrfs-tests.c +++ b/fs/btrfs/tests/btrfs-tests.c @@ -160,8 +160,7 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info) if (!fs_info) return; - if (WARN_ON(!test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, - &fs_info->fs_state))) + if (WARN_ON(!btrfs_is_testing(fs_info))) return; test_mnt->mnt_sb->s_fs_info = NULL; diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 32604e9b31c3..a2c3651a3d8f 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -2021,7 +2021,7 @@ int btrfs_check_eb_owner(const struct extent_buffer *eb, u64 root_owner) * Skip dummy fs, as selftests don't create unique ebs for each dummy * root. */ - if (test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &eb->fs_info->fs_state)) + if (btrfs_is_testing(eb->fs_info)) return 0; /* * There are several call sites (backref walking, qgroup, and data -- cgit v1.2.3-59-g8ed1b From 9a7b68d32afc4e92909c21e166ad993801236be3 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 16 Apr 2024 10:06:58 +0800 Subject: btrfs: report filemap_fdata_range() error In the function btrfs_write_marked_extents() and in __btrfs_wait_marked_extents() return the actual error if when filemap_fdata_range() fails. Suggested-by: Josef Bacik Reviewed-by: Qu Wenruo Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 3e3bcc5f64c6..8c3b3cda1390 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1156,6 +1156,8 @@ int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info, else if (wait_writeback) werr = filemap_fdatawait_range(mapping, start, end); free_extent_state(cached_state); + if (werr) + break; cached_state = NULL; cond_resched(); start = end + 1; @@ -1198,6 +1200,8 @@ static int __btrfs_wait_marked_extents(struct btrfs_fs_info *fs_info, if (err) werr = err; free_extent_state(cached_state); + if (werr) + break; cached_state = NULL; cond_resched(); start = end + 1; -- cgit v1.2.3-59-g8ed1b From ce87531120aea7953706c63f1622ebcdfcd257d1 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:20 +0530 Subject: btrfs: rename werr and err to ret in btrfs_write_marked_extents() Rename the function's local variable werr and err to ret. Reviewed-by: Josef Bacik Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 8c3b3cda1390..defdb0979d68 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1118,8 +1118,7 @@ int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans) int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info, struct extent_io_tree *dirty_pages, int mark) { - int err = 0; - int werr = 0; + int ret = 0; struct address_space *mapping = fs_info->btree_inode->i_mapping; struct extent_state *cached_state = NULL; u64 start = 0; @@ -1129,7 +1128,7 @@ int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info, mark, &cached_state)) { bool wait_writeback = false; - err = convert_extent_bit(dirty_pages, start, end, + ret = convert_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT, mark, &cached_state); /* @@ -1145,24 +1144,22 @@ int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info, * We cleanup any entries left in the io tree when committing * the transaction (through extent_io_tree_release()). */ - if (err == -ENOMEM) { - err = 0; + if (ret == -ENOMEM) { + ret = 0; wait_writeback = true; } - if (!err) - err = filemap_fdatawrite_range(mapping, start, end); - if (err) - werr = err; - else if (wait_writeback) - werr = filemap_fdatawait_range(mapping, start, end); + if (!ret) + ret = filemap_fdatawrite_range(mapping, start, end); + if (!ret && wait_writeback) + ret = filemap_fdatawait_range(mapping, start, end); free_extent_state(cached_state); - if (werr) + if (ret) break; cached_state = NULL; cond_resched(); start = end + 1; } - return werr; + return ret; } /* -- cgit v1.2.3-59-g8ed1b From 1e8a42375f5f16bba814f132d802b3cf2ac357bd Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:21 +0530 Subject: btrfs: rename werr and err to ret in __btrfs_wait_marked_extents() Rename the function's local return variables err and werr to ret. Also, align the variable declarations with the other declarations in the function for better function space alignment. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 21 ++++++++------------- 1 file changed, 8 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index defdb0979d68..3388c836b9a5 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1171,12 +1171,11 @@ int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info, static int __btrfs_wait_marked_extents(struct btrfs_fs_info *fs_info, struct extent_io_tree *dirty_pages) { - int err = 0; - int werr = 0; struct address_space *mapping = fs_info->btree_inode->i_mapping; struct extent_state *cached_state = NULL; u64 start = 0; u64 end; + int ret = 0; while (find_first_extent_bit(dirty_pages, start, &start, &end, EXTENT_NEED_WAIT, &cached_state)) { @@ -1188,24 +1187,20 @@ static int __btrfs_wait_marked_extents(struct btrfs_fs_info *fs_info, * concurrently - we do it only at transaction commit time when * it's safe to do it (through extent_io_tree_release()). */ - err = clear_extent_bit(dirty_pages, start, end, + ret = clear_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT, &cached_state); - if (err == -ENOMEM) - err = 0; - if (!err) - err = filemap_fdatawait_range(mapping, start, end); - if (err) - werr = err; + if (ret == -ENOMEM) + ret = 0; + if (!ret) + ret = filemap_fdatawait_range(mapping, start, end); free_extent_state(cached_state); - if (werr) + if (ret) break; cached_state = NULL; cond_resched(); start = end + 1; } - if (err) - werr = err; - return werr; + return ret; } static int btrfs_wait_extents(struct btrfs_fs_info *fs_info, -- cgit v1.2.3-59-g8ed1b From 2daca1e4194eb5651d3516d8c0c6e12ce4674e2a Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:22 +0530 Subject: btrfs: rename err and ret to ret in build_backref_tree() Code style fix in the function build_backref_tree(). Drop the ret initialization 0, as we don't need it. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/relocation.c | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 7e7799b4560b..aef7d286252b 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -473,20 +473,19 @@ static noinline_for_stack struct btrfs_backref_node *build_backref_tree( struct btrfs_backref_node *node = NULL; struct btrfs_backref_edge *edge; int ret; - int err = 0; iter = btrfs_backref_iter_alloc(rc->extent_root->fs_info); if (!iter) return ERR_PTR(-ENOMEM); path = btrfs_alloc_path(); if (!path) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } node = btrfs_backref_alloc_node(cache, bytenr, level); if (!node) { - err = -ENOMEM; + ret = -ENOMEM; goto out; } @@ -497,10 +496,9 @@ static noinline_for_stack struct btrfs_backref_node *build_backref_tree( do { ret = btrfs_backref_add_tree_node(trans, cache, path, iter, node_key, cur); - if (ret < 0) { - err = ret; + if (ret < 0) goto out; - } + edge = list_first_entry_or_null(&cache->pending_edge, struct btrfs_backref_edge, list[UPPER]); /* @@ -515,10 +513,8 @@ static noinline_for_stack struct btrfs_backref_node *build_backref_tree( /* Finish the upper linkage of newly added edges/nodes */ ret = btrfs_backref_finish_upper_links(cache, node); - if (ret < 0) { - err = ret; + if (ret < 0) goto out; - } if (handle_useless_nodes(rc, node)) node = NULL; @@ -526,9 +522,9 @@ out: btrfs_free_path(iter->path); kfree(iter); btrfs_free_path(path); - if (err) { + if (ret) { btrfs_backref_error_cleanup(cache, node); - return ERR_PTR(err); + return ERR_PTR(ret); } ASSERT(!node || !node->detached); ASSERT(list_empty(&cache->useless_node) && -- cgit v1.2.3-59-g8ed1b From acde0e86093a7aa3b25250cd4f43f4c3fb780c8a Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:23 +0530 Subject: btrfs: reuse ret instead of err in relocate_tree_blocks() Coding style fixes the function relocate_tree_blocks(). After the fix, ret is the return value variable. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/relocation.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index aef7d286252b..bd573a0ec270 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2742,12 +2742,11 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans, struct btrfs_path *path; struct tree_block *block; struct tree_block *next; - int ret; - int err = 0; + int ret = 0; path = btrfs_alloc_path(); if (!path) { - err = -ENOMEM; + ret = -ENOMEM; goto out_free_blocks; } @@ -2762,8 +2761,8 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans, /* Get first keys */ rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) { if (!block->key_ready) { - err = get_tree_block_key(fs_info, block); - if (err) + ret = get_tree_block_key(fs_info, block); + if (ret) goto out_free_path; } } @@ -2773,25 +2772,23 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans, node = build_backref_tree(trans, rc, &block->key, block->level, block->bytenr); if (IS_ERR(node)) { - err = PTR_ERR(node); + ret = PTR_ERR(node); goto out; } ret = relocate_tree_block(trans, rc, node, &block->key, path); - if (ret < 0) { - err = ret; + if (ret < 0) break; - } } out: - err = finish_pending_nodes(trans, rc, path, err); + ret = finish_pending_nodes(trans, rc, path, ret); out_free_path: btrfs_free_path(path); out_free_blocks: free_block_list(blocks); - return err; + return ret; } static noinline_for_stack int prealloc_file_extent_cluster( -- cgit v1.2.3-59-g8ed1b From 5e8fb9b84b0b460659d3e037fcc61148601c0aea Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:29 +0530 Subject: btrfs: drop variable err in quick_update_accounting() In quick_update_accounting() err is used as 2nd return value, which could be achieved just with ret. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 6488d9b888e4..c44fb306f99f 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1536,18 +1536,15 @@ static int quick_update_accounting(struct btrfs_fs_info *fs_info, { struct btrfs_qgroup *qgroup; int ret = 1; - int err = 0; qgroup = find_qgroup_rb(fs_info, src); if (!qgroup) goto out; if (qgroup->excl == qgroup->rfer) { - ret = 0; - err = __qgroup_excl_accounting(fs_info, dst, qgroup, sign); - if (err < 0) { - ret = err; + ret = __qgroup_excl_accounting(fs_info, dst, qgroup, sign); + if (ret < 0) goto out; - } + ret = 0; } out: if (ret) -- cgit v1.2.3-59-g8ed1b From 6e812a9c656b32d9ec3d71ab58376d3d4f457e06 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:30 +0530 Subject: btrfs: rename return variables in btrfs_qgroup_rescan_worker() Rename ret to ret2 compile and then err to ret. Also, new ret2 is found to be localized within the 'if (trans)' statement, so move its declaration there. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index c44fb306f99f..1768fc6f232f 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3700,7 +3700,6 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work) qgroup_rescan_work); struct btrfs_path *path; struct btrfs_trans_handle *trans = NULL; - int err = -ENOMEM; int ret = 0; bool stopped = false; bool did_leaf_rescans = false; @@ -3709,8 +3708,10 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work) return; path = btrfs_alloc_path(); - if (!path) + if (!path) { + ret = -ENOMEM; goto out; + } /* * Rescan should only search for commit root, and any later difference * should be recorded by qgroup @@ -3718,18 +3719,17 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work) path->search_commit_root = 1; path->skip_locking = 1; - err = 0; - while (!err && !(stopped = rescan_should_stop(fs_info))) { + while (!ret && !(stopped = rescan_should_stop(fs_info))) { trans = btrfs_start_transaction(fs_info->fs_root, 0); if (IS_ERR(trans)) { - err = PTR_ERR(trans); + ret = PTR_ERR(trans); break; } - err = qgroup_rescan_leaf(trans, path); + ret = qgroup_rescan_leaf(trans, path); did_leaf_rescans = true; - if (err > 0) + if (ret > 0) btrfs_commit_transaction(trans); else btrfs_end_transaction(trans); @@ -3739,10 +3739,10 @@ out: btrfs_free_path(path); mutex_lock(&fs_info->qgroup_rescan_lock); - if (err > 0 && + if (ret > 0 && fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT) { fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT; - } else if (err < 0 || stopped) { + } else if (ret < 0 || stopped) { fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT; } mutex_unlock(&fs_info->qgroup_rescan_lock); @@ -3757,11 +3757,11 @@ out: if (did_leaf_rescans) { trans = btrfs_start_transaction(fs_info->quota_root, 1); if (IS_ERR(trans)) { - err = PTR_ERR(trans); + ret = PTR_ERR(trans); trans = NULL; btrfs_err(fs_info, "fail to start transaction for status update: %d", - err); + ret); } } else { trans = NULL; @@ -3772,11 +3772,11 @@ out: fs_info->qgroup_flags & BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN) fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_RESCAN; if (trans) { - ret = update_qgroup_status_item(trans); - if (ret < 0) { - err = ret; - btrfs_err(fs_info, "fail to update qgroup status: %d", - err); + int ret2 = update_qgroup_status_item(trans); + + if (ret2 < 0) { + ret = ret2; + btrfs_err(fs_info, "fail to update qgroup status: %d", ret); } } fs_info->qgroup_rescan_running = false; @@ -3793,11 +3793,11 @@ out: btrfs_info(fs_info, "qgroup scan paused"); } else if (fs_info->qgroup_flags & BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN) { btrfs_info(fs_info, "qgroup scan cancelled"); - } else if (err >= 0) { + } else if (ret >= 0) { btrfs_info(fs_info, "qgroup scan completed%s", - err > 0 ? " (inconsistency flag cleared)" : ""); + ret > 0 ? " (inconsistency flag cleared)" : ""); } else { - btrfs_err(fs_info, "qgroup scan failed with %d", err); + btrfs_err(fs_info, "qgroup scan failed with %d", ret); } } -- cgit v1.2.3-59-g8ed1b From 1618aa3c2e0163f5ac34d514ae89474521910536 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:31 +0530 Subject: btrfs: simplify return variables in lookup_extent_data_ref() First, drop err instead reuse ret, choose to return the error instead of goto fail and then return the same error. Do not initialize the ret until where it has to be initialized. Slight logic change in handling the btrfs_search_slot() and btrfs_next_leaf() return value. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 29 ++++++++++++++--------------- 1 file changed, 14 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 023920d0d971..78dc94a97e35 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -446,9 +446,8 @@ static noinline int lookup_extent_data_ref(struct btrfs_trans_handle *trans, struct btrfs_extent_data_ref *ref; struct extent_buffer *leaf; u32 nritems; - int ret; int recow; - int err = -ENOENT; + int ret; key.objectid = bytenr; if (parent) { @@ -462,26 +461,26 @@ static noinline int lookup_extent_data_ref(struct btrfs_trans_handle *trans, again: recow = 0; ret = btrfs_search_slot(trans, root, &key, path, -1, 1); - if (ret < 0) { - err = ret; - goto fail; - } + if (ret < 0) + return ret; if (parent) { - if (!ret) - return 0; - goto fail; + if (ret) + return -ENOENT; + return 0; } + ret = -ENOENT; leaf = path->nodes[0]; nritems = btrfs_header_nritems(leaf); while (1) { if (path->slots[0] >= nritems) { ret = btrfs_next_leaf(root, path); - if (ret < 0) - err = ret; - if (ret) - goto fail; + if (ret) { + if (ret > 1) + return -ENOENT; + return ret; + } leaf = path->nodes[0]; nritems = btrfs_header_nritems(leaf); @@ -502,13 +501,13 @@ again: btrfs_release_path(path); goto again; } - err = 0; + ret = 0; break; } path->slots[0]++; } fail: - return err; + return ret; } static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans, -- cgit v1.2.3-59-g8ed1b From 33a44f3760509c6b4df595774f1232f4c1b81e65 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 19 Mar 2024 20:25:33 +0530 Subject: btrfs: simplify return variables in btrfs_drop_subtree() There's another return variable wret that is only passed to ret on error, we can simply use ret. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 78dc94a97e35..47d48233b592 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -6099,7 +6099,6 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans, int level; int parent_level; int ret = 0; - int wret; BUG_ON(btrfs_root_id(root) != BTRFS_TREE_RELOC_OBJECTID); @@ -6135,17 +6134,16 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans, wc->reada_count = BTRFS_NODEPTRS_PER_BLOCK(fs_info); while (1) { - wret = walk_down_tree(trans, root, path, wc); - if (wret < 0) { - ret = wret; + ret = walk_down_tree(trans, root, path, wc); + if (ret < 0) break; - } - wret = walk_up_tree(trans, root, path, wc, parent_level); - if (wret < 0) - ret = wret; - if (wret != 0) + ret = walk_up_tree(trans, root, path, wc, parent_level); + if (ret) { + if (ret > 0) + ret = 0; break; + } } kfree(wc); -- cgit v1.2.3-59-g8ed1b From 30704a0d562bbed00b3ca4bb14108d75c3ee9108 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Tue, 23 Apr 2024 12:10:29 +0200 Subject: btrfs: drop unused argument of calcu_metadata_size() calcu_metadata_size() has a "reserve" argument, but the only caller always set it to "1". The other usage (reserve = 0) is dropped by a commit 0647bf564f1e ("Btrfs: improve forever loop when doing balance relocation"), which is more than 10 years ago. Drop the argument and simplify the code. Reviewed-by: Qu Wenruo Reviewed-by: Johannes Thumshirn Signed-off-by: Naohiro Aota Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/relocation.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index bd573a0ec270..5cf5718c3b1f 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2280,9 +2280,8 @@ struct btrfs_root *select_one_root(struct btrfs_backref_node *node) return fs_root; } -static noinline_for_stack -u64 calcu_metadata_size(struct reloc_control *rc, - struct btrfs_backref_node *node, int reserve) +static noinline_for_stack u64 calcu_metadata_size(struct reloc_control *rc, + struct btrfs_backref_node *node) { struct btrfs_fs_info *fs_info = rc->extent_root->fs_info; struct btrfs_backref_node *next = node; @@ -2291,12 +2290,12 @@ u64 calcu_metadata_size(struct reloc_control *rc, u64 num_bytes = 0; int index = 0; - BUG_ON(reserve && node->processed); + BUG_ON(node->processed); while (next) { cond_resched(); while (1) { - if (next->processed && (reserve || next != node)) + if (next->processed) break; num_bytes += fs_info->nodesize; @@ -2324,7 +2323,7 @@ static int reserve_metadata_space(struct btrfs_trans_handle *trans, int ret; u64 tmp; - num_bytes = calcu_metadata_size(rc, node, 1) * 2; + num_bytes = calcu_metadata_size(rc, node) * 2; trans->block_rsv = rc->block_rsv; rc->reserved_bytes += num_bytes; -- cgit v1.2.3-59-g8ed1b From 319d91ee7265505d96f085c6bf5b781667b71090 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Tue, 2 Apr 2024 16:00:21 +1030 Subject: btrfs: add extra comments on extent_map members The extent_map structure is very critical to btrfs, as it is involved for both read and write paths. Unfortunately the structure is not properly explained, making it pretty hard to understand nor to do further improvement. This patch adds extra comments explaining the major members based on my code reading. Hopefully we can find more members to cleanup in the future. Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Signed-off-by: David Sterba --- fs/btrfs/extent_map.h | 55 ++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 54 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index 8306a8f294d0..6d587111f73a 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -35,19 +35,72 @@ enum { }; /* + * This structure represents file extents and holes. + * + * Unlike on-disk file extent items, extent maps can be merged to save memory. + * This means members only match file extent items before any merging. + * * Keep this structure as compact as possible, as we can have really large * amounts of allocated extent maps at any time. */ struct extent_map { struct rb_node rb_node; - /* all of these are in bytes */ + /* All of these are in bytes. */ + + /* File offset matching the offset of a BTRFS_EXTENT_ITEM_KEY key. */ u64 start; + + /* + * Length of the file extent. + * + * For non-inlined file extents it's btrfs_file_extent_item::num_bytes. + * For inline extents it's sectorsize, since inline data starts at + * offsetof(struct btrfs_file_extent_item, disk_bytenr) thus + * btrfs_file_extent_item::num_bytes is not valid. + */ u64 len; + + /* + * The file offset of the original file extent before splitting. + * + * This is an in-memory only member, matching + * extent_map::start - btrfs_file_extent_item::offset for + * regular/preallocated extents. EXTENT_MAP_HOLE otherwise. + */ u64 orig_start; + + /* + * The full on-disk extent length, matching + * btrfs_file_extent_item::disk_num_bytes. + */ u64 orig_block_len; + + /* + * The decompressed size of the whole on-disk extent, matching + * btrfs_file_extent_item::ram_bytes. + */ u64 ram_bytes; + + /* + * The on-disk logical bytenr for the file extent. + * + * For compressed extents it matches btrfs_file_extent_item::disk_bytenr. + * For uncompressed extents it matches + * btrfs_file_extent_item::disk_bytenr + btrfs_file_extent_item::offset + * + * For holes it is EXTENT_MAP_HOLE and for inline extents it is + * EXTENT_MAP_INLINE. + */ u64 block_start; + + /* + * The on-disk length for the file extent. + * + * For compressed extents it matches btrfs_file_extent_item::disk_num_bytes. + * For uncompressed extents it matches extent_map::len. + * For holes and inline extents it's -1 and shouldn't be used. + */ u64 block_len; /* -- cgit v1.2.3-59-g8ed1b From 4bdc558bf906ce4e3e33cb33a9368383a22e2151 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Tue, 2 Apr 2024 16:30:15 +1030 Subject: btrfs: simplify the inline extent map creation With the tree-checker ensuring all inline file extents starts at file offset 0 and has a length no larger than sectorsize, we can simplify the calculation to assigned those fixes values directly. Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/file-item.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 5fead7a253df..bce95f871750 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -1279,20 +1279,19 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode, struct extent_buffer *leaf = path->nodes[0]; const int slot = path->slots[0]; struct btrfs_key key; - u64 extent_start, extent_end; + u64 extent_start; u64 bytenr; u8 type = btrfs_file_extent_type(leaf, fi); int compress_type = btrfs_file_extent_compression(leaf, fi); btrfs_item_key_to_cpu(leaf, &key, slot); extent_start = key.offset; - extent_end = btrfs_file_extent_end(path); em->ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); em->generation = btrfs_file_extent_generation(leaf, fi); if (type == BTRFS_FILE_EXTENT_REG || type == BTRFS_FILE_EXTENT_PREALLOC) { em->start = extent_start; - em->len = extent_end - extent_start; + em->len = btrfs_file_extent_end(path) - extent_start; em->orig_start = extent_start - btrfs_file_extent_offset(leaf, fi); em->orig_block_len = btrfs_file_extent_disk_num_bytes(leaf, fi); @@ -1313,9 +1312,12 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode, em->flags |= EXTENT_FLAG_PREALLOC; } } else if (type == BTRFS_FILE_EXTENT_INLINE) { + /* Tree-checker has ensured this. */ + ASSERT(extent_start == 0); + em->block_start = EXTENT_MAP_INLINE; - em->start = extent_start; - em->len = extent_end - extent_start; + em->start = 0; + em->len = fs_info->sectorsize; /* * Initialize orig_start and block_len with the same values * as in inode.c:btrfs_get_extent(). @@ -1348,12 +1350,10 @@ u64 btrfs_file_extent_end(const struct btrfs_path *path) ASSERT(key.type == BTRFS_EXTENT_DATA_KEY); fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); - if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) { - end = btrfs_file_extent_ram_bytes(leaf, fi); - end = ALIGN(key.offset + end, leaf->fs_info->sectorsize); - } else { + if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) + end = leaf->fs_info->sectorsize; + else end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); - } return end; } -- cgit v1.2.3-59-g8ed1b From e98bf64f7aaca09d5e6780f60ca79fda3003aecb Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Wed, 3 Apr 2024 10:30:06 +1030 Subject: btrfs: add extra sanity checks for create_io_em() The function create_io_em() is called before we submit an IO, to update the in-memory extent map for the involved range. This patch changes the following aspects: - Does not allow BTRFS_ORDERED_NOCOW type For real NOCOW (excluding NOCOW writes into preallocated ranges) writes, we never call create_io_em(), as we does not need to update the extent map at all. So remove the sanity check allowing BTRFS_ORDERED_NOCOW type. - Add extra sanity checks * PREALLOC - @block_len == len For uncompressed writes. * REGULAR - @block_len == @orig_block_len == @ram_bytes == @len We're creating a new uncompressed extent, and referring all of it. - @orig_start == @start We haven no offset inside the extent. * COMPRESSED - valid @compress_type - @len <= @ram_bytes This is to co-operate with encoded writes, which can cause a new file extent referring only part of a uncompressed extent. Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 40 +++++++++++++++++++++++++++++++++++++++- 1 file changed, 39 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index eb0dc913c33b..3d587b517809 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7258,11 +7258,49 @@ static struct extent_map *create_io_em(struct btrfs_inode *inode, u64 start, struct extent_map *em; int ret; + /* + * Note the missing NOCOW type. + * + * For pure NOCOW writes, we should not create an io extent map, but + * just reusing the existing one. + * Only PREALLOC writes (NOCOW write into preallocated range) can + * create an io extent map. + */ ASSERT(type == BTRFS_ORDERED_PREALLOC || type == BTRFS_ORDERED_COMPRESSED || - type == BTRFS_ORDERED_NOCOW || type == BTRFS_ORDERED_REGULAR); + switch (type) { + case BTRFS_ORDERED_PREALLOC: + /* Uncompressed extents. */ + ASSERT(block_len == len); + + /* We're only referring part of a larger preallocated extent. */ + ASSERT(block_len <= ram_bytes); + break; + case BTRFS_ORDERED_REGULAR: + /* Uncompressed extents. */ + ASSERT(block_len == len); + + /* COW results a new extent matching our file extent size. */ + ASSERT(orig_block_len == len); + ASSERT(ram_bytes == len); + + /* Since it's a new extent, we should not have any offset. */ + ASSERT(orig_start == start); + break; + case BTRFS_ORDERED_COMPRESSED: + /* Must be compressed. */ + ASSERT(compress_type != BTRFS_COMPRESS_NONE); + + /* + * Encoded write can make us to refer to part of the + * uncompressed extent. + */ + ASSERT(len <= ram_bytes); + break; + } + em = alloc_extent_map(); if (!em) return ERR_PTR(-ENOMEM); -- cgit v1.2.3-59-g8ed1b From aa5ccf29173acfaa8aa2fdd1421aa6aca1a50cf2 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 14:49:20 -0400 Subject: btrfs: handle errors in btrfs_reloc_clone_csums properly In the cow path we will clone the reloc csums for relocated data extents, and if there's an error we already have an ordered extent and rely on the ordered extent finishing to clean everything up. There's a problem however, we don't mark the ordered extent with an error, we pretend like everything was just fine. If we were at the end of our range we won't actually bubble up this error anywhere, and we could end up inserting an extent that doesn't have csums where it should have them. Fix this by adding a helper to mark the ordered extent with an error, and then use this when we fail to lookup the csums in btrfs_reloc_clone_csums. Use this helper in the other place where we use the same pattern while we're here. This will prevent us from erroneously inserting the extent that doesn't have the required checksums. Reviewed-by: Johannes Thumshirn Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 5 ++--- fs/btrfs/ordered-data.c | 6 ++++++ fs/btrfs/ordered-data.h | 1 + fs/btrfs/relocation.c | 4 +++- 4 files changed, 12 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 3d587b517809..7a709a884864 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3183,9 +3183,8 @@ out: * set the mapping error, so we need to set it if we're the ones * marking this ordered extent as failed. */ - if (ret && !test_and_set_bit(BTRFS_ORDERED_IOERR, - &ordered_extent->flags)) - mapping_set_error(ordered_extent->inode->i_mapping, -EIO); + if (ret) + btrfs_mark_ordered_extent_error(ordered_extent); if (truncated) unwritten_start += logical_len; diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 3c2921e32843..c5bdd674f55c 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -294,6 +294,12 @@ void btrfs_add_ordered_sum(struct btrfs_ordered_extent *entry, spin_unlock_irq(&inode->ordered_tree_lock); } +void btrfs_mark_ordered_extent_error(struct btrfs_ordered_extent *ordered) +{ + if (!test_and_set_bit(BTRFS_ORDERED_IOERR, &ordered->flags)) + mapping_set_error(ordered->inode->i_mapping, -EIO); +} + static void finish_ordered_fn(struct btrfs_work *work) { struct btrfs_ordered_extent *ordered_extent; diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h index 34413fc5b4bd..b6f6c6b91732 100644 --- a/fs/btrfs/ordered-data.h +++ b/fs/btrfs/ordered-data.h @@ -203,6 +203,7 @@ bool btrfs_try_lock_ordered_range(struct btrfs_inode *inode, u64 start, u64 end, struct extent_state **cached_state); struct btrfs_ordered_extent *btrfs_split_ordered_extent( struct btrfs_ordered_extent *ordered, u64 len); +void btrfs_mark_ordered_extent_error(struct btrfs_ordered_extent *ordered); int __init ordered_data_init(void); void __cold ordered_data_exit(void); diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 5cf5718c3b1f..8b24bb5a0aa1 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4403,8 +4403,10 @@ int btrfs_reloc_clone_csums(struct btrfs_ordered_extent *ordered) ret = btrfs_lookup_csums_list(csum_root, disk_bytenr, disk_bytenr + ordered->num_bytes - 1, &list, false); - if (ret < 0) + if (ret < 0) { + btrfs_mark_ordered_extent_error(ordered); return ret; + } while (!list_empty(&list)) { struct btrfs_ordered_sum *sums = -- cgit v1.2.3-59-g8ed1b From 6eecfa22403e2c322a91f7654ade7f126adc55c2 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 20 Mar 2024 16:40:51 -0400 Subject: btrfs: push all inline logic into cow_file_range Currently we have a lot of duplicated checks of if (start == 0 && fs_info->sectorsize == PAGE_SIZE) cow_file_range_inline(); Instead of duplicating this check everywhere, consolidate all of the inline extent logic into a helper which documents all of the checks and then use that helper inside of cow_file_range_inline(). With this we can clean up all of the calls to either unconditionally call cow_file_range_inline(), or at least reduce the checks we're doing before we call cow_file_range_inline(); Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 143 +++++++++++++++++++++++++++++++------------------------ 1 file changed, 81 insertions(+), 62 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7a709a884864..fffa3d4015a3 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -614,14 +614,56 @@ fail: return ret; } +static bool can_cow_file_range_inline(struct btrfs_inode *inode, + u64 offset, u64 size, + size_t compressed_size) +{ + struct btrfs_fs_info *fs_info = inode->root->fs_info; + u64 data_len = (compressed_size ?: size); + + /* Inline extents must start at offset 0. */ + if (offset != 0) + return false; + + /* + * Due to the page size limit, for subpage we can only trigger the + * writeback for the dirty sectors of page, that means data writeback + * is doing more writeback than what we want. + * + * This is especially unexpected for some call sites like fallocate, + * where we only increase i_size after everything is done. + * This means we can trigger inline extent even if we didn't want to. + * So here we skip inline extent creation completely. + */ + if (fs_info->sectorsize != PAGE_SIZE) + return false; + + /* Inline extents are limited to sectorsize. */ + if (size > fs_info->sectorsize) + return false; + + /* We cannot exceed the maximum inline data size. */ + if (data_len > BTRFS_MAX_INLINE_DATA_SIZE(fs_info)) + return false; + + /* We cannot exceed the user specified max_inline size. */ + if (data_len > fs_info->max_inline) + return false; + + /* Inline extents must be the entirety of the file. */ + if (size < i_size_read(&inode->vfs_inode)) + return false; + + return true; +} /* * conditionally insert an inline extent into the file. This * does the checks required to make sure the data is small enough * to fit as an inline extent. */ -static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 size, - size_t compressed_size, +static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, + u64 size, size_t compressed_size, int compress_type, struct folio *compressed_folio, bool update_i_size) @@ -634,16 +676,7 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 size, int ret; struct btrfs_path *path; - /* - * We can create an inline extent if it ends at or beyond the current - * i_size, is no larger than a sector (decompressed), and the (possibly - * compressed) data fits in a leaf and the configured maximum inline - * size. - */ - if (size < i_size_read(&inode->vfs_inode) || - size > fs_info->sectorsize || - data_len > BTRFS_MAX_INLINE_DATA_SIZE(fs_info) || - data_len > fs_info->max_inline) + if (!can_cow_file_range_inline(inode, offset, size, compressed_size)) return 1; path = btrfs_alloc_path(); @@ -971,43 +1004,38 @@ again: * Check cow_file_range() for why we don't even try to create inline * extent for the subpage case. */ - if (start == 0 && fs_info->sectorsize == PAGE_SIZE) { - if (total_in < actual_end) { - ret = cow_file_range_inline(inode, actual_end, 0, - BTRFS_COMPRESS_NONE, NULL, - false); - } else { - ret = cow_file_range_inline(inode, actual_end, - total_compressed, - compress_type, folios[0], - false); - } - if (ret <= 0) { - unsigned long clear_flags = EXTENT_DELALLOC | - EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | - EXTENT_DO_ACCOUNTING; + if (total_in < actual_end) + ret = cow_file_range_inline(inode, start, actual_end, 0, + BTRFS_COMPRESS_NONE, NULL, false); + else + ret = cow_file_range_inline(inode, start, actual_end, + total_compressed, compress_type, + folios[0], false); + if (ret <= 0) { + unsigned long clear_flags = EXTENT_DELALLOC | + EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | + EXTENT_DO_ACCOUNTING; - if (ret < 0) - mapping_set_error(mapping, -EIO); + if (ret < 0) + mapping_set_error(mapping, -EIO); - /* - * inline extent creation worked or returned error, - * we don't need to create any more async work items. - * Unlock and free up our temp pages. - * - * We use DO_ACCOUNTING here because we need the - * delalloc_release_metadata to be done _after_ we drop - * our outstanding extent for clearing delalloc for this - * range. - */ - extent_clear_unlock_delalloc(inode, start, end, - NULL, - clear_flags, - PAGE_UNLOCK | - PAGE_START_WRITEBACK | - PAGE_END_WRITEBACK); - goto free_pages; - } + /* + * inline extent creation worked or returned error, + * we don't need to create any more async work items. + * Unlock and free up our temp pages. + * + * We use DO_ACCOUNTING here because we need the + * delalloc_release_metadata to be done _after_ we drop + * our outstanding extent for clearing delalloc for this + * range. + */ + extent_clear_unlock_delalloc(inode, start, end, + NULL, + clear_flags, + PAGE_UNLOCK | + PAGE_START_WRITEBACK | + PAGE_END_WRITEBACK); + goto free_pages; } /* @@ -1315,22 +1343,12 @@ static noinline int cow_file_range(struct btrfs_inode *inode, inode_should_defrag(inode, start, end, num_bytes, SZ_64K); - /* - * Due to the page size limit, for subpage we can only trigger the - * writeback for the dirty sectors of page, that means data writeback - * is doing more writeback than what we want. - * - * This is especially unexpected for some call sites like fallocate, - * where we only increase i_size after everything is done. - * This means we can trigger inline extent even if we didn't want to. - * So here we skip inline extent creation completely. - */ - if (start == 0 && fs_info->sectorsize == PAGE_SIZE && !no_inline) { + if (!no_inline) { u64 actual_end = min_t(u64, i_size_read(&inode->vfs_inode), end + 1); /* lets try to make an inline extent */ - ret = cow_file_range_inline(inode, actual_end, 0, + ret = cow_file_range_inline(inode, start, actual_end, 0, BTRFS_COMPRESS_NONE, NULL, false); if (ret == 0) { /* @@ -10266,10 +10284,11 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, goto out_qgroup_free_data; /* Try an inline extent first. */ - if (start == 0 && encoded->unencoded_len == encoded->len && + if (encoded->unencoded_len == encoded->len && encoded->unencoded_offset == 0) { - ret = cow_file_range_inline(inode, encoded->len, orig_count, - compression, folios[0], true); + ret = cow_file_range_inline(inode, start, encoded->len, + orig_count, compression, folios[0], + true); if (ret <= 0) { if (ret == 0) ret = orig_count; -- cgit v1.2.3-59-g8ed1b From 0332967b4d657754c914b2a1951ae2d4ed3d99bd Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 20 Mar 2024 17:06:10 -0400 Subject: btrfs: unlock all the pages with successful inline extent creation Since 4750af3bbe5d ("btrfs: prevent extent_clear_unlock_delalloc() to unlock page not locked by __process_pages_contig()") we have been unlocking the locked page manually instead of via extent_clear_unlock_delalloc() because of subpage blocksize support. However we actually disable inline extent creation for subpage blocksize support, so this behavior isn't necessary. Remove this code and comment, if at some point the subpage blocksize code grows support for inline extents this can be re-evaluated. Reviewed-by: Johannes Thumshirn Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index fffa3d4015a3..477836cae06f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1358,24 +1358,11 @@ static noinline int cow_file_range(struct btrfs_inode *inode, * range. */ extent_clear_unlock_delalloc(inode, start, end, - locked_page, + NULL, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); - /* - * locked_page is locked by the caller of - * writepage_delalloc(), not locked by - * __process_pages_contig(). - * - * We can't let __process_pages_contig() to unlock it, - * as it doesn't have any subpage::writers recorded. - * - * Here we manually unlock the page, since the caller - * can't determine if it's an inline extent or a - * compressed extent. - */ - unlock_page(locked_page); ret = 1; goto done; } else if (ret < 0) { -- cgit v1.2.3-59-g8ed1b From 0586d0a89e77d717da14df42648ace4a9fd67981 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 20 Mar 2024 17:24:13 -0400 Subject: btrfs: move extent bit and page cleanup into cow_file_range_inline We duplicate the extent cleanup for cow_file_range_inline() in the cow and compressed case. The encoded case doesn't need to do cleanup the same way, so rename cow_file_range_inline to __cow_file_range_inline and then make cow_file_range_inline handle the extent cleanup appropriately, and update the callers. Reviewed-by: Johannes Thumshirn Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 104 +++++++++++++++++++++++++++---------------------------- 1 file changed, 51 insertions(+), 53 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 477836cae06f..be079e4971e5 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -662,11 +662,11 @@ static bool can_cow_file_range_inline(struct btrfs_inode *inode, * does the checks required to make sure the data is small enough * to fit as an inline extent. */ -static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, - u64 size, size_t compressed_size, - int compress_type, - struct folio *compressed_folio, - bool update_i_size) +static noinline int __cow_file_range_inline(struct btrfs_inode *inode, u64 offset, + u64 size, size_t compressed_size, + int compress_type, + struct folio *compressed_folio, + bool update_i_size) { struct btrfs_drop_extents_args drop_args = { 0 }; struct btrfs_root *root = inode->root; @@ -737,6 +737,33 @@ out: return ret; } +static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, + u64 end, + size_t compressed_size, + int compress_type, + struct folio *compressed_folio, + bool update_i_size, bool locked) +{ + unsigned long clear_flags = EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | + EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING; + u64 size = min_t(u64, i_size_read(&inode->vfs_inode), end + 1); + int ret; + + ret = __cow_file_range_inline(inode, offset, size, compressed_size, + compress_type, compressed_folio, + update_i_size); + if (ret > 0) + return ret; + + if (locked) + clear_flags |= EXTENT_LOCKED; + + extent_clear_unlock_delalloc(inode, offset, end, NULL, clear_flags, + PAGE_UNLOCK | PAGE_START_WRITEBACK | + PAGE_END_WRITEBACK); + return ret; +} + struct async_extent { u64 start; u64 ram_size; @@ -1005,36 +1032,15 @@ again: * extent for the subpage case. */ if (total_in < actual_end) - ret = cow_file_range_inline(inode, start, actual_end, 0, - BTRFS_COMPRESS_NONE, NULL, false); + ret = cow_file_range_inline(inode, start, end, 0, + BTRFS_COMPRESS_NONE, NULL, false, + false); else - ret = cow_file_range_inline(inode, start, actual_end, - total_compressed, compress_type, - folios[0], false); + ret = cow_file_range_inline(inode, start, end, total_compressed, + compress_type, folios[0], false, false); if (ret <= 0) { - unsigned long clear_flags = EXTENT_DELALLOC | - EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | - EXTENT_DO_ACCOUNTING; - if (ret < 0) mapping_set_error(mapping, -EIO); - - /* - * inline extent creation worked or returned error, - * we don't need to create any more async work items. - * Unlock and free up our temp pages. - * - * We use DO_ACCOUNTING here because we need the - * delalloc_release_metadata to be done _after_ we drop - * our outstanding extent for clearing delalloc for this - * range. - */ - extent_clear_unlock_delalloc(inode, start, end, - NULL, - clear_flags, - PAGE_UNLOCK | - PAGE_START_WRITEBACK | - PAGE_END_WRITEBACK); goto free_pages; } @@ -1344,29 +1350,21 @@ static noinline int cow_file_range(struct btrfs_inode *inode, inode_should_defrag(inode, start, end, num_bytes, SZ_64K); if (!no_inline) { - u64 actual_end = min_t(u64, i_size_read(&inode->vfs_inode), - end + 1); - /* lets try to make an inline extent */ - ret = cow_file_range_inline(inode, start, actual_end, 0, - BTRFS_COMPRESS_NONE, NULL, false); - if (ret == 0) { + ret = cow_file_range_inline(inode, start, end, 0, + BTRFS_COMPRESS_NONE, NULL, false, + true); + if (ret <= 0) { /* - * We use DO_ACCOUNTING here because we need the - * delalloc_release_metadata to be run _after_ we drop - * our outstanding extent for clearing delalloc for this - * range. + * We succeeded, return 1 so the caller knows we're done + * with this page and already handled the IO. + * + * If there was an error then cow_file_range_inline() has + * already done the cleanup. */ - extent_clear_unlock_delalloc(inode, start, end, - NULL, - EXTENT_LOCKED | EXTENT_DELALLOC | - EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | - EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | - PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); - ret = 1; + if (ret == 0) + ret = 1; goto done; - } else if (ret < 0) { - goto out_unlock; } } @@ -10273,9 +10271,9 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, /* Try an inline extent first. */ if (encoded->unencoded_len == encoded->len && encoded->unencoded_offset == 0) { - ret = cow_file_range_inline(inode, start, encoded->len, - orig_count, compression, folios[0], - true); + ret = __cow_file_range_inline(inode, start, encoded->len, + orig_count, compression, folios[0], + true); if (ret <= 0) { if (ret == 0) ret = orig_count; -- cgit v1.2.3-59-g8ed1b From 7034674b8a070361b3149fdae377f5b80d5f860f Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 13:42:45 -0400 Subject: btrfs: lock extent when doing inline extent in compression We currently don't lock the extent when we're doing a cow_file_range_inline() for a compressed extent. This isn't a problem necessarily, but it's inconsistent with the rest of our usage of cow_file_range_inline(). This also leads to some extra weird logic around whether the extent is locked or not. Fix this to lock the extent before calling cow_file_range_inline() in compression to make it consistent with the rest of the inline users. In future patches this will be pushed down into the cow_file_range_inline() helper, so we're fine with the quick and dirty locking here. This patch exists to make the behavior change obvious. Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index be079e4971e5..d82c453ae9d7 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -742,10 +742,10 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, size_t compressed_size, int compress_type, struct folio *compressed_folio, - bool update_i_size, bool locked) + bool update_i_size) { unsigned long clear_flags = EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | - EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING; + EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING | EXTENT_LOCKED; u64 size = min_t(u64, i_size_read(&inode->vfs_inode), end + 1); int ret; @@ -755,9 +755,6 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, if (ret > 0) return ret; - if (locked) - clear_flags |= EXTENT_LOCKED; - extent_clear_unlock_delalloc(inode, offset, end, NULL, clear_flags, PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); @@ -1031,18 +1028,19 @@ again: * Check cow_file_range() for why we don't even try to create inline * extent for the subpage case. */ + lock_extent(&inode->io_tree, start, end, NULL); if (total_in < actual_end) ret = cow_file_range_inline(inode, start, end, 0, - BTRFS_COMPRESS_NONE, NULL, false, - false); + BTRFS_COMPRESS_NONE, NULL, false); else ret = cow_file_range_inline(inode, start, end, total_compressed, - compress_type, folios[0], false, false); + compress_type, folios[0], false); if (ret <= 0) { if (ret < 0) mapping_set_error(mapping, -EIO); goto free_pages; } + unlock_extent(&inode->io_tree, start, end, NULL); /* * We aren't doing an inline extent. Round the compressed size up to a @@ -1352,8 +1350,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode, if (!no_inline) { /* lets try to make an inline extent */ ret = cow_file_range_inline(inode, start, end, 0, - BTRFS_COMPRESS_NONE, NULL, false, - true); + BTRFS_COMPRESS_NONE, NULL, false); if (ret <= 0) { /* * We succeeded, return 1 so the caller knows we're done -- cgit v1.2.3-59-g8ed1b From c0707c9e1e36d56cef7b3c8de5c5fdcb14f34aa5 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 12 Feb 2024 16:10:44 -0500 Subject: btrfs: push the extent lock into btrfs_run_delalloc_range We want to limit the scope of the extent lock to be around operations that can change in flight. Currently we hold the extent lock through the entire writepage operation, which isn't really necessary. We want to protect to make sure nobody has updated DELALLOC. In find_lock_delalloc_range we must lock the range in order to validate the contents of our io_tree. However once we've done that we're safe to unlock the range and continue, as we have the page lock already held for the range. We are protected from all operations at this point. * mmap() - we're holding the page lock, thus are protected. * buffered writes - again, we're protected because we take the page lock for the first and last page in our range for buffered writes so we won't create new delalloc ranges in this area. * direct IO - we invalidate pagecache before attempting to write a new area, which requires the page lock, so again are protected once we're holding the page lock on this range. Additionally this behavior actually already exists for compressed, we unlock the range as soon as we start to process the async extents, and re-lock it during compression. So this is completely safe, and makes the locking more consistent. Make this simple by just pushing the extent lock into btrfs_run_delalloc_range. From there followup patches will push the lock further down into its users. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 5 ++--- fs/btrfs/inode.c | 5 +++++ 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index a9f9f5abdf53..d76ba4099b79 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -396,15 +396,14 @@ again: /* then test to make sure it is all still delalloc */ ret = test_range_bit(tree, delalloc_start, delalloc_end, EXTENT_DELALLOC, cached_state); + + unlock_extent(tree, delalloc_start, delalloc_end, &cached_state); if (!ret) { - unlock_extent(tree, delalloc_start, delalloc_end, - &cached_state); __unlock_for_delalloc(inode, locked_page, delalloc_start, delalloc_end); cond_resched(); goto again; } - free_extent_state(cached_state); *start = delalloc_start; *end = delalloc_end; out_failed: diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d82c453ae9d7..347de4e3a331 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2249,6 +2249,11 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page const bool zoned = btrfs_is_zoned(inode->root->fs_info); int ret; + /* + * We're unlocked by the different fill functions below. + */ + lock_extent(&inode->io_tree, start, end, NULL); + /* * The range must cover part of the @locked_page, or a return of 1 * can confuse the caller. -- cgit v1.2.3-59-g8ed1b From 7c9acd440f4d1124122639928ac4ff69082bbd3a Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 12 Feb 2024 16:21:19 -0500 Subject: btrfs: push extent lock into run_delalloc_nocow run_delalloc_nocow is a bit special as it walks through the file extents for the inode and determines what it can nocow and what it can't. This is the more complicated area for extent locking, so start with this function. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 347de4e3a331..d8037775ab75 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1977,6 +1977,8 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode, */ ASSERT(!btrfs_is_zoned(fs_info) || btrfs_is_data_reloc_root(root)); + lock_extent(&inode->io_tree, start, end, NULL); + path = btrfs_alloc_path(); if (!path) { ret = -ENOMEM; @@ -2249,11 +2251,6 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page const bool zoned = btrfs_is_zoned(inode->root->fs_info); int ret; - /* - * We're unlocked by the different fill functions below. - */ - lock_extent(&inode->io_tree, start, end, NULL); - /* * The range must cover part of the @locked_page, or a return of 1 * can confuse the caller. @@ -2266,6 +2263,11 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page goto out; } + /* + * We're unlocked by the different fill functions below. + */ + lock_extent(&inode->io_tree, start, end, NULL); + if (btrfs_inode_can_compress(inode) && inode_need_compress(inode, start, end) && run_delalloc_compressed(inode, locked_page, start, end, wbc)) -- cgit v1.2.3-59-g8ed1b From 0ed30c17f699d5df73c445999b0114c5859d1145 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 12 Feb 2024 16:27:15 -0500 Subject: btrfs: adjust while loop condition in run_delalloc_nocow We have the following pattern while (1) { if (cur_offset > end) break; } Which is just while (cur_offset <= end) { ... } so adjust the code to be more clear. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d8037775ab75..2fe82614521c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1988,7 +1988,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode, nocow_args.end = end; nocow_args.writeback_path = true; - while (1) { + while (cur_offset <= end) { struct btrfs_block_group *nocow_bg = NULL; struct btrfs_ordered_extent *ordered; struct btrfs_key found_key; @@ -2192,8 +2192,6 @@ must_cow: */ if (ret) goto error; - if (cur_offset > end) - break; } btrfs_release_path(path); -- cgit v1.2.3-59-g8ed1b From aa56b0aa91692c51142c3c7e4ab7665e84f792f5 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 12 Feb 2024 16:52:53 -0500 Subject: btrfs: push extent lock down in run_delalloc_nocow run_delalloc_nocow is a little special because we use the file extents to see if we can nocow a range. We don't actually need the protection of the extent lock to look at the file extents at this point however. We are currently holding the page lock for this range, so we are protected from anybody who would simultaneously be modifying the file extent items for this range. * mmap() - we're holding the page lock. * buffered writes - we're holding the page lock. * direct writes - we're holding the page lock and direct IO has to flush page cache before it's able to continue. * fallocate() - all callers flush the range and wait on ordered extents while holding the inode lock and the mmap lock, so we are again saved by the page lock. We want to use the extent lock to protect 1) The mapping tree for the given range. 2) The ordered extents for the given range. 3) The io_tree for the given range. Push the extent lock down to cover these operations. In the fallback_to_cow() case we simply lock before doing anything and rely on the cow_file_range() helper to handle it's range properly. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 2fe82614521c..7c4ebe061abb 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1747,6 +1747,8 @@ static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page, u64 count; int ret; + lock_extent(io_tree, start, end, NULL); + /* * If EXTENT_NORESERVE is set it means that when the buffered write was * made we had not enough available data space and therefore we did not @@ -1977,8 +1979,6 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode, */ ASSERT(!btrfs_is_zoned(fs_info) || btrfs_is_data_reloc_root(root)); - lock_extent(&inode->io_tree, start, end, NULL); - path = btrfs_alloc_path(); if (!path) { ret = -ENOMEM; @@ -1994,6 +1994,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode, struct btrfs_key found_key; struct btrfs_file_extent_item *fi; struct extent_buffer *leaf; + struct extent_state *cached_state = NULL; u64 extent_end; u64 ram_bytes; u64 nocow_end; @@ -2131,6 +2132,8 @@ must_cow: } nocow_end = cur_offset + nocow_args.num_bytes - 1; + lock_extent(&inode->io_tree, cur_offset, nocow_end, &cached_state); + is_prealloc = extent_type == BTRFS_FILE_EXTENT_PREALLOC; if (is_prealloc) { u64 orig_start = found_key.offset - nocow_args.extent_offset; @@ -2144,6 +2147,8 @@ must_cow: ram_bytes, BTRFS_COMPRESS_NONE, BTRFS_ORDERED_PREALLOC); if (IS_ERR(em)) { + unlock_extent(&inode->io_tree, cur_offset, + nocow_end, &cached_state); btrfs_dec_nocow_writers(nocow_bg); ret = PTR_ERR(em); goto error; @@ -2164,6 +2169,8 @@ must_cow: btrfs_drop_extent_map_range(inode, cur_offset, nocow_end, false); } + unlock_extent(&inode->io_tree, cur_offset, + nocow_end, &cached_state); ret = PTR_ERR(ordered); goto error; } @@ -2182,6 +2189,7 @@ must_cow: EXTENT_DELALLOC | EXTENT_CLEAR_DATA_RESV, PAGE_UNLOCK | PAGE_SET_ORDERED); + free_extent_state(cached_state); cur_offset = extent_end; @@ -2217,13 +2225,20 @@ error: */ if (cow_start != (u64)-1) cur_offset = cow_start; - if (cur_offset < end) + + /* + * We need to lock the extent here because we're clearing DELALLOC and + * we're not locked at this point. + */ + if (cur_offset < end) { + lock_extent(&inode->io_tree, cur_offset, end, NULL); extent_clear_unlock_delalloc(inode, cur_offset, end, locked_page, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); + } btrfs_free_path(path); return ret; } -- cgit v1.2.3-59-g8ed1b From 0e128d4e41735b37d9450b4e622b427189afb191 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 12 Feb 2024 16:59:43 -0500 Subject: btrfs: remove unlock_extent from run_delalloc_compressed Since we immediately unlock the extent range when we enter run_delalloc_compressed() simply move the lock_extent() down to cover cow_file_range() and then remove the unlock_extent() from run_delalloc_compressed. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7c4ebe061abb..5337b54148e3 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1642,7 +1642,6 @@ static bool run_delalloc_compressed(struct btrfs_inode *inode, if (!ctx) return false; - unlock_extent(&inode->io_tree, start, end, NULL); set_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, &inode->runtime_flags); async_chunk = ctx->chunks; @@ -2276,16 +2275,16 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page goto out; } - /* - * We're unlocked by the different fill functions below. - */ - lock_extent(&inode->io_tree, start, end, NULL); - if (btrfs_inode_can_compress(inode) && inode_need_compress(inode, start, end) && run_delalloc_compressed(inode, locked_page, start, end, wbc)) return 1; + /* + * We're unlocked by the different fill functions below. + */ + lock_extent(&inode->io_tree, start, end, NULL); + if (zoned) ret = run_delalloc_cow(inode, locked_page, start, end, wbc, true); -- cgit v1.2.3-59-g8ed1b From 00009d7bcb0ccc4479e1178790751e95cb36d7c4 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 15:31:17 -0400 Subject: btrfs: push extent lock into run_delalloc_cow This is used by zoned but also as the fallback for uncompressed extents when we fail to compress the ranges. Push the extent lock into run_dealloc_cow(), and adjust the compression case to take the extent lock after calling run_delalloc_cow(). Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 5337b54148e3..fea5a44e2719 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1167,13 +1167,13 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, if (!(start >= locked_page_end || end <= locked_page_start)) locked_page = async_chunk->locked_page; } - lock_extent(io_tree, start, end, NULL); if (async_extent->compress_type == BTRFS_COMPRESS_NONE) { submit_uncompressed_range(inode, async_extent, locked_page); goto done; } + lock_extent(io_tree, start, end, NULL); ret = btrfs_reserve_extent(root, async_extent->ram_size, async_extent->compressed_size, async_extent->compressed_size, @@ -1722,6 +1722,8 @@ static noinline int run_delalloc_cow(struct btrfs_inode *inode, u64 done_offset = end; int ret; + lock_extent(&inode->io_tree, start, end, NULL); + while (start <= end) { ret = cow_file_range(inode, locked_page, start, end, &done_offset, true, false); @@ -2280,17 +2282,14 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page run_delalloc_compressed(inode, locked_page, start, end, wbc)) return 1; - /* - * We're unlocked by the different fill functions below. - */ - lock_extent(&inode->io_tree, start, end, NULL); - - if (zoned) + if (zoned) { ret = run_delalloc_cow(inode, locked_page, start, end, wbc, true); - else + } else { + lock_extent(&inode->io_tree, start, end, NULL); ret = cow_file_range(inode, locked_page, start, end, NULL, false, false); + } out: if (ret < 0) -- cgit v1.2.3-59-g8ed1b From a0766d8f35d8c7b3df319ccfd6e85b4a07c1f0bd Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 12 Feb 2024 17:07:58 -0500 Subject: btrfs: push extent lock into cow_file_range Now that cow_file_range is the only function that is called with the range locked, push this call into cow_file_range so we can further narrow the scope. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index fea5a44e2719..7f431151af94 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1336,6 +1336,8 @@ static noinline int cow_file_range(struct btrfs_inode *inode, bool extent_reserved = false; int ret = 0; + lock_extent(&inode->io_tree, start, end, NULL); + if (btrfs_is_free_space_inode(inode)) { ret = -EINVAL; goto out_unlock; @@ -1722,8 +1724,6 @@ static noinline int run_delalloc_cow(struct btrfs_inode *inode, u64 done_offset = end; int ret; - lock_extent(&inode->io_tree, start, end, NULL); - while (start <= end) { ret = cow_file_range(inode, locked_page, start, end, &done_offset, true, false); @@ -1744,12 +1744,11 @@ static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page, const bool is_reloc_ino = btrfs_is_data_reloc_root(inode->root); const u64 range_bytes = end + 1 - start; struct extent_io_tree *io_tree = &inode->io_tree; + struct extent_state *cached_state = NULL; u64 range_start = start; u64 count; int ret; - lock_extent(io_tree, start, end, NULL); - /* * If EXTENT_NORESERVE is set it means that when the buffered write was * made we had not enough available data space and therefore we did not @@ -1782,6 +1781,7 @@ static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page, * group that contains that extent to RO mode and therefore force COW * when starting writeback. */ + lock_extent(io_tree, start, end, &cached_state); count = count_range_bits(io_tree, &range_start, end, range_bytes, EXTENT_NORESERVE, 0, NULL); if (count > 0 || is_space_ino || is_reloc_ino) { @@ -1800,6 +1800,7 @@ static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page, clear_extent_bit(io_tree, start, end, EXTENT_NORESERVE, NULL); } + unlock_extent(io_tree, start, end, &cached_state); /* * Don't try to create inline extents, as a mix of inline extent that @@ -2282,14 +2283,12 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page run_delalloc_compressed(inode, locked_page, start, end, wbc)) return 1; - if (zoned) { + if (zoned) ret = run_delalloc_cow(inode, locked_page, start, end, wbc, true); - } else { - lock_extent(&inode->io_tree, start, end, NULL); + else ret = cow_file_range(inode, locked_page, start, end, NULL, false, false); - } out: if (ret < 0) -- cgit v1.2.3-59-g8ed1b From 0ab540995ae049f1536260b0f6124483e685b593 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 13:53:57 -0400 Subject: btrfs: push lock_extent into cow_file_range_inline Now that we've pushed the lock_extent() into cow_file_range() we can push the extent locking into cow_file_range_inline() and move the lock_extent in cow_file_range() to after we call cow_file_range_inline(). Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7f431151af94..d6bad01c0cdf 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -744,17 +744,22 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, struct folio *compressed_folio, bool update_i_size) { + struct extent_state *cached = NULL; unsigned long clear_flags = EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING | EXTENT_LOCKED; u64 size = min_t(u64, i_size_read(&inode->vfs_inode), end + 1); int ret; + lock_extent(&inode->io_tree, offset, end, &cached); ret = __cow_file_range_inline(inode, offset, size, compressed_size, compress_type, compressed_folio, update_i_size); - if (ret > 0) + if (ret > 0) { + unlock_extent(&inode->io_tree, offset, end, &cached); return ret; + } + free_extent_state(cached); extent_clear_unlock_delalloc(inode, offset, end, NULL, clear_flags, PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); @@ -1028,7 +1033,6 @@ again: * Check cow_file_range() for why we don't even try to create inline * extent for the subpage case. */ - lock_extent(&inode->io_tree, start, end, NULL); if (total_in < actual_end) ret = cow_file_range_inline(inode, start, end, 0, BTRFS_COMPRESS_NONE, NULL, false); @@ -1040,7 +1044,6 @@ again: mapping_set_error(mapping, -EIO); goto free_pages; } - unlock_extent(&inode->io_tree, start, end, NULL); /* * We aren't doing an inline extent. Round the compressed size up to a @@ -1336,8 +1339,6 @@ static noinline int cow_file_range(struct btrfs_inode *inode, bool extent_reserved = false; int ret = 0; - lock_extent(&inode->io_tree, start, end, NULL); - if (btrfs_is_free_space_inode(inode)) { ret = -EINVAL; goto out_unlock; @@ -1367,6 +1368,8 @@ static noinline int cow_file_range(struct btrfs_inode *inode, } } + lock_extent(&inode->io_tree, start, end, NULL); + alloc_hint = get_extent_allocation_hint(inode, start, num_bytes); /* -- cgit v1.2.3-59-g8ed1b From cd241a8f554c258e92a694f6aa07e90b89ddebe6 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 13:57:30 -0400 Subject: btrfs: move can_cow_file_range_inline() outside of the extent lock These checks aren't reliant on the extent lock. Move this up into cow_file_range_inline(), and then update encoded writes to call this check before calling __cow_file_range_inline(). This will allow us to skip the extent lock if we're not able to inline the given extent. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d6bad01c0cdf..4b728f30da5d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -661,6 +661,9 @@ static bool can_cow_file_range_inline(struct btrfs_inode *inode, * conditionally insert an inline extent into the file. This * does the checks required to make sure the data is small enough * to fit as an inline extent. + * + * If being used directly, you must have already checked we're allowed to cow + * the range by getting true from can_cow_file_range_inline(). */ static noinline int __cow_file_range_inline(struct btrfs_inode *inode, u64 offset, u64 size, size_t compressed_size, @@ -676,9 +679,6 @@ static noinline int __cow_file_range_inline(struct btrfs_inode *inode, u64 offse int ret; struct btrfs_path *path; - if (!can_cow_file_range_inline(inode, offset, size, compressed_size)) - return 1; - path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -750,6 +750,9 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, u64 size = min_t(u64, i_size_read(&inode->vfs_inode), end + 1); int ret; + if (!can_cow_file_range_inline(inode, offset, size, compressed_size)) + return 1; + lock_extent(&inode->io_tree, offset, end, &cached); ret = __cow_file_range_inline(inode, offset, size, compressed_size, compress_type, compressed_folio, @@ -10287,7 +10290,8 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, /* Try an inline extent first. */ if (encoded->unencoded_len == encoded->len && - encoded->unencoded_offset == 0) { + encoded->unencoded_offset == 0 && + can_cow_file_range_inline(inode, start, encoded->len, orig_count)) { ret = __cow_file_range_inline(inode, start, encoded->len, orig_count, compression, folios[0], true); -- cgit v1.2.3-59-g8ed1b From d456c25dbb2385fa081e615793c797cbcd412724 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 15:06:09 -0400 Subject: btrfs: push lock_extent down in cow_file_range() Now that we've got the extent lock pushed into cow_file_range() we can push it further down into the allocation loop. This allows us to only hold the extent lock during the dropping of the extent map range and inserting the ordered extent. This makes the error case a little trickier as we'll now have to lock the range before clearing any of the other extent bits for the range, but this is the error path so is less performance critical. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 4b728f30da5d..549a5f327428 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1371,8 +1371,6 @@ static noinline int cow_file_range(struct btrfs_inode *inode, } } - lock_extent(&inode->io_tree, start, end, NULL); - alloc_hint = get_extent_allocation_hint(inode, start, num_bytes); /* @@ -1429,6 +1427,9 @@ static noinline int cow_file_range(struct btrfs_inode *inode, extent_reserved = true; ram_size = ins.offset; + + lock_extent(&inode->io_tree, start, start + ram_size - 1, NULL); + em = create_io_em(inode, start, ins.offset, /* len */ start, /* orig_start */ ins.objectid, /* block_start */ @@ -1438,6 +1439,8 @@ static noinline int cow_file_range(struct btrfs_inode *inode, BTRFS_COMPRESS_NONE, /* compress_type */ BTRFS_ORDERED_REGULAR /* type */); if (IS_ERR(em)) { + unlock_extent(&inode->io_tree, start, + start + ram_size - 1, NULL); ret = PTR_ERR(em); goto out_reserve; } @@ -1448,6 +1451,8 @@ static noinline int cow_file_range(struct btrfs_inode *inode, 0, 1 << BTRFS_ORDERED_REGULAR, BTRFS_COMPRESS_NONE); if (IS_ERR(ordered)) { + unlock_extent(&inode->io_tree, start, + start + ram_size - 1, NULL); ret = PTR_ERR(ordered); goto out_drop_extent_cache; } @@ -1549,6 +1554,13 @@ out_unlock: locked_page, 0, page_ops); } + /* + * At this point we're unlocked, we want to make sure we're only + * clearing these flags under the extent lock, so lock the rest of the + * range and clear everything up. + */ + lock_extent(&inode->io_tree, start, end, NULL); + /* * For the range (2). If we reserved an extent for our delalloc range * (or a subrange) and failed to create the respective ordered extent, -- cgit v1.2.3-59-g8ed1b From 8325f41a564acf39f1251c0423e97eb73c85a56f Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 16:28:45 -0400 Subject: btrfs: push extent lock down in submit_one_async_extent We don't need to include the time we spend in the allocator under our extent lock protection, move it after the allocator and make sure we lock the extent in the error case to ensure we're not clearing these bits without the extent lock held. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/inode.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 549a5f327428..944877a363fa 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1179,7 +1179,6 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, goto done; } - lock_extent(io_tree, start, end, NULL); ret = btrfs_reserve_extent(root, async_extent->ram_size, async_extent->compressed_size, async_extent->compressed_size, @@ -1195,6 +1194,8 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, goto done; } + lock_extent(io_tree, start, end, NULL); + /* Here we're doing allocation and writeback of the compressed pages */ em = create_io_em(inode, start, async_extent->ram_size, /* len */ -- cgit v1.2.3-59-g8ed1b From 6b0a63a4fa3142d1cb0069b9c7bf02270412d96f Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 3 Apr 2024 17:29:40 -0400 Subject: btrfs: add a cached state to extent_clear_unlock_delalloc Now that we have the lock_extent tightly coupled with extent_clear_unlock_delalloc we can add a cached state to extent_clear_unlock_delalloc and benefit from skipping the extra lookup when we're doing cow. Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 3 ++- fs/btrfs/extent_io.h | 2 ++ fs/btrfs/inode.c | 42 ++++++++++++++++++++++++------------------ 3 files changed, 28 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index d76ba4099b79..47a5bb95a994 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -412,9 +412,10 @@ out_failed: void extent_clear_unlock_delalloc(struct btrfs_inode *inode, u64 start, u64 end, struct page *locked_page, + struct extent_state **cached, u32 clear_bits, unsigned long page_ops) { - clear_extent_bit(&inode->io_tree, start, end, clear_bits, NULL); + clear_extent_bit(&inode->io_tree, start, end, clear_bits, cached); __process_pages_contig(inode->vfs_inode.i_mapping, locked_page, start, end, page_ops); diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index f38397765e90..dca6b12769ec 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -27,6 +27,7 @@ struct address_space; struct writeback_control; struct extent_io_tree; struct extent_map_tree; +struct extent_state; struct btrfs_block_group; struct btrfs_fs_info; struct btrfs_inode; @@ -352,6 +353,7 @@ void clear_extent_buffer_uptodate(struct extent_buffer *eb); void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end); void extent_clear_unlock_delalloc(struct btrfs_inode *inode, u64 start, u64 end, struct page *locked_page, + struct extent_state **cached, u32 bits_to_clear, unsigned long page_ops); int extent_invalidate_folio(struct extent_io_tree *tree, struct folio *folio, size_t offset); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 944877a363fa..d0274324c75a 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -762,8 +762,8 @@ static noinline int cow_file_range_inline(struct btrfs_inode *inode, u64 offset, return ret; } - free_extent_state(cached); - extent_clear_unlock_delalloc(inode, offset, end, NULL, clear_flags, + extent_clear_unlock_delalloc(inode, offset, end, NULL, &cached, + clear_flags, PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); return ret; @@ -1154,6 +1154,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, struct btrfs_ordered_extent *ordered; struct btrfs_key ins; struct page *locked_page = NULL; + struct extent_state *cached = NULL; struct extent_map *em; int ret = 0; u64 start = async_extent->start; @@ -1194,7 +1195,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, goto done; } - lock_extent(io_tree, start, end, NULL); + lock_extent(io_tree, start, end, &cached); /* Here we're doing allocation and writeback of the compressed pages */ em = create_io_em(inode, start, @@ -1229,7 +1230,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, /* Clear dirty, set writeback and unlock the pages. */ extent_clear_unlock_delalloc(inode, start, end, - NULL, EXTENT_LOCKED | EXTENT_DELALLOC, + NULL, &cached, EXTENT_LOCKED | EXTENT_DELALLOC, PAGE_UNLOCK | PAGE_START_WRITEBACK); btrfs_submit_compressed_write(ordered, async_extent->folios, /* compressed_folios */ @@ -1247,7 +1248,8 @@ out_free_reserve: btrfs_free_reserved_extent(fs_info, ins.objectid, ins.offset, 1); mapping_set_error(inode->vfs_inode.i_mapping, -EIO); extent_clear_unlock_delalloc(inode, start, end, - NULL, EXTENT_LOCKED | EXTENT_DELALLOC | + NULL, &cached, + EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | PAGE_START_WRITEBACK | @@ -1329,6 +1331,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode, { struct btrfs_root *root = inode->root; struct btrfs_fs_info *fs_info = root->fs_info; + struct extent_state *cached = NULL; u64 alloc_hint = 0; u64 orig_start = start; u64 num_bytes; @@ -1429,7 +1432,8 @@ static noinline int cow_file_range(struct btrfs_inode *inode, ram_size = ins.offset; - lock_extent(&inode->io_tree, start, start + ram_size - 1, NULL); + lock_extent(&inode->io_tree, start, start + ram_size - 1, + &cached); em = create_io_em(inode, start, ins.offset, /* len */ start, /* orig_start */ @@ -1441,7 +1445,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode, BTRFS_ORDERED_REGULAR /* type */); if (IS_ERR(em)) { unlock_extent(&inode->io_tree, start, - start + ram_size - 1, NULL); + start + ram_size - 1, &cached); ret = PTR_ERR(em); goto out_reserve; } @@ -1453,7 +1457,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode, BTRFS_COMPRESS_NONE); if (IS_ERR(ordered)) { unlock_extent(&inode->io_tree, start, - start + ram_size - 1, NULL); + start + ram_size - 1, &cached); ret = PTR_ERR(ordered); goto out_drop_extent_cache; } @@ -1493,7 +1497,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode, page_ops |= PAGE_SET_ORDERED; extent_clear_unlock_delalloc(inode, start, start + ram_size - 1, - locked_page, + locked_page, &cached, EXTENT_LOCKED | EXTENT_DELALLOC, page_ops); if (num_bytes < cur_alloc_size) @@ -1552,7 +1556,7 @@ out_unlock: if (!locked_page) mapping_set_error(inode->vfs_inode.i_mapping, ret); extent_clear_unlock_delalloc(inode, orig_start, start - 1, - locked_page, 0, page_ops); + locked_page, NULL, 0, page_ops); } /* @@ -1575,7 +1579,7 @@ out_unlock: if (extent_reserved) { extent_clear_unlock_delalloc(inode, start, start + cur_alloc_size - 1, - locked_page, + locked_page, &cached, clear_bits, page_ops); start += cur_alloc_size; @@ -1590,7 +1594,7 @@ out_unlock: if (start < end) { clear_bits |= EXTENT_CLEAR_DATA_RESV; extent_clear_unlock_delalloc(inode, start, end, locked_page, - clear_bits, page_ops); + &cached, clear_bits, page_ops); } return ret; } @@ -2206,11 +2210,10 @@ must_cow: btrfs_put_ordered_extent(ordered); extent_clear_unlock_delalloc(inode, cur_offset, nocow_end, - locked_page, EXTENT_LOCKED | - EXTENT_DELALLOC | + locked_page, &cached_state, + EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_CLEAR_DATA_RESV, PAGE_UNLOCK | PAGE_SET_ORDERED); - free_extent_state(cached_state); cur_offset = extent_end; @@ -2252,10 +2255,13 @@ error: * we're not locked at this point. */ if (cur_offset < end) { - lock_extent(&inode->io_tree, cur_offset, end, NULL); + struct extent_state *cached = NULL; + + lock_extent(&inode->io_tree, cur_offset, end, &cached); extent_clear_unlock_delalloc(inode, cur_offset, end, - locked_page, EXTENT_LOCKED | - EXTENT_DELALLOC | EXTENT_DEFRAG | + locked_page, &cached, + EXTENT_LOCKED | EXTENT_DELALLOC | + EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); -- cgit v1.2.3-59-g8ed1b From 58a774ca168dfb2a3513be90d56a6a04eeae53ad Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Thu, 2 May 2024 23:26:28 +0200 Subject: btrfs: remove duplicate included header from fs.h Remove duplicate included header file linux/blkdev.h . Signed-off-by: Thorsten Blum Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/fs.h | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h index 9711c9c0e78f..89f0650631cd 100644 --- a/fs/btrfs/fs.h +++ b/fs/btrfs/fs.h @@ -9,7 +9,6 @@ #include #include #include -#include #include #include #include -- cgit v1.2.3-59-g8ed1b From c94b7349b8592d2787284d470c10a17583c9d205 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:49:56 +0100 Subject: btrfs: convert super block writes to folio in wait_dev_supers() This is a direct conversion from pages to folios, assuming single page folio. Also removes a few calls to compound_head() and calls to obsolete APIs. Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index e0bdf3ed2449..5e2293c5b747 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3833,8 +3833,7 @@ static int write_dev_supers(struct btrfs_device *device, * Wait for write completion of superblocks done by write_dev_supers, * @max_mirrors same for write and wait phases. * - * Return number of errors when page is not found or not marked up to - * date. + * Return number of errors when folio is not found or not marked up to date. */ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) { @@ -3848,7 +3847,7 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) max_mirrors = BTRFS_SUPER_MIRROR_MAX; for (i = 0; i < max_mirrors; i++) { - struct page *page; + struct folio *folio; ret = btrfs_sb_log_location(device, i, READ, &bytenr); if (ret == -ENOENT) { @@ -3863,27 +3862,29 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) device->commit_total_bytes) break; - page = find_get_page(device->bdev->bd_inode->i_mapping, - bytenr >> PAGE_SHIFT); - if (!page) { + folio = filemap_get_folio(device->bdev->bd_inode->i_mapping, + bytenr >> PAGE_SHIFT); + if (IS_ERR(folio)) { errors++; if (i == 0) primary_failed = true; continue; } - /* Page is submitted locked and unlocked once the IO completes */ - wait_on_page_locked(page); - if (PageError(page)) { + ASSERT(folio_order(folio) == 0); + + /* Folio will be unlocked once the write completes. */ + folio_wait_locked(folio); + if (folio_test_error(folio)) { errors++; if (i == 0) primary_failed = true; } /* Drop our reference */ - put_page(page); + folio_put(folio); /* Drop the reference from the writing run */ - put_page(page); + folio_put(folio); } /* log error, force error return */ -- cgit v1.2.3-59-g8ed1b From f93ee0df51397a9432250798a93d2271ab86bb69 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:49:57 +0100 Subject: btrfs: convert super block writes to folio in write_dev_supers() This is a direct conversion from pages to folios, assuming single page folio. Also removes some calls to obsolete APIs and some hidden calls to compound_head(). Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 5e2293c5b747..9e8d0b86d3e0 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3734,13 +3734,13 @@ struct btrfs_super_block *btrfs_read_dev_super(struct block_device *bdev) /* * Write superblock @sb to the @device. Do not wait for completion, all the - * pages we use for writing are locked. + * folios we use for writing are locked. * * Write @max_mirrors copies of the superblock, where 0 means default that fit * the expected device size at commit time. Note that max_mirrors must be * same for write and wait phases. * - * Return number of errors when page is not found or submission fails. + * Return number of errors when folio is not found or submission fails. */ static int write_dev_supers(struct btrfs_device *device, struct btrfs_super_block *sb, int max_mirrors) @@ -3759,9 +3759,10 @@ static int write_dev_supers(struct btrfs_device *device, shash->tfm = fs_info->csum_shash; for (i = 0; i < max_mirrors; i++) { - struct page *page; + struct folio *folio; struct bio *bio; struct btrfs_super_block *disk_super; + size_t offset; bytenr_orig = btrfs_sb_offset(i); ret = btrfs_sb_log_location(device, i, WRITE, &bytenr); @@ -3784,20 +3785,23 @@ static int write_dev_supers(struct btrfs_device *device, BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE, sb->csum); - page = find_or_create_page(mapping, bytenr >> PAGE_SHIFT, - GFP_NOFS); - if (!page) { + folio = __filemap_get_folio(mapping, bytenr >> PAGE_SHIFT, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, + GFP_NOFS); + if (IS_ERR(folio)) { btrfs_err(device->fs_info, "couldn't get super block page for bytenr %llu", bytenr); errors++; continue; } + ASSERT(folio_order(folio) == 0); /* Bump the refcount for wait_dev_supers() */ - get_page(page); + folio_get(folio); - disk_super = page_address(page); + offset = offset_in_folio(folio, bytenr); + disk_super = folio_address(folio) + offset; memcpy(disk_super, sb, BTRFS_SUPER_INFO_SIZE); /* @@ -3811,8 +3815,7 @@ static int write_dev_supers(struct btrfs_device *device, bio->bi_iter.bi_sector = bytenr >> SECTOR_SHIFT; bio->bi_private = device; bio->bi_end_io = btrfs_end_super_write; - __bio_add_page(bio, page, BTRFS_SUPER_INFO_SIZE, - offset_in_page(bytenr)); + bio_add_folio_nofail(bio, folio, BTRFS_SUPER_INFO_SIZE, offset); /* * We FUA only the first super block. The others we allow to -- cgit v1.2.3-59-g8ed1b From 617fb10ea838f48c88c7f010eb68d150350914c8 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:49:58 +0100 Subject: btrfs: use the folio iterator in btrfs_end_super_write() Iterate over folios instead of bvecs. Switch the order of unlock and put to be the usual order; we know this folio can't be put until it's been waited for, but that's fragile. Remove the calls to ClearPageUptodate / SetPageUptodate -- if PAGE_SIZE is larger than BTRFS_SUPER_INFO_SIZE, we'd be marking the entire folio uptodate without having actually initialised all the bytes in the page. Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 9e8d0b86d3e0..90c54466ecc3 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3626,28 +3626,21 @@ ALLOW_ERROR_INJECTION(open_ctree, ERRNO); static void btrfs_end_super_write(struct bio *bio) { struct btrfs_device *device = bio->bi_private; - struct bio_vec *bvec; - struct bvec_iter_all iter_all; - struct page *page; - - bio_for_each_segment_all(bvec, bio, iter_all) { - page = bvec->bv_page; + struct folio_iter fi; + bio_for_each_folio_all(fi, bio) { if (bio->bi_status) { btrfs_warn_rl_in_rcu(device->fs_info, - "lost page write due to IO error on %s (%d)", + "lost super block write due to IO error on %s (%d)", btrfs_dev_name(device), blk_status_to_errno(bio->bi_status)); - ClearPageUptodate(page); - SetPageError(page); + folio_set_error(fi.folio); btrfs_dev_stat_inc_and_print(device, BTRFS_DEV_STAT_WRITE_ERRS); - } else { - SetPageUptodate(page); } - put_page(page); - unlock_page(page); + folio_unlock(fi.folio); + folio_put(fi.folio); } bio_put(bio); -- cgit v1.2.3-59-g8ed1b From bc00965dbff7a8612c8ec0005b3bc943d7196629 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:49:59 +0100 Subject: btrfs: count super block write errors in device instead of tracking folio error state Currently the error status of super block write is tracked in page/folio status bit Error. For that we need to keep the reference for the whole duration of write and wait. Count the number of superblock writeback errors in the btrfs_device. That means we don't need the folio to stay around until it's waited for, and can avoid the extra call to folio_get/put. Also remove a mention of PageError in a comment as it's the last mention of the page Error state. Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 46 +++++++++++++++++++--------------------------- fs/btrfs/extent_io.c | 2 +- fs/btrfs/volumes.h | 9 +++++++++ 3 files changed, 29 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 90c54466ecc3..a91a8056758a 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3634,11 +3634,15 @@ static void btrfs_end_super_write(struct bio *bio) "lost super block write due to IO error on %s (%d)", btrfs_dev_name(device), blk_status_to_errno(bio->bi_status)); - folio_set_error(fi.folio); btrfs_dev_stat_inc_and_print(device, BTRFS_DEV_STAT_WRITE_ERRS); + /* Ensure failure if the primary sb fails. */ + if (bio->bi_opf & REQ_FUA) + atomic_add(BTRFS_SUPER_PRIMARY_WRITE_ERROR, + &device->sb_write_errors); + else + atomic_inc(&device->sb_write_errors); } - folio_unlock(fi.folio); folio_put(fi.folio); } @@ -3742,10 +3746,11 @@ static int write_dev_supers(struct btrfs_device *device, struct address_space *mapping = device->bdev->bd_inode->i_mapping; SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); int i; - int errors = 0; int ret; u64 bytenr, bytenr_orig; + atomic_set(&device->sb_write_errors, 0); + if (max_mirrors == 0) max_mirrors = BTRFS_SUPER_MIRROR_MAX; @@ -3765,7 +3770,7 @@ static int write_dev_supers(struct btrfs_device *device, btrfs_err(device->fs_info, "couldn't get super block location for mirror %d", i); - errors++; + atomic_inc(&device->sb_write_errors); continue; } if (bytenr + BTRFS_SUPER_INFO_SIZE >= @@ -3785,14 +3790,11 @@ static int write_dev_supers(struct btrfs_device *device, btrfs_err(device->fs_info, "couldn't get super block page for bytenr %llu", bytenr); - errors++; + atomic_inc(&device->sb_write_errors); continue; } ASSERT(folio_order(folio) == 0); - /* Bump the refcount for wait_dev_supers() */ - folio_get(folio); - offset = offset_in_folio(folio, bytenr); disk_super = folio_address(folio) + offset; memcpy(disk_super, sb, BTRFS_SUPER_INFO_SIZE); @@ -3820,16 +3822,17 @@ static int write_dev_supers(struct btrfs_device *device, submit_bio(bio); if (btrfs_advance_sb_log(device, i)) - errors++; + atomic_inc(&device->sb_write_errors); } - return errors < i ? 0 : -1; + return atomic_read(&device->sb_write_errors) < i ? 0 : -1; } /* * Wait for write completion of superblocks done by write_dev_supers, * @max_mirrors same for write and wait phases. * - * Return number of errors when folio is not found or not marked up to date. + * Return -1 if primary super block write failed or when there were no super block + * copies written. Otherwise 0. */ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) { @@ -3860,30 +3863,19 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) folio = filemap_get_folio(device->bdev->bd_inode->i_mapping, bytenr >> PAGE_SHIFT); - if (IS_ERR(folio)) { - errors++; - if (i == 0) - primary_failed = true; + /* If the folio has been removed, then we know it completed. */ + if (IS_ERR(folio)) continue; - } ASSERT(folio_order(folio) == 0); /* Folio will be unlocked once the write completes. */ folio_wait_locked(folio); - if (folio_test_error(folio)) { - errors++; - if (i == 0) - primary_failed = true; - } - - /* Drop our reference */ - folio_put(folio); - - /* Drop the reference from the writing run */ folio_put(folio); } - /* log error, force error return */ + errors += atomic_read(&device->sb_write_errors); + if (errors >= BTRFS_SUPER_PRIMARY_WRITE_ERROR) + primary_failed = true; if (primary_failed) { btrfs_err(device->fs_info, "error writing primary super block to device %llu", device->devid); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 47a5bb95a994..597387e9f040 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -1602,7 +1602,7 @@ static void set_btree_ioerr(struct extent_buffer *eb) * can be no longer dirty nor marked anymore for writeback (if a * subsequent modification to the extent buffer didn't happen before the * transaction commit), which makes filemap_fdata[write|wait]_range not - * able to find the pages tagged with SetPageError at transaction + * able to find the pages which contain errors at transaction * commit time. So if this happens we must abort the transaction, * otherwise we commit a super block with btree roots that point to * btree nodes/leafs whose content on disk is invalid - either garbage diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index cf555f5b47ce..66e6fc481ecd 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -92,6 +92,9 @@ enum btrfs_raid_types { #define BTRFS_DEV_STATE_FLUSH_SENT (4) #define BTRFS_DEV_STATE_NO_READA (5) +/* Special value encoding failure to write primary super block. */ +#define BTRFS_SUPER_PRIMARY_WRITE_ERROR (INT_MAX / 2) + struct btrfs_fs_devices; struct btrfs_device { @@ -142,6 +145,12 @@ struct btrfs_device { /* type and info about this device */ u64 type; + /* + * Counter of super block write errors, values larger than + * BTRFS_SUPER_PRIMARY_WRITE_ERROR encode primary super block write failure. + */ + atomic_t sb_write_errors; + /* minimal io size for this device */ u32 sector_size; -- cgit v1.2.3-59-g8ed1b From 0e39c9e524479b85c1b83134df0cfc6e3cb5353a Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sat, 4 May 2024 14:38:41 +0300 Subject: btrfs: qgroup: fix initialization of auto inherit array The "i++" was accidentally left out so it just sets qgids[0] over and over. This can lead to unexpected problems, as the groups[1:] would be all 0, leading to later find_qgroup_rb() unable to find a qgroup and cause snapshot creation failure. Fixes: 5343cd9364ea ("btrfs: qgroup: simple quota auto hierarchy for nested subvolumes") CC: stable@vger.kernel.org # 6.7+ Reviewed-by: Qu Wenruo Signed-off-by: Dan Carpenter Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 1768fc6f232f..eb28141d5c37 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3123,7 +3123,7 @@ static int qgroup_auto_inherit(struct btrfs_fs_info *fs_info, qgids = res->qgroups; list_for_each_entry(qg_list, &inode_qg->groups, next_group) - qgids[i] = qg_list->group->qgroupid; + qgids[i++] = qg_list->group->qgroupid; *inherit = res; return 0; -- cgit v1.2.3-59-g8ed1b From 99b150d84e4939735cfce245e32e3d29312c68ec Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 18:28:54 +0100 Subject: ext4: convert bd_bitmap_page to bd_bitmap_folio There is no need to make this a multi-page folio, so leave all the infrastructure around it in pages. But since we're locking it, playing with its refcount and checking whether it's uptodate, it needs to move to the folio API. Signed-off-by: Matthew Wilcox (Oracle) Link: https://lore.kernel.org/r/20240416172900.244637-2-willy@infradead.org Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 98 +++++++++++++++++++++++++++++-------------------------- fs/ext4/mballoc.h | 2 +- 2 files changed, 52 insertions(+), 48 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index e89c1bfb7ce3..174c029a79f6 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1452,9 +1452,10 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, int block, pnum, poff; int blocks_per_page; struct page *page; + struct folio *folio; e4b->bd_buddy_page = NULL; - e4b->bd_bitmap_page = NULL; + e4b->bd_bitmap_folio = NULL; blocks_per_page = PAGE_SIZE / sb->s_blocksize; /* @@ -1465,12 +1466,13 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, block = group * 2; pnum = block / blocks_per_page; poff = block % blocks_per_page; - page = find_or_create_page(inode->i_mapping, pnum, gfp); - if (!page) - return -ENOMEM; - BUG_ON(page->mapping != inode->i_mapping); - e4b->bd_bitmap_page = page; - e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize); + folio = __filemap_get_folio(inode->i_mapping, pnum, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, gfp); + if (IS_ERR(folio)) + return PTR_ERR(folio); + BUG_ON(folio->mapping != inode->i_mapping); + e4b->bd_bitmap_folio = folio; + e4b->bd_bitmap = folio_address(folio) + (poff * sb->s_blocksize); if (blocks_per_page >= 2) { /* buddy and bitmap are on the same page */ @@ -1488,9 +1490,9 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, static void ext4_mb_put_buddy_page_lock(struct ext4_buddy *e4b) { - if (e4b->bd_bitmap_page) { - unlock_page(e4b->bd_bitmap_page); - put_page(e4b->bd_bitmap_page); + if (e4b->bd_bitmap_folio) { + folio_unlock(e4b->bd_bitmap_folio); + folio_put(e4b->bd_bitmap_folio); } if (e4b->bd_buddy_page) { unlock_page(e4b->bd_buddy_page); @@ -1510,6 +1512,7 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp) struct ext4_group_info *this_grp; struct ext4_buddy e4b; struct page *page; + struct folio *folio; int ret = 0; might_sleep(); @@ -1536,11 +1539,11 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp) goto err; } - page = e4b.bd_bitmap_page; - ret = ext4_mb_init_cache(page, NULL, gfp); + folio = e4b.bd_bitmap_folio; + ret = ext4_mb_init_cache(&folio->page, NULL, gfp); if (ret) goto err; - if (!PageUptodate(page)) { + if (!folio_test_uptodate(folio)) { ret = -EIO; goto err; } @@ -1582,6 +1585,7 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, int pnum; int poff; struct page *page; + struct folio *folio; int ret; struct ext4_group_info *grp; struct ext4_sb_info *sbi = EXT4_SB(sb); @@ -1600,7 +1604,7 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, e4b->bd_sb = sb; e4b->bd_group = group; e4b->bd_buddy_page = NULL; - e4b->bd_bitmap_page = NULL; + e4b->bd_bitmap_folio = NULL; if (unlikely(EXT4_MB_GRP_NEED_INIT(grp))) { /* @@ -1621,53 +1625,53 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, pnum = block / blocks_per_page; poff = block % blocks_per_page; - /* we could use find_or_create_page(), but it locks page - * what we'd like to avoid in fast path ... */ - page = find_get_page_flags(inode->i_mapping, pnum, FGP_ACCESSED); - if (page == NULL || !PageUptodate(page)) { - if (page) + /* Avoid locking the folio in the fast path ... */ + folio = __filemap_get_folio(inode->i_mapping, pnum, FGP_ACCESSED, 0); + if (IS_ERR(folio) || !folio_test_uptodate(folio)) { + if (!IS_ERR(folio)) /* - * drop the page reference and try - * to get the page with lock. If we + * drop the folio reference and try + * to get the folio with lock. If we * are not uptodate that implies - * somebody just created the page but - * is yet to initialize the same. So + * somebody just created the folio but + * is yet to initialize it. So * wait for it to initialize. */ - put_page(page); - page = find_or_create_page(inode->i_mapping, pnum, gfp); - if (page) { - if (WARN_RATELIMIT(page->mapping != inode->i_mapping, - "ext4: bitmap's paging->mapping != inode->i_mapping\n")) { + folio_put(folio); + folio = __filemap_get_folio(inode->i_mapping, pnum, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, gfp); + if (!IS_ERR(folio)) { + if (WARN_RATELIMIT(folio->mapping != inode->i_mapping, + "ext4: bitmap's mapping != inode->i_mapping\n")) { /* should never happen */ - unlock_page(page); + folio_unlock(folio); ret = -EINVAL; goto err; } - if (!PageUptodate(page)) { - ret = ext4_mb_init_cache(page, NULL, gfp); + if (!folio_test_uptodate(folio)) { + ret = ext4_mb_init_cache(&folio->page, NULL, gfp); if (ret) { - unlock_page(page); + folio_unlock(folio); goto err; } - mb_cmp_bitmaps(e4b, page_address(page) + + mb_cmp_bitmaps(e4b, folio_address(folio) + (poff * sb->s_blocksize)); } - unlock_page(page); + folio_unlock(folio); } } - if (page == NULL) { - ret = -ENOMEM; + if (IS_ERR(folio)) { + ret = PTR_ERR(folio); goto err; } - if (!PageUptodate(page)) { + if (!folio_test_uptodate(folio)) { ret = -EIO; goto err; } /* Pages marked accessed already */ - e4b->bd_bitmap_page = page; - e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize); + e4b->bd_bitmap_folio = folio; + e4b->bd_bitmap = folio_address(folio) + (poff * sb->s_blocksize); block++; pnum = block / blocks_per_page; @@ -1715,8 +1719,8 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, err: if (page) put_page(page); - if (e4b->bd_bitmap_page) - put_page(e4b->bd_bitmap_page); + if (e4b->bd_bitmap_folio) + folio_put(e4b->bd_bitmap_folio); e4b->bd_buddy = NULL; e4b->bd_bitmap = NULL; @@ -1731,8 +1735,8 @@ static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, static void ext4_mb_unload_buddy(struct ext4_buddy *e4b) { - if (e4b->bd_bitmap_page) - put_page(e4b->bd_bitmap_page); + if (e4b->bd_bitmap_folio) + folio_put(e4b->bd_bitmap_folio); if (e4b->bd_buddy_page) put_page(e4b->bd_buddy_page); } @@ -2157,7 +2161,7 @@ static void ext4_mb_use_best_found(struct ext4_allocation_context *ac, * double allocate blocks. The reference is dropped * in ext4_mb_release_context */ - ac->ac_bitmap_page = e4b->bd_bitmap_page; + ac->ac_bitmap_page = &e4b->bd_bitmap_folio->page; get_page(ac->ac_bitmap_page); ac->ac_buddy_page = e4b->bd_buddy_page; get_page(ac->ac_buddy_page); @@ -3894,7 +3898,7 @@ static void ext4_free_data_in_buddy(struct super_block *sb, * balance refcounts from ext4_mb_free_metadata() */ put_page(e4b.bd_buddy_page); - put_page(e4b.bd_bitmap_page); + folio_put(e4b.bd_bitmap_folio); } ext4_unlock_group(sb, entry->efd_group); ext4_mb_unload_buddy(&e4b); @@ -6316,7 +6320,7 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, struct rb_node *parent = NULL, *new_node; BUG_ON(!ext4_handle_valid(handle)); - BUG_ON(e4b->bd_bitmap_page == NULL); + BUG_ON(e4b->bd_bitmap_folio == NULL); BUG_ON(e4b->bd_buddy_page == NULL); new_node = &new_entry->efd_node; @@ -6329,7 +6333,7 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, * on-disk bitmap and lose not-yet-available * blocks */ get_page(e4b->bd_buddy_page); - get_page(e4b->bd_bitmap_page); + folio_get(e4b->bd_bitmap_folio); } while (*n) { parent = *n; diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 7ee13d8eecd7..f8b7e9d2e463 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -217,7 +217,7 @@ struct ext4_allocation_context { struct ext4_buddy { struct page *bd_buddy_page; void *bd_buddy; - struct page *bd_bitmap_page; + struct folio *bd_bitmap_folio; void *bd_bitmap; struct ext4_group_info *bd_info; struct super_block *bd_sb; -- cgit v1.2.3-59-g8ed1b From 5eea586b47f05b5f5518cf8f9dd9283a01a8066d Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 18:28:55 +0100 Subject: ext4: convert bd_buddy_page to bd_buddy_folio There is no need to make this a multi-page folio, so leave all the infrastructure around it in pages. But since we're locking it, playing with its refcount and checking whether it's uptodate, it needs to move to the folio API. Signed-off-by: Matthew Wilcox (Oracle) Link: https://lore.kernel.org/r/20240416172900.244637-3-willy@infradead.org Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 91 +++++++++++++++++++++++++++---------------------------- fs/ext4/mballoc.h | 2 +- 2 files changed, 46 insertions(+), 47 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 174c029a79f6..1a76e5741d36 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1443,7 +1443,7 @@ out: * Lock the buddy and bitmap pages. This make sure other parallel init_group * on the same buddy page doesn't happen whild holding the buddy page lock. * Return locked buddy and bitmap pages on e4b struct. If buddy and bitmap - * are on the same page e4b->bd_buddy_page is NULL and return value is 0. + * are on the same page e4b->bd_buddy_folio is NULL and return value is 0. */ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, ext4_group_t group, struct ext4_buddy *e4b, gfp_t gfp) @@ -1451,10 +1451,9 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, struct inode *inode = EXT4_SB(sb)->s_buddy_cache; int block, pnum, poff; int blocks_per_page; - struct page *page; struct folio *folio; - e4b->bd_buddy_page = NULL; + e4b->bd_buddy_folio = NULL; e4b->bd_bitmap_folio = NULL; blocks_per_page = PAGE_SIZE / sb->s_blocksize; @@ -1480,11 +1479,12 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb, } /* blocks_per_page == 1, hence we need another page for the buddy */ - page = find_or_create_page(inode->i_mapping, block + 1, gfp); - if (!page) - return -ENOMEM; - BUG_ON(page->mapping != inode->i_mapping); - e4b->bd_buddy_page = page; + folio = __filemap_get_folio(inode->i_mapping, block + 1, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, gfp); + if (IS_ERR(folio)) + return PTR_ERR(folio); + BUG_ON(folio->mapping != inode->i_mapping); + e4b->bd_buddy_folio = folio; return 0; } @@ -1494,9 +1494,9 @@ static void ext4_mb_put_buddy_page_lock(struct ext4_buddy *e4b) folio_unlock(e4b->bd_bitmap_folio); folio_put(e4b->bd_bitmap_folio); } - if (e4b->bd_buddy_page) { - unlock_page(e4b->bd_buddy_page); - put_page(e4b->bd_buddy_page); + if (e4b->bd_buddy_folio) { + folio_unlock(e4b->bd_buddy_folio); + folio_put(e4b->bd_buddy_folio); } } @@ -1511,7 +1511,6 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp) struct ext4_group_info *this_grp; struct ext4_buddy e4b; - struct page *page; struct folio *folio; int ret = 0; @@ -1548,7 +1547,7 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp) goto err; } - if (e4b.bd_buddy_page == NULL) { + if (e4b.bd_buddy_folio == NULL) { /* * If both the bitmap and buddy are in * the same page we don't need to force @@ -1558,11 +1557,11 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp) goto err; } /* init buddy cache */ - page = e4b.bd_buddy_page; - ret = ext4_mb_init_cache(page, e4b.bd_bitmap, gfp); + folio = e4b.bd_buddy_folio; + ret = ext4_mb_init_cache(&folio->page, e4b.bd_bitmap, gfp); if (ret) goto err; - if (!PageUptodate(page)) { + if (!folio_test_uptodate(folio)) { ret = -EIO; goto err; } @@ -1584,7 +1583,6 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, int block; int pnum; int poff; - struct page *page; struct folio *folio; int ret; struct ext4_group_info *grp; @@ -1603,7 +1601,7 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, e4b->bd_info = grp; e4b->bd_sb = sb; e4b->bd_group = group; - e4b->bd_buddy_page = NULL; + e4b->bd_buddy_folio = NULL; e4b->bd_bitmap_folio = NULL; if (unlikely(EXT4_MB_GRP_NEED_INIT(grp))) { @@ -1669,7 +1667,7 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, goto err; } - /* Pages marked accessed already */ + /* Folios marked accessed already */ e4b->bd_bitmap_folio = folio; e4b->bd_bitmap = folio_address(folio) + (poff * sb->s_blocksize); @@ -1677,48 +1675,49 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, pnum = block / blocks_per_page; poff = block % blocks_per_page; - page = find_get_page_flags(inode->i_mapping, pnum, FGP_ACCESSED); - if (page == NULL || !PageUptodate(page)) { - if (page) - put_page(page); - page = find_or_create_page(inode->i_mapping, pnum, gfp); - if (page) { - if (WARN_RATELIMIT(page->mapping != inode->i_mapping, - "ext4: buddy bitmap's page->mapping != inode->i_mapping\n")) { + folio = __filemap_get_folio(inode->i_mapping, pnum, FGP_ACCESSED, 0); + if (IS_ERR(folio) || !folio_test_uptodate(folio)) { + if (!IS_ERR(folio)) + folio_put(folio); + folio = __filemap_get_folio(inode->i_mapping, pnum, + FGP_LOCK | FGP_ACCESSED | FGP_CREAT, gfp); + if (!IS_ERR(folio)) { + if (WARN_RATELIMIT(folio->mapping != inode->i_mapping, + "ext4: buddy bitmap's mapping != inode->i_mapping\n")) { /* should never happen */ - unlock_page(page); + folio_unlock(folio); ret = -EINVAL; goto err; } - if (!PageUptodate(page)) { - ret = ext4_mb_init_cache(page, e4b->bd_bitmap, + if (!folio_test_uptodate(folio)) { + ret = ext4_mb_init_cache(&folio->page, e4b->bd_bitmap, gfp); if (ret) { - unlock_page(page); + folio_unlock(folio); goto err; } } - unlock_page(page); + folio_unlock(folio); } } - if (page == NULL) { - ret = -ENOMEM; + if (IS_ERR(folio)) { + ret = PTR_ERR(folio); goto err; } - if (!PageUptodate(page)) { + if (!folio_test_uptodate(folio)) { ret = -EIO; goto err; } - /* Pages marked accessed already */ - e4b->bd_buddy_page = page; - e4b->bd_buddy = page_address(page) + (poff * sb->s_blocksize); + /* Folios marked accessed already */ + e4b->bd_buddy_folio = folio; + e4b->bd_buddy = folio_address(folio) + (poff * sb->s_blocksize); return 0; err: - if (page) - put_page(page); + if (folio) + folio_put(folio); if (e4b->bd_bitmap_folio) folio_put(e4b->bd_bitmap_folio); @@ -1737,8 +1736,8 @@ static void ext4_mb_unload_buddy(struct ext4_buddy *e4b) { if (e4b->bd_bitmap_folio) folio_put(e4b->bd_bitmap_folio); - if (e4b->bd_buddy_page) - put_page(e4b->bd_buddy_page); + if (e4b->bd_buddy_folio) + folio_put(e4b->bd_buddy_folio); } @@ -2163,7 +2162,7 @@ static void ext4_mb_use_best_found(struct ext4_allocation_context *ac, */ ac->ac_bitmap_page = &e4b->bd_bitmap_folio->page; get_page(ac->ac_bitmap_page); - ac->ac_buddy_page = e4b->bd_buddy_page; + ac->ac_buddy_page = &e4b->bd_buddy_folio->page; get_page(ac->ac_buddy_page); /* store last allocated for subsequent stream allocation */ if (ac->ac_flags & EXT4_MB_STREAM_ALLOC) { @@ -3897,7 +3896,7 @@ static void ext4_free_data_in_buddy(struct super_block *sb, /* No more items in the per group rb tree * balance refcounts from ext4_mb_free_metadata() */ - put_page(e4b.bd_buddy_page); + folio_put(e4b.bd_buddy_folio); folio_put(e4b.bd_bitmap_folio); } ext4_unlock_group(sb, entry->efd_group); @@ -6321,7 +6320,7 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, BUG_ON(!ext4_handle_valid(handle)); BUG_ON(e4b->bd_bitmap_folio == NULL); - BUG_ON(e4b->bd_buddy_page == NULL); + BUG_ON(e4b->bd_buddy_folio == NULL); new_node = &new_entry->efd_node; cluster = new_entry->efd_start_cluster; @@ -6332,7 +6331,7 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, * otherwise we'll refresh it from * on-disk bitmap and lose not-yet-available * blocks */ - get_page(e4b->bd_buddy_page); + folio_get(e4b->bd_buddy_folio); folio_get(e4b->bd_bitmap_folio); } while (*n) { diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index f8b7e9d2e463..268c838ddc2d 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -215,7 +215,7 @@ struct ext4_allocation_context { #define AC_STATUS_BREAK 3 struct ext4_buddy { - struct page *bd_buddy_page; + struct folio *bd_buddy_folio; void *bd_buddy; struct folio *bd_bitmap_folio; void *bd_bitmap; -- cgit v1.2.3-59-g8ed1b From e1622a0d558200b0ccdfbf69ee29b9ac1f5c2442 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 18:28:56 +0100 Subject: ext4: convert ext4_mb_init_cache() to take a folio All callers now have a folio, so convert this function from operating on a page to operating on a folio. The folio is assumed to be a single page. Signe-off-by: Matthew Wilcox (Oracle) Link: https://lore.kernel.org/r/20240416172900.244637-4-willy@infradead.org Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 37 +++++++++++++++++++------------------ 1 file changed, 19 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 1a76e5741d36..71950abb457e 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1274,7 +1274,7 @@ static void mb_regenerate_buddy(struct ext4_buddy *e4b) * for this page; do not hold this lock when calling this routine! */ -static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) +static int ext4_mb_init_cache(struct folio *folio, char *incore, gfp_t gfp) { ext4_group_t ngroups; unsigned int blocksize; @@ -1292,13 +1292,13 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) char *bitmap; struct ext4_group_info *grinfo; - inode = page->mapping->host; + inode = folio->mapping->host; sb = inode->i_sb; ngroups = ext4_get_groups_count(sb); blocksize = i_blocksize(inode); blocks_per_page = PAGE_SIZE / blocksize; - mb_debug(sb, "init page %lu\n", page->index); + mb_debug(sb, "init folio %lu\n", folio->index); groups_per_page = blocks_per_page >> 1; if (groups_per_page == 0) @@ -1313,9 +1313,9 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) } else bh = &bhs; - first_group = page->index * blocks_per_page / 2; + first_group = folio->index * blocks_per_page / 2; - /* read all groups the page covers into the cache */ + /* read all groups the folio covers into the cache */ for (i = 0, group = first_group; i < groups_per_page; i++, group++) { if (group >= ngroups) break; @@ -1326,10 +1326,11 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) /* * If page is uptodate then we came here after online resize * which added some new uninitialized group info structs, so - * we must skip all initialized uptodate buddies on the page, + * we must skip all initialized uptodate buddies on the folio, * which may be currently in use by an allocating task. */ - if (PageUptodate(page) && !EXT4_MB_GRP_NEED_INIT(grinfo)) { + if (folio_test_uptodate(folio) && + !EXT4_MB_GRP_NEED_INIT(grinfo)) { bh[i] = NULL; continue; } @@ -1353,7 +1354,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) err = err2; } - first_block = page->index * blocks_per_page; + first_block = folio->index * blocks_per_page; for (i = 0; i < blocks_per_page; i++) { group = (first_block + i) >> 1; if (group >= ngroups) @@ -1374,7 +1375,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) * above * */ - data = page_address(page) + (i * blocksize); + data = folio_address(folio) + (i * blocksize); bitmap = bh[group - first_group]->b_data; /* @@ -1389,8 +1390,8 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) if ((first_block + i) & 1) { /* this is block of buddy */ BUG_ON(incore == NULL); - mb_debug(sb, "put buddy for group %u in page %lu/%x\n", - group, page->index, i * blocksize); + mb_debug(sb, "put buddy for group %u in folio %lu/%x\n", + group, folio->index, i * blocksize); trace_ext4_mb_buddy_bitmap_load(sb, group); grinfo->bb_fragments = 0; memset(grinfo->bb_counters, 0, @@ -1408,8 +1409,8 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) } else { /* this is block of bitmap */ BUG_ON(incore != NULL); - mb_debug(sb, "put bitmap for group %u in page %lu/%x\n", - group, page->index, i * blocksize); + mb_debug(sb, "put bitmap for group %u in folio %lu/%x\n", + group, folio->index, i * blocksize); trace_ext4_mb_bitmap_load(sb, group); /* see comments in ext4_mb_put_pa() */ @@ -1427,7 +1428,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp) incore = data; } } - SetPageUptodate(page); + folio_mark_uptodate(folio); out: if (bh) { @@ -1539,7 +1540,7 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp) } folio = e4b.bd_bitmap_folio; - ret = ext4_mb_init_cache(&folio->page, NULL, gfp); + ret = ext4_mb_init_cache(folio, NULL, gfp); if (ret) goto err; if (!folio_test_uptodate(folio)) { @@ -1558,7 +1559,7 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp) } /* init buddy cache */ folio = e4b.bd_buddy_folio; - ret = ext4_mb_init_cache(&folio->page, e4b.bd_bitmap, gfp); + ret = ext4_mb_init_cache(folio, e4b.bd_bitmap, gfp); if (ret) goto err; if (!folio_test_uptodate(folio)) { @@ -1647,7 +1648,7 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, goto err; } if (!folio_test_uptodate(folio)) { - ret = ext4_mb_init_cache(&folio->page, NULL, gfp); + ret = ext4_mb_init_cache(folio, NULL, gfp); if (ret) { folio_unlock(folio); goto err; @@ -1690,7 +1691,7 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, goto err; } if (!folio_test_uptodate(folio)) { - ret = ext4_mb_init_cache(&folio->page, e4b->bd_bitmap, + ret = ext4_mb_init_cache(folio, e4b->bd_bitmap, gfp); if (ret) { folio_unlock(folio); -- cgit v1.2.3-59-g8ed1b From ccedf35b5daa429c0e731eac6fb32b0208302c6b Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 18:28:57 +0100 Subject: ext4: convert ac_bitmap_page to ac_bitmap_folio This just carries around the bd_bitmap_folio so should also be a folio. Signed-off-by: Matthew Wilcox (Oracle) Link: https://lore.kernel.org/r/20240416172900.244637-5-willy@infradead.org Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 8 ++++---- fs/ext4/mballoc.h | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 71950abb457e..218480b20058 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2161,8 +2161,8 @@ static void ext4_mb_use_best_found(struct ext4_allocation_context *ac, * double allocate blocks. The reference is dropped * in ext4_mb_release_context */ - ac->ac_bitmap_page = &e4b->bd_bitmap_folio->page; - get_page(ac->ac_bitmap_page); + ac->ac_bitmap_folio = e4b->bd_bitmap_folio; + folio_get(ac->ac_bitmap_folio); ac->ac_buddy_page = &e4b->bd_buddy_folio->page; get_page(ac->ac_buddy_page); /* store last allocated for subsequent stream allocation */ @@ -6002,8 +6002,8 @@ static void ext4_mb_release_context(struct ext4_allocation_context *ac) ext4_mb_put_pa(ac, ac->ac_sb, pa); } - if (ac->ac_bitmap_page) - put_page(ac->ac_bitmap_page); + if (ac->ac_bitmap_folio) + folio_put(ac->ac_bitmap_folio); if (ac->ac_buddy_page) put_page(ac->ac_buddy_page); if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 268c838ddc2d..469e5516fb77 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -204,7 +204,7 @@ struct ext4_allocation_context { __u8 ac_2order; /* if request is to allocate 2^N blocks and * N > 0, the field stores N, otherwise 0 */ __u8 ac_op; /* operation, for history only */ - struct page *ac_bitmap_page; + struct folio *ac_bitmap_folio; struct page *ac_buddy_page; struct ext4_prealloc_space *ac_pa; struct ext4_locality_group *ac_lg; -- cgit v1.2.3-59-g8ed1b From c84f1510fba9fb181e6a1aa7b5fcfc67381b39c9 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 16 Apr 2024 18:28:58 +0100 Subject: ext4: convert ac_buddy_page to ac_buddy_folio This just carries around the bd_buddy_folio so should also be a folio. Signed-off-by: Matthew Wilcox (Oracle) Link: https://lore.kernel.org/r/20240416172900.244637-6-willy@infradead.org Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 8 ++++---- fs/ext4/mballoc.h | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 218480b20058..ac0ee747f4b9 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2163,8 +2163,8 @@ static void ext4_mb_use_best_found(struct ext4_allocation_context *ac, */ ac->ac_bitmap_folio = e4b->bd_bitmap_folio; folio_get(ac->ac_bitmap_folio); - ac->ac_buddy_page = &e4b->bd_buddy_folio->page; - get_page(ac->ac_buddy_page); + ac->ac_buddy_folio = e4b->bd_buddy_folio; + folio_get(ac->ac_buddy_folio); /* store last allocated for subsequent stream allocation */ if (ac->ac_flags & EXT4_MB_STREAM_ALLOC) { spin_lock(&sbi->s_md_lock); @@ -6004,8 +6004,8 @@ static void ext4_mb_release_context(struct ext4_allocation_context *ac) } if (ac->ac_bitmap_folio) folio_put(ac->ac_bitmap_folio); - if (ac->ac_buddy_page) - put_page(ac->ac_buddy_page); + if (ac->ac_buddy_folio) + folio_put(ac->ac_buddy_folio); if (ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC) mutex_unlock(&ac->ac_lg->lg_mutex); ext4_mb_collect_stats(ac); diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 469e5516fb77..d8553f1498d3 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -205,7 +205,7 @@ struct ext4_allocation_context { * N > 0, the field stores N, otherwise 0 */ __u8 ac_op; /* operation, for history only */ struct folio *ac_bitmap_folio; - struct page *ac_buddy_page; + struct folio *ac_buddy_folio; struct ext4_prealloc_space *ac_pa; struct ext4_locality_group *ac_lg; }; -- cgit v1.2.3-59-g8ed1b From 3f4830abd236d0428e50451e1ecb62e14c365e9b Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 17 Apr 2024 21:10:40 +0300 Subject: ext4: fix potential unnitialized variable Smatch complains "err" can be uninitialized in the caller. fs/ext4/indirect.c:349 ext4_alloc_branch() error: uninitialized symbol 'err'. Set the error to zero on the success path. Fixes: 8016e29f4362 ("ext4: fast commit recovery path") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/363a4673-0fb8-4adf-b4fb-90a499077276@moroto.mountain Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index ac0ee747f4b9..648989c125f2 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -6126,6 +6126,7 @@ ext4_mb_new_blocks_simple(struct ext4_allocation_request *ar, int *errp) ext4_mb_mark_bb(sb, block, 1, true); ar->len = 1; + *errp = 0; return block; } -- cgit v1.2.3-59-g8ed1b From df0b5afc62f3368d657a8fe4a8d393ac481474c2 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Fri, 19 Apr 2024 10:30:05 +0800 Subject: ext4: remove the redundant folio_wait_stable() __filemap_get_folio() with FGP_WRITEBEGIN parameter has already wait for stable folio, so remove the redundant folio_wait_stable() in ext4_da_write_begin(), it was left over from the commit cc883236b792 ("ext4: drop unnecessary journal handle in delalloc write") that removed the retry getting page logic. Fixes: cc883236b792 ("ext4: drop unnecessary journal handle in delalloc write") Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240419023005.2719050-1-yi.zhang@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 989e28c24a8c..4bae9ccf5fe0 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2887,9 +2887,6 @@ retry: if (IS_ERR(folio)) return PTR_ERR(folio); - /* In case writeback began while the folio was unlocked */ - folio_wait_stable(folio); - #ifdef CONFIG_FS_ENCRYPTION ret = ext4_block_write_begin(folio, pos, len, ext4_da_get_block_prep); #else -- cgit v1.2.3-59-g8ed1b From 8b57de1c5edde3faf8a4f6a440b7ec16bb3c81d4 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 10 Apr 2024 12:28:03 +0100 Subject: jbd2: remove redundant assignement to variable err The variable err is being assigned a value that is never read, it is being re-assigned inside the following while loop and also after the while loop. The assignment is redundant and can be removed. Cleans up clang scan build warning: fs/jbd2/commit.c:574:2: warning: Value stored to 'err' is never read [deadcode.DeadStores] Signed-off-by: Colin Ian King Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240410112803.232993-1-colin.i.king@gmail.com Signed-off-by: Theodore Ts'o --- fs/jbd2/commit.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 5e122586e06e..78a9d08ae9f8 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -571,7 +571,6 @@ void jbd2_journal_commit_transaction(journal_t *journal) J_ASSERT(commit_transaction->t_nr_buffers <= atomic_read(&commit_transaction->t_outstanding_credits)); - err = 0; bufs = 0; descriptor = NULL; while (commit_transaction->t_buffers) { -- cgit v1.2.3-59-g8ed1b From 0c0b4a49d3e7f49690a6827a41faeffad5df7e21 Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Sat, 4 May 2024 15:55:25 +0800 Subject: ext4: fix mb_cache_entry's e_refcnt leak in ext4_xattr_block_cache_find() Syzbot reports a warning as follows: ============================================ WARNING: CPU: 0 PID: 5075 at fs/mbcache.c:419 mb_cache_destroy+0x224/0x290 Modules linked in: CPU: 0 PID: 5075 Comm: syz-executor199 Not tainted 6.9.0-rc6-gb947cc5bf6d7 RIP: 0010:mb_cache_destroy+0x224/0x290 fs/mbcache.c:419 Call Trace: ext4_put_super+0x6d4/0xcd0 fs/ext4/super.c:1375 generic_shutdown_super+0x136/0x2d0 fs/super.c:641 kill_block_super+0x44/0x90 fs/super.c:1675 ext4_kill_sb+0x68/0xa0 fs/ext4/super.c:7327 [...] ============================================ This is because when finding an entry in ext4_xattr_block_cache_find(), if ext4_sb_bread() returns -ENOMEM, the ce's e_refcnt, which has already grown in the __entry_find(), won't be put away, and eventually trigger the above issue in mb_cache_destroy() due to reference count leakage. So call mb_cache_entry_put() on the -ENOMEM error branch as a quick fix. Reported-by: syzbot+dd43bd0f7474512edc47@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=dd43bd0f7474512edc47 Fixes: fb265c9cb49e ("ext4: add ext4_sb_bread() to disambiguate ENOMEM cases") Cc: stable@kernel.org Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240504075526.2254349-2-libaokun@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 04f90df8dbae..3e9b088800b1 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -3117,8 +3117,10 @@ ext4_xattr_block_cache_find(struct inode *inode, bh = ext4_sb_bread(inode->i_sb, ce->e_value, REQ_PRIO); if (IS_ERR(bh)) { - if (PTR_ERR(bh) == -ENOMEM) + if (PTR_ERR(bh) == -ENOMEM) { + mb_cache_entry_put(ea_block_cache, ce); return NULL; + } bh = NULL; EXT4_ERROR_INODE(inode, "block %lu read error", (unsigned long)ce->e_value); -- cgit v1.2.3-59-g8ed1b From dc1c4663bc493f323d6b2f9dd55c044ea920dacf Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Sat, 4 May 2024 15:55:26 +0800 Subject: ext4: propagate errors from ext4_sb_bread() in ext4_xattr_block_cache_find() In ext4_xattr_block_cache_find(), when ext4_sb_bread() returns an error, we will either continue to find the next ea block or return NULL to try to insert a new ea block. But whether ext4_sb_bread() returns -EIO or -ENOMEM, the next operation is most likely to fail with the same error. So propagate the error returned by ext4_sb_bread() to make ext4_xattr_block_set() fail to reduce pointless operations. Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240504075526.2254349-3-libaokun@huaweicloud.com Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 3e9b088800b1..6460879b9fcb 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -2028,8 +2028,13 @@ clone_block: inserted: if (!IS_LAST_ENTRY(s->first)) { - new_bh = ext4_xattr_block_cache_find(inode, header(s->base), - &ce); + new_bh = ext4_xattr_block_cache_find(inode, header(s->base), &ce); + if (IS_ERR(new_bh)) { + error = PTR_ERR(new_bh); + new_bh = NULL; + goto cleanup; + } + if (new_bh) { /* We found an identical block in the cache. */ if (new_bh == bs->bh) @@ -3094,8 +3099,8 @@ ext4_xattr_cmp(struct ext4_xattr_header *header1, * * Find an identical extended attribute block. * - * Returns a pointer to the block found, or NULL if such a block was - * not found or an error occurred. + * Returns a pointer to the block found, or NULL if such a block was not + * found, or an error pointer if an error occurred while reading ea block. */ static struct buffer_head * ext4_xattr_block_cache_find(struct inode *inode, @@ -3117,13 +3122,11 @@ ext4_xattr_block_cache_find(struct inode *inode, bh = ext4_sb_bread(inode->i_sb, ce->e_value, REQ_PRIO); if (IS_ERR(bh)) { - if (PTR_ERR(bh) == -ENOMEM) { - mb_cache_entry_put(ea_block_cache, ce); - return NULL; - } - bh = NULL; - EXT4_ERROR_INODE(inode, "block %lu read error", - (unsigned long)ce->e_value); + if (PTR_ERR(bh) != -ENOMEM) + EXT4_ERROR_INODE(inode, "block %lu read error", + (unsigned long)ce->e_value); + mb_cache_entry_put(ea_block_cache, ce); + return bh; } else if (ext4_xattr_cmp(header, BHDR(bh)) == 0) { *pce = ce; return bh; -- cgit v1.2.3-59-g8ed1b From 413e8f014c8b848e4ce939156f210df59fbd1c24 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:50:06 +0100 Subject: fuse: Convert fuse_readpages_end() to use folio_end_read() Nobody checks the error flag on fuse folios, so stop setting it. Optimise the (optional) setting of the uptodate flag and clearing of the lock flag by using folio_end_read(). Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Miklos Szeredi --- fs/fuse/file.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index b57ce4157640..f39456c65ed7 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -935,14 +935,10 @@ static void fuse_readpages_end(struct fuse_mount *fm, struct fuse_args *args, } for (i = 0; i < ap->num_pages; i++) { - struct page *page = ap->pages[i]; + struct folio *folio = page_folio(ap->pages[i]); - if (!err) - SetPageUptodate(page); - else - SetPageError(page); - unlock_page(page); - put_page(page); + folio_end_read(folio, !err); + folio_put(folio); } if (ia->ff) fuse_file_put(ia->ff, false); -- cgit v1.2.3-59-g8ed1b From 96d88f65adfbcaca153afd7d3e20d74ba379c599 Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Thu, 25 Apr 2024 06:44:00 -0400 Subject: virtiofs: include a newline in sysfs tag The internal tag string doesn't contain a newline. Append one when emitting the tag via sysfs. [Stefan] Orthogonal to the newline issue, sysfs_emit(buf, "%s", fs->tag) is needed to prevent format string injection. Signed-off-by: Brian Foster Fixes: a8f62f50b4e4 ("virtiofs: export filesystem tags through sysfs") Signed-off-by: Miklos Szeredi --- fs/fuse/virtio_fs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index 322af827a232..bb3e941b9503 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -170,7 +170,7 @@ static ssize_t tag_show(struct kobject *kobj, { struct virtio_fs *fs = container_of(kobj, struct virtio_fs, kobj); - return sysfs_emit(buf, fs->tag); + return sysfs_emit(buf, "%s\n", fs->tag); } static struct kobj_attribute virtio_fs_tag_attr = __ATTR_RO(tag); -- cgit v1.2.3-59-g8ed1b From 9fe2a036a23ceeac402c4fde8ec37c02ab25f133 Mon Sep 17 00:00:00 2001 From: Richard Fung Date: Tue, 16 Apr 2024 00:16:39 +0000 Subject: fuse: Add initial support for fs-verity MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This adds support for the FS_IOC_ENABLE_VERITY and FS_IOC_MEASURE_VERITY ioctls. The FS_IOC_READ_VERITY_METADATA is missing but from the documentation, "This is a fairly specialized use case, and most fs-verity users won’t need this ioctl." Signed-off-by: Richard Fung Acked-by: Eric Biggers Signed-off-by: Miklos Szeredi --- fs/fuse/ioctl.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+) (limited to 'fs') diff --git a/fs/fuse/ioctl.c b/fs/fuse/ioctl.c index 726640fa439e..572ce8a82ceb 100644 --- a/fs/fuse/ioctl.c +++ b/fs/fuse/ioctl.c @@ -8,6 +8,7 @@ #include #include #include +#include static ssize_t fuse_send_ioctl(struct fuse_mount *fm, struct fuse_args *args, struct fuse_ioctl_out *outarg) @@ -117,6 +118,53 @@ static int fuse_copy_ioctl_iovec(struct fuse_conn *fc, struct iovec *dst, return 0; } +/* For fs-verity, determine iov lengths from input */ +static int fuse_setup_measure_verity(unsigned long arg, struct iovec *iov) +{ + __u16 digest_size; + struct fsverity_digest __user *uarg = (void __user *)arg; + + if (copy_from_user(&digest_size, &uarg->digest_size, sizeof(digest_size))) + return -EFAULT; + + if (digest_size > SIZE_MAX - sizeof(struct fsverity_digest)) + return -EINVAL; + + iov->iov_len = sizeof(struct fsverity_digest) + digest_size; + + return 0; +} + +static int fuse_setup_enable_verity(unsigned long arg, struct iovec *iov, + unsigned int *in_iovs) +{ + struct fsverity_enable_arg enable; + struct fsverity_enable_arg __user *uarg = (void __user *)arg; + const __u32 max_buffer_len = FUSE_MAX_MAX_PAGES * PAGE_SIZE; + + if (copy_from_user(&enable, uarg, sizeof(enable))) + return -EFAULT; + + if (enable.salt_size > max_buffer_len || enable.sig_size > max_buffer_len) + return -ENOMEM; + + if (enable.salt_size > 0) { + iov++; + (*in_iovs)++; + + iov->iov_base = u64_to_user_ptr(enable.salt_ptr); + iov->iov_len = enable.salt_size; + } + + if (enable.sig_size > 0) { + iov++; + (*in_iovs)++; + + iov->iov_base = u64_to_user_ptr(enable.sig_ptr); + iov->iov_len = enable.sig_size; + } + return 0; +} /* * For ioctls, there is no generic way to determine how much memory @@ -227,6 +275,18 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg, out_iov = iov; out_iovs = 1; } + + err = 0; + switch (cmd) { + case FS_IOC_MEASURE_VERITY: + err = fuse_setup_measure_verity(arg, iov); + break; + case FS_IOC_ENABLE_VERITY: + err = fuse_setup_enable_verity(arg, iov, &in_iovs); + break; + } + if (err) + goto out; } retry: -- cgit v1.2.3-59-g8ed1b From cacd5b04e24c74a813c694ec7b26a1a370b5d666 Mon Sep 17 00:00:00 2001 From: Chunhai Guo Date: Mon, 1 Apr 2024 07:55:50 -0600 Subject: erofs: rename utils.c to zutil.c Currently, utils.c is only useful if CONFIG_EROFS_FS_ZIP is on. So let's rename it to zutil.c as well as avoid its inclusion if CONFIG_EROFS_FS_ZIP is explicitly disabled. Signed-off-by: Chunhai Guo Reviewed-by: Gao Xiang Link: https://lore.kernel.org/r/20240401135550.2550043-1-guochunhai@vivo.com Signed-off-by: Gao Xiang --- fs/erofs/Makefile | 4 +- fs/erofs/utils.c | 287 ------------------------------------------------------ fs/erofs/zutil.c | 279 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 281 insertions(+), 289 deletions(-) delete mode 100644 fs/erofs/utils.c create mode 100644 fs/erofs/zutil.c (limited to 'fs') diff --git a/fs/erofs/Makefile b/fs/erofs/Makefile index 994d0b9deddf..845eafdcee4a 100644 --- a/fs/erofs/Makefile +++ b/fs/erofs/Makefile @@ -1,9 +1,9 @@ # SPDX-License-Identifier: GPL-2.0-only obj-$(CONFIG_EROFS_FS) += erofs.o -erofs-objs := super.o inode.o data.o namei.o dir.o utils.o sysfs.o +erofs-objs := super.o inode.o data.o namei.o dir.o sysfs.o erofs-$(CONFIG_EROFS_FS_XATTR) += xattr.o -erofs-$(CONFIG_EROFS_FS_ZIP) += decompressor.o zmap.o zdata.o pcpubuf.o +erofs-$(CONFIG_EROFS_FS_ZIP) += decompressor.o zmap.o zdata.o pcpubuf.o zutil.o erofs-$(CONFIG_EROFS_FS_ZIP_LZMA) += decompressor_lzma.o erofs-$(CONFIG_EROFS_FS_ZIP_DEFLATE) += decompressor_deflate.o erofs-$(CONFIG_EROFS_FS_ONDEMAND) += fscache.o diff --git a/fs/erofs/utils.c b/fs/erofs/utils.c deleted file mode 100644 index 518bdd69c823..000000000000 --- a/fs/erofs/utils.c +++ /dev/null @@ -1,287 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Copyright (C) 2018 HUAWEI, Inc. - * https://www.huawei.com/ - */ -#include "internal.h" - -struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp) -{ - struct page *page = *pagepool; - - if (page) { - DBG_BUGON(page_ref_count(page) != 1); - *pagepool = (struct page *)page_private(page); - } else { - page = alloc_page(gfp); - } - return page; -} - -void erofs_release_pages(struct page **pagepool) -{ - while (*pagepool) { - struct page *page = *pagepool; - - *pagepool = (struct page *)page_private(page); - put_page(page); - } -} - -#ifdef CONFIG_EROFS_FS_ZIP -/* global shrink count (for all mounted EROFS instances) */ -static atomic_long_t erofs_global_shrink_cnt; - -static bool erofs_workgroup_get(struct erofs_workgroup *grp) -{ - if (lockref_get_not_zero(&grp->lockref)) - return true; - - spin_lock(&grp->lockref.lock); - if (__lockref_is_dead(&grp->lockref)) { - spin_unlock(&grp->lockref.lock); - return false; - } - - if (!grp->lockref.count++) - atomic_long_dec(&erofs_global_shrink_cnt); - spin_unlock(&grp->lockref.lock); - return true; -} - -struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb, - pgoff_t index) -{ - struct erofs_sb_info *sbi = EROFS_SB(sb); - struct erofs_workgroup *grp; - -repeat: - rcu_read_lock(); - grp = xa_load(&sbi->managed_pslots, index); - if (grp) { - if (!erofs_workgroup_get(grp)) { - /* prefer to relax rcu read side */ - rcu_read_unlock(); - goto repeat; - } - - DBG_BUGON(index != grp->index); - } - rcu_read_unlock(); - return grp; -} - -struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb, - struct erofs_workgroup *grp) -{ - struct erofs_sb_info *const sbi = EROFS_SB(sb); - struct erofs_workgroup *pre; - - DBG_BUGON(grp->lockref.count < 1); -repeat: - xa_lock(&sbi->managed_pslots); - pre = __xa_cmpxchg(&sbi->managed_pslots, grp->index, - NULL, grp, GFP_KERNEL); - if (pre) { - if (xa_is_err(pre)) { - pre = ERR_PTR(xa_err(pre)); - } else if (!erofs_workgroup_get(pre)) { - /* try to legitimize the current in-tree one */ - xa_unlock(&sbi->managed_pslots); - cond_resched(); - goto repeat; - } - grp = pre; - } - xa_unlock(&sbi->managed_pslots); - return grp; -} - -static void __erofs_workgroup_free(struct erofs_workgroup *grp) -{ - atomic_long_dec(&erofs_global_shrink_cnt); - erofs_workgroup_free_rcu(grp); -} - -void erofs_workgroup_put(struct erofs_workgroup *grp) -{ - if (lockref_put_or_lock(&grp->lockref)) - return; - - DBG_BUGON(__lockref_is_dead(&grp->lockref)); - if (grp->lockref.count == 1) - atomic_long_inc(&erofs_global_shrink_cnt); - --grp->lockref.count; - spin_unlock(&grp->lockref.lock); -} - -static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi, - struct erofs_workgroup *grp) -{ - int free = false; - - spin_lock(&grp->lockref.lock); - if (grp->lockref.count) - goto out; - - /* - * Note that all cached pages should be detached before deleted from - * the XArray. Otherwise some cached pages could be still attached to - * the orphan old workgroup when the new one is available in the tree. - */ - if (erofs_try_to_free_all_cached_folios(sbi, grp)) - goto out; - - /* - * It's impossible to fail after the workgroup is freezed, - * however in order to avoid some race conditions, add a - * DBG_BUGON to observe this in advance. - */ - DBG_BUGON(__xa_erase(&sbi->managed_pslots, grp->index) != grp); - - lockref_mark_dead(&grp->lockref); - free = true; -out: - spin_unlock(&grp->lockref.lock); - if (free) - __erofs_workgroup_free(grp); - return free; -} - -static unsigned long erofs_shrink_workstation(struct erofs_sb_info *sbi, - unsigned long nr_shrink) -{ - struct erofs_workgroup *grp; - unsigned int freed = 0; - unsigned long index; - - xa_lock(&sbi->managed_pslots); - xa_for_each(&sbi->managed_pslots, index, grp) { - /* try to shrink each valid workgroup */ - if (!erofs_try_to_release_workgroup(sbi, grp)) - continue; - xa_unlock(&sbi->managed_pslots); - - ++freed; - if (!--nr_shrink) - return freed; - xa_lock(&sbi->managed_pslots); - } - xa_unlock(&sbi->managed_pslots); - return freed; -} - -/* protected by 'erofs_sb_list_lock' */ -static unsigned int shrinker_run_no; - -/* protects the mounted 'erofs_sb_list' */ -static DEFINE_SPINLOCK(erofs_sb_list_lock); -static LIST_HEAD(erofs_sb_list); - -void erofs_shrinker_register(struct super_block *sb) -{ - struct erofs_sb_info *sbi = EROFS_SB(sb); - - mutex_init(&sbi->umount_mutex); - - spin_lock(&erofs_sb_list_lock); - list_add(&sbi->list, &erofs_sb_list); - spin_unlock(&erofs_sb_list_lock); -} - -void erofs_shrinker_unregister(struct super_block *sb) -{ - struct erofs_sb_info *const sbi = EROFS_SB(sb); - - mutex_lock(&sbi->umount_mutex); - /* clean up all remaining workgroups in memory */ - erofs_shrink_workstation(sbi, ~0UL); - - spin_lock(&erofs_sb_list_lock); - list_del(&sbi->list); - spin_unlock(&erofs_sb_list_lock); - mutex_unlock(&sbi->umount_mutex); -} - -static unsigned long erofs_shrink_count(struct shrinker *shrink, - struct shrink_control *sc) -{ - return atomic_long_read(&erofs_global_shrink_cnt); -} - -static unsigned long erofs_shrink_scan(struct shrinker *shrink, - struct shrink_control *sc) -{ - struct erofs_sb_info *sbi; - struct list_head *p; - - unsigned long nr = sc->nr_to_scan; - unsigned int run_no; - unsigned long freed = 0; - - spin_lock(&erofs_sb_list_lock); - do { - run_no = ++shrinker_run_no; - } while (run_no == 0); - - /* Iterate over all mounted superblocks and try to shrink them */ - p = erofs_sb_list.next; - while (p != &erofs_sb_list) { - sbi = list_entry(p, struct erofs_sb_info, list); - - /* - * We move the ones we do to the end of the list, so we stop - * when we see one we have already done. - */ - if (sbi->shrinker_run_no == run_no) - break; - - if (!mutex_trylock(&sbi->umount_mutex)) { - p = p->next; - continue; - } - - spin_unlock(&erofs_sb_list_lock); - sbi->shrinker_run_no = run_no; - - freed += erofs_shrink_workstation(sbi, nr - freed); - - spin_lock(&erofs_sb_list_lock); - /* Get the next list element before we move this one */ - p = p->next; - - /* - * Move this one to the end of the list to provide some - * fairness. - */ - list_move_tail(&sbi->list, &erofs_sb_list); - mutex_unlock(&sbi->umount_mutex); - - if (freed >= nr) - break; - } - spin_unlock(&erofs_sb_list_lock); - return freed; -} - -static struct shrinker *erofs_shrinker_info; - -int __init erofs_init_shrinker(void) -{ - erofs_shrinker_info = shrinker_alloc(0, "erofs-shrinker"); - if (!erofs_shrinker_info) - return -ENOMEM; - - erofs_shrinker_info->count_objects = erofs_shrink_count; - erofs_shrinker_info->scan_objects = erofs_shrink_scan; - - shrinker_register(erofs_shrinker_info); - - return 0; -} - -void erofs_exit_shrinker(void) -{ - shrinker_free(erofs_shrinker_info); -} -#endif /* !CONFIG_EROFS_FS_ZIP */ diff --git a/fs/erofs/zutil.c b/fs/erofs/zutil.c new file mode 100644 index 000000000000..8cd30ac2091f --- /dev/null +++ b/fs/erofs/zutil.c @@ -0,0 +1,279 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2018 HUAWEI, Inc. + * https://www.huawei.com/ + */ +#include "internal.h" + +static atomic_long_t erofs_global_shrink_cnt; /* for all mounted instances */ +/* protected by 'erofs_sb_list_lock' */ +static unsigned int shrinker_run_no; + +/* protects the mounted 'erofs_sb_list' */ +static DEFINE_SPINLOCK(erofs_sb_list_lock); +static LIST_HEAD(erofs_sb_list); +static struct shrinker *erofs_shrinker_info; + +struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp) +{ + struct page *page = *pagepool; + + if (page) { + DBG_BUGON(page_ref_count(page) != 1); + *pagepool = (struct page *)page_private(page); + return page; + } + return alloc_page(gfp); +} + +void erofs_release_pages(struct page **pagepool) +{ + while (*pagepool) { + struct page *page = *pagepool; + + *pagepool = (struct page *)page_private(page); + put_page(page); + } +} + +static bool erofs_workgroup_get(struct erofs_workgroup *grp) +{ + if (lockref_get_not_zero(&grp->lockref)) + return true; + + spin_lock(&grp->lockref.lock); + if (__lockref_is_dead(&grp->lockref)) { + spin_unlock(&grp->lockref.lock); + return false; + } + + if (!grp->lockref.count++) + atomic_long_dec(&erofs_global_shrink_cnt); + spin_unlock(&grp->lockref.lock); + return true; +} + +struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb, + pgoff_t index) +{ + struct erofs_sb_info *sbi = EROFS_SB(sb); + struct erofs_workgroup *grp; + +repeat: + rcu_read_lock(); + grp = xa_load(&sbi->managed_pslots, index); + if (grp) { + if (!erofs_workgroup_get(grp)) { + /* prefer to relax rcu read side */ + rcu_read_unlock(); + goto repeat; + } + + DBG_BUGON(index != grp->index); + } + rcu_read_unlock(); + return grp; +} + +struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb, + struct erofs_workgroup *grp) +{ + struct erofs_sb_info *const sbi = EROFS_SB(sb); + struct erofs_workgroup *pre; + + DBG_BUGON(grp->lockref.count < 1); +repeat: + xa_lock(&sbi->managed_pslots); + pre = __xa_cmpxchg(&sbi->managed_pslots, grp->index, + NULL, grp, GFP_KERNEL); + if (pre) { + if (xa_is_err(pre)) { + pre = ERR_PTR(xa_err(pre)); + } else if (!erofs_workgroup_get(pre)) { + /* try to legitimize the current in-tree one */ + xa_unlock(&sbi->managed_pslots); + cond_resched(); + goto repeat; + } + grp = pre; + } + xa_unlock(&sbi->managed_pslots); + return grp; +} + +static void __erofs_workgroup_free(struct erofs_workgroup *grp) +{ + atomic_long_dec(&erofs_global_shrink_cnt); + erofs_workgroup_free_rcu(grp); +} + +void erofs_workgroup_put(struct erofs_workgroup *grp) +{ + if (lockref_put_or_lock(&grp->lockref)) + return; + + DBG_BUGON(__lockref_is_dead(&grp->lockref)); + if (grp->lockref.count == 1) + atomic_long_inc(&erofs_global_shrink_cnt); + --grp->lockref.count; + spin_unlock(&grp->lockref.lock); +} + +static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi, + struct erofs_workgroup *grp) +{ + int free = false; + + spin_lock(&grp->lockref.lock); + if (grp->lockref.count) + goto out; + + /* + * Note that all cached pages should be detached before deleted from + * the XArray. Otherwise some cached pages could be still attached to + * the orphan old workgroup when the new one is available in the tree. + */ + if (erofs_try_to_free_all_cached_folios(sbi, grp)) + goto out; + + /* + * It's impossible to fail after the workgroup is freezed, + * however in order to avoid some race conditions, add a + * DBG_BUGON to observe this in advance. + */ + DBG_BUGON(__xa_erase(&sbi->managed_pslots, grp->index) != grp); + + lockref_mark_dead(&grp->lockref); + free = true; +out: + spin_unlock(&grp->lockref.lock); + if (free) + __erofs_workgroup_free(grp); + return free; +} + +static unsigned long erofs_shrink_workstation(struct erofs_sb_info *sbi, + unsigned long nr_shrink) +{ + struct erofs_workgroup *grp; + unsigned int freed = 0; + unsigned long index; + + xa_lock(&sbi->managed_pslots); + xa_for_each(&sbi->managed_pslots, index, grp) { + /* try to shrink each valid workgroup */ + if (!erofs_try_to_release_workgroup(sbi, grp)) + continue; + xa_unlock(&sbi->managed_pslots); + + ++freed; + if (!--nr_shrink) + return freed; + xa_lock(&sbi->managed_pslots); + } + xa_unlock(&sbi->managed_pslots); + return freed; +} + +void erofs_shrinker_register(struct super_block *sb) +{ + struct erofs_sb_info *sbi = EROFS_SB(sb); + + mutex_init(&sbi->umount_mutex); + + spin_lock(&erofs_sb_list_lock); + list_add(&sbi->list, &erofs_sb_list); + spin_unlock(&erofs_sb_list_lock); +} + +void erofs_shrinker_unregister(struct super_block *sb) +{ + struct erofs_sb_info *const sbi = EROFS_SB(sb); + + mutex_lock(&sbi->umount_mutex); + /* clean up all remaining workgroups in memory */ + erofs_shrink_workstation(sbi, ~0UL); + + spin_lock(&erofs_sb_list_lock); + list_del(&sbi->list); + spin_unlock(&erofs_sb_list_lock); + mutex_unlock(&sbi->umount_mutex); +} + +static unsigned long erofs_shrink_count(struct shrinker *shrink, + struct shrink_control *sc) +{ + return atomic_long_read(&erofs_global_shrink_cnt); +} + +static unsigned long erofs_shrink_scan(struct shrinker *shrink, + struct shrink_control *sc) +{ + struct erofs_sb_info *sbi; + struct list_head *p; + + unsigned long nr = sc->nr_to_scan; + unsigned int run_no; + unsigned long freed = 0; + + spin_lock(&erofs_sb_list_lock); + do { + run_no = ++shrinker_run_no; + } while (run_no == 0); + + /* Iterate over all mounted superblocks and try to shrink them */ + p = erofs_sb_list.next; + while (p != &erofs_sb_list) { + sbi = list_entry(p, struct erofs_sb_info, list); + + /* + * We move the ones we do to the end of the list, so we stop + * when we see one we have already done. + */ + if (sbi->shrinker_run_no == run_no) + break; + + if (!mutex_trylock(&sbi->umount_mutex)) { + p = p->next; + continue; + } + + spin_unlock(&erofs_sb_list_lock); + sbi->shrinker_run_no = run_no; + + freed += erofs_shrink_workstation(sbi, nr - freed); + + spin_lock(&erofs_sb_list_lock); + /* Get the next list element before we move this one */ + p = p->next; + + /* + * Move this one to the end of the list to provide some + * fairness. + */ + list_move_tail(&sbi->list, &erofs_sb_list); + mutex_unlock(&sbi->umount_mutex); + + if (freed >= nr) + break; + } + spin_unlock(&erofs_sb_list_lock); + return freed; +} + +int __init erofs_init_shrinker(void) +{ + erofs_shrinker_info = shrinker_alloc(0, "erofs-shrinker"); + if (!erofs_shrinker_info) + return -ENOMEM; + + erofs_shrinker_info->count_objects = erofs_shrink_count; + erofs_shrinker_info->scan_objects = erofs_shrink_scan; + shrinker_register(erofs_shrinker_info); + return 0; +} + +void erofs_exit_shrinker(void) +{ + shrinker_free(erofs_shrinker_info); +} -- cgit v1.2.3-59-g8ed1b From f36f3010f67611a45d66e773bc91e4c66a9abab5 Mon Sep 17 00:00:00 2001 From: Chunhai Guo Date: Tue, 2 Apr 2024 04:00:36 -0600 Subject: erofs: rename per-CPU buffers to global buffer pool and make it configurable It will cost more time if compressed buffers are allocated on demand for low-latency algorithms (like lz4) so EROFS uses per-CPU buffers to keep compressed data if in-place decompression is unfulfilled. While it is kind of wasteful of memory for a device with hundreds of CPUs, and only a small number of CPUs concurrently decompress most of the time. This patch renames it as 'global buffer pool' and makes it configurable. This allows two or more CPUs to share a common buffer to reduce memory occupation. Suggested-by: Gao Xiang Reviewed-by: Gao Xiang Signed-off-by: Chunhai Guo Link: https://lore.kernel.org/r/20240402100036.2673604-1-guochunhai@vivo.com Signed-off-by: Sandeep Dhavale Link: https://lore.kernel.org/r/20240408215231.3376659-1-dhavale@google.com Signed-off-by: Gao Xiang --- fs/erofs/Makefile | 2 +- fs/erofs/decompressor.c | 6 +- fs/erofs/internal.h | 14 ++--- fs/erofs/pcpubuf.c | 148 ------------------------------------------------ fs/erofs/super.c | 9 ++- fs/erofs/zutil.c | 148 ++++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 166 insertions(+), 161 deletions(-) delete mode 100644 fs/erofs/pcpubuf.c (limited to 'fs') diff --git a/fs/erofs/Makefile b/fs/erofs/Makefile index 845eafdcee4a..20d1ec422443 100644 --- a/fs/erofs/Makefile +++ b/fs/erofs/Makefile @@ -3,7 +3,7 @@ obj-$(CONFIG_EROFS_FS) += erofs.o erofs-objs := super.o inode.o data.o namei.o dir.o sysfs.o erofs-$(CONFIG_EROFS_FS_XATTR) += xattr.o -erofs-$(CONFIG_EROFS_FS_ZIP) += decompressor.o zmap.o zdata.o pcpubuf.o zutil.o +erofs-$(CONFIG_EROFS_FS_ZIP) += decompressor.o zmap.o zdata.o zutil.o erofs-$(CONFIG_EROFS_FS_ZIP_LZMA) += decompressor_lzma.o erofs-$(CONFIG_EROFS_FS_ZIP_DEFLATE) += decompressor_deflate.o erofs-$(CONFIG_EROFS_FS_ONDEMAND) += fscache.o diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c index 2ec9b2bb628d..e1239d886984 100644 --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -54,7 +54,7 @@ static int z_erofs_load_lz4_config(struct super_block *sb, sbi->lz4.max_distance_pages = distance ? DIV_ROUND_UP(distance, PAGE_SIZE) + 1 : LZ4_MAX_DISTANCE_PAGES; - return erofs_pcpubuf_growsize(sbi->lz4.max_pclusterblks); + return z_erofs_gbuf_growsize(sbi->lz4.max_pclusterblks); } /* @@ -159,7 +159,7 @@ static void *z_erofs_lz4_handle_overlap(struct z_erofs_lz4_decompress_ctx *ctx, docopy: /* Or copy compressed data which can be overlapped to per-CPU buffer */ in = rq->in; - src = erofs_get_pcpubuf(ctx->inpages); + src = z_erofs_get_gbuf(ctx->inpages); if (!src) { DBG_BUGON(1); kunmap_local(inpage); @@ -260,7 +260,7 @@ static int z_erofs_lz4_decompress_mem(struct z_erofs_lz4_decompress_ctx *ctx, } else if (maptype == 1) { vm_unmap_ram(src, ctx->inpages); } else if (maptype == 2) { - erofs_put_pcpubuf(src); + z_erofs_put_gbuf(src); } else if (maptype != 3) { DBG_BUGON(1); return -EFAULT; diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index d28ccfc0352b..ee080d042ab3 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -463,11 +463,11 @@ int erofs_try_to_free_all_cached_folios(struct erofs_sb_info *sbi, struct erofs_workgroup *egrp); int z_erofs_map_blocks_iter(struct inode *inode, struct erofs_map_blocks *map, int flags); -void *erofs_get_pcpubuf(unsigned int requiredpages); -void erofs_put_pcpubuf(void *ptr); -int erofs_pcpubuf_growsize(unsigned int nrpages); -void __init erofs_pcpubuf_init(void); -void erofs_pcpubuf_exit(void); +void *z_erofs_get_gbuf(unsigned int requiredpages); +void z_erofs_put_gbuf(void *ptr); +int z_erofs_gbuf_growsize(unsigned int nrpages); +int __init z_erofs_gbuf_init(void); +void z_erofs_gbuf_exit(void); int erofs_init_managed_cache(struct super_block *sb); int z_erofs_parse_cfgs(struct super_block *sb, struct erofs_super_block *dsb); #else @@ -477,8 +477,8 @@ static inline int erofs_init_shrinker(void) { return 0; } static inline void erofs_exit_shrinker(void) {} static inline int z_erofs_init_zip_subsystem(void) { return 0; } static inline void z_erofs_exit_zip_subsystem(void) {} -static inline void erofs_pcpubuf_init(void) {} -static inline void erofs_pcpubuf_exit(void) {} +static inline int z_erofs_gbuf_init(void) { return 0; } +static inline void z_erofs_gbuf_exit(void) {} static inline int erofs_init_managed_cache(struct super_block *sb) { return 0; } #endif /* !CONFIG_EROFS_FS_ZIP */ diff --git a/fs/erofs/pcpubuf.c b/fs/erofs/pcpubuf.c deleted file mode 100644 index c7a4b1d77069..000000000000 --- a/fs/erofs/pcpubuf.c +++ /dev/null @@ -1,148 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Copyright (C) Gao Xiang - * - * For low-latency decompression algorithms (e.g. lz4), reserve consecutive - * per-CPU virtual memory (in pages) in advance to store such inplace I/O - * data if inplace decompression is failed (due to unmet inplace margin for - * example). - */ -#include "internal.h" - -struct erofs_pcpubuf { - raw_spinlock_t lock; - void *ptr; - struct page **pages; - unsigned int nrpages; -}; - -static DEFINE_PER_CPU(struct erofs_pcpubuf, erofs_pcb); - -void *erofs_get_pcpubuf(unsigned int requiredpages) - __acquires(pcb->lock) -{ - struct erofs_pcpubuf *pcb = &get_cpu_var(erofs_pcb); - - raw_spin_lock(&pcb->lock); - /* check if the per-CPU buffer is too small */ - if (requiredpages > pcb->nrpages) { - raw_spin_unlock(&pcb->lock); - put_cpu_var(erofs_pcb); - /* (for sparse checker) pretend pcb->lock is still taken */ - __acquire(pcb->lock); - return NULL; - } - return pcb->ptr; -} - -void erofs_put_pcpubuf(void *ptr) __releases(pcb->lock) -{ - struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, smp_processor_id()); - - DBG_BUGON(pcb->ptr != ptr); - raw_spin_unlock(&pcb->lock); - put_cpu_var(erofs_pcb); -} - -/* the next step: support per-CPU page buffers hotplug */ -int erofs_pcpubuf_growsize(unsigned int nrpages) -{ - static DEFINE_MUTEX(pcb_resize_mutex); - static unsigned int pcb_nrpages; - struct page *pagepool = NULL; - int delta, cpu, ret, i; - - mutex_lock(&pcb_resize_mutex); - delta = nrpages - pcb_nrpages; - ret = 0; - /* avoid shrinking pcpubuf, since no idea how many fses rely on */ - if (delta <= 0) - goto out; - - for_each_possible_cpu(cpu) { - struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, cpu); - struct page **pages, **oldpages; - void *ptr, *old_ptr; - - pages = kmalloc_array(nrpages, sizeof(*pages), GFP_KERNEL); - if (!pages) { - ret = -ENOMEM; - break; - } - - for (i = 0; i < nrpages; ++i) { - pages[i] = erofs_allocpage(&pagepool, GFP_KERNEL); - if (!pages[i]) { - ret = -ENOMEM; - oldpages = pages; - goto free_pagearray; - } - } - ptr = vmap(pages, nrpages, VM_MAP, PAGE_KERNEL); - if (!ptr) { - ret = -ENOMEM; - oldpages = pages; - goto free_pagearray; - } - raw_spin_lock(&pcb->lock); - old_ptr = pcb->ptr; - pcb->ptr = ptr; - oldpages = pcb->pages; - pcb->pages = pages; - i = pcb->nrpages; - pcb->nrpages = nrpages; - raw_spin_unlock(&pcb->lock); - - if (!oldpages) { - DBG_BUGON(old_ptr); - continue; - } - - if (old_ptr) - vunmap(old_ptr); -free_pagearray: - while (i) - erofs_pagepool_add(&pagepool, oldpages[--i]); - kfree(oldpages); - if (ret) - break; - } - pcb_nrpages = nrpages; - erofs_release_pages(&pagepool); -out: - mutex_unlock(&pcb_resize_mutex); - return ret; -} - -void __init erofs_pcpubuf_init(void) -{ - int cpu; - - for_each_possible_cpu(cpu) { - struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, cpu); - - raw_spin_lock_init(&pcb->lock); - } -} - -void erofs_pcpubuf_exit(void) -{ - int cpu, i; - - for_each_possible_cpu(cpu) { - struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, cpu); - - if (pcb->ptr) { - vunmap(pcb->ptr); - pcb->ptr = NULL; - } - if (!pcb->pages) - continue; - - for (i = 0; i < pcb->nrpages; ++i) - if (pcb->pages[i]) - put_page(pcb->pages[i]); - kfree(pcb->pages); - pcb->pages = NULL; - } -} diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 30b49b2eee53..c1dae1fb949b 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -859,7 +859,10 @@ static int __init erofs_module_init(void) if (err) goto deflate_err; - erofs_pcpubuf_init(); + err = z_erofs_gbuf_init(); + if (err) + goto gbuf_err; + err = z_erofs_init_zip_subsystem(); if (err) goto zip_err; @@ -879,6 +882,8 @@ fs_err: sysfs_err: z_erofs_exit_zip_subsystem(); zip_err: + z_erofs_gbuf_exit(); +gbuf_err: z_erofs_deflate_exit(); deflate_err: z_erofs_lzma_exit(); @@ -902,7 +907,7 @@ static void __exit erofs_module_exit(void) z_erofs_lzma_exit(); erofs_exit_shrinker(); kmem_cache_destroy(erofs_inode_cachep); - erofs_pcpubuf_exit(); + z_erofs_gbuf_exit(); } static int erofs_statfs(struct dentry *dentry, struct kstatfs *buf) diff --git a/fs/erofs/zutil.c b/fs/erofs/zutil.c index 8cd30ac2091f..2fa90b10b985 100644 --- a/fs/erofs/zutil.c +++ b/fs/erofs/zutil.c @@ -5,6 +5,18 @@ */ #include "internal.h" +struct z_erofs_gbuf { + spinlock_t lock; + void *ptr; + struct page **pages; + unsigned int nrpages; +}; + +static struct z_erofs_gbuf *z_erofs_gbufpool; +static unsigned int z_erofs_gbuf_count, z_erofs_gbuf_nrpages; + +module_param_named(global_buffers, z_erofs_gbuf_count, uint, 0444); + static atomic_long_t erofs_global_shrink_cnt; /* for all mounted instances */ /* protected by 'erofs_sb_list_lock' */ static unsigned int shrinker_run_no; @@ -14,6 +26,142 @@ static DEFINE_SPINLOCK(erofs_sb_list_lock); static LIST_HEAD(erofs_sb_list); static struct shrinker *erofs_shrinker_info; +static unsigned int z_erofs_gbuf_id(void) +{ + return raw_smp_processor_id() % z_erofs_gbuf_count; +} + +void *z_erofs_get_gbuf(unsigned int requiredpages) + __acquires(gbuf->lock) +{ + struct z_erofs_gbuf *gbuf; + + gbuf = &z_erofs_gbufpool[z_erofs_gbuf_id()]; + spin_lock(&gbuf->lock); + /* check if the buffer is too small */ + if (requiredpages > gbuf->nrpages) { + spin_unlock(&gbuf->lock); + /* (for sparse checker) pretend gbuf->lock is still taken */ + __acquire(gbuf->lock); + return NULL; + } + return gbuf->ptr; +} + +void z_erofs_put_gbuf(void *ptr) __releases(gbuf->lock) +{ + struct z_erofs_gbuf *gbuf; + + gbuf = &z_erofs_gbufpool[z_erofs_gbuf_id()]; + DBG_BUGON(gbuf->ptr != ptr); + spin_unlock(&gbuf->lock); +} + +int z_erofs_gbuf_growsize(unsigned int nrpages) +{ + static DEFINE_MUTEX(gbuf_resize_mutex); + struct page *pagepool = NULL; + int delta, ret, i, j; + + mutex_lock(&gbuf_resize_mutex); + delta = nrpages - z_erofs_gbuf_nrpages; + ret = 0; + /* avoid shrinking gbufs, since no idea how many fses rely on */ + if (delta <= 0) + goto out; + + for (i = 0; i < z_erofs_gbuf_count; ++i) { + struct z_erofs_gbuf *gbuf = &z_erofs_gbufpool[i]; + struct page **pages, **tmp_pages; + void *ptr, *old_ptr = NULL; + + ret = -ENOMEM; + tmp_pages = kcalloc(nrpages, sizeof(*tmp_pages), GFP_KERNEL); + if (!tmp_pages) + break; + for (j = 0; j < nrpages; ++j) { + tmp_pages[j] = erofs_allocpage(&pagepool, GFP_KERNEL); + if (!tmp_pages[j]) + goto free_pagearray; + } + ptr = vmap(tmp_pages, nrpages, VM_MAP, PAGE_KERNEL); + if (!ptr) + goto free_pagearray; + + pages = tmp_pages; + spin_lock(&gbuf->lock); + old_ptr = gbuf->ptr; + gbuf->ptr = ptr; + tmp_pages = gbuf->pages; + gbuf->pages = pages; + j = gbuf->nrpages; + gbuf->nrpages = nrpages; + spin_unlock(&gbuf->lock); + ret = 0; + if (!tmp_pages) { + DBG_BUGON(old_ptr); + continue; + } + + if (old_ptr) + vunmap(old_ptr); +free_pagearray: + while (j) + erofs_pagepool_add(&pagepool, tmp_pages[--j]); + kfree(tmp_pages); + if (ret) + break; + } + z_erofs_gbuf_nrpages = nrpages; + erofs_release_pages(&pagepool); +out: + mutex_unlock(&gbuf_resize_mutex); + return ret; +} + +int __init z_erofs_gbuf_init(void) +{ + unsigned int i = num_possible_cpus(); + + if (!z_erofs_gbuf_count) + z_erofs_gbuf_count = i; + else + z_erofs_gbuf_count = min(z_erofs_gbuf_count, i); + + z_erofs_gbufpool = kcalloc(z_erofs_gbuf_count, + sizeof(*z_erofs_gbufpool), GFP_KERNEL); + if (!z_erofs_gbufpool) + return -ENOMEM; + + for (i = 0; i < z_erofs_gbuf_count; ++i) + spin_lock_init(&z_erofs_gbufpool[i].lock); + return 0; +} + +void z_erofs_gbuf_exit(void) +{ + int i; + + for (i = 0; i < z_erofs_gbuf_count; ++i) { + struct z_erofs_gbuf *gbuf = &z_erofs_gbufpool[i]; + + if (gbuf->ptr) { + vunmap(gbuf->ptr); + gbuf->ptr = NULL; + } + + if (!gbuf->pages) + continue; + + for (i = 0; i < gbuf->nrpages; ++i) + if (gbuf->pages[i]) + put_page(gbuf->pages[i]); + kfree(gbuf->pages); + gbuf->pages = NULL; + } + kfree(z_erofs_gbufpool); +} + struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp) { struct page *page = *pagepool; -- cgit v1.2.3-59-g8ed1b From d6db47e571dcaecaeaafa8840d00ae849ae3907b Mon Sep 17 00:00:00 2001 From: Chunhai Guo Date: Tue, 2 Apr 2024 03:27:57 -0600 Subject: erofs: do not use pagepool in z_erofs_gbuf_growsize() Let's use alloc_pages_bulk_array() for simplicity and get rid of unnecessary pagepool. Signed-off-by: Chunhai Guo Reviewed-by: Gao Xiang Link: https://lore.kernel.org/r/20240402092757.2635257-1-guochunhai@vivo.com Signed-off-by: Gao Xiang --- fs/erofs/zutil.c | 67 ++++++++++++++++++++++++++------------------------------ 1 file changed, 31 insertions(+), 36 deletions(-) (limited to 'fs') diff --git a/fs/erofs/zutil.c b/fs/erofs/zutil.c index 2fa90b10b985..bc88dc4fe8bd 100644 --- a/fs/erofs/zutil.c +++ b/fs/erofs/zutil.c @@ -60,63 +60,58 @@ void z_erofs_put_gbuf(void *ptr) __releases(gbuf->lock) int z_erofs_gbuf_growsize(unsigned int nrpages) { static DEFINE_MUTEX(gbuf_resize_mutex); - struct page *pagepool = NULL; - int delta, ret, i, j; + struct page **tmp_pages = NULL; + struct z_erofs_gbuf *gbuf; + void *ptr, *old_ptr; + int last, i, j; mutex_lock(&gbuf_resize_mutex); - delta = nrpages - z_erofs_gbuf_nrpages; - ret = 0; /* avoid shrinking gbufs, since no idea how many fses rely on */ - if (delta <= 0) - goto out; + if (nrpages <= z_erofs_gbuf_nrpages) { + mutex_unlock(&gbuf_resize_mutex); + return 0; + } for (i = 0; i < z_erofs_gbuf_count; ++i) { - struct z_erofs_gbuf *gbuf = &z_erofs_gbufpool[i]; - struct page **pages, **tmp_pages; - void *ptr, *old_ptr = NULL; - - ret = -ENOMEM; + gbuf = &z_erofs_gbufpool[i]; tmp_pages = kcalloc(nrpages, sizeof(*tmp_pages), GFP_KERNEL); if (!tmp_pages) - break; - for (j = 0; j < nrpages; ++j) { - tmp_pages[j] = erofs_allocpage(&pagepool, GFP_KERNEL); - if (!tmp_pages[j]) - goto free_pagearray; - } + goto out; + + for (j = 0; j < gbuf->nrpages; ++j) + tmp_pages[j] = gbuf->pages[j]; + do { + last = j; + j = alloc_pages_bulk_array(GFP_KERNEL, nrpages, + tmp_pages); + if (last == j) + goto out; + } while (j != nrpages); + ptr = vmap(tmp_pages, nrpages, VM_MAP, PAGE_KERNEL); if (!ptr) - goto free_pagearray; + goto out; - pages = tmp_pages; spin_lock(&gbuf->lock); + kfree(gbuf->pages); + gbuf->pages = tmp_pages; old_ptr = gbuf->ptr; gbuf->ptr = ptr; - tmp_pages = gbuf->pages; - gbuf->pages = pages; - j = gbuf->nrpages; gbuf->nrpages = nrpages; spin_unlock(&gbuf->lock); - ret = 0; - if (!tmp_pages) { - DBG_BUGON(old_ptr); - continue; - } - if (old_ptr) vunmap(old_ptr); -free_pagearray: - while (j) - erofs_pagepool_add(&pagepool, tmp_pages[--j]); - kfree(tmp_pages); - if (ret) - break; } z_erofs_gbuf_nrpages = nrpages; - erofs_release_pages(&pagepool); out: + if (i < z_erofs_gbuf_count && tmp_pages) { + for (j = 0; j < nrpages; ++j) + if (tmp_pages[j] && tmp_pages[j] != gbuf->pages[j]) + __free_page(tmp_pages[j]); + kfree(tmp_pages); + } mutex_unlock(&gbuf_resize_mutex); - return ret; + return i < z_erofs_gbuf_count ? -ENOMEM : 0; } int __init z_erofs_gbuf_init(void) -- cgit v1.2.3-59-g8ed1b From 0f6273ab46375b62c8dd5c987ce7c15877602831 Mon Sep 17 00:00:00 2001 From: Chunhai Guo Date: Tue, 2 Apr 2024 07:15:23 -0600 Subject: erofs: add a reserved buffer pool for lz4 decompression This adds a special global buffer pool (in the end) for reserved pages. Using a reserved pool for LZ4 decompression significantly reduces the time spent on extra temporary page allocation for the extreme cases in low memory scenarios. The table below shows the reduction in time spent on page allocation for LZ4 decompression when using a reserved pool. The results were obtained from multi-app launch benchmarks on ARM64 Android devices running the 5.15 kernel with an 8-core CPU and 8GB of memory. In the benchmark, we launched 16 frequently-used apps, and the camera app was the last one in each round. The data in the table is the average time of camera app for each round. After using the reserved pool, there was an average improvement of 150ms in the overall launch time of our camera app, which was obtained from the systrace log. +--------------+---------------+--------------+---------+ | | w/o page pool | w/ page pool | diff | +--------------+---------------+--------------+---------+ | Average (ms) | 3434 | 21 | -99.38% | +--------------+---------------+--------------+---------+ Based on the benchmark logs, 64 pages are sufficient for 95% of scenarios. This value can be adjusted with a module parameter `reserved_pages`. The default value is 0. This pool is currently only used for the LZ4 decompressor, but it can be applied to more decompressors if needed. Signed-off-by: Chunhai Guo Reviewed-by: Gao Xiang Link: https://lore.kernel.org/r/20240402131523.2703948-1-guochunhai@vivo.com Signed-off-by: Gao Xiang --- fs/erofs/decompressor.c | 2 +- fs/erofs/internal.h | 6 ++++- fs/erofs/zutil.c | 61 +++++++++++++++++++++++++++++++++++++------------ 3 files changed, 52 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c index e1239d886984..d2fe8130819e 100644 --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -111,7 +111,7 @@ static int z_erofs_lz4_prepare_dstpages(struct z_erofs_lz4_decompress_ctx *ctx, victim = availables[--top]; get_page(victim); } else { - victim = erofs_allocpage(pagepool, rq->gfp); + victim = __erofs_allocpage(pagepool, rq->gfp, true); if (!victim) return -ENOMEM; set_page_private(victim, Z_EROFS_SHORTLIVED_PAGE); diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index ee080d042ab3..53ebba952a2f 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -438,7 +438,11 @@ void erofs_unregister_sysfs(struct super_block *sb); int __init erofs_init_sysfs(void); void erofs_exit_sysfs(void); -struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp); +struct page *__erofs_allocpage(struct page **pagepool, gfp_t gfp, bool tryrsv); +static inline struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp) +{ + return __erofs_allocpage(pagepool, gfp, false); +} static inline void erofs_pagepool_add(struct page **pagepool, struct page *page) { set_page_private(page, (unsigned long)*pagepool); diff --git a/fs/erofs/zutil.c b/fs/erofs/zutil.c index bc88dc4fe8bd..036024bce9f7 100644 --- a/fs/erofs/zutil.c +++ b/fs/erofs/zutil.c @@ -12,10 +12,12 @@ struct z_erofs_gbuf { unsigned int nrpages; }; -static struct z_erofs_gbuf *z_erofs_gbufpool; -static unsigned int z_erofs_gbuf_count, z_erofs_gbuf_nrpages; +static struct z_erofs_gbuf *z_erofs_gbufpool, *z_erofs_rsvbuf; +static unsigned int z_erofs_gbuf_count, z_erofs_gbuf_nrpages, + z_erofs_rsv_nrpages; module_param_named(global_buffers, z_erofs_gbuf_count, uint, 0444); +module_param_named(reserved_pages, z_erofs_rsv_nrpages, uint, 0444); static atomic_long_t erofs_global_shrink_cnt; /* for all mounted instances */ /* protected by 'erofs_sb_list_lock' */ @@ -116,19 +118,30 @@ out: int __init z_erofs_gbuf_init(void) { - unsigned int i = num_possible_cpus(); + unsigned int i, total = num_possible_cpus(); - if (!z_erofs_gbuf_count) - z_erofs_gbuf_count = i; - else - z_erofs_gbuf_count = min(z_erofs_gbuf_count, i); + if (z_erofs_gbuf_count) + total = min(z_erofs_gbuf_count, total); + z_erofs_gbuf_count = total; - z_erofs_gbufpool = kcalloc(z_erofs_gbuf_count, - sizeof(*z_erofs_gbufpool), GFP_KERNEL); + /* The last (special) global buffer is the reserved buffer */ + total += !!z_erofs_rsv_nrpages; + + z_erofs_gbufpool = kcalloc(total, sizeof(*z_erofs_gbufpool), + GFP_KERNEL); if (!z_erofs_gbufpool) return -ENOMEM; - for (i = 0; i < z_erofs_gbuf_count; ++i) + if (z_erofs_rsv_nrpages) { + z_erofs_rsvbuf = &z_erofs_gbufpool[total - 1]; + z_erofs_rsvbuf->pages = kcalloc(z_erofs_rsv_nrpages, + sizeof(*z_erofs_rsvbuf->pages), GFP_KERNEL); + if (!z_erofs_rsvbuf->pages) { + z_erofs_rsvbuf = NULL; + z_erofs_rsv_nrpages = 0; + } + } + for (i = 0; i < total; ++i) spin_lock_init(&z_erofs_gbufpool[i].lock); return 0; } @@ -137,7 +150,7 @@ void z_erofs_gbuf_exit(void) { int i; - for (i = 0; i < z_erofs_gbuf_count; ++i) { + for (i = 0; i < z_erofs_gbuf_count + (!!z_erofs_rsvbuf); ++i) { struct z_erofs_gbuf *gbuf = &z_erofs_gbufpool[i]; if (gbuf->ptr) { @@ -157,16 +170,22 @@ void z_erofs_gbuf_exit(void) kfree(z_erofs_gbufpool); } -struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp) +struct page *__erofs_allocpage(struct page **pagepool, gfp_t gfp, bool tryrsv) { struct page *page = *pagepool; if (page) { - DBG_BUGON(page_ref_count(page) != 1); *pagepool = (struct page *)page_private(page); - return page; + } else if (tryrsv && z_erofs_rsvbuf && z_erofs_rsvbuf->nrpages) { + spin_lock(&z_erofs_rsvbuf->lock); + if (z_erofs_rsvbuf->nrpages) + page = z_erofs_rsvbuf->pages[--z_erofs_rsvbuf->nrpages]; + spin_unlock(&z_erofs_rsvbuf->lock); } - return alloc_page(gfp); + if (!page) + page = alloc_page(gfp); + DBG_BUGON(page && page_ref_count(page) != 1); + return page; } void erofs_release_pages(struct page **pagepool) @@ -175,6 +194,18 @@ void erofs_release_pages(struct page **pagepool) struct page *page = *pagepool; *pagepool = (struct page *)page_private(page); + /* try to fill reserved global pool first */ + if (z_erofs_rsvbuf && z_erofs_rsvbuf->nrpages < + z_erofs_rsv_nrpages) { + spin_lock(&z_erofs_rsvbuf->lock); + if (z_erofs_rsvbuf->nrpages < z_erofs_rsv_nrpages) { + z_erofs_rsvbuf->pages[z_erofs_rsvbuf->nrpages++] + = page; + spin_unlock(&z_erofs_rsvbuf->lock); + continue; + } + spin_unlock(&z_erofs_rsvbuf->lock); + } put_page(page); } } -- cgit v1.2.3-59-g8ed1b From 1872df8dcd87bd1e623e1a70076d08636b9c473d Mon Sep 17 00:00:00 2001 From: Hongzhen Luo Date: Tue, 9 Apr 2024 19:30:22 +0800 Subject: erofs: derive fsid from on-disk UUID for .statfs() if possible Use the superblock's UUID to generate the fsid when it's non-null. Reviewed-by: Gao Xiang Reviewed-by: Jingbo Xu Signed-off-by: Hongzhen Luo Link: https://lore.kernel.org/r/20240409113022.74720-1-hongzhen@linux.alibaba.com Signed-off-by: Gao Xiang --- fs/erofs/super.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/erofs/super.c b/fs/erofs/super.c index c1dae1fb949b..e3438f1a7bac 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -914,22 +914,20 @@ static int erofs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; struct erofs_sb_info *sbi = EROFS_SB(sb); - u64 id = 0; - - if (!erofs_is_fscache_mode(sb)) - id = huge_encode_dev(sb->s_bdev->bd_dev); buf->f_type = sb->s_magic; buf->f_bsize = sb->s_blocksize; buf->f_blocks = sbi->total_blocks; buf->f_bfree = buf->f_bavail = 0; - buf->f_files = ULLONG_MAX; buf->f_ffree = ULLONG_MAX - sbi->inos; - buf->f_namelen = EROFS_NAME_LEN; - buf->f_fsid = u64_to_fsid(id); + if (uuid_is_null(&sb->s_uuid)) + buf->f_fsid = u64_to_fsid(erofs_is_fscache_mode(sb) ? 0 : + huge_encode_dev(sb->s_bdev->bd_dev)); + else + buf->f_fsid = uuid_to_fsid(sb->s_uuid.b); return 0; } -- cgit v1.2.3-59-g8ed1b From d69189428d50c9a8de6475f3c8c241f86f4e764a Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Wed, 8 May 2024 20:33:57 +0800 Subject: erofs: clean up z_erofs_load_full_lcluster() Only four lcluster types here, remove redundant code. No real logic changes. Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20240508123357.3266173-1-hsiangkao@linux.alibaba.com --- fs/erofs/erofs_fs.h | 5 +---- fs/erofs/zmap.c | 21 +++++---------------- 2 files changed, 6 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h index a03ec70ba6f2..550baf1729d4 100644 --- a/fs/erofs/erofs_fs.h +++ b/fs/erofs/erofs_fs.h @@ -396,8 +396,7 @@ enum { Z_EROFS_LCLUSTER_TYPE_MAX }; -#define Z_EROFS_LI_LCLUSTER_TYPE_BITS 2 -#define Z_EROFS_LI_LCLUSTER_TYPE_BIT 0 +#define Z_EROFS_LI_LCLUSTER_TYPE_MASK (Z_EROFS_LCLUSTER_TYPE_MAX - 1) /* (noncompact only, HEAD) This pcluster refers to partial decompressed data */ #define Z_EROFS_LI_PARTIAL_REF (1 << 15) @@ -451,8 +450,6 @@ static inline void erofs_check_ondisk_layout_definitions(void) sizeof(struct z_erofs_lcluster_index)); BUILD_BUG_ON(sizeof(struct erofs_deviceslot) != 128); - BUILD_BUG_ON(BIT(Z_EROFS_LI_LCLUSTER_TYPE_BITS) < - Z_EROFS_LCLUSTER_TYPE_MAX - 1); /* exclude old compiler versions like gcc 7.5.0 */ BUILD_BUG_ON(__builtin_constant_p(fmh) ? fmh != cpu_to_le64(1ULL << 63) : 0); diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index e313c936351d..26637a60eba5 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -31,7 +31,7 @@ static int z_erofs_load_full_lcluster(struct z_erofs_maprecorder *m, vi->inode_isize + vi->xattr_isize) + lcn * sizeof(struct z_erofs_lcluster_index); struct z_erofs_lcluster_index *di; - unsigned int advise, type; + unsigned int advise; m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb, erofs_blknr(inode->i_sb, pos), EROFS_KMAP); @@ -43,10 +43,8 @@ static int z_erofs_load_full_lcluster(struct z_erofs_maprecorder *m, di = m->kaddr + erofs_blkoff(inode->i_sb, pos); advise = le16_to_cpu(di->di_advise); - type = (advise >> Z_EROFS_LI_LCLUSTER_TYPE_BIT) & - ((1 << Z_EROFS_LI_LCLUSTER_TYPE_BITS) - 1); - switch (type) { - case Z_EROFS_LCLUSTER_TYPE_NONHEAD: + m->type = advise & Z_EROFS_LI_LCLUSTER_TYPE_MASK; + if (m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) { m->clusterofs = 1 << vi->z_logical_clusterbits; m->delta[0] = le16_to_cpu(di->di_u.delta[0]); if (m->delta[0] & Z_EROFS_LI_D0_CBLKCNT) { @@ -60,24 +58,15 @@ static int z_erofs_load_full_lcluster(struct z_erofs_maprecorder *m, m->delta[0] = 1; } m->delta[1] = le16_to_cpu(di->di_u.delta[1]); - break; - case Z_EROFS_LCLUSTER_TYPE_PLAIN: - case Z_EROFS_LCLUSTER_TYPE_HEAD1: - case Z_EROFS_LCLUSTER_TYPE_HEAD2: - if (advise & Z_EROFS_LI_PARTIAL_REF) - m->partialref = true; + } else { + m->partialref = !!(advise & Z_EROFS_LI_PARTIAL_REF); m->clusterofs = le16_to_cpu(di->di_clusterofs); if (m->clusterofs >= 1 << vi->z_logical_clusterbits) { DBG_BUGON(1); return -EFSCORRUPTED; } m->pblk = le32_to_cpu(di->di_u.blkaddr); - break; - default: - DBG_BUGON(1); - return -EOPNOTSUPP; } - m->type = type; return 0; } -- cgit v1.2.3-59-g8ed1b From 36defdd9d7c605ba042962ef1eeb66a9d3ff5884 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 25 Apr 2024 03:27:16 +0900 Subject: nilfs2: convert to use the new mount API Convert nilfs2 to use the new mount API. [sandeen@redhat.com: v2] Link: https://lkml.kernel.org/r/33d078a7-9072-4d8e-a3a9-dec23d4191da@redhat.com Link: https://lkml.kernel.org/r/20240425190526.10905-1-konishi.ryusuke@gmail.com [konishi.ryusuke: fixed missing SB_RDONLY flag repair in nilfs_reconfigure] Link: https://lkml.kernel.org/r/33d078a7-9072-4d8e-a3a9-dec23d4191da@redhat.com Link: https://lkml.kernel.org/r/20240424182716.6024-1-konishi.ryusuke@gmail.com Signed-off-by: Eric Sandeen Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton --- fs/nilfs2/nilfs.h | 4 +- fs/nilfs2/super.c | 388 ++++++++++++++++++++++---------------------------- fs/nilfs2/the_nilfs.c | 5 +- fs/nilfs2/the_nilfs.h | 6 +- 4 files changed, 174 insertions(+), 229 deletions(-) (limited to 'fs') diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index 2e29b98ba8ba..728e90be3570 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -335,8 +335,8 @@ void __nilfs_error(struct super_block *sb, const char *function, extern struct nilfs_super_block * nilfs_read_super_block(struct super_block *, u64, int, struct buffer_head **); -extern int nilfs_store_magic_and_option(struct super_block *, - struct nilfs_super_block *, char *); +extern int nilfs_store_magic(struct super_block *sb, + struct nilfs_super_block *sbp); extern int nilfs_check_feature_compatibility(struct super_block *, struct nilfs_super_block *); extern void nilfs_set_log_cursor(struct nilfs_super_block *, diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index ac24ed109ce9..e835e1f5a712 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -29,13 +29,13 @@ #include #include #include -#include #include #include #include #include #include #include +#include #include "nilfs.h" #include "export.h" #include "mdt.h" @@ -61,7 +61,6 @@ struct kmem_cache *nilfs_segbuf_cachep; struct kmem_cache *nilfs_btree_path_cache; static int nilfs_setup_super(struct super_block *sb, int is_mount); -static int nilfs_remount(struct super_block *sb, int *flags, char *data); void __nilfs_msg(struct super_block *sb, const char *fmt, ...) { @@ -702,105 +701,98 @@ static const struct super_operations nilfs_sops = { .freeze_fs = nilfs_freeze, .unfreeze_fs = nilfs_unfreeze, .statfs = nilfs_statfs, - .remount_fs = nilfs_remount, .show_options = nilfs_show_options }; enum { - Opt_err_cont, Opt_err_panic, Opt_err_ro, - Opt_barrier, Opt_nobarrier, Opt_snapshot, Opt_order, Opt_norecovery, - Opt_discard, Opt_nodiscard, Opt_err, + Opt_err, Opt_barrier, Opt_snapshot, Opt_order, Opt_norecovery, + Opt_discard, }; -static match_table_t tokens = { - {Opt_err_cont, "errors=continue"}, - {Opt_err_panic, "errors=panic"}, - {Opt_err_ro, "errors=remount-ro"}, - {Opt_barrier, "barrier"}, - {Opt_nobarrier, "nobarrier"}, - {Opt_snapshot, "cp=%u"}, - {Opt_order, "order=%s"}, - {Opt_norecovery, "norecovery"}, - {Opt_discard, "discard"}, - {Opt_nodiscard, "nodiscard"}, - {Opt_err, NULL} +static const struct constant_table nilfs_param_err[] = { + {"continue", NILFS_MOUNT_ERRORS_CONT}, + {"panic", NILFS_MOUNT_ERRORS_PANIC}, + {"remount-ro", NILFS_MOUNT_ERRORS_RO}, + {} }; -static int parse_options(char *options, struct super_block *sb, int is_remount) -{ - struct the_nilfs *nilfs = sb->s_fs_info; - char *p; - substring_t args[MAX_OPT_ARGS]; - - if (!options) - return 1; - - while ((p = strsep(&options, ",")) != NULL) { - int token; +static const struct fs_parameter_spec nilfs_param_spec[] = { + fsparam_enum ("errors", Opt_err, nilfs_param_err), + fsparam_flag_no ("barrier", Opt_barrier), + fsparam_u64 ("cp", Opt_snapshot), + fsparam_string ("order", Opt_order), + fsparam_flag ("norecovery", Opt_norecovery), + fsparam_flag_no ("discard", Opt_discard), + {} +}; - if (!*p) - continue; +struct nilfs_fs_context { + unsigned long ns_mount_opt; + __u64 cno; +}; - token = match_token(p, tokens, args); - switch (token) { - case Opt_barrier: - nilfs_set_opt(nilfs, BARRIER); - break; - case Opt_nobarrier: +static int nilfs_parse_param(struct fs_context *fc, struct fs_parameter *param) +{ + struct nilfs_fs_context *nilfs = fc->fs_private; + int is_remount = fc->purpose == FS_CONTEXT_FOR_RECONFIGURE; + struct fs_parse_result result; + int opt; + + opt = fs_parse(fc, nilfs_param_spec, param, &result); + if (opt < 0) + return opt; + + switch (opt) { + case Opt_barrier: + if (result.negated) nilfs_clear_opt(nilfs, BARRIER); - break; - case Opt_order: - if (strcmp(args[0].from, "relaxed") == 0) - /* Ordered data semantics */ - nilfs_clear_opt(nilfs, STRICT_ORDER); - else if (strcmp(args[0].from, "strict") == 0) - /* Strict in-order semantics */ - nilfs_set_opt(nilfs, STRICT_ORDER); - else - return 0; - break; - case Opt_err_panic: - nilfs_write_opt(nilfs, ERROR_MODE, ERRORS_PANIC); - break; - case Opt_err_ro: - nilfs_write_opt(nilfs, ERROR_MODE, ERRORS_RO); - break; - case Opt_err_cont: - nilfs_write_opt(nilfs, ERROR_MODE, ERRORS_CONT); - break; - case Opt_snapshot: - if (is_remount) { - nilfs_err(sb, - "\"%s\" option is invalid for remount", - p); - return 0; - } - break; - case Opt_norecovery: - nilfs_set_opt(nilfs, NORECOVERY); - break; - case Opt_discard: - nilfs_set_opt(nilfs, DISCARD); - break; - case Opt_nodiscard: - nilfs_clear_opt(nilfs, DISCARD); - break; - default: - nilfs_err(sb, "unrecognized mount option \"%s\"", p); - return 0; + else + nilfs_set_opt(nilfs, BARRIER); + break; + case Opt_order: + if (strcmp(param->string, "relaxed") == 0) + /* Ordered data semantics */ + nilfs_clear_opt(nilfs, STRICT_ORDER); + else if (strcmp(param->string, "strict") == 0) + /* Strict in-order semantics */ + nilfs_set_opt(nilfs, STRICT_ORDER); + else + return -EINVAL; + break; + case Opt_err: + nilfs->ns_mount_opt &= ~NILFS_MOUNT_ERROR_MODE; + nilfs->ns_mount_opt |= result.uint_32; + break; + case Opt_snapshot: + if (is_remount) { + struct super_block *sb = fc->root->d_sb; + + nilfs_err(sb, + "\"%s\" option is invalid for remount", + param->key); + return -EINVAL; + } + if (result.uint_64 == 0) { + nilfs_err(NULL, + "invalid option \"cp=0\": invalid checkpoint number 0"); + return -EINVAL; } + nilfs->cno = result.uint_64; + break; + case Opt_norecovery: + nilfs_set_opt(nilfs, NORECOVERY); + break; + case Opt_discard: + if (result.negated) + nilfs_clear_opt(nilfs, DISCARD); + else + nilfs_set_opt(nilfs, DISCARD); + break; + default: + return -EINVAL; } - return 1; -} - -static inline void -nilfs_set_default_options(struct super_block *sb, - struct nilfs_super_block *sbp) -{ - struct the_nilfs *nilfs = sb->s_fs_info; - nilfs->ns_mount_opt = - NILFS_MOUNT_ERRORS_RO | NILFS_MOUNT_BARRIER; + return 0; } static int nilfs_setup_super(struct super_block *sb, int is_mount) @@ -857,9 +849,8 @@ struct nilfs_super_block *nilfs_read_super_block(struct super_block *sb, return (struct nilfs_super_block *)((char *)(*pbh)->b_data + offset); } -int nilfs_store_magic_and_option(struct super_block *sb, - struct nilfs_super_block *sbp, - char *data) +int nilfs_store_magic(struct super_block *sb, + struct nilfs_super_block *sbp) { struct the_nilfs *nilfs = sb->s_fs_info; @@ -870,14 +861,12 @@ int nilfs_store_magic_and_option(struct super_block *sb, sb->s_flags |= SB_NOATIME; #endif - nilfs_set_default_options(sb, sbp); - nilfs->ns_resuid = le16_to_cpu(sbp->s_def_resuid); nilfs->ns_resgid = le16_to_cpu(sbp->s_def_resgid); nilfs->ns_interval = le32_to_cpu(sbp->s_c_interval); nilfs->ns_watermark = le32_to_cpu(sbp->s_c_block_max); - return !parse_options(data, sb, 0) ? -EINVAL : 0; + return 0; } int nilfs_check_feature_compatibility(struct super_block *sb, @@ -1035,17 +1024,17 @@ int nilfs_checkpoint_is_mounted(struct super_block *sb, __u64 cno) /** * nilfs_fill_super() - initialize a super block instance * @sb: super_block - * @data: mount options - * @silent: silent mode flag + * @fc: filesystem context * * This function is called exclusively by nilfs->ns_mount_mutex. * So, the recovery process is protected from other simultaneous mounts. */ static int -nilfs_fill_super(struct super_block *sb, void *data, int silent) +nilfs_fill_super(struct super_block *sb, struct fs_context *fc) { struct the_nilfs *nilfs; struct nilfs_root *fsroot; + struct nilfs_fs_context *ctx = fc->fs_private; __u64 cno; int err; @@ -1055,10 +1044,13 @@ nilfs_fill_super(struct super_block *sb, void *data, int silent) sb->s_fs_info = nilfs; - err = init_nilfs(nilfs, sb, (char *)data); + err = init_nilfs(nilfs, sb); if (err) goto failed_nilfs; + /* Copy in parsed mount options */ + nilfs->ns_mount_opt = ctx->ns_mount_opt; + sb->s_op = &nilfs_sops; sb->s_export_op = &nilfs_export_ops; sb->s_root = NULL; @@ -1117,34 +1109,25 @@ nilfs_fill_super(struct super_block *sb, void *data, int silent) return err; } -static int nilfs_remount(struct super_block *sb, int *flags, char *data) +static int nilfs_reconfigure(struct fs_context *fc) { + struct nilfs_fs_context *ctx = fc->fs_private; + struct super_block *sb = fc->root->d_sb; struct the_nilfs *nilfs = sb->s_fs_info; - unsigned long old_sb_flags; - unsigned long old_mount_opt; int err; sync_filesystem(sb); - old_sb_flags = sb->s_flags; - old_mount_opt = nilfs->ns_mount_opt; - - if (!parse_options(data, sb, 1)) { - err = -EINVAL; - goto restore_opts; - } - sb->s_flags = (sb->s_flags & ~SB_POSIXACL); err = -EINVAL; if (!nilfs_valid_fs(nilfs)) { nilfs_warn(sb, "couldn't remount because the filesystem is in an incomplete recovery state"); - goto restore_opts; + goto ignore_opts; } - - if ((bool)(*flags & SB_RDONLY) == sb_rdonly(sb)) + if ((bool)(fc->sb_flags & SB_RDONLY) == sb_rdonly(sb)) goto out; - if (*flags & SB_RDONLY) { + if (fc->sb_flags & SB_RDONLY) { sb->s_flags |= SB_RDONLY; /* @@ -1172,138 +1155,67 @@ static int nilfs_remount(struct super_block *sb, int *flags, char *data) "couldn't remount RDWR because of unsupported optional features (%llx)", (unsigned long long)features); err = -EROFS; - goto restore_opts; + goto ignore_opts; } sb->s_flags &= ~SB_RDONLY; root = NILFS_I(d_inode(sb->s_root))->i_root; err = nilfs_attach_log_writer(sb, root); - if (err) - goto restore_opts; + if (err) { + sb->s_flags |= SB_RDONLY; + goto ignore_opts; + } down_write(&nilfs->ns_sem); nilfs_setup_super(sb, true); up_write(&nilfs->ns_sem); } out: - return 0; - - restore_opts: - sb->s_flags = old_sb_flags; - nilfs->ns_mount_opt = old_mount_opt; - return err; -} - -struct nilfs_super_data { - __u64 cno; - int flags; -}; - -static int nilfs_parse_snapshot_option(const char *option, - const substring_t *arg, - struct nilfs_super_data *sd) -{ - unsigned long long val; - const char *msg = NULL; - int err; - - if (!(sd->flags & SB_RDONLY)) { - msg = "read-only option is not specified"; - goto parse_error; - } - - err = kstrtoull(arg->from, 0, &val); - if (err) { - if (err == -ERANGE) - msg = "too large checkpoint number"; - else - msg = "malformed argument"; - goto parse_error; - } else if (val == 0) { - msg = "invalid checkpoint number 0"; - goto parse_error; - } - sd->cno = val; - return 0; - -parse_error: - nilfs_err(NULL, "invalid option \"%s\": %s", option, msg); - return 1; -} - -/** - * nilfs_identify - pre-read mount options needed to identify mount instance - * @data: mount options - * @sd: nilfs_super_data - */ -static int nilfs_identify(char *data, struct nilfs_super_data *sd) -{ - char *p, *options = data; - substring_t args[MAX_OPT_ARGS]; - int token; - int ret = 0; - - do { - p = strsep(&options, ","); - if (p != NULL && *p) { - token = match_token(p, tokens, args); - if (token == Opt_snapshot) - ret = nilfs_parse_snapshot_option(p, &args[0], - sd); - } - if (!options) - break; - BUG_ON(options == data); - *(options - 1) = ','; - } while (!ret); - return ret; -} + sb->s_flags = (sb->s_flags & ~SB_POSIXACL); + /* Copy over parsed remount options */ + nilfs->ns_mount_opt = ctx->ns_mount_opt; -static int nilfs_set_bdev_super(struct super_block *s, void *data) -{ - s->s_dev = *(dev_t *)data; return 0; -} -static int nilfs_test_bdev_super(struct super_block *s, void *data) -{ - return !(s->s_iflags & SB_I_RETIRED) && s->s_dev == *(dev_t *)data; + ignore_opts: + return err; } -static struct dentry * -nilfs_mount(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data) +static int +nilfs_get_tree(struct fs_context *fc) { - struct nilfs_super_data sd = { .flags = flags }; + struct nilfs_fs_context *ctx = fc->fs_private; struct super_block *s; dev_t dev; int err; - if (nilfs_identify(data, &sd)) - return ERR_PTR(-EINVAL); + if (ctx->cno && !(fc->sb_flags & SB_RDONLY)) { + nilfs_err(NULL, + "invalid option \"cp=%llu\": read-only option is not specified", + ctx->cno); + return -EINVAL; + } - err = lookup_bdev(dev_name, &dev); + err = lookup_bdev(fc->source, &dev); if (err) - return ERR_PTR(err); + return err; - s = sget(fs_type, nilfs_test_bdev_super, nilfs_set_bdev_super, flags, - &dev); + s = sget_dev(fc, dev); if (IS_ERR(s)) - return ERR_CAST(s); + return PTR_ERR(s); if (!s->s_root) { - err = setup_bdev_super(s, flags, NULL); + err = setup_bdev_super(s, fc->sb_flags, fc); if (!err) - err = nilfs_fill_super(s, data, - flags & SB_SILENT ? 1 : 0); + err = nilfs_fill_super(s, fc); if (err) goto failed_super; s->s_flags |= SB_ACTIVE; - } else if (!sd.cno) { + } else if (!ctx->cno) { if (nilfs_tree_is_busy(s->s_root)) { - if ((flags ^ s->s_flags) & SB_RDONLY) { + if ((fc->sb_flags ^ s->s_flags) & SB_RDONLY) { nilfs_err(s, "the device already has a %s mount.", sb_rdonly(s) ? "read-only" : "read/write"); @@ -1312,37 +1224,75 @@ nilfs_mount(struct file_system_type *fs_type, int flags, } } else { /* - * Try remount to setup mount states if the current + * Try reconfigure to setup mount states if the current * tree is not mounted and only snapshots use this sb. + * + * Since nilfs_reconfigure() requires fc->root to be + * set, set it first and release it on failure. */ - err = nilfs_remount(s, &flags, data); - if (err) + fc->root = dget(s->s_root); + err = nilfs_reconfigure(fc); + if (err) { + dput(fc->root); + fc->root = NULL; /* prevent double release */ goto failed_super; + } + return 0; } } - if (sd.cno) { + if (ctx->cno) { struct dentry *root_dentry; - err = nilfs_attach_snapshot(s, sd.cno, &root_dentry); + err = nilfs_attach_snapshot(s, ctx->cno, &root_dentry); if (err) goto failed_super; - return root_dentry; + fc->root = root_dentry; + return 0; } - return dget(s->s_root); + fc->root = dget(s->s_root); + return 0; failed_super: deactivate_locked_super(s); - return ERR_PTR(err); + return err; +} + +static void nilfs_free_fc(struct fs_context *fc) +{ + kfree(fc->fs_private); +} + +static const struct fs_context_operations nilfs_context_ops = { + .parse_param = nilfs_parse_param, + .get_tree = nilfs_get_tree, + .reconfigure = nilfs_reconfigure, + .free = nilfs_free_fc, +}; + +static int nilfs_init_fs_context(struct fs_context *fc) +{ + struct nilfs_fs_context *ctx; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + ctx->ns_mount_opt = NILFS_MOUNT_ERRORS_RO | NILFS_MOUNT_BARRIER; + fc->fs_private = ctx; + fc->ops = &nilfs_context_ops; + + return 0; } struct file_system_type nilfs_fs_type = { .owner = THIS_MODULE, .name = "nilfs2", - .mount = nilfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, + .init_fs_context = nilfs_init_fs_context, + .parameters = nilfs_param_spec, }; MODULE_ALIAS_FS("nilfs2"); diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index 2ae2c1bbf6d1..db322068678f 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -659,7 +659,6 @@ static int nilfs_load_super_block(struct the_nilfs *nilfs, * init_nilfs - initialize a NILFS instance. * @nilfs: the_nilfs structure * @sb: super block - * @data: mount options * * init_nilfs() performs common initialization per block device (e.g. * reading the super block, getting disk layout information, initializing @@ -668,7 +667,7 @@ static int nilfs_load_super_block(struct the_nilfs *nilfs, * Return Value: On success, 0 is returned. On error, a negative error * code is returned. */ -int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb, char *data) +int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb) { struct nilfs_super_block *sbp; int blocksize; @@ -686,7 +685,7 @@ int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb, char *data) if (err) goto out; - err = nilfs_store_magic_and_option(sb, sbp, data); + err = nilfs_store_magic(sb, sbp); if (err) goto failed_sbh; diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h index cd4ae1b8ae16..85da0629415d 100644 --- a/fs/nilfs2/the_nilfs.h +++ b/fs/nilfs2/the_nilfs.h @@ -219,10 +219,6 @@ THE_NILFS_FNS(PURGING, purging) #define nilfs_set_opt(nilfs, opt) \ ((nilfs)->ns_mount_opt |= NILFS_MOUNT_##opt) #define nilfs_test_opt(nilfs, opt) ((nilfs)->ns_mount_opt & NILFS_MOUNT_##opt) -#define nilfs_write_opt(nilfs, mask, opt) \ - ((nilfs)->ns_mount_opt = \ - (((nilfs)->ns_mount_opt & ~NILFS_MOUNT_##mask) | \ - NILFS_MOUNT_##opt)) \ /** * struct nilfs_root - nilfs root object @@ -276,7 +272,7 @@ static inline int nilfs_sb_will_flip(struct the_nilfs *nilfs) void nilfs_set_last_segment(struct the_nilfs *, sector_t, u64, __u64); struct the_nilfs *alloc_nilfs(struct super_block *sb); void destroy_nilfs(struct the_nilfs *nilfs); -int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb, char *data); +int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb); int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb); unsigned long nilfs_nrsvsegs(struct the_nilfs *nilfs, unsigned long nsegs); void nilfs_set_nsegments(struct the_nilfs *nilfs, unsigned long nsegs); -- cgit v1.2.3-59-g8ed1b From f492fb365699448949a2eb4aa295e0e8fbf79b10 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Tue, 23 Apr 2024 23:30:18 +0100 Subject: ocfs2: remove redundant assignment to variable status Variable status is being assigned and error code that is never read, it is being assigned inside of a do-while loop. The assignment is redundant and can be removed. Cleans up clang scan build warning: fs/ocfs2/dlm/dlmdomain.c:1530:2: warning: Value stored to 'status' is never read [deadcode.DeadStores] Link: https://lkml.kernel.org/r/20240423223018.1573213-1-colin.i.king@gmail.com Signed-off-by: Colin Ian King Acked-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Gang He Cc: Jun Piao Cc: Heming Zhao Cc: Dan Carpenter Signed-off-by: Andrew Morton --- fs/ocfs2/dlm/dlmdomain.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 2e0a2f338282..2018501b2249 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -1527,7 +1527,6 @@ static void dlm_send_join_asserts(struct dlm_ctxt *dlm, { int status, node, live; - status = 0; node = -1; while ((node = find_next_bit(node_map, O2NM_MAX_NODES, node + 1)) < O2NM_MAX_NODES) { -- cgit v1.2.3-59-g8ed1b From 675f02e5e65651553c945a464e4f7630859f2032 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:50:17 +0100 Subject: squashfs: convert squashfs_symlink_read_folio to use folio APIs Remove use of page APIs, return the errno instead of 0, switch from kmap_atomic to kmap_local and use folio_end_read() to unify the two exit paths. Link: https://lkml.kernel.org/r/20240420025029.2166544-23-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Tested-by: Phillip Lougher Reviewed-by: Phillip Lougher Signed-off-by: Andrew Morton --- fs/squashfs/symlink.c | 35 ++++++++++++++++------------------- 1 file changed, 16 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/squashfs/symlink.c b/fs/squashfs/symlink.c index 2bf977a52c2c..6ef735bd841a 100644 --- a/fs/squashfs/symlink.c +++ b/fs/squashfs/symlink.c @@ -32,20 +32,19 @@ static int squashfs_symlink_read_folio(struct file *file, struct folio *folio) { - struct page *page = &folio->page; - struct inode *inode = page->mapping->host; + struct inode *inode = folio->mapping->host; struct super_block *sb = inode->i_sb; struct squashfs_sb_info *msblk = sb->s_fs_info; - int index = page->index << PAGE_SHIFT; + int index = folio_pos(folio); u64 block = squashfs_i(inode)->start; int offset = squashfs_i(inode)->offset; int length = min_t(int, i_size_read(inode) - index, PAGE_SIZE); - int bytes, copied; + int bytes, copied, error; void *pageaddr; struct squashfs_cache_entry *entry; TRACE("Entered squashfs_symlink_readpage, page index %ld, start block " - "%llx, offset %x\n", page->index, block, offset); + "%llx, offset %x\n", folio->index, block, offset); /* * Skip index bytes into symlink metadata. @@ -57,14 +56,15 @@ static int squashfs_symlink_read_folio(struct file *file, struct folio *folio) ERROR("Unable to read symlink [%llx:%x]\n", squashfs_i(inode)->start, squashfs_i(inode)->offset); - goto error_out; + error = bytes; + goto out; } } /* * Read length bytes from symlink metadata. Squashfs_read_metadata * is not used here because it can sleep and we want to use - * kmap_atomic to map the page. Instead call the underlying + * kmap_local to map the folio. Instead call the underlying * squashfs_cache_get routine. As length bytes may overlap metadata * blocks, we may need to call squashfs_cache_get multiple times. */ @@ -75,29 +75,26 @@ static int squashfs_symlink_read_folio(struct file *file, struct folio *folio) squashfs_i(inode)->start, squashfs_i(inode)->offset); squashfs_cache_put(entry); - goto error_out; + error = entry->error; + goto out; } - pageaddr = kmap_atomic(page); + pageaddr = kmap_local_folio(folio, 0); copied = squashfs_copy_data(pageaddr + bytes, entry, offset, length - bytes); if (copied == length - bytes) memset(pageaddr + length, 0, PAGE_SIZE - length); else block = entry->next_index; - kunmap_atomic(pageaddr); + kunmap_local(pageaddr); squashfs_cache_put(entry); } - flush_dcache_page(page); - SetPageUptodate(page); - unlock_page(page); - return 0; - -error_out: - SetPageError(page); - unlock_page(page); - return 0; + flush_dcache_folio(folio); + error = 0; +out: + folio_end_read(folio, error == 0); + return error; } -- cgit v1.2.3-59-g8ed1b From bbf45b7e68555569489ab2428dd9c23960cdc9bf Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:50:18 +0100 Subject: squashfs: remove calls to set the folio error flag Nobody checks the error flag on squashfs folios, so stop setting it. Link: https://lkml.kernel.org/r/20240420025029.2166544-24-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Tested-by: Phillip Lougher Reviewed-by: Phillip Lougher Signed-off-by: Andrew Morton --- fs/squashfs/file.c | 6 +----- fs/squashfs/file_direct.c | 3 +-- 2 files changed, 2 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c index e8df6430444b..a8c1e7f9a609 100644 --- a/fs/squashfs/file.c +++ b/fs/squashfs/file.c @@ -375,8 +375,6 @@ void squashfs_fill_page(struct page *page, struct squashfs_cache_entry *buffer, flush_dcache_page(page); if (copied == avail) SetPageUptodate(page); - else - SetPageError(page); } /* Copy data into page cache */ @@ -471,7 +469,7 @@ static int squashfs_read_folio(struct file *file, struct folio *folio) res = read_blocklist(inode, index, &block); if (res < 0) - goto error_out; + goto out; if (res == 0) res = squashfs_readpage_sparse(page, expected); @@ -483,8 +481,6 @@ static int squashfs_read_folio(struct file *file, struct folio *folio) if (!res) return 0; -error_out: - SetPageError(page); out: pageaddr = kmap_atomic(page); memset(pageaddr, 0, PAGE_SIZE); diff --git a/fs/squashfs/file_direct.c b/fs/squashfs/file_direct.c index 763a3f7a75f6..2a689ce71de9 100644 --- a/fs/squashfs/file_direct.c +++ b/fs/squashfs/file_direct.c @@ -106,14 +106,13 @@ int squashfs_readpage_block(struct page *target_page, u64 block, int bsize, return 0; mark_errored: - /* Decompression failed, mark pages as errored. Target_page is + /* Decompression failed. Target_page is * dealt with by the caller */ for (i = 0; i < pages; i++) { if (page[i] == NULL || page[i] == target_page) continue; flush_dcache_page(page[i]); - SetPageError(page[i]); unlock_page(page[i]); put_page(page[i]); } -- cgit v1.2.3-59-g8ed1b From 91d743a9c8299de1fc1b47428d8bb4c85face00f Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Tue, 30 Apr 2024 17:00:19 +0900 Subject: nilfs2: make superblock data array index computation sparse friendly Upon running sparse, "warning: dubious: x & !y" is output at an array index calculation within nilfs_load_super_block(). The calculation is not wrong, but to eliminate the sparse warning, replace it with an equivalent calculation. Also, add a comment to make it easier to understand what the unintuitive array index calculation is doing and whether it's correct. Link: https://lkml.kernel.org/r/20240430080019.4242-3-konishi.ryusuke@gmail.com Fixes: e339ad31f599 ("nilfs2: introduce secondary super block") Signed-off-by: Ryusuke Konishi Cc: Bart Van Assche Cc: Jens Axboe Cc: kernel test robot Signed-off-by: Andrew Morton --- fs/nilfs2/the_nilfs.c | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index db322068678f..f41d7b6d432c 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -592,7 +592,7 @@ static int nilfs_load_super_block(struct the_nilfs *nilfs, struct nilfs_super_block **sbp = nilfs->ns_sbp; struct buffer_head **sbh = nilfs->ns_sbh; u64 sb2off, devsize = bdev_nr_bytes(nilfs->ns_bdev); - int valid[2], swp = 0; + int valid[2], swp = 0, older; if (devsize < NILFS_SEG_MIN_BLOCKS * NILFS_MIN_BLOCK_SIZE + 4096) { nilfs_err(sb, "device size too small"); @@ -648,9 +648,25 @@ static int nilfs_load_super_block(struct the_nilfs *nilfs, if (swp) nilfs_swap_super_block(nilfs); + /* + * Calculate the array index of the older superblock data. + * If one has been dropped, set index 0 pointing to the remaining one, + * otherwise set index 1 pointing to the old one (including if both + * are the same). + * + * Divided case valid[0] valid[1] swp -> older + * ------------------------------------------------------------- + * Both SBs are invalid 0 0 N/A (Error) + * SB1 is invalid 0 1 1 0 + * SB2 is invalid 1 0 0 0 + * SB2 is newer 1 1 1 0 + * SB2 is older or the same 1 1 0 1 + */ + older = valid[1] ^ swp; + nilfs->ns_sbwcount = 0; nilfs->ns_sbwtime = le64_to_cpu(sbp[0]->s_wtime); - nilfs->ns_prot_seq = le64_to_cpu(sbp[valid[1] & !swp]->s_last_seq); + nilfs->ns_prot_seq = le64_to_cpu(sbp[older]->s_last_seq); *sbpp = sbp[0]; return 0; } -- cgit v1.2.3-59-g8ed1b From 4bbf9c3b53e637eb3a14ee27b996300ce88e752a Mon Sep 17 00:00:00 2001 From: Allen Pais Date: Mon, 6 May 2024 19:37:00 +0000 Subject: fs/coredump: Enable dynamic configuration of max file note size Introduce the capability to dynamically configure the maximum file note size for ELF core dumps via sysctl. Why is this being done? We have observed that during a crash when there are more than 65k mmaps in memory, the existing fixed limit on the size of the ELF notes section becomes a bottleneck. The notes section quickly reaches its capacity, leading to incomplete memory segment information in the resulting coredump. This truncation compromises the utility of the coredumps, as crucial information about the memory state at the time of the crash might be omitted. This enhancement removes the previous static limit of 4MB, allowing system administrators to adjust the size based on system-specific requirements or constraints. Eg: $ sysctl -a | grep core_file_note_size_limit kernel.core_file_note_size_limit = 4194304 $ sysctl -n kernel.core_file_note_size_limit 4194304 $echo 519304 > /proc/sys/kernel/core_file_note_size_limit $sysctl -n kernel.core_file_note_size_limit 519304 Attempting to write beyond the ceiling value of 16MB $echo 17194304 > /proc/sys/kernel/core_file_note_size_limit bash: echo: write error: Invalid argument Signed-off-by: Vijay Nag Signed-off-by: Allen Pais Link: https://lore.kernel.org/r/20240506193700.7884-1-apais@linux.microsoft.com Signed-off-by: Kees Cook --- fs/binfmt_elf.c | 7 +++++-- fs/coredump.c | 17 +++++++++++++++++ include/linux/coredump.h | 2 ++ 3 files changed, 24 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 7862962f7a85..b5a25ee49eea 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1567,7 +1567,6 @@ static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata, fill_note(note, "CORE", NT_SIGINFO, sizeof(*csigdata), csigdata); } -#define MAX_FILE_NOTE_SIZE (4*1024*1024) /* * Format of NT_FILE note: * @@ -1595,8 +1594,12 @@ static int fill_files_note(struct memelfnote *note, struct coredump_params *cprm names_ofs = (2 + 3 * count) * sizeof(data[0]); alloc: - if (size >= MAX_FILE_NOTE_SIZE) /* paranoia check */ + /* paranoia check */ + if (size >= core_file_note_size_limit) { + pr_warn_once("coredump Note size too large: %u (does kernel.core_file_note_size_limit sysctl need adjustment?\n", + size); return -EINVAL; + } size = round_up(size, PAGE_SIZE); /* * "size" can be 0 here legitimately. diff --git a/fs/coredump.c b/fs/coredump.c index be6403b4b14b..317065e3eb9b 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -56,10 +56,15 @@ static bool dump_vma_snapshot(struct coredump_params *cprm); static void free_vma_snapshot(struct coredump_params *cprm); +#define CORE_FILE_NOTE_SIZE_DEFAULT (4*1024*1024) +/* Define a reasonable max cap */ +#define CORE_FILE_NOTE_SIZE_MAX (16*1024*1024) + static int core_uses_pid; static unsigned int core_pipe_limit; static char core_pattern[CORENAME_MAX_SIZE] = "core"; static int core_name_size = CORENAME_MAX_SIZE; +unsigned int core_file_note_size_limit = CORE_FILE_NOTE_SIZE_DEFAULT; struct core_name { char *corename; @@ -998,6 +1003,9 @@ static int proc_dostring_coredump(struct ctl_table *table, int write, return error; } +static const unsigned int core_file_note_size_min = CORE_FILE_NOTE_SIZE_DEFAULT; +static const unsigned int core_file_note_size_max = CORE_FILE_NOTE_SIZE_MAX; + static struct ctl_table coredump_sysctls[] = { { .procname = "core_uses_pid", @@ -1020,6 +1028,15 @@ static struct ctl_table coredump_sysctls[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "core_file_note_size_limit", + .data = &core_file_note_size_limit, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_douintvec_minmax, + .extra1 = (unsigned int *)&core_file_note_size_min, + .extra2 = (unsigned int *)&core_file_note_size_max, + }, }; static int __init init_fs_coredump_sysctls(void) diff --git a/include/linux/coredump.h b/include/linux/coredump.h index d3eba4360150..0904ba010341 100644 --- a/include/linux/coredump.h +++ b/include/linux/coredump.h @@ -30,6 +30,8 @@ struct coredump_params { struct core_vma_metadata *vma_meta; }; +extern unsigned int core_file_note_size_limit; + /* * These are the only things you should do on a core-file: use only these * functions to write out all the necessary info. -- cgit v1.2.3-59-g8ed1b From 7f016edaa0f385da0c37eee1ebb64c7f6929c533 Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Wed, 8 May 2024 10:14:00 +0200 Subject: fscrypt: try to avoid refing parent dentry in fscrypt_file_open Merely checking if the directory is encrypted happens for every open when using ext4, at the moment refing and unrefing the parent, costing 2 atomics and serializing opens of different files. The most common case of encryption not being used can be checked for with RCU instead. Sample result from open1_processes -t 20 ("Separate file open/close") from will-it-scale on Sapphire Rapids (ops/s): before: 12539898 after: 25575494 (+103%) v2: - add a comment justifying rcu usage, submitted by Eric Biggers - whack spurious IS_ENCRYPTED check from the refed case Signed-off-by: Mateusz Guzik Link: https://lore.kernel.org/r/20240508081400.422212-1-mjguzik@gmail.com Signed-off-by: Eric Biggers --- fs/crypto/hooks.c | 32 ++++++++++++++++++++++++++------ 1 file changed, 26 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/crypto/hooks.c b/fs/crypto/hooks.c index 104771c3d3f6..d8d5049b8fe1 100644 --- a/fs/crypto/hooks.c +++ b/fs/crypto/hooks.c @@ -30,21 +30,41 @@ int fscrypt_file_open(struct inode *inode, struct file *filp) { int err; - struct dentry *dir; + struct dentry *dentry, *dentry_parent; + struct inode *inode_parent; err = fscrypt_require_key(inode); if (err) return err; - dir = dget_parent(file_dentry(filp)); - if (IS_ENCRYPTED(d_inode(dir)) && - !fscrypt_has_permitted_context(d_inode(dir), inode)) { + dentry = file_dentry(filp); + + /* + * Getting a reference to the parent dentry is needed for the actual + * encryption policy comparison, but it's expensive on multi-core + * systems. Since this function runs on unencrypted files too, start + * with a lightweight RCU-mode check for the parent directory being + * unencrypted (in which case it's fine for the child to be either + * unencrypted, or encrypted with any policy). Only continue on to the + * full policy check if the parent directory is actually encrypted. + */ + rcu_read_lock(); + dentry_parent = READ_ONCE(dentry->d_parent); + inode_parent = d_inode_rcu(dentry_parent); + if (inode_parent != NULL && !IS_ENCRYPTED(inode_parent)) { + rcu_read_unlock(); + return 0; + } + rcu_read_unlock(); + + dentry_parent = dget_parent(dentry); + if (!fscrypt_has_permitted_context(d_inode(dentry_parent), inode)) { fscrypt_warn(inode, "Inconsistent encryption context (parent directory: %lu)", - d_inode(dir)->i_ino); + d_inode(dentry_parent)->i_ino); err = -EPERM; } - dput(dir); + dput(dentry_parent); return err; } EXPORT_SYMBOL_GPL(fscrypt_file_open); -- cgit v1.2.3-59-g8ed1b From 5ad1f33c29c311787943a92c56903da74f19fd5d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 10:17:12 -0400 Subject: bcachefs: Fix sb_clean_validate endianness conversion Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-clean.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index 194e55b11137..d27804d4987a 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -283,7 +283,7 @@ static int bch2_sb_clean_validate(struct bch_sb *sb, entry = vstruct_next(entry)) { if ((void *) vstruct_next(entry) > vstruct_end(&clean->field)) { prt_str(err, "entry type "); - bch2_prt_jset_entry_type(err, le16_to_cpu(entry->type)); + bch2_prt_jset_entry_type(err, entry->type); prt_str(err, " overruns end of section"); return -BCH_ERR_invalid_sb_clean; } -- cgit v1.2.3-59-g8ed1b From 5dfd3746b6c486db18bc75de89c7abce41c7826c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 00:29:24 -0400 Subject: bcachefs: Fix needs_whiteout BUG_ON() in bkey_sort() Btree nodes are log structured; thus, we need to emit whiteouts when we're deleting a key that's been written out to disk. k->needs_whiteout tracks whether a key will need a whiteout when it's deleted, and this requires some careful handling; e.g. the key we're deleting may not have been written out to disk, but it may have overwritten a key that was - thus we need to carry this flag around on overwrites. Invariants: There may be multiple key for the same position in a given node (because of overwrites), but only one of them will be a live (non deleted) key, and only one key for a given position will have the needs_whiteout flag set. Additionally, we don't want to carry around whiteouts that need to be written in the main searchable part of a btree node - btree_iter_peek() will have to skip past them, and this can lead to an O(n^2) issues when doing sequential deletions (e.g. inode rm/truncate). So there's a separate region in the btree node buffer for unwritten whiteouts; these are merge sorted with the rest of the keys we're writing in the btree node write path. The unwritten whiteouts was a later optimization that bch2_sort_keys() didn't take into account; the unwritten whiteouts area means that we never have deleted keys with needs_whiteout set in the main searchable part of a btree node. That means we can simplify and optimize some sort paths, and eliminate an assertion that syzbot found: - Unless we're in the btree node write path, it's always ok to drop whiteouts when sorting - When sorting for a btree node write, we drop the whiteout if it's not from the unwritten whiteouts area, or if it's overwritten by a real key at the same position. This completely eliminates some tricky logic for propagating the needs_whiteout flag: syzbot was able to hit the assertion that checked that there shouldn't be more than one key at the same pos with needs_whiteout set, likely due to a combination of flipping on needs_whiteout on all written keys (they need whiteouts if overwritten), combined with not always dropping unneeded whiteouts, and the tricky logic in the sort path for preserving needs_whiteout that wasn't really needed. Signed-off-by: Kent Overstreet --- fs/bcachefs/bkey_sort.c | 79 ++++++++++++++++++++++++++++--------------------- fs/bcachefs/bkey_sort.h | 4 +-- fs/bcachefs/btree_io.c | 18 +++++------ 3 files changed, 56 insertions(+), 45 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bkey_sort.c b/fs/bcachefs/bkey_sort.c index bcca9e76a0b4..4536eb50fc40 100644 --- a/fs/bcachefs/bkey_sort.c +++ b/fs/bcachefs/bkey_sort.c @@ -6,9 +6,9 @@ #include "bset.h" #include "extents.h" -typedef int (*sort_cmp_fn)(struct btree *, - struct bkey_packed *, - struct bkey_packed *); +typedef int (*sort_cmp_fn)(const struct btree *, + const struct bkey_packed *, + const struct bkey_packed *); static inline bool sort_iter_end(struct sort_iter *iter) { @@ -70,9 +70,9 @@ static inline struct bkey_packed *sort_iter_next(struct sort_iter *iter, /* * If keys compare equal, compare by pointer order: */ -static inline int key_sort_fix_overlapping_cmp(struct btree *b, - struct bkey_packed *l, - struct bkey_packed *r) +static inline int key_sort_fix_overlapping_cmp(const struct btree *b, + const struct bkey_packed *l, + const struct bkey_packed *r) { return bch2_bkey_cmp_packed(b, l, r) ?: cmp_int((unsigned long) l, (unsigned long) r); @@ -154,46 +154,59 @@ bch2_sort_repack(struct bset *dst, struct btree *src, return nr; } -static inline int sort_keys_cmp(struct btree *b, - struct bkey_packed *l, - struct bkey_packed *r) +static inline int keep_unwritten_whiteouts_cmp(const struct btree *b, + const struct bkey_packed *l, + const struct bkey_packed *r) { return bch2_bkey_cmp_packed_inlined(b, l, r) ?: (int) bkey_deleted(r) - (int) bkey_deleted(l) ?: - (int) l->needs_whiteout - (int) r->needs_whiteout; + (long) l - (long) r; } -unsigned bch2_sort_keys(struct bkey_packed *dst, - struct sort_iter *iter, - bool filter_whiteouts) +#include "btree_update_interior.h" + +/* + * For sorting in the btree node write path: whiteouts not in the unwritten + * whiteouts area are dropped, whiteouts in the unwritten whiteouts area are + * dropped if overwritten by real keys: + */ +unsigned bch2_sort_keys_keep_unwritten_whiteouts(struct bkey_packed *dst, struct sort_iter *iter) { - const struct bkey_format *f = &iter->b->format; struct bkey_packed *in, *next, *out = dst; - sort_iter_sort(iter, sort_keys_cmp); + sort_iter_sort(iter, keep_unwritten_whiteouts_cmp); - while ((in = sort_iter_next(iter, sort_keys_cmp))) { - bool needs_whiteout = false; + while ((in = sort_iter_next(iter, keep_unwritten_whiteouts_cmp))) { + if (bkey_deleted(in) && in < unwritten_whiteouts_start(iter->b)) + continue; - if (bkey_deleted(in) && - (filter_whiteouts || !in->needs_whiteout)) + if ((next = sort_iter_peek(iter)) && + !bch2_bkey_cmp_packed_inlined(iter->b, in, next)) continue; - while ((next = sort_iter_peek(iter)) && - !bch2_bkey_cmp_packed_inlined(iter->b, in, next)) { - BUG_ON(in->needs_whiteout && - next->needs_whiteout); - needs_whiteout |= in->needs_whiteout; - in = sort_iter_next(iter, sort_keys_cmp); - } + bkey_p_copy(out, in); + out = bkey_p_next(out); + } - if (bkey_deleted(in)) { - memcpy_u64s_small(out, in, bkeyp_key_u64s(f, in)); - set_bkeyp_val_u64s(f, out, 0); - } else { - bkey_p_copy(out, in); - } - out->needs_whiteout |= needs_whiteout; + return (u64 *) out - (u64 *) dst; +} + +/* + * Main sort routine for compacting a btree node in memory: we always drop + * whiteouts because any whiteouts that need to be written are in the unwritten + * whiteouts area: + */ +unsigned bch2_sort_keys(struct bkey_packed *dst, struct sort_iter *iter) +{ + struct bkey_packed *in, *out = dst; + + sort_iter_sort(iter, bch2_bkey_cmp_packed_inlined); + + while ((in = sort_iter_next(iter, bch2_bkey_cmp_packed_inlined))) { + if (bkey_deleted(in)) + continue; + + bkey_p_copy(out, in); out = bkey_p_next(out); } diff --git a/fs/bcachefs/bkey_sort.h b/fs/bcachefs/bkey_sort.h index 7c0f0b160f18..9be969d46890 100644 --- a/fs/bcachefs/bkey_sort.h +++ b/fs/bcachefs/bkey_sort.h @@ -48,7 +48,7 @@ bch2_sort_repack(struct bset *, struct btree *, struct btree_node_iter *, struct bkey_format *, bool); -unsigned bch2_sort_keys(struct bkey_packed *, - struct sort_iter *, bool); +unsigned bch2_sort_keys_keep_unwritten_whiteouts(struct bkey_packed *, struct sort_iter *); +unsigned bch2_sort_keys(struct bkey_packed *, struct sort_iter *); #endif /* _BCACHEFS_BKEY_SORT_H */ diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index debb0edc3455..94753c5b0e89 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -288,8 +288,7 @@ bool bch2_compact_whiteouts(struct bch_fs *c, struct btree *b, static void btree_node_sort(struct bch_fs *c, struct btree *b, unsigned start_idx, - unsigned end_idx, - bool filter_whiteouts) + unsigned end_idx) { struct btree_node *out; struct sort_iter_stack sort_iter; @@ -320,7 +319,7 @@ static void btree_node_sort(struct bch_fs *c, struct btree *b, start_time = local_clock(); - u64s = bch2_sort_keys(out->keys.start, &sort_iter.iter, filter_whiteouts); + u64s = bch2_sort_keys(out->keys.start, &sort_iter.iter); out->keys.u64s = cpu_to_le16(u64s); @@ -426,13 +425,12 @@ static bool btree_node_compact(struct bch_fs *c, struct btree *b) break; if (b->nsets - unwritten_idx > 1) { - btree_node_sort(c, b, unwritten_idx, - b->nsets, false); + btree_node_sort(c, b, unwritten_idx, b->nsets); ret = true; } if (unwritten_idx > 1) { - btree_node_sort(c, b, 0, unwritten_idx, false); + btree_node_sort(c, b, 0, unwritten_idx); ret = true; } @@ -2095,11 +2093,11 @@ do_write: unwritten_whiteouts_end(b)); SET_BSET_SEPARATE_WHITEOUTS(i, false); - b->whiteout_u64s = 0; - - u64s = bch2_sort_keys(i->start, &sort_iter.iter, false); + u64s = bch2_sort_keys_keep_unwritten_whiteouts(i->start, &sort_iter.iter); le16_add_cpu(&i->u64s, u64s); + b->whiteout_u64s = 0; + BUG_ON(!b->written && i->u64s != b->data->keys.u64s); set_needs_whiteout(i, false); @@ -2249,7 +2247,7 @@ bool bch2_btree_post_write_cleanup(struct bch_fs *c, struct btree *b) * single bset: */ if (b->nsets > 1) { - btree_node_sort(c, b, 0, b->nsets, true); + btree_node_sort(c, b, 0, b->nsets); invalidated_iter = true; } else { invalidated_iter = bch2_drop_whiteouts(b, COMPACT_ALL); -- cgit v1.2.3-59-g8ed1b From 61692c7812ab2aca17a3751f6e7798acbdae4b6b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 10:58:26 -0400 Subject: bcachefs: bch2_bkey_format_field_overflows() Fix another shift-by-64 by factoring out a common helper for bch2_bkey_format_invalid() and bformat_needs_redo() (where it was already fixed). Reported-by: syzbot+9833a1d29d4a44361e2c@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/bkey.c | 13 +++++-------- fs/bcachefs/bkey.h | 23 +++++++++++++++++++++++ fs/bcachefs/move.c | 20 ++------------------ 3 files changed, 30 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bkey.c b/fs/bcachefs/bkey.c index 76e79a15ba08..952299a2e416 100644 --- a/fs/bcachefs/bkey.c +++ b/fs/bcachefs/bkey.c @@ -656,20 +656,17 @@ int bch2_bkey_format_invalid(struct bch_fs *c, * unpacked format: */ for (i = 0; i < f->nr_fields; i++) { - if (!c || c->sb.version_min >= bcachefs_metadata_version_snapshot) { + if ((!c || c->sb.version_min >= bcachefs_metadata_version_snapshot) && + bch2_bkey_format_field_overflows(f, i)) { unsigned unpacked_bits = bch2_bkey_format_current.bits_per_field[i]; u64 unpacked_max = ~((~0ULL << 1) << (unpacked_bits - 1)); u64 packed_max = f->bits_per_field[i] ? ~((~0ULL << 1) << (f->bits_per_field[i] - 1)) : 0; - u64 field_offset = le64_to_cpu(f->field_offset[i]); - if (packed_max + field_offset < packed_max || - packed_max + field_offset > unpacked_max) { - prt_printf(err, "field %u too large: %llu + %llu > %llu", - i, packed_max, field_offset, unpacked_max); - return -BCH_ERR_invalid; - } + prt_printf(err, "field %u too large: %llu + %llu > %llu", + i, packed_max, le64_to_cpu(f->field_offset[i]), unpacked_max); + return -BCH_ERR_invalid; } bits += f->bits_per_field[i]; diff --git a/fs/bcachefs/bkey.h b/fs/bcachefs/bkey.h index 3a45d128f608..1bb0bd4371b0 100644 --- a/fs/bcachefs/bkey.h +++ b/fs/bcachefs/bkey.h @@ -574,6 +574,29 @@ static inline void bch2_bkey_format_add_key(struct bkey_format_state *s, const s void bch2_bkey_format_add_pos(struct bkey_format_state *, struct bpos); struct bkey_format bch2_bkey_format_done(struct bkey_format_state *); + +static inline bool bch2_bkey_format_field_overflows(struct bkey_format *f, unsigned i) +{ + unsigned f_bits = f->bits_per_field[i]; + unsigned unpacked_bits = bch2_bkey_format_current.bits_per_field[i]; + u64 unpacked_mask = ~((~0ULL << 1) << (unpacked_bits - 1)); + u64 field_offset = le64_to_cpu(f->field_offset[i]); + + if (f_bits > unpacked_bits) + return true; + + if ((f_bits == unpacked_bits) && field_offset) + return true; + + u64 f_mask = f_bits + ? ~((~0ULL << (f_bits - 1)) << 1) + : 0; + + if (((field_offset + f_mask) & unpacked_mask) < field_offset) + return true; + return false; +} + int bch2_bkey_format_invalid(struct bch_fs *, struct bkey_format *, enum bkey_invalid_flags, struct printbuf *); void bch2_bkey_format_to_text(struct printbuf *, const struct bkey_format *); diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index 4d94b7742dbb..45984a688e5b 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -975,26 +975,10 @@ static bool migrate_btree_pred(struct bch_fs *c, void *arg, */ static bool bformat_needs_redo(struct bkey_format *f) { - for (unsigned i = 0; i < f->nr_fields; i++) { - unsigned f_bits = f->bits_per_field[i]; - unsigned unpacked_bits = bch2_bkey_format_current.bits_per_field[i]; - u64 unpacked_mask = ~((~0ULL << 1) << (unpacked_bits - 1)); - u64 field_offset = le64_to_cpu(f->field_offset[i]); - - if (f_bits > unpacked_bits) - return true; - - if ((f_bits == unpacked_bits) && field_offset) + for (unsigned i = 0; i < f->nr_fields; i++) + if (bch2_bkey_format_field_overflows(f, i)) return true; - u64 f_mask = f_bits - ? ~((~0ULL << (f_bits - 1)) << 1) - : 0; - - if (((field_offset + f_mask) & unpacked_mask) < field_offset) - return true; - } - return false; } -- cgit v1.2.3-59-g8ed1b From 74768337de7f95666fb68a70c73eb1728126cff7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 02:50:42 -0400 Subject: bcachefs: Fix xattr_to_text() unsafety Signed-off-by: Kent Overstreet --- fs/bcachefs/xattr.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/xattr.c b/fs/bcachefs/xattr.c index 754f17bba68e..4b3a2df2da32 100644 --- a/fs/bcachefs/xattr.c +++ b/fs/bcachefs/xattr.c @@ -118,11 +118,17 @@ void bch2_xattr_to_text(struct printbuf *out, struct bch_fs *c, else prt_printf(out, "(unknown type %u)", xattr.v->x_type); + unsigned name_len = xattr.v->x_name_len; + unsigned val_len = le16_to_cpu(xattr.v->x_val_len); + unsigned max_name_val_bytes = bkey_val_bytes(xattr.k) - + offsetof(struct bch_xattr, x_name); + + val_len = min_t(int, val_len, max_name_val_bytes - name_len); + name_len = min(name_len, max_name_val_bytes); + prt_printf(out, "%.*s:%.*s", - xattr.v->x_name_len, - xattr.v->x_name, - le16_to_cpu(xattr.v->x_val_len), - (char *) xattr_val(xattr.v)); + name_len, xattr.v->x_name, + val_len, (char *) xattr_val(xattr.v)); if (xattr.v->x_type == KEY_TYPE_XATTR_INDEX_POSIX_ACL_ACCESS || xattr.v->x_type == KEY_TYPE_XATTR_INDEX_POSIX_ACL_DEFAULT) { -- cgit v1.2.3-59-g8ed1b From 6d828691858396e95e9856fad6a08c57f06b6e2d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 14:17:54 -0400 Subject: bcachefs: Better write_super() error messages When a superblock write is silently dropped or it's been modified by another process we need to know which device it was. Signed-off-by: Kent Overstreet --- fs/bcachefs/super-io.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index bfdb15e7d778..1cadc99898f5 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -1020,26 +1020,35 @@ int bch2_write_super(struct bch_fs *c) continue; if (le64_to_cpu(ca->sb_read_scratch->seq) < ca->disk_sb.seq) { - bch2_fs_fatal_error(c, + struct printbuf buf = PRINTBUF; + prt_char(&buf, ' '); + prt_bdevname(&buf, ca->disk_sb.bdev); + prt_printf(&buf, ": Superblock write was silently dropped! (seq %llu expected %llu)", le64_to_cpu(ca->sb_read_scratch->seq), ca->disk_sb.seq); - percpu_ref_put(&ca->io_ref); + bch2_fs_fatal_error(c, "%s", buf.buf); + printbuf_exit(&buf); ret = -BCH_ERR_erofs_sb_err; - goto out; } if (le64_to_cpu(ca->sb_read_scratch->seq) > ca->disk_sb.seq) { - bch2_fs_fatal_error(c, + struct printbuf buf = PRINTBUF; + prt_char(&buf, ' '); + prt_bdevname(&buf, ca->disk_sb.bdev); + prt_printf(&buf, ": Superblock modified by another process (seq %llu expected %llu)", le64_to_cpu(ca->sb_read_scratch->seq), ca->disk_sb.seq); - percpu_ref_put(&ca->io_ref); + bch2_fs_fatal_error(c, "%s", buf.buf); + printbuf_exit(&buf); ret = -BCH_ERR_erofs_sb_err; - goto out; } } + if (ret) + goto out; + do { wrote = false; darray_for_each(online_devices, cap) { -- cgit v1.2.3-59-g8ed1b From 62606398d5d718186d914d8f30be34177dc3f26d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 28 Apr 2024 21:24:45 -0400 Subject: bcachefs: Run upgrade/downgrade even in -o nochanges mode We need to be able to test these paths in dry run mode. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 84 ++++++++++++++++++++++++-------------------------- 1 file changed, 41 insertions(+), 43 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 8091d0686029..08c0422a722d 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -597,56 +597,54 @@ int bch2_fs_recovery(struct bch_fs *c) if (c->opts.norecovery) c->opts.recovery_pass_last = BCH_RECOVERY_PASS_journal_replay - 1; - if (!c->opts.nochanges) { - mutex_lock(&c->sb_lock); - struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); - bool write_sb = false; - - if (BCH_SB_HAS_TOPOLOGY_ERRORS(c->disk_sb.sb)) { - ext->recovery_passes_required[0] |= - cpu_to_le64(bch2_recovery_passes_to_stable(BIT_ULL(BCH_RECOVERY_PASS_check_topology))); - write_sb = true; - } + mutex_lock(&c->sb_lock); + struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); + bool write_sb = false; - u64 sb_passes = bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); - if (sb_passes) { - struct printbuf buf = PRINTBUF; - prt_str(&buf, "superblock requires following recovery passes to be run:\n "); - prt_bitflags(&buf, bch2_recovery_passes, sb_passes); - bch_info(c, "%s", buf.buf); - printbuf_exit(&buf); - } + if (BCH_SB_HAS_TOPOLOGY_ERRORS(c->disk_sb.sb)) { + ext->recovery_passes_required[0] |= + cpu_to_le64(bch2_recovery_passes_to_stable(BIT_ULL(BCH_RECOVERY_PASS_check_topology))); + write_sb = true; + } - if (bch2_check_version_downgrade(c)) { - struct printbuf buf = PRINTBUF; + u64 sb_passes = bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); + if (sb_passes) { + struct printbuf buf = PRINTBUF; + prt_str(&buf, "superblock requires following recovery passes to be run:\n "); + prt_bitflags(&buf, bch2_recovery_passes, sb_passes); + bch_info(c, "%s", buf.buf); + printbuf_exit(&buf); + } - prt_str(&buf, "Version downgrade required:"); + if (bch2_check_version_downgrade(c)) { + struct printbuf buf = PRINTBUF; - __le64 passes = ext->recovery_passes_required[0]; - bch2_sb_set_downgrade(c, - BCH_VERSION_MINOR(bcachefs_metadata_version_current), - BCH_VERSION_MINOR(c->sb.version)); - passes = ext->recovery_passes_required[0] & ~passes; - if (passes) { - prt_str(&buf, "\n running recovery passes: "); - prt_bitflags(&buf, bch2_recovery_passes, - bch2_recovery_passes_from_stable(le64_to_cpu(passes))); - } + prt_str(&buf, "Version downgrade required:"); - bch_info(c, "%s", buf.buf); - printbuf_exit(&buf); - write_sb = true; + __le64 passes = ext->recovery_passes_required[0]; + bch2_sb_set_downgrade(c, + BCH_VERSION_MINOR(bcachefs_metadata_version_current), + BCH_VERSION_MINOR(c->sb.version)); + passes = ext->recovery_passes_required[0] & ~passes; + if (passes) { + prt_str(&buf, "\n running recovery passes: "); + prt_bitflags(&buf, bch2_recovery_passes, + bch2_recovery_passes_from_stable(le64_to_cpu(passes))); } - if (check_version_upgrade(c)) - write_sb = true; + bch_info(c, "%s", buf.buf); + printbuf_exit(&buf); + write_sb = true; + } - if (write_sb) - bch2_write_super(c); + if (check_version_upgrade(c)) + write_sb = true; - c->recovery_passes_explicit |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); - mutex_unlock(&c->sb_lock); - } + if (write_sb) + bch2_write_super(c); + + c->recovery_passes_explicit |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); + mutex_unlock(&c->sb_lock); if (c->opts.fsck && IS_ENABLED(CONFIG_BCACHEFS_DEBUG)) c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); @@ -832,8 +830,8 @@ use_clean: } mutex_lock(&c->sb_lock); - struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); - bool write_sb = false; + ext = bch2_sb_field_get(c->disk_sb.sb, ext); + write_sb = false; if (BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb) != le16_to_cpu(c->disk_sb.sb->version)) { SET_BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb, le16_to_cpu(c->disk_sb.sb->version)); -- cgit v1.2.3-59-g8ed1b From acce32a51e6c32e0ad8f921f9af2226e6750bb51 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 14:32:45 -0400 Subject: bcachefs: printbuf improvements - fix assorted (harmless) off-by-one errors - we were inconsistent on whether out->pos stays <= out->size on overflow; now it does, and printbuf.overflow exists to indicate if a printbuf has overflowed - factor out printbuf_advance_pos() - printbuf_nul_terminate_reserved(); use this to reduce the number of printbuf_make_room() calls Signed-off-by: Kent Overstreet --- fs/bcachefs/printbuf.c | 65 ++++++++++++++++++++++++-------------------------- fs/bcachefs/printbuf.h | 53 +++++++++++++++++++--------------------- 2 files changed, 55 insertions(+), 63 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/printbuf.c b/fs/bcachefs/printbuf.c index b27d22925929..46ce63312a3f 100644 --- a/fs/bcachefs/printbuf.c +++ b/fs/bcachefs/printbuf.c @@ -17,28 +17,28 @@ static inline unsigned printbuf_linelen(struct printbuf *buf) int bch2_printbuf_make_room(struct printbuf *out, unsigned extra) { - unsigned new_size; - char *buf; - - if (!out->heap_allocated) - return 0; - /* Reserved space for terminating nul: */ extra += 1; - if (out->pos + extra < out->size) + if (out->pos + extra <= out->size) return 0; - new_size = roundup_pow_of_two(out->size + extra); + if (!out->heap_allocated) { + out->overflow = true; + return 0; + } + + unsigned new_size = roundup_pow_of_two(out->size + extra); /* * Note: output buffer must be freeable with kfree(), it's not required * that the user use printbuf_exit(). */ - buf = krealloc(out->buf, new_size, !out->atomic ? GFP_KERNEL : GFP_NOWAIT); + char *buf = krealloc(out->buf, new_size, !out->atomic ? GFP_KERNEL : GFP_NOWAIT); if (!buf) { out->allocation_failure = true; + out->overflow = true; return -ENOMEM; } @@ -47,6 +47,11 @@ int bch2_printbuf_make_room(struct printbuf *out, unsigned extra) return 0; } +static void printbuf_advance_pos(struct printbuf *out, unsigned len) +{ + out->pos += min(len, printbuf_remaining(out)); +} + void bch2_prt_vprintf(struct printbuf *out, const char *fmt, va_list args) { int len; @@ -55,14 +60,12 @@ void bch2_prt_vprintf(struct printbuf *out, const char *fmt, va_list args) va_list args2; va_copy(args2, args); - len = vsnprintf(out->buf + out->pos, printbuf_remaining(out), fmt, args2); + len = vsnprintf(out->buf + out->pos, printbuf_remaining_size(out), fmt, args2); va_end(args2); - } while (len + 1 >= printbuf_remaining(out) && - !bch2_printbuf_make_room(out, len + 1)); + } while (len > printbuf_remaining(out) && + !bch2_printbuf_make_room(out, len)); - len = min_t(size_t, len, - printbuf_remaining(out) ? printbuf_remaining(out) - 1 : 0); - out->pos += len; + printbuf_advance_pos(out, len); } void bch2_prt_printf(struct printbuf *out, const char *fmt, ...) @@ -72,14 +75,12 @@ void bch2_prt_printf(struct printbuf *out, const char *fmt, ...) do { va_start(args, fmt); - len = vsnprintf(out->buf + out->pos, printbuf_remaining(out), fmt, args); + len = vsnprintf(out->buf + out->pos, printbuf_remaining_size(out), fmt, args); va_end(args); - } while (len + 1 >= printbuf_remaining(out) && - !bch2_printbuf_make_room(out, len + 1)); + } while (len > printbuf_remaining(out) && + !bch2_printbuf_make_room(out, len)); - len = min_t(size_t, len, - printbuf_remaining(out) ? printbuf_remaining(out) - 1 : 0); - out->pos += len; + printbuf_advance_pos(out, len); } /** @@ -194,18 +195,15 @@ void bch2_printbuf_indent_sub(struct printbuf *buf, unsigned spaces) void bch2_prt_newline(struct printbuf *buf) { - unsigned i; - bch2_printbuf_make_room(buf, 1 + buf->indent); - __prt_char(buf, '\n'); + __prt_char_reserved(buf, '\n'); buf->last_newline = buf->pos; - for (i = 0; i < buf->indent; i++) - __prt_char(buf, ' '); + __prt_chars_reserved(buf, ' ', buf->indent); - printbuf_nul_terminate(buf); + printbuf_nul_terminate_reserved(buf); buf->last_field = buf->pos; buf->cur_tabstop = 0; @@ -262,7 +260,7 @@ static void __prt_tab_rjust(struct printbuf *buf) memset(buf->buf + buf->last_field, ' ', min((unsigned) pad, buf->size - buf->last_field)); - buf->pos += pad; + printbuf_advance_pos(buf, pad); printbuf_nul_terminate(buf); } @@ -348,9 +346,10 @@ void bch2_prt_bytes_indented(struct printbuf *out, const char *str, unsigned cou void bch2_prt_human_readable_u64(struct printbuf *out, u64 v) { bch2_printbuf_make_room(out, 10); - out->pos += string_get_size(v, 1, !out->si_units, - out->buf + out->pos, - printbuf_remaining_size(out)); + unsigned len = string_get_size(v, 1, !out->si_units, + out->buf + out->pos, + printbuf_remaining_size(out)); + printbuf_advance_pos(out, len); } /** @@ -402,9 +401,7 @@ void bch2_prt_string_option(struct printbuf *out, const char * const list[], size_t selected) { - size_t i; - - for (i = 0; list[i]; i++) + for (size_t i = 0; list[i]; i++) bch2_prt_printf(out, i == selected ? "[%s] " : "%s ", list[i]); } diff --git a/fs/bcachefs/printbuf.h b/fs/bcachefs/printbuf.h index 9a4a56c40937..9ecc56bc9635 100644 --- a/fs/bcachefs/printbuf.h +++ b/fs/bcachefs/printbuf.h @@ -86,6 +86,7 @@ struct printbuf { u8 atomic; bool allocation_failure:1; bool heap_allocated:1; + bool overflow:1; enum printbuf_si si_units:1; bool human_readable_units:1; bool has_indent_or_tabstops:1; @@ -142,7 +143,9 @@ void bch2_prt_bitflags_vector(struct printbuf *, const char * const[], */ static inline unsigned printbuf_remaining_size(struct printbuf *out) { - return out->pos < out->size ? out->size - out->pos : 0; + if (WARN_ON(out->size && out->pos >= out->size)) + out->pos = out->size - 1; + return out->size - out->pos; } /* @@ -151,7 +154,7 @@ static inline unsigned printbuf_remaining_size(struct printbuf *out) */ static inline unsigned printbuf_remaining(struct printbuf *out) { - return out->pos < out->size ? out->size - out->pos - 1 : 0; + return out->size ? printbuf_remaining_size(out) - 1 : 0; } static inline unsigned printbuf_written(struct printbuf *out) @@ -159,30 +162,25 @@ static inline unsigned printbuf_written(struct printbuf *out) return out->size ? min(out->pos, out->size - 1) : 0; } -/* - * Returns true if output was truncated: - */ -static inline bool printbuf_overflowed(struct printbuf *out) +static inline void printbuf_nul_terminate_reserved(struct printbuf *out) { - return out->pos >= out->size; + if (WARN_ON(out->size && out->pos >= out->size)) + out->pos = out->size - 1; + if (out->size) + out->buf[out->pos] = 0; } static inline void printbuf_nul_terminate(struct printbuf *out) { bch2_printbuf_make_room(out, 1); - - if (out->pos < out->size) - out->buf[out->pos] = 0; - else if (out->size) - out->buf[out->size - 1] = 0; + printbuf_nul_terminate_reserved(out); } /* Doesn't call bch2_printbuf_make_room(), doesn't nul terminate: */ static inline void __prt_char_reserved(struct printbuf *out, char c) { if (printbuf_remaining(out)) - out->buf[out->pos] = c; - out->pos++; + out->buf[out->pos++] = c; } /* Doesn't nul terminate: */ @@ -194,37 +192,34 @@ static inline void __prt_char(struct printbuf *out, char c) static inline void prt_char(struct printbuf *out, char c) { - __prt_char(out, c); - printbuf_nul_terminate(out); + bch2_printbuf_make_room(out, 2); + __prt_char_reserved(out, c); + printbuf_nul_terminate_reserved(out); } static inline void __prt_chars_reserved(struct printbuf *out, char c, unsigned n) { - unsigned i, can_print = min(n, printbuf_remaining(out)); + unsigned can_print = min(n, printbuf_remaining(out)); - for (i = 0; i < can_print; i++) + for (unsigned i = 0; i < can_print; i++) out->buf[out->pos++] = c; - out->pos += n - can_print; } static inline void prt_chars(struct printbuf *out, char c, unsigned n) { bch2_printbuf_make_room(out, n); __prt_chars_reserved(out, c, n); - printbuf_nul_terminate(out); + printbuf_nul_terminate_reserved(out); } static inline void prt_bytes(struct printbuf *out, const void *b, unsigned n) { - unsigned i, can_print; - bch2_printbuf_make_room(out, n); - can_print = min(n, printbuf_remaining(out)); + unsigned can_print = min(n, printbuf_remaining(out)); - for (i = 0; i < can_print; i++) + for (unsigned i = 0; i < can_print; i++) out->buf[out->pos++] = ((char *) b)[i]; - out->pos += n - can_print; printbuf_nul_terminate(out); } @@ -241,18 +236,18 @@ static inline void prt_str_indented(struct printbuf *out, const char *str) static inline void prt_hex_byte(struct printbuf *out, u8 byte) { - bch2_printbuf_make_room(out, 2); + bch2_printbuf_make_room(out, 3); __prt_char_reserved(out, hex_asc_hi(byte)); __prt_char_reserved(out, hex_asc_lo(byte)); - printbuf_nul_terminate(out); + printbuf_nul_terminate_reserved(out); } static inline void prt_hex_byte_upper(struct printbuf *out, u8 byte) { - bch2_printbuf_make_room(out, 2); + bch2_printbuf_make_room(out, 3); __prt_char_reserved(out, hex_asc_upper_hi(byte)); __prt_char_reserved(out, hex_asc_upper_lo(byte)); - printbuf_nul_terminate(out); + printbuf_nul_terminate_reserved(out); } /** -- cgit v1.2.3-59-g8ed1b From 2dcb605e86891f17bea3d77ba05f5d12cdf38573 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 14:32:45 -0400 Subject: bcachefs: printbufs: prt_printf() now handles \t\r\n Signed-off-by: Kent Overstreet --- fs/bcachefs/printbuf.c | 169 +++++++++++++++++++++++++++++++------------------ 1 file changed, 106 insertions(+), 63 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/printbuf.c b/fs/bcachefs/printbuf.c index 46ce63312a3f..8b0369185f5c 100644 --- a/fs/bcachefs/printbuf.c +++ b/fs/bcachefs/printbuf.c @@ -10,9 +10,24 @@ #include "printbuf.h" +static inline unsigned __printbuf_linelen(struct printbuf *buf, unsigned pos) +{ + return pos - buf->last_newline; +} + static inline unsigned printbuf_linelen(struct printbuf *buf) { - return buf->pos - buf->last_newline; + return __printbuf_linelen(buf, buf->pos); +} + +/* + * Returns spaces from start of line, if set, or 0 if unset: + */ +static inline unsigned cur_tabstop(struct printbuf *buf) +{ + return buf->cur_tabstop < buf->nr_tabstops + ? buf->_tabstops[buf->cur_tabstop] + : 0; } int bch2_printbuf_make_room(struct printbuf *out, unsigned extra) @@ -52,6 +67,87 @@ static void printbuf_advance_pos(struct printbuf *out, unsigned len) out->pos += min(len, printbuf_remaining(out)); } +static void printbuf_insert_spaces(struct printbuf *out, unsigned pos, unsigned nr) +{ + unsigned move = out->pos - pos; + + bch2_printbuf_make_room(out, nr); + + if (pos + nr < out->size) + memmove(out->buf + pos + nr, + out->buf + pos, + min(move, out->size - 1 - pos - nr)); + + if (pos < out->size) + memset(out->buf + pos, ' ', min(nr, out->size - pos)); + + printbuf_advance_pos(out, nr); + printbuf_nul_terminate_reserved(out); +} + +static void __printbuf_do_indent(struct printbuf *out, unsigned pos) +{ + while (true) { + int pad; + unsigned len = out->pos - pos; + char *p = out->buf + pos; + char *n = memscan(p, '\n', len); + if (cur_tabstop(out)) { + n = min(n, (char *) memscan(p, '\r', len)); + n = min(n, (char *) memscan(p, '\t', len)); + } + + pos = n - out->buf; + if (pos == out->pos) + break; + + switch (*n) { + case '\n': + pos++; + out->last_newline = pos; + + printbuf_insert_spaces(out, pos, out->indent); + + pos = min(pos + out->indent, out->pos); + out->last_field = pos; + out->cur_tabstop = 0; + break; + case '\r': + memmove(n, n + 1, out->pos - pos); + --out->pos; + pad = (int) cur_tabstop(out) - (int) __printbuf_linelen(out, pos); + if (pad > 0) { + printbuf_insert_spaces(out, out->last_field, pad); + pos += pad; + } + + out->last_field = pos; + out->cur_tabstop++; + break; + case '\t': + pad = (int) cur_tabstop(out) - (int) __printbuf_linelen(out, pos) - 1; + if (pad > 0) { + *n = ' '; + printbuf_insert_spaces(out, pos, pad - 1); + pos += pad; + } else { + memmove(n, n + 1, out->pos - pos); + --out->pos; + } + + out->last_field = pos; + out->cur_tabstop++; + break; + } + } +} + +static inline void printbuf_do_indent(struct printbuf *out, unsigned pos) +{ + if (out->has_indent_or_tabstops && !out->suppress_indent_tabstop_handling) + __printbuf_do_indent(out, pos); +} + void bch2_prt_vprintf(struct printbuf *out, const char *fmt, va_list args) { int len; @@ -65,7 +161,9 @@ void bch2_prt_vprintf(struct printbuf *out, const char *fmt, va_list args) } while (len > printbuf_remaining(out) && !bch2_printbuf_make_room(out, len)); + unsigned indent_pos = out->pos; printbuf_advance_pos(out, len); + printbuf_do_indent(out, indent_pos); } void bch2_prt_printf(struct printbuf *out, const char *fmt, ...) @@ -80,7 +178,9 @@ void bch2_prt_printf(struct printbuf *out, const char *fmt, ...) } while (len > printbuf_remaining(out) && !bch2_printbuf_make_room(out, len)); + unsigned indent_pos = out->pos; printbuf_advance_pos(out, len); + printbuf_do_indent(out, indent_pos); } /** @@ -209,16 +309,6 @@ void bch2_prt_newline(struct printbuf *buf) buf->cur_tabstop = 0; } -/* - * Returns spaces from start of line, if set, or 0 if unset: - */ -static inline unsigned cur_tabstop(struct printbuf *buf) -{ - return buf->cur_tabstop < buf->nr_tabstops - ? buf->_tabstops[buf->cur_tabstop] - : 0; -} - static void __prt_tab(struct printbuf *out) { int spaces = max_t(int, 0, cur_tabstop(out) - printbuf_linelen(out)); @@ -245,24 +335,9 @@ void bch2_prt_tab(struct printbuf *out) static void __prt_tab_rjust(struct printbuf *buf) { - unsigned move = buf->pos - buf->last_field; int pad = (int) cur_tabstop(buf) - (int) printbuf_linelen(buf); - - if (pad > 0) { - bch2_printbuf_make_room(buf, pad); - - if (buf->last_field + pad < buf->size) - memmove(buf->buf + buf->last_field + pad, - buf->buf + buf->last_field, - min(move, buf->size - 1 - buf->last_field - pad)); - - if (buf->last_field < buf->size) - memset(buf->buf + buf->last_field, ' ', - min((unsigned) pad, buf->size - buf->last_field)); - - printbuf_advance_pos(buf, pad); - printbuf_nul_terminate(buf); - } + if (pad > 0) + printbuf_insert_spaces(buf, buf->last_field, pad); buf->last_field = buf->pos; buf->cur_tabstop++; @@ -299,41 +374,9 @@ void bch2_prt_tab_rjust(struct printbuf *buf) */ void bch2_prt_bytes_indented(struct printbuf *out, const char *str, unsigned count) { - const char *unprinted_start = str; - const char *end = str + count; - - if (!out->has_indent_or_tabstops || out->suppress_indent_tabstop_handling) { - prt_bytes(out, str, count); - return; - } - - while (str != end) { - switch (*str) { - case '\n': - prt_bytes(out, unprinted_start, str - unprinted_start); - unprinted_start = str + 1; - bch2_prt_newline(out); - break; - case '\t': - if (likely(cur_tabstop(out))) { - prt_bytes(out, unprinted_start, str - unprinted_start); - unprinted_start = str + 1; - __prt_tab(out); - } - break; - case '\r': - if (likely(cur_tabstop(out))) { - prt_bytes(out, unprinted_start, str - unprinted_start); - unprinted_start = str + 1; - __prt_tab_rjust(out); - } - break; - } - - str++; - } - - prt_bytes(out, unprinted_start, str - unprinted_start); + unsigned indent_pos = out->pos; + prt_bytes(out, str, count); + printbuf_do_indent(out, indent_pos); } /** -- cgit v1.2.3-59-g8ed1b From 7423330e30ab916d56ce9a9d49e37e40ead5b153 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 16:08:24 -0400 Subject: bcachefs: prt_printf() now respects \r\n\t Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 32 ++++------- fs/bcachefs/backpointers.c | 3 +- fs/bcachefs/btree_io.c | 15 ++---- fs/bcachefs/btree_iter.c | 18 +++---- fs/bcachefs/btree_key_cache.c | 9 ++-- fs/bcachefs/btree_locking.c | 6 +-- fs/bcachefs/btree_trans_commit.c | 6 +-- fs/bcachefs/buckets.c | 21 ++------ fs/bcachefs/checksum.c | 12 ++--- fs/bcachefs/debug.c | 64 ++++++---------------- fs/bcachefs/inode.c | 24 +++------ fs/bcachefs/io_read.c | 3 +- fs/bcachefs/io_write.c | 3 +- fs/bcachefs/journal.c | 97 ++++++++++++++------------------- fs/bcachefs/move.c | 37 +++++-------- fs/bcachefs/quota.c | 112 +++++++-------------------------------- fs/bcachefs/sb-clean.c | 6 +-- fs/bcachefs/sb-counters.c | 15 ++---- fs/bcachefs/sb-downgrade.c | 9 ++-- fs/bcachefs/sb-members.c | 96 +++++++++------------------------ fs/bcachefs/super-io.c | 70 ++++++++---------------- fs/bcachefs/sysfs.c | 96 +++++++-------------------------- fs/bcachefs/util.c | 61 ++++++--------------- 23 files changed, 226 insertions(+), 589 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 534ba2b02bd6..6a7d249f108c 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -330,27 +330,17 @@ void bch2_alloc_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c prt_printf(out, "gen %u oldest_gen %u data_type ", a->gen, a->oldest_gen); bch2_prt_data_type(out, a->data_type); prt_newline(out); - prt_printf(out, "journal_seq %llu", a->journal_seq); - prt_newline(out); - prt_printf(out, "need_discard %llu", BCH_ALLOC_V4_NEED_DISCARD(a)); - prt_newline(out); - prt_printf(out, "need_inc_gen %llu", BCH_ALLOC_V4_NEED_INC_GEN(a)); - prt_newline(out); - prt_printf(out, "dirty_sectors %u", a->dirty_sectors); - prt_newline(out); - prt_printf(out, "cached_sectors %u", a->cached_sectors); - prt_newline(out); - prt_printf(out, "stripe %u", a->stripe); - prt_newline(out); - prt_printf(out, "stripe_redundancy %u", a->stripe_redundancy); - prt_newline(out); - prt_printf(out, "io_time[READ] %llu", a->io_time[READ]); - prt_newline(out); - prt_printf(out, "io_time[WRITE] %llu", a->io_time[WRITE]); - prt_newline(out); - prt_printf(out, "fragmentation %llu", a->fragmentation_lru); - prt_newline(out); - prt_printf(out, "bp_start %llu", BCH_ALLOC_V4_BACKPOINTERS_START(a)); + prt_printf(out, "journal_seq %llu\n", a->journal_seq); + prt_printf(out, "need_discard %llu\n", BCH_ALLOC_V4_NEED_DISCARD(a)); + prt_printf(out, "need_inc_gen %llu\n", BCH_ALLOC_V4_NEED_INC_GEN(a)); + prt_printf(out, "dirty_sectors %u\n", a->dirty_sectors); + prt_printf(out, "cached_sectors %u\n", a->cached_sectors); + prt_printf(out, "stripe %u\n", a->stripe); + prt_printf(out, "stripe_redundancy %u\n", a->stripe_redundancy); + prt_printf(out, "io_time[READ] %llu\n", a->io_time[READ]); + prt_printf(out, "io_time[WRITE] %llu\n", a->io_time[WRITE]); + prt_printf(out, "fragmentation %llu\n", a->fragmentation_lru); + prt_printf(out, "bp_start %llu\n", BCH_ALLOC_V4_BACKPOINTERS_START(a)); printbuf_indent_sub(out, 2); } diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index af7a71de1bdf..fc9506c71eae 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -117,8 +117,7 @@ static noinline int backpointer_mod_err(struct btree_trans *trans, bch_err(c, "%s", buf.buf); } else if (c->curr_recovery_pass > BCH_RECOVERY_PASS_check_extents_to_backpointers) { - prt_printf(&buf, "backpointer not found when deleting"); - prt_newline(&buf); + prt_printf(&buf, "backpointer not found when deleting\n"); printbuf_indent_add(&buf, 2); prt_printf(&buf, "searching for "); diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 94753c5b0e89..fc046c9ccb45 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -522,7 +522,9 @@ static void btree_err_msg(struct printbuf *out, struct bch_fs *c, prt_printf(out, "at btree "); bch2_btree_pos_to_text(out, c, b); - prt_printf(out, "\n node offset %u/%u", + printbuf_indent_add(out, 2); + + prt_printf(out, "\nnode offset %u/%u", b->written, btree_ptr_sectors_written(&b->key)); if (i) prt_printf(out, " bset u64s %u", le16_to_cpu(i->u64s)); @@ -2344,20 +2346,13 @@ void bch2_btree_write_stats_to_text(struct printbuf *out, struct bch_fs *c) printbuf_tabstop_push(out, 20); printbuf_tabstop_push(out, 10); - prt_tab(out); - prt_str(out, "nr"); - prt_tab(out); - prt_str(out, "size"); - prt_newline(out); + prt_printf(out, "\tnr\tsize\n"); for (unsigned i = 0; i < BTREE_WRITE_TYPE_NR; i++) { u64 nr = atomic64_read(&c->btree_write_stats[i].nr); u64 bytes = atomic64_read(&c->btree_write_stats[i].bytes); - prt_printf(out, "%s:", bch2_btree_write_types[i]); - prt_tab(out); - prt_u64(out, nr); - prt_tab(out); + prt_printf(out, "%s:\t%llu\t", bch2_btree_write_types[i], nr); prt_human_readable_u64(out, nr ? div64_u64(bytes, nr) : 0); prt_newline(out); } diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 2a211a4bebd1..080cb50168b0 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1387,19 +1387,17 @@ void __noreturn bch2_trans_in_restart_error(struct btree_trans *trans) noinline __cold void bch2_trans_updates_to_text(struct printbuf *buf, struct btree_trans *trans) { - prt_printf(buf, "transaction updates for %s journal seq %llu", + prt_printf(buf, "transaction updates for %s journal seq %llu\n", trans->fn, trans->journal_res.seq); - prt_newline(buf); printbuf_indent_add(buf, 2); trans_for_each_update(trans, i) { struct bkey_s_c old = { &i->old_k, i->old_v }; - prt_printf(buf, "update: btree=%s cached=%u %pS", + prt_printf(buf, "update: btree=%s cached=%u %pS\n", bch2_btree_id_str(i->btree_id), i->cached, (void *) i->ip_allocated); - prt_newline(buf); prt_printf(buf, " old "); bch2_bkey_val_to_text(buf, trans->c, old); @@ -3166,13 +3164,11 @@ bch2_btree_bkey_cached_common_to_text(struct printbuf *out, pid = owner ? owner->pid : 0; rcu_read_unlock(); - prt_tab(out); - prt_printf(out, "%px %c l=%u %s:", b, b->cached ? 'c' : 'b', + prt_printf(out, "\t%px %c l=%u %s:", b, b->cached ? 'c' : 'b', b->level, bch2_btree_id_str(b->btree_id)); bch2_bpos_to_text(out, btree_node_pos(b)); - prt_tab(out); - prt_printf(out, " locks %u:%u:%u held by pid %u", + prt_printf(out, "\t locks %u:%u:%u held by pid %u", c.n[0], c.n[1], c.n[2], pid); } @@ -3229,10 +3225,8 @@ void bch2_btree_trans_to_text(struct printbuf *out, struct btree_trans *trans) b = READ_ONCE(trans->locking); if (b) { - prt_printf(out, " blocked for %lluus on", - div_u64(local_clock() - trans->locking_wait.start_time, - 1000)); - prt_newline(out); + prt_printf(out, " blocked for %lluus on\n", + div_u64(local_clock() - trans->locking_wait.start_time, 1000)); prt_printf(out, " %c", lock_types[trans->locking_wait.lock_want]); bch2_btree_bkey_cached_common_to_text(out, b); prt_newline(out); diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 7dafa1accec2..3dea89c6cf7e 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -1039,12 +1039,9 @@ int bch2_fs_btree_key_cache_init(struct btree_key_cache *bc) void bch2_btree_key_cache_to_text(struct printbuf *out, struct btree_key_cache *c) { - prt_printf(out, "nr_freed:\t%lu", atomic_long_read(&c->nr_freed)); - prt_newline(out); - prt_printf(out, "nr_keys:\t%lu", atomic_long_read(&c->nr_keys)); - prt_newline(out); - prt_printf(out, "nr_dirty:\t%lu", atomic_long_read(&c->nr_dirty)); - prt_newline(out); + prt_printf(out, "nr_freed:\t%lu\n", atomic_long_read(&c->nr_freed)); + prt_printf(out, "nr_keys:\t%lu\n", atomic_long_read(&c->nr_keys)); + prt_printf(out, "nr_dirty:\t%lu\n", atomic_long_read(&c->nr_dirty)); } void bch2_btree_key_cache_exit(void) diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index f2caf491957e..4ee4855395b9 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -83,8 +83,7 @@ static noinline void print_cycle(struct printbuf *out, struct lock_graph *g) { struct trans_waiting_for_lock *i; - prt_printf(out, "Found lock cycle (%u entries):", g->nr); - prt_newline(out); + prt_printf(out, "Found lock cycle (%u entries):\n", g->nr); for (i = g->g; i < g->g + g->nr; i++) { struct task_struct *task = READ_ONCE(i->trans->locking_wait.task); @@ -224,8 +223,7 @@ static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle) bch2_btree_trans_to_text(&buf, trans); - prt_printf(&buf, "backtrace:"); - prt_newline(&buf); + prt_printf(&buf, "backtrace:\n"); printbuf_indent_add(&buf, 2); bch2_prt_task_backtrace(&buf, trans->locking_wait.task, 2, GFP_NOWAIT); printbuf_indent_sub(&buf, 2); diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index bbec91e8e650..79cbf54c8912 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -773,9 +773,8 @@ static noinline int bch2_trans_commit_bkey_invalid(struct btree_trans *trans, struct bch_fs *c = trans->c; printbuf_reset(err); - prt_printf(err, "invalid bkey on insert from %s -> %ps", + prt_printf(err, "invalid bkey on insert from %s -> %ps\n", trans->fn, (void *) i->ip_allocated); - prt_newline(err); printbuf_indent_add(err, 2); bch2_bkey_val_to_text(err, c, bkey_i_to_s_c(i->k)); @@ -796,8 +795,7 @@ static noinline int bch2_trans_commit_journal_entry_invalid(struct btree_trans * struct bch_fs *c = trans->c; struct printbuf buf = PRINTBUF; - prt_printf(&buf, "invalid bkey on insert from %s", trans->fn); - prt_newline(&buf); + prt_printf(&buf, "invalid bkey on insert from %s\n", trans->fn); printbuf_indent_add(&buf, 2); bch2_journal_entry_to_text(&buf, c, i); diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 82f179258867..6ac6f61cfbc4 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -274,25 +274,14 @@ void bch2_dev_usage_init(struct bch_dev *ca) void bch2_dev_usage_to_text(struct printbuf *out, struct bch_dev_usage *usage) { - prt_tab(out); - prt_str(out, "buckets"); - prt_tab_rjust(out); - prt_str(out, "sectors"); - prt_tab_rjust(out); - prt_str(out, "fragmented"); - prt_tab_rjust(out); - prt_newline(out); + prt_printf(out, "\tbuckets\rsectors\rfragmented\r\n"); for (unsigned i = 0; i < BCH_DATA_NR; i++) { bch2_prt_data_type(out, i); - prt_tab(out); - prt_u64(out, usage->d[i].buckets); - prt_tab_rjust(out); - prt_u64(out, usage->d[i].sectors); - prt_tab_rjust(out); - prt_u64(out, usage->d[i].fragmented); - prt_tab_rjust(out); - prt_newline(out); + prt_printf(out, "\t%llu\r%llu\r%llu\r\n", + usage->d[i].buckets, + usage->d[i].sectors, + usage->d[i].fragmented); } } diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index 088fd2e7bdf1..2b3677ae0942 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -494,14 +494,10 @@ static void bch2_sb_crypt_to_text(struct printbuf *out, struct bch_sb *sb, { struct bch_sb_field_crypt *crypt = field_to_type(f, crypt); - prt_printf(out, "KFD: %llu", BCH_CRYPT_KDF_TYPE(crypt)); - prt_newline(out); - prt_printf(out, "scrypt n: %llu", BCH_KDF_SCRYPT_N(crypt)); - prt_newline(out); - prt_printf(out, "scrypt r: %llu", BCH_KDF_SCRYPT_R(crypt)); - prt_newline(out); - prt_printf(out, "scrypt p: %llu", BCH_KDF_SCRYPT_P(crypt)); - prt_newline(out); + prt_printf(out, "KFD: %llu\n", BCH_CRYPT_KDF_TYPE(crypt)); + prt_printf(out, "scrypt n: %llu\n", BCH_KDF_SCRYPT_N(crypt)); + prt_printf(out, "scrypt r: %llu\n", BCH_KDF_SCRYPT_R(crypt)); + prt_printf(out, "scrypt p: %llu\n", BCH_KDF_SCRYPT_P(crypt)); } const struct bch_sb_field_ops bch_sb_field_ops_crypt = { diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index cd99b7399414..d0ec74e6dc95 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -492,51 +492,26 @@ static void bch2_cached_btree_node_to_text(struct printbuf *out, struct bch_fs * if (!out->nr_tabstops) printbuf_tabstop_push(out, 32); - prt_printf(out, "%px btree=%s l=%u ", - b, - bch2_btree_id_str(b->c.btree_id), - b->c.level); - prt_newline(out); + prt_printf(out, "%px btree=%s l=%u\n", b, bch2_btree_id_str(b->c.btree_id), b->c.level); printbuf_indent_add(out, 2); bch2_bkey_val_to_text(out, c, bkey_i_to_s_c(&b->key)); prt_newline(out); - prt_printf(out, "flags: "); - prt_tab(out); + prt_printf(out, "flags:\t"); prt_bitflags(out, bch2_btree_node_flags, b->flags); prt_newline(out); - prt_printf(out, "pcpu read locks: "); - prt_tab(out); - prt_printf(out, "%u", b->c.lock.readers != NULL); - prt_newline(out); - - prt_printf(out, "written:"); - prt_tab(out); - prt_printf(out, "%u", b->written); - prt_newline(out); - - prt_printf(out, "writes blocked:"); - prt_tab(out); - prt_printf(out, "%u", !list_empty_careful(&b->write_blocked)); - prt_newline(out); - - prt_printf(out, "will make reachable:"); - prt_tab(out); - prt_printf(out, "%lx", b->will_make_reachable); - prt_newline(out); - - prt_printf(out, "journal pin %px:", &b->writes[0].journal); - prt_tab(out); - prt_printf(out, "%llu", b->writes[0].journal.seq); - prt_newline(out); + prt_printf(out, "pcpu read locks:\t%u\n", b->c.lock.readers != NULL); + prt_printf(out, "written:\t%u\n", b->written); + prt_printf(out, "writes blocked:\t%u\n", !list_empty_careful(&b->write_blocked)); + prt_printf(out, "will make reachable:\t%lx\n", b->will_make_reachable); - prt_printf(out, "journal pin %px:", &b->writes[1].journal); - prt_tab(out); - prt_printf(out, "%llu", b->writes[1].journal.seq); - prt_newline(out); + prt_printf(out, "journal pin %px:\t%llu\n", + &b->writes[0].journal, b->writes[0].journal.seq); + prt_printf(out, "journal pin %px:\t%llu\n", + &b->writes[1].journal, b->writes[1].journal.seq); printbuf_indent_sub(out, 2); } @@ -625,8 +600,7 @@ restart: bch2_btree_trans_to_text(&i->buf, trans); - prt_printf(&i->buf, "backtrace:"); - prt_newline(&i->buf); + prt_printf(&i->buf, "backtrace:\n"); printbuf_indent_add(&i->buf, 2); bch2_prt_task_backtrace(&i->buf, task, 0, GFP_KERNEL); printbuf_indent_sub(&i->buf, 2); @@ -782,25 +756,20 @@ static ssize_t btree_transaction_stats_read(struct file *file, char __user *buf, !bch2_btree_transaction_fns[i->iter]) break; - prt_printf(&i->buf, "%s: ", bch2_btree_transaction_fns[i->iter]); - prt_newline(&i->buf); + prt_printf(&i->buf, "%s:\n", bch2_btree_transaction_fns[i->iter]); printbuf_indent_add(&i->buf, 2); mutex_lock(&s->lock); - prt_printf(&i->buf, "Max mem used: %u", s->max_mem); - prt_newline(&i->buf); - - prt_printf(&i->buf, "Transaction duration:"); - prt_newline(&i->buf); + prt_printf(&i->buf, "Max mem used: %u\n", s->max_mem); + prt_printf(&i->buf, "Transaction duration:\n"); printbuf_indent_add(&i->buf, 2); bch2_time_stats_to_text(&i->buf, &s->duration); printbuf_indent_sub(&i->buf, 2); if (IS_ENABLED(CONFIG_BCACHEFS_LOCK_TIME_STATS)) { - prt_printf(&i->buf, "Lock hold times:"); - prt_newline(&i->buf); + prt_printf(&i->buf, "Lock hold times:\n"); printbuf_indent_add(&i->buf, 2); bch2_time_stats_to_text(&i->buf, &s->lock_hold_times); @@ -808,8 +777,7 @@ static ssize_t btree_transaction_stats_read(struct file *file, char __user *buf, } if (s->max_paths_text) { - prt_printf(&i->buf, "Maximum allocated btree paths (%u):", s->nr_max_paths); - prt_newline(&i->buf); + prt_printf(&i->buf, "Maximum allocated btree paths (%u):\n", s->nr_max_paths); printbuf_indent_add(&i->buf, 2); prt_str_indented(&i->buf, s->max_paths_text); diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index 0f95d7fb5ec0..dc72c59773fe 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -535,29 +535,19 @@ static void __bch2_inode_unpacked_to_text(struct printbuf *out, struct bch_inode_unpacked *inode) { printbuf_indent_add(out, 2); - prt_printf(out, "mode=%o", inode->bi_mode); - prt_newline(out); + prt_printf(out, "mode=%o\n", inode->bi_mode); prt_str(out, "flags="); prt_bitflags(out, bch2_inode_flag_strs, inode->bi_flags & ((1U << 20) - 1)); - prt_printf(out, " (%x)", inode->bi_flags); - prt_newline(out); + prt_printf(out, " (%x)\n", inode->bi_flags); - prt_printf(out, "journal_seq=%llu", inode->bi_journal_seq); - prt_newline(out); - - prt_printf(out, "bi_size=%llu", inode->bi_size); - prt_newline(out); - - prt_printf(out, "bi_sectors=%llu", inode->bi_sectors); - prt_newline(out); - - prt_printf(out, "bi_version=%llu", inode->bi_version); - prt_newline(out); + prt_printf(out, "journal_seq=%llu\n", inode->bi_journal_seq); + prt_printf(out, "bi_size=%llu\n", inode->bi_size); + prt_printf(out, "bi_sectors=%llu\n", inode->bi_sectors); + prt_printf(out, "bi_version=%llu\n", inode->bi_version); #define x(_name, _bits) \ - prt_printf(out, #_name "=%llu", (u64) inode->_name); \ - prt_newline(out); + prt_printf(out, #_name "=%llu\n", (u64) inode->_name); BCH_INODE_FIELDS_v3() #undef x printbuf_indent_sub(out, 2); diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 8a556e6d1ab6..c07902b6c759 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -771,9 +771,8 @@ static noinline void read_from_stale_dirty_pointer(struct btree_trans *trans, PTR_BUCKET_POS(c, &ptr), BTREE_ITER_CACHED); - prt_printf(&buf, "Attempting to read from stale dirty pointer:"); + prt_printf(&buf, "Attempting to read from stale dirty pointer:\n"); printbuf_indent_add(&buf, 2); - prt_newline(&buf); bch2_bkey_val_to_text(&buf, c, k); prt_newline(&buf); diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 40d7df7607df..c11e0944b069 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1649,8 +1649,7 @@ void bch2_write_op_to_text(struct printbuf *out, struct bch_write_op *op) prt_bitflags(out, bch2_write_flags, op->flags); prt_newline(out); - prt_printf(out, "ref: %u", closure_nr_remaining(&op->cl)); - prt_newline(out); + prt_printf(out, "ref: %u\n", closure_nr_remaining(&op->cl)); printbuf_indent_sub(out, 2); } diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index a8b08e76d0d0..c40b5c40d63f 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -53,29 +53,19 @@ static void bch2_journal_buf_to_text(struct printbuf *out, struct journal *j, u6 unsigned i = seq & JOURNAL_BUF_MASK; struct journal_buf *buf = j->buf + i; - prt_str(out, "seq:"); - prt_tab(out); - prt_printf(out, "%llu", seq); - prt_newline(out); + prt_printf(out, "seq:\t%llu\n", seq); printbuf_indent_add(out, 2); - prt_str(out, "refcount:"); - prt_tab(out); - prt_printf(out, "%u", journal_state_count(s, i)); - prt_newline(out); + prt_printf(out, "refcount:\t%u\n", journal_state_count(s, i)); - prt_str(out, "size:"); - prt_tab(out); + prt_printf(out, "size:\t"); prt_human_readable_u64(out, vstruct_bytes(buf->data)); prt_newline(out); - prt_str(out, "expires:"); - prt_tab(out); - prt_printf(out, "%li jiffies", buf->expires - jiffies); - prt_newline(out); + prt_printf(out, "expires:\t"); + prt_printf(out, "%li jiffies\n", buf->expires - jiffies); - prt_str(out, "flags:"); - prt_tab(out); + prt_printf(out, "flags:\t"); if (buf->noflush) prt_str(out, "noflush "); if (buf->must_flush) @@ -1422,12 +1412,12 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) s = READ_ONCE(j->reservations); prt_printf(out, "dirty journal entries:\t%llu/%llu\n", fifo_used(&j->pin), j->pin.size); - prt_printf(out, "seq:\t\t\t%llu\n", journal_cur_seq(j)); - prt_printf(out, "seq_ondisk:\t\t%llu\n", j->seq_ondisk); - prt_printf(out, "last_seq:\t\t%llu\n", journal_last_seq(j)); + prt_printf(out, "seq:\t%llu\n", journal_cur_seq(j)); + prt_printf(out, "seq_ondisk:\t%llu\n", j->seq_ondisk); + prt_printf(out, "last_seq:\t%llu\n", journal_last_seq(j)); prt_printf(out, "last_seq_ondisk:\t%llu\n", j->last_seq_ondisk); prt_printf(out, "flushed_seq_ondisk:\t%llu\n", j->flushed_seq_ondisk); - prt_printf(out, "watermark:\t\t%s\n", bch2_watermarks[j->watermark]); + prt_printf(out, "watermark:\t%s\n", bch2_watermarks[j->watermark]); prt_printf(out, "each entry reserved:\t%u\n", j->entry_u64s_reserved); prt_printf(out, "nr flush writes:\t%llu\n", j->nr_flush_writes); prt_printf(out, "nr noflush writes:\t%llu\n", j->nr_noflush_writes); @@ -1436,48 +1426,48 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) prt_newline(out); prt_printf(out, "nr direct reclaim:\t%llu\n", j->nr_direct_reclaim); prt_printf(out, "nr background reclaim:\t%llu\n", j->nr_background_reclaim); - prt_printf(out, "reclaim kicked:\t\t%u\n", j->reclaim_kicked); + prt_printf(out, "reclaim kicked:\t%u\n", j->reclaim_kicked); prt_printf(out, "reclaim runs in:\t%u ms\n", time_after(j->next_reclaim, now) ? jiffies_to_msecs(j->next_reclaim - jiffies) : 0); - prt_printf(out, "blocked:\t\t%u\n", j->blocked); + prt_printf(out, "blocked:\t%u\n", j->blocked); prt_printf(out, "current entry sectors:\t%u\n", j->cur_entry_sectors); prt_printf(out, "current entry error:\t%s\n", bch2_journal_errors[j->cur_entry_error]); - prt_printf(out, "current entry:\t\t"); + prt_printf(out, "current entry:\t"); switch (s.cur_entry_offset) { case JOURNAL_ENTRY_ERROR_VAL: - prt_printf(out, "error"); + prt_printf(out, "error\n"); break; case JOURNAL_ENTRY_CLOSED_VAL: - prt_printf(out, "closed"); + prt_printf(out, "closed\n"); break; default: - prt_printf(out, "%u/%u", s.cur_entry_offset, j->cur_entry_u64s); + prt_printf(out, "%u/%u\n", s.cur_entry_offset, j->cur_entry_u64s); break; } - prt_newline(out); - prt_printf(out, "unwritten entries:"); - prt_newline(out); + prt_printf(out, "unwritten entries:\n"); bch2_journal_bufs_to_text(out, j); prt_printf(out, - "replay done:\t\t%i\n", + "replay done:\t%i\n", test_bit(JOURNAL_REPLAY_DONE, &j->flags)); prt_printf(out, "space:\n"); - prt_printf(out, "\tdiscarded\t%u:%u\n", + printbuf_indent_add(out, 2); + prt_printf(out, "discarded\t%u:%u\n", j->space[journal_space_discarded].next_entry, j->space[journal_space_discarded].total); - prt_printf(out, "\tclean ondisk\t%u:%u\n", + prt_printf(out, "clean ondisk\t%u:%u\n", j->space[journal_space_clean_ondisk].next_entry, j->space[journal_space_clean_ondisk].total); - prt_printf(out, "\tclean\t\t%u:%u\n", + prt_printf(out, "clean\t%u:%u\n", j->space[journal_space_clean].next_entry, j->space[journal_space_clean].total); - prt_printf(out, "\ttotal\t\t%u:%u\n", + prt_printf(out, "total\t%u:%u\n", j->space[journal_space_total].next_entry, j->space[journal_space_total].total); + printbuf_indent_sub(out, 2); for_each_member_device_rcu(c, ca, &c->rw_devs[BCH_DATA_journal]) { struct journal_device *ja = &ca->journal; @@ -1488,14 +1478,16 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) if (!ja->nr) continue; - prt_printf(out, "dev %u:\n", ca->dev_idx); - prt_printf(out, "\tnr\t\t%u\n", ja->nr); - prt_printf(out, "\tbucket size\t%u\n", ca->mi.bucket_size); - prt_printf(out, "\tavailable\t%u:%u\n", bch2_journal_dev_buckets_available(j, ja, journal_space_discarded), ja->sectors_free); - prt_printf(out, "\tdiscard_idx\t%u\n", ja->discard_idx); - prt_printf(out, "\tdirty_ondisk\t%u (seq %llu)\n", ja->dirty_idx_ondisk, ja->bucket_seq[ja->dirty_idx_ondisk]); - prt_printf(out, "\tdirty_idx\t%u (seq %llu)\n", ja->dirty_idx, ja->bucket_seq[ja->dirty_idx]); - prt_printf(out, "\tcur_idx\t\t%u (seq %llu)\n", ja->cur_idx, ja->bucket_seq[ja->cur_idx]); + prt_printf(out, "dev %u:\n", ca->dev_idx); + printbuf_indent_add(out, 2); + prt_printf(out, "nr\t%u\n", ja->nr); + prt_printf(out, "bucket size\t%u\n", ca->mi.bucket_size); + prt_printf(out, "available\t%u:%u\n", bch2_journal_dev_buckets_available(j, ja, journal_space_discarded), ja->sectors_free); + prt_printf(out, "discard_idx\t%u\n", ja->discard_idx); + prt_printf(out, "dirty_ondisk\t%u (seq %llu)\n",ja->dirty_idx_ondisk, ja->bucket_seq[ja->dirty_idx_ondisk]); + prt_printf(out, "dirty_idx\t%u (seq %llu)\n", ja->dirty_idx, ja->bucket_seq[ja->dirty_idx]); + prt_printf(out, "cur_idx\t%u (seq %llu)\n", ja->cur_idx, ja->bucket_seq[ja->cur_idx]); + printbuf_indent_sub(out, 2); } rcu_read_unlock(); @@ -1527,25 +1519,18 @@ bool bch2_journal_seq_pins_to_text(struct printbuf *out, struct journal *j, u64 pin_list = journal_seq_pin(j, *seq); - prt_printf(out, "%llu: count %u", *seq, atomic_read(&pin_list->count)); - prt_newline(out); + prt_printf(out, "%llu: count %u\n", *seq, atomic_read(&pin_list->count)); printbuf_indent_add(out, 2); for (unsigned i = 0; i < ARRAY_SIZE(pin_list->list); i++) - list_for_each_entry(pin, &pin_list->list[i], list) { - prt_printf(out, "\t%px %ps", pin, pin->flush); - prt_newline(out); - } + list_for_each_entry(pin, &pin_list->list[i], list) + prt_printf(out, "\t%px %ps\n", pin, pin->flush); - if (!list_empty(&pin_list->flushed)) { - prt_printf(out, "flushed:"); - prt_newline(out); - } + if (!list_empty(&pin_list->flushed)) + prt_printf(out, "flushed:\n"); - list_for_each_entry(pin, &pin_list->flushed, list) { - prt_printf(out, "\t%px %ps", pin, pin->flush); - prt_newline(out); - } + list_for_each_entry(pin, &pin_list->flushed, list) + prt_printf(out, "\t%px %ps\n", pin, pin->flush); printbuf_indent_sub(out, 2); diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index 45984a688e5b..e491203f21c1 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -41,28 +41,23 @@ static void bch2_data_update_opts_to_text(struct printbuf *out, struct bch_fs *c struct data_update_opts *data_opts) { printbuf_tabstop_push(out, 20); - prt_str(out, "rewrite ptrs:"); - prt_tab(out); + prt_str(out, "rewrite ptrs:\t"); bch2_prt_u64_base2(out, data_opts->rewrite_ptrs); prt_newline(out); - prt_str(out, "kill ptrs: "); - prt_tab(out); + prt_str(out, "kill ptrs:\t"); bch2_prt_u64_base2(out, data_opts->kill_ptrs); prt_newline(out); - prt_str(out, "target: "); - prt_tab(out); + prt_str(out, "target:\t"); bch2_target_to_text(out, c, data_opts->target); prt_newline(out); - prt_str(out, "compression: "); - prt_tab(out); + prt_str(out, "compression:\t"); bch2_compression_opt_to_text(out, background_compression(*io_opts)); prt_newline(out); - prt_str(out, "extra replicas: "); - prt_tab(out); + prt_str(out, "extra replicas:\t"); prt_u64(out, data_opts->extra_replicas); } @@ -1127,23 +1122,17 @@ void bch2_move_stats_to_text(struct printbuf *out, struct bch_move_stats *stats) prt_newline(out); printbuf_indent_add(out, 2); - prt_str(out, "keys moved: "); - prt_u64(out, atomic64_read(&stats->keys_moved)); - prt_newline(out); - - prt_str(out, "keys raced: "); - prt_u64(out, atomic64_read(&stats->keys_raced)); - prt_newline(out); - - prt_str(out, "bytes seen: "); + prt_printf(out, "keys moved: %llu\n", atomic64_read(&stats->keys_moved)); + prt_printf(out, "keys raced: %llu\n", atomic64_read(&stats->keys_raced)); + prt_printf(out, "bytes seen: "); prt_human_readable_u64(out, atomic64_read(&stats->sectors_seen) << 9); prt_newline(out); - prt_str(out, "bytes moved: "); + prt_printf(out, "bytes moved: "); prt_human_readable_u64(out, atomic64_read(&stats->sectors_moved) << 9); prt_newline(out); - prt_str(out, "bytes raced: "); + prt_printf(out, "bytes raced: "); prt_human_readable_u64(out, atomic64_read(&stats->sectors_raced) << 9); prt_newline(out); @@ -1157,19 +1146,17 @@ static void bch2_moving_ctxt_to_text(struct printbuf *out, struct bch_fs *c, str bch2_move_stats_to_text(out, ctxt->stats); printbuf_indent_add(out, 2); - prt_printf(out, "reads: ios %u/%u sectors %u/%u", + prt_printf(out, "reads: ios %u/%u sectors %u/%u\n", atomic_read(&ctxt->read_ios), c->opts.move_ios_in_flight, atomic_read(&ctxt->read_sectors), c->opts.move_bytes_in_flight >> 9); - prt_newline(out); - prt_printf(out, "writes: ios %u/%u sectors %u/%u", + prt_printf(out, "writes: ios %u/%u sectors %u/%u\n", atomic_read(&ctxt->write_ios), c->opts.move_ios_in_flight, atomic_read(&ctxt->write_sectors), c->opts.move_bytes_in_flight >> 9); - prt_newline(out); printbuf_indent_add(out, 2); diff --git a/fs/bcachefs/quota.c b/fs/bcachefs/quota.c index 556da0738106..5660798c6453 100644 --- a/fs/bcachefs/quota.c +++ b/fs/bcachefs/quota.c @@ -97,45 +97,14 @@ static void qc_info_to_text(struct printbuf *out, struct qc_info *i) printbuf_tabstops_reset(out); printbuf_tabstop_push(out, 20); - prt_str(out, "i_fieldmask"); - prt_tab(out); - prt_printf(out, "%x", i->i_fieldmask); - prt_newline(out); - - prt_str(out, "i_flags"); - prt_tab(out); - prt_printf(out, "%u", i->i_flags); - prt_newline(out); - - prt_str(out, "i_spc_timelimit"); - prt_tab(out); - prt_printf(out, "%u", i->i_spc_timelimit); - prt_newline(out); - - prt_str(out, "i_ino_timelimit"); - prt_tab(out); - prt_printf(out, "%u", i->i_ino_timelimit); - prt_newline(out); - - prt_str(out, "i_rt_spc_timelimit"); - prt_tab(out); - prt_printf(out, "%u", i->i_rt_spc_timelimit); - prt_newline(out); - - prt_str(out, "i_spc_warnlimit"); - prt_tab(out); - prt_printf(out, "%u", i->i_spc_warnlimit); - prt_newline(out); - - prt_str(out, "i_ino_warnlimit"); - prt_tab(out); - prt_printf(out, "%u", i->i_ino_warnlimit); - prt_newline(out); - - prt_str(out, "i_rt_spc_warnlimit"); - prt_tab(out); - prt_printf(out, "%u", i->i_rt_spc_warnlimit); - prt_newline(out); + prt_printf(out, "i_fieldmask\t%x\n", i->i_fieldmask); + prt_printf(out, "i_flags\t%u\n", i->i_flags); + prt_printf(out, "i_spc_timelimit\t%u\n", i->i_spc_timelimit); + prt_printf(out, "i_ino_timelimit\t%u\n", i->i_ino_timelimit); + prt_printf(out, "i_rt_spc_timelimit\t%u\n", i->i_rt_spc_timelimit); + prt_printf(out, "i_spc_warnlimit\t%u\n", i->i_spc_warnlimit); + prt_printf(out, "i_ino_warnlimit\t%u\n", i->i_ino_warnlimit); + prt_printf(out, "i_rt_spc_warnlimit\t%u\n", i->i_rt_spc_warnlimit); } static void qc_dqblk_to_text(struct printbuf *out, struct qc_dqblk *q) @@ -143,60 +112,17 @@ static void qc_dqblk_to_text(struct printbuf *out, struct qc_dqblk *q) printbuf_tabstops_reset(out); printbuf_tabstop_push(out, 20); - prt_str(out, "d_fieldmask"); - prt_tab(out); - prt_printf(out, "%x", q->d_fieldmask); - prt_newline(out); - - prt_str(out, "d_spc_hardlimit"); - prt_tab(out); - prt_printf(out, "%llu", q->d_spc_hardlimit); - prt_newline(out); - - prt_str(out, "d_spc_softlimit"); - prt_tab(out); - prt_printf(out, "%llu", q->d_spc_softlimit); - prt_newline(out); - - prt_str(out, "d_ino_hardlimit"); - prt_tab(out); - prt_printf(out, "%llu", q->d_ino_hardlimit); - prt_newline(out); - - prt_str(out, "d_ino_softlimit"); - prt_tab(out); - prt_printf(out, "%llu", q->d_ino_softlimit); - prt_newline(out); - - prt_str(out, "d_space"); - prt_tab(out); - prt_printf(out, "%llu", q->d_space); - prt_newline(out); - - prt_str(out, "d_ino_count"); - prt_tab(out); - prt_printf(out, "%llu", q->d_ino_count); - prt_newline(out); - - prt_str(out, "d_ino_timer"); - prt_tab(out); - prt_printf(out, "%llu", q->d_ino_timer); - prt_newline(out); - - prt_str(out, "d_spc_timer"); - prt_tab(out); - prt_printf(out, "%llu", q->d_spc_timer); - prt_newline(out); - - prt_str(out, "d_ino_warns"); - prt_tab(out); - prt_printf(out, "%i", q->d_ino_warns); - prt_newline(out); - - prt_str(out, "d_spc_warns"); - prt_tab(out); - prt_printf(out, "%i", q->d_spc_warns); - prt_newline(out); + prt_printf(out, "d_fieldmask\t%x\n", q->d_fieldmask); + prt_printf(out, "d_spc_hardlimit\t%llu\n", q->d_spc_hardlimit); + prt_printf(out, "d_spc_softlimit\t%llu\n", q->d_spc_softlimit); + prt_printf(out, "d_ino_hardlimit\%llu\n", q->d_ino_hardlimit); + prt_printf(out, "d_ino_softlimit\t%llu\n", q->d_ino_softlimit); + prt_printf(out, "d_space\t%llu\n", q->d_space); + prt_printf(out, "d_ino_count\t%llu\n", q->d_ino_count); + prt_printf(out, "d_ino_timer\t%llu\n", q->d_ino_timer); + prt_printf(out, "d_spc_timer\t%llu\n", q->d_spc_timer); + prt_printf(out, "d_ino_warns\t%i\n", q->d_ino_warns); + prt_printf(out, "d_spc_warns\t%i\n", q->d_spc_warns); } static inline unsigned __next_qtype(unsigned i, unsigned qtypes) diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index d27804d4987a..c2d4be52c8c5 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -298,10 +298,8 @@ static void bch2_sb_clean_to_text(struct printbuf *out, struct bch_sb *sb, struct bch_sb_field_clean *clean = field_to_type(f, clean); struct jset_entry *entry; - prt_printf(out, "flags: %x", le32_to_cpu(clean->flags)); - prt_newline(out); - prt_printf(out, "journal_seq: %llu", le64_to_cpu(clean->journal_seq)); - prt_newline(out); + prt_printf(out, "flags: %x\n", le32_to_cpu(clean->flags)); + prt_printf(out, "journal_seq: %llu\n", le64_to_cpu(clean->journal_seq)); for (entry = clean->start; entry != vstruct_end(&clean->field); diff --git a/fs/bcachefs/sb-counters.c b/fs/bcachefs/sb-counters.c index 7dc898761bb3..7282f7886192 100644 --- a/fs/bcachefs/sb-counters.c +++ b/fs/bcachefs/sb-counters.c @@ -31,19 +31,12 @@ static void bch2_sb_counters_to_text(struct printbuf *out, struct bch_sb *sb, struct bch_sb_field *f) { struct bch_sb_field_counters *ctrs = field_to_type(f, counters); - unsigned int i; unsigned int nr = bch2_sb_counter_nr_entries(ctrs); - for (i = 0; i < nr; i++) { - if (i < BCH_COUNTER_NR) - prt_printf(out, "%s ", bch2_counter_names[i]); - else - prt_printf(out, "(unknown)"); - - prt_tab(out); - prt_printf(out, "%llu", le64_to_cpu(ctrs->d[i])); - prt_newline(out); - } + for (unsigned i = 0; i < nr; i++) + prt_printf(out, "%s \t%llu\n", + i < BCH_COUNTER_NR ? bch2_counter_names[i] : "(unknown)", + le64_to_cpu(ctrs->d[i])); }; int bch2_sb_counters_to_cpu(struct bch_fs *c) diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c index a98ef940b7a3..90b06f8aa1df 100644 --- a/fs/bcachefs/sb-downgrade.c +++ b/fs/bcachefs/sb-downgrade.c @@ -164,19 +164,16 @@ static void bch2_sb_downgrade_to_text(struct printbuf *out, struct bch_sb *sb, printbuf_tabstop_push(out, 16); for_each_downgrade_entry(e, i) { - prt_str(out, "version:"); - prt_tab(out); + prt_str(out, "version:\t"); bch2_version_to_text(out, le16_to_cpu(i->version)); prt_newline(out); - prt_str(out, "recovery passes:"); - prt_tab(out); + prt_str(out, "recovery passes:\t"); prt_bitflags(out, bch2_recovery_passes, bch2_recovery_passes_from_stable(le64_to_cpu(i->recovery_passes[0]))); prt_newline(out); - prt_str(out, "errors:"); - prt_tab(out); + prt_str(out, "errors:\t"); bool first = true; for (unsigned j = 0; j < le16_to_cpu(i->nr_errors); j++) { if (!first) diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 44b3f0cb7b49..0007fcad5baa 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -167,15 +167,11 @@ static void member_to_text(struct printbuf *out, if (!bch2_member_exists(&m)) return; - prt_printf(out, "Device:"); - prt_tab(out); - prt_printf(out, "%u", i); - prt_newline(out); + prt_printf(out, "Device:\t%u\n", i); printbuf_indent_add(out, 2); - prt_printf(out, "Label:"); - prt_tab(out); + prt_printf(out, "Label:\t"); if (BCH_MEMBER_GROUP(&m)) { unsigned idx = BCH_MEMBER_GROUP(&m) - 1; @@ -189,96 +185,59 @@ static void member_to_text(struct printbuf *out, } prt_newline(out); - prt_printf(out, "UUID:"); - prt_tab(out); + prt_printf(out, "UUID:\t"); pr_uuid(out, m.uuid.b); prt_newline(out); - prt_printf(out, "Size:"); - prt_tab(out); + prt_printf(out, "Size:\t"); prt_units_u64(out, device_size << 9); prt_newline(out); - for (unsigned i = 0; i < BCH_MEMBER_ERROR_NR; i++) { - prt_printf(out, "%s errors:", bch2_member_error_strs[i]); - prt_tab(out); - prt_u64(out, le64_to_cpu(m.errors[i])); - prt_newline(out); - } + for (unsigned i = 0; i < BCH_MEMBER_ERROR_NR; i++) + prt_printf(out, "%s errors:\t%llu\n", bch2_member_error_strs[i], le64_to_cpu(m.errors[i])); - for (unsigned i = 0; i < BCH_IOPS_NR; i++) { - prt_printf(out, "%s iops:", bch2_iops_measurements[i]); - prt_tab(out); - prt_printf(out, "%u", le32_to_cpu(m.iops[i])); - prt_newline(out); - } + for (unsigned i = 0; i < BCH_IOPS_NR; i++) + prt_printf(out, "%s iops:\t%u\n", bch2_iops_measurements[i], le32_to_cpu(m.iops[i])); - prt_printf(out, "Bucket size:"); - prt_tab(out); + prt_printf(out, "Bucket size:\t"); prt_units_u64(out, bucket_size << 9); prt_newline(out); - prt_printf(out, "First bucket:"); - prt_tab(out); - prt_printf(out, "%u", le16_to_cpu(m.first_bucket)); - prt_newline(out); + prt_printf(out, "First bucket:\t%u\n", le16_to_cpu(m.first_bucket)); + prt_printf(out, "Buckets:\t%llu\n", le64_to_cpu(m.nbuckets)); - prt_printf(out, "Buckets:"); - prt_tab(out); - prt_printf(out, "%llu", le64_to_cpu(m.nbuckets)); - prt_newline(out); - - prt_printf(out, "Last mount:"); - prt_tab(out); + prt_printf(out, "Last mount:\t"); if (m.last_mount) bch2_prt_datetime(out, le64_to_cpu(m.last_mount)); else prt_printf(out, "(never)"); prt_newline(out); - prt_printf(out, "Last superblock write:"); - prt_tab(out); - prt_u64(out, le64_to_cpu(m.seq)); - prt_newline(out); + prt_printf(out, "Last superblock write:\t%llu\n", le64_to_cpu(m.seq)); - prt_printf(out, "State:"); - prt_tab(out); - prt_printf(out, "%s", + prt_printf(out, "State:\t%s\n", BCH_MEMBER_STATE(&m) < BCH_MEMBER_STATE_NR ? bch2_member_states[BCH_MEMBER_STATE(&m)] : "unknown"); - prt_newline(out); - prt_printf(out, "Data allowed:"); - prt_tab(out); + prt_printf(out, "Data allowed:\t"); if (BCH_MEMBER_DATA_ALLOWED(&m)) prt_bitflags(out, __bch2_data_types, BCH_MEMBER_DATA_ALLOWED(&m)); else prt_printf(out, "(none)"); prt_newline(out); - prt_printf(out, "Has data:"); - prt_tab(out); + prt_printf(out, "Has data:\t"); if (data_have) prt_bitflags(out, __bch2_data_types, data_have); else prt_printf(out, "(none)"); prt_newline(out); - prt_str(out, "Durability:"); - prt_tab(out); - prt_printf(out, "%llu", BCH_MEMBER_DURABILITY(&m) ? BCH_MEMBER_DURABILITY(&m) - 1 : 1); - prt_newline(out); + prt_printf(out, "Durability:\t%llu\n", BCH_MEMBER_DURABILITY(&m) ? BCH_MEMBER_DURABILITY(&m) - 1 : 1); - prt_printf(out, "Discard:"); - prt_tab(out); - prt_printf(out, "%llu", BCH_MEMBER_DISCARD(&m)); - prt_newline(out); - - prt_printf(out, "Freespace initialized:"); - prt_tab(out); - prt_printf(out, "%llu", BCH_MEMBER_FREESPACE_INITIALIZED(&m)); - prt_newline(out); + prt_printf(out, "Discard:\t%llu\n", BCH_MEMBER_DISCARD(&m)); + prt_printf(out, "Freespace initialized:\t%llu\n", BCH_MEMBER_FREESPACE_INITIALIZED(&m)); printbuf_indent_sub(out, 2); } @@ -390,12 +349,8 @@ void bch2_dev_io_errors_to_text(struct printbuf *out, struct bch_dev *ca) prt_newline(out); printbuf_indent_add(out, 2); - for (unsigned i = 0; i < BCH_MEMBER_ERROR_NR; i++) { - prt_printf(out, "%s:", bch2_member_error_strs[i]); - prt_tab(out); - prt_u64(out, atomic64_read(&ca->errors[i])); - prt_newline(out); - } + for (unsigned i = 0; i < BCH_MEMBER_ERROR_NR; i++) + prt_printf(out, "%s:\t%llu\n", bch2_member_error_strs[i], atomic64_read(&ca->errors[i])); printbuf_indent_sub(out, 2); prt_str(out, "IO errors since "); @@ -404,12 +359,9 @@ void bch2_dev_io_errors_to_text(struct printbuf *out, struct bch_dev *ca) prt_newline(out); printbuf_indent_add(out, 2); - for (unsigned i = 0; i < BCH_MEMBER_ERROR_NR; i++) { - prt_printf(out, "%s:", bch2_member_error_strs[i]); - prt_tab(out); - prt_u64(out, atomic64_read(&ca->errors[i]) - le64_to_cpu(m.errors_at_reset[i])); - prt_newline(out); - } + for (unsigned i = 0; i < BCH_MEMBER_ERROR_NR; i++) + prt_printf(out, "%s:\t%llu\n", bch2_member_error_strs[i], + atomic64_read(&ca->errors[i]) - le64_to_cpu(m.errors_at_reset[i])); printbuf_indent_sub(out, 2); } diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index 1cadc99898f5..52f0ffd4a0d2 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -1176,8 +1176,7 @@ static void bch2_sb_ext_to_text(struct printbuf *out, struct bch_sb *sb, { struct bch_sb_field_ext *e = field_to_type(f, ext); - prt_printf(out, "Recovery passes required:"); - prt_tab(out); + prt_printf(out, "Recovery passes required:\t"); prt_bitflags(out, bch2_recovery_passes, bch2_recovery_passes_from_stable(le64_to_cpu(e->recovery_passes_required[0]))); prt_newline(out); @@ -1186,16 +1185,14 @@ static void bch2_sb_ext_to_text(struct printbuf *out, struct bch_sb *sb, if (errors_silent) { le_bitvector_to_cpu(errors_silent, (void *) e->errors_silent, sizeof(e->errors_silent) * 8); - prt_printf(out, "Errors to silently fix:"); - prt_tab(out); + prt_printf(out, "Errors to silently fix:\t"); prt_bitflags_vector(out, bch2_sb_error_strs, errors_silent, sizeof(e->errors_silent) * 8); prt_newline(out); kfree(errors_silent); } - prt_printf(out, "Btrees with missing data:"); - prt_tab(out); + prt_printf(out, "Btrees with missing data:\t"); prt_bitflags(out, __bch2_btree_ids, le64_to_cpu(e->btrees_lost_data)); prt_newline(out); } @@ -1305,95 +1302,71 @@ void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb, for (int i = 0; i < sb->nr_devices; i++) nr_devices += bch2_dev_exists(sb, i); - prt_printf(out, "External UUID:"); - prt_tab(out); + prt_printf(out, "External UUID:\t"); pr_uuid(out, sb->user_uuid.b); prt_newline(out); - prt_printf(out, "Internal UUID:"); - prt_tab(out); + prt_printf(out, "Internal UUID:\t"); pr_uuid(out, sb->uuid.b); prt_newline(out); - prt_printf(out, "Magic number:"); - prt_tab(out); + prt_printf(out, "Magic number:\t"); pr_uuid(out, sb->magic.b); prt_newline(out); - prt_str(out, "Device index:"); - prt_tab(out); - prt_printf(out, "%u", sb->dev_idx); - prt_newline(out); + prt_printf(out, "Device index:\t%u\n", sb->dev_idx); - prt_str(out, "Label:"); - prt_tab(out); + prt_str(out, "Label:\t"); prt_printf(out, "%.*s", (int) sizeof(sb->label), sb->label); prt_newline(out); - prt_str(out, "Version:"); - prt_tab(out); + prt_str(out, "Version:\t"); bch2_version_to_text(out, le16_to_cpu(sb->version)); prt_newline(out); - prt_str(out, "Version upgrade complete:"); - prt_tab(out); + prt_str(out, "Version upgrade complete:\t"); bch2_version_to_text(out, BCH_SB_VERSION_UPGRADE_COMPLETE(sb)); prt_newline(out); - prt_printf(out, "Oldest version on disk:"); - prt_tab(out); + prt_printf(out, "Oldest version on disk:\t"); bch2_version_to_text(out, le16_to_cpu(sb->version_min)); prt_newline(out); - prt_printf(out, "Created:"); - prt_tab(out); + prt_printf(out, "Created:\t"); if (sb->time_base_lo) bch2_prt_datetime(out, div_u64(le64_to_cpu(sb->time_base_lo), NSEC_PER_SEC)); else prt_printf(out, "(not set)"); prt_newline(out); - prt_printf(out, "Sequence number:"); - prt_tab(out); + prt_printf(out, "Sequence number:\t"); prt_printf(out, "%llu", le64_to_cpu(sb->seq)); prt_newline(out); - prt_printf(out, "Time of last write:"); - prt_tab(out); + prt_printf(out, "Time of last write:\t"); bch2_prt_datetime(out, le64_to_cpu(sb->write_time)); prt_newline(out); - prt_printf(out, "Superblock size:"); - prt_tab(out); + prt_printf(out, "Superblock size:\t"); prt_units_u64(out, vstruct_bytes(sb)); prt_str(out, "/"); prt_units_u64(out, 512ULL << sb->layout.sb_max_size_bits); prt_newline(out); - prt_printf(out, "Clean:"); - prt_tab(out); - prt_printf(out, "%llu", BCH_SB_CLEAN(sb)); - prt_newline(out); - - prt_printf(out, "Devices:"); - prt_tab(out); - prt_printf(out, "%u", nr_devices); - prt_newline(out); + prt_printf(out, "Clean:\t%llu\n", BCH_SB_CLEAN(sb)); + prt_printf(out, "Devices:\t%u\n", nr_devices); - prt_printf(out, "Sections:"); + prt_printf(out, "Sections:\t"); vstruct_for_each(sb, f) fields_have |= 1 << le32_to_cpu(f->type); - prt_tab(out); prt_bitflags(out, bch2_sb_fields, fields_have); prt_newline(out); - prt_printf(out, "Features:"); - prt_tab(out); + prt_printf(out, "Features:\t"); prt_bitflags(out, bch2_sb_features, le64_to_cpu(sb->features[0])); prt_newline(out); - prt_printf(out, "Compat features:"); - prt_tab(out); + prt_printf(out, "Compat features:\t"); prt_bitflags(out, bch2_sb_compat, le64_to_cpu(sb->compat[0])); prt_newline(out); @@ -1410,8 +1383,7 @@ void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb, if (opt->get_sb != BCH2_NO_SB_OPT) { u64 v = bch2_opt_from_sb(sb, id); - prt_printf(out, "%s:", opt->attr.name); - prt_tab(out); + prt_printf(out, "%s:\t", opt->attr.name); bch2_opt_to_text(out, NULL, sb, opt, v, OPT_HUMAN_READABLE|OPT_SHOW_FULL_LIST); prt_newline(out); diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 5be92fe3f4ea..b6bfd7cbce18 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -189,12 +189,8 @@ static void bch2_write_refs_to_text(struct printbuf *out, struct bch_fs *c) { bch2_printbuf_tabstop_push(out, 24); - for (unsigned i = 0; i < ARRAY_SIZE(c->writes); i++) { - prt_str(out, bch2_write_refs[i]); - prt_tab(out); - prt_printf(out, "%li", atomic_long_read(&c->writes[i])); - prt_newline(out); - } + for (unsigned i = 0; i < ARRAY_SIZE(c->writes); i++) + prt_printf(out, "%s\t%li\n", bch2_write_refs[i], atomic_long_read(&c->writes[i])); } #endif @@ -313,22 +309,11 @@ static int bch2_compression_stats_to_text(struct printbuf *out, struct bch_fs *c if (ret) return ret; - prt_str(out, "type"); printbuf_tabstop_push(out, 12); - prt_tab(out); - - prt_str(out, "compressed"); printbuf_tabstop_push(out, 16); - prt_tab_rjust(out); - - prt_str(out, "uncompressed"); printbuf_tabstop_push(out, 16); - prt_tab_rjust(out); - - prt_str(out, "average extent size"); printbuf_tabstop_push(out, 24); - prt_tab_rjust(out); - prt_newline(out); + prt_printf(out, "type\tcompressed\runcompressed\raverage extent size\r\n"); for (unsigned i = 0; i < ARRAY_SIZE(s); i++) { bch2_prt_compression_type(out, i); @@ -594,13 +579,11 @@ SHOW(bch2_fs_counters) if (attr == &sysfs_##t) { \ counter = percpu_u64_get(&c->counters[BCH_COUNTER_##t]);\ counter_since_mount = counter - c->counters_on_mount[BCH_COUNTER_##t];\ - prt_printf(out, "since mount:"); \ - prt_tab(out); \ + prt_printf(out, "since mount:\t"); \ prt_human_readable_u64(out, counter_since_mount); \ prt_newline(out); \ \ - prt_printf(out, "since filesystem creation:"); \ - prt_tab(out); \ + prt_printf(out, "since filesystem creation:\t"); \ prt_human_readable_u64(out, counter); \ prt_newline(out); \ } @@ -796,11 +779,11 @@ static void dev_alloc_debug_to_text(struct printbuf *out, struct bch_dev *ca) { struct bch_fs *c = ca->fs; struct bch_dev_usage stats = bch2_dev_usage_read(ca); - unsigned i, nr[BCH_DATA_NR]; + unsigned nr[BCH_DATA_NR]; memset(nr, 0, sizeof(nr)); - for (i = 0; i < ARRAY_SIZE(c->open_buckets); i++) + for (unsigned i = 0; i < ARRAY_SIZE(c->open_buckets); i++) nr[c->open_buckets[i].data_type]++; printbuf_tabstop_push(out, 8); @@ -813,65 +796,24 @@ static void dev_alloc_debug_to_text(struct printbuf *out, struct bch_dev *ca) prt_newline(out); - prt_printf(out, "reserves:"); - prt_newline(out); - for (i = 0; i < BCH_WATERMARK_NR; i++) { - prt_str(out, bch2_watermarks[i]); - prt_tab(out); - prt_u64(out, bch2_dev_buckets_reserved(ca, i)); - prt_tab_rjust(out); - prt_newline(out); - } + prt_printf(out, "reserves:\n"); + for (unsigned i = 0; i < BCH_WATERMARK_NR; i++) + prt_printf(out, "%s\t%llu\r\n", bch2_watermarks[i], bch2_dev_buckets_reserved(ca, i)); prt_newline(out); printbuf_tabstops_reset(out); printbuf_tabstop_push(out, 24); - prt_str(out, "freelist_wait"); - prt_tab(out); - prt_str(out, c->freelist_wait.list.first ? "waiting" : "empty"); - prt_newline(out); - - prt_str(out, "open buckets allocated"); - prt_tab(out); - prt_u64(out, OPEN_BUCKETS_COUNT - c->open_buckets_nr_free); - prt_newline(out); - - prt_str(out, "open buckets this dev"); - prt_tab(out); - prt_u64(out, ca->nr_open_buckets); - prt_newline(out); - - prt_str(out, "open buckets total"); - prt_tab(out); - prt_u64(out, OPEN_BUCKETS_COUNT); - prt_newline(out); - - prt_str(out, "open_buckets_wait"); - prt_tab(out); - prt_str(out, c->open_buckets_wait.list.first ? "waiting" : "empty"); - prt_newline(out); - - prt_str(out, "open_buckets_btree"); - prt_tab(out); - prt_u64(out, nr[BCH_DATA_btree]); - prt_newline(out); - - prt_str(out, "open_buckets_user"); - prt_tab(out); - prt_u64(out, nr[BCH_DATA_user]); - prt_newline(out); - - prt_str(out, "buckets_to_invalidate"); - prt_tab(out); - prt_u64(out, should_invalidate_buckets(ca, stats)); - prt_newline(out); - - prt_str(out, "btree reserve cache"); - prt_tab(out); - prt_u64(out, c->btree_reserve_cache_nr); - prt_newline(out); + prt_printf(out, "freelist_wait\t%s\n", c->freelist_wait.list.first ? "waiting" : "empty"); + prt_printf(out, "open buckets allocated\t%i\n", OPEN_BUCKETS_COUNT - c->open_buckets_nr_free); + prt_printf(out, "open buckets this dev\t%i\n", ca->nr_open_buckets); + prt_printf(out, "open buckets total\t%u\n", OPEN_BUCKETS_COUNT); + prt_printf(out, "open_buckets_wait\t%s\n", c->open_buckets_wait.list.first ? "waiting" : "empty"); + prt_printf(out, "open_buckets_btree\t%u\n", nr[BCH_DATA_btree]); + prt_printf(out, "open_buckets_user\t%u\n", nr[BCH_DATA_user]); + prt_printf(out, "buckets_to_invalidate\t%llu\n", should_invalidate_buckets(ca, stats)); + prt_printf(out, "btree reserve cache\t%u\n", c->btree_reserve_cache_nr); } static const char * const bch2_rw[] = { diff --git a/fs/bcachefs/util.c b/fs/bcachefs/util.c index 92c6ad75e702..de331dec2a99 100644 --- a/fs/bcachefs/util.c +++ b/fs/bcachefs/util.c @@ -348,15 +348,12 @@ static void bch2_pr_time_units_aligned(struct printbuf *out, u64 ns) { const struct time_unit *u = bch2_pick_time_units(ns); - prt_printf(out, "%llu ", div64_u64(ns, u->nsecs)); - prt_tab_rjust(out); - prt_printf(out, "%s", u->name); + prt_printf(out, "%llu \r%s", div64_u64(ns, u->nsecs), u->name); } static inline void pr_name_and_units(struct printbuf *out, const char *name, u64 ns) { - prt_str(out, name); - prt_tab(out); + prt_printf(out, "%s\t", name); bch2_pr_time_units_aligned(out, ns); prt_newline(out); } @@ -389,12 +386,8 @@ void bch2_time_stats_to_text(struct printbuf *out, struct bch2_time_stats *stats } printbuf_tabstop_push(out, out->indent + TABSTOP_SIZE); - prt_printf(out, "count:"); - prt_tab(out); - prt_printf(out, "%llu ", - stats->duration_stats.n); + prt_printf(out, "count:\t%llu\n", stats->duration_stats.n); printbuf_tabstop_pop(out); - prt_newline(out); printbuf_tabstops_reset(out); @@ -403,13 +396,8 @@ void bch2_time_stats_to_text(struct printbuf *out, struct bch2_time_stats *stats printbuf_tabstop_push(out, 0); printbuf_tabstop_push(out, TABSTOP_SIZE + 2); - prt_tab(out); - prt_printf(out, "since mount"); - prt_tab_rjust(out); - prt_tab(out); + prt_printf(out, "\tsince mount\r\trecent\r\n"); prt_printf(out, "recent"); - prt_tab_rjust(out); - prt_newline(out); printbuf_tabstops_reset(out); printbuf_tabstop_push(out, out->indent + 20); @@ -417,23 +405,20 @@ void bch2_time_stats_to_text(struct printbuf *out, struct bch2_time_stats *stats printbuf_tabstop_push(out, 2); printbuf_tabstop_push(out, TABSTOP_SIZE); - prt_printf(out, "duration of events"); - prt_newline(out); + prt_printf(out, "duration of events\n"); printbuf_indent_add(out, 2); pr_name_and_units(out, "min:", stats->min_duration); pr_name_and_units(out, "max:", stats->max_duration); pr_name_and_units(out, "total:", stats->total_duration); - prt_printf(out, "mean:"); - prt_tab(out); + prt_printf(out, "mean:\t"); bch2_pr_time_units_aligned(out, d_mean); prt_tab(out); bch2_pr_time_units_aligned(out, mean_and_variance_weighted_get_mean(stats->duration_stats_weighted, TIME_STATS_MV_WEIGHT)); prt_newline(out); - prt_printf(out, "stddev:"); - prt_tab(out); + prt_printf(out, "stddev:\t"); bch2_pr_time_units_aligned(out, d_stddev); prt_tab(out); bch2_pr_time_units_aligned(out, mean_and_variance_weighted_get_stddev(stats->duration_stats_weighted, TIME_STATS_MV_WEIGHT)); @@ -441,22 +426,19 @@ void bch2_time_stats_to_text(struct printbuf *out, struct bch2_time_stats *stats printbuf_indent_sub(out, 2); prt_newline(out); - prt_printf(out, "time between events"); - prt_newline(out); + prt_printf(out, "time between events\n"); printbuf_indent_add(out, 2); pr_name_and_units(out, "min:", stats->min_freq); pr_name_and_units(out, "max:", stats->max_freq); - prt_printf(out, "mean:"); - prt_tab(out); + prt_printf(out, "mean:\t"); bch2_pr_time_units_aligned(out, f_mean); prt_tab(out); bch2_pr_time_units_aligned(out, mean_and_variance_weighted_get_mean(stats->freq_stats_weighted, TIME_STATS_MV_WEIGHT)); prt_newline(out); - prt_printf(out, "stddev:"); - prt_tab(out); + prt_printf(out, "stddev:\t"); bch2_pr_time_units_aligned(out, f_stddev); prt_tab(out); bch2_pr_time_units_aligned(out, mean_and_variance_weighted_get_stddev(stats->freq_stats_weighted, TIME_STATS_MV_WEIGHT)); @@ -589,40 +571,31 @@ void bch2_pd_controller_debug_to_text(struct printbuf *out, struct bch_pd_contro if (!out->nr_tabstops) printbuf_tabstop_push(out, 20); - prt_printf(out, "rate:"); - prt_tab(out); + prt_printf(out, "rate:\t"); prt_human_readable_s64(out, pd->rate.rate); prt_newline(out); - prt_printf(out, "target:"); - prt_tab(out); + prt_printf(out, "target:\t"); prt_human_readable_u64(out, pd->last_target); prt_newline(out); - prt_printf(out, "actual:"); - prt_tab(out); + prt_printf(out, "actual:\t"); prt_human_readable_u64(out, pd->last_actual); prt_newline(out); - prt_printf(out, "proportional:"); - prt_tab(out); + prt_printf(out, "proportional:\t"); prt_human_readable_s64(out, pd->last_proportional); prt_newline(out); - prt_printf(out, "derivative:"); - prt_tab(out); + prt_printf(out, "derivative:\t"); prt_human_readable_s64(out, pd->last_derivative); prt_newline(out); - prt_printf(out, "change:"); - prt_tab(out); + prt_printf(out, "change:\t"); prt_human_readable_s64(out, pd->last_change); prt_newline(out); - prt_printf(out, "next io:"); - prt_tab(out); - prt_printf(out, "%llims", div64_s64(pd->rate.next - local_clock(), NSEC_PER_MSEC)); - prt_newline(out); + prt_printf(out, "next io:\t%llims\n", div64_s64(pd->rate.next - local_clock(), NSEC_PER_MSEC)); } /* misc: */ -- cgit v1.2.3-59-g8ed1b From 9089376f709e726e9bd2914284b9b98af65de749 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 14:41:17 -0400 Subject: bcachefs: bch2_btree_node_header_to_text() better btree node read path error messages Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_io.c | 27 ++++++++++++++++++++------- 1 file changed, 20 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index fc046c9ccb45..ed80556b80fe 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -23,6 +23,18 @@ #include +static void bch2_btree_node_header_to_text(struct printbuf *out, struct btree_node *bn) +{ + prt_printf(out, "btree=%s l=%u seq %llux\n", + bch2_btree_id_str(BTREE_NODE_ID(bn)), + (unsigned) BTREE_NODE_LEVEL(bn), bn->keys.seq); + prt_str(out, "min: "); + bch2_bpos_to_text(out, bn->min_key); + prt_newline(out); + prt_str(out, "max: "); + bch2_bpos_to_text(out, bn->max_key); +} + void bch2_btree_node_io_unlock(struct btree *b) { EBUG_ON(!btree_node_write_in_flight(b)); @@ -1021,18 +1033,19 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, -BCH_ERR_btree_node_read_err_must_retry, c, ca, b, NULL, btree_node_bad_seq, - "got wrong btree node (want %llx got %llx)\n" - "got btree %s level %llu pos %s", - bp->seq, b->data->keys.seq, - bch2_btree_id_str(BTREE_NODE_ID(b->data)), - BTREE_NODE_LEVEL(b->data), - buf.buf); + "got wrong btree node: got\n%s", + (printbuf_reset(&buf), + bch2_btree_node_header_to_text(&buf, b->data), + buf.buf)); } else { btree_err_on(!b->data->keys.seq, -BCH_ERR_btree_node_read_err_must_retry, c, ca, b, NULL, btree_node_bad_seq, - "bad btree header: seq 0"); + "bad btree header: seq 0\n%s", + (printbuf_reset(&buf), + bch2_btree_node_header_to_text(&buf, b->data), + buf.buf)); } while (b->written < (ptr_written ?: btree_sectors(c))) { -- cgit v1.2.3-59-g8ed1b From 6ab71b4a8e281f30a3dfd6f7831d9a2c67e162e4 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 00:04:31 -0400 Subject: bcachefs: bch2_journal_keys_dump() debug helper Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_journal_iter.c | 17 +++++++++++++++++ fs/bcachefs/btree_journal_iter.h | 2 ++ 2 files changed, 19 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index 1e8cf49a6935..332dbf164929 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -623,3 +623,20 @@ void bch2_shoot_down_journal_keys(struct bch_fs *c, enum btree_id btree, keys->data[dst++] = *i; keys->nr = keys->gap = dst; } + +void bch2_journal_keys_dump(struct bch_fs *c) +{ + struct journal_keys *keys = &c->journal_keys; + struct printbuf buf = PRINTBUF; + + pr_info("%zu keys:", keys->nr); + + move_gap(keys, keys->nr); + + darray_for_each(*keys, i) { + printbuf_reset(&buf); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(i->k)); + pr_err("%s l=%u %s", bch2_btree_id_str(i->btree_id), i->level, buf.buf); + } + printbuf_exit(&buf); +} diff --git a/fs/bcachefs/btree_journal_iter.h b/fs/bcachefs/btree_journal_iter.h index af25046ebcaa..1ba4a79b0ef9 100644 --- a/fs/bcachefs/btree_journal_iter.h +++ b/fs/bcachefs/btree_journal_iter.h @@ -70,4 +70,6 @@ void bch2_shoot_down_journal_keys(struct bch_fs *, enum btree_id, unsigned, unsigned, struct bpos, struct bpos); +void bch2_journal_keys_dump(struct bch_fs *); + #endif /* _BCACHEFS_BTREE_JOURNAL_ITER_H */ -- cgit v1.2.3-59-g8ed1b From ac01928b8e9688aca3c23134d4c889c4839c360a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 16:16:49 -0400 Subject: bcachefs: bch2_hash_lookup() now returns bkey_s_c small cleanup Signed-off-by: Kent Overstreet --- fs/bcachefs/acl.c | 37 +++++++++++-------------------------- fs/bcachefs/dirent.c | 30 +++++++++--------------------- fs/bcachefs/fs.c | 10 +++------- fs/bcachefs/fsck.c | 8 ++++---- fs/bcachefs/str_hash.h | 27 +++++++++++++-------------- fs/bcachefs/xattr.c | 25 +++++++++---------------- 6 files changed, 49 insertions(+), 88 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/acl.c b/fs/bcachefs/acl.c index 5c180fdc3efb..b4b112e33be4 100644 --- a/fs/bcachefs/acl.c +++ b/fs/bcachefs/acl.c @@ -282,18 +282,12 @@ struct posix_acl *bch2_get_acl(struct mnt_idmap *idmap, struct btree_trans *trans = bch2_trans_get(c); struct btree_iter iter = { NULL }; struct posix_acl *acl = NULL; - struct bkey_s_c k; - int ret; retry: bch2_trans_begin(trans); - ret = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, - &hash, inode_inum(inode), &search, 0); - if (ret) - goto err; - - k = bch2_btree_iter_peek_slot(&iter); - ret = bkey_err(k); + struct bkey_s_c k = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, + &hash, inode_inum(inode), &search, 0); + int ret = bkey_err(k); if (ret) goto err; @@ -414,39 +408,30 @@ int bch2_acl_chmod(struct btree_trans *trans, subvol_inum inum, struct bch_hash_info hash_info = bch2_hash_info_init(trans->c, inode); struct xattr_search_key search = X_SEARCH(KEY_TYPE_XATTR_INDEX_POSIX_ACL_ACCESS, "", 0); struct btree_iter iter; - struct bkey_s_c_xattr xattr; - struct bkey_i_xattr *new; struct posix_acl *acl = NULL; - struct bkey_s_c k; - int ret; - ret = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, + struct bkey_s_c k = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, &hash_info, inum, &search, BTREE_ITER_INTENT); + int ret = bkey_err(k); if (ret) return bch2_err_matches(ret, ENOENT) ? 0 : ret; - k = bch2_btree_iter_peek_slot(&iter); - ret = bkey_err(k); - if (ret) - goto err; - xattr = bkey_s_c_to_xattr(k); + struct bkey_s_c_xattr xattr = bkey_s_c_to_xattr(k); acl = bch2_acl_from_disk(trans, xattr_val(xattr.v), le16_to_cpu(xattr.v->x_val_len)); ret = PTR_ERR_OR_ZERO(acl); - if (IS_ERR_OR_NULL(acl)) + if (ret) goto err; - ret = allocate_dropping_locks_errcode(trans, - __posix_acl_chmod(&acl, _gfp, mode)); + ret = allocate_dropping_locks_errcode(trans, __posix_acl_chmod(&acl, _gfp, mode)); if (ret) goto err; - new = bch2_acl_to_xattr(trans, acl, ACL_TYPE_ACCESS); - if (IS_ERR(new)) { - ret = PTR_ERR(new); + struct bkey_i_xattr *new = bch2_acl_to_xattr(trans, acl, ACL_TYPE_ACCESS); + ret = PTR_ERR_OR_ZERO(new); + if (ret) goto err; - } new->k.p = iter.pos; ret = bch2_trans_update(trans, &iter, &new->k_i, 0); diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c index d37bd07afbfe..d4e3ca5045d6 100644 --- a/fs/bcachefs/dirent.c +++ b/fs/bcachefs/dirent.c @@ -301,13 +301,9 @@ int bch2_dirent_rename(struct btree_trans *trans, memset(dst_inum, 0, sizeof(*dst_inum)); /* Lookup src: */ - ret = bch2_hash_lookup(trans, &src_iter, bch2_dirent_hash_desc, - src_hash, src_dir, src_name, - BTREE_ITER_INTENT); - if (ret) - goto out; - - old_src = bch2_btree_iter_peek_slot(&src_iter); + old_src = bch2_hash_lookup(trans, &src_iter, bch2_dirent_hash_desc, + src_hash, src_dir, src_name, + BTREE_ITER_INTENT); ret = bkey_err(old_src); if (ret) goto out; @@ -329,13 +325,9 @@ int bch2_dirent_rename(struct btree_trans *trans, if (ret) goto out; } else { - ret = bch2_hash_lookup(trans, &dst_iter, bch2_dirent_hash_desc, - dst_hash, dst_dir, dst_name, - BTREE_ITER_INTENT); - if (ret) - goto out; - - old_dst = bch2_btree_iter_peek_slot(&dst_iter); + old_dst = bch2_hash_lookup(trans, &dst_iter, bch2_dirent_hash_desc, + dst_hash, dst_dir, dst_name, + BTREE_ITER_INTENT); ret = bkey_err(old_dst); if (ret) goto out; @@ -479,13 +471,9 @@ int bch2_dirent_lookup_trans(struct btree_trans *trans, const struct qstr *name, subvol_inum *inum, unsigned flags) { - int ret = bch2_hash_lookup(trans, iter, bch2_dirent_hash_desc, - hash_info, dir, name, flags); - if (ret) - return ret; - - struct bkey_s_c k = bch2_btree_iter_peek_slot(iter); - ret = bkey_err(k); + struct bkey_s_c k = bch2_hash_lookup(trans, iter, bch2_dirent_hash_desc, + hash_info, dir, name, flags); + int ret = bkey_err(k); if (ret) goto err; diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 6f114803c6f2..3d15275909bd 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -377,16 +377,12 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, struct btree_iter dirent_iter = {}; subvol_inum inum = {}; - int ret = bch2_hash_lookup(trans, &dirent_iter, bch2_dirent_hash_desc, - dir_hash_info, dir, name, 0); + struct bkey_s_c k = bch2_hash_lookup(trans, &dirent_iter, bch2_dirent_hash_desc, + dir_hash_info, dir, name, 0); + int ret = bkey_err(k); if (ret) return ERR_PTR(ret); - struct bkey_s_c k = bch2_btree_iter_peek_slot(&dirent_iter); - ret = bkey_err(k); - if (ret) - goto err; - ret = bch2_dirent_read_target(trans, dir, bkey_s_c_to_dirent(k), &inum); if (ret > 0) ret = -ENOENT; diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 8e2010212cc3..caef81aa618c 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -127,13 +127,13 @@ static int lookup_dirent_in_snapshot(struct btree_trans *trans, u64 *target, unsigned *type, u32 snapshot) { struct btree_iter iter; - struct bkey_s_c_dirent d; - int ret = bch2_hash_lookup_in_snapshot(trans, &iter, bch2_dirent_hash_desc, - &hash_info, dir, name, 0, snapshot); + struct bkey_s_c k = bch2_hash_lookup_in_snapshot(trans, &iter, bch2_dirent_hash_desc, + &hash_info, dir, name, 0, snapshot); + int ret = bkey_err(k); if (ret) return ret; - d = bkey_s_c_to_dirent(bch2_btree_iter_peek_slot(&iter)); + struct bkey_s_c_dirent d = bkey_s_c_to_dirent(bch2_btree_iter_peek_slot(&iter)); *target = le64_to_cpu(d.v->d_inum); *type = d.v->d_type; bch2_trans_iter_exit(trans, &iter); diff --git a/fs/bcachefs/str_hash.h b/fs/bcachefs/str_hash.h index 3976f80721bf..11f5f5d27d23 100644 --- a/fs/bcachefs/str_hash.h +++ b/fs/bcachefs/str_hash.h @@ -159,7 +159,7 @@ static inline bool is_visible_key(struct bch_hash_desc desc, subvol_inum inum, s desc.is_visible(inum, k)); } -static __always_inline int +static __always_inline struct bkey_s_c bch2_hash_lookup_in_snapshot(struct btree_trans *trans, struct btree_iter *iter, const struct bch_hash_desc desc, @@ -176,7 +176,7 @@ bch2_hash_lookup_in_snapshot(struct btree_trans *trans, BTREE_ITER_SLOTS|flags, k, ret) { if (is_visible_key(desc, inum, k)) { if (!desc.cmp_key(k, key)) - return 0; + return k; } else if (k.k->type == KEY_TYPE_hash_whiteout) { ; } else { @@ -186,10 +186,10 @@ bch2_hash_lookup_in_snapshot(struct btree_trans *trans, } bch2_trans_iter_exit(trans, iter); - return ret ?: -BCH_ERR_ENOENT_str_hash_lookup; + return bkey_s_c_err(ret ?: -BCH_ERR_ENOENT_str_hash_lookup); } -static __always_inline int +static __always_inline struct bkey_s_c bch2_hash_lookup(struct btree_trans *trans, struct btree_iter *iter, const struct bch_hash_desc desc, @@ -198,8 +198,11 @@ bch2_hash_lookup(struct btree_trans *trans, unsigned flags) { u32 snapshot; - return bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot) ?: - bch2_hash_lookup_in_snapshot(trans, iter, desc, info, inum, key, flags, snapshot); + int ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); + if (ret) + return bkey_s_c_err(ret); + + return bch2_hash_lookup_in_snapshot(trans, iter, desc, info, inum, key, flags, snapshot); } static __always_inline int @@ -369,14 +372,10 @@ int bch2_hash_delete(struct btree_trans *trans, subvol_inum inum, const void *key) { struct btree_iter iter; - int ret; - - ret = bch2_hash_lookup(trans, &iter, desc, info, inum, key, - BTREE_ITER_INTENT); - if (ret) - return ret; - - ret = bch2_hash_delete_at(trans, desc, info, &iter, 0); + struct bkey_s_c k = bch2_hash_lookup(trans, &iter, desc, info, inum, key, + BTREE_ITER_INTENT); + int ret = bkey_err(k) ?: + bch2_hash_delete_at(trans, desc, info, &iter, 0); bch2_trans_iter_exit(trans, &iter); return ret; } diff --git a/fs/bcachefs/xattr.c b/fs/bcachefs/xattr.c index 4b3a2df2da32..d787f1d4cc78 100644 --- a/fs/bcachefs/xattr.c +++ b/fs/bcachefs/xattr.c @@ -144,21 +144,13 @@ static int bch2_xattr_get_trans(struct btree_trans *trans, struct bch_inode_info struct bch_hash_info hash = bch2_hash_info_init(trans->c, &inode->ei_inode); struct xattr_search_key search = X_SEARCH(type, name, strlen(name)); struct btree_iter iter; - struct bkey_s_c_xattr xattr; - struct bkey_s_c k; - int ret; - - ret = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, &hash, - inode_inum(inode), &search, 0); - if (ret) - goto err1; - - k = bch2_btree_iter_peek_slot(&iter); - ret = bkey_err(k); + struct bkey_s_c k = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, &hash, + inode_inum(inode), &search, 0); + int ret = bkey_err(k); if (ret) - goto err2; + return ret; - xattr = bkey_s_c_to_xattr(k); + struct bkey_s_c_xattr xattr = bkey_s_c_to_xattr(k); ret = le16_to_cpu(xattr.v->x_val_len); if (buffer) { if (ret > size) @@ -166,10 +158,8 @@ static int bch2_xattr_get_trans(struct btree_trans *trans, struct bch_inode_info else memcpy(buffer, xattr_val(xattr.v), ret); } -err2: bch2_trans_iter_exit(trans, &iter); -err1: - return ret < 0 && bch2_err_matches(ret, ENOENT) ? -ENODATA : ret; + return ret; } int bch2_xattr_set(struct btree_trans *trans, subvol_inum inum, @@ -365,6 +355,9 @@ static int bch2_xattr_get_handler(const struct xattr_handler *handler, int ret = bch2_trans_do(c, NULL, NULL, 0, bch2_xattr_get_trans(trans, inode, name, buffer, size, handler->flags)); + if (ret < 0 && bch2_err_matches(ret, ENOENT)) + ret = -ENODATA; + return bch2_err_class(ret); } -- cgit v1.2.3-59-g8ed1b From 5577881455cdfe0c39781e43215138faf5ed2118 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 5 Apr 2024 16:21:39 -0400 Subject: bcachefs: add btree_node_merging_disabled debug param Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 2 ++ fs/bcachefs/btree_update_interior.h | 3 +++ 2 files changed, 5 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 91c3c1fef233..6366a6a90944 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -359,6 +359,8 @@ do { \ #define BCH_DEBUG_PARAMS_ALWAYS() \ BCH_DEBUG_PARAM(key_merging_disabled, \ "Disables merging of extents") \ + BCH_DEBUG_PARAM(btree_node_merging_disabled, \ + "Disables merging of btree nodes") \ BCH_DEBUG_PARAM(btree_gc_always_rewrite, \ "Causes mark and sweep to compact and rewrite every " \ "btree node it traverses") \ diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index c1a479ebaad1..ca1a3be43af8 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -144,6 +144,9 @@ static inline int bch2_foreground_maybe_merge_sibling(struct btree_trans *trans, EBUG_ON(!btree_node_locked(path, level)); + if (bch2_btree_node_merging_disabled) + return 0; + b = path->l[level].b; if (b->sib_u64s[sib] > trans->c->btree_foreground_merge_threshold) return 0; -- cgit v1.2.3-59-g8ed1b From 00589cadb1fb12548e096a29b1b9c0ccb313f2fa Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 5 Apr 2024 21:32:06 -0400 Subject: bcachefs: bch2_btree_path_to_text() Long form version of bch2_btree_path_to_text() - useful in error messages and tracepoints. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 52 ++++++++++++++++++++++++++++++++++++++++----- fs/bcachefs/btree_iter.h | 1 + fs/bcachefs/btree_locking.c | 5 +++++ 3 files changed, 53 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 080cb50168b0..ca3b6a0ea1f8 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1426,23 +1426,63 @@ void bch2_dump_trans_updates(struct btree_trans *trans) printbuf_exit(&buf); } -static void bch2_btree_path_to_text(struct printbuf *out, struct btree_trans *trans, btree_path_idx_t path_idx) +static void bch2_btree_path_to_text_short(struct printbuf *out, struct btree_trans *trans, btree_path_idx_t path_idx) { struct btree_path *path = trans->paths + path_idx; - prt_printf(out, "path: idx %2u ref %u:%u %c %c btree=%s l=%u pos ", + prt_printf(out, "path: idx %2u ref %u:%u %c %c %c btree=%s l=%u pos ", path_idx, path->ref, path->intent_ref, path->preserve ? 'P' : ' ', path->should_be_locked ? 'S' : ' ', + path->cached ? 'C' : 'B', bch2_btree_id_str(path->btree_id), path->level); bch2_bpos_to_text(out, path->pos); - prt_printf(out, " locks %u", path->nodes_locked); #ifdef TRACK_PATH_ALLOCATED prt_printf(out, " %pS", (void *) path->ip_allocated); #endif +} + +static const char *btree_node_locked_str(enum btree_node_locked_type t) +{ + switch (t) { + case BTREE_NODE_UNLOCKED: + return "unlocked"; + case BTREE_NODE_READ_LOCKED: + return "read"; + case BTREE_NODE_INTENT_LOCKED: + return "intent"; + case BTREE_NODE_WRITE_LOCKED: + return "write"; + default: + return NULL; + } +} + +void bch2_btree_path_to_text(struct printbuf *out, struct btree_trans *trans, btree_path_idx_t path_idx) +{ + bch2_btree_path_to_text_short(out, trans, path_idx); + + struct btree_path *path = trans->paths + path_idx; + + prt_printf(out, " uptodate %u locks_want %u", path->uptodate, path->locks_want); prt_newline(out); + + printbuf_indent_add(out, 2); + for (unsigned l = 0; l < BTREE_MAX_DEPTH; l++) { + prt_printf(out, "l=%u locks %s seq %u node ", l, + btree_node_locked_str(btree_node_locked_type(path, l)), + path->l[l].lock_seq); + + int ret = PTR_ERR_OR_ZERO(path->l[l].b); + if (ret) + prt_str(out, bch2_err_str(ret)); + else + prt_printf(out, "%px", path->l[l].b); + prt_newline(out); + } + printbuf_indent_sub(out, 2); } static noinline __cold @@ -1454,8 +1494,10 @@ void __bch2_trans_paths_to_text(struct printbuf *out, struct btree_trans *trans, if (!nosort) btree_trans_sort_paths(trans); - trans_for_each_path_idx_inorder(trans, iter) - bch2_btree_path_to_text(out, trans, iter.path_idx); + trans_for_each_path_idx_inorder(trans, iter) { + bch2_btree_path_to_text_short(out, trans, iter.path_idx); + prt_newline(out); + } } noinline __cold diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 1c70836dd7cc..fcf2742b6ad9 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -861,6 +861,7 @@ __bch2_btree_iter_peek_and_restart(struct btree_trans *trans, }) void bch2_trans_updates_to_text(struct printbuf *, struct btree_trans *); +void bch2_btree_path_to_text(struct printbuf *, struct btree_trans *, btree_path_idx_t); void bch2_trans_paths_to_text(struct printbuf *, struct btree_trans *); void bch2_dump_trans_updates(struct btree_trans *); void bch2_dump_trans_paths_updates(struct btree_trans *); diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index 4ee4855395b9..8d1c4f78db5e 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -836,6 +836,11 @@ void bch2_btree_path_verify_locks(struct btree_path *path) { unsigned l; + /* + * A path may be uptodate and yet have nothing locked if and only if + * there is no node at path->level, which generally means we were + * iterating over all nodes and got to the end of the btree + */ if (!path->nodes_locked) { BUG_ON(path->uptodate == BTREE_ITER_UPTODATE && btree_path_node(path, path->level)); -- cgit v1.2.3-59-g8ed1b From 497c982f057d3a20af4df313d997e81ca903ce4e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 20 Feb 2024 21:17:15 -0500 Subject: bcachefs: New assertion for writing to the journal after shutdown Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update.c | 2 +- fs/bcachefs/journal.c | 6 ++++-- fs/bcachefs/journal.h | 4 ++-- fs/bcachefs/journal_reclaim.c | 2 +- fs/bcachefs/journal_types.h | 2 +- fs/bcachefs/super.c | 1 + 6 files changed, 10 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update.c b/fs/bcachefs/btree_update.c index 8e47e260eba5..c46ee4c813ae 100644 --- a/fs/bcachefs/btree_update.c +++ b/fs/bcachefs/btree_update.c @@ -852,7 +852,7 @@ __bch2_fs_log_msg(struct bch_fs *c, unsigned commit_flags, const char *fmt, if (ret) goto err; - if (!test_bit(JOURNAL_STARTED, &c->journal.flags)) { + if (!test_bit(JOURNAL_RUNNING, &c->journal.flags)) { ret = darray_make_room(&c->journal.early_journal_entries, jset_u64s(u64s)); if (ret) goto err; diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index c40b5c40d63f..d302d8e83678 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -1177,12 +1177,14 @@ void bch2_fs_journal_stop(struct journal *j) bch2_journal_meta(j); journal_quiesce(j); + cancel_delayed_work_sync(&j->write_work); BUG_ON(!bch2_journal_error(j) && test_bit(JOURNAL_REPLAY_DONE, &j->flags) && j->last_empty_seq != journal_cur_seq(j)); - cancel_delayed_work_sync(&j->write_work); + if (!bch2_journal_error(j)) + clear_bit(JOURNAL_RUNNING, &j->flags); } int bch2_fs_journal_start(struct journal *j, u64 cur_seq) @@ -1256,7 +1258,7 @@ int bch2_fs_journal_start(struct journal *j, u64 cur_seq) spin_lock(&j->lock); - set_bit(JOURNAL_STARTED, &j->flags); + set_bit(JOURNAL_RUNNING, &j->flags); j->last_flush_write = jiffies; j->reservations.idx = j->reservations.unwritten_idx = journal_cur_seq(j); diff --git a/fs/bcachefs/journal.h b/fs/bcachefs/journal.h index 7c7528f839c5..4b8c709bc317 100644 --- a/fs/bcachefs/journal.h +++ b/fs/bcachefs/journal.h @@ -372,7 +372,7 @@ static inline int bch2_journal_res_get(struct journal *j, struct journal_res *re int ret; EBUG_ON(res->ref); - EBUG_ON(!test_bit(JOURNAL_STARTED, &j->flags)); + EBUG_ON(!test_bit(JOURNAL_RUNNING, &j->flags)); res->u64s = u64s; @@ -418,7 +418,7 @@ struct bch_dev; static inline void bch2_journal_set_replay_done(struct journal *j) { - BUG_ON(!test_bit(JOURNAL_STARTED, &j->flags)); + BUG_ON(!test_bit(JOURNAL_RUNNING, &j->flags)); set_bit(JOURNAL_REPLAY_DONE, &j->flags); } diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index 04a577848b01..a0c9f7ac611d 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -833,7 +833,7 @@ bool bch2_journal_flush_pins(struct journal *j, u64 seq_to_flush) /* time_stats this */ bool did_work = false; - if (!test_bit(JOURNAL_STARTED, &j->flags)) + if (!test_bit(JOURNAL_RUNNING, &j->flags)) return false; closure_wait_event(&j->async_wait, diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index b5161b5d76a0..08debe6bfeef 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -131,7 +131,7 @@ enum journal_space_from { enum journal_flags { JOURNAL_REPLAY_DONE, - JOURNAL_STARTED, + JOURNAL_RUNNING, JOURNAL_MAY_SKIP_FLUSH, JOURNAL_NEED_FLUSH_WRITE, JOURNAL_SPACE_LOW, diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index dddf57ec4511..ca2f3debd7fd 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -468,6 +468,7 @@ static int __bch2_fs_read_write(struct bch_fs *c, bool early) * at least one non-flush write in the journal or recovery will fail: */ set_bit(JOURNAL_NEED_FLUSH_WRITE, &c->journal.flags); + set_bit(JOURNAL_RUNNING, &c->journal.flags); for_each_rw_member(c, ca) bch2_dev_allocator_add(c, ca); -- cgit v1.2.3-59-g8ed1b From 19391b92947c75267cdacb1acefe8ad0d278a9dd Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 28 Mar 2024 00:39:11 -0400 Subject: bcachefs: allow for custom action in fsck error messages Be more explicit to the user about what we're doing. Signed-off-by: Kent Overstreet --- fs/bcachefs/error.c | 59 +++++++++++++++++++++++++++++++++++++++--------- fs/bcachefs/journal_io.c | 2 +- fs/bcachefs/snapshot.c | 2 +- 3 files changed, 50 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/error.c b/fs/bcachefs/error.c index 82a6656c941c..c66eeffcd7f2 100644 --- a/fs/bcachefs/error.c +++ b/fs/bcachefs/error.c @@ -176,6 +176,21 @@ static struct fsck_err_state *fsck_err_get(struct bch_fs *c, const char *fmt) return s; } +/* s/fix?/fixing/ s/recreate?/recreating/ */ +static void prt_actioning(struct printbuf *out, const char *action) +{ + unsigned len = strlen(action); + + BUG_ON(action[len - 1] != '?'); + --len; + + if (action[len - 1] == 'e') + --len; + + prt_bytes(out, action, len); + prt_str(out, "ing"); +} + int bch2_fsck_err(struct bch_fs *c, enum bch_fsck_flags flags, enum bch_sb_error_id err, @@ -186,6 +201,7 @@ int bch2_fsck_err(struct bch_fs *c, bool print = true, suppressing = false, inconsistent = false; struct printbuf buf = PRINTBUF, *out = &buf; int ret = -BCH_ERR_fsck_ignore; + const char *action_orig = "fix?", *action = action_orig; if ((flags & FSCK_CAN_FIX) && test_bit(err, c->sb.errors_silent)) @@ -197,6 +213,19 @@ int bch2_fsck_err(struct bch_fs *c, prt_vprintf(out, fmt, args); va_end(args); + /* Custom fix/continue/recreate/etc.? */ + if (out->buf[out->pos - 1] == '?') { + const char *p = strrchr(out->buf, ','); + if (p) { + out->pos = p - out->buf; + action = kstrdup(p + 2, GFP_KERNEL); + if (!action) { + ret = -ENOMEM; + goto err; + } + } + } + mutex_lock(&c->fsck_error_msgs_lock); s = fsck_err_get(c, fmt); if (s) { @@ -208,12 +237,16 @@ int bch2_fsck_err(struct bch_fs *c, if (s->last_msg && !strcmp(buf.buf, s->last_msg)) { ret = s->ret; mutex_unlock(&c->fsck_error_msgs_lock); - printbuf_exit(&buf); - return ret; + goto err; } kfree(s->last_msg); s->last_msg = kstrdup(buf.buf, GFP_KERNEL); + if (!s->last_msg) { + mutex_unlock(&c->fsck_error_msgs_lock); + ret = -ENOMEM; + goto err; + } if (c->opts.ratelimit_errors && !(flags & FSCK_NO_RATELIMIT) && @@ -239,7 +272,8 @@ int bch2_fsck_err(struct bch_fs *c, inconsistent = true; ret = -BCH_ERR_fsck_errors_not_fixed; } else if (flags & FSCK_CAN_FIX) { - prt_str(out, ", fixing"); + prt_str(out, ", "); + prt_actioning(out, action); ret = -BCH_ERR_fsck_fix; } else { prt_str(out, ", continuing"); @@ -254,16 +288,16 @@ int bch2_fsck_err(struct bch_fs *c, : c->opts.fix_errors; if (fix == FSCK_FIX_ask) { - int ask; + prt_str(out, ", "); + prt_str(out, action); - prt_str(out, ": fix?"); if (bch2_fs_stdio_redirect(c)) bch2_print(c, "%s", out->buf); else bch2_print_string_as_lines(KERN_ERR, out->buf); print = false; - ask = bch2_fsck_ask_yn(c); + int ask = bch2_fsck_ask_yn(c); if (ask >= YN_ALLNO && s) s->fix = ask == YN_ALLNO @@ -276,10 +310,12 @@ int bch2_fsck_err(struct bch_fs *c, } else if (fix == FSCK_FIX_yes || (c->opts.nochanges && !(flags & FSCK_CAN_IGNORE))) { - prt_str(out, ", fixing"); + prt_str(out, ", "); + prt_actioning(out, action); ret = -BCH_ERR_fsck_fix; } else { - prt_str(out, ", not fixing"); + prt_str(out, ", not "); + prt_actioning(out, action); } } else if (flags & FSCK_NEED_FSCK) { prt_str(out, " (run fsck to correct)"); @@ -311,8 +347,6 @@ int bch2_fsck_err(struct bch_fs *c, mutex_unlock(&c->fsck_error_msgs_lock); - printbuf_exit(&buf); - if (inconsistent) bch2_inconsistent_error(c); @@ -322,7 +356,10 @@ int bch2_fsck_err(struct bch_fs *c, set_bit(BCH_FS_errors_not_fixed, &c->flags); set_bit(BCH_FS_error, &c->flags); } - +err: + if (action != action_orig) + kfree(action); + printbuf_exit(&buf); return ret; } diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index eb1f9d6f5a19..cd14636bc2ab 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1366,7 +1366,7 @@ int bch2_journal_read(struct bch_fs *c, fsck_err(c, journal_entries_missing, "journal entries %llu-%llu missing! (replaying %llu-%llu)\n" " prev at %s\n" - " next at %s", + " next at %s, continue?", missing_start, missing_end, *last_seq, *blacklist_seq - 1, buf1.buf, buf2.buf); diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 544322d5c251..c415ea0a649b 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -1018,7 +1018,7 @@ int bch2_reconstruct_snapshots(struct bch_fs *c) darray_for_each(*t, id) { if (fsck_err_on(!bch2_snapshot_equiv(c, *id), c, snapshot_node_missing, - "snapshot node %u from tree %s missing", *id, buf.buf)) { + "snapshot node %u from tree %s missing, recreate?", *id, buf.buf)) { if (t->nr > 1) { bch_err(c, "cannot reconstruct snapshot trees with multiple nodes"); ret = -BCH_ERR_fsck_repair_unimplemented; -- cgit v1.2.3-59-g8ed1b From 4dcd90b6d1b76b87722059db0f368aa48e7703da Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 30 Mar 2024 22:42:29 -0400 Subject: bcachefs: Don't read journal just for fsck reading the journal can take a decent amount of time compared to the rest of fsck, let's only read it when required. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 08c0422a722d..9cf262f1a3cd 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -658,7 +658,7 @@ int bch2_fs_recovery(struct bch_fs *c) goto err; } - if (!c->sb.clean || c->opts.fsck || c->opts.retain_recovery_info) { + if (!c->sb.clean || c->opts.retain_recovery_info) { struct genradix_iter iter; struct journal_replay **i; -- cgit v1.2.3-59-g8ed1b From be31bf439c2154e4c1174186cf2bd6fd1e6b7a88 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 21 Mar 2024 23:21:56 -0400 Subject: bcachefs: When traversing to interior nodes, propagate result to paths to same leaf node Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index ca3b6a0ea1f8..39c89914cb53 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1162,6 +1162,7 @@ int bch2_btree_path_traverse_one(struct btree_trans *trans, goto out_uptodate; path->level = btree_path_up_until_good_node(trans, path, 0); + unsigned max_level = path->level; EBUG_ON(btree_path_node(path, path->level) && !btree_node_locked(path, path->level)); @@ -1192,6 +1193,16 @@ int bch2_btree_path_traverse_one(struct btree_trans *trans, goto out; } } + + if (unlikely(max_level > path->level)) { + struct btree_path *linked; + unsigned iter; + + trans_for_each_path_with_node(trans, path_l(path)->b, linked, iter) + for (unsigned j = path->level + 1; j < max_level; j++) + linked->l[j] = path->l[j]; + } + out_uptodate: path->uptodate = BTREE_ITER_UPTODATE; out: -- cgit v1.2.3-59-g8ed1b From 0ddb5f0854a8f42d949f5d8a49980d8174f266a3 Mon Sep 17 00:00:00 2001 From: Kuan-Wei Chiu Date: Sun, 7 Apr 2024 11:39:04 +0800 Subject: bcachefs: Optimize eytzinger0_sort() with bottom-up heapsort This optimization reduces the average number of comparisons required from 2*n*log2(n) - 3*n + o(n) to n*log2(n) + 0.37*n + o(n). When n is sufficiently large, it results in approximately 50% fewer comparisons. Currently, eytzinger0_sort employs the textbook version of heapsort, where during the heapify process, each level requires two comparisons to determine the maximum among three elements. In contrast, the bottom-up heapsort, during heapify, only compares two children at each level until reaching a leaf node. Then, it backtracks from the leaf node to find the correct position. Since heapify typically continues until very close to the leaf node, the standard heapify requires about 2*log2(n) comparisons, while the bottom-up variant only needs log2(n) comparisons. The experimental data presented below is based on an array generated by get_random_u32(). | N | comparisons(old) | comparisons(new) | time(old) | time(new) | |-------|------------------|------------------|-----------|-----------| | 10000 | 235381 | 136615 | 25545 us | 20366 us | | 20000 | 510694 | 293425 | 31336 us | 18312 us | | 30000 | 800384 | 457412 | 35042 us | 27386 us | | 40000 | 1101617 | 626831 | 48779 us | 38253 us | | 50000 | 1409762 | 799637 | 62238 us | 46950 us | | 60000 | 1721191 | 974521 | 75588 us | 58367 us | | 70000 | 2038536 | 1152171 | 90823 us | 68778 us | | 80000 | 2362958 | 1333472 | 104165 us | 78625 us | | 90000 | 2690900 | 1516065 | 116111 us | 89573 us | | 100000| 3019413 | 1699879 | 133638 us | 100998 us | Refs: BOTTOM-UP-HEAPSORT, a new variant of HEAPSORT beating, on an average, QUICKSORT (if n is not very small) Ingo Wegener Theoretical Computer Science, 118(1); Pages 81-98, 13 September 1993 https://doi.org/10.1016/0304-3975(93)90364-Y Signed-off-by: Kuan-Wei Chiu Signed-off-by: Kent Overstreet --- fs/bcachefs/eytzinger.c | 105 ++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 88 insertions(+), 17 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/eytzinger.c b/fs/bcachefs/eytzinger.c index 0f955c3c76a7..2eaffe37b5e7 100644 --- a/fs/bcachefs/eytzinger.c +++ b/fs/bcachefs/eytzinger.c @@ -171,7 +171,7 @@ void eytzinger0_sort_r(void *base, size_t n, size_t size, swap_r_func_t swap_func, const void *priv) { - int i, c, r; + int i, j, k; /* called from 'sort' without swap function, let's pick the default */ if (swap_func == SWAP_WRAPPER && !((struct wrapper *)priv)->swap_func) @@ -188,17 +188,22 @@ void eytzinger0_sort_r(void *base, size_t n, size_t size, /* heapify */ for (i = n / 2 - 1; i >= 0; --i) { - for (r = i; r * 2 + 1 < n; r = c) { - c = r * 2 + 1; + /* Find the sift-down path all the way to the leaves. */ + for (j = i; k = j * 2 + 1, k + 1 < n;) + j = eytzinger0_do_cmp(base, n, size, cmp_func, priv, k, k + 1) > 0 ? k : k + 1; - if (c + 1 < n && - eytzinger0_do_cmp(base, n, size, cmp_func, priv, c, c + 1) < 0) - c++; + /* Special case for the last leaf with no sibling. */ + if (j * 2 + 2 == n) + j = j * 2 + 1; - if (eytzinger0_do_cmp(base, n, size, cmp_func, priv, r, c) >= 0) - break; + /* Backtrack to the correct location. */ + while (j != i && eytzinger0_do_cmp(base, n, size, cmp_func, priv, i, j) >= 0) + j = (j - 1) / 2; - eytzinger0_do_swap(base, n, size, swap_func, priv, r, c); + /* Shift the element into its correct place. */ + for (k = j; j != i;) { + j = (j - 1) / 2; + eytzinger0_do_swap(base, n, size, swap_func, priv, j, k); } } @@ -206,17 +211,22 @@ void eytzinger0_sort_r(void *base, size_t n, size_t size, for (i = n - 1; i > 0; --i) { eytzinger0_do_swap(base, n, size, swap_func, priv, 0, i); - for (r = 0; r * 2 + 1 < i; r = c) { - c = r * 2 + 1; + /* Find the sift-down path all the way to the leaves. */ + for (j = 0; k = j * 2 + 1, k + 1 < i;) + j = eytzinger0_do_cmp(base, n, size, cmp_func, priv, k, k + 1) > 0 ? k : k + 1; - if (c + 1 < i && - eytzinger0_do_cmp(base, n, size, cmp_func, priv, c, c + 1) < 0) - c++; + /* Special case for the last leaf with no sibling. */ + if (j * 2 + 2 == i) + j = j * 2 + 1; - if (eytzinger0_do_cmp(base, n, size, cmp_func, priv, r, c) >= 0) - break; + /* Backtrack to the correct location. */ + while (j && eytzinger0_do_cmp(base, n, size, cmp_func, priv, 0, j) >= 0) + j = (j - 1) / 2; - eytzinger0_do_swap(base, n, size, swap_func, priv, r, c); + /* Shift the element into its correct place. */ + for (k = j; j;) { + j = (j - 1) / 2; + eytzinger0_do_swap(base, n, size, swap_func, priv, j, k); } } } @@ -232,3 +242,64 @@ void eytzinger0_sort(void *base, size_t n, size_t size, return eytzinger0_sort_r(base, n, size, _CMP_WRAPPER, SWAP_WRAPPER, &w); } + +#if 0 +#include +#include +#include + +static u64 cmp_count; + +static int mycmp(const void *a, const void *b) +{ + u32 _a = *(u32 *)a; + u32 _b = *(u32 *)b; + + cmp_count++; + if (_a < _b) + return -1; + else if (_a > _b) + return 1; + else + return 0; +} + +static int test(void) +{ + size_t N, i; + ktime_t start, end; + s64 delta; + u32 *arr; + + for (N = 10000; N <= 100000; N += 10000) { + arr = kmalloc_array(N, sizeof(u32), GFP_KERNEL); + cmp_count = 0; + + for (i = 0; i < N; i++) + arr[i] = get_random_u32(); + + start = ktime_get(); + eytzinger0_sort(arr, N, sizeof(u32), mycmp, NULL); + end = ktime_get(); + + delta = ktime_us_delta(end, start); + printk(KERN_INFO "time: %lld\n", delta); + printk(KERN_INFO "comparisons: %lld\n", cmp_count); + + u32 prev = 0; + + eytzinger0_for_each(i, N) { + if (prev > arr[i]) + goto err; + prev = arr[i]; + } + + kfree(arr); + } + return 0; + +err: + kfree(arr); + return -1; +} +#endif -- cgit v1.2.3-59-g8ed1b From a21107eeb17a348a20248b343c80c2aea3c65b9f Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 01:44:05 -0400 Subject: bcachefs: kill for_each_btree_key_old() Dead code Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.h | 8 -------- 1 file changed, 8 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index fcf2742b6ad9..0427097bb966 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -794,14 +794,6 @@ __bch2_btree_iter_peek_and_restart(struct btree_trans *trans, return k; } -#define for_each_btree_key_old(_trans, _iter, _btree_id, \ - _start, _flags, _k, _ret) \ - for (bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ - (_start), (_flags)); \ - (_k) = __bch2_btree_iter_peek_and_restart((_trans), &(_iter), _flags),\ - !((_ret) = bkey_err(_k)) && (_k).k; \ - bch2_btree_iter_advance(&(_iter))) - #define for_each_btree_key_upto_norestart(_trans, _iter, _btree_id, \ _start, _end, _flags, _k, _ret) \ for (bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ -- cgit v1.2.3-59-g8ed1b From 665e8b32393523f0fb4752ccd4831ae6a10b1145 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 00:52:47 -0400 Subject: bcachefs: for_each_btree_key_continue() Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 28 +++++----------------------- fs/bcachefs/btree_iter.h | 21 +++++++++++++++------ 2 files changed, 20 insertions(+), 29 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index fc9506c71eae..0ca53392f787 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -793,31 +793,13 @@ static int bch2_check_extents_to_backpointers_pass(struct btree_trans *trans, while (level >= depth) { struct btree_iter iter; - bch2_trans_node_iter_init(trans, &iter, btree_id, POS_MIN, 0, - level, + bch2_trans_node_iter_init(trans, &iter, btree_id, POS_MIN, 0, level, BTREE_ITER_PREFETCH); - while (1) { - bch2_trans_begin(trans); - - struct bkey_s_c k = bch2_btree_iter_peek(&iter); - if (!k.k) - break; - ret = bkey_err(k) ?: - check_extent_to_backpointers(trans, s, btree_id, level, k) ?: - bch2_trans_commit(trans, NULL, NULL, - BCH_TRANS_COMMIT_no_enospc); - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) { - ret = 0; - continue; - } - if (ret) - break; - if (bpos_eq(iter.pos, SPOS_MAX)) - break; - bch2_btree_iter_advance(&iter); - } - bch2_trans_iter_exit(trans, &iter); + ret = for_each_btree_key_continue(trans, iter, 0, k, ({ + check_extent_to_backpointers(trans, s, btree_id, level, k) ?: + bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); + })); if (ret) return ret; diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 0427097bb966..52030929da0a 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -699,16 +699,12 @@ transaction_restart: \ _ret2 ?: trans_was_restarted(_trans, _restart_count); \ }) -#define for_each_btree_key_upto(_trans, _iter, _btree_id, \ - _start, _end, _flags, _k, _do) \ +#define for_each_btree_key_upto_continue(_trans, _iter, \ + _end, _flags, _k, _do) \ ({ \ - struct btree_iter _iter; \ struct bkey_s_c _k; \ int _ret3 = 0; \ \ - bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ - (_start), (_flags)); \ - \ do { \ _ret3 = lockrestart_do(_trans, ({ \ (_k) = bch2_btree_iter_peek_upto_type(&(_iter), \ @@ -724,6 +720,19 @@ transaction_restart: \ _ret3; \ }) +#define for_each_btree_key_continue(_trans, _iter, _flags, _k, _do) \ + for_each_btree_key_upto_continue(_trans, _iter, SPOS_MAX, _flags, _k, _do) + +#define for_each_btree_key_upto(_trans, _iter, _btree_id, \ + _start, _end, _flags, _k, _do) \ +({ \ + struct btree_iter _iter; \ + bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ + (_start), (_flags)); \ + \ + for_each_btree_key_upto_continue(_trans, _iter, _end, _flags, _k, _do);\ +}) + #define for_each_btree_key(_trans, _iter, _btree_id, \ _start, _flags, _k, _do) \ for_each_btree_key_upto(_trans, _iter, _btree_id, _start, \ -- cgit v1.2.3-59-g8ed1b From 68e142405cf4cf01461012ec72d675038c514b92 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 00:11:01 -0400 Subject: bcachefs: bch2_gc() is now private to btree_gc.c Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 14 ++++++-------- fs/bcachefs/btree_gc.h | 2 +- fs/bcachefs/recovery_passes.c | 5 ----- fs/bcachefs/sysfs.c | 12 +----------- 4 files changed, 8 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 791470b0c654..606e1120ab7c 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -1756,7 +1756,7 @@ static void bch2_gc_stripes_reset(struct bch_fs *c, bool metadata_only) * move around - if references move backwards in the ordering GC * uses, GC could skip past them */ -int bch2_gc(struct bch_fs *c, bool initial, bool metadata_only) +static int bch2_gc(struct bch_fs *c, bool initial, bool metadata_only) { unsigned iter = 0; int ret; @@ -1843,6 +1843,11 @@ out: return ret; } +int bch2_check_allocations(struct bch_fs *c) +{ + return bch2_gc(c, true, false); +} + static int gc_btree_gens_key(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k) @@ -2024,14 +2029,7 @@ static int bch2_gc_thread(void *arg) last = atomic64_read(&clock->now); last_kick = atomic_read(&c->kick_gc); - /* - * Full gc is currently incompatible with btree key cache: - */ -#if 0 - ret = bch2_gc(c, false, false); -#else bch2_gc_gens(c); -#endif debug_check_no_locks_held(); } diff --git a/fs/bcachefs/btree_gc.h b/fs/bcachefs/btree_gc.h index 607575f83a00..0d6c0a2df613 100644 --- a/fs/bcachefs/btree_gc.h +++ b/fs/bcachefs/btree_gc.h @@ -6,7 +6,7 @@ #include "btree_types.h" int bch2_check_topology(struct bch_fs *); -int bch2_gc(struct bch_fs *, bool, bool); +int bch2_check_allocations(struct bch_fs *); int bch2_gc_gens(struct bch_fs *); void bch2_gc_thread_stop(struct bch_fs *); int bch2_gc_thread_start(struct bch_fs *); diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index 0cec0f7d9703..c065c0d01212 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -26,11 +26,6 @@ const char * const bch2_recovery_passes[] = { NULL }; -static int bch2_check_allocations(struct bch_fs *c) -{ - return bch2_gc(c, true, false); -} - static int bch2_set_may_go_rw(struct bch_fs *c) { struct journal_keys *keys = &c->journal_keys; diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index b6bfd7cbce18..4011ab4e710b 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -501,18 +501,8 @@ STORE(bch2_fs) if (attr == &sysfs_btree_wakeup) bch2_btree_wakeup_all(c); - if (attr == &sysfs_trigger_gc) { - /* - * Full gc is currently incompatible with btree key cache: - */ -#if 0 - down_read(&c->state_lock); - bch2_gc(c, false, false); - up_read(&c->state_lock); -#else + if (attr == &sysfs_trigger_gc) bch2_gc_gens(c); -#endif - } if (attr == &sysfs_trigger_discards) bch2_do_discards(c); -- cgit v1.2.3-59-g8ed1b From d1b213a00ddc1166bc210868371d4f619286d69b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 17:05:01 -0400 Subject: bcachefs: Finish converting reconstruct_alloc to errors_silent with errors_silent, reconstruct_alloc no longer requires fsck and fix_errors to work Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 4 +--- fs/bcachefs/recovery.c | 11 +++++++++++ 2 files changed, 12 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 606e1120ab7c..130f4a3885b2 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -1209,9 +1209,7 @@ static int bch2_gc_done(struct bch_fs *c, { struct bch_dev *ca = NULL; struct printbuf buf = PRINTBUF; - bool verify = !metadata_only && - !c->opts.reconstruct_alloc && - (!initial || (c->sb.compat & (1ULL << BCH_COMPAT_alloc_info))); + bool verify = !metadata_only; unsigned i; int ret = 0; diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 9cf262f1a3cd..d22579a036c7 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -65,9 +65,20 @@ static void bch2_reconstruct_alloc(struct bch_fs *c) __set_bit_le64(BCH_FSCK_ERR_ptr_to_missing_alloc_key, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_ptr_gen_newer_than_bucket_gen, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_stale_dirty_ptr, ext->errors_silent); + + __set_bit_le64(BCH_FSCK_ERR_dev_usage_buckets_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_dev_usage_sectors_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_dev_usage_fragmented_wrong, ext->errors_silent); + + __set_bit_le64(BCH_FSCK_ERR_fs_usage_btree_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_fs_usage_cached_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_fs_usage_persistent_reserved_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_fs_usage_replicas_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_data_type_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_alloc_key_gen_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_alloc_key_dirty_sectors_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_cached_sectors_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_alloc_key_stripe_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_alloc_key_stripe_redundancy_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_need_discard_key_wrong, ext->errors_silent); -- cgit v1.2.3-59-g8ed1b From 58dda9c10e3fdea488180a74de622432de6b4568 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 22:40:12 -0400 Subject: bcachefs: kill metadata only gc Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 188 ++++++++++++++++--------------------------------- 1 file changed, 59 insertions(+), 129 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 130f4a3885b2..614959f0f669 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -899,18 +899,18 @@ static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool i } static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id, - bool initial, bool metadata_only) + bool initial) { struct bch_fs *c = trans->c; struct btree_iter iter; struct btree *b; - unsigned depth = metadata_only ? 1 : 0; + unsigned target_depth = btree_type_has_ptrs(btree_id) ? 0 : 1; int ret = 0; gc_pos_set(c, gc_pos_btree(btree_id, POS_MIN, 0)); __for_each_btree_node(trans, iter, btree_id, POS_MIN, - 0, depth, BTREE_ITER_PREFETCH, b, ret) { + 0, target_depth, BTREE_ITER_PREFETCH, b, ret) { bch2_verify_btree_nr_keys(b); gc_pos_set(c, gc_pos_btree_node(b)); @@ -1029,12 +1029,15 @@ fsck_err: } static int bch2_gc_btree_init(struct btree_trans *trans, - enum btree_id btree_id, - bool metadata_only) + enum btree_id btree_id) { struct bch_fs *c = trans->c; struct btree *b; - unsigned target_depth = metadata_only ? 1 : 0; + /* + * We need to make sure every leaf node is readable before going RW + unsigned target_depth = btree_node_type_needs_gc(__btree_node_type(0, btree_id)) ? 0 : 1; + */ + unsigned target_depth = 0; struct printbuf buf = PRINTBUF; int ret = 0; @@ -1046,8 +1049,7 @@ static int bch2_gc_btree_init(struct btree_trans *trans, if (mustfix_fsck_err_on(!bpos_eq(b->data->min_key, POS_MIN), c, btree_root_bad_min_key, "btree root with incorrect min_key: %s", buf.buf)) { - bch_err(c, "repair unimplemented"); - ret = -BCH_ERR_fsck_repair_unimplemented; + ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); goto fsck_err; } @@ -1056,8 +1058,7 @@ static int bch2_gc_btree_init(struct btree_trans *trans, if (mustfix_fsck_err_on(!bpos_eq(b->data->max_key, SPOS_MAX), c, btree_root_bad_max_key, "btree root with incorrect max_key: %s", buf.buf)) { - bch_err(c, "repair unimplemented"); - ret = -BCH_ERR_fsck_repair_unimplemented; + ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); goto fsck_err; } @@ -1084,7 +1085,7 @@ static inline int btree_id_gc_phase_cmp(enum btree_id l, enum btree_id r) (int) btree_id_to_gc_phase(r); } -static int bch2_gc_btrees(struct bch_fs *c, bool initial, bool metadata_only) +static int bch2_gc_btrees(struct bch_fs *c, bool initial) { struct btree_trans *trans = bch2_trans_get(c); enum btree_id ids[BTREE_ID_NR]; @@ -1095,18 +1096,15 @@ static int bch2_gc_btrees(struct bch_fs *c, bool initial, bool metadata_only) ids[i] = i; bubble_sort(ids, BTREE_ID_NR, btree_id_gc_phase_cmp); - for (i = 0; i < BTREE_ID_NR && !ret; i++) - ret = initial - ? bch2_gc_btree_init(trans, ids[i], metadata_only) - : bch2_gc_btree(trans, ids[i], initial, metadata_only); + for (i = 0; i < btree_id_nr_alive(c) && !ret; i++) { + unsigned btree = i < BTREE_ID_NR ? ids[i] : i; - for (i = BTREE_ID_NR; i < btree_id_nr_alive(c) && !ret; i++) { - if (!bch2_btree_id_root(c, i)->alive) + if (IS_ERR_OR_NULL(bch2_btree_id_root(c, btree)->b)) continue; ret = initial - ? bch2_gc_btree_init(trans, i, metadata_only) - : bch2_gc_btree(trans, i, initial, metadata_only); + ? bch2_gc_btree_init(trans, btree) + : bch2_gc_btree(trans, btree, initial); } bch2_trans_put(trans); @@ -1169,24 +1167,6 @@ static void bch2_mark_superblocks(struct bch_fs *c) mutex_unlock(&c->sb_lock); } -#if 0 -/* Also see bch2_pending_btree_node_free_insert_done() */ -static void bch2_mark_pending_btree_node_frees(struct bch_fs *c) -{ - struct btree_update *as; - struct pending_btree_node_free *d; - - mutex_lock(&c->btree_interior_update_lock); - gc_pos_set(c, gc_phase(GC_PHASE_PENDING_DELETE)); - - for_each_pending_btree_node_free(c, as, d) - if (d->index_update_done) - bch2_mark_key(c, bkey_i_to_s_c(&d->key), BTREE_TRIGGER_GC); - - mutex_unlock(&c->btree_interior_update_lock); -} -#endif - static void bch2_gc_free(struct bch_fs *c) { genradix_free(&c->reflink_gc_table); @@ -1204,26 +1184,23 @@ static void bch2_gc_free(struct bch_fs *c) c->usage_gc = NULL; } -static int bch2_gc_done(struct bch_fs *c, - bool initial, bool metadata_only) +static int bch2_gc_done(struct bch_fs *c) { struct bch_dev *ca = NULL; struct printbuf buf = PRINTBUF; - bool verify = !metadata_only; unsigned i; int ret = 0; percpu_down_write(&c->mark_lock); -#define copy_field(_err, _f, _msg, ...) \ - if (dst->_f != src->_f && \ - (!verify || \ - fsck_err(c, _err, _msg ": got %llu, should be %llu" \ - , ##__VA_ARGS__, dst->_f, src->_f))) \ +#define copy_field(_err, _f, _msg, ...) \ + if (fsck_err_on(dst->_f != src->_f, c, _err, \ + _msg ": got %llu, should be %llu" , ##__VA_ARGS__, \ + dst->_f, src->_f)) \ dst->_f = src->_f -#define copy_dev_field(_err, _f, _msg, ...) \ +#define copy_dev_field(_err, _f, _msg, ...) \ copy_field(_err, _f, "dev %u has wrong " _msg, ca->dev_idx, ##__VA_ARGS__) -#define copy_fs_field(_err, _f, _msg, ...) \ +#define copy_fs_field(_err, _f, _msg, ...) \ copy_field(_err, _f, "fs has wrong " _msg, ##__VA_ARGS__) for (i = 0; i < ARRAY_SIZE(c->usage); i++) @@ -1256,31 +1233,24 @@ static int bch2_gc_done(struct bch_fs *c, copy_fs_field(fs_usage_btree_wrong, b.btree, "btree"); - if (!metadata_only) { - copy_fs_field(fs_usage_data_wrong, - b.data, "data"); - copy_fs_field(fs_usage_cached_wrong, - b.cached, "cached"); - copy_fs_field(fs_usage_reserved_wrong, - b.reserved, "reserved"); - copy_fs_field(fs_usage_nr_inodes_wrong, - b.nr_inodes,"nr_inodes"); - - for (i = 0; i < BCH_REPLICAS_MAX; i++) - copy_fs_field(fs_usage_persistent_reserved_wrong, - persistent_reserved[i], - "persistent_reserved[%i]", i); - } + copy_fs_field(fs_usage_data_wrong, + b.data, "data"); + copy_fs_field(fs_usage_cached_wrong, + b.cached, "cached"); + copy_fs_field(fs_usage_reserved_wrong, + b.reserved, "reserved"); + copy_fs_field(fs_usage_nr_inodes_wrong, + b.nr_inodes,"nr_inodes"); + + for (i = 0; i < BCH_REPLICAS_MAX; i++) + copy_fs_field(fs_usage_persistent_reserved_wrong, + persistent_reserved[i], + "persistent_reserved[%i]", i); for (i = 0; i < c->replicas.nr; i++) { struct bch_replicas_entry_v1 *e = cpu_replicas_entry(&c->replicas, i); - if (metadata_only && - (e->data_type == BCH_DATA_user || - e->data_type == BCH_DATA_cached)) - continue; - printbuf_reset(&buf); bch2_replicas_entry_to_text(&buf, e); @@ -1359,8 +1329,7 @@ static inline bool bch2_alloc_v4_cmp(struct bch_alloc_v4 l, static int bch2_alloc_write_key(struct btree_trans *trans, struct btree_iter *iter, - struct bkey_s_c k, - bool metadata_only) + struct bkey_s_c k) { struct bch_fs *c = trans->c; struct bch_dev *ca = bch_dev_bkey_exists(c, iter->pos.inode); @@ -1400,12 +1369,6 @@ static int bch2_alloc_write_key(struct btree_trans *trans, bch2_dev_usage_update_m(c, ca, &old_gc, &gc); percpu_up_read(&c->mark_lock); - if (metadata_only && - gc.data_type != BCH_DATA_sb && - gc.data_type != BCH_DATA_journal && - gc.data_type != BCH_DATA_btree) - return 0; - if (gen_after(old->gen, gc.gen)) return 0; @@ -1463,7 +1426,7 @@ fsck_err: return ret; } -static int bch2_gc_alloc_done(struct bch_fs *c, bool metadata_only) +static int bch2_gc_alloc_done(struct bch_fs *c) { int ret = 0; @@ -1474,7 +1437,7 @@ static int bch2_gc_alloc_done(struct bch_fs *c, bool metadata_only) POS(ca->dev_idx, ca->mi.nbuckets - 1), BTREE_ITER_SLOTS|BTREE_ITER_PREFETCH, k, NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, - bch2_alloc_write_key(trans, &iter, k, metadata_only))); + bch2_alloc_write_key(trans, &iter, k))); if (ret) { percpu_ref_put(&ca->ref); break; @@ -1485,7 +1448,7 @@ static int bch2_gc_alloc_done(struct bch_fs *c, bool metadata_only) return ret; } -static int bch2_gc_alloc_start(struct bch_fs *c, bool metadata_only) +static int bch2_gc_alloc_start(struct bch_fs *c) { for_each_member_device(c, ca) { struct bucket_array *buckets = kvmalloc(sizeof(struct bucket_array) + @@ -1513,36 +1476,19 @@ static int bch2_gc_alloc_start(struct bch_fs *c, bool metadata_only) g->gen_valid = 1; g->gen = a->gen; - - if (metadata_only && - (a->data_type == BCH_DATA_user || - a->data_type == BCH_DATA_cached || - a->data_type == BCH_DATA_parity)) { - g->data_type = a->data_type; - g->dirty_sectors = a->dirty_sectors; - g->cached_sectors = a->cached_sectors; - g->stripe = a->stripe; - g->stripe_redundancy = a->stripe_redundancy; - } - 0; }))); bch_err_fn(c, ret); return ret; } -static void bch2_gc_alloc_reset(struct bch_fs *c, bool metadata_only) +static void bch2_gc_alloc_reset(struct bch_fs *c) { for_each_member_device(c, ca) { struct bucket_array *buckets = gc_bucket_array(ca); struct bucket *g; for_each_bucket(g, buckets) { - if (metadata_only && - (g->data_type == BCH_DATA_user || - g->data_type == BCH_DATA_cached || - g->data_type == BCH_DATA_parity)) - continue; g->data_type = 0; g->dirty_sectors = 0; g->cached_sectors = 0; @@ -1599,13 +1545,10 @@ fsck_err: return ret; } -static int bch2_gc_reflink_done(struct bch_fs *c, bool metadata_only) +static int bch2_gc_reflink_done(struct bch_fs *c) { size_t idx = 0; - if (metadata_only) - return 0; - int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_reflink, POS_MIN, @@ -1616,13 +1559,8 @@ static int bch2_gc_reflink_done(struct bch_fs *c, bool metadata_only) return ret; } -static int bch2_gc_reflink_start(struct bch_fs *c, - bool metadata_only) +static int bch2_gc_reflink_start(struct bch_fs *c) { - - if (metadata_only) - return 0; - c->reflink_gc_nr = 0; int ret = bch2_trans_run(c, @@ -1650,7 +1588,7 @@ static int bch2_gc_reflink_start(struct bch_fs *c, return ret; } -static void bch2_gc_reflink_reset(struct bch_fs *c, bool metadata_only) +static void bch2_gc_reflink_reset(struct bch_fs *c) { struct genradix_iter iter; struct reflink_gc *r; @@ -1712,11 +1650,8 @@ fsck_err: return ret; } -static int bch2_gc_stripes_done(struct bch_fs *c, bool metadata_only) +static int bch2_gc_stripes_done(struct bch_fs *c) { - if (metadata_only) - return 0; - return bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_stripes, POS_MIN, @@ -1725,7 +1660,7 @@ static int bch2_gc_stripes_done(struct bch_fs *c, bool metadata_only) bch2_gc_write_stripes_key(trans, &iter, k))); } -static void bch2_gc_stripes_reset(struct bch_fs *c, bool metadata_only) +static void bch2_gc_stripes_reset(struct bch_fs *c) { genradix_free(&c->gc_stripes); } @@ -1735,7 +1670,6 @@ static void bch2_gc_stripes_reset(struct bch_fs *c, bool metadata_only) * * @c: filesystem object * @initial: are we in recovery? - * @metadata_only: are we just checking metadata references, or everything? * * Returns: 0 on success, or standard errcode on failure * @@ -1754,7 +1688,7 @@ static void bch2_gc_stripes_reset(struct bch_fs *c, bool metadata_only) * move around - if references move backwards in the ordering GC * uses, GC could skip past them */ -static int bch2_gc(struct bch_fs *c, bool initial, bool metadata_only) +static int bch2_gc(struct bch_fs *c, bool initial) { unsigned iter = 0; int ret; @@ -1766,8 +1700,8 @@ static int bch2_gc(struct bch_fs *c, bool initial, bool metadata_only) bch2_btree_interior_updates_flush(c); ret = bch2_gc_start(c) ?: - bch2_gc_alloc_start(c, metadata_only) ?: - bch2_gc_reflink_start(c, metadata_only); + bch2_gc_alloc_start(c) ?: + bch2_gc_reflink_start(c); if (ret) goto out; again: @@ -1775,14 +1709,10 @@ again: bch2_mark_superblocks(c); - ret = bch2_gc_btrees(c, initial, metadata_only); - + ret = bch2_gc_btrees(c, initial); if (ret) goto out; -#if 0 - bch2_mark_pending_btree_node_frees(c); -#endif c->gc_count++; if (test_bit(BCH_FS_need_another_gc, &c->flags) || @@ -1800,9 +1730,9 @@ again: clear_bit(BCH_FS_need_another_gc, &c->flags); __gc_pos_set(c, gc_phase(GC_PHASE_NOT_RUNNING)); - bch2_gc_stripes_reset(c, metadata_only); - bch2_gc_alloc_reset(c, metadata_only); - bch2_gc_reflink_reset(c, metadata_only); + bch2_gc_stripes_reset(c); + bch2_gc_alloc_reset(c); + bch2_gc_reflink_reset(c); ret = bch2_gc_reset(c); if (ret) goto out; @@ -1815,10 +1745,10 @@ out: if (!ret) { bch2_journal_block(&c->journal); - ret = bch2_gc_alloc_done(c, metadata_only) ?: - bch2_gc_done(c, initial, metadata_only) ?: - bch2_gc_stripes_done(c, metadata_only) ?: - bch2_gc_reflink_done(c, metadata_only); + ret = bch2_gc_alloc_done(c) ?: + bch2_gc_done(c) ?: + bch2_gc_stripes_done(c) ?: + bch2_gc_reflink_done(c); bch2_journal_unblock(&c->journal); } @@ -1843,7 +1773,7 @@ out: int bch2_check_allocations(struct bch_fs *c) { - return bch2_gc(c, true, false); + return bch2_gc(c, true); } static int gc_btree_gens_key(struct btree_trans *trans, -- cgit v1.2.3-59-g8ed1b From b982d645a40e29296a5dbc3c5da067e93e7abd62 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 23:20:49 -0400 Subject: bcachefs: move topology repair kick to gc_btrees() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 48 +++++++++--------------------------------------- 1 file changed, 9 insertions(+), 39 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 614959f0f669..b86a821f388a 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -52,12 +52,6 @@ static struct bkey_s unsafe_bkey_s_c_to_s(struct bkey_s_c k) }}}; } -static bool should_restart_for_topology_repair(struct bch_fs *c) -{ - return c->opts.fix_errors != FSCK_FIX_no && - !(c->recovery_passes_complete & BIT_ULL(BCH_RECOVERY_PASS_check_topology)); -} - static inline void __gc_pos_set(struct bch_fs *c, struct gc_pos new_pos) { preempt_disable(); @@ -881,9 +875,6 @@ static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool i if (ret) return ret; - if (!btree_node_type_needs_gc(btree_node_type(b))) - return 0; - bch2_btree_node_iter_init_from_start(&iter, b); while ((k = bch2_btree_node_iter_peek_unpack(&iter, b, &unpacked)).k) { @@ -982,36 +973,9 @@ static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b b->c.btree_id, b->c.level - 1, false); ret = PTR_ERR_OR_ZERO(child); - - if (bch2_err_matches(ret, EIO)) { - bch2_topology_error(c); - - if (__fsck_err(c, - FSCK_CAN_FIX| - FSCK_CAN_IGNORE| - FSCK_NO_RATELIMIT, - btree_node_read_error, - "Unreadable btree node at btree %s level %u:\n" - " %s", - bch2_btree_id_str(b->c.btree_id), - b->c.level - 1, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(cur.k)), buf.buf)) && - should_restart_for_topology_repair(c)) { - bch_info(c, "Halting mark and sweep to start topology repair pass"); - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); - goto fsck_err; - } else { - /* Continue marking when opted to not - * fix the error: */ - ret = 0; - set_bit(BCH_FS_initial_gc_unfixed, &c->flags); - continue; - } - } else if (ret) { - bch_err_msg(c, ret, "getting btree node"); + bch_err_msg(c, ret, "getting btree node"); + if (ret) break; - } ret = bch2_gc_btree_init_recurse(trans, child, target_depth); @@ -1105,8 +1069,14 @@ static int bch2_gc_btrees(struct bch_fs *c, bool initial) ret = initial ? bch2_gc_btree_init(trans, btree) : bch2_gc_btree(trans, btree, initial); - } + if (mustfix_fsck_err_on(bch2_err_matches(ret, EIO), + c, btree_node_read_error, + "btree node read error for %s", + bch2_btree_id_str(btree))) + ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); + } +fsck_err: bch2_trans_put(trans); bch_err_fn(c, ret); return ret; -- cgit v1.2.3-59-g8ed1b From d1adfe4e7e4e7ea225547a07c4b79c314c50c6fb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 23:26:36 -0400 Subject: bcachefs: move root node topo checks to node_check_topology() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 24 +----------------------- fs/bcachefs/btree_update_interior.c | 18 ++++++++++++++++++ 2 files changed, 19 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index b86a821f388a..6b466b2ebebe 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -996,36 +996,16 @@ static int bch2_gc_btree_init(struct btree_trans *trans, enum btree_id btree_id) { struct bch_fs *c = trans->c; - struct btree *b; /* * We need to make sure every leaf node is readable before going RW unsigned target_depth = btree_node_type_needs_gc(__btree_node_type(0, btree_id)) ? 0 : 1; */ unsigned target_depth = 0; - struct printbuf buf = PRINTBUF; int ret = 0; - b = bch2_btree_id_root(c, btree_id)->b; + struct btree *b = bch2_btree_id_root(c, btree_id)->b; six_lock_read(&b->c.lock, NULL, NULL); - printbuf_reset(&buf); - bch2_bpos_to_text(&buf, b->data->min_key); - if (mustfix_fsck_err_on(!bpos_eq(b->data->min_key, POS_MIN), c, - btree_root_bad_min_key, - "btree root with incorrect min_key: %s", buf.buf)) { - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); - goto fsck_err; - } - - printbuf_reset(&buf); - bch2_bpos_to_text(&buf, b->data->max_key); - if (mustfix_fsck_err_on(!bpos_eq(b->data->max_key, SPOS_MAX), c, - btree_root_bad_max_key, - "btree root with incorrect max_key: %s", buf.buf)) { - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); - goto fsck_err; - } - if (b->c.level >= target_depth) ret = bch2_gc_btree_init_recurse(trans, b, target_depth); @@ -1035,11 +1015,9 @@ static int bch2_gc_btree_init(struct btree_trans *trans, ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, true, &k, true); } -fsck_err: six_unlock_read(&b->c.lock); bch_err_fn(c, ret); - printbuf_exit(&buf); return ret; } diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index b4efd8cc4d1a..875b1795e408 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -73,6 +73,24 @@ int bch2_btree_node_check_topology(struct btree_trans *trans, struct btree *b) !bpos_eq(bkey_i_to_btree_ptr_v2(&b->key)->v.min_key, b->data->min_key)); + if (b == btree_node_root(c, b)) { + if (!bpos_eq(b->data->min_key, POS_MIN)) { + printbuf_reset(&buf); + bch2_bpos_to_text(&buf, b->data->min_key); + need_fsck_err(c, btree_root_bad_min_key, + "btree root with incorrect min_key: %s", buf.buf); + goto topology_repair; + } + + if (!bpos_eq(b->data->max_key, SPOS_MAX)) { + printbuf_reset(&buf); + bch2_bpos_to_text(&buf, b->data->max_key); + need_fsck_err(c, btree_root_bad_max_key, + "btree root with incorrect max_key: %s", buf.buf); + goto topology_repair; + } + } + if (!b->c.level) return 0; -- cgit v1.2.3-59-g8ed1b From ba665494fbf8ea63656b7d45755595cfc6214b50 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 23:39:12 -0400 Subject: bcachefs: gc_btree_init_recurse() uses gc_mark_node() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 64 +++++++++++++++++--------------------------------- 1 file changed, 22 insertions(+), 42 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 6b466b2ebebe..18b33811ae40 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -866,8 +866,7 @@ err: static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool initial) { - struct btree_node_iter iter; - struct bkey unpacked; + struct btree_and_journal_iter iter; struct bkey_s_c k; int ret = 0; @@ -875,18 +874,18 @@ static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool i if (ret) return ret; - bch2_btree_node_iter_init_from_start(&iter, b); + bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); - while ((k = bch2_btree_node_iter_peek_unpack(&iter, b, &unpacked)).k) { - ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false, - &k, initial); + while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { + ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false, &k, initial); if (ret) - return ret; + break; - bch2_btree_node_iter_advance(&iter, b); + bch2_btree_and_journal_iter_advance(&iter); } - return 0; + bch2_btree_and_journal_iter_exit(&iter); + return ret; } static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id, @@ -932,44 +931,26 @@ static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id, static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b, unsigned target_depth) { - struct bch_fs *c = trans->c; - struct btree_and_journal_iter iter; - struct bkey_s_c k; - struct bkey_buf cur; - struct printbuf buf = PRINTBUF; - int ret = 0; - - ret = bch2_btree_node_check_topology(trans, b); + int ret = btree_gc_mark_node(trans, b, true); if (ret) return ret; - bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); - bch2_bkey_buf_init(&cur); - - while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { - BUG_ON(bpos_lt(k.k->p, b->data->min_key)); - BUG_ON(bpos_gt(k.k->p, b->data->max_key)); - - ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, - false, &k, true); - if (ret) - goto fsck_err; - - bch2_btree_and_journal_iter_advance(&iter); - } - if (b->c.level > target_depth) { - bch2_btree_and_journal_iter_exit(&iter); + struct bch_fs *c = trans->c; + struct btree_and_journal_iter iter; + struct bkey_s_c k; + struct bkey_buf cur; + + bch2_bkey_buf_init(&cur); bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); iter.prefetch = true; while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { - struct btree *child; - bch2_bkey_buf_reassemble(&cur, c, k); bch2_btree_and_journal_iter_advance(&iter); - child = bch2_btree_node_get_noiter(trans, cur.k, + struct btree *child = + bch2_btree_node_get_noiter(trans, cur.k, b->c.btree_id, b->c.level - 1, false); ret = PTR_ERR_OR_ZERO(child); @@ -977,18 +958,17 @@ static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b if (ret) break; - ret = bch2_gc_btree_init_recurse(trans, child, - target_depth); + ret = bch2_gc_btree_init_recurse(trans, child, target_depth); six_unlock_read(&child->c.lock); if (ret) break; } + + bch2_bkey_buf_exit(&cur, c); + bch2_btree_and_journal_iter_exit(&iter); } -fsck_err: - bch2_bkey_buf_exit(&cur, c); - bch2_btree_and_journal_iter_exit(&iter); - printbuf_exit(&buf); + return ret; } -- cgit v1.2.3-59-g8ed1b From c281db0fa59180bd8901ed88c526616143ef284b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 02:11:03 -0400 Subject: bcachefs: mark_superblock cleanup Consolidate mark_superblock() and trans_mark_superblock(), like we did with the other trigger paths. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 54 ++--------------- fs/bcachefs/buckets.c | 160 +++++++++++++++++++++++++------------------------ fs/bcachefs/buckets.h | 11 ++-- fs/bcachefs/journal.c | 5 +- fs/bcachefs/super.c | 6 +- 5 files changed, 97 insertions(+), 139 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 18b33811ae40..512be84aec23 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -1040,59 +1040,14 @@ fsck_err: return ret; } -static void mark_metadata_sectors(struct bch_fs *c, struct bch_dev *ca, - u64 start, u64 end, - enum bch_data_type type, - unsigned flags) -{ - u64 b = sector_to_bucket(ca, start); - - do { - unsigned sectors = - min_t(u64, bucket_to_sector(ca, b + 1), end) - start; - - bch2_mark_metadata_bucket(c, ca, b, type, sectors, - gc_phase(GC_PHASE_SB), flags); - b++; - start += sectors; - } while (start < end); -} - -static void bch2_mark_dev_superblock(struct bch_fs *c, struct bch_dev *ca, - unsigned flags) -{ - struct bch_sb_layout *layout = &ca->disk_sb.sb->layout; - unsigned i; - u64 b; - - for (i = 0; i < layout->nr_superblocks; i++) { - u64 offset = le64_to_cpu(layout->sb_offset[i]); - - if (offset == BCH_SB_SECTOR) - mark_metadata_sectors(c, ca, 0, BCH_SB_SECTOR, - BCH_DATA_sb, flags); - - mark_metadata_sectors(c, ca, offset, - offset + (1 << layout->sb_max_size_bits), - BCH_DATA_sb, flags); - } - - for (i = 0; i < ca->journal.nr; i++) { - b = ca->journal.buckets[i]; - bch2_mark_metadata_bucket(c, ca, b, BCH_DATA_journal, - ca->mi.bucket_size, - gc_phase(GC_PHASE_SB), flags); - } -} - -static void bch2_mark_superblocks(struct bch_fs *c) +static int bch2_mark_superblocks(struct bch_fs *c) { mutex_lock(&c->sb_lock); gc_pos_set(c, gc_phase(GC_PHASE_SB)); - for_each_online_member(c, ca) - bch2_mark_dev_superblock(c, ca, BTREE_TRIGGER_GC); + int ret = bch2_trans_mark_dev_sbs_flags(c, BTREE_TRIGGER_GC); mutex_unlock(&c->sb_lock); + return ret; } static void bch2_gc_free(struct bch_fs *c) @@ -1635,7 +1590,8 @@ static int bch2_gc(struct bch_fs *c, bool initial) again: gc_pos_set(c, gc_phase(GC_PHASE_START)); - bch2_mark_superblocks(c); + ret = bch2_mark_superblocks(c); + BUG_ON(ret); ret = bch2_gc_btrees(c, initial); if (ret) diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 6ac6f61cfbc4..015321a87170 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -485,59 +485,6 @@ int bch2_update_cached_sectors_list(struct btree_trans *trans, unsigned dev, s64 return bch2_update_replicas_list(trans, &r.e, sectors); } -int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca, - size_t b, enum bch_data_type data_type, - unsigned sectors, struct gc_pos pos, - unsigned flags) -{ - struct bucket old, new, *g; - int ret = 0; - - BUG_ON(!(flags & BTREE_TRIGGER_GC)); - BUG_ON(data_type != BCH_DATA_sb && - data_type != BCH_DATA_journal); - - /* - * Backup superblock might be past the end of our normal usable space: - */ - if (b >= ca->mi.nbuckets) - return 0; - - percpu_down_read(&c->mark_lock); - g = gc_bucket(ca, b); - - bucket_lock(g); - old = *g; - - if (bch2_fs_inconsistent_on(g->data_type && - g->data_type != data_type, c, - "different types of data in same bucket: %s, %s", - bch2_data_type_str(g->data_type), - bch2_data_type_str(data_type))) { - ret = -EIO; - goto err; - } - - if (bch2_fs_inconsistent_on((u64) g->dirty_sectors + sectors > ca->mi.bucket_size, c, - "bucket %u:%zu gen %u data type %s sector count overflow: %u + %u > bucket size", - ca->dev_idx, b, g->gen, - bch2_data_type_str(g->data_type ?: data_type), - g->dirty_sectors, sectors)) { - ret = -EIO; - goto err; - } - - g->data_type = data_type; - g->dirty_sectors += sectors; - new = *g; -err: - bucket_unlock(g); - if (!ret) - bch2_dev_usage_update_m(c, ca, &old, &new); - percpu_up_read(&c->mark_lock); - return ret; -} - int bch2_check_bucket_ref(struct btree_trans *trans, struct bkey_s_c k, const struct bch_extent_ptr *ptr, @@ -1107,22 +1054,16 @@ int bch2_trigger_reservation(struct btree_trans *trans, /* Mark superblocks: */ static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans, - struct bch_dev *ca, size_t b, + struct bch_dev *ca, u64 b, enum bch_data_type type, unsigned sectors) { struct bch_fs *c = trans->c; struct btree_iter iter; - struct bkey_i_alloc_v4 *a; int ret = 0; - /* - * Backup superblock might be past the end of our normal usable space: - */ - if (b >= ca->mi.nbuckets) - return 0; - - a = bch2_trans_start_alloc_update(trans, &iter, POS(ca->dev_idx, b)); + struct bkey_i_alloc_v4 *a = + bch2_trans_start_alloc_update(trans, &iter, POS(ca->dev_idx, b)); if (IS_ERR(a)) return PTR_ERR(a); @@ -1150,20 +1091,78 @@ err: return ret; } +static int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca, + u64 b, enum bch_data_type data_type, + unsigned sectors, unsigned flags) +{ + struct bucket old, new, *g; + int ret = 0; + + percpu_down_read(&c->mark_lock); + g = gc_bucket(ca, b); + + bucket_lock(g); + old = *g; + + if (bch2_fs_inconsistent_on(g->data_type && + g->data_type != data_type, c, + "different types of data in same bucket: %s, %s", + bch2_data_type_str(g->data_type), + bch2_data_type_str(data_type))) { + ret = -EIO; + goto err; + } + + if (bch2_fs_inconsistent_on((u64) g->dirty_sectors + sectors > ca->mi.bucket_size, c, + "bucket %u:%llu gen %u data type %s sector count overflow: %u + %u > bucket size", + ca->dev_idx, b, g->gen, + bch2_data_type_str(g->data_type ?: data_type), + g->dirty_sectors, sectors)) { + ret = -EIO; + goto err; + } + + g->data_type = data_type; + g->dirty_sectors += sectors; + new = *g; +err: + bucket_unlock(g); + if (!ret) + bch2_dev_usage_update_m(c, ca, &old, &new); + percpu_up_read(&c->mark_lock); + return ret; +} + int bch2_trans_mark_metadata_bucket(struct btree_trans *trans, - struct bch_dev *ca, size_t b, + struct bch_dev *ca, u64 b, enum bch_data_type type, - unsigned sectors) + unsigned sectors, unsigned flags) { - return commit_do(trans, NULL, NULL, 0, - __bch2_trans_mark_metadata_bucket(trans, ca, b, type, sectors)); + BUG_ON(type != BCH_DATA_free && + type != BCH_DATA_sb && + type != BCH_DATA_journal); + + /* + * Backup superblock might be past the end of our normal usable space: + */ + if (b >= ca->mi.nbuckets) + return 0; + + if (flags & BTREE_TRIGGER_GC) + return bch2_mark_metadata_bucket(trans->c, ca, b, type, sectors, flags); + else if (flags & BTREE_TRIGGER_TRANSACTIONAL) + return commit_do(trans, NULL, NULL, 0, + __bch2_trans_mark_metadata_bucket(trans, ca, b, type, sectors)); + else + BUG(); } static int bch2_trans_mark_metadata_sectors(struct btree_trans *trans, struct bch_dev *ca, u64 start, u64 end, enum bch_data_type type, - u64 *bucket, unsigned *bucket_sectors) + u64 *bucket, unsigned *bucket_sectors, + unsigned flags) { do { u64 b = sector_to_bucket(ca, start); @@ -1172,7 +1171,7 @@ static int bch2_trans_mark_metadata_sectors(struct btree_trans *trans, if (b != *bucket && *bucket_sectors) { int ret = bch2_trans_mark_metadata_bucket(trans, ca, *bucket, - type, *bucket_sectors); + type, *bucket_sectors, flags); if (ret) return ret; @@ -1188,7 +1187,7 @@ static int bch2_trans_mark_metadata_sectors(struct btree_trans *trans, } static int __bch2_trans_mark_dev_sb(struct btree_trans *trans, - struct bch_dev *ca) + struct bch_dev *ca, unsigned flags) { struct bch_sb_layout *layout = &ca->disk_sb.sb->layout; u64 bucket = 0; @@ -1201,21 +1200,21 @@ static int __bch2_trans_mark_dev_sb(struct btree_trans *trans, if (offset == BCH_SB_SECTOR) { ret = bch2_trans_mark_metadata_sectors(trans, ca, 0, BCH_SB_SECTOR, - BCH_DATA_sb, &bucket, &bucket_sectors); + BCH_DATA_sb, &bucket, &bucket_sectors, flags); if (ret) return ret; } ret = bch2_trans_mark_metadata_sectors(trans, ca, offset, offset + (1 << layout->sb_max_size_bits), - BCH_DATA_sb, &bucket, &bucket_sectors); + BCH_DATA_sb, &bucket, &bucket_sectors, flags); if (ret) return ret; } if (bucket_sectors) { ret = bch2_trans_mark_metadata_bucket(trans, ca, - bucket, BCH_DATA_sb, bucket_sectors); + bucket, BCH_DATA_sb, bucket_sectors, flags); if (ret) return ret; } @@ -1223,7 +1222,7 @@ static int __bch2_trans_mark_dev_sb(struct btree_trans *trans, for (i = 0; i < ca->journal.nr; i++) { ret = bch2_trans_mark_metadata_bucket(trans, ca, ca->journal.buckets[i], - BCH_DATA_journal, ca->mi.bucket_size); + BCH_DATA_journal, ca->mi.bucket_size, flags); if (ret) return ret; } @@ -1231,18 +1230,18 @@ static int __bch2_trans_mark_dev_sb(struct btree_trans *trans, return 0; } -int bch2_trans_mark_dev_sb(struct bch_fs *c, struct bch_dev *ca) +int bch2_trans_mark_dev_sb(struct bch_fs *c, struct bch_dev *ca, unsigned flags) { - int ret = bch2_trans_run(c, __bch2_trans_mark_dev_sb(trans, ca)); - + int ret = bch2_trans_run(c, + __bch2_trans_mark_dev_sb(trans, ca, flags)); bch_err_fn(c, ret); return ret; } -int bch2_trans_mark_dev_sbs(struct bch_fs *c) +int bch2_trans_mark_dev_sbs_flags(struct bch_fs *c, unsigned flags) { for_each_online_member(c, ca) { - int ret = bch2_trans_mark_dev_sb(c, ca); + int ret = bch2_trans_mark_dev_sb(c, ca, flags); if (ret) { percpu_ref_put(&ca->ref); return ret; @@ -1252,6 +1251,11 @@ int bch2_trans_mark_dev_sbs(struct bch_fs *c) return 0; } +int bch2_trans_mark_dev_sbs(struct bch_fs *c) +{ + return bch2_trans_mark_dev_sbs_flags(c, BTREE_TRIGGER_TRANSACTIONAL); +} + /* Disk reservations: */ #define SECTORS_CACHE 1024 diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index f9af5adabe83..bae76e58311d 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -337,10 +337,6 @@ int bch2_check_bucket_ref(struct btree_trans *, struct bkey_s_c, const struct bch_extent_ptr *, s64, enum bch_data_type, u8, u8, u32); -int bch2_mark_metadata_bucket(struct bch_fs *, struct bch_dev *, - size_t, enum bch_data_type, unsigned, - struct gc_pos, unsigned); - int bch2_trigger_extent(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, unsigned); int bch2_trigger_reservation(struct btree_trans *, enum btree_id, unsigned, @@ -362,9 +358,10 @@ void bch2_trans_account_disk_usage_change(struct btree_trans *); void bch2_trans_fs_usage_revert(struct btree_trans *, struct replicas_delta_list *); int bch2_trans_fs_usage_apply(struct btree_trans *, struct replicas_delta_list *); -int bch2_trans_mark_metadata_bucket(struct btree_trans *, struct bch_dev *, - size_t, enum bch_data_type, unsigned); -int bch2_trans_mark_dev_sb(struct bch_fs *, struct bch_dev *); +int bch2_trans_mark_metadata_bucket(struct btree_trans *, struct bch_dev *, u64, + enum bch_data_type, unsigned, unsigned); +int bch2_trans_mark_dev_sb(struct bch_fs *, struct bch_dev *, unsigned); +int bch2_trans_mark_dev_sbs_flags(struct bch_fs *, unsigned); int bch2_trans_mark_dev_sbs(struct bch_fs *); static inline bool is_superblock_bucket(struct bch_dev *ca, u64 b) diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index d302d8e83678..fc02cb4ee74f 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -946,7 +946,7 @@ static int __bch2_set_nr_journal_buckets(struct bch_dev *ca, unsigned nr, ret = bch2_trans_run(c, bch2_trans_mark_metadata_bucket(trans, ca, ob[nr_got]->bucket, BCH_DATA_journal, - ca->mi.bucket_size)); + ca->mi.bucket_size, BTREE_TRIGGER_TRANSACTIONAL)); if (ret) { bch2_open_bucket_put(c, ob[nr_got]); bch_err_msg(c, ret, "marking new journal buckets"); @@ -1026,7 +1026,8 @@ err_unblock: for (i = 0; i < nr_got; i++) bch2_trans_run(c, bch2_trans_mark_metadata_bucket(trans, ca, - bu[i], BCH_DATA_free, 0)); + bu[i], BCH_DATA_free, 0, + BTREE_TRIGGER_TRANSACTIONAL)); err_free: if (!new_fs) for (i = 0; i < nr_got; i++) diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index ca2f3debd7fd..9073915f22a1 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -1822,7 +1822,7 @@ have_slot: bch2_dev_usage_journal_reserve(c); - ret = bch2_trans_mark_dev_sb(c, ca); + ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_TRANSACTIONAL); bch_err_msg(ca, ret, "marking new superblock"); if (ret) goto err_late; @@ -1887,7 +1887,7 @@ int bch2_dev_online(struct bch_fs *c, const char *path) ca = bch_dev_locked(c, dev_idx); - ret = bch2_trans_mark_dev_sb(c, ca); + ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_TRANSACTIONAL); bch_err_msg(c, ret, "bringing %s online: error from bch2_trans_mark_dev_sb", path); if (ret) goto err; @@ -1980,7 +1980,7 @@ int bch2_dev_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) if (ret) goto err; - ret = bch2_trans_mark_dev_sb(c, ca); + ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_TRANSACTIONAL); if (ret) goto err; -- cgit v1.2.3-59-g8ed1b From bf5f6a689b6037cb164e110dca37042aab9c0e09 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 17:54:00 -0400 Subject: bcachefs: __BTREE_ITER_ALL_SNAPSHOTS -> BTREE_ITER_SNAPSHOT_FIELD Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 4 ++-- fs/bcachefs/btree_iter.h | 2 +- fs/bcachefs/btree_types.h | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 39c89914cb53..ccf2e41b8aa7 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -258,7 +258,7 @@ static void bch2_btree_iter_verify(struct btree_iter *iter) BUG_ON((iter->flags & BTREE_ITER_IS_EXTENTS) && (iter->flags & BTREE_ITER_ALL_SNAPSHOTS)); - BUG_ON(!(iter->flags & __BTREE_ITER_ALL_SNAPSHOTS) && + BUG_ON(!(iter->flags & BTREE_ITER_SNAPSHOT_FIELD) && (iter->flags & BTREE_ITER_ALL_SNAPSHOTS) && !btree_type_has_snapshot_field(iter->btree_id)); @@ -2809,7 +2809,7 @@ void bch2_trans_node_iter_init(struct btree_trans *trans, unsigned flags) { flags |= BTREE_ITER_NOT_EXTENTS; - flags |= __BTREE_ITER_ALL_SNAPSHOTS; + flags |= BTREE_ITER_SNAPSHOT_FIELD; flags |= BTREE_ITER_ALL_SNAPSHOTS; bch2_trans_iter_init_common(trans, iter, btree_id, pos, locks_want, depth, diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 52030929da0a..57323d5e94a9 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -420,7 +420,7 @@ static inline unsigned __bch2_btree_iter_flags(struct btree_trans *trans, btree_id_is_extents(btree_id)) flags |= BTREE_ITER_IS_EXTENTS; - if (!(flags & __BTREE_ITER_ALL_SNAPSHOTS) && + if (!(flags & BTREE_ITER_SNAPSHOT_FIELD) && !btree_type_has_snapshot_field(btree_id)) flags &= ~BTREE_ITER_ALL_SNAPSHOTS; diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index c69b233c41bb..d4649129e5d3 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -210,7 +210,7 @@ static const __maybe_unused u16 BTREE_ITER_CACHED = 1 << 5; static const __maybe_unused u16 BTREE_ITER_WITH_KEY_CACHE = 1 << 6; static const __maybe_unused u16 BTREE_ITER_WITH_UPDATES = 1 << 7; static const __maybe_unused u16 BTREE_ITER_WITH_JOURNAL = 1 << 8; -static const __maybe_unused u16 __BTREE_ITER_ALL_SNAPSHOTS = 1 << 9; +static const __maybe_unused u16 BTREE_ITER_SNAPSHOT_FIELD = 1 << 9; static const __maybe_unused u16 BTREE_ITER_ALL_SNAPSHOTS = 1 << 10; static const __maybe_unused u16 BTREE_ITER_FILTER_SNAPSHOTS = 1 << 11; static const __maybe_unused u16 BTREE_ITER_NOPRESERVE = 1 << 12; -- cgit v1.2.3-59-g8ed1b From 5dd8c60e1e044816d789098ce2454a130e06b03d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 18:05:34 -0400 Subject: bcachefs: iter/update/trigger/str_hash flag cleanup Combine iter/update/trigger/str_hash flags into a single enum, and x-macroize them for a to_text() function later. These flags are all for a specific iter/key/update context, so it makes sense to group them together - iter/update/trigger flags were already given distinct bits, this cleans up and unifies that handling. Signed-off-by: Kent Overstreet --- fs/bcachefs/acl.c | 4 +- fs/bcachefs/alloc_background.c | 60 ++++++------- fs/bcachefs/alloc_foreground.c | 8 +- fs/bcachefs/backpointers.c | 14 +-- fs/bcachefs/bkey_methods.h | 65 +++----------- fs/bcachefs/btree_gc.c | 22 ++--- fs/bcachefs/btree_iter.c | 166 ++++++++++++++++++------------------ fs/bcachefs/btree_iter.h | 34 ++++---- fs/bcachefs/btree_key_cache.c | 28 +++--- fs/bcachefs/btree_trans_commit.c | 24 +++--- fs/bcachefs/btree_types.h | 109 ++++++++++++++++------- fs/bcachefs/btree_update.c | 88 +++++++++---------- fs/bcachefs/btree_update.h | 12 +-- fs/bcachefs/btree_update_interior.c | 20 ++--- fs/bcachefs/btree_write_buffer.c | 8 +- fs/bcachefs/buckets.c | 77 +++++++++-------- fs/bcachefs/buckets.h | 19 +++-- fs/bcachefs/data_update.c | 12 +-- fs/bcachefs/debug.c | 8 +- fs/bcachefs/dirent.c | 21 +++-- fs/bcachefs/dirent.h | 4 +- fs/bcachefs/ec.c | 26 +++--- fs/bcachefs/ec.h | 3 +- fs/bcachefs/extent_update.c | 2 +- fs/bcachefs/fs-common.c | 34 ++++---- fs/bcachefs/fs-io-buffered.c | 2 +- fs/bcachefs/fs-io-direct.c | 2 +- fs/bcachefs/fs-io-pagecache.c | 2 +- fs/bcachefs/fs-io.c | 4 +- fs/bcachefs/fs.c | 6 +- fs/bcachefs/fsck.c | 74 ++++++++-------- fs/bcachefs/inode.c | 30 +++---- fs/bcachefs/inode.h | 2 +- fs/bcachefs/io_misc.c | 10 +-- fs/bcachefs/io_read.c | 10 +-- fs/bcachefs/io_write.c | 16 ++-- fs/bcachefs/journal.c | 4 +- fs/bcachefs/journal_seq_blacklist.c | 2 +- fs/bcachefs/logged_ops.c | 2 +- fs/bcachefs/lru.c | 2 +- fs/bcachefs/migrate.c | 8 +- fs/bcachefs/move.c | 14 +-- fs/bcachefs/movinggc.c | 2 +- fs/bcachefs/opts.h | 2 +- fs/bcachefs/quota.c | 6 +- fs/bcachefs/rebalance.c | 8 +- fs/bcachefs/recovery.c | 14 +-- fs/bcachefs/reflink.c | 55 ++++++------ fs/bcachefs/reflink.h | 6 +- fs/bcachefs/snapshot.c | 40 ++++----- fs/bcachefs/str_hash.h | 45 ++++------ fs/bcachefs/subvolume.c | 18 ++-- fs/bcachefs/super.c | 18 ++-- fs/bcachefs/sysfs.c | 2 +- fs/bcachefs/tests.c | 16 ++-- fs/bcachefs/xattr.c | 6 +- 56 files changed, 652 insertions(+), 644 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/acl.c b/fs/bcachefs/acl.c index b4b112e33be4..250d6c6d3a3a 100644 --- a/fs/bcachefs/acl.c +++ b/fs/bcachefs/acl.c @@ -360,7 +360,7 @@ retry: ret = bch2_subvol_is_ro_trans(trans, inode->ei_subvol) ?: bch2_inode_peek(trans, &inode_iter, &inode_u, inode_inum(inode), - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto btree_err; @@ -411,7 +411,7 @@ int bch2_acl_chmod(struct btree_trans *trans, subvol_inum inum, struct posix_acl *acl = NULL; struct bkey_s_c k = bch2_hash_lookup(trans, &iter, bch2_xattr_hash_desc, - &hash_info, inum, &search, BTREE_ITER_INTENT); + &hash_info, inum, &search, BTREE_ITER_intent); int ret = bkey_err(k); if (ret) return bch2_err_matches(ret, ENOENT) ? 0 : ret; diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 6a7d249f108c..d3b86ce0fae6 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -437,9 +437,9 @@ bch2_trans_start_alloc_update(struct btree_trans *trans, struct btree_iter *iter int ret; k = bch2_bkey_get_iter(trans, iter, BTREE_ID_alloc, pos, - BTREE_ITER_WITH_UPDATES| - BTREE_ITER_CACHED| - BTREE_ITER_INTENT); + BTREE_ITER_with_updates| + BTREE_ITER_cached| + BTREE_ITER_intent); ret = bkey_err(k); if (unlikely(ret)) return ERR_PTR(ret); @@ -510,7 +510,7 @@ int bch2_bucket_gens_init(struct bch_fs *c) int ret; ret = for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, - BTREE_ITER_PREFETCH, k, ({ + BTREE_ITER_prefetch, k, ({ /* * Not a fsck error because this is checked/repaired by * bch2_check_alloc_key() which runs later: @@ -563,7 +563,7 @@ int bch2_alloc_read(struct bch_fs *c) if (c->sb.version_upgrade_complete >= bcachefs_metadata_version_bucket_gens) { ret = for_each_btree_key(trans, iter, BTREE_ID_bucket_gens, POS_MIN, - BTREE_ITER_PREFETCH, k, ({ + BTREE_ITER_prefetch, k, ({ u64 start = bucket_gens_pos_to_alloc(k.k->p, 0).offset; u64 end = bucket_gens_pos_to_alloc(bpos_nosnap_successor(k.k->p), 0).offset; @@ -589,7 +589,7 @@ int bch2_alloc_read(struct bch_fs *c) })); } else { ret = for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, - BTREE_ITER_PREFETCH, k, ({ + BTREE_ITER_prefetch, k, ({ /* * Not a fsck error because this is checked/repaired by * bch2_check_alloc_key() which runs later: @@ -657,7 +657,7 @@ static int bch2_bucket_do_index(struct btree_trans *trans, old = bch2_bkey_get_iter(trans, &iter, btree, bkey_start_pos(&k->k), - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = bkey_err(old); if (ret) return ret; @@ -701,8 +701,8 @@ static noinline int bch2_bucket_gen_update(struct btree_trans *trans, return ret; k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_bucket_gens, pos, - BTREE_ITER_INTENT| - BTREE_ITER_WITH_UPDATES); + BTREE_ITER_intent| + BTREE_ITER_with_updates); ret = bkey_err(k); if (ret) return ret; @@ -738,7 +738,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, struct bch_alloc_v4 old_a_convert; const struct bch_alloc_v4 *old_a = bch2_alloc_to_v4(old, &old_a_convert); - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { struct bch_alloc_v4 *new_a = bkey_s_to_alloc_v4(new).v; new_a->data_type = alloc_data_type(*new_a, new_a->data_type); @@ -802,7 +802,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, * not: */ - if ((flags & BTREE_TRIGGER_BUCKET_INVALIDATE) && + if ((flags & BTREE_TRIGGER_bucket_invalidate) && old_a->cached_sectors) { ret = bch2_update_cached_sectors_list(trans, new.k->p.inode, -((s64) old_a->cached_sectors)); @@ -811,12 +811,12 @@ int bch2_trigger_alloc(struct btree_trans *trans, } } - if ((flags & BTREE_TRIGGER_ATOMIC) && (flags & BTREE_TRIGGER_INSERT)) { + if ((flags & BTREE_TRIGGER_atomic) && (flags & BTREE_TRIGGER_insert)) { struct bch_alloc_v4 *new_a = bkey_s_to_alloc_v4(new).v; u64 journal_seq = trans->journal_res.seq; u64 bucket_journal_seq = new_a->journal_seq; - if ((flags & BTREE_TRIGGER_INSERT) && + if ((flags & BTREE_TRIGGER_insert) && data_type_is_empty(old_a->data_type) != data_type_is_empty(new_a->data_type) && new.k->type == KEY_TYPE_alloc_v4) { @@ -877,8 +877,8 @@ int bch2_trigger_alloc(struct btree_trans *trans, bch2_do_gc_gens(c); } - if ((flags & BTREE_TRIGGER_GC) && - (flags & BTREE_TRIGGER_BUCKET_INVALIDATE)) { + if ((flags & BTREE_TRIGGER_gc) && + (flags & BTREE_TRIGGER_bucket_invalidate)) { struct bch_alloc_v4 new_a_convert; const struct bch_alloc_v4 *new_a = bch2_alloc_to_v4(new.s_c, &new_a_convert); @@ -903,7 +903,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, } /* - * This synthesizes deleted extents for holes, similar to BTREE_ITER_SLOTS for + * This synthesizes deleted extents for holes, similar to BTREE_ITER_slots for * extents style btrees, but works on non-extents btrees: */ static struct bkey_s_c bch2_get_key_or_hole(struct btree_iter *iter, struct bpos end, struct bkey *hole) @@ -1401,13 +1401,13 @@ int bch2_check_alloc_info(struct bch_fs *c) int ret = 0; bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, POS_MIN, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); bch2_trans_iter_init(trans, &discard_iter, BTREE_ID_need_discard, POS_MIN, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); bch2_trans_iter_init(trans, &freespace_iter, BTREE_ID_freespace, POS_MIN, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); bch2_trans_iter_init(trans, &bucket_gens_iter, BTREE_ID_bucket_gens, POS_MIN, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); while (1) { struct bpos next; @@ -1469,13 +1469,13 @@ bkey_err: ret = for_each_btree_key(trans, iter, BTREE_ID_need_discard, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, bch2_check_discard_freespace_key(trans, &iter)); if (ret) goto err; bch2_trans_iter_init(trans, &iter, BTREE_ID_freespace, POS_MIN, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); while (1) { bch2_trans_begin(trans); k = bch2_btree_iter_peek(&iter); @@ -1505,7 +1505,7 @@ bkey_err: ret = for_each_btree_key_commit(trans, iter, BTREE_ID_bucket_gens, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_check_bucket_gens_key(trans, &iter, k)); err: @@ -1552,7 +1552,7 @@ static int bch2_check_alloc_to_lru_ref(struct btree_trans *trans, a_mut->v.io_time[READ] = atomic64_read(&c->io_clock[READ].now); ret = bch2_trans_update(trans, alloc_iter, - &a_mut->k_i, BTREE_TRIGGER_NORUN); + &a_mut->k_i, BTREE_TRIGGER_norun); if (ret) goto err; @@ -1591,7 +1591,7 @@ int bch2_check_alloc_to_lru_refs(struct bch_fs *c) { int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_alloc, - POS_MIN, BTREE_ITER_PREFETCH, k, + POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_check_alloc_to_lru_ref(trans, &iter))); bch_err_fn(c, ret); @@ -1693,7 +1693,7 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_alloc, need_discard_iter->pos, - BTREE_ITER_CACHED); + BTREE_ITER_cached); ret = bkey_err(k); if (ret) goto out; @@ -1817,7 +1817,7 @@ void bch2_do_discards(struct bch_fs *c) static int bch2_clear_bucket_needs_discard(struct btree_trans *trans, struct bpos bucket) { struct btree_iter iter; - bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, bucket, BTREE_ITER_INTENT); + bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, bucket, BTREE_ITER_intent); struct bkey_s_c k = bch2_btree_iter_peek_slot(&iter); int ret = bkey_err(k); if (ret) @@ -1952,7 +1952,7 @@ static int invalidate_one_bucket(struct btree_trans *trans, a->v.io_time[WRITE] = atomic64_read(&c->io_clock[WRITE].now); ret = bch2_trans_update(trans, &alloc_iter, &a->k_i, - BTREE_TRIGGER_BUCKET_INVALIDATE) ?: + BTREE_TRIGGER_bucket_invalidate) ?: bch2_trans_commit(trans, NULL, NULL, BCH_WATERMARK_btree| BCH_TRANS_COMMIT_no_enospc); @@ -2004,7 +2004,7 @@ static void bch2_do_invalidates_work(struct work_struct *work) ret = for_each_btree_key_upto(trans, iter, BTREE_ID_lru, lru_pos(ca->dev_idx, 0, 0), lru_pos(ca->dev_idx, U64_MAX, LRU_TIME_MAX), - BTREE_ITER_INTENT, k, + BTREE_ITER_intent, k, invalidate_one_bucket(trans, &iter, k, &nr_to_invalidate)); if (ret < 0) { @@ -2041,7 +2041,7 @@ int bch2_dev_freespace_init(struct bch_fs *c, struct bch_dev *ca, bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, POS(ca->dev_idx, max_t(u64, ca->mi.first_bucket, bucket_start)), - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); /* * Scan the alloc btree for every bucket on @ca, and add buckets to the * freespace/need_discard/need_gc_gens btrees as needed: diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index a1fc30adf912..b68e1fd782f3 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -300,7 +300,7 @@ static struct open_bucket *try_alloc_bucket(struct btree_trans *trans, struct bc k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_alloc, POS(ca->dev_idx, b), - BTREE_ITER_CACHED); + BTREE_ITER_cached); ret = bkey_err(k); if (ret) { ob = ERR_PTR(ret); @@ -344,7 +344,7 @@ static struct open_bucket *try_alloc_bucket(struct btree_trans *trans, struct bc ret = bch2_get_next_backpointer(trans, POS(ca->dev_idx, b), -1, &bp_pos, &bp, - BTREE_ITER_NOPRESERVE); + BTREE_ITER_nopreserve); if (ret) { ob = ERR_PTR(ret); goto err; @@ -404,7 +404,7 @@ bch2_bucket_alloc_early(struct btree_trans *trans, */ again: for_each_btree_key_norestart(trans, iter, BTREE_ID_alloc, POS(ca->dev_idx, alloc_cursor), - BTREE_ITER_SLOTS, k, ret) { + BTREE_ITER_slots, k, ret) { struct bch_alloc_v4 a_convert; const struct bch_alloc_v4 *a; @@ -420,7 +420,7 @@ again: continue; /* now check the cached key to serialize concurrent allocs of the bucket */ - ck = bch2_bkey_get_iter(trans, &citer, BTREE_ID_alloc, k.k->p, BTREE_ITER_CACHED); + ck = bch2_bkey_get_iter(trans, &citer, BTREE_ID_alloc, k.k->p, BTREE_ITER_cached); ret = bkey_err(ck); if (ret) break; diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 0ca53392f787..5586820fcf2c 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -170,9 +170,9 @@ int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *trans, k = bch2_bkey_get_iter(trans, &bp_iter, BTREE_ID_backpointers, bp_k->k.p, - BTREE_ITER_INTENT| - BTREE_ITER_SLOTS| - BTREE_ITER_WITH_UPDATES); + BTREE_ITER_intent| + BTREE_ITER_slots| + BTREE_ITER_with_updates); ret = bkey_err(k); if (ret) goto err; @@ -212,7 +212,7 @@ int bch2_get_next_backpointer(struct btree_trans *trans, if (gen >= 0) { k = bch2_bkey_get_iter(trans, &alloc_iter, BTREE_ID_alloc, - bucket, BTREE_ITER_CACHED|iter_flags); + bucket, BTREE_ITER_cached|iter_flags); ret = bkey_err(k); if (ret) goto out; @@ -759,7 +759,7 @@ static int bch2_get_btree_in_memory_pos(struct btree_trans *trans, __for_each_btree_node(trans, iter, btree, btree == start.btree ? start.pos : POS_MIN, - 0, depth, BTREE_ITER_PREFETCH, b, ret) { + 0, depth, BTREE_ITER_prefetch, b, ret) { mem_may_pin -= btree_buf_bytes(b); if (mem_may_pin <= 0) { c->btree_cache.pinned_nodes_end = *end = @@ -794,7 +794,7 @@ static int bch2_check_extents_to_backpointers_pass(struct btree_trans *trans, while (level >= depth) { struct btree_iter iter; bch2_trans_node_iter_init(trans, &iter, btree_id, POS_MIN, 0, level, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); ret = for_each_btree_key_continue(trans, iter, 0, k, ({ check_extent_to_backpointers(trans, s, btree_id, level, k) ?: @@ -917,7 +917,7 @@ static int bch2_check_backpointers_to_extents_pass(struct btree_trans *trans, struct bpos last_flushed_pos = SPOS_MAX; return for_each_btree_key_commit(trans, iter, BTREE_ID_backpointers, - POS_MIN, BTREE_ITER_PREFETCH, k, + POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_one_backpointer(trans, start, end, bkey_s_c_to_backpointer(k), diff --git a/fs/bcachefs/bkey_methods.h b/fs/bcachefs/bkey_methods.h index 03efe8ee565a..7351c0d38534 100644 --- a/fs/bcachefs/bkey_methods.h +++ b/fs/bcachefs/bkey_methods.h @@ -29,7 +29,8 @@ struct bkey_ops { bool (*key_normalize)(struct bch_fs *, struct bkey_s); bool (*key_merge)(struct bch_fs *, struct bkey_s, struct bkey_s_c); int (*trigger)(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); void (*compat)(enum btree_id id, unsigned version, unsigned big_endian, int write, struct bkey_s); @@ -76,56 +77,10 @@ static inline bool bch2_bkey_maybe_mergable(const struct bkey *l, const struct b bool bch2_bkey_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); -enum btree_update_flags { - __BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE = __BTREE_ITER_FLAGS_END, - __BTREE_UPDATE_NOJOURNAL, - __BTREE_UPDATE_KEY_CACHE_RECLAIM, - - __BTREE_TRIGGER_NORUN, - __BTREE_TRIGGER_TRANSACTIONAL, - __BTREE_TRIGGER_ATOMIC, - __BTREE_TRIGGER_GC, - __BTREE_TRIGGER_INSERT, - __BTREE_TRIGGER_OVERWRITE, - __BTREE_TRIGGER_BUCKET_INVALIDATE, -}; - -#define BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE (1U << __BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) -#define BTREE_UPDATE_NOJOURNAL (1U << __BTREE_UPDATE_NOJOURNAL) -#define BTREE_UPDATE_KEY_CACHE_RECLAIM (1U << __BTREE_UPDATE_KEY_CACHE_RECLAIM) - -/* Don't run triggers at all */ -#define BTREE_TRIGGER_NORUN (1U << __BTREE_TRIGGER_NORUN) - -/* - * If set, we're running transactional triggers as part of a transaction commit: - * triggers may generate new updates - * - * If cleared, and either BTREE_TRIGGER_INSERT|BTREE_TRIGGER_OVERWRITE are set, - * we're running atomic triggers during a transaction commit: we have our - * journal reservation, we're holding btree node write locks, and we know the - * transaction is going to commit (returning an error here is a fatal error, - * causing us to go emergency read-only) - */ -#define BTREE_TRIGGER_TRANSACTIONAL (1U << __BTREE_TRIGGER_TRANSACTIONAL) -#define BTREE_TRIGGER_ATOMIC (1U << __BTREE_TRIGGER_ATOMIC) - -/* We're in gc/fsck: running triggers to recalculate e.g. disk usage */ -#define BTREE_TRIGGER_GC (1U << __BTREE_TRIGGER_GC) - -/* @new is entering the btree */ -#define BTREE_TRIGGER_INSERT (1U << __BTREE_TRIGGER_INSERT) - -/* @old is leaving the btree */ -#define BTREE_TRIGGER_OVERWRITE (1U << __BTREE_TRIGGER_OVERWRITE) - -/* signal from bucket invalidate path to alloc trigger */ -#define BTREE_TRIGGER_BUCKET_INVALIDATE (1U << __BTREE_TRIGGER_BUCKET_INVALIDATE) - static inline int bch2_key_trigger(struct btree_trans *trans, enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { const struct bkey_ops *ops = bch2_bkey_type_ops(old.k->type ?: new.k->type); @@ -135,8 +90,9 @@ static inline int bch2_key_trigger(struct btree_trans *trans, } static inline int bch2_key_trigger_old(struct btree_trans *trans, - enum btree_id btree_id, unsigned level, - struct bkey_s_c old, unsigned flags) + enum btree_id btree_id, unsigned level, + struct bkey_s_c old, + enum btree_iter_update_trigger_flags flags) { struct bkey_i deleted; @@ -144,12 +100,13 @@ static inline int bch2_key_trigger_old(struct btree_trans *trans, deleted.k.p = old.k->p; return bch2_key_trigger(trans, btree_id, level, old, bkey_i_to_s(&deleted), - BTREE_TRIGGER_OVERWRITE|flags); + BTREE_TRIGGER_overwrite|flags); } static inline int bch2_key_trigger_new(struct btree_trans *trans, - enum btree_id btree_id, unsigned level, - struct bkey_s new, unsigned flags) + enum btree_id btree_id, unsigned level, + struct bkey_s new, + enum btree_iter_update_trigger_flags flags) { struct bkey_i deleted; @@ -157,7 +114,7 @@ static inline int bch2_key_trigger_new(struct btree_trans *trans, deleted.k.p = new.k->p; return bch2_key_trigger(trans, btree_id, level, bkey_i_to_s_c(&deleted), new, - BTREE_TRIGGER_INSERT|flags); + BTREE_TRIGGER_insert|flags); } void bch2_bkey_renumber(enum btree_node_type, struct bkey_packed *, int); diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 512be84aec23..c98b0ce45524 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -856,7 +856,7 @@ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, ret = commit_do(trans, NULL, NULL, 0, bch2_key_trigger(trans, btree_id, level, old, - unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_GC)); + unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_gc)); fsck_err: err: printbuf_exit(&buf); @@ -900,7 +900,7 @@ static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id, gc_pos_set(c, gc_pos_btree(btree_id, POS_MIN, 0)); __for_each_btree_node(trans, iter, btree_id, POS_MIN, - 0, target_depth, BTREE_ITER_PREFETCH, b, ret) { + 0, target_depth, BTREE_ITER_prefetch, b, ret) { bch2_verify_btree_nr_keys(b); gc_pos_set(c, gc_pos_btree_node(b)); @@ -1045,7 +1045,7 @@ static int bch2_mark_superblocks(struct bch_fs *c) mutex_lock(&c->sb_lock); gc_pos_set(c, gc_phase(GC_PHASE_SB)); - int ret = bch2_trans_mark_dev_sbs_flags(c, BTREE_TRIGGER_GC); + int ret = bch2_trans_mark_dev_sbs_flags(c, BTREE_TRIGGER_gc); mutex_unlock(&c->sb_lock); return ret; } @@ -1304,7 +1304,7 @@ static int bch2_alloc_write_key(struct btree_trans *trans, if (a->v.data_type == BCH_DATA_cached && !a->v.io_time[READ]) a->v.io_time[READ] = max_t(u64, 1, atomic64_read(&c->io_clock[READ].now)); - ret = bch2_trans_update(trans, iter, &a->k_i, BTREE_TRIGGER_NORUN); + ret = bch2_trans_update(trans, iter, &a->k_i, BTREE_TRIGGER_norun); fsck_err: return ret; } @@ -1318,7 +1318,7 @@ static int bch2_gc_alloc_done(struct bch_fs *c) for_each_btree_key_upto_commit(trans, iter, BTREE_ID_alloc, POS(ca->dev_idx, ca->mi.first_bucket), POS(ca->dev_idx, ca->mi.nbuckets - 1), - BTREE_ITER_SLOTS|BTREE_ITER_PREFETCH, k, + BTREE_ITER_slots|BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, bch2_alloc_write_key(trans, &iter, k))); if (ret) { @@ -1350,7 +1350,7 @@ static int bch2_gc_alloc_start(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, - BTREE_ITER_PREFETCH, k, ({ + BTREE_ITER_prefetch, k, ({ struct bch_dev *ca = bch_dev_bkey_exists(c, k.k->p.inode); struct bucket *g = gc_bucket(ca, k.k->p.offset); @@ -1435,7 +1435,7 @@ static int bch2_gc_reflink_done(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_reflink, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_gc_write_reflink_key(trans, &iter, k, &idx))); c->reflink_gc_nr = 0; @@ -1448,7 +1448,7 @@ static int bch2_gc_reflink_start(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_reflink, POS_MIN, - BTREE_ITER_PREFETCH, k, ({ + BTREE_ITER_prefetch, k, ({ const __le64 *refcount = bkey_refcount_c(k); if (!refcount) @@ -1538,7 +1538,7 @@ static int bch2_gc_stripes_done(struct bch_fs *c) return bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_stripes, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_gc_write_stripes_key(trans, &iter, k))); } @@ -1762,7 +1762,7 @@ int bch2_gc_gens(struct bch_fs *c) ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, i, POS_MIN, - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, @@ -1774,7 +1774,7 @@ int bch2_gc_gens(struct bch_fs *c) ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_alloc, POS_MIN, - BTREE_ITER_PREFETCH, + BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index ccf2e41b8aa7..db47063e5844 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -61,7 +61,7 @@ static inline int btree_path_cmp(const struct btree_path *l, static inline struct bpos bkey_successor(struct btree_iter *iter, struct bpos p) { /* Are we iterating over keys in all snapshots? */ - if (iter->flags & BTREE_ITER_ALL_SNAPSHOTS) { + if (iter->flags & BTREE_ITER_all_snapshots) { p = bpos_successor(p); } else { p = bpos_nosnap_successor(p); @@ -74,7 +74,7 @@ static inline struct bpos bkey_successor(struct btree_iter *iter, struct bpos p) static inline struct bpos bkey_predecessor(struct btree_iter *iter, struct bpos p) { /* Are we iterating over keys in all snapshots? */ - if (iter->flags & BTREE_ITER_ALL_SNAPSHOTS) { + if (iter->flags & BTREE_ITER_all_snapshots) { p = bpos_predecessor(p); } else { p = bpos_nosnap_predecessor(p); @@ -88,7 +88,7 @@ static inline struct bpos btree_iter_search_key(struct btree_iter *iter) { struct bpos pos = iter->pos; - if ((iter->flags & BTREE_ITER_IS_EXTENTS) && + if ((iter->flags & BTREE_ITER_is_extents) && !bkey_eq(pos, POS_MAX)) pos = bkey_successor(iter, pos); return pos; @@ -253,13 +253,13 @@ static void bch2_btree_iter_verify(struct btree_iter *iter) BUG_ON(iter->btree_id >= BTREE_ID_NR); - BUG_ON(!!(iter->flags & BTREE_ITER_CACHED) != btree_iter_path(trans, iter)->cached); + BUG_ON(!!(iter->flags & BTREE_ITER_cached) != btree_iter_path(trans, iter)->cached); - BUG_ON((iter->flags & BTREE_ITER_IS_EXTENTS) && - (iter->flags & BTREE_ITER_ALL_SNAPSHOTS)); + BUG_ON((iter->flags & BTREE_ITER_is_extents) && + (iter->flags & BTREE_ITER_all_snapshots)); - BUG_ON(!(iter->flags & BTREE_ITER_SNAPSHOT_FIELD) && - (iter->flags & BTREE_ITER_ALL_SNAPSHOTS) && + BUG_ON(!(iter->flags & BTREE_ITER_snapshot_field) && + (iter->flags & BTREE_ITER_all_snapshots) && !btree_type_has_snapshot_field(iter->btree_id)); if (iter->update_path) @@ -269,10 +269,10 @@ static void bch2_btree_iter_verify(struct btree_iter *iter) static void bch2_btree_iter_verify_entry_exit(struct btree_iter *iter) { - BUG_ON((iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) && + BUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && !iter->pos.snapshot); - BUG_ON(!(iter->flags & BTREE_ITER_ALL_SNAPSHOTS) && + BUG_ON(!(iter->flags & BTREE_ITER_all_snapshots) && iter->pos.snapshot != iter->snapshot); BUG_ON(bkey_lt(iter->pos, bkey_start_pos(&iter->k)) || @@ -289,7 +289,7 @@ static int bch2_btree_iter_verify_ret(struct btree_iter *iter, struct bkey_s_c k if (!bch2_debug_check_iterators) return 0; - if (!(iter->flags & BTREE_ITER_FILTER_SNAPSHOTS)) + if (!(iter->flags & BTREE_ITER_filter_snapshots)) return 0; if (bkey_err(k) || !k.k) @@ -300,8 +300,8 @@ static int bch2_btree_iter_verify_ret(struct btree_iter *iter, struct bkey_s_c k k.k->p.snapshot)); bch2_trans_iter_init(trans, ©, iter->btree_id, iter->pos, - BTREE_ITER_NOPRESERVE| - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_nopreserve| + BTREE_ITER_all_snapshots); prev = bch2_btree_iter_prev(©); if (!prev.k) goto out; @@ -897,7 +897,7 @@ static noinline int btree_node_iter_and_journal_peek(struct btree_trans *trans, bch2_bkey_buf_reassemble(out, c, k); - if ((flags & BTREE_ITER_PREFETCH) && + if ((flags & BTREE_ITER_prefetch) && c->opts.btree_node_prefetch) ret = btree_path_prefetch_j(trans, path, &jiter); @@ -944,7 +944,7 @@ static __always_inline int btree_path_down(struct btree_trans *trans, bch2_bkey_buf_unpack(&tmp, c, l->b, k); - if ((flags & BTREE_ITER_PREFETCH) && + if ((flags & BTREE_ITER_prefetch) && c->opts.btree_node_prefetch) { ret = btree_path_prefetch(trans, path); if (ret) @@ -1659,8 +1659,8 @@ btree_path_idx_t bch2_path_get(struct btree_trans *trans, unsigned flags, unsigned long ip) { struct btree_path *path; - bool cached = flags & BTREE_ITER_CACHED; - bool intent = flags & BTREE_ITER_INTENT; + bool cached = flags & BTREE_ITER_cached; + bool intent = flags & BTREE_ITER_intent; struct trans_for_each_path_inorder_iter iter; btree_path_idx_t path_pos = 0, path_idx; @@ -1708,7 +1708,7 @@ btree_path_idx_t bch2_path_get(struct btree_trans *trans, trans->paths_sorted = false; } - if (!(flags & BTREE_ITER_NOPRESERVE)) + if (!(flags & BTREE_ITER_nopreserve)) path->preserve = true; if (path->intent_ref) @@ -1786,7 +1786,7 @@ bch2_btree_iter_traverse(struct btree_iter *iter) iter->path = bch2_btree_path_set_pos(trans, iter->path, btree_iter_search_key(iter), - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); ret = bch2_btree_path_traverse(iter->trans, iter->path, iter->flags); @@ -1825,7 +1825,7 @@ struct btree *bch2_btree_iter_peek_node(struct btree_iter *iter) iter->k.p = iter->pos = b->key.k.p; iter->path = bch2_btree_path_set_pos(trans, iter->path, b->key.k.p, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); btree_path_set_should_be_locked(btree_iter_path(trans, iter)); out: @@ -1892,7 +1892,7 @@ struct btree *bch2_btree_iter_next_node(struct btree_iter *iter) */ iter->path = bch2_btree_path_set_pos(trans, iter->path, bpos_successor(iter->pos), - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); path = btree_iter_path(trans, iter); @@ -1910,7 +1910,7 @@ struct btree *bch2_btree_iter_next_node(struct btree_iter *iter) iter->k.p = iter->pos = b->key.k.p; iter->path = bch2_btree_path_set_pos(trans, iter->path, b->key.k.p, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); btree_path_set_should_be_locked(btree_iter_path(trans, iter)); EBUG_ON(btree_iter_path(trans, iter)->uptodate); @@ -1929,11 +1929,11 @@ err: inline bool bch2_btree_iter_advance(struct btree_iter *iter) { struct bpos pos = iter->k.p; - bool ret = !(iter->flags & BTREE_ITER_ALL_SNAPSHOTS + bool ret = !(iter->flags & BTREE_ITER_all_snapshots ? bpos_eq(pos, SPOS_MAX) : bkey_eq(pos, SPOS_MAX)); - if (ret && !(iter->flags & BTREE_ITER_IS_EXTENTS)) + if (ret && !(iter->flags & BTREE_ITER_is_extents)) pos = bkey_successor(iter, pos); bch2_btree_iter_set_pos(iter, pos); return ret; @@ -1942,11 +1942,11 @@ inline bool bch2_btree_iter_advance(struct btree_iter *iter) inline bool bch2_btree_iter_rewind(struct btree_iter *iter) { struct bpos pos = bkey_start_pos(&iter->k); - bool ret = !(iter->flags & BTREE_ITER_ALL_SNAPSHOTS + bool ret = !(iter->flags & BTREE_ITER_all_snapshots ? bpos_eq(pos, POS_MIN) : bkey_eq(pos, POS_MIN)); - if (ret && !(iter->flags & BTREE_ITER_IS_EXTENTS)) + if (ret && !(iter->flags & BTREE_ITER_is_extents)) pos = bkey_predecessor(iter, pos); bch2_btree_iter_set_pos(iter, pos); return ret; @@ -2057,7 +2057,7 @@ struct bkey_s_c btree_trans_peek_key_cache(struct btree_iter *iter, struct bpos struct bkey_s_c k; int ret; - if ((iter->flags & BTREE_ITER_KEY_CACHE_FILL) && + if ((iter->flags & BTREE_ITER_key_cache_fill) && bpos_eq(iter->pos, pos)) return bkey_s_c_null; @@ -2066,17 +2066,17 @@ struct bkey_s_c btree_trans_peek_key_cache(struct btree_iter *iter, struct bpos if (!iter->key_cache_path) iter->key_cache_path = bch2_path_get(trans, iter->btree_id, pos, - iter->flags & BTREE_ITER_INTENT, 0, - iter->flags|BTREE_ITER_CACHED| - BTREE_ITER_CACHED_NOFILL, + iter->flags & BTREE_ITER_intent, 0, + iter->flags|BTREE_ITER_cached| + BTREE_ITER_cached_nofill, _THIS_IP_); iter->key_cache_path = bch2_btree_path_set_pos(trans, iter->key_cache_path, pos, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); ret = bch2_btree_path_traverse(trans, iter->key_cache_path, - iter->flags|BTREE_ITER_CACHED) ?: + iter->flags|BTREE_ITER_cached) ?: bch2_btree_path_relock(trans, btree_iter_path(trans, iter), _THIS_IP_); if (unlikely(ret)) return bkey_s_c_err(ret); @@ -2104,7 +2104,7 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp struct btree_path_level *l; iter->path = bch2_btree_path_set_pos(trans, iter->path, search_key, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); ret = bch2_btree_path_traverse(trans, iter->path, iter->flags); @@ -2129,7 +2129,7 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp k = btree_path_level_peek_all(trans->c, l, &iter->k); - if (unlikely(iter->flags & BTREE_ITER_WITH_KEY_CACHE) && + if (unlikely(iter->flags & BTREE_ITER_with_key_cache) && k.k && (k2 = btree_trans_peek_key_cache(iter, k.k->p)).k) { k = k2; @@ -2140,10 +2140,10 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp } } - if (unlikely(iter->flags & BTREE_ITER_WITH_JOURNAL)) + if (unlikely(iter->flags & BTREE_ITER_with_journal)) k = btree_trans_peek_journal(trans, iter, k); - if (unlikely((iter->flags & BTREE_ITER_WITH_UPDATES) && + if (unlikely((iter->flags & BTREE_ITER_with_updates) && trans->nr_updates)) bch2_btree_trans_peek_updates(trans, iter, &k); @@ -2195,11 +2195,11 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e struct bpos iter_pos; int ret; - EBUG_ON((iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) && bkey_eq(end, POS_MAX)); + EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bkey_eq(end, POS_MAX)); if (iter->update_path) { bch2_path_put_nokeep(trans, iter->update_path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); iter->update_path = 0; } @@ -2222,7 +2222,7 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e * isn't monotonically increasing before FILTER_SNAPSHOTS, and * that's what we check against in extents mode: */ - if (unlikely(!(iter->flags & BTREE_ITER_IS_EXTENTS) + if (unlikely(!(iter->flags & BTREE_ITER_is_extents) ? bkey_gt(k.k->p, end) : k.k->p.inode > end.inode)) goto end; @@ -2230,13 +2230,13 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e if (iter->update_path && !bkey_eq(trans->paths[iter->update_path].pos, k.k->p)) { bch2_path_put_nokeep(trans, iter->update_path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); iter->update_path = 0; } - if ((iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) && - (iter->flags & BTREE_ITER_INTENT) && - !(iter->flags & BTREE_ITER_IS_EXTENTS) && + if ((iter->flags & BTREE_ITER_filter_snapshots) && + (iter->flags & BTREE_ITER_intent) && + !(iter->flags & BTREE_ITER_is_extents) && !iter->update_path) { struct bpos pos = k.k->p; @@ -2251,12 +2251,12 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e * advance, same as on exit for iter->path, but only up * to snapshot */ - __btree_path_get(trans->paths + iter->path, iter->flags & BTREE_ITER_INTENT); + __btree_path_get(trans->paths + iter->path, iter->flags & BTREE_ITER_intent); iter->update_path = iter->path; iter->update_path = bch2_btree_path_set_pos(trans, iter->update_path, pos, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, _THIS_IP_); ret = bch2_btree_path_traverse(trans, iter->update_path, iter->flags); if (unlikely(ret)) { @@ -2269,7 +2269,7 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e * We can never have a key in a leaf node at POS_MAX, so * we don't have to check these successor() calls: */ - if ((iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) && + if ((iter->flags & BTREE_ITER_filter_snapshots) && !bch2_snapshot_is_ancestor(trans->c, iter->snapshot, k.k->p.snapshot)) { @@ -2278,7 +2278,7 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e } if (bkey_whiteout(k.k) && - !(iter->flags & BTREE_ITER_ALL_SNAPSHOTS)) { + !(iter->flags & BTREE_ITER_all_snapshots)) { search_key = bkey_successor(iter, k.k->p); continue; } @@ -2288,12 +2288,12 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e * equal to the key we just returned - except extents can * straddle iter->pos: */ - if (!(iter->flags & BTREE_ITER_IS_EXTENTS)) + if (!(iter->flags & BTREE_ITER_is_extents)) iter_pos = k.k->p; else iter_pos = bkey_max(iter->pos, bkey_start_pos(k.k)); - if (unlikely(!(iter->flags & BTREE_ITER_IS_EXTENTS) + if (unlikely(!(iter->flags & BTREE_ITER_is_extents) ? bkey_gt(iter_pos, end) : bkey_ge(iter_pos, end))) goto end; @@ -2304,7 +2304,7 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e iter->pos = iter_pos; iter->path = bch2_btree_path_set_pos(trans, iter->path, k.k->p, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); btree_path_set_should_be_locked(btree_iter_path(trans, iter)); @@ -2317,7 +2317,7 @@ out_no_locked: btree_path_set_should_be_locked(trans->paths + iter->update_path); } - if (!(iter->flags & BTREE_ITER_ALL_SNAPSHOTS)) + if (!(iter->flags & BTREE_ITER_all_snapshots)) iter->pos.snapshot = iter->snapshot; ret = bch2_btree_iter_verify_ret(iter, k); @@ -2370,18 +2370,18 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) EBUG_ON(btree_iter_path(trans, iter)->cached || btree_iter_path(trans, iter)->level); - if (iter->flags & BTREE_ITER_WITH_JOURNAL) + if (iter->flags & BTREE_ITER_with_journal) return bkey_s_c_err(-BCH_ERR_btree_iter_with_journal_not_supported); bch2_btree_iter_verify(iter); bch2_btree_iter_verify_entry_exit(iter); - if (iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) + if (iter->flags & BTREE_ITER_filter_snapshots) search_key.snapshot = U32_MAX; while (1) { iter->path = bch2_btree_path_set_pos(trans, iter->path, search_key, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); ret = bch2_btree_path_traverse(trans, iter->path, iter->flags); @@ -2396,17 +2396,17 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) k = btree_path_level_peek(trans, path, &path->l[0], &iter->k); if (!k.k || - ((iter->flags & BTREE_ITER_IS_EXTENTS) + ((iter->flags & BTREE_ITER_is_extents) ? bpos_ge(bkey_start_pos(k.k), search_key) : bpos_gt(k.k->p, search_key))) k = btree_path_level_prev(trans, path, &path->l[0], &iter->k); - if (unlikely((iter->flags & BTREE_ITER_WITH_UPDATES) && + if (unlikely((iter->flags & BTREE_ITER_with_updates) && trans->nr_updates)) bch2_btree_trans_peek_prev_updates(trans, iter, &k); if (likely(k.k)) { - if (iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) { + if (iter->flags & BTREE_ITER_filter_snapshots) { if (k.k->p.snapshot == iter->snapshot) goto got_key; @@ -2417,7 +2417,7 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) */ if (saved_path && !bkey_eq(k.k->p, saved_k.p)) { bch2_path_put_nokeep(trans, iter->path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); iter->path = saved_path; saved_path = 0; iter->k = saved_k; @@ -2430,9 +2430,9 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) k.k->p.snapshot)) { if (saved_path) bch2_path_put_nokeep(trans, saved_path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); saved_path = btree_path_clone(trans, iter->path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); path = btree_iter_path(trans, iter); saved_k = *k.k; saved_v = k.v; @@ -2443,9 +2443,9 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) } got_key: if (bkey_whiteout(k.k) && - !(iter->flags & BTREE_ITER_ALL_SNAPSHOTS)) { + !(iter->flags & BTREE_ITER_all_snapshots)) { search_key = bkey_predecessor(iter, k.k->p); - if (iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) + if (iter->flags & BTREE_ITER_filter_snapshots) search_key.snapshot = U32_MAX; continue; } @@ -2469,11 +2469,11 @@ got_key: if (bkey_lt(k.k->p, iter->pos)) iter->pos = k.k->p; - if (iter->flags & BTREE_ITER_FILTER_SNAPSHOTS) + if (iter->flags & BTREE_ITER_filter_snapshots) iter->pos.snapshot = iter->snapshot; out_no_locked: if (saved_path) - bch2_path_put_nokeep(trans, saved_path, iter->flags & BTREE_ITER_INTENT); + bch2_path_put_nokeep(trans, saved_path, iter->flags & BTREE_ITER_intent); bch2_btree_iter_verify_entry_exit(iter); bch2_btree_iter_verify(iter); @@ -2505,10 +2505,10 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) bch2_btree_iter_verify(iter); bch2_btree_iter_verify_entry_exit(iter); - EBUG_ON(btree_iter_path(trans, iter)->level && (iter->flags & BTREE_ITER_WITH_KEY_CACHE)); + EBUG_ON(btree_iter_path(trans, iter)->level && (iter->flags & BTREE_ITER_with_key_cache)); /* extents can't span inode numbers: */ - if ((iter->flags & BTREE_ITER_IS_EXTENTS) && + if ((iter->flags & BTREE_ITER_is_extents) && unlikely(iter->pos.offset == KEY_OFFSET_MAX)) { if (iter->pos.inode == KEY_INODE_MAX) return bkey_s_c_null; @@ -2518,7 +2518,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) search_key = btree_iter_search_key(iter); iter->path = bch2_btree_path_set_pos(trans, iter->path, search_key, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); ret = bch2_btree_path_traverse(trans, iter->path, iter->flags); @@ -2527,22 +2527,22 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) goto out_no_locked; } - if ((iter->flags & BTREE_ITER_CACHED) || - !(iter->flags & (BTREE_ITER_IS_EXTENTS|BTREE_ITER_FILTER_SNAPSHOTS))) { + if ((iter->flags & BTREE_ITER_cached) || + !(iter->flags & (BTREE_ITER_is_extents|BTREE_ITER_filter_snapshots))) { k = bkey_s_c_null; - if (unlikely((iter->flags & BTREE_ITER_WITH_UPDATES) && + if (unlikely((iter->flags & BTREE_ITER_with_updates) && trans->nr_updates)) { bch2_btree_trans_peek_slot_updates(trans, iter, &k); if (k.k) goto out; } - if (unlikely(iter->flags & BTREE_ITER_WITH_JOURNAL) && + if (unlikely(iter->flags & BTREE_ITER_with_journal) && (k = btree_trans_peek_slot_journal(trans, iter)).k) goto out; - if (unlikely(iter->flags & BTREE_ITER_WITH_KEY_CACHE) && + if (unlikely(iter->flags & BTREE_ITER_with_key_cache) && (k = btree_trans_peek_key_cache(iter, iter->pos)).k) { if (!bkey_err(k)) iter->k = *k.k; @@ -2557,12 +2557,12 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) struct bpos next; struct bpos end = iter->pos; - if (iter->flags & BTREE_ITER_IS_EXTENTS) + if (iter->flags & BTREE_ITER_is_extents) end.offset = U64_MAX; EBUG_ON(btree_iter_path(trans, iter)->level); - if (iter->flags & BTREE_ITER_INTENT) { + if (iter->flags & BTREE_ITER_intent) { struct btree_iter iter2; bch2_trans_copy_iter(&iter2, iter); @@ -2593,7 +2593,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) bkey_init(&iter->k); iter->k.p = iter->pos; - if (iter->flags & BTREE_ITER_IS_EXTENTS) { + if (iter->flags & BTREE_ITER_is_extents) { bch2_key_resize(&iter->k, min_t(u64, KEY_SIZE_MAX, (next.inode == iter->pos.inode @@ -2777,13 +2777,13 @@ void bch2_trans_iter_exit(struct btree_trans *trans, struct btree_iter *iter) { if (iter->update_path) bch2_path_put_nokeep(trans, iter->update_path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); if (iter->path) bch2_path_put(trans, iter->path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); if (iter->key_cache_path) bch2_path_put(trans, iter->key_cache_path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); iter->path = 0; iter->update_path = 0; iter->key_cache_path = 0; @@ -2808,9 +2808,9 @@ void bch2_trans_node_iter_init(struct btree_trans *trans, unsigned depth, unsigned flags) { - flags |= BTREE_ITER_NOT_EXTENTS; - flags |= BTREE_ITER_SNAPSHOT_FIELD; - flags |= BTREE_ITER_ALL_SNAPSHOTS; + flags |= BTREE_ITER_not_extents; + flags |= BTREE_ITER_snapshot_field; + flags |= BTREE_ITER_all_snapshots; bch2_trans_iter_init_common(trans, iter, btree_id, pos, locks_want, depth, __bch2_btree_iter_flags(trans, btree_id, flags), @@ -2833,9 +2833,9 @@ void bch2_trans_copy_iter(struct btree_iter *dst, struct btree_iter *src) dst->ip_allocated = _RET_IP_; #endif if (src->path) - __btree_path_get(trans->paths + src->path, src->flags & BTREE_ITER_INTENT); + __btree_path_get(trans->paths + src->path, src->flags & BTREE_ITER_intent); if (src->update_path) - __btree_path_get(trans->paths + src->update_path, src->flags & BTREE_ITER_INTENT); + __btree_path_get(trans->paths + src->update_path, src->flags & BTREE_ITER_intent); dst->key_cache_path = 0; } diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 57323d5e94a9..abbf7bc8d9c5 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -386,10 +386,10 @@ static inline void bch2_btree_iter_set_pos(struct btree_iter *iter, struct bpos if (unlikely(iter->update_path)) bch2_path_put(trans, iter->update_path, - iter->flags & BTREE_ITER_INTENT); + iter->flags & BTREE_ITER_intent); iter->update_path = 0; - if (!(iter->flags & BTREE_ITER_ALL_SNAPSHOTS)) + if (!(iter->flags & BTREE_ITER_all_snapshots)) new_pos.snapshot = iter->snapshot; __bch2_btree_iter_set_pos(iter, new_pos); @@ -397,7 +397,7 @@ static inline void bch2_btree_iter_set_pos(struct btree_iter *iter, struct bpos static inline void bch2_btree_iter_set_pos_to_extent_start(struct btree_iter *iter) { - BUG_ON(!(iter->flags & BTREE_ITER_IS_EXTENTS)); + BUG_ON(!(iter->flags & BTREE_ITER_is_extents)); iter->pos = bkey_start_pos(&iter->k); } @@ -416,20 +416,20 @@ static inline unsigned __bch2_btree_iter_flags(struct btree_trans *trans, unsigned btree_id, unsigned flags) { - if (!(flags & (BTREE_ITER_ALL_SNAPSHOTS|BTREE_ITER_NOT_EXTENTS)) && + if (!(flags & (BTREE_ITER_all_snapshots|BTREE_ITER_not_extents)) && btree_id_is_extents(btree_id)) - flags |= BTREE_ITER_IS_EXTENTS; + flags |= BTREE_ITER_is_extents; - if (!(flags & BTREE_ITER_SNAPSHOT_FIELD) && + if (!(flags & BTREE_ITER_snapshot_field) && !btree_type_has_snapshot_field(btree_id)) - flags &= ~BTREE_ITER_ALL_SNAPSHOTS; + flags &= ~BTREE_ITER_all_snapshots; - if (!(flags & BTREE_ITER_ALL_SNAPSHOTS) && + if (!(flags & BTREE_ITER_all_snapshots) && btree_type_has_snapshots(btree_id)) - flags |= BTREE_ITER_FILTER_SNAPSHOTS; + flags |= BTREE_ITER_filter_snapshots; if (trans->journal_replay_not_finished) - flags |= BTREE_ITER_WITH_JOURNAL; + flags |= BTREE_ITER_with_journal; return flags; } @@ -439,10 +439,10 @@ static inline unsigned bch2_btree_iter_flags(struct btree_trans *trans, unsigned flags) { if (!btree_id_cached(trans->c, btree_id)) { - flags &= ~BTREE_ITER_CACHED; - flags &= ~BTREE_ITER_WITH_KEY_CACHE; - } else if (!(flags & BTREE_ITER_CACHED)) - flags |= BTREE_ITER_WITH_KEY_CACHE; + flags &= ~BTREE_ITER_cached; + flags &= ~BTREE_ITER_with_key_cache; + } else if (!(flags & BTREE_ITER_cached)) + flags |= BTREE_ITER_with_key_cache; return __bch2_btree_iter_flags(trans, btree_id, flags); } @@ -619,14 +619,14 @@ u32 bch2_trans_begin(struct btree_trans *); static inline struct bkey_s_c bch2_btree_iter_peek_prev_type(struct btree_iter *iter, unsigned flags) { - return flags & BTREE_ITER_SLOTS ? bch2_btree_iter_peek_slot(iter) : + return flags & BTREE_ITER_slots ? bch2_btree_iter_peek_slot(iter) : bch2_btree_iter_peek_prev(iter); } static inline struct bkey_s_c bch2_btree_iter_peek_type(struct btree_iter *iter, unsigned flags) { - return flags & BTREE_ITER_SLOTS ? bch2_btree_iter_peek_slot(iter) : + return flags & BTREE_ITER_slots ? bch2_btree_iter_peek_slot(iter) : bch2_btree_iter_peek(iter); } @@ -634,7 +634,7 @@ static inline struct bkey_s_c bch2_btree_iter_peek_upto_type(struct btree_iter * struct bpos end, unsigned flags) { - if (!(flags & BTREE_ITER_SLOTS)) + if (!(flags & BTREE_ITER_slots)) return bch2_btree_iter_peek_upto(iter, end); if (bkey_gt(iter->pos, end)) diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 3dea89c6cf7e..547ba9329678 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -383,9 +383,9 @@ static int btree_key_cache_fill(struct btree_trans *trans, int ret; bch2_trans_iter_init(trans, &iter, ck->key.btree_id, ck->key.pos, - BTREE_ITER_KEY_CACHE_FILL| - BTREE_ITER_CACHED_NOFILL); - iter.flags &= ~BTREE_ITER_WITH_JOURNAL; + BTREE_ITER_key_cache_fill| + BTREE_ITER_cached_nofill); + iter.flags &= ~BTREE_ITER_with_journal; k = bch2_btree_iter_peek_slot(&iter); ret = bkey_err(k); if (ret) @@ -515,7 +515,7 @@ retry: fill: path->uptodate = BTREE_ITER_UPTODATE; - if (!ck->valid && !(flags & BTREE_ITER_CACHED_NOFILL)) { + if (!ck->valid && !(flags & BTREE_ITER_cached_nofill)) { /* * Using the underscore version because we haven't set * path->uptodate yet: @@ -622,13 +622,13 @@ static int btree_key_cache_flush_pos(struct btree_trans *trans, int ret; bch2_trans_iter_init(trans, &b_iter, key.btree_id, key.pos, - BTREE_ITER_SLOTS| - BTREE_ITER_INTENT| - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_slots| + BTREE_ITER_intent| + BTREE_ITER_all_snapshots); bch2_trans_iter_init(trans, &c_iter, key.btree_id, key.pos, - BTREE_ITER_CACHED| - BTREE_ITER_INTENT); - b_iter.flags &= ~BTREE_ITER_WITH_KEY_CACHE; + BTREE_ITER_cached| + BTREE_ITER_intent); + b_iter.flags &= ~BTREE_ITER_with_key_cache; ret = bch2_btree_iter_traverse(&c_iter); if (ret) @@ -666,9 +666,9 @@ static int btree_key_cache_flush_pos(struct btree_trans *trans, ret = bch2_btree_iter_traverse(&b_iter) ?: bch2_trans_update(trans, &b_iter, ck->k, - BTREE_UPDATE_KEY_CACHE_RECLAIM| - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE| - BTREE_TRIGGER_NORUN) ?: + BTREE_UPDATE_key_cache_reclaim| + BTREE_UPDATE_internal_snapshot_node| + BTREE_TRIGGER_norun) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_check_rw| BCH_TRANS_COMMIT_no_enospc| @@ -790,7 +790,7 @@ bool bch2_btree_insert_key_cached(struct btree_trans *trans, * flushing. The flush callback will not proceed unless ->seq matches * the latest pin, so make sure it starts with a consistent value. */ - if (!(insert_entry->flags & BTREE_UPDATE_NOJOURNAL) || + if (!(insert_entry->flags & BTREE_UPDATE_nojournal) || !journal_pin_active(&ck->journal)) { ck->seq = trans->journal_res.seq; } diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 79cbf54c8912..c5317e74ee06 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -315,7 +315,7 @@ static inline void btree_insert_entry_checks(struct btree_trans *trans, BUG_ON(i->btree_id != path->btree_id); EBUG_ON(!i->level && btree_type_has_snapshots(i->btree_id) && - !(i->flags & BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) && + !(i->flags & BTREE_UPDATE_internal_snapshot_node) && test_bit(JOURNAL_REPLAY_DONE, &trans->c->journal.flags) && i->k->k.p.snapshot && bch2_snapshot_is_internal_node(trans->c, i->k->k.p.snapshot) > 0); @@ -443,13 +443,13 @@ static int run_one_mem_trigger(struct btree_trans *trans, verify_update_old_key(trans, i); - if (unlikely(flags & BTREE_TRIGGER_NORUN)) + if (unlikely(flags & BTREE_TRIGGER_norun)) return 0; if (old_ops->trigger == new_ops->trigger) { ret = bch2_key_trigger(trans, i->btree_id, i->level, old, bkey_i_to_s(new), - BTREE_TRIGGER_INSERT|BTREE_TRIGGER_OVERWRITE|flags); + BTREE_TRIGGER_insert|BTREE_TRIGGER_overwrite|flags); } else { ret = bch2_key_trigger_new(trans, i->btree_id, i->level, bkey_i_to_s(new), flags) ?: @@ -472,11 +472,11 @@ static int run_one_trans_trigger(struct btree_trans *trans, struct btree_insert_ struct bkey_s_c old = { &old_k, i->old_v }; const struct bkey_ops *old_ops = bch2_bkey_type_ops(old.k->type); const struct bkey_ops *new_ops = bch2_bkey_type_ops(i->k->k.type); - unsigned flags = i->flags|BTREE_TRIGGER_TRANSACTIONAL; + unsigned flags = i->flags|BTREE_TRIGGER_transactional; verify_update_old_key(trans, i); - if ((i->flags & BTREE_TRIGGER_NORUN) || + if ((i->flags & BTREE_TRIGGER_norun) || !(BTREE_NODE_TYPE_HAS_TRANS_TRIGGERS & (1U << i->bkey_type))) return 0; @@ -486,8 +486,8 @@ static int run_one_trans_trigger(struct btree_trans *trans, struct btree_insert_ i->overwrite_trigger_run = true; i->insert_trigger_run = true; return bch2_key_trigger(trans, i->btree_id, i->level, old, bkey_i_to_s(i->k), - BTREE_TRIGGER_INSERT| - BTREE_TRIGGER_OVERWRITE|flags) ?: 1; + BTREE_TRIGGER_insert| + BTREE_TRIGGER_overwrite|flags) ?: 1; } else if (overwrite && !i->overwrite_trigger_run) { i->overwrite_trigger_run = true; return bch2_key_trigger_old(trans, i->btree_id, i->level, old, flags) ?: 1; @@ -572,7 +572,7 @@ static int bch2_trans_commit_run_triggers(struct btree_trans *trans) #ifdef CONFIG_BCACHEFS_DEBUG trans_for_each_update(trans, i) - BUG_ON(!(i->flags & BTREE_TRIGGER_NORUN) && + BUG_ON(!(i->flags & BTREE_TRIGGER_norun) && (BTREE_NODE_TYPE_HAS_TRANS_TRIGGERS & (1U << i->bkey_type)) && (!i->insert_trigger_run || !i->overwrite_trigger_run)); #endif @@ -590,7 +590,7 @@ static noinline int bch2_trans_commit_run_gc_triggers(struct btree_trans *trans) if (btree_node_type_needs_gc(__btree_node_type(i->level, i->btree_id)) && gc_visited(trans->c, gc_pos_btree_node(insert_l(trans, i)->b))) { - int ret = run_one_mem_trigger(trans, i, i->flags|BTREE_TRIGGER_GC); + int ret = run_one_mem_trigger(trans, i, i->flags|BTREE_TRIGGER_gc); if (ret) return ret; } @@ -686,7 +686,7 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, trans_for_each_update(trans, i) if (BTREE_NODE_TYPE_HAS_ATOMIC_TRIGGERS & (1U << i->bkey_type)) { - ret = run_one_mem_trigger(trans, i, BTREE_TRIGGER_ATOMIC|i->flags); + ret = run_one_mem_trigger(trans, i, BTREE_TRIGGER_atomic|i->flags); if (ret) goto fatal_err; } @@ -705,7 +705,7 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, if (i->key_cache_already_flushed) continue; - if (i->flags & BTREE_UPDATE_NOJOURNAL) + if (i->flags & BTREE_UPDATE_nojournal) continue; verify_update_old_key(trans, i); @@ -1063,7 +1063,7 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) if (i->key_cache_already_flushed) continue; - if (i->flags & BTREE_UPDATE_NOJOURNAL) + if (i->flags & BTREE_UPDATE_nojournal) continue; /* we're going to journal the key being updated: */ diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index d4649129e5d3..123abeec4ce9 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -187,36 +187,87 @@ struct btree_node_iter { } data[MAX_BSETS]; }; +#define BTREE_ITER_FLAGS() \ + x(slots) \ + x(intent) \ + x(prefetch) \ + x(is_extents) \ + x(not_extents) \ + x(cached) \ + x(with_key_cache) \ + x(with_updates) \ + x(with_journal) \ + x(snapshot_field) \ + x(all_snapshots) \ + x(filter_snapshots) \ + x(nopreserve) \ + x(cached_nofill) \ + x(key_cache_fill) \ + +#define STR_HASH_FLAGS() \ + x(must_create) \ + x(must_replace) + +#define BTREE_UPDATE_FLAGS() \ + x(internal_snapshot_node) \ + x(nojournal) \ + x(key_cache_reclaim) + + /* - * Iterate over all possible positions, synthesizing deleted keys for holes: - */ -static const __maybe_unused u16 BTREE_ITER_SLOTS = 1 << 0; -/* - * Indicates that intent locks should be taken on leaf nodes, because we expect - * to be doing updates: - */ -static const __maybe_unused u16 BTREE_ITER_INTENT = 1 << 1; -/* - * Causes the btree iterator code to prefetch additional btree nodes from disk: - */ -static const __maybe_unused u16 BTREE_ITER_PREFETCH = 1 << 2; -/* - * Used in bch2_btree_iter_traverse(), to indicate whether we're searching for - * @pos or the first key strictly greater than @pos + * BTREE_TRIGGER_norun - don't run triggers at all + * + * BTREE_TRIGGER_transactional - we're running transactional triggers as part of + * a transaction commit: triggers may generate new updates + * + * BTREE_TRIGGER_atomic - we're running atomic triggers during a transaction + * commit: we have our journal reservation, we're holding btree node write + * locks, and we know the transaction is going to commit (returning an error + * here is a fatal error, causing us to go emergency read-only) + * + * BTREE_TRIGGER_gc - we're in gc/fsck: running triggers to recalculate e.g. disk usage + * + * BTREE_TRIGGER_insert - @new is entering the btree + * BTREE_TRIGGER_overwrite - @old is leaving the btree + * + * BTREE_TRIGGER_bucket_invalidate - signal from bucket invalidate path to alloc + * trigger */ -static const __maybe_unused u16 BTREE_ITER_IS_EXTENTS = 1 << 3; -static const __maybe_unused u16 BTREE_ITER_NOT_EXTENTS = 1 << 4; -static const __maybe_unused u16 BTREE_ITER_CACHED = 1 << 5; -static const __maybe_unused u16 BTREE_ITER_WITH_KEY_CACHE = 1 << 6; -static const __maybe_unused u16 BTREE_ITER_WITH_UPDATES = 1 << 7; -static const __maybe_unused u16 BTREE_ITER_WITH_JOURNAL = 1 << 8; -static const __maybe_unused u16 BTREE_ITER_SNAPSHOT_FIELD = 1 << 9; -static const __maybe_unused u16 BTREE_ITER_ALL_SNAPSHOTS = 1 << 10; -static const __maybe_unused u16 BTREE_ITER_FILTER_SNAPSHOTS = 1 << 11; -static const __maybe_unused u16 BTREE_ITER_NOPRESERVE = 1 << 12; -static const __maybe_unused u16 BTREE_ITER_CACHED_NOFILL = 1 << 13; -static const __maybe_unused u16 BTREE_ITER_KEY_CACHE_FILL = 1 << 14; -#define __BTREE_ITER_FLAGS_END 15 +#define BTREE_TRIGGER_FLAGS() \ + x(norun) \ + x(transactional) \ + x(atomic) \ + x(gc) \ + x(insert) \ + x(overwrite) \ + x(bucket_invalidate) + +enum { +#define x(n) BTREE_ITER_FLAG_BIT_##n, + BTREE_ITER_FLAGS() + STR_HASH_FLAGS() + BTREE_UPDATE_FLAGS() + BTREE_TRIGGER_FLAGS() +#undef x +}; + +/* iter flags must fit in a u16: */ +//BUILD_BUG_ON(BTREE_ITER_FLAG_BIT_key_cache_fill > 15); + +enum btree_iter_update_trigger_flags { +#define x(n) BTREE_ITER_##n = 1U << BTREE_ITER_FLAG_BIT_##n, + BTREE_ITER_FLAGS() +#undef x +#define x(n) STR_HASH_##n = 1U << BTREE_ITER_FLAG_BIT_##n, + STR_HASH_FLAGS() +#undef x +#define x(n) BTREE_UPDATE_##n = 1U << BTREE_ITER_FLAG_BIT_##n, + BTREE_UPDATE_FLAGS() +#undef x +#define x(n) BTREE_TRIGGER_##n = 1U << BTREE_ITER_FLAG_BIT_##n, + BTREE_TRIGGER_FLAGS() +#undef x +}; enum btree_path_uptodate { BTREE_ITER_UPTODATE = 0, @@ -307,7 +358,7 @@ struct btree_iter { */ struct bkey k; - /* BTREE_ITER_WITH_JOURNAL: */ + /* BTREE_ITER_with_journal: */ size_t journal_idx; #ifdef TRACK_PATH_ALLOCATED unsigned long ip_allocated; diff --git a/fs/bcachefs/btree_update.c b/fs/bcachefs/btree_update.c index c46ee4c813ae..da4b17704296 100644 --- a/fs/bcachefs/btree_update.c +++ b/fs/bcachefs/btree_update.c @@ -25,14 +25,14 @@ static inline int btree_insert_entry_cmp(const struct btree_insert_entry *l, static int __must_check bch2_trans_update_by_path(struct btree_trans *, btree_path_idx_t, - struct bkey_i *, enum btree_update_flags, + struct bkey_i *, enum btree_iter_update_trigger_flags, unsigned long ip); static noinline int extent_front_merge(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k, struct bkey_i **insert, - enum btree_update_flags flags) + enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; struct bkey_i *update; @@ -104,8 +104,8 @@ static int need_whiteout_for_snapshot(struct btree_trans *trans, pos.snapshot++; for_each_btree_key_norestart(trans, iter, btree_id, pos, - BTREE_ITER_ALL_SNAPSHOTS| - BTREE_ITER_NOPRESERVE, k, ret) { + BTREE_ITER_all_snapshots| + BTREE_ITER_nopreserve, k, ret) { if (!bkey_eq(k.k->p, pos)) break; @@ -138,8 +138,8 @@ int __bch2_insert_snapshot_whiteouts(struct btree_trans *trans, darray_init(&s); bch2_trans_iter_init(trans, &old_iter, id, old_pos, - BTREE_ITER_NOT_EXTENTS| - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_not_extents| + BTREE_ITER_all_snapshots); while ((old_k = bch2_btree_iter_prev(&old_iter)).k && !(ret = bkey_err(old_k)) && bkey_eq(old_pos, old_k.k->p)) { @@ -151,8 +151,8 @@ int __bch2_insert_snapshot_whiteouts(struct btree_trans *trans, continue; new_k = bch2_bkey_get_iter(trans, &new_iter, id, whiteout_pos, - BTREE_ITER_NOT_EXTENTS| - BTREE_ITER_INTENT); + BTREE_ITER_not_extents| + BTREE_ITER_intent); ret = bkey_err(new_k); if (ret) break; @@ -168,7 +168,7 @@ int __bch2_insert_snapshot_whiteouts(struct btree_trans *trans, update->k.type = KEY_TYPE_whiteout; ret = bch2_trans_update(trans, &new_iter, update, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); } bch2_trans_iter_exit(trans, &new_iter); @@ -185,7 +185,7 @@ int __bch2_insert_snapshot_whiteouts(struct btree_trans *trans, int bch2_trans_update_extent_overwrite(struct btree_trans *trans, struct btree_iter *iter, - enum btree_update_flags flags, + enum btree_iter_update_trigger_flags flags, struct bkey_s_c old, struct bkey_s_c new) { @@ -218,7 +218,7 @@ int bch2_trans_update_extent_overwrite(struct btree_trans *trans, ret = bch2_insert_snapshot_whiteouts(trans, btree_id, old.k->p, update->k.p) ?: bch2_btree_insert_nonextent(trans, btree_id, update, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE|flags); + BTREE_UPDATE_internal_snapshot_node|flags); if (ret) return ret; } @@ -235,7 +235,7 @@ int bch2_trans_update_extent_overwrite(struct btree_trans *trans, ret = bch2_insert_snapshot_whiteouts(trans, btree_id, old.k->p, update->k.p) ?: bch2_btree_insert_nonextent(trans, btree_id, update, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE|flags); + BTREE_UPDATE_internal_snapshot_node|flags); if (ret) return ret; } @@ -260,7 +260,7 @@ int bch2_trans_update_extent_overwrite(struct btree_trans *trans, } ret = bch2_btree_insert_nonextent(trans, btree_id, update, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE|flags); + BTREE_UPDATE_internal_snapshot_node|flags); if (ret) return ret; } @@ -273,7 +273,7 @@ int bch2_trans_update_extent_overwrite(struct btree_trans *trans, bch2_cut_front(new.k->p, update); ret = bch2_trans_update_by_path(trans, iter->path, update, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE| + BTREE_UPDATE_internal_snapshot_node| flags, _RET_IP_); if (ret) return ret; @@ -285,7 +285,7 @@ int bch2_trans_update_extent_overwrite(struct btree_trans *trans, static int bch2_trans_update_extent(struct btree_trans *trans, struct btree_iter *orig_iter, struct bkey_i *insert, - enum btree_update_flags flags) + enum btree_iter_update_trigger_flags flags) { struct btree_iter iter; struct bkey_s_c k; @@ -293,9 +293,9 @@ static int bch2_trans_update_extent(struct btree_trans *trans, int ret = 0; bch2_trans_iter_init(trans, &iter, btree_id, bkey_start_pos(&insert->k), - BTREE_ITER_INTENT| - BTREE_ITER_WITH_UPDATES| - BTREE_ITER_NOT_EXTENTS); + BTREE_ITER_intent| + BTREE_ITER_with_updates| + BTREE_ITER_not_extents); k = bch2_btree_iter_peek_upto(&iter, POS(insert->k.p.inode, U64_MAX)); if ((ret = bkey_err(k))) goto err; @@ -346,7 +346,7 @@ err: static noinline int flush_new_cached_update(struct btree_trans *trans, struct btree_insert_entry *i, - enum btree_update_flags flags, + enum btree_iter_update_trigger_flags flags, unsigned long ip) { struct bkey k; @@ -354,7 +354,7 @@ static noinline int flush_new_cached_update(struct btree_trans *trans, btree_path_idx_t path_idx = bch2_path_get(trans, i->btree_id, i->old_k.p, 1, 0, - BTREE_ITER_INTENT, _THIS_IP_); + BTREE_ITER_intent, _THIS_IP_); ret = bch2_btree_path_traverse(trans, path_idx, 0); if (ret) goto out; @@ -372,7 +372,7 @@ static noinline int flush_new_cached_update(struct btree_trans *trans, goto out; i->key_cache_already_flushed = true; - i->flags |= BTREE_TRIGGER_NORUN; + i->flags |= BTREE_TRIGGER_norun; btree_path_set_should_be_locked(btree_path); ret = bch2_trans_update_by_path(trans, path_idx, i->k, flags, ip); @@ -383,7 +383,7 @@ out: static int __must_check bch2_trans_update_by_path(struct btree_trans *trans, btree_path_idx_t path_idx, - struct bkey_i *k, enum btree_update_flags flags, + struct bkey_i *k, enum btree_iter_update_trigger_flags flags, unsigned long ip) { struct bch_fs *c = trans->c; @@ -479,15 +479,15 @@ static noinline int bch2_trans_update_get_key_cache(struct btree_trans *trans, if (!iter->key_cache_path) iter->key_cache_path = bch2_path_get(trans, path->btree_id, path->pos, 1, 0, - BTREE_ITER_INTENT| - BTREE_ITER_CACHED, _THIS_IP_); + BTREE_ITER_intent| + BTREE_ITER_cached, _THIS_IP_); iter->key_cache_path = bch2_btree_path_set_pos(trans, iter->key_cache_path, path->pos, - iter->flags & BTREE_ITER_INTENT, + iter->flags & BTREE_ITER_intent, _THIS_IP_); - ret = bch2_btree_path_traverse(trans, iter->key_cache_path, BTREE_ITER_CACHED); + ret = bch2_btree_path_traverse(trans, iter->key_cache_path, BTREE_ITER_cached); if (unlikely(ret)) return ret; @@ -505,17 +505,17 @@ static noinline int bch2_trans_update_get_key_cache(struct btree_trans *trans, } int __must_check bch2_trans_update(struct btree_trans *trans, struct btree_iter *iter, - struct bkey_i *k, enum btree_update_flags flags) + struct bkey_i *k, enum btree_iter_update_trigger_flags flags) { btree_path_idx_t path_idx = iter->update_path ?: iter->path; int ret; - if (iter->flags & BTREE_ITER_IS_EXTENTS) + if (iter->flags & BTREE_ITER_is_extents) return bch2_trans_update_extent(trans, iter, k, flags); if (bkey_deleted(&k->k) && - !(flags & BTREE_UPDATE_KEY_CACHE_RECLAIM) && - (iter->flags & BTREE_ITER_FILTER_SNAPSHOTS)) { + !(flags & BTREE_UPDATE_key_cache_reclaim) && + (iter->flags & BTREE_ITER_filter_snapshots)) { ret = need_whiteout_for_snapshot(trans, iter->btree_id, k->k.p); if (unlikely(ret < 0)) return ret; @@ -528,7 +528,7 @@ int __must_check bch2_trans_update(struct btree_trans *trans, struct btree_iter * Ensure that updates to cached btrees go to the key cache: */ struct btree_path *path = trans->paths + path_idx; - if (!(flags & BTREE_UPDATE_KEY_CACHE_RECLAIM) && + if (!(flags & BTREE_UPDATE_key_cache_reclaim) && !path->cached && !path->level && btree_id_cached(trans->c, path->btree_id)) { @@ -587,7 +587,7 @@ int bch2_bkey_get_empty_slot(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k; int ret = 0; - bch2_trans_iter_init(trans, iter, btree, POS_MAX, BTREE_ITER_INTENT); + bch2_trans_iter_init(trans, iter, btree, POS_MAX, BTREE_ITER_intent); k = bch2_btree_iter_prev(iter); ret = bkey_err(k); if (ret) @@ -621,15 +621,15 @@ void bch2_trans_commit_hook(struct btree_trans *trans, int bch2_btree_insert_nonextent(struct btree_trans *trans, enum btree_id btree, struct bkey_i *k, - enum btree_update_flags flags) + enum btree_iter_update_trigger_flags flags) { struct btree_iter iter; int ret; bch2_trans_iter_init(trans, &iter, btree, k->k.p, - BTREE_ITER_CACHED| - BTREE_ITER_NOT_EXTENTS| - BTREE_ITER_INTENT); + BTREE_ITER_cached| + BTREE_ITER_not_extents| + BTREE_ITER_intent); ret = bch2_btree_iter_traverse(&iter) ?: bch2_trans_update(trans, &iter, k, flags); bch2_trans_iter_exit(trans, &iter); @@ -637,14 +637,14 @@ int bch2_btree_insert_nonextent(struct btree_trans *trans, } int bch2_btree_insert_trans(struct btree_trans *trans, enum btree_id id, - struct bkey_i *k, enum btree_update_flags flags) + struct bkey_i *k, enum btree_iter_update_trigger_flags flags) { struct btree_iter iter; int ret; bch2_trans_iter_init(trans, &iter, id, bkey_start_pos(&k->k), - BTREE_ITER_CACHED| - BTREE_ITER_INTENT); + BTREE_ITER_cached| + BTREE_ITER_intent); ret = bch2_btree_iter_traverse(&iter) ?: bch2_trans_update(trans, &iter, k, flags); bch2_trans_iter_exit(trans, &iter); @@ -698,8 +698,8 @@ int bch2_btree_delete(struct btree_trans *trans, int ret; bch2_trans_iter_init(trans, &iter, btree, pos, - BTREE_ITER_CACHED| - BTREE_ITER_INTENT); + BTREE_ITER_cached| + BTREE_ITER_intent); ret = bch2_btree_iter_traverse(&iter) ?: bch2_btree_delete_at(trans, &iter, update_flags); bch2_trans_iter_exit(trans, &iter); @@ -717,7 +717,7 @@ int bch2_btree_delete_range_trans(struct btree_trans *trans, enum btree_id id, struct bkey_s_c k; int ret = 0; - bch2_trans_iter_init(trans, &iter, id, start, BTREE_ITER_INTENT); + bch2_trans_iter_init(trans, &iter, id, start, BTREE_ITER_intent); while ((k = bch2_btree_iter_peek_upto(&iter, end)).k) { struct disk_reservation disk_res = bch2_disk_reservation_init(trans->c, 0); @@ -745,7 +745,7 @@ int bch2_btree_delete_range_trans(struct btree_trans *trans, enum btree_id id, */ delete.k.p = iter.pos; - if (iter.flags & BTREE_ITER_IS_EXTENTS) + if (iter.flags & BTREE_ITER_is_extents) bch2_key_resize(&delete.k, bpos_min(end, k.k->p).offset - iter.pos.offset); @@ -804,7 +804,7 @@ int bch2_btree_bit_mod(struct btree_trans *trans, enum btree_id btree, k->k.p = pos; struct btree_iter iter; - bch2_trans_iter_init(trans, &iter, btree, pos, BTREE_ITER_INTENT); + bch2_trans_iter_init(trans, &iter, btree, pos, BTREE_ITER_intent); ret = bch2_btree_iter_traverse(&iter) ?: bch2_trans_update(trans, &iter, k, 0); diff --git a/fs/bcachefs/btree_update.h b/fs/bcachefs/btree_update.h index cc7c53e83f89..3a04ede35475 100644 --- a/fs/bcachefs/btree_update.h +++ b/fs/bcachefs/btree_update.h @@ -50,10 +50,10 @@ int bch2_btree_delete_at(struct btree_trans *, struct btree_iter *, unsigned); int bch2_btree_delete(struct btree_trans *, enum btree_id, struct bpos, unsigned); int bch2_btree_insert_nonextent(struct btree_trans *, enum btree_id, - struct bkey_i *, enum btree_update_flags); + struct bkey_i *, enum btree_iter_update_trigger_flags); int bch2_btree_insert_trans(struct btree_trans *, enum btree_id, struct bkey_i *, - enum btree_update_flags); + enum btree_iter_update_trigger_flags); int bch2_btree_insert(struct bch_fs *, enum btree_id, struct bkey_i *, struct disk_reservation *, int flags); @@ -94,14 +94,14 @@ static inline int bch2_insert_snapshot_whiteouts(struct btree_trans *trans, } int bch2_trans_update_extent_overwrite(struct btree_trans *, struct btree_iter *, - enum btree_update_flags, + enum btree_iter_update_trigger_flags, struct bkey_s_c, struct bkey_s_c); int bch2_bkey_get_empty_slot(struct btree_trans *, struct btree_iter *, enum btree_id, struct bpos); int __must_check bch2_trans_update(struct btree_trans *, struct btree_iter *, - struct bkey_i *, enum btree_update_flags); + struct bkey_i *, enum btree_iter_update_trigger_flags); struct jset_entry *__bch2_trans_jset_entry_alloc(struct btree_trans *, unsigned); @@ -276,7 +276,7 @@ static inline struct bkey_i *__bch2_bkey_get_mut_noupdate(struct btree_trans *tr unsigned flags, unsigned type, unsigned min_bytes) { struct bkey_s_c k = __bch2_bkey_get_iter(trans, iter, - btree_id, pos, flags|BTREE_ITER_INTENT, type); + btree_id, pos, flags|BTREE_ITER_intent, type); struct bkey_i *ret = IS_ERR(k.k) ? ERR_CAST(k.k) : __bch2_bkey_make_mut_noupdate(trans, k, 0, min_bytes); @@ -299,7 +299,7 @@ static inline struct bkey_i *__bch2_bkey_get_mut(struct btree_trans *trans, unsigned flags, unsigned type, unsigned min_bytes) { struct bkey_i *mut = __bch2_bkey_get_mut_noupdate(trans, iter, - btree_id, pos, flags|BTREE_ITER_INTENT, type, min_bytes); + btree_id, pos, flags|BTREE_ITER_intent, type, min_bytes); int ret; if (IS_ERR(mut)) diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 875b1795e408..0b6645c2266f 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -44,8 +44,8 @@ static btree_path_idx_t get_unlocked_mut_path(struct btree_trans *trans, struct bpos pos) { btree_path_idx_t path_idx = bch2_path_get(trans, btree_id, pos, level + 1, level, - BTREE_ITER_NOPRESERVE| - BTREE_ITER_INTENT, _RET_IP_); + BTREE_ITER_nopreserve| + BTREE_ITER_intent, _RET_IP_); path_idx = bch2_btree_path_make_mut(trans, path_idx, true, _RET_IP_); struct btree_path *path = trans->paths + path_idx; @@ -664,7 +664,7 @@ static int btree_update_nodes_written_trans(struct btree_trans *trans, unsigned level = bkey_i_to_btree_ptr_v2(k)->v.mem_ptr; ret = bch2_key_trigger_old(trans, as->btree_id, level, bkey_i_to_s_c(k), - BTREE_TRIGGER_TRANSACTIONAL); + BTREE_TRIGGER_transactional); if (ret) return ret; } @@ -673,7 +673,7 @@ static int btree_update_nodes_written_trans(struct btree_trans *trans, unsigned level = bkey_i_to_btree_ptr_v2(k)->v.mem_ptr; ret = bch2_key_trigger_new(trans, as->btree_id, level, bkey_i_to_s(k), - BTREE_TRIGGER_TRANSACTIONAL); + BTREE_TRIGGER_transactional); if (ret) return ret; } @@ -1997,7 +1997,7 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, : bpos_successor(b->data->max_key); sib_path = bch2_path_get(trans, btree, sib_pos, - U8_MAX, level, BTREE_ITER_INTENT, _THIS_IP_); + U8_MAX, level, BTREE_ITER_intent, _THIS_IP_); ret = bch2_btree_path_traverse(trans, sib_path, false); if (ret) goto err; @@ -2351,10 +2351,10 @@ static int __bch2_btree_node_update_key(struct btree_trans *trans, if (!skip_triggers) { ret = bch2_key_trigger_old(trans, b->c.btree_id, b->c.level + 1, bkey_i_to_s_c(&b->key), - BTREE_TRIGGER_TRANSACTIONAL) ?: + BTREE_TRIGGER_transactional) ?: bch2_key_trigger_new(trans, b->c.btree_id, b->c.level + 1, bkey_i_to_s(new_key), - BTREE_TRIGGER_TRANSACTIONAL); + BTREE_TRIGGER_transactional); if (ret) return ret; } @@ -2371,7 +2371,7 @@ static int __bch2_btree_node_update_key(struct btree_trans *trans, bch2_trans_copy_iter(&iter2, iter); iter2.path = bch2_btree_path_make_mut(trans, iter2.path, - iter2.flags & BTREE_ITER_INTENT, + iter2.flags & BTREE_ITER_intent, _THIS_IP_); struct btree_path *path2 = btree_iter_path(trans, &iter2); @@ -2383,7 +2383,7 @@ static int __bch2_btree_node_update_key(struct btree_trans *trans, trans->paths_sorted = false; ret = bch2_btree_iter_traverse(&iter2) ?: - bch2_trans_update(trans, &iter2, new_key, BTREE_TRIGGER_NORUN); + bch2_trans_update(trans, &iter2, new_key, BTREE_TRIGGER_norun); if (ret) goto err; } else { @@ -2491,7 +2491,7 @@ int bch2_btree_node_update_key_get_iter(struct btree_trans *trans, bch2_trans_node_iter_init(trans, &iter, b->c.btree_id, b->key.k.p, BTREE_MAX_DEPTH, b->c.level, - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = bch2_btree_iter_traverse(&iter); if (ret) goto out; diff --git a/fs/bcachefs/btree_write_buffer.c b/fs/bcachefs/btree_write_buffer.c index 36a6f42aba5e..75c8a196b3f6 100644 --- a/fs/bcachefs/btree_write_buffer.c +++ b/fs/bcachefs/btree_write_buffer.c @@ -122,7 +122,7 @@ static noinline int wb_flush_one_slowpath(struct btree_trans *trans, trans->journal_res.seq = wb->journal_seq; return bch2_trans_update(trans, iter, &wb->k, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?: + BTREE_UPDATE_internal_snapshot_node) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc| BCH_TRANS_COMMIT_no_check_rw| @@ -191,13 +191,13 @@ btree_write_buffered_insert(struct btree_trans *trans, int ret; bch2_trans_iter_init(trans, &iter, wb->btree, bkey_start_pos(&wb->k.k), - BTREE_ITER_CACHED|BTREE_ITER_INTENT); + BTREE_ITER_cached|BTREE_ITER_intent); trans->journal_res.seq = wb->journal_seq; ret = bch2_btree_iter_traverse(&iter) ?: bch2_trans_update(trans, &iter, &wb->k, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); bch2_trans_iter_exit(trans, &iter); return ret; } @@ -332,7 +332,7 @@ static int bch2_btree_write_buffer_flush_locked(struct btree_trans *trans) if (!iter.path || iter.btree_id != k->btree) { bch2_trans_iter_exit(trans, &iter); bch2_trans_iter_init(trans, &iter, k->btree, k->k.k.p, - BTREE_ITER_INTENT|BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_intent|BTREE_ITER_all_snapshots); } bch2_btree_iter_set_pos(&iter, k->k.k.p); diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 015321a87170..20f471c08b2e 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -752,16 +752,17 @@ static int bch2_trigger_pointer(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c k, struct extent_ptr_decoded p, const union bch_extent_entry *entry, - s64 *sectors, unsigned flags) + s64 *sectors, + enum btree_iter_update_trigger_flags flags) { - bool insert = !(flags & BTREE_TRIGGER_OVERWRITE); + bool insert = !(flags & BTREE_TRIGGER_overwrite); struct bpos bucket; struct bch_backpointer bp; bch2_extent_ptr_to_bp(trans->c, btree_id, level, k, p, entry, &bucket, &bp); *sectors = insert ? bp.bucket_len : -((s64) bp.bucket_len); - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { struct btree_iter iter; struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, &iter, bucket); int ret = PTR_ERR_OR_ZERO(a); @@ -784,7 +785,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, } } - if (flags & BTREE_TRIGGER_GC) { + if (flags & BTREE_TRIGGER_gc) { struct bch_fs *c = trans->c; struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev); enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry); @@ -820,13 +821,14 @@ static int bch2_trigger_stripe_ptr(struct btree_trans *trans, struct bkey_s_c k, struct extent_ptr_decoded p, enum bch_data_type data_type, - s64 sectors, unsigned flags) + s64 sectors, + enum btree_iter_update_trigger_flags flags) { - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { struct btree_iter iter; struct bkey_i_stripe *s = bch2_bkey_get_mut_typed(trans, &iter, BTREE_ID_stripes, POS(0, p.ec.idx), - BTREE_ITER_WITH_UPDATES, stripe); + BTREE_ITER_with_updates, stripe); int ret = PTR_ERR_OR_ZERO(s); if (unlikely(ret)) { bch2_trans_inconsistent_on(bch2_err_matches(ret, ENOENT), trans, @@ -856,10 +858,10 @@ err: return ret; } - if (flags & BTREE_TRIGGER_GC) { + if (flags & BTREE_TRIGGER_gc) { struct bch_fs *c = trans->c; - BUG_ON(!(flags & BTREE_TRIGGER_GC)); + BUG_ON(!(flags & BTREE_TRIGGER_gc)); struct gc_stripe *m = genradix_ptr_alloc(&c->gc_stripes, p.ec.idx, GFP_KERNEL); if (!m) { @@ -895,9 +897,10 @@ err: static int __trigger_extent(struct btree_trans *trans, enum btree_id btree_id, unsigned level, - struct bkey_s_c k, unsigned flags) + struct bkey_s_c k, + enum btree_iter_update_trigger_flags flags) { - bool gc = flags & BTREE_TRIGGER_GC; + bool gc = flags & BTREE_TRIGGER_gc; struct bch_fs *c = trans->c; struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); const union bch_extent_entry *entry; @@ -969,7 +972,7 @@ static int __trigger_extent(struct btree_trans *trans, int bch2_trigger_extent(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { struct bkey_ptrs_c new_ptrs = bch2_bkey_ptrs_c(new.s_c); struct bkey_ptrs_c old_ptrs = bch2_bkey_ptrs_c(old); @@ -983,7 +986,7 @@ int bch2_trigger_extent(struct btree_trans *trans, new_ptrs_bytes)) return 0; - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { struct bch_fs *c = trans->c; int mod = (int) bch2_bkey_needs_rebalance(c, new.s_c) - (int) bch2_bkey_needs_rebalance(c, old); @@ -996,7 +999,7 @@ int bch2_trigger_extent(struct btree_trans *trans, } } - if (flags & (BTREE_TRIGGER_TRANSACTIONAL|BTREE_TRIGGER_GC)) + if (flags & (BTREE_TRIGGER_transactional|BTREE_TRIGGER_gc)) return trigger_run_overwrite_then_insert(__trigger_extent, trans, btree_id, level, old, new, flags); return 0; @@ -1005,17 +1008,17 @@ int bch2_trigger_extent(struct btree_trans *trans, /* KEY_TYPE_reservation */ static int __trigger_reservation(struct btree_trans *trans, - enum btree_id btree_id, unsigned level, - struct bkey_s_c k, unsigned flags) + enum btree_id btree_id, unsigned level, struct bkey_s_c k, + enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; unsigned replicas = bkey_s_c_to_reservation(k).v->nr_replicas; s64 sectors = (s64) k.k->size * replicas; - if (flags & BTREE_TRIGGER_OVERWRITE) + if (flags & BTREE_TRIGGER_overwrite) sectors = -sectors; - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { int ret = bch2_replicas_deltas_realloc(trans, 0); if (ret) return ret; @@ -1026,7 +1029,7 @@ static int __trigger_reservation(struct btree_trans *trans, d->persistent_reserved[replicas - 1] += sectors; } - if (flags & BTREE_TRIGGER_GC) { + if (flags & BTREE_TRIGGER_gc) { percpu_down_read(&c->mark_lock); preempt_disable(); @@ -1046,7 +1049,7 @@ static int __trigger_reservation(struct btree_trans *trans, int bch2_trigger_reservation(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { return trigger_run_overwrite_then_insert(__trigger_reservation, trans, btree_id, level, old, new, flags); } @@ -1092,8 +1095,8 @@ err: } static int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca, - u64 b, enum bch_data_type data_type, - unsigned sectors, unsigned flags) + u64 b, enum bch_data_type data_type, unsigned sectors, + enum btree_iter_update_trigger_flags flags) { struct bucket old, new, *g; int ret = 0; @@ -1134,9 +1137,9 @@ err: } int bch2_trans_mark_metadata_bucket(struct btree_trans *trans, - struct bch_dev *ca, u64 b, - enum bch_data_type type, - unsigned sectors, unsigned flags) + struct bch_dev *ca, u64 b, + enum bch_data_type type, unsigned sectors, + enum btree_iter_update_trigger_flags flags) { BUG_ON(type != BCH_DATA_free && type != BCH_DATA_sb && @@ -1148,9 +1151,9 @@ int bch2_trans_mark_metadata_bucket(struct btree_trans *trans, if (b >= ca->mi.nbuckets) return 0; - if (flags & BTREE_TRIGGER_GC) + if (flags & BTREE_TRIGGER_gc) return bch2_mark_metadata_bucket(trans->c, ca, b, type, sectors, flags); - else if (flags & BTREE_TRIGGER_TRANSACTIONAL) + else if (flags & BTREE_TRIGGER_transactional) return commit_do(trans, NULL, NULL, 0, __bch2_trans_mark_metadata_bucket(trans, ca, b, type, sectors)); else @@ -1158,11 +1161,9 @@ int bch2_trans_mark_metadata_bucket(struct btree_trans *trans, } static int bch2_trans_mark_metadata_sectors(struct btree_trans *trans, - struct bch_dev *ca, - u64 start, u64 end, - enum bch_data_type type, - u64 *bucket, unsigned *bucket_sectors, - unsigned flags) + struct bch_dev *ca, u64 start, u64 end, + enum bch_data_type type, u64 *bucket, unsigned *bucket_sectors, + enum btree_iter_update_trigger_flags flags) { do { u64 b = sector_to_bucket(ca, start); @@ -1186,8 +1187,8 @@ static int bch2_trans_mark_metadata_sectors(struct btree_trans *trans, return 0; } -static int __bch2_trans_mark_dev_sb(struct btree_trans *trans, - struct bch_dev *ca, unsigned flags) +static int __bch2_trans_mark_dev_sb(struct btree_trans *trans, struct bch_dev *ca, + enum btree_iter_update_trigger_flags flags) { struct bch_sb_layout *layout = &ca->disk_sb.sb->layout; u64 bucket = 0; @@ -1230,7 +1231,8 @@ static int __bch2_trans_mark_dev_sb(struct btree_trans *trans, return 0; } -int bch2_trans_mark_dev_sb(struct bch_fs *c, struct bch_dev *ca, unsigned flags) +int bch2_trans_mark_dev_sb(struct bch_fs *c, struct bch_dev *ca, + enum btree_iter_update_trigger_flags flags) { int ret = bch2_trans_run(c, __bch2_trans_mark_dev_sb(trans, ca, flags)); @@ -1238,7 +1240,8 @@ int bch2_trans_mark_dev_sb(struct bch_fs *c, struct bch_dev *ca, unsigned flags) return ret; } -int bch2_trans_mark_dev_sbs_flags(struct bch_fs *c, unsigned flags) +int bch2_trans_mark_dev_sbs_flags(struct bch_fs *c, + enum btree_iter_update_trigger_flags flags) { for_each_online_member(c, ca) { int ret = bch2_trans_mark_dev_sb(c, ca, flags); @@ -1253,7 +1256,7 @@ int bch2_trans_mark_dev_sbs_flags(struct bch_fs *c, unsigned flags) int bch2_trans_mark_dev_sbs(struct bch_fs *c) { - return bch2_trans_mark_dev_sbs_flags(c, BTREE_TRIGGER_TRANSACTIONAL); + return bch2_trans_mark_dev_sbs_flags(c, BTREE_TRIGGER_transactional); } /* Disk reservations: */ diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index bae76e58311d..a84e14468276 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -338,18 +338,20 @@ int bch2_check_bucket_ref(struct btree_trans *, struct bkey_s_c, s64, enum bch_data_type, u8, u8, u32); int bch2_trigger_extent(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); int bch2_trigger_reservation(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); #define trigger_run_overwrite_then_insert(_fn, _trans, _btree_id, _level, _old, _new, _flags)\ ({ \ int ret = 0; \ \ if (_old.k->type) \ - ret = _fn(_trans, _btree_id, _level, _old, _flags & ~BTREE_TRIGGER_INSERT); \ + ret = _fn(_trans, _btree_id, _level, _old, _flags & ~BTREE_TRIGGER_insert); \ if (!ret && _new.k->type) \ - ret = _fn(_trans, _btree_id, _level, _new.s_c, _flags & ~BTREE_TRIGGER_OVERWRITE);\ + ret = _fn(_trans, _btree_id, _level, _new.s_c, _flags & ~BTREE_TRIGGER_overwrite);\ ret; \ }) @@ -359,9 +361,12 @@ void bch2_trans_fs_usage_revert(struct btree_trans *, struct replicas_delta_list int bch2_trans_fs_usage_apply(struct btree_trans *, struct replicas_delta_list *); int bch2_trans_mark_metadata_bucket(struct btree_trans *, struct bch_dev *, u64, - enum bch_data_type, unsigned, unsigned); -int bch2_trans_mark_dev_sb(struct bch_fs *, struct bch_dev *, unsigned); -int bch2_trans_mark_dev_sbs_flags(struct bch_fs *, unsigned); + enum bch_data_type, unsigned, + enum btree_iter_update_trigger_flags); +int bch2_trans_mark_dev_sb(struct bch_fs *, struct bch_dev *, + enum btree_iter_update_trigger_flags); +int bch2_trans_mark_dev_sbs_flags(struct bch_fs *, + enum btree_iter_update_trigger_flags); int bch2_trans_mark_dev_sbs(struct bch_fs *); static inline bool is_superblock_bucket(struct bch_dev *ca, u64 b) diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 0022b51ce3c0..3c0f0801468e 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -106,7 +106,7 @@ static int __bch2_data_update_index_update(struct btree_trans *trans, bch2_trans_iter_init(trans, &iter, m->btree_id, bkey_start_pos(&bch2_keylist_front(keys)->k), - BTREE_ITER_SLOTS|BTREE_ITER_INTENT); + BTREE_ITER_slots|BTREE_ITER_intent); while (1) { struct bkey_s_c k; @@ -288,7 +288,7 @@ restart_drop_extra_replicas: k.k->p, insert->k.p) ?: bch2_bkey_set_needs_rebalance(c, insert, &op->opts) ?: bch2_trans_update(trans, &iter, insert, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?: + BTREE_UPDATE_internal_snapshot_node) ?: bch2_trans_commit(trans, &op->res, NULL, BCH_TRANS_COMMIT_no_check_rw| @@ -387,7 +387,7 @@ static void bch2_update_unwritten_extent(struct btree_trans *trans, unsigned sectors = bio_sectors(bio); bch2_trans_iter_init(trans, &iter, update->btree_id, update->op.pos, - BTREE_ITER_SLOTS); + BTREE_ITER_slots); ret = lockrestart_do(trans, ({ k = bch2_btree_iter_peek_slot(&iter); bkey_err(k); @@ -480,15 +480,15 @@ int bch2_extent_drop_ptrs(struct btree_trans *trans, /* * Since we're not inserting through an extent iterator - * (BTREE_ITER_ALL_SNAPSHOTS iterators aren't extent iterators), + * (BTREE_ITER_all_snapshots iterators aren't extent iterators), * we aren't using the extent overwrite path to delete, we're * just using the normal key deletion path: */ - if (bkey_deleted(&n->k) && !(iter->flags & BTREE_ITER_IS_EXTENTS)) + if (bkey_deleted(&n->k) && !(iter->flags & BTREE_ITER_is_extents)) n->k.size = 0; return bch2_trans_relock(trans) ?: - bch2_trans_update(trans, iter, n, BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?: + bch2_trans_update(trans, iter, n, BTREE_UPDATE_internal_snapshot_node) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); } diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index d0ec74e6dc95..8310eff12a17 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -375,8 +375,8 @@ static ssize_t bch2_read_btree(struct file *file, char __user *buf, return flush_buf(i) ?: bch2_trans_run(i->c, for_each_btree_key(trans, iter, i->id, i->from, - BTREE_ITER_PREFETCH| - BTREE_ITER_ALL_SNAPSHOTS, k, ({ + BTREE_ITER_prefetch| + BTREE_ITER_all_snapshots, k, ({ bch2_bkey_val_to_text(&i->buf, i->c, k); prt_newline(&i->buf); bch2_trans_unlock(trans); @@ -459,8 +459,8 @@ static ssize_t bch2_read_bfloat_failed(struct file *file, char __user *buf, return flush_buf(i) ?: bch2_trans_run(i->c, for_each_btree_key(trans, iter, i->id, i->from, - BTREE_ITER_PREFETCH| - BTREE_ITER_ALL_SNAPSHOTS, k, ({ + BTREE_ITER_prefetch| + BTREE_ITER_all_snapshots, k, ({ struct btree_path_level *l = &btree_iter_path(trans, &iter)->l[0]; struct bkey_packed *_k = diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c index d4e3ca5045d6..ed256c3aa8a5 100644 --- a/fs/bcachefs/dirent.c +++ b/fs/bcachefs/dirent.c @@ -205,7 +205,7 @@ int bch2_dirent_create_snapshot(struct btree_trans *trans, const struct bch_hash_info *hash_info, u8 type, const struct qstr *name, u64 dst_inum, u64 *dir_offset, - bch_str_hash_flags_t str_hash_flags) + enum btree_iter_update_trigger_flags flags) { subvol_inum dir_inum = { .subvol = dir_subvol, .inum = dir }; struct bkey_i_dirent *dirent; @@ -220,9 +220,8 @@ int bch2_dirent_create_snapshot(struct btree_trans *trans, dirent->k.p.snapshot = snapshot; ret = bch2_hash_set_in_snapshot(trans, bch2_dirent_hash_desc, hash_info, - dir_inum, snapshot, - &dirent->k_i, str_hash_flags, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + dir_inum, snapshot, &dirent->k_i, + flags|BTREE_UPDATE_internal_snapshot_node); *dir_offset = dirent->k.p.offset; return ret; @@ -232,7 +231,7 @@ int bch2_dirent_create(struct btree_trans *trans, subvol_inum dir, const struct bch_hash_info *hash_info, u8 type, const struct qstr *name, u64 dst_inum, u64 *dir_offset, - bch_str_hash_flags_t str_hash_flags) + enum btree_iter_update_trigger_flags flags) { struct bkey_i_dirent *dirent; int ret; @@ -243,7 +242,7 @@ int bch2_dirent_create(struct btree_trans *trans, subvol_inum dir, return ret; ret = bch2_hash_set(trans, bch2_dirent_hash_desc, hash_info, - dir, &dirent->k_i, str_hash_flags); + dir, &dirent->k_i, flags); *dir_offset = dirent->k.p.offset; return ret; @@ -272,7 +271,7 @@ int bch2_dirent_read_target(struct btree_trans *trans, subvol_inum dir, } else { target->subvol = le32_to_cpu(d.v->d_child_subvol); - ret = bch2_subvolume_get(trans, target->subvol, true, BTREE_ITER_CACHED, &s); + ret = bch2_subvolume_get(trans, target->subvol, true, BTREE_ITER_cached, &s); target->inum = le64_to_cpu(s.inode); } @@ -303,7 +302,7 @@ int bch2_dirent_rename(struct btree_trans *trans, /* Lookup src: */ old_src = bch2_hash_lookup(trans, &src_iter, bch2_dirent_hash_desc, src_hash, src_dir, src_name, - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = bkey_err(old_src); if (ret) goto out; @@ -327,7 +326,7 @@ int bch2_dirent_rename(struct btree_trans *trans, } else { old_dst = bch2_hash_lookup(trans, &dst_iter, bch2_dirent_hash_desc, dst_hash, dst_dir, dst_name, - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = bkey_err(old_dst); if (ret) goto out; @@ -442,7 +441,7 @@ out_set_src: if (delete_src) { bch2_btree_iter_set_snapshot(&src_iter, old_src.k->p.snapshot); ret = bch2_btree_iter_traverse(&src_iter) ?: - bch2_btree_delete_at(trans, &src_iter, BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + bch2_btree_delete_at(trans, &src_iter, BTREE_UPDATE_internal_snapshot_node); if (ret) goto out; } @@ -450,7 +449,7 @@ out_set_src: if (delete_dst) { bch2_btree_iter_set_snapshot(&dst_iter, old_dst.k->p.snapshot); ret = bch2_btree_iter_traverse(&dst_iter) ?: - bch2_btree_delete_at(trans, &dst_iter, BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + bch2_btree_delete_at(trans, &dst_iter, BTREE_UPDATE_internal_snapshot_node); if (ret) goto out; } diff --git a/fs/bcachefs/dirent.h b/fs/bcachefs/dirent.h index bee55cca2aa0..635ceb3c07fb 100644 --- a/fs/bcachefs/dirent.h +++ b/fs/bcachefs/dirent.h @@ -38,11 +38,11 @@ int bch2_dirent_read_target(struct btree_trans *, subvol_inum, int bch2_dirent_create_snapshot(struct btree_trans *, u32, u64, u32, const struct bch_hash_info *, u8, const struct qstr *, u64, u64 *, - bch_str_hash_flags_t); + enum btree_iter_update_trigger_flags); int bch2_dirent_create(struct btree_trans *, subvol_inum, const struct bch_hash_info *, u8, const struct qstr *, u64, u64 *, - bch_str_hash_flags_t); + enum btree_iter_update_trigger_flags); static inline unsigned vfs_d_type(unsigned type) { diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 556a217108d3..86999ed80439 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -244,7 +244,7 @@ err: static int mark_stripe_bucket(struct btree_trans *trans, struct bkey_s_c k, unsigned ptr_idx, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; const struct bch_stripe *s = bkey_s_c_to_stripe(k).v; @@ -258,7 +258,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, struct printbuf buf = PRINTBUF; int ret = 0; - BUG_ON(!(flags & BTREE_TRIGGER_GC)); + BUG_ON(!(flags & BTREE_TRIGGER_gc)); /* * XXX doesn't handle deletion */ @@ -302,7 +302,7 @@ err: int bch2_trigger_stripe(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s _new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { struct bkey_s_c new = _new.s_c; struct bch_fs *c = trans->c; @@ -312,7 +312,7 @@ int bch2_trigger_stripe(struct btree_trans *trans, const struct bch_stripe *new_s = new.k->type == KEY_TYPE_stripe ? bkey_s_c_to_stripe(new).v : NULL; - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { /* * If the pointers aren't changing, we don't need to do anything: */ @@ -371,7 +371,7 @@ int bch2_trigger_stripe(struct btree_trans *trans, } } - if (flags & BTREE_TRIGGER_ATOMIC) { + if (flags & BTREE_TRIGGER_atomic) { struct stripe *m = genradix_ptr(&c->stripes, idx); if (!m) { @@ -410,7 +410,7 @@ int bch2_trigger_stripe(struct btree_trans *trans, } } - if (flags & BTREE_TRIGGER_GC) { + if (flags & BTREE_TRIGGER_gc) { struct gc_stripe *m = genradix_ptr_alloc(&c->gc_stripes, idx, GFP_KERNEL); @@ -769,7 +769,7 @@ static int get_stripe_key_trans(struct btree_trans *trans, u64 idx, int ret; k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_stripes, - POS(0, idx), BTREE_ITER_SLOTS); + POS(0, idx), BTREE_ITER_slots); ret = bkey_err(k); if (ret) goto err; @@ -1060,7 +1060,7 @@ static int ec_stripe_delete(struct btree_trans *trans, u64 idx) int ret; k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_stripes, POS(0, idx), - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = bkey_err(k); if (ret) goto err; @@ -1131,7 +1131,7 @@ static int ec_stripe_key_update(struct btree_trans *trans, int ret; k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_stripes, - new->k.p, BTREE_ITER_INTENT); + new->k.p, BTREE_ITER_intent); ret = bkey_err(k); if (ret) goto err; @@ -1189,7 +1189,7 @@ static int ec_stripe_update_extent(struct btree_trans *trans, int ret, dev, block; ret = bch2_get_next_backpointer(trans, bucket, gen, - bp_pos, &bp, BTREE_ITER_CACHED); + bp_pos, &bp, BTREE_ITER_cached); if (ret) return ret; if (bpos_eq(*bp_pos, SPOS_MAX)) @@ -1214,7 +1214,7 @@ static int ec_stripe_update_extent(struct btree_trans *trans, return -EIO; } - k = bch2_backpointer_get_key(trans, &iter, *bp_pos, bp, BTREE_ITER_INTENT); + k = bch2_backpointer_get_key(trans, &iter, *bp_pos, bp, BTREE_ITER_intent); ret = bkey_err(k); if (ret) return ret; @@ -1937,7 +1937,7 @@ static int __bch2_ec_stripe_head_reserve(struct btree_trans *trans, struct ec_st } for_each_btree_key_norestart(trans, iter, BTREE_ID_stripes, start_pos, - BTREE_ITER_SLOTS|BTREE_ITER_INTENT, k, ret) { + BTREE_ITER_slots|BTREE_ITER_intent, k, ret) { if (bkey_gt(k.k->p, POS(0, U32_MAX))) { if (start_pos.offset) { start_pos = min_pos; @@ -2127,7 +2127,7 @@ int bch2_stripes_read(struct bch_fs *c) { int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_stripes, POS_MIN, - BTREE_ITER_PREFETCH, k, ({ + BTREE_ITER_prefetch, k, ({ if (k.k->type != KEY_TYPE_stripe) continue; diff --git a/fs/bcachefs/ec.h b/fs/bcachefs/ec.h index f042616888b0..a1c2bc74ef45 100644 --- a/fs/bcachefs/ec.h +++ b/fs/bcachefs/ec.h @@ -13,7 +13,8 @@ int bch2_stripe_invalid(struct bch_fs *, struct bkey_s_c, void bch2_stripe_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_stripe(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); #define bch2_bkey_ops_stripe ((struct bkey_ops) { \ .key_invalid = bch2_stripe_invalid, \ diff --git a/fs/bcachefs/extent_update.c b/fs/bcachefs/extent_update.c index b9033bb4f11c..5f4fecb358da 100644 --- a/fs/bcachefs/extent_update.c +++ b/fs/bcachefs/extent_update.c @@ -72,7 +72,7 @@ static int count_iters_for_insert(struct btree_trans *trans, for_each_btree_key_norestart(trans, iter, BTREE_ID_reflink, POS(0, idx + offset), - BTREE_ITER_SLOTS, r_k, ret2) { + BTREE_ITER_slots, r_k, ret2) { if (bkey_ge(bkey_start_pos(r_k.k), POS(0, idx + sectors))) break; diff --git a/fs/bcachefs/fs-common.c b/fs/bcachefs/fs-common.c index 624e6f963240..6c737def357e 100644 --- a/fs/bcachefs/fs-common.c +++ b/fs/bcachefs/fs-common.c @@ -42,7 +42,7 @@ int bch2_create_trans(struct btree_trans *trans, if (ret) goto err; - ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_INTENT); + ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_intent); if (ret) goto err; @@ -70,7 +70,7 @@ int bch2_create_trans(struct btree_trans *trans, struct bch_subvolume s; ret = bch2_subvolume_get(trans, snapshot_src.subvol, true, - BTREE_ITER_CACHED, &s); + BTREE_ITER_cached, &s); if (ret) goto err; @@ -78,7 +78,7 @@ int bch2_create_trans(struct btree_trans *trans, } ret = bch2_inode_peek(trans, &inode_iter, new_inode, snapshot_src, - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto err; @@ -163,7 +163,7 @@ int bch2_create_trans(struct btree_trans *trans, name, dir_target, &dir_offset, - BCH_HASH_SET_MUST_CREATE); + STR_HASH_must_create); if (ret) goto err; @@ -171,7 +171,7 @@ int bch2_create_trans(struct btree_trans *trans, new_inode->bi_dir_offset = dir_offset; } - inode_iter.flags &= ~BTREE_ITER_ALL_SNAPSHOTS; + inode_iter.flags &= ~BTREE_ITER_all_snapshots; bch2_btree_iter_set_snapshot(&inode_iter, snapshot); ret = bch2_btree_iter_traverse(&inode_iter) ?: @@ -198,7 +198,7 @@ int bch2_link_trans(struct btree_trans *trans, if (dir.subvol != inum.subvol) return -EXDEV; - ret = bch2_inode_peek(trans, &inode_iter, inode_u, inum, BTREE_ITER_INTENT); + ret = bch2_inode_peek(trans, &inode_iter, inode_u, inum, BTREE_ITER_intent); if (ret) goto err; @@ -207,7 +207,7 @@ int bch2_link_trans(struct btree_trans *trans, if (ret) return ret; - ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_INTENT); + ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_intent); if (ret) goto err; @@ -223,7 +223,7 @@ int bch2_link_trans(struct btree_trans *trans, ret = bch2_dirent_create(trans, dir, &dir_hash, mode_to_type(inode_u->bi_mode), name, inum.inum, &dir_offset, - BCH_HASH_SET_MUST_CREATE); + STR_HASH_must_create); if (ret) goto err; @@ -255,19 +255,19 @@ int bch2_unlink_trans(struct btree_trans *trans, struct bkey_s_c k; int ret; - ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_INTENT); + ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_intent); if (ret) goto err; dir_hash = bch2_hash_info_init(c, dir_u); ret = bch2_dirent_lookup_trans(trans, &dirent_iter, dir, &dir_hash, - name, &inum, BTREE_ITER_INTENT); + name, &inum, BTREE_ITER_intent); if (ret) goto err; ret = bch2_inode_peek(trans, &inode_iter, inode_u, inum, - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto err; @@ -322,7 +322,7 @@ int bch2_unlink_trans(struct btree_trans *trans, ret = bch2_hash_delete_at(trans, bch2_dirent_hash_desc, &dir_hash, &dirent_iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?: + BTREE_UPDATE_internal_snapshot_node) ?: bch2_inode_write(trans, &dir_iter, dir_u) ?: bch2_inode_write(trans, &inode_iter, inode_u); err: @@ -363,7 +363,7 @@ static int subvol_update_parent(struct btree_trans *trans, u32 subvol, u32 new_p struct bkey_i_subvolume *s = bch2_bkey_get_mut_typed(trans, &iter, BTREE_ID_subvolumes, POS(0, subvol), - BTREE_ITER_CACHED, subvolume); + BTREE_ITER_cached, subvolume); int ret = PTR_ERR_OR_ZERO(s); if (ret) return ret; @@ -394,7 +394,7 @@ int bch2_rename_trans(struct btree_trans *trans, int ret; ret = bch2_inode_peek(trans, &src_dir_iter, src_dir_u, src_dir, - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto err; @@ -403,7 +403,7 @@ int bch2_rename_trans(struct btree_trans *trans, if (dst_dir.inum != src_dir.inum || dst_dir.subvol != src_dir.subvol) { ret = bch2_inode_peek(trans, &dst_dir_iter, dst_dir_u, dst_dir, - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto err; @@ -423,13 +423,13 @@ int bch2_rename_trans(struct btree_trans *trans, goto err; ret = bch2_inode_peek(trans, &src_inode_iter, src_inode_u, src_inum, - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto err; if (dst_inum.inum) { ret = bch2_inode_peek(trans, &dst_inode_iter, dst_inode_u, dst_inum, - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto err; } diff --git a/fs/bcachefs/fs-io-buffered.c b/fs/bcachefs/fs-io-buffered.c index 39292e7ef342..120430abdae7 100644 --- a/fs/bcachefs/fs-io-buffered.c +++ b/fs/bcachefs/fs-io-buffered.c @@ -176,7 +176,7 @@ retry: bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, SPOS(inum.inum, rbio->bio.bi_iter.bi_sector, snapshot), - BTREE_ITER_SLOTS); + BTREE_ITER_slots); while (1) { struct bkey_s_c k; unsigned bytes, sectors, offset_into_extent; diff --git a/fs/bcachefs/fs-io-direct.c b/fs/bcachefs/fs-io-direct.c index b889370a5088..09d21aef879a 100644 --- a/fs/bcachefs/fs-io-direct.c +++ b/fs/bcachefs/fs-io-direct.c @@ -254,7 +254,7 @@ retry: for_each_btree_key_norestart(trans, iter, BTREE_ID_extents, SPOS(inum.inum, offset, snapshot), - BTREE_ITER_SLOTS, k, err) { + BTREE_ITER_slots, k, err) { if (bkey_ge(bkey_start_pos(k.k), POS(inum.inum, end))) break; diff --git a/fs/bcachefs/fs-io-pagecache.c b/fs/bcachefs/fs-io-pagecache.c index d359aa9b33b8..872283e5bd1e 100644 --- a/fs/bcachefs/fs-io-pagecache.c +++ b/fs/bcachefs/fs-io-pagecache.c @@ -214,7 +214,7 @@ retry: for_each_btree_key_norestart(trans, iter, BTREE_ID_extents, SPOS(inum.inum, offset, snapshot), - BTREE_ITER_SLOTS, k, ret) { + BTREE_ITER_slots, k, ret) { unsigned nr_ptrs = bch2_bkey_nr_ptrs_fully_allocated(k); unsigned state = bkey_to_sector_state(k); diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c index 20b40477425f..442bcb0793c4 100644 --- a/fs/bcachefs/fs-io.c +++ b/fs/bcachefs/fs-io.c @@ -594,7 +594,7 @@ static int __bchfs_fallocate(struct bch_inode_info *inode, int mode, bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, POS(inode->v.i_ino, start_sector), - BTREE_ITER_SLOTS|BTREE_ITER_INTENT); + BTREE_ITER_slots|BTREE_ITER_intent); while (!ret && bkey_lt(iter.pos, end_pos)) { s64 i_sectors_delta = 0; @@ -1009,7 +1009,7 @@ retry: for_each_btree_key_norestart(trans, iter, BTREE_ID_extents, SPOS(inode->v.i_ino, offset >> 9, snapshot), - BTREE_ITER_SLOTS, k, ret) { + BTREE_ITER_slots, k, ret) { if (k.k->p.inode != inode->v.i_ino) { next_hole = bch2_seek_pagecache_hole(&inode->v, offset, MAX_LFS_FILESIZE, 0, false); diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 3d15275909bd..81cfc74828fd 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -90,7 +90,7 @@ retry: bch2_trans_begin(trans); ret = bch2_inode_peek(trans, &iter, &inode_u, inode_inum(inode), - BTREE_ITER_INTENT) ?: + BTREE_ITER_intent) ?: (set ? set(trans, inode, &inode_u, p) : 0) ?: bch2_inode_write(trans, &iter, &inode_u) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); @@ -323,7 +323,7 @@ retry: inum.inum = inode_u.bi_inum; ret = bch2_subvolume_get(trans, inum.subvol, true, - BTREE_ITER_WITH_UPDATES, &subvol) ?: + BTREE_ITER_with_updates, &subvol) ?: bch2_trans_commit(trans, NULL, &journal_seq, 0); if (unlikely(ret)) { bch2_quota_acct(c, bch_qid(&inode_u), Q_INO, -1, @@ -783,7 +783,7 @@ retry: acl = NULL; ret = bch2_inode_peek(trans, &inode_iter, &inode_u, inode_inum(inode), - BTREE_ITER_INTENT); + BTREE_ITER_intent); if (ret) goto btree_err; diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index caef81aa618c..d166c17109dd 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -79,7 +79,7 @@ static int lookup_first_inode(struct btree_trans *trans, u64 inode_nr, bch2_trans_iter_init(trans, &iter, BTREE_ID_inodes, POS(0, inode_nr), - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_all_snapshots); k = bch2_btree_iter_peek(&iter); ret = bkey_err(k); if (ret) @@ -154,12 +154,12 @@ static int __remove_dirent(struct btree_trans *trans, struct bpos pos) dir_hash_info = bch2_hash_info_init(c, &dir_inode); - bch2_trans_iter_init(trans, &iter, BTREE_ID_dirents, pos, BTREE_ITER_INTENT); + bch2_trans_iter_init(trans, &iter, BTREE_ID_dirents, pos, BTREE_ITER_intent); ret = bch2_btree_iter_traverse(&iter) ?: bch2_hash_delete_at(trans, bch2_dirent_hash_desc, &dir_hash_info, &iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); bch2_trans_iter_exit(trans, &iter); err: bch_err_fn(c, ret); @@ -274,9 +274,9 @@ create_lostfound: &lostfound_str, lostfound->bi_inum, &lostfound->bi_dir_offset, - BCH_HASH_SET_MUST_CREATE) ?: + STR_HASH_must_create) ?: bch2_inode_write_flags(trans, &lostfound_iter, lostfound, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); err: bch_err_msg(c, ret, "creating lost+found"); bch2_trans_iter_exit(trans, &lostfound_iter); @@ -333,7 +333,7 @@ static int reattach_inode(struct btree_trans *trans, &name, inode->bi_subvol ?: inode->bi_inum, &dir_offset, - BCH_HASH_SET_MUST_CREATE); + STR_HASH_must_create); if (ret) return ret; @@ -708,7 +708,7 @@ static int get_inodes_all_snapshots(struct btree_trans *trans, w->inodes.nr = 0; for_each_btree_key_norestart(trans, iter, BTREE_ID_inodes, POS(0, inum), - BTREE_ITER_ALL_SNAPSHOTS, k, ret) { + BTREE_ITER_all_snapshots, k, ret) { if (k.k->p.offset != inum) break; @@ -799,7 +799,7 @@ static int __get_visible_inodes(struct btree_trans *trans, w->inodes.nr = 0; for_each_btree_key_norestart(trans, iter, BTREE_ID_inodes, POS(0, inum), - BTREE_ITER_ALL_SNAPSHOTS, k, ret) { + BTREE_ITER_all_snapshots, k, ret) { u32 equiv = bch2_snapshot_equiv(c, k.k->p.snapshot); if (k.k->p.offset != inum) @@ -832,7 +832,7 @@ static int check_key_has_snapshot(struct btree_trans *trans, "key in missing snapshot: %s", (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) ret = bch2_btree_delete_at(trans, iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?: 1; + BTREE_UPDATE_internal_snapshot_node) ?: 1; fsck_err: printbuf_exit(&buf); return ret; @@ -861,8 +861,8 @@ static int hash_redo_key(struct btree_trans *trans, bch2_hash_set_in_snapshot(trans, desc, hash_info, (subvol_inum) { 0, k.k->p.inode }, k.k->p.snapshot, tmp, - BCH_HASH_SET_MUST_CREATE, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?: + STR_HASH_must_create| + BTREE_UPDATE_internal_snapshot_node) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); } @@ -891,7 +891,7 @@ static int hash_check_key(struct btree_trans *trans, for_each_btree_key_norestart(trans, iter, desc.btree_id, SPOS(hash_k.k->p.inode, hash, hash_k.k->p.snapshot), - BTREE_ITER_SLOTS, k, ret) { + BTREE_ITER_slots, k, ret) { if (bkey_eq(k.k->p, hash_k.k->p)) break; @@ -1233,7 +1233,7 @@ int bch2_check_inodes(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_inodes, POS_MIN, - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_inode(trans, &iter, k, &prev, &s, full))); @@ -1362,8 +1362,8 @@ static int overlapping_extents_found(struct btree_trans *trans, BUG_ON(bkey_le(pos1, bkey_start_pos(&pos2))); bch2_trans_iter_init(trans, &iter1, btree, pos1, - BTREE_ITER_ALL_SNAPSHOTS| - BTREE_ITER_NOT_EXTENTS); + BTREE_ITER_all_snapshots| + BTREE_ITER_not_extents); k1 = bch2_btree_iter_peek_upto(&iter1, POS(pos1.inode, U64_MAX)); ret = bkey_err(k1); if (ret) @@ -1425,7 +1425,7 @@ static int overlapping_extents_found(struct btree_trans *trans, trans->extra_disk_res += bch2_bkey_sectors_compressed(k2); ret = bch2_trans_update_extent_overwrite(trans, old_iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE, + BTREE_UPDATE_internal_snapshot_node, k1, k2) ?: bch2_trans_commit(trans, &res, NULL, BCH_TRANS_COMMIT_no_enospc); bch2_disk_reservation_put(c, &res); @@ -1625,7 +1625,7 @@ static int check_extent(struct btree_trans *trans, struct btree_iter *iter, bch2_btree_iter_set_snapshot(&iter2, i->snapshot); ret = bch2_btree_iter_traverse(&iter2) ?: bch2_btree_delete_at(trans, &iter2, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); bch2_trans_iter_exit(trans, &iter2); if (ret) goto err; @@ -1652,7 +1652,7 @@ fsck_err: bch_err_fn(c, ret); return ret; delete: - ret = bch2_btree_delete_at(trans, iter, BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + ret = bch2_btree_delete_at(trans, iter, BTREE_UPDATE_internal_snapshot_node); goto out; } @@ -1673,7 +1673,7 @@ int bch2_check_extents(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_extents, POS(BCACHEFS_ROOT_INO, 0), - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, &res, NULL, BCH_TRANS_COMMIT_no_enospc, ({ bch2_disk_reservation_put(c, &res); @@ -1698,7 +1698,7 @@ int bch2_check_indirect_extents(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_reflink, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, &res, NULL, BCH_TRANS_COMMIT_no_enospc, ({ bch2_disk_reservation_put(c, &res); @@ -2104,7 +2104,7 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { ret = bch2_btree_delete_at(trans, iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); goto out; } @@ -2191,7 +2191,7 @@ int bch2_check_dirents(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_dirents, POS(BCACHEFS_ROOT_INO, 0), - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, @@ -2255,7 +2255,7 @@ int bch2_check_xattrs(struct bch_fs *c) ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_xattrs, POS(BCACHEFS_ROOT_INO, 0), - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, @@ -2422,7 +2422,7 @@ int bch2_check_subvolume_structure(struct bch_fs *c) { int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, - BTREE_ID_subvolumes, POS_MIN, BTREE_ITER_PREFETCH, k, + BTREE_ID_subvolumes, POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_subvol_path(trans, &iter, k))); bch_err_fn(c, ret); @@ -2559,9 +2559,9 @@ int bch2_check_directory_structure(struct bch_fs *c) ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_inodes, POS_MIN, - BTREE_ITER_INTENT| - BTREE_ITER_PREFETCH| - BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_intent| + BTREE_ITER_prefetch| + BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ if (!bkey_is_inode(k.k)) continue; @@ -2661,9 +2661,9 @@ static int check_nlinks_find_hardlinks(struct bch_fs *c, int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_inodes, POS(0, start), - BTREE_ITER_INTENT| - BTREE_ITER_PREFETCH| - BTREE_ITER_ALL_SNAPSHOTS, k, ({ + BTREE_ITER_intent| + BTREE_ITER_prefetch| + BTREE_ITER_all_snapshots, k, ({ if (!bkey_is_inode(k.k)) continue; @@ -2704,9 +2704,9 @@ static int check_nlinks_walk_dirents(struct bch_fs *c, struct nlink_table *links int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_dirents, POS_MIN, - BTREE_ITER_INTENT| - BTREE_ITER_PREFETCH| - BTREE_ITER_ALL_SNAPSHOTS, k, ({ + BTREE_ITER_intent| + BTREE_ITER_prefetch| + BTREE_ITER_all_snapshots, k, ({ ret = snapshots_seen_update(c, &s, iter.btree_id, k.k->p); if (ret) break; @@ -2781,7 +2781,7 @@ static int check_nlinks_update_hardlinks(struct bch_fs *c, int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_inodes, POS(0, range_start), - BTREE_ITER_INTENT|BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_intent|BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_nlinks_update_inode(trans, &iter, k, links, &idx, range_end))); if (ret < 0) { @@ -2849,7 +2849,7 @@ static int fix_reflink_p_key(struct btree_trans *trans, struct btree_iter *iter, u->v.front_pad = 0; u->v.back_pad = 0; - return bch2_trans_update(trans, iter, &u->k_i, BTREE_TRIGGER_NORUN); + return bch2_trans_update(trans, iter, &u->k_i, BTREE_TRIGGER_norun); } int bch2_fix_reflink_p(struct bch_fs *c) @@ -2860,8 +2860,8 @@ int bch2_fix_reflink_p(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_extents, POS_MIN, - BTREE_ITER_INTENT|BTREE_ITER_PREFETCH| - BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_intent|BTREE_ITER_prefetch| + BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, fix_reflink_p_key(trans, &iter, k))); bch_err_fn(c, ret); diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index dc72c59773fe..fc8b30441bbc 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -339,7 +339,7 @@ int bch2_inode_peek_nowarn(struct btree_trans *trans, k = bch2_bkey_get_iter(trans, iter, BTREE_ID_inodes, SPOS(0, inum.inum, snapshot), - flags|BTREE_ITER_CACHED); + flags|BTREE_ITER_cached); ret = bkey_err(k); if (ret) return ret; @@ -371,7 +371,7 @@ int bch2_inode_peek(struct btree_trans *trans, int bch2_inode_write_flags(struct btree_trans *trans, struct btree_iter *iter, struct bch_inode_unpacked *inode, - enum btree_update_flags flags) + enum btree_iter_update_trigger_flags flags) { struct bkey_inode_buf *inode_p; @@ -399,7 +399,7 @@ int __bch2_fsck_write_inode(struct btree_trans *trans, return bch2_btree_insert_nonextent(trans, BTREE_ID_inodes, &inode_p->inode.k_i, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); } int bch2_fsck_write_inode(struct btree_trans *trans, @@ -598,7 +598,7 @@ int bch2_trigger_inode(struct btree_trans *trans, { s64 nr = (s64) bkey_is_inode(new.k) - (s64) bkey_is_inode(old.k); - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { if (nr) { int ret = bch2_replicas_deltas_realloc(trans, 0); if (ret) @@ -617,13 +617,13 @@ int bch2_trigger_inode(struct btree_trans *trans, } } - if ((flags & BTREE_TRIGGER_ATOMIC) && (flags & BTREE_TRIGGER_INSERT)) { + if ((flags & BTREE_TRIGGER_atomic) && (flags & BTREE_TRIGGER_insert)) { BUG_ON(!trans->journal_res.seq); bkey_s_to_inode_v3(new).v->bi_journal_seq = cpu_to_le64(trans->journal_res.seq); } - if (flags & BTREE_TRIGGER_GC) { + if (flags & BTREE_TRIGGER_gc) { struct bch_fs *c = trans->c; percpu_down_read(&c->mark_lock); @@ -752,8 +752,8 @@ int bch2_inode_create(struct btree_trans *trans, pos = start; bch2_trans_iter_init(trans, iter, BTREE_ID_inodes, POS(0, pos), - BTREE_ITER_ALL_SNAPSHOTS| - BTREE_ITER_INTENT); + BTREE_ITER_all_snapshots| + BTREE_ITER_intent); again: while ((k = bch2_btree_iter_peek(iter)).k && !(ret = bkey_err(k)) && @@ -814,7 +814,7 @@ static int bch2_inode_delete_keys(struct btree_trans *trans, * extent iterator: */ bch2_trans_iter_init(trans, &iter, id, POS(inum.inum, 0), - BTREE_ITER_INTENT); + BTREE_ITER_intent); while (1) { bch2_trans_begin(trans); @@ -836,7 +836,7 @@ static int bch2_inode_delete_keys(struct btree_trans *trans, bkey_init(&delete.k); delete.k.p = iter.pos; - if (iter.flags & BTREE_ITER_IS_EXTENTS) + if (iter.flags & BTREE_ITER_is_extents) bch2_key_resize(&delete.k, bpos_min(end, k.k->p).offset - iter.pos.offset); @@ -885,7 +885,7 @@ retry: k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, SPOS(0, inum.inum, snapshot), - BTREE_ITER_INTENT|BTREE_ITER_CACHED); + BTREE_ITER_intent|BTREE_ITER_cached); ret = bkey_err(k); if (ret) goto err; @@ -1045,7 +1045,7 @@ retry: bch2_trans_begin(trans); k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, - SPOS(0, inum, snapshot), BTREE_ITER_INTENT); + SPOS(0, inum, snapshot), BTREE_ITER_intent); ret = bkey_err(k); if (ret) goto err; @@ -1090,7 +1090,7 @@ static int may_delete_deleted_inode(struct btree_trans *trans, struct bch_inode_unpacked inode; int ret; - k = bch2_bkey_get_iter(trans, &inode_iter, BTREE_ID_inodes, pos, BTREE_ITER_CACHED); + k = bch2_bkey_get_iter(trans, &inode_iter, BTREE_ID_inodes, pos, BTREE_ITER_cached); ret = bkey_err(k); if (ret) return ret; @@ -1142,7 +1142,7 @@ static int may_delete_deleted_inode(struct btree_trans *trans, inode.bi_flags &= ~BCH_INODE_unlinked; ret = bch2_inode_write_flags(trans, &inode_iter, &inode, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); bch_err_msg(c, ret, "clearing inode unlinked flag"); if (ret) goto out; @@ -1189,7 +1189,7 @@ again: * flushed and we'd spin: */ ret = for_each_btree_key_commit(trans, iter, BTREE_ID_deleted_inodes, POS_MIN, - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ ret = may_delete_deleted_inode(trans, &iter, k.k->p, &need_another_pass); if (ret > 0) { diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h index 056298050550..08e8505dfa33 100644 --- a/fs/bcachefs/inode.h +++ b/fs/bcachefs/inode.h @@ -101,7 +101,7 @@ int bch2_inode_peek(struct btree_trans *, struct btree_iter *, struct bch_inode_unpacked *, subvol_inum, unsigned); int bch2_inode_write_flags(struct btree_trans *, struct btree_iter *, - struct bch_inode_unpacked *, enum btree_update_flags); + struct bch_inode_unpacked *, enum btree_iter_update_trigger_flags); static inline int bch2_inode_write(struct btree_trans *trans, struct btree_iter *iter, diff --git a/fs/bcachefs/io_misc.c b/fs/bcachefs/io_misc.c index 82f9170dab3f..4ec979b4b23e 100644 --- a/fs/bcachefs/io_misc.c +++ b/fs/bcachefs/io_misc.c @@ -198,7 +198,7 @@ int bch2_fpunch(struct bch_fs *c, subvol_inum inum, u64 start, u64 end, bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, POS(inum.inum, start), - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = bch2_fpunch_at(trans, &iter, inum, end, i_sectors_delta); @@ -230,7 +230,7 @@ static int truncate_set_isize(struct btree_trans *trans, struct bch_inode_unpacked inode_u; int ret; - ret = bch2_inode_peek(trans, &iter, &inode_u, inum, BTREE_ITER_INTENT) ?: + ret = bch2_inode_peek(trans, &iter, &inode_u, inum, BTREE_ITER_intent) ?: (inode_u.bi_size = new_i_size, 0) ?: bch2_inode_write(trans, &iter, &inode_u); @@ -256,7 +256,7 @@ static int __bch2_resume_logged_op_truncate(struct btree_trans *trans, bch2_trans_iter_init(trans, &fpunch_iter, BTREE_ID_extents, POS(inum.inum, round_up(new_i_size, block_bytes(c)) >> 9), - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = bch2_fpunch_at(trans, &fpunch_iter, inum, U64_MAX, i_sectors_delta); bch2_trans_iter_exit(trans, &fpunch_iter); @@ -317,7 +317,7 @@ static int adjust_i_size(struct btree_trans *trans, subvol_inum inum, u64 offset offset <<= 9; len <<= 9; - ret = bch2_inode_peek(trans, &iter, &inode_u, inum, BTREE_ITER_INTENT); + ret = bch2_inode_peek(trans, &iter, &inode_u, inum, BTREE_ITER_intent); if (ret) return ret; @@ -365,7 +365,7 @@ static int __bch2_resume_logged_op_finsert(struct btree_trans *trans, bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, POS(inum.inum, 0), - BTREE_ITER_INTENT); + BTREE_ITER_intent); switch (op->v.state) { case LOGGED_OP_FINSERT_start: diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index c07902b6c759..475ab6c02dd1 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -378,7 +378,7 @@ static void bch2_read_retry_nodecode(struct bch_fs *c, struct bch_read_bio *rbio bch2_bkey_buf_init(&sk); bch2_trans_iter_init(trans, &iter, rbio->data_btree, - rbio->read_pos, BTREE_ITER_SLOTS); + rbio->read_pos, BTREE_ITER_slots); retry: rbio->bio.bi_status = 0; @@ -487,7 +487,7 @@ static int __bch2_rbio_narrow_crcs(struct btree_trans *trans, return 0; k = bch2_bkey_get_iter(trans, &iter, rbio->data_btree, rbio->data_pos, - BTREE_ITER_SLOTS|BTREE_ITER_INTENT); + BTREE_ITER_slots|BTREE_ITER_intent); if ((ret = bkey_err(k))) goto out; @@ -523,7 +523,7 @@ static int __bch2_rbio_narrow_crcs(struct btree_trans *trans, goto out; ret = bch2_trans_update(trans, &iter, new, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); out: bch2_trans_iter_exit(trans, &iter); return ret; @@ -769,7 +769,7 @@ static noinline void read_from_stale_dirty_pointer(struct btree_trans *trans, bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, PTR_BUCKET_POS(c, &ptr), - BTREE_ITER_CACHED); + BTREE_ITER_cached); prt_printf(&buf, "Attempting to read from stale dirty pointer:\n"); printbuf_indent_add(&buf, 2); @@ -1112,7 +1112,7 @@ retry: bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, SPOS(inum.inum, bvec_iter.bi_sector, snapshot), - BTREE_ITER_SLOTS); + BTREE_ITER_slots); while (1) { unsigned bytes, sectors, offset_into_extent; enum btree_id data_btree = BTREE_ID_extents; diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index c11e0944b069..9826772c0eb4 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -166,7 +166,7 @@ int bch2_sum_sector_overwrites(struct btree_trans *trans, bch2_trans_copy_iter(&iter, extent_iter); for_each_btree_key_upto_continue_norestart(iter, - new->k.p, BTREE_ITER_SLOTS, old, ret) { + new->k.p, BTREE_ITER_slots, old, ret) { s64 sectors = min(new->k.p.offset, old.k->p.offset) - max(bkey_start_offset(&new->k), bkey_start_offset(old.k)); @@ -210,14 +210,14 @@ static inline int bch2_extent_update_i_size_sectors(struct btree_trans *trans, * to be journalled - if we crash, the bi_journal_seq update will be * lost, but that's fine. */ - unsigned inode_update_flags = BTREE_UPDATE_NOJOURNAL; + unsigned inode_update_flags = BTREE_UPDATE_nojournal; struct btree_iter iter; struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, SPOS(0, extent_iter->pos.inode, extent_iter->snapshot), - BTREE_ITER_CACHED); + BTREE_ITER_cached); int ret = bkey_err(k); if (unlikely(ret)) return ret; @@ -259,7 +259,7 @@ static inline int bch2_extent_update_i_size_sectors(struct btree_trans *trans, } ret = bch2_trans_update(trans, &iter, &inode->k_i, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE| + BTREE_UPDATE_internal_snapshot_node| inode_update_flags); err: bch2_trans_iter_exit(trans, &iter); @@ -368,7 +368,7 @@ static int bch2_write_index_default(struct bch_write_op *op) bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, bkey_start_pos(&sk.k->k), - BTREE_ITER_SLOTS|BTREE_ITER_INTENT); + BTREE_ITER_slots|BTREE_ITER_intent); ret = bch2_bkey_set_needs_rebalance(c, sk.k, &op->opts) ?: bch2_extent_update(trans, inum, &iter, sk.k, @@ -1158,7 +1158,7 @@ static int bch2_nocow_write_convert_one_unwritten(struct btree_trans *trans, return bch2_extent_update_i_size_sectors(trans, iter, min(new->k.p.offset << 9, new_i_size), 0) ?: bch2_trans_update(trans, iter, new, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); } static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op) @@ -1169,7 +1169,7 @@ static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op) for_each_keylist_key(&op->insert_keys, orig) { int ret = for_each_btree_key_upto_commit(trans, iter, BTREE_ID_extents, bkey_start_pos(&orig->k), orig->k.p, - BTREE_ITER_INTENT, k, + BTREE_ITER_intent, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ bch2_nocow_write_convert_one_unwritten(trans, &iter, orig, k, op->new_i_size); })); @@ -1242,7 +1242,7 @@ retry: bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, SPOS(op->pos.inode, op->pos.offset, snapshot), - BTREE_ITER_SLOTS); + BTREE_ITER_slots); while (1) { struct bio *bio = &op->wbio.bio; diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index fc02cb4ee74f..1b62034e197d 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -946,7 +946,7 @@ static int __bch2_set_nr_journal_buckets(struct bch_dev *ca, unsigned nr, ret = bch2_trans_run(c, bch2_trans_mark_metadata_bucket(trans, ca, ob[nr_got]->bucket, BCH_DATA_journal, - ca->mi.bucket_size, BTREE_TRIGGER_TRANSACTIONAL)); + ca->mi.bucket_size, BTREE_TRIGGER_transactional)); if (ret) { bch2_open_bucket_put(c, ob[nr_got]); bch_err_msg(c, ret, "marking new journal buckets"); @@ -1027,7 +1027,7 @@ err_unblock: bch2_trans_run(c, bch2_trans_mark_metadata_bucket(trans, ca, bu[i], BCH_DATA_free, 0, - BTREE_TRIGGER_TRANSACTIONAL)); + BTREE_TRIGGER_transactional)); err_free: if (!new_fs) for (i = 0; i < nr_got; i++) diff --git a/fs/bcachefs/journal_seq_blacklist.c b/fs/bcachefs/journal_seq_blacklist.c index 37a024e034d4..c87c68ae27f4 100644 --- a/fs/bcachefs/journal_seq_blacklist.c +++ b/fs/bcachefs/journal_seq_blacklist.c @@ -233,7 +233,7 @@ void bch2_blacklist_entries_gc(struct work_struct *work) struct btree *b; bch2_trans_node_iter_init(trans, &iter, i, POS_MIN, - 0, 0, BTREE_ITER_PREFETCH); + 0, 0, BTREE_ITER_prefetch); retry: bch2_trans_begin(trans); diff --git a/fs/bcachefs/logged_ops.c b/fs/bcachefs/logged_ops.c index b82f8209041f..f49fdca1d07d 100644 --- a/fs/bcachefs/logged_ops.c +++ b/fs/bcachefs/logged_ops.c @@ -56,7 +56,7 @@ int bch2_resume_logged_ops(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_logged_ops, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, resume_logged_op(trans, &iter, k))); bch_err_fn(c, ret); return ret; diff --git a/fs/bcachefs/lru.c b/fs/bcachefs/lru.c index 26569043e368..19285fcd252c 100644 --- a/fs/bcachefs/lru.c +++ b/fs/bcachefs/lru.c @@ -149,7 +149,7 @@ int bch2_check_lrus(struct bch_fs *c) struct bpos last_flushed_pos = POS_MIN; int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, - BTREE_ID_lru, POS_MIN, BTREE_ITER_PREFETCH, k, + BTREE_ID_lru, POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc|BCH_TRANS_COMMIT_lazy_rw, bch2_check_lru_key(trans, &iter, k, &last_flushed_pos))); bch_err_fn(c, ret); diff --git a/fs/bcachefs/migrate.c b/fs/bcachefs/migrate.c index 69098eeb5d48..ddc187fb693d 100644 --- a/fs/bcachefs/migrate.c +++ b/fs/bcachefs/migrate.c @@ -49,7 +49,7 @@ static int bch2_dev_usrdata_drop_key(struct btree_trans *trans, if (!bch2_bkey_has_device_c(k, dev_idx)) return 0; - n = bch2_bkey_make_mut(trans, iter, &k, BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + n = bch2_bkey_make_mut(trans, iter, &k, BTREE_UPDATE_internal_snapshot_node); ret = PTR_ERR_OR_ZERO(n); if (ret) return ret; @@ -67,7 +67,7 @@ static int bch2_dev_usrdata_drop_key(struct btree_trans *trans, /* * Since we're not inserting through an extent iterator - * (BTREE_ITER_ALL_SNAPSHOTS iterators aren't extent iterators), + * (BTREE_ITER_all_snapshots iterators aren't extent iterators), * we aren't using the extent overwrite path to delete, we're * just using the normal key deletion path: */ @@ -87,7 +87,7 @@ static int bch2_dev_usrdata_drop(struct bch_fs *c, unsigned dev_idx, int flags) continue; ret = for_each_btree_key_commit(trans, iter, id, POS_MIN, - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_dev_usrdata_drop_key(trans, &iter, k, dev_idx, flags)); if (ret) @@ -119,7 +119,7 @@ static int bch2_dev_metadata_drop(struct bch_fs *c, unsigned dev_idx, int flags) for (id = 0; id < BTREE_ID_NR; id++) { bch2_trans_node_iter_init(trans, &iter, id, POS_MIN, 0, 0, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); retry: ret = 0; while (bch2_trans_begin(trans), diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index e491203f21c1..db383dc239bb 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -416,7 +416,7 @@ struct bch_io_opts *bch2_move_get_io_opts(struct btree_trans *trans, io_opts->d.nr = 0; ret = for_each_btree_key(trans, iter, BTREE_ID_inodes, POS(0, extent_k.k->p.inode), - BTREE_ITER_ALL_SNAPSHOTS, k, ({ + BTREE_ITER_all_snapshots, k, ({ if (k.k->p.offset != extent_k.k->p.inode) break; @@ -462,7 +462,7 @@ int bch2_move_get_io_opts_one(struct btree_trans *trans, k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, SPOS(0, extent_k.k->p.inode, extent_k.k->p.snapshot), - BTREE_ITER_CACHED); + BTREE_ITER_cached); ret = bkey_err(k); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) return ret; @@ -548,8 +548,8 @@ static int bch2_move_data_btree(struct moving_context *ctxt, } bch2_trans_iter_init(trans, &iter, btree_id, start, - BTREE_ITER_PREFETCH| - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_prefetch| + BTREE_ITER_all_snapshots); if (ctxt->rate) bch2_ratelimit_reset(ctxt->rate); @@ -700,7 +700,7 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, bch2_trans_begin(trans); bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, - bucket, BTREE_ITER_CACHED); + bucket, BTREE_ITER_cached); ret = lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_slot(&iter))); bch2_trans_iter_exit(trans, &iter); @@ -727,7 +727,7 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, ret = bch2_get_next_backpointer(trans, bucket, gen, &bp_pos, &bp, - BTREE_ITER_CACHED); + BTREE_ITER_cached); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) continue; if (ret) @@ -863,7 +863,7 @@ static int bch2_move_btree(struct bch_fs *c, continue; bch2_trans_node_iter_init(trans, &iter, btree, POS_MIN, 0, 0, - BTREE_ITER_PREFETCH); + BTREE_ITER_prefetch); retry: ret = 0; while (bch2_trans_begin(trans), diff --git a/fs/bcachefs/movinggc.c b/fs/bcachefs/movinggc.c index 0d2b82d8d11f..d0afd8bc0193 100644 --- a/fs/bcachefs/movinggc.c +++ b/fs/bcachefs/movinggc.c @@ -84,7 +84,7 @@ static int bch2_bucket_is_movable(struct btree_trans *trans, return 0; k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_alloc, - b->k.bucket, BTREE_ITER_CACHED); + b->k.bucket, BTREE_ITER_cached); ret = bkey_err(k); if (ret) return ret; diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index 84e452835a17..74f4d92849bb 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -480,7 +480,7 @@ enum fsck_err_opts { OPT_FS|OPT_MOUNT|OPT_RUNTIME, \ OPT_BOOL(), \ BCH2_NO_SB_OPT, true, \ - NULL, "BTREE_ITER_PREFETCH casuse btree nodes to be\n"\ + NULL, "BTREE_ITER_prefetch casuse btree nodes to be\n"\ " prefetched sequentially") struct bch_opts { diff --git a/fs/bcachefs/quota.c b/fs/bcachefs/quota.c index 5660798c6453..9d7498e774e2 100644 --- a/fs/bcachefs/quota.c +++ b/fs/bcachefs/quota.c @@ -536,10 +536,10 @@ int bch2_fs_quota_read(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_quotas, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, __bch2_quota_set(c, k, NULL)) ?: for_each_btree_key(trans, iter, BTREE_ID_inodes, POS_MIN, - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, bch2_fs_quota_read_inode(trans, &iter, k))); bch_err_fn(c, ret); return ret; @@ -826,7 +826,7 @@ static int bch2_set_quota_trans(struct btree_trans *trans, int ret; k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_quotas, new_quota->k.p, - BTREE_ITER_SLOTS|BTREE_ITER_INTENT); + BTREE_ITER_slots|BTREE_ITER_intent); ret = bkey_err(k); if (unlikely(ret)) return ret; diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c index 56336f3dd1d0..c640c0f13976 100644 --- a/fs/bcachefs/rebalance.c +++ b/fs/bcachefs/rebalance.c @@ -42,7 +42,7 @@ static int __bch2_set_rebalance_needs_scan(struct btree_trans *trans, u64 inum) bch2_trans_iter_init(trans, &iter, BTREE_ID_rebalance_work, SPOS(inum, REBALANCE_WORK_SCAN_OFFSET, U32_MAX), - BTREE_ITER_INTENT); + BTREE_ITER_intent); k = bch2_btree_iter_peek_slot(&iter); ret = bkey_err(k); if (ret) @@ -89,7 +89,7 @@ static int bch2_clear_rebalance_needs_scan(struct btree_trans *trans, u64 inum, bch2_trans_iter_init(trans, &iter, BTREE_ID_rebalance_work, SPOS(inum, REBALANCE_WORK_SCAN_OFFSET, U32_MAX), - BTREE_ITER_INTENT); + BTREE_ITER_intent); k = bch2_btree_iter_peek_slot(&iter); ret = bkey_err(k); if (ret) @@ -140,7 +140,7 @@ static struct bkey_s_c next_rebalance_extent(struct btree_trans *trans, bch2_trans_iter_init(trans, extent_iter, work_pos.inode ? BTREE_ID_extents : BTREE_ID_reflink, work_pos, - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_all_snapshots); k = bch2_btree_iter_peek_slot(extent_iter); if (bkey_err(k)) return k; @@ -328,7 +328,7 @@ static int do_rebalance(struct moving_context *ctxt) bch2_trans_iter_init(trans, &rebalance_work_iter, BTREE_ID_rebalance_work, POS_MIN, - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_all_snapshots); while (!bch2_move_ratelimit(ctxt)) { if (!r->enabled) { diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index d22579a036c7..e997217374c6 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -136,9 +136,9 @@ static int bch2_journal_replay_key(struct btree_trans *trans, { struct btree_iter iter; unsigned iter_flags = - BTREE_ITER_INTENT| - BTREE_ITER_NOT_EXTENTS; - unsigned update_flags = BTREE_TRIGGER_NORUN; + BTREE_ITER_intent| + BTREE_ITER_not_extents; + unsigned update_flags = BTREE_TRIGGER_norun; int ret; if (k->overwritten) @@ -147,17 +147,17 @@ static int bch2_journal_replay_key(struct btree_trans *trans, trans->journal_res.seq = k->journal_seq; /* - * BTREE_UPDATE_KEY_CACHE_RECLAIM disables key cache lookup/update to + * BTREE_UPDATE_key_cache_reclaim disables key cache lookup/update to * keep the key cache coherent with the underlying btree. Nothing * besides the allocator is doing updates yet so we don't need key cache * coherency for non-alloc btrees, and key cache fills for snapshots - * btrees use BTREE_ITER_FILTER_SNAPSHOTS, which isn't available until + * btrees use BTREE_ITER_filter_snapshots, which isn't available until * the snapshots recovery pass runs. */ if (!k->level && k->btree_id == BTREE_ID_alloc) - iter_flags |= BTREE_ITER_CACHED; + iter_flags |= BTREE_ITER_cached; else - update_flags |= BTREE_UPDATE_KEY_CACHE_RECLAIM; + update_flags |= BTREE_UPDATE_key_cache_reclaim; bch2_trans_node_iter_init(trans, &iter, k->btree_id, k->k->k.p, BTREE_MAX_DEPTH, k->level, diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index ff7864731a07..a0cf5d97513f 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -74,20 +74,20 @@ bool bch2_reflink_p_merge(struct bch_fs *c, struct bkey_s _l, struct bkey_s_c _r } static int trans_trigger_reflink_p_segment(struct btree_trans *trans, - struct bkey_s_c_reflink_p p, - u64 *idx, unsigned flags) + struct bkey_s_c_reflink_p p, u64 *idx, + enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; struct btree_iter iter; struct bkey_i *k; __le64 *refcount; - int add = !(flags & BTREE_TRIGGER_OVERWRITE) ? 1 : -1; + int add = !(flags & BTREE_TRIGGER_overwrite) ? 1 : -1; struct printbuf buf = PRINTBUF; int ret; k = bch2_bkey_get_mut_noupdate(trans, &iter, BTREE_ID_reflink, POS(0, *idx), - BTREE_ITER_WITH_UPDATES); + BTREE_ITER_with_updates); ret = PTR_ERR_OR_ZERO(k); if (ret) goto err; @@ -102,7 +102,7 @@ static int trans_trigger_reflink_p_segment(struct btree_trans *trans, goto err; } - if (!*refcount && (flags & BTREE_TRIGGER_OVERWRITE)) { + if (!*refcount && (flags & BTREE_TRIGGER_overwrite)) { bch2_bkey_val_to_text(&buf, c, p.s_c); bch2_trans_inconsistent(trans, "indirect extent refcount underflow at %llu while marking\n %s", @@ -111,7 +111,7 @@ static int trans_trigger_reflink_p_segment(struct btree_trans *trans, goto err; } - if (flags & BTREE_TRIGGER_INSERT) { + if (flags & BTREE_TRIGGER_insert) { struct bch_reflink_p *v = (struct bch_reflink_p *) p.v; u64 pad; @@ -141,12 +141,13 @@ err: } static s64 gc_trigger_reflink_p_segment(struct btree_trans *trans, - struct bkey_s_c_reflink_p p, - u64 *idx, unsigned flags, size_t r_idx) + struct bkey_s_c_reflink_p p, u64 *idx, + enum btree_iter_update_trigger_flags flags, + size_t r_idx) { struct bch_fs *c = trans->c; struct reflink_gc *r; - int add = !(flags & BTREE_TRIGGER_OVERWRITE) ? 1 : -1; + int add = !(flags & BTREE_TRIGGER_overwrite) ? 1 : -1; u64 start = le64_to_cpu(p.v->idx); u64 end = le64_to_cpu(p.v->idx) + p.k->size; u64 next_idx = end + le32_to_cpu(p.v->back_pad); @@ -189,7 +190,7 @@ not_found: set_bkey_val_u64s(&update->k, 0); } - ret = bch2_btree_insert_trans(trans, BTREE_ID_extents, update, BTREE_TRIGGER_NORUN); + ret = bch2_btree_insert_trans(trans, BTREE_ID_extents, update, BTREE_TRIGGER_norun); } *idx = next_idx; @@ -200,8 +201,8 @@ fsck_err: } static int __trigger_reflink_p(struct btree_trans *trans, - enum btree_id btree_id, unsigned level, - struct bkey_s_c k, unsigned flags) + enum btree_id btree_id, unsigned level, struct bkey_s_c k, + enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k); @@ -210,12 +211,12 @@ static int __trigger_reflink_p(struct btree_trans *trans, u64 idx = le64_to_cpu(p.v->idx) - le32_to_cpu(p.v->front_pad); u64 end = le64_to_cpu(p.v->idx) + p.k->size + le32_to_cpu(p.v->back_pad); - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { while (idx < end && !ret) ret = trans_trigger_reflink_p_segment(trans, p, &idx, flags); } - if (flags & BTREE_TRIGGER_GC) { + if (flags & BTREE_TRIGGER_gc) { size_t l = 0, r = c->reflink_gc_nr; while (l < r) { @@ -238,10 +239,10 @@ int bch2_trigger_reflink_p(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { - if ((flags & BTREE_TRIGGER_TRANSACTIONAL) && - (flags & BTREE_TRIGGER_INSERT)) { + if ((flags & BTREE_TRIGGER_transactional) && + (flags & BTREE_TRIGGER_insert)) { struct bch_reflink_p *v = bkey_s_to_reflink_p(new).v; v->front_pad = v->back_pad = 0; @@ -283,21 +284,21 @@ bool bch2_reflink_v_merge(struct bch_fs *c, struct bkey_s _l, struct bkey_s_c _r static inline void check_indirect_extent_deleting(struct bkey_s new, unsigned *flags) { - if ((*flags & BTREE_TRIGGER_INSERT) && !*bkey_refcount(new)) { + if ((*flags & BTREE_TRIGGER_insert) && !*bkey_refcount(new)) { new.k->type = KEY_TYPE_deleted; new.k->size = 0; set_bkey_val_u64s(new.k, 0); - *flags &= ~BTREE_TRIGGER_INSERT; + *flags &= ~BTREE_TRIGGER_insert; } } int bch2_trigger_reflink_v(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { - if ((flags & BTREE_TRIGGER_TRANSACTIONAL) && - (flags & BTREE_TRIGGER_INSERT)) + if ((flags & BTREE_TRIGGER_transactional) && + (flags & BTREE_TRIGGER_insert)) check_indirect_extent_deleting(new, &flags); return bch2_trigger_extent(trans, btree_id, level, old, new, flags); @@ -349,7 +350,7 @@ static int bch2_make_extent_indirect(struct btree_trans *trans, bch2_check_set_feature(c, BCH_FEATURE_reflink_inline_data); bch2_trans_iter_init(trans, &reflink_iter, BTREE_ID_reflink, POS_MAX, - BTREE_ITER_INTENT); + BTREE_ITER_intent); k = bch2_btree_iter_peek_prev(&reflink_iter); ret = bkey_err(k); if (ret) @@ -394,7 +395,7 @@ static int bch2_make_extent_indirect(struct btree_trans *trans, r_p->v.idx = cpu_to_le64(bkey_start_offset(&r_v->k)); ret = bch2_trans_update(trans, extent_iter, &r_p->k_i, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); err: bch2_trans_iter_exit(trans, &reflink_iter); @@ -455,9 +456,9 @@ s64 bch2_remap_range(struct bch_fs *c, goto err; bch2_trans_iter_init(trans, &src_iter, BTREE_ID_extents, src_start, - BTREE_ITER_INTENT); + BTREE_ITER_intent); bch2_trans_iter_init(trans, &dst_iter, BTREE_ID_extents, dst_start, - BTREE_ITER_INTENT); + BTREE_ITER_intent); while ((ret == 0 || bch2_err_matches(ret, BCH_ERR_transaction_restart)) && @@ -567,7 +568,7 @@ s64 bch2_remap_range(struct bch_fs *c, bch2_trans_begin(trans); ret2 = bch2_inode_peek(trans, &inode_iter, &inode_u, - dst_inum, BTREE_ITER_INTENT); + dst_inum, BTREE_ITER_intent); if (!ret2 && inode_u.bi_size < new_i_size) { diff --git a/fs/bcachefs/reflink.h b/fs/bcachefs/reflink.h index 4d8867289717..e34c4bdf7a76 100644 --- a/fs/bcachefs/reflink.h +++ b/fs/bcachefs/reflink.h @@ -10,7 +10,8 @@ void bch2_reflink_p_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); bool bch2_reflink_p_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); int bch2_trigger_reflink_p(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); #define bch2_bkey_ops_reflink_p ((struct bkey_ops) { \ .key_invalid = bch2_reflink_p_invalid, \ @@ -25,7 +26,8 @@ int bch2_reflink_v_invalid(struct bch_fs *, struct bkey_s_c, void bch2_reflink_v_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_reflink_v(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); #define bch2_bkey_ops_reflink_v ((struct bkey_ops) { \ .key_invalid = bch2_reflink_v_invalid, \ diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index c415ea0a649b..0b26dee17a5a 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -49,7 +49,7 @@ int bch2_snapshot_tree_lookup(struct btree_trans *trans, u32 id, struct bch_snapshot_tree *s) { int ret = bch2_bkey_get_val_typed(trans, BTREE_ID_snapshot_trees, POS(0, id), - BTREE_ITER_WITH_UPDATES, snapshot_tree, s); + BTREE_ITER_with_updates, snapshot_tree, s); if (bch2_err_matches(ret, ENOENT)) ret = -BCH_ERR_ENOENT_snapshot_tree; @@ -361,7 +361,7 @@ int bch2_snapshot_lookup(struct btree_trans *trans, u32 id, struct bch_snapshot *s) { return bch2_bkey_get_val_typed(trans, BTREE_ID_snapshots, POS(0, id), - BTREE_ITER_WITH_UPDATES, snapshot, s); + BTREE_ITER_with_updates, snapshot, s); } static int bch2_snapshot_live(struct btree_trans *trans, u32 id) @@ -618,7 +618,7 @@ int bch2_check_snapshot_trees(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_snapshot_trees, POS_MIN, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_snapshot_tree(trans, &iter, k))); bch_err_fn(c, ret); @@ -695,7 +695,7 @@ static int snapshot_tree_ptr_repair(struct btree_trans *trans, root = bch2_bkey_get_iter_typed(trans, &root_iter, BTREE_ID_snapshots, POS(0, root_id), - BTREE_ITER_WITH_UPDATES, snapshot); + BTREE_ITER_with_updates, snapshot); ret = bkey_err(root); if (ret) goto err; @@ -886,7 +886,7 @@ int bch2_check_snapshots(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_reverse_commit(trans, iter, BTREE_ID_snapshots, POS_MAX, - BTREE_ITER_PREFETCH, k, + BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_snapshot(trans, &iter, k))); bch_err_fn(c, ret); @@ -1001,7 +1001,7 @@ int bch2_reconstruct_snapshots(struct bch_fs *c) r.btree = btree; ret = for_each_btree_key(trans, iter, btree, POS_MIN, - BTREE_ITER_ALL_SNAPSHOTS|BTREE_ITER_PREFETCH, k, ({ + BTREE_ITER_all_snapshots|BTREE_ITER_prefetch, k, ({ get_snapshot_trees(c, &r, k.k->p); })); if (ret) @@ -1090,7 +1090,7 @@ static int bch2_snapshot_node_delete(struct btree_trans *trans, u32 id) int ret = 0; s = bch2_bkey_get_iter_typed(trans, &iter, BTREE_ID_snapshots, POS(0, id), - BTREE_ITER_INTENT, snapshot); + BTREE_ITER_intent, snapshot); ret = bkey_err(s); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), c, "missing snapshot %u", id); @@ -1199,7 +1199,7 @@ static int create_snapids(struct btree_trans *trans, u32 parent, u32 tree, int ret; bch2_trans_iter_init(trans, &iter, BTREE_ID_snapshots, - POS_MIN, BTREE_ITER_INTENT); + POS_MIN, BTREE_ITER_intent); k = bch2_btree_iter_peek(&iter); ret = bkey_err(k); if (ret) @@ -1367,7 +1367,7 @@ static int snapshot_delete_key(struct btree_trans *trans, if (snapshot_list_has_id(deleted, k.k->p.snapshot) || snapshot_list_has_id(equiv_seen, equiv)) { return bch2_btree_delete_at(trans, iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); } else { return snapshot_list_add(c, equiv_seen, equiv); } @@ -1404,15 +1404,15 @@ static int move_key_to_correct_snapshot(struct btree_trans *trans, new->k.p.snapshot = equiv; bch2_trans_iter_init(trans, &new_iter, iter->btree_id, new->k.p, - BTREE_ITER_ALL_SNAPSHOTS| - BTREE_ITER_CACHED| - BTREE_ITER_INTENT); + BTREE_ITER_all_snapshots| + BTREE_ITER_cached| + BTREE_ITER_intent); ret = bch2_btree_iter_traverse(&new_iter) ?: bch2_trans_update(trans, &new_iter, new, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?: + BTREE_UPDATE_internal_snapshot_node) ?: bch2_btree_delete_at(trans, iter, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE); + BTREE_UPDATE_internal_snapshot_node); bch2_trans_iter_exit(trans, &new_iter); if (ret) return ret; @@ -1603,12 +1603,12 @@ int bch2_delete_dead_snapshots(struct bch_fs *c) ret = for_each_btree_key_commit(trans, iter, id, POS_MIN, - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, &res, NULL, BCH_TRANS_COMMIT_no_enospc, snapshot_delete_key(trans, &iter, k, &deleted, &equiv_seen, &last_pos)) ?: for_each_btree_key_commit(trans, iter, id, POS_MIN, - BTREE_ITER_PREFETCH|BTREE_ITER_ALL_SNAPSHOTS, k, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, &res, NULL, BCH_TRANS_COMMIT_no_enospc, move_key_to_correct_snapshot(trans, &iter, k)); @@ -1643,7 +1643,7 @@ int bch2_delete_dead_snapshots(struct bch_fs *c) * nodes some depth fields will be off: */ ret = for_each_btree_key_commit(trans, iter, BTREE_ID_snapshots, POS_MIN, - BTREE_ITER_INTENT, k, + BTREE_ITER_intent, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_fix_child_of_deleted_snapshot(trans, &iter, k, &deleted_interior)); if (ret) @@ -1699,8 +1699,8 @@ int __bch2_key_has_snapshot_overwrites(struct btree_trans *trans, int ret; bch2_trans_iter_init(trans, &iter, id, pos, - BTREE_ITER_NOT_EXTENTS| - BTREE_ITER_ALL_SNAPSHOTS); + BTREE_ITER_not_extents| + BTREE_ITER_all_snapshots); while (1) { k = bch2_btree_iter_prev(&iter); ret = bkey_err(k); @@ -1752,7 +1752,7 @@ static int bch2_propagate_key_to_snapshot_leaf(struct btree_trans *trans, pos.snapshot = leaf_id; - bch2_trans_iter_init(trans, &iter, btree, pos, BTREE_ITER_INTENT); + bch2_trans_iter_init(trans, &iter, btree, pos, BTREE_ITER_intent); k = bch2_btree_iter_peek_slot(&iter); ret = bkey_err(k); if (ret) diff --git a/fs/bcachefs/str_hash.h b/fs/bcachefs/str_hash.h index 11f5f5d27d23..cbad9b27874f 100644 --- a/fs/bcachefs/str_hash.h +++ b/fs/bcachefs/str_hash.h @@ -15,16 +15,6 @@ #include #include -typedef unsigned __bitwise bch_str_hash_flags_t; - -enum bch_str_hash_flags { - __BCH_HASH_SET_MUST_CREATE, - __BCH_HASH_SET_MUST_REPLACE, -}; - -#define BCH_HASH_SET_MUST_CREATE (__force bch_str_hash_flags_t) BIT(__BCH_HASH_SET_MUST_CREATE) -#define BCH_HASH_SET_MUST_REPLACE (__force bch_str_hash_flags_t) BIT(__BCH_HASH_SET_MUST_REPLACE) - static inline enum bch_str_hash_type bch2_str_hash_opt_to_type(struct bch_fs *c, enum bch_str_hash_opts opt) { @@ -165,7 +155,8 @@ bch2_hash_lookup_in_snapshot(struct btree_trans *trans, const struct bch_hash_desc desc, const struct bch_hash_info *info, subvol_inum inum, const void *key, - unsigned flags, u32 snapshot) + enum btree_iter_update_trigger_flags flags, + u32 snapshot) { struct bkey_s_c k; int ret; @@ -173,7 +164,7 @@ bch2_hash_lookup_in_snapshot(struct btree_trans *trans, for_each_btree_key_upto_norestart(trans, *iter, desc.btree_id, SPOS(inum.inum, desc.hash_key(info, key), snapshot), POS(inum.inum, U64_MAX), - BTREE_ITER_SLOTS|flags, k, ret) { + BTREE_ITER_slots|flags, k, ret) { if (is_visible_key(desc, inum, k)) { if (!desc.cmp_key(k, key)) return k; @@ -195,7 +186,7 @@ bch2_hash_lookup(struct btree_trans *trans, const struct bch_hash_desc desc, const struct bch_hash_info *info, subvol_inum inum, const void *key, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { u32 snapshot; int ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); @@ -223,7 +214,7 @@ bch2_hash_hole(struct btree_trans *trans, for_each_btree_key_upto_norestart(trans, *iter, desc.btree_id, SPOS(inum.inum, desc.hash_key(info, key), snapshot), POS(inum.inum, U64_MAX), - BTREE_ITER_SLOTS|BTREE_ITER_INTENT, k, ret) + BTREE_ITER_slots|BTREE_ITER_intent, k, ret) if (!is_visible_key(desc, inum, k)) return 0; bch2_trans_iter_exit(trans, iter); @@ -245,7 +236,7 @@ int bch2_hash_needs_whiteout(struct btree_trans *trans, bch2_btree_iter_advance(&iter); - for_each_btree_key_continue_norestart(iter, BTREE_ITER_SLOTS, k, ret) { + for_each_btree_key_continue_norestart(iter, BTREE_ITER_slots, k, ret) { if (k.k->type != desc.key_type && k.k->type != KEY_TYPE_hash_whiteout) break; @@ -267,8 +258,7 @@ int bch2_hash_set_in_snapshot(struct btree_trans *trans, const struct bch_hash_info *info, subvol_inum inum, u32 snapshot, struct bkey_i *insert, - bch_str_hash_flags_t str_hash_flags, - int update_flags) + enum btree_iter_update_trigger_flags flags) { struct btree_iter iter, slot = { NULL }; struct bkey_s_c k; @@ -280,7 +270,7 @@ int bch2_hash_set_in_snapshot(struct btree_trans *trans, desc.hash_bkey(info, bkey_i_to_s_c(insert)), snapshot), POS(insert->k.p.inode, U64_MAX), - BTREE_ITER_SLOTS|BTREE_ITER_INTENT, k, ret) { + BTREE_ITER_slots|BTREE_ITER_intent, k, ret) { if (is_visible_key(desc, inum, k)) { if (!desc.cmp_bkey(k, bkey_i_to_s_c(insert))) goto found; @@ -289,8 +279,7 @@ int bch2_hash_set_in_snapshot(struct btree_trans *trans, continue; } - if (!slot.path && - !(str_hash_flags & BCH_HASH_SET_MUST_REPLACE)) + if (!slot.path && !(flags & STR_HASH_must_replace)) bch2_trans_copy_iter(&slot, &iter); if (k.k->type != KEY_TYPE_hash_whiteout) @@ -308,16 +297,16 @@ found: found = true; not_found: - if (!found && (str_hash_flags & BCH_HASH_SET_MUST_REPLACE)) { + if (!found && (flags & STR_HASH_must_replace)) { ret = -BCH_ERR_ENOENT_str_hash_set_must_replace; - } else if (found && (str_hash_flags & BCH_HASH_SET_MUST_CREATE)) { + } else if (found && (flags & STR_HASH_must_create)) { ret = -EEXIST; } else { if (!found && slot.path) swap(iter, slot); insert->k.p = iter.pos; - ret = bch2_trans_update(trans, &iter, insert, update_flags); + ret = bch2_trans_update(trans, &iter, insert, flags); } goto out; @@ -329,14 +318,14 @@ int bch2_hash_set(struct btree_trans *trans, const struct bch_hash_info *info, subvol_inum inum, struct bkey_i *insert, - bch_str_hash_flags_t str_hash_flags) + enum btree_iter_update_trigger_flags flags) { insert->k.p.inode = inum.inum; u32 snapshot; return bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot) ?: bch2_hash_set_in_snapshot(trans, desc, info, inum, - snapshot, insert, str_hash_flags, 0); + snapshot, insert, flags); } static __always_inline @@ -344,7 +333,7 @@ int bch2_hash_delete_at(struct btree_trans *trans, const struct bch_hash_desc desc, const struct bch_hash_info *info, struct btree_iter *iter, - unsigned update_flags) + enum btree_iter_update_trigger_flags flags) { struct bkey_i *delete; int ret; @@ -362,7 +351,7 @@ int bch2_hash_delete_at(struct btree_trans *trans, delete->k.p = iter->pos; delete->k.type = ret ? KEY_TYPE_hash_whiteout : KEY_TYPE_deleted; - return bch2_trans_update(trans, iter, delete, update_flags); + return bch2_trans_update(trans, iter, delete, flags); } static __always_inline @@ -373,7 +362,7 @@ int bch2_hash_delete(struct btree_trans *trans, { struct btree_iter iter; struct bkey_s_c k = bch2_hash_lookup(trans, &iter, desc, info, inum, key, - BTREE_ITER_INTENT); + BTREE_ITER_intent); int ret = bkey_err(k) ?: bch2_hash_delete_at(trans, desc, info, &iter, 0); bch2_trans_iter_exit(trans, &iter); diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index 88a79c823276..33292dae4d50 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -162,7 +162,7 @@ int bch2_check_subvols(struct bch_fs *c) { int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, - BTREE_ID_subvolumes, POS_MIN, BTREE_ITER_PREFETCH, k, + BTREE_ID_subvolumes, POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_subvol(trans, &iter, k))); bch_err_fn(c, ret); @@ -198,7 +198,7 @@ int bch2_check_subvol_children(struct bch_fs *c) { int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, - BTREE_ID_subvolume_children, POS_MIN, BTREE_ITER_PREFETCH, k, + BTREE_ID_subvolume_children, POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, check_subvol_child(trans, &iter, k))); bch_err_fn(c, ret); @@ -247,7 +247,7 @@ int bch2_subvolume_trigger(struct btree_trans *trans, struct bkey_s_c old, struct bkey_s new, unsigned flags) { - if (flags & BTREE_TRIGGER_TRANSACTIONAL) { + if (flags & BTREE_TRIGGER_transactional) { struct bpos children_pos_old = subvolume_children_pos(old); struct bpos children_pos_new = subvolume_children_pos(new.s_c); @@ -333,7 +333,7 @@ int bch2_subvolume_get_snapshot(struct btree_trans *trans, u32 subvolid, subvol = bch2_bkey_get_iter_typed(trans, &iter, BTREE_ID_subvolumes, POS(0, subvolid), - BTREE_ITER_CACHED|BTREE_ITER_WITH_UPDATES, + BTREE_ITER_cached|BTREE_ITER_with_updates, subvolume); ret = bkey_err(subvol); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, @@ -383,9 +383,9 @@ static int bch2_subvolumes_reparent(struct btree_trans *trans, u32 subvolid_to_d return lockrestart_do(trans, bch2_subvolume_get(trans, subvolid_to_delete, true, - BTREE_ITER_CACHED, &s)) ?: + BTREE_ITER_cached, &s)) ?: for_each_btree_key_commit(trans, iter, - BTREE_ID_subvolumes, POS_MIN, BTREE_ITER_PREFETCH, k, + BTREE_ID_subvolumes, POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_subvolume_reparent(trans, &iter, k, subvolid_to_delete, le32_to_cpu(s.creation_parent))); @@ -404,7 +404,7 @@ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid) subvol = bch2_bkey_get_iter_typed(trans, &iter, BTREE_ID_subvolumes, POS(0, subvolid), - BTREE_ITER_CACHED|BTREE_ITER_INTENT, + BTREE_ITER_cached|BTREE_ITER_intent, subvolume); ret = bkey_err(subvol); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, @@ -505,7 +505,7 @@ int bch2_subvolume_unlink(struct btree_trans *trans, u32 subvolid) n = bch2_bkey_get_mut_typed(trans, &iter, BTREE_ID_subvolumes, POS(0, subvolid), - BTREE_ITER_CACHED, subvolume); + BTREE_ITER_cached, subvolume); ret = PTR_ERR_OR_ZERO(n); if (unlikely(ret)) { bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, @@ -547,7 +547,7 @@ int bch2_subvolume_create(struct btree_trans *trans, u64 inode, src_subvol = bch2_bkey_get_mut_typed(trans, &src_iter, BTREE_ID_subvolumes, POS(0, src_subvolid), - BTREE_ITER_CACHED, subvolume); + BTREE_ITER_cached, subvolume); ret = PTR_ERR_OR_ZERO(src_subvol); if (unlikely(ret)) { bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), c, diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 9073915f22a1..c16131391300 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -1600,17 +1600,17 @@ static int bch2_dev_remove_alloc(struct bch_fs *c, struct bch_dev *ca) * with bch2_do_invalidates() and bch2_do_discards() */ ret = bch2_btree_delete_range(c, BTREE_ID_lru, start, end, - BTREE_TRIGGER_NORUN, NULL) ?: + BTREE_TRIGGER_norun, NULL) ?: bch2_btree_delete_range(c, BTREE_ID_need_discard, start, end, - BTREE_TRIGGER_NORUN, NULL) ?: + BTREE_TRIGGER_norun, NULL) ?: bch2_btree_delete_range(c, BTREE_ID_freespace, start, end, - BTREE_TRIGGER_NORUN, NULL) ?: + BTREE_TRIGGER_norun, NULL) ?: bch2_btree_delete_range(c, BTREE_ID_backpointers, start, end, - BTREE_TRIGGER_NORUN, NULL) ?: + BTREE_TRIGGER_norun, NULL) ?: bch2_btree_delete_range(c, BTREE_ID_alloc, start, end, - BTREE_TRIGGER_NORUN, NULL) ?: + BTREE_TRIGGER_norun, NULL) ?: bch2_btree_delete_range(c, BTREE_ID_bucket_gens, start, end, - BTREE_TRIGGER_NORUN, NULL); + BTREE_TRIGGER_norun, NULL); bch_err_msg(c, ret, "removing dev alloc info"); return ret; } @@ -1822,7 +1822,7 @@ have_slot: bch2_dev_usage_journal_reserve(c); - ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_TRANSACTIONAL); + ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_transactional); bch_err_msg(ca, ret, "marking new superblock"); if (ret) goto err_late; @@ -1887,7 +1887,7 @@ int bch2_dev_online(struct bch_fs *c, const char *path) ca = bch_dev_locked(c, dev_idx); - ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_TRANSACTIONAL); + ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_transactional); bch_err_msg(c, ret, "bringing %s online: error from bch2_trans_mark_dev_sb", path); if (ret) goto err; @@ -1980,7 +1980,7 @@ int bch2_dev_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) if (ret) goto err; - ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_TRANSACTIONAL); + ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_transactional); if (ret) goto err; diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 4011ab4e710b..607c930831bf 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -274,7 +274,7 @@ static int bch2_compression_stats_to_text(struct printbuf *out, struct bch_fs *c continue; ret = for_each_btree_key(trans, iter, id, POS_MIN, - BTREE_ITER_ALL_SNAPSHOTS, k, ({ + BTREE_ITER_all_snapshots, k, ({ struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); struct bch_extent_crc_unpacked crc; const union bch_extent_entry *entry; diff --git a/fs/bcachefs/tests.c b/fs/bcachefs/tests.c index bfec656f94c0..68104b2056d9 100644 --- a/fs/bcachefs/tests.c +++ b/fs/bcachefs/tests.c @@ -40,7 +40,7 @@ static int test_delete(struct bch_fs *c, u64 nr) k.k.p.snapshot = U32_MAX; bch2_trans_iter_init(trans, &iter, BTREE_ID_xattrs, k.k.p, - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = commit_do(trans, NULL, NULL, 0, bch2_btree_iter_traverse(&iter) ?: @@ -81,7 +81,7 @@ static int test_delete_written(struct bch_fs *c, u64 nr) k.k.p.snapshot = U32_MAX; bch2_trans_iter_init(trans, &iter, BTREE_ID_xattrs, k.k.p, - BTREE_ITER_INTENT); + BTREE_ITER_intent); ret = commit_do(trans, NULL, NULL, 0, bch2_btree_iter_traverse(&iter) ?: @@ -261,7 +261,7 @@ static int test_iterate_slots(struct bch_fs *c, u64 nr) ret = bch2_trans_run(c, for_each_btree_key_upto(trans, iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), - BTREE_ITER_SLOTS, k, ({ + BTREE_ITER_slots, k, ({ if (i >= nr * 2) break; @@ -322,7 +322,7 @@ static int test_iterate_slots_extents(struct bch_fs *c, u64 nr) ret = bch2_trans_run(c, for_each_btree_key_upto(trans, iter, BTREE_ID_extents, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), - BTREE_ITER_SLOTS, k, ({ + BTREE_ITER_slots, k, ({ if (i == nr) break; BUG_ON(bkey_deleted(k.k) != !(i % 16)); @@ -452,7 +452,7 @@ static int insert_test_overlapping_extent(struct bch_fs *c, u64 inum, u64 start, ret = bch2_trans_do(c, NULL, NULL, 0, bch2_btree_insert_nonextent(trans, BTREE_ID_extents, &k.k_i, - BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE)); + BTREE_UPDATE_internal_snapshot_node)); bch_err_fn(c, ret); return ret; } @@ -671,7 +671,7 @@ static int __do_delete(struct btree_trans *trans, struct bpos pos) int ret = 0; bch2_trans_iter_init(trans, &iter, BTREE_ID_xattrs, pos, - BTREE_ITER_INTENT); + BTREE_ITER_intent); k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)); ret = bkey_err(k); if (ret) @@ -714,7 +714,7 @@ static int seq_insert(struct bch_fs *c, u64 nr) return bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), - BTREE_ITER_SLOTS|BTREE_ITER_INTENT, k, + BTREE_ITER_slots|BTREE_ITER_intent, k, NULL, NULL, 0, ({ if (iter.pos.offset >= nr) break; @@ -737,7 +737,7 @@ static int seq_overwrite(struct bch_fs *c, u64 nr) return bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), - BTREE_ITER_INTENT, k, + BTREE_ITER_intent, k, NULL, NULL, 0, ({ struct bkey_i_cookie u; diff --git a/fs/bcachefs/xattr.c b/fs/bcachefs/xattr.c index d787f1d4cc78..786ccd7a7662 100644 --- a/fs/bcachefs/xattr.c +++ b/fs/bcachefs/xattr.c @@ -173,7 +173,7 @@ int bch2_xattr_set(struct btree_trans *trans, subvol_inum inum, int ret; ret = bch2_subvol_is_ro_trans(trans, inum.subvol) ?: - bch2_inode_peek(trans, &inode_iter, inode_u, inum, BTREE_ITER_INTENT); + bch2_inode_peek(trans, &inode_iter, inode_u, inum, BTREE_ITER_intent); if (ret) return ret; @@ -208,8 +208,8 @@ int bch2_xattr_set(struct btree_trans *trans, subvol_inum inum, ret = bch2_hash_set(trans, bch2_xattr_hash_desc, hash_info, inum, &xattr->k_i, - (flags & XATTR_CREATE ? BCH_HASH_SET_MUST_CREATE : 0)| - (flags & XATTR_REPLACE ? BCH_HASH_SET_MUST_REPLACE : 0)); + (flags & XATTR_CREATE ? STR_HASH_must_create : 0)| + (flags & XATTR_REPLACE ? STR_HASH_must_replace : 0)); } else { struct xattr_search_key search = X_SEARCH(type, name, strlen(name)); -- cgit v1.2.3-59-g8ed1b From 65bd442397274347e721a89c2c4882a392bae982 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 19:13:45 -0400 Subject: bcachefs: bch2_btree_insert_trans() no longer specifies BTREE_ITER_cached Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update.c b/fs/bcachefs/btree_update.c index da4b17704296..ee10c1f845b8 100644 --- a/fs/bcachefs/btree_update.c +++ b/fs/bcachefs/btree_update.c @@ -640,13 +640,10 @@ int bch2_btree_insert_trans(struct btree_trans *trans, enum btree_id id, struct bkey_i *k, enum btree_iter_update_trigger_flags flags) { struct btree_iter iter; - int ret; - bch2_trans_iter_init(trans, &iter, id, bkey_start_pos(&k->k), - BTREE_ITER_cached| - BTREE_ITER_intent); - ret = bch2_btree_iter_traverse(&iter) ?: - bch2_trans_update(trans, &iter, k, flags); + BTREE_ITER_intent|flags); + int ret = bch2_btree_iter_traverse(&iter) ?: + bch2_trans_update(trans, &iter, k, flags); bch2_trans_iter_exit(trans, &iter); return ret; } -- cgit v1.2.3-59-g8ed1b From 0c0cbfdb84725e9933a24ecf47c61bdeeda06ba2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 13:18:22 -0400 Subject: bcachefs: bch2_dir_emit() - drop_locks_do() conversion Add a new helper that calls dir_emit() and updates ctx->pos on success; this lets us convert bch2_readdir() to drop_locks_do(). Signed-off-by: Kent Overstreet --- fs/bcachefs/dirent.c | 46 ++++++++++++++++++++++++++-------------------- 1 file changed, 26 insertions(+), 20 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c index ed256c3aa8a5..dd9de8e5a28a 100644 --- a/fs/bcachefs/dirent.c +++ b/fs/bcachefs/dirent.c @@ -528,16 +528,26 @@ int bch2_empty_dir_trans(struct btree_trans *trans, subvol_inum dir) bch2_empty_dir_snapshot(trans, dir.inum, dir.subvol, snapshot); } +static int bch2_dir_emit(struct dir_context *ctx, struct bkey_s_c_dirent d, subvol_inum target) +{ + struct qstr name = bch2_dirent_get_name(d); + bool ret = dir_emit(ctx, name.name, + name.len, + target.inum, + vfs_d_type(d.v->d_type)); + if (ret) + ctx->pos = d.k->p.offset + 1; + return ret; +} + int bch2_readdir(struct bch_fs *c, subvol_inum inum, struct dir_context *ctx) { struct btree_trans *trans = bch2_trans_get(c); struct btree_iter iter; struct bkey_s_c k; - struct bkey_s_c_dirent dirent; subvol_inum target; u32 snapshot; struct bkey_buf sk; - struct qstr name; int ret; bch2_bkey_buf_init(&sk); @@ -554,7 +564,9 @@ retry: if (k.k->type != KEY_TYPE_dirent) continue; - dirent = bkey_s_c_to_dirent(k); + /* dir_emit() can fault and block: */ + bch2_bkey_buf_reassemble(&sk, c, k); + struct bkey_s_c_dirent dirent = bkey_i_to_s_c_dirent(sk.k); ret = bch2_dirent_read_target(trans, inum, dirent, &target); if (ret < 0) @@ -562,28 +574,22 @@ retry: if (ret) continue; - /* dir_emit() can fault and block: */ - bch2_bkey_buf_reassemble(&sk, c, k); - dirent = bkey_i_to_s_c_dirent(sk.k); - bch2_trans_unlock(trans); - - name = bch2_dirent_get_name(dirent); - - ctx->pos = dirent.k->p.offset; - if (!dir_emit(ctx, name.name, - name.len, - target.inum, - vfs_d_type(dirent.v->d_type))) - break; - ctx->pos = dirent.k->p.offset + 1; - /* * read_target looks up subvolumes, we can overflow paths if the * directory has many subvolumes in it + * + * XXX: btree_trans_too_many_iters() is something we'd like to + * get rid of, and there's no good reason to be using it here + * except that we don't yet have a for_each_btree_key() helper + * that does subvolume_get_snapshot(). */ - ret = btree_trans_too_many_iters(trans); - if (ret) + ret = drop_locks_do(trans, + bch2_dir_emit(ctx, dirent, target)) ?: + btree_trans_too_many_iters(trans); + if (ret) { + ret = ret < 0 ? ret : 0; break; + } } bch2_trans_iter_exit(trans, &iter); err: -- cgit v1.2.3-59-g8ed1b From 923ed0ae5ebd0bb0dbc64a88348fa1835a3644ee Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 19:45:41 -0400 Subject: bcachefs: bch2_trans_relock_fail() - factor out slowpath Factor out slowpath into a separate helper Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_locking.c | 98 ++++++++++++++++++++++++--------------------- 1 file changed, 52 insertions(+), 46 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index 8d1c4f78db5e..e6adb2df3a57 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -723,51 +723,51 @@ void bch2_trans_downgrade(struct btree_trans *trans) bch2_btree_path_downgrade(trans, path); } -int bch2_trans_relock(struct btree_trans *trans) +static inline void __bch2_trans_unlock(struct btree_trans *trans) { struct btree_path *path; unsigned i; - if (unlikely(trans->restarted)) - return -((int) trans->restarted); + trans_for_each_path(trans, path, i) + __bch2_btree_path_unlock(trans, path); +} - trans_for_each_path(trans, path, i) { - struct get_locks_fail f; +static noinline __cold int bch2_trans_relock_fail(struct btree_trans *trans, struct btree_path *path, + struct get_locks_fail *f, bool trace) +{ + if (!trace) + goto out; - if (path->should_be_locked && - !btree_path_get_locks(trans, path, false, &f)) { - if (trace_trans_restart_relock_enabled()) { - struct printbuf buf = PRINTBUF; - - bch2_bpos_to_text(&buf, path->pos); - prt_printf(&buf, " l=%u seq=%u node seq=", - f.l, path->l[f.l].lock_seq); - if (IS_ERR_OR_NULL(f.b)) { - prt_str(&buf, bch2_err_str(PTR_ERR(f.b))); - } else { - prt_printf(&buf, "%u", f.b->c.lock.seq); - - struct six_lock_count c = - bch2_btree_node_lock_counts(trans, NULL, &f.b->c, f.l); - prt_printf(&buf, " self locked %u.%u.%u", c.n[0], c.n[1], c.n[2]); - - c = six_lock_counts(&f.b->c.lock); - prt_printf(&buf, " total locked %u.%u.%u", c.n[0], c.n[1], c.n[2]); - } + if (trace_trans_restart_relock_enabled()) { + struct printbuf buf = PRINTBUF; - trace_trans_restart_relock(trans, _RET_IP_, buf.buf); - printbuf_exit(&buf); - } + bch2_bpos_to_text(&buf, path->pos); + prt_printf(&buf, " l=%u seq=%u node seq=", f->l, path->l[f->l].lock_seq); + if (IS_ERR_OR_NULL(f->b)) { + prt_str(&buf, bch2_err_str(PTR_ERR(f->b))); + } else { + prt_printf(&buf, "%u", f->b->c.lock.seq); - count_event(trans->c, trans_restart_relock); - return btree_trans_restart(trans, BCH_ERR_transaction_restart_relock); + struct six_lock_count c = + bch2_btree_node_lock_counts(trans, NULL, &f->b->c, f->l); + prt_printf(&buf, " self locked %u.%u.%u", c.n[0], c.n[1], c.n[2]); + + c = six_lock_counts(&f->b->c.lock); + prt_printf(&buf, " total locked %u.%u.%u", c.n[0], c.n[1], c.n[2]); } + + trace_trans_restart_relock(trans, _RET_IP_, buf.buf); + printbuf_exit(&buf); } - return 0; + count_event(trans->c, trans_restart_relock); +out: + __bch2_trans_unlock(trans); + bch2_trans_verify_locks(trans); + return btree_trans_restart(trans, BCH_ERR_transaction_restart_relock); } -int bch2_trans_relock_notrace(struct btree_trans *trans) +static inline int __bch2_trans_relock(struct btree_trans *trans, bool trace) { struct btree_path *path; unsigned i; @@ -775,30 +775,36 @@ int bch2_trans_relock_notrace(struct btree_trans *trans) if (unlikely(trans->restarted)) return -((int) trans->restarted); - trans_for_each_path(trans, path, i) + trans_for_each_path(trans, path, i) { + struct get_locks_fail f; + if (path->should_be_locked && - !bch2_btree_path_relock_norestart(trans, path)) { - return btree_trans_restart(trans, BCH_ERR_transaction_restart_relock); - } + !btree_path_get_locks(trans, path, false, &f)) + return bch2_trans_relock_fail(trans, path, &f, trace); + } + + bch2_trans_verify_locks(trans); return 0; } -void bch2_trans_unlock_noassert(struct btree_trans *trans) +int bch2_trans_relock(struct btree_trans *trans) { - struct btree_path *path; - unsigned i; + return __bch2_trans_relock(trans, true); +} - trans_for_each_path(trans, path, i) - __bch2_btree_path_unlock(trans, path); +int bch2_trans_relock_notrace(struct btree_trans *trans) +{ + return __bch2_trans_relock(trans, false); } -void bch2_trans_unlock(struct btree_trans *trans) +void bch2_trans_unlock_noassert(struct btree_trans *trans) { - struct btree_path *path; - unsigned i; + __bch2_trans_unlock(trans); +} - trans_for_each_path(trans, path, i) - __bch2_btree_path_unlock(trans, path); +void bch2_trans_unlock(struct btree_trans *trans) +{ + __bch2_trans_unlock(trans); } void bch2_trans_unlock_long(struct btree_trans *trans) -- cgit v1.2.3-59-g8ed1b From d155272b6e58a4b372982de777fde7dc503a8f96 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 21:18:35 -0400 Subject: bcachefs: bucket_valid() cut out a branch from doing it the obvious way Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.h | 3 +-- fs/bcachefs/bcachefs.h | 1 + fs/bcachefs/buckets.h | 14 +++++++++----- fs/bcachefs/extents.c | 4 +--- fs/bcachefs/sb-members.h | 2 ++ fs/bcachefs/sb-members_types.h | 21 +++++++++++++++++++++ fs/bcachefs/super_types.h | 15 --------------- 7 files changed, 35 insertions(+), 25 deletions(-) create mode 100644 fs/bcachefs/sb-members_types.h (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 2790e516383d..2bcc648081c3 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -21,8 +21,7 @@ static inline bool bch2_dev_bucket_exists(struct bch_fs *c, struct bpos pos) return false; ca = bch_dev_bkey_exists(c, pos.inode); - return pos.offset >= ca->mi.first_bucket && - pos.offset < ca->mi.nbuckets; + return bucket_valid(ca, pos.offset); } static inline u64 bucket_to_u64(struct bpos bucket) diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 6366a6a90944..3dad47533491 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -470,6 +470,7 @@ enum bch_time_stats { #include "quota_types.h" #include "rebalance_types.h" #include "replicas_types.h" +#include "sb-members_types.h" #include "subvolume_types.h" #include "super_types.h" #include "thread_with_file_types.h" diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index a84e14468276..97b4bf8961e2 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -12,7 +12,7 @@ #include "extents.h" #include "sb-members.h" -static inline size_t sector_to_bucket(const struct bch_dev *ca, sector_t s) +static inline u64 sector_to_bucket(const struct bch_dev *ca, sector_t s) { return div_u64(s, ca->mi.bucket_size); } @@ -30,12 +30,16 @@ static inline sector_t bucket_remainder(const struct bch_dev *ca, sector_t s) return remainder; } -static inline size_t sector_to_bucket_and_offset(const struct bch_dev *ca, sector_t s, - u32 *offset) +static inline u64 sector_to_bucket_and_offset(const struct bch_dev *ca, sector_t s, u32 *offset) { return div_u64_rem(s, ca->mi.bucket_size, offset); } +static inline bool bucket_valid(const struct bch_dev *ca, u64 b) +{ + return b - ca->mi.first_bucket < ca->mi.nbuckets_minus_first; +} + #define for_each_bucket(_b, _buckets) \ for (_b = (_buckets)->b + (_buckets)->first_bucket; \ _b < (_buckets)->b + (_buckets)->nbuckets; _b++) @@ -94,7 +98,7 @@ static inline struct bucket *gc_bucket(struct bch_dev *ca, size_t b) { struct bucket_array *buckets = gc_bucket_array(ca); - BUG_ON(b < buckets->first_bucket || b >= buckets->nbuckets); + BUG_ON(!bucket_valid(ca, b)); return buckets->b + b; } @@ -111,7 +115,7 @@ static inline u8 *bucket_gen(struct bch_dev *ca, size_t b) { struct bucket_gens *gens = bucket_gens(ca); - BUG_ON(b < gens->first_bucket || b >= gens->nbuckets); + BUG_ON(!bucket_valid(ca, b)); return gens->b + b; } diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 1a331e539204..e9db8f6d6bc1 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -998,9 +998,7 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struc prt_str(out, " cached"); if (ptr->unwritten) prt_str(out, " unwritten"); - if (b >= ca->mi.first_bucket && - b < ca->mi.nbuckets && - ptr_stale(ca, ptr)) + if (bucket_valid(ca, b) && ptr_stale(ca, ptr)) prt_printf(out, " stale"); } } diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index 5bf27d30ca29..fa55b9666241 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -210,6 +210,8 @@ static inline struct bch_member_cpu bch2_mi_to_cpu(struct bch_member *mi) { return (struct bch_member_cpu) { .nbuckets = le64_to_cpu(mi->nbuckets), + .nbuckets_minus_first = le64_to_cpu(mi->nbuckets) - + le16_to_cpu(mi->first_bucket), .first_bucket = le16_to_cpu(mi->first_bucket), .bucket_size = le16_to_cpu(mi->bucket_size), .group = BCH_MEMBER_GROUP(mi), diff --git a/fs/bcachefs/sb-members_types.h b/fs/bcachefs/sb-members_types.h new file mode 100644 index 000000000000..c0eda888fe39 --- /dev/null +++ b/fs/bcachefs/sb-members_types.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _BCACHEFS_SB_MEMBERS_TYPES_H +#define _BCACHEFS_SB_MEMBERS_TYPES_H + +struct bch_member_cpu { + u64 nbuckets; /* device size */ + u64 nbuckets_minus_first; + u16 first_bucket; /* index of first bucket used */ + u16 bucket_size; /* sectors */ + u16 group; + u8 state; + u8 discard; + u8 data_allowed; + u8 durability; + u8 freespace_initialized; + u8 valid; + u8 btree_bitmap_shift; + u64 btree_allocated_bitmap; +}; + +#endif /* _BCACHEFS_SB_MEMBERS_H */ diff --git a/fs/bcachefs/super_types.h b/fs/bcachefs/super_types.h index 11bcef170c2c..368a63d938cf 100644 --- a/fs/bcachefs/super_types.h +++ b/fs/bcachefs/super_types.h @@ -26,19 +26,4 @@ struct bch_devs_list { u8 data[BCH_BKEY_PTRS_MAX]; }; -struct bch_member_cpu { - u64 nbuckets; /* device size */ - u16 first_bucket; /* index of first bucket used */ - u16 bucket_size; /* sectors */ - u16 group; - u8 state; - u8 discard; - u8 data_allowed; - u8 durability; - u8 freespace_initialized; - u8 valid; - u8 btree_bitmap_shift; - u64 btree_allocated_bitmap; -}; - #endif /* _BCACHEFS_SUPER_TYPES_H */ -- cgit v1.2.3-59-g8ed1b From 2f724563fcd76166b9922c506078a4afa4e3a90a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 23:31:55 -0400 Subject: bcachefs: member helper cleanups Some renaming for better consistency bch2_member_exists -> bch2_member_alive bch2_dev_exists -> bch2_member_exists bch2_dev_exsits2 -> bch2_dev_exists bch_dev_locked -> bch2_dev_locked bch_dev_bkey_exists -> bch2_dev_bkey_exists new helper - bch2_dev_safe Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 32 ++++++++++++++++---------------- fs/bcachefs/alloc_background.h | 4 ++-- fs/bcachefs/alloc_foreground.c | 14 +++++++------- fs/bcachefs/alloc_foreground.h | 2 +- fs/bcachefs/backpointers.c | 10 +++++----- fs/bcachefs/backpointers.h | 4 ++-- fs/bcachefs/bcachefs.h | 5 ----- fs/bcachefs/btree_gc.c | 16 ++++++++-------- fs/bcachefs/btree_io.c | 16 ++++++++-------- fs/bcachefs/buckets.c | 4 ++-- fs/bcachefs/buckets.h | 4 ++-- fs/bcachefs/data_update.c | 6 +++--- fs/bcachefs/debug.c | 4 ++-- fs/bcachefs/disk_groups.c | 4 ++-- fs/bcachefs/ec.c | 10 +++++----- fs/bcachefs/extents.c | 29 +++++++++++++++-------------- fs/bcachefs/io_read.c | 8 ++++---- fs/bcachefs/io_write.c | 12 ++++++------ fs/bcachefs/journal_io.c | 10 +++++----- fs/bcachefs/move.c | 2 +- fs/bcachefs/recovery.c | 2 +- fs/bcachefs/replicas.c | 6 +++--- fs/bcachefs/sb-members.c | 4 ++-- fs/bcachefs/sb-members.h | 32 ++++++++++++++++++++++---------- fs/bcachefs/super-io.c | 2 +- fs/bcachefs/super.c | 17 ++++++++--------- 26 files changed, 133 insertions(+), 126 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index d3b86ce0fae6..2f91ff67453f 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -576,10 +576,10 @@ int bch2_alloc_read(struct bch_fs *c) * Not a fsck error because this is checked/repaired by * bch2_check_alloc_key() which runs later: */ - if (!bch2_dev_exists2(c, k.k->p.inode)) + if (!bch2_dev_exists(c, k.k->p.inode)) continue; - struct bch_dev *ca = bch_dev_bkey_exists(c, k.k->p.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, k.k->p.inode); for (u64 b = max_t(u64, ca->mi.first_bucket, start); b < min_t(u64, ca->mi.nbuckets, end); @@ -597,7 +597,7 @@ int bch2_alloc_read(struct bch_fs *c) if (!bch2_dev_bucket_exists(c, k.k->p)) continue; - struct bch_dev *ca = bch_dev_bkey_exists(c, k.k->p.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, k.k->p.inode); struct bch_alloc_v4 a; *bucket_gen(ca, k.k->p.offset) = bch2_alloc_to_v4(k, &a)->gen; @@ -620,7 +620,7 @@ static int bch2_bucket_do_index(struct btree_trans *trans, bool set) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, alloc_k.k->p.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, alloc_k.k->p.inode); struct btree_iter iter; struct bkey_s_c old; struct bkey_i *k; @@ -733,7 +733,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, "alloc key for invalid device or bucket")) return -EIO; - struct bch_dev *ca = bch_dev_bkey_exists(c, new.k->p.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, new.k->p.inode); struct bch_alloc_v4 old_a_convert; const struct bch_alloc_v4 *old_a = bch2_alloc_to_v4(old, &old_a_convert); @@ -781,7 +781,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, } new_a->fragmentation_lru = alloc_lru_idx_fragmentation(*new_a, - bch_dev_bkey_exists(c, new.k->p.inode)); + bch2_dev_bkey_exists(c, new.k->p.inode)); if (old_a->fragmentation_lru != new_a->fragmentation_lru) { ret = bch2_lru_change(trans, BCH_LRU_FRAGMENTATION_START, @@ -955,8 +955,8 @@ static bool next_bucket(struct bch_fs *c, struct bpos *bucket) if (bch2_dev_bucket_exists(c, *bucket)) return true; - if (bch2_dev_exists2(c, bucket->inode)) { - ca = bch_dev_bkey_exists(c, bucket->inode); + if (bch2_dev_exists(c, bucket->inode)) { + ca = bch2_dev_bkey_exists(c, bucket->inode); if (bucket->offset < ca->mi.first_bucket) { bucket->offset = ca->mi.first_bucket; @@ -997,7 +997,7 @@ again: } if (!bch2_dev_bucket_exists(c, k.k->p)) { - struct bch_dev *ca = bch_dev_bkey_exists(c, bucket.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, bucket.inode); bch2_key_resize(hole, ca->mi.nbuckets - bucket.offset); } @@ -1030,7 +1030,7 @@ int bch2_check_alloc_key(struct btree_trans *trans, alloc_k.k->p.inode, alloc_k.k->p.offset)) return bch2_btree_delete_at(trans, alloc_iter, 0); - ca = bch_dev_bkey_exists(c, alloc_k.k->p.inode); + ca = bch2_dev_bkey_exists(c, alloc_k.k->p.inode); if (!ca->mi.freespace_initialized) return 0; @@ -1149,7 +1149,7 @@ int bch2_check_alloc_hole_freespace(struct btree_trans *trans, struct printbuf buf = PRINTBUF; int ret; - ca = bch_dev_bkey_exists(c, start.inode); + ca = bch2_dev_bkey_exists(c, start.inode); if (!ca->mi.freespace_initialized) return 0; @@ -1339,7 +1339,7 @@ int bch2_check_bucket_gens_key(struct btree_trans *trans, bkey_reassemble(&g.k_i, k); /* if no bch_dev, skip out whether we repair or not */ - dev_exists = bch2_dev_exists2(c, k.k->p.inode); + dev_exists = bch2_dev_exists(c, k.k->p.inode); if (!dev_exists) { if (fsck_err_on(!dev_exists, c, bucket_gens_to_invalid_dev, @@ -1350,7 +1350,7 @@ int bch2_check_bucket_gens_key(struct btree_trans *trans, goto out; } - ca = bch_dev_bkey_exists(c, k.k->p.inode); + ca = bch2_dev_bkey_exists(c, k.k->p.inode); if (fsck_err_on(end <= ca->mi.first_bucket || start >= ca->mi.nbuckets, c, bucket_gens_to_invalid_buckets, @@ -1669,7 +1669,7 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, bool discard_locked = false; int ret = 0; - ca = bch_dev_bkey_exists(c, pos.inode); + ca = bch2_dev_bkey_exists(c, pos.inode); if (!percpu_ref_tryget(&ca->io_ref)) { bch2_btree_iter_set_pos(need_discard_iter, POS(pos.inode + 1, 0)); @@ -1852,7 +1852,7 @@ static void bch2_do_discards_fast_work(struct work_struct *work) if (i->snapshot) continue; - ca = bch_dev_bkey_exists(c, i->inode); + ca = bch2_dev_bkey_exists(c, i->inode); if (!percpu_ref_tryget(&ca->io_ref)) { darray_remove_item(&c->discard_buckets_in_flight, i); @@ -1893,7 +1893,7 @@ static void bch2_do_discards_fast_work(struct work_struct *work) static void bch2_discard_one_bucket_fast(struct bch_fs *c, struct bpos bucket) { - struct bch_dev *ca = bch_dev_bkey_exists(c, bucket.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, bucket.inode); if (!percpu_ref_is_dying(&ca->io_ref) && !discard_in_flight_add(c, bucket) && diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 2bcc648081c3..44db71ccc953 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -17,10 +17,10 @@ static inline bool bch2_dev_bucket_exists(struct bch_fs *c, struct bpos pos) { struct bch_dev *ca; - if (!bch2_dev_exists2(c, pos.inode)) + if (!bch2_dev_exists(c, pos.inode)) return false; - ca = bch_dev_bkey_exists(c, pos.inode); + ca = bch2_dev_bkey_exists(c, pos.inode); return bucket_valid(ca, pos.offset); } diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index b68e1fd782f3..df4439a38df5 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -100,7 +100,7 @@ static void bch2_open_bucket_hash_remove(struct bch_fs *c, struct open_bucket *o void __bch2_open_bucket_put(struct bch_fs *c, struct open_bucket *ob) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); if (ob->ec) { ec_stripe_new_put(c, ob->ec, STRIPE_REF_io); @@ -679,7 +679,7 @@ static int add_new_bucket(struct bch_fs *c, struct open_bucket *ob) { unsigned durability = - bch_dev_bkey_exists(c, ob->dev)->mi.durability; + bch2_dev_bkey_exists(c, ob->dev)->mi.durability; BUG_ON(*nr_effective >= nr_replicas); @@ -836,7 +836,7 @@ static bool want_bucket(struct bch_fs *c, bool *have_cache, bool ec, struct open_bucket *ob) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); if (!test_bit(ob->dev, devs_may_alloc->d)) return false; @@ -906,7 +906,7 @@ static int bucket_alloc_set_partial(struct bch_fs *c, struct open_bucket *ob = c->open_buckets + c->open_buckets_partial[i]; if (want_bucket(c, wp, devs_may_alloc, have_cache, ec, ob)) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); struct bch_dev_usage usage; u64 avail; @@ -1291,7 +1291,7 @@ deallocate_extra_replicas(struct bch_fs *c, unsigned i; open_bucket_for_each(c, ptrs, ob, i) { - unsigned d = bch_dev_bkey_exists(c, ob->dev)->mi.durability; + unsigned d = bch2_dev_bkey_exists(c, ob->dev)->mi.durability; if (d && d <= extra_replicas) { extra_replicas -= d; @@ -1444,7 +1444,7 @@ err: struct bch_extent_ptr bch2_ob_ptr(struct bch_fs *c, struct open_bucket *ob) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); return (struct bch_extent_ptr) { .type = 1 << BCH_EXTENT_ENTRY_ptr, @@ -1520,7 +1520,7 @@ void bch2_fs_allocator_foreground_init(struct bch_fs *c) static void bch2_open_bucket_to_text(struct printbuf *out, struct bch_fs *c, struct open_bucket *ob) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); unsigned data_type = ob->data_type; barrier(); /* READ_ONCE() doesn't work on bitfields */ diff --git a/fs/bcachefs/alloc_foreground.h b/fs/bcachefs/alloc_foreground.h index 7aaeec44c746..c101b2c36743 100644 --- a/fs/bcachefs/alloc_foreground.h +++ b/fs/bcachefs/alloc_foreground.h @@ -184,7 +184,7 @@ bch2_alloc_sectors_append_ptrs_inlined(struct bch_fs *c, struct write_point *wp, wp->sectors_allocated += sectors; open_bucket_for_each(c, &wp->ptrs, ob, i) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); struct bch_extent_ptr ptr = bch2_ob_ptr(c, ob); ptr.cached = cached || diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 5586820fcf2c..36e5e63ec3d5 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -46,10 +46,10 @@ int bch2_backpointer_invalid(struct bch_fs *c, struct bkey_s_c k, struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); /* these will be caught by fsck */ - if (!bch2_dev_exists2(c, bp.k->p.inode)) + if (!bch2_dev_exists(c, bp.k->p.inode)) return 0; - struct bch_dev *ca = bch_dev_bkey_exists(c, bp.k->p.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, bp.k->p.inode); struct bpos bucket = bp_pos_to_bucket(c, bp.k->p); int ret = 0; @@ -75,7 +75,7 @@ void bch2_backpointer_to_text(struct printbuf *out, const struct bch_backpointer void bch2_backpointer_k_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) { - if (bch2_dev_exists2(c, k.k->p.inode)) { + if (bch2_dev_exists(c, k.k->p.inode)) { prt_str(out, "bucket="); bch2_bpos_to_text(out, bp_pos_to_bucket(c, k.k->p)); prt_str(out, " "); @@ -366,7 +366,7 @@ static int bch2_check_btree_backpointer(struct btree_trans *trans, struct btree_ struct printbuf buf = PRINTBUF; int ret = 0; - if (fsck_err_on(!bch2_dev_exists2(c, k.k->p.inode), c, + if (fsck_err_on(!bch2_dev_exists(c, k.k->p.inode), c, backpointer_to_missing_device, "backpointer for missing device:\n%s", (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { @@ -459,7 +459,7 @@ found: bytes = p.crc.compressed_size << 9; - struct bch_dev *ca = bch_dev_bkey_exists(c, dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, dev); if (!bch2_dev_get_ioref(ca, READ)) return false; diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index c1b274eadda1..e7f1eddbf4f4 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -39,7 +39,7 @@ void bch2_backpointer_swab(struct bkey_s); static inline struct bpos bp_pos_to_bucket(const struct bch_fs *c, struct bpos bp_pos) { - struct bch_dev *ca = bch_dev_bkey_exists(c, bp_pos.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, bp_pos.inode); u64 bucket_sector = bp_pos.offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT; return POS(bp_pos.inode, sector_to_bucket(ca, bucket_sector)); @@ -61,7 +61,7 @@ static inline struct bpos bucket_pos_to_bp(const struct bch_fs *c, struct bpos bucket, u64 bucket_offset) { - struct bch_dev *ca = bch_dev_bkey_exists(c, bucket.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, bucket.inode); struct bpos ret = bucket_pos_to_bp_noerror(ca, bucket, bucket_offset); EBUG_ON(!bkey_eq(bucket, bp_pos_to_bucket(c, ret))); return ret; diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 3dad47533491..7562446f2d2a 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -1253,11 +1253,6 @@ static inline s64 bch2_current_time(const struct bch_fs *c) return timespec_to_bch2_time(c, now); } -static inline bool bch2_dev_exists2(const struct bch_fs *c, unsigned dev) -{ - return dev < c->sb.nr_devices && c->devs[dev]; -} - static inline struct stdio_redirect *bch2_fs_stdio_redirect(struct bch_fs *c) { struct stdio_redirect *stdio = c->stdio; diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index c98b0ce45524..4c5bd256d251 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -597,7 +597,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id * use check_bucket_ref here */ bkey_for_each_ptr_decode(k->k, ptrs_c, p, entry_c) { - struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, p, entry_c); @@ -730,7 +730,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id */ struct bkey_ptrs ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); struct bucket *g = PTR_GC_BUCKET(ca, ptr); ptr->gen = g->gen; @@ -741,7 +741,7 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id restart_drop_ptrs: ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); bkey_for_each_ptr_decode(bkey_i_to_s(new).k, ptrs, p, entry) { - struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); enum bch_data_type data_type = bch2_bkey_ptr_data_type(bkey_i_to_s_c(new), p, entry); @@ -1215,7 +1215,7 @@ static int bch2_alloc_write_key(struct btree_trans *trans, struct bkey_s_c k) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, iter->pos.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, iter->pos.inode); struct bucket old_gc, gc, *b; struct bkey_i_alloc_v4 *a; struct bch_alloc_v4 old_convert, new; @@ -1351,7 +1351,7 @@ static int bch2_gc_alloc_start(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, BTREE_ITER_prefetch, k, ({ - struct bch_dev *ca = bch_dev_bkey_exists(c, k.k->p.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, k.k->p.inode); struct bucket *g = gc_bucket(ca, k.k->p.offset); struct bch_alloc_v4 a_convert; @@ -1671,7 +1671,7 @@ static int gc_btree_gens_key(struct btree_trans *trans, percpu_down_read(&c->mark_lock); bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); if (ptr_stale(ca, ptr) > 16) { percpu_up_read(&c->mark_lock); @@ -1680,7 +1680,7 @@ static int gc_btree_gens_key(struct btree_trans *trans, } bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); u8 *gen = &ca->oldest_gen[PTR_BUCKET_NR(ca, ptr)]; if (gen_after(*gen, ptr->gen)) @@ -1701,7 +1701,7 @@ update: static int bch2_alloc_write_oldest_gen(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k) { - struct bch_dev *ca = bch_dev_bkey_exists(trans->c, iter->pos.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(trans->c, iter->pos.inode); struct bch_alloc_v4 a_convert; const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert); struct bkey_i_alloc_v4 *a_mut; diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index ed80556b80fe..37f00d844edd 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1263,7 +1263,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, btree_node_reset_sib_u64s(b); bkey_for_each_ptr(bch2_bkey_ptrs(bkey_i_to_s(&b->key)), ptr) { - struct bch_dev *ca2 = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca2 = bch2_dev_bkey_exists(c, ptr->dev); if (ca2->mi.state != BCH_MEMBER_STATE_rw) set_btree_node_need_rewrite(b); @@ -1293,7 +1293,7 @@ static void btree_node_read_work(struct work_struct *work) container_of(work, struct btree_read_bio, work); struct bch_fs *c = rb->c; struct btree *b = rb->b; - struct bch_dev *ca = bch_dev_bkey_exists(c, rb->pick.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); struct bio *bio = &rb->bio; struct bch_io_failures failed = { .nr = 0 }; struct printbuf buf = PRINTBUF; @@ -1305,7 +1305,7 @@ static void btree_node_read_work(struct work_struct *work) while (1) { retry = true; bch_info(c, "retrying read"); - ca = bch_dev_bkey_exists(c, rb->pick.ptr.dev); + ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); rb->have_ioref = bch2_dev_get_ioref(ca, READ); bio_reset(bio, NULL, REQ_OP_READ|REQ_SYNC|REQ_META); bio->bi_iter.bi_sector = rb->pick.ptr.offset; @@ -1376,7 +1376,7 @@ static void btree_node_read_endio(struct bio *bio) struct bch_fs *c = rb->c; if (rb->have_ioref) { - struct bch_dev *ca = bch_dev_bkey_exists(c, rb->pick.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); bch2_latency_acct(ca, rb->start_time, READ); } @@ -1573,7 +1573,7 @@ static void btree_node_read_all_replicas_endio(struct bio *bio) struct btree_node_read_all *ra = rb->ra; if (rb->have_ioref) { - struct bch_dev *ca = bch_dev_bkey_exists(c, rb->pick.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); bch2_latency_acct(ca, rb->start_time, READ); } @@ -1615,7 +1615,7 @@ static int btree_node_read_all_replicas(struct bch_fs *c, struct btree *b, bool i = 0; bkey_for_each_ptr_decode(k.k, ptrs, pick, entry) { - struct bch_dev *ca = bch_dev_bkey_exists(c, pick.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, pick.ptr.dev); struct btree_read_bio *rb = container_of(ra->bio[i], struct btree_read_bio, bio); rb->c = c; @@ -1692,7 +1692,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, return; } - ca = bch_dev_bkey_exists(c, pick.ptr.dev); + ca = bch2_dev_bkey_exists(c, pick.ptr.dev); bio = bio_alloc_bioset(NULL, buf_pages(b->data, btree_buf_bytes(b)), @@ -1909,7 +1909,7 @@ static void btree_node_write_endio(struct bio *bio) struct btree_write_bio *wb = container_of(orig, struct btree_write_bio, wbio); struct bch_fs *c = wbio->c; struct btree *b = wbio->bio.bi_private; - struct bch_dev *ca = bch_dev_bkey_exists(c, wbio->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, wbio->dev); unsigned long flags; if (wbio->have_ioref) diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 20f471c08b2e..74e2098dfde5 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -493,7 +493,7 @@ int bch2_check_bucket_ref(struct btree_trans *trans, u32 bucket_sectors) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); size_t bucket_nr = PTR_BUCKET_NR(ca, ptr); struct printbuf buf = PRINTBUF; int ret = 0; @@ -787,7 +787,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, if (flags & BTREE_TRIGGER_gc) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry); percpu_down_read(&c->mark_lock); diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index 97b4bf8961e2..a88b9033349f 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -128,7 +128,7 @@ static inline size_t PTR_BUCKET_NR(const struct bch_dev *ca, static inline struct bpos PTR_BUCKET_POS(const struct bch_fs *c, const struct bch_extent_ptr *ptr) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); return POS(ptr->dev, PTR_BUCKET_NR(ca, ptr)); } @@ -137,7 +137,7 @@ static inline struct bpos PTR_BUCKET_POS_OFFSET(const struct bch_fs *c, const struct bch_extent_ptr *ptr, u32 *bucket_offset) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); return POS(ptr->dev, sector_to_bucket_and_offset(ca, ptr->offset, bucket_offset)); } diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 3c0f0801468e..cae4e4eb5329 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -360,7 +360,7 @@ void bch2_data_update_exit(struct data_update *update) if (c->opts.nocow_enabled) bch2_bucket_nocow_unlock(&c->nocow_locks, PTR_BUCKET_POS(c, ptr), 0); - percpu_ref_put(&bch_dev_bkey_exists(c, ptr->dev)->ref); + percpu_ref_put(&bch2_dev_bkey_exists(c, ptr->dev)->ref); } bch2_bkey_buf_exit(&update->k, c); @@ -540,7 +540,7 @@ int bch2_data_update_init(struct btree_trans *trans, m->op.watermark = m->data_opts.btree_insert_flags & BCH_WATERMARK_MASK; bkey_for_each_ptr(ptrs, ptr) - percpu_ref_get(&bch_dev_bkey_exists(c, ptr->dev)->ref); + percpu_ref_get(&bch2_dev_bkey_exists(c, ptr->dev)->ref); unsigned durability_have = 0, durability_removing = 0; @@ -652,7 +652,7 @@ err: if ((1U << i) & ptrs_locked) bch2_bucket_nocow_unlock(&c->nocow_locks, PTR_BUCKET_POS(c, &p.ptr), 0); - percpu_ref_put(&bch_dev_bkey_exists(c, p.ptr.dev)->ref); + percpu_ref_put(&bch2_dev_bkey_exists(c, p.ptr.dev)->ref); i++; } diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index 8310eff12a17..b5aa4e5b224b 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -37,7 +37,7 @@ static bool bch2_btree_verify_replica(struct bch_fs *c, struct btree *b, struct btree_node *n_ondisk = c->verify_ondisk; struct btree_node *n_sorted = c->verify_data->data; struct bset *sorted, *inmemory = &b->data->keys; - struct bch_dev *ca = bch_dev_bkey_exists(c, pick.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, pick.ptr.dev); struct bio *bio; bool failed = false, saw_error = false; @@ -194,7 +194,7 @@ void bch2_btree_node_ondisk_to_text(struct printbuf *out, struct bch_fs *c, return; } - ca = bch_dev_bkey_exists(c, pick.ptr.dev); + ca = bch2_dev_bkey_exists(c, pick.ptr.dev); if (!bch2_dev_get_ioref(ca, READ)) { prt_printf(out, "error getting device to read from: not online\n"); return; diff --git a/fs/bcachefs/disk_groups.c b/fs/bcachefs/disk_groups.c index 06a7df529b40..04b6aa414476 100644 --- a/fs/bcachefs/disk_groups.c +++ b/fs/bcachefs/disk_groups.c @@ -177,7 +177,7 @@ int bch2_sb_disk_groups_to_cpu(struct bch_fs *c) struct bch_member m = bch2_sb_member_get(c->disk_sb.sb, i); struct bch_disk_group_cpu *dst; - if (!bch2_member_exists(&m)) + if (!bch2_member_alive(&m)) continue; g = BCH_MEMBER_GROUP(&m); @@ -588,7 +588,7 @@ static void bch2_target_to_text_sb(struct printbuf *out, struct bch_sb *sb, unsi case TARGET_DEV: { struct bch_member m = bch2_sb_member_get(sb, t.dev); - if (bch2_dev_exists(sb, t.dev)) { + if (bch2_member_exists(sb, t.dev)) { prt_printf(out, "Device "); pr_uuid(out, m.uuid.b); prt_printf(out, " (%u)", t.dev); diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 86999ed80439..97ed598a55a5 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -253,7 +253,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, enum bch_data_type data_type = parity ? BCH_DATA_parity : BCH_DATA_stripe; s64 sectors = parity ? le16_to_cpu(s->sectors) : 0; const struct bch_extent_ptr *ptr = s->ptrs + ptr_idx; - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); struct bucket old, new, *g; struct printbuf buf = PRINTBUF; int ret = 0; @@ -609,7 +609,7 @@ static void ec_validate_checksums(struct bch_fs *c, struct ec_stripe_buf *buf) if (bch2_crc_cmp(want, got)) { struct printbuf err = PRINTBUF; - struct bch_dev *ca = bch_dev_bkey_exists(c, v->ptrs[i].dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, v->ptrs[i].dev); prt_str(&err, "stripe "); bch2_csum_err_msg(&err, v->csum_type, want, got); @@ -705,7 +705,7 @@ static void ec_block_io(struct bch_fs *c, struct ec_stripe_buf *buf, struct bch_stripe *v = &bkey_i_to_stripe(&buf->key)->v; unsigned offset = 0, bytes = buf->size << 9; struct bch_extent_ptr *ptr = &v->ptrs[idx]; - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); enum bch_data_type data_type = idx < v->nr_blocks - v->nr_redundant ? BCH_DATA_user : BCH_DATA_parity; @@ -1321,7 +1321,7 @@ static void zero_out_rest_of_ec_bucket(struct bch_fs *c, unsigned block, struct open_bucket *ob) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); unsigned offset = ca->mi.bucket_size - ob->sectors_free; int ret; @@ -1527,7 +1527,7 @@ void *bch2_writepoint_ec_buf(struct bch_fs *c, struct write_point *wp) BUG_ON(!ob->ec->new_stripe.data[ob->ec_idx]); - ca = bch_dev_bkey_exists(c, ob->dev); + ca = bch2_dev_bkey_exists(c, ob->dev); offset = ca->mi.bucket_size - ob->sectors_free; return ob->ec->new_stripe.data[ob->ec_idx] + (offset << 9); diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index e9db8f6d6bc1..cb8d52436ab1 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -79,8 +79,8 @@ static inline bool ptr_better(struct bch_fs *c, const struct extent_ptr_decoded p2) { if (likely(!p1.idx && !p2.idx)) { - struct bch_dev *dev1 = bch_dev_bkey_exists(c, p1.ptr.dev); - struct bch_dev *dev2 = bch_dev_bkey_exists(c, p2.ptr.dev); + struct bch_dev *dev1 = bch2_dev_bkey_exists(c, p1.ptr.dev); + struct bch_dev *dev2 = bch2_dev_bkey_exists(c, p2.ptr.dev); u64 l1 = atomic64_read(&dev1->cur_latency[READ]); u64 l2 = atomic64_read(&dev2->cur_latency[READ]); @@ -123,7 +123,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, if (p.ptr.unwritten) return 0; - ca = bch_dev_bkey_exists(c, p.ptr.dev); + ca = bch2_dev_bkey_exists(c, p.ptr.dev); /* * If there are any dirty pointers it's an error if we can't @@ -278,7 +278,7 @@ bool bch2_extent_merge(struct bch_fs *c, struct bkey_s l, struct bkey_s_c r) return false; /* Extents may not straddle buckets: */ - ca = bch_dev_bkey_exists(c, lp.ptr.dev); + ca = bch2_dev_bkey_exists(c, lp.ptr.dev); if (PTR_BUCKET_NR(ca, &lp.ptr) != PTR_BUCKET_NR(ca, &rp.ptr)) return false; @@ -667,14 +667,14 @@ static inline unsigned __extent_ptr_durability(struct bch_dev *ca, struct extent unsigned bch2_extent_ptr_desired_durability(struct bch_fs *c, struct extent_ptr_decoded *p) { - struct bch_dev *ca = bch_dev_bkey_exists(c, p->ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, p->ptr.dev); return __extent_ptr_durability(ca, p); } unsigned bch2_extent_ptr_durability(struct bch_fs *c, struct extent_ptr_decoded *p) { - struct bch_dev *ca = bch_dev_bkey_exists(c, p->ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, p->ptr.dev); if (ca->mi.state == BCH_MEMBER_STATE_failed) return 0; @@ -864,7 +864,7 @@ bool bch2_bkey_has_target(struct bch_fs *c, struct bkey_s_c k, unsigned target) bkey_for_each_ptr(ptrs, ptr) if (bch2_dev_in_target(c, ptr->dev, target) && (!ptr->cached || - !ptr_stale(bch_dev_bkey_exists(c, ptr->dev), ptr))) + !ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr))) return true; return false; @@ -973,17 +973,16 @@ bool bch2_extent_normalize(struct bch_fs *c, struct bkey_s k) bch2_bkey_drop_ptrs(k, ptr, ptr->cached && - ptr_stale(bch_dev_bkey_exists(c, ptr->dev), ptr)); + ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr)); return bkey_deleted(k.k); } void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struct bch_extent_ptr *ptr) { - struct bch_dev *ca = c && ptr->dev < c->sb.nr_devices && c->devs[ptr->dev] - ? bch_dev_bkey_exists(c, ptr->dev) - : NULL; - + out->atomic++; + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_safe(c, ptr->dev); if (!ca) { prt_printf(out, "ptr: %u:%llu gen %u%s", ptr->dev, (u64) ptr->offset, ptr->gen, @@ -1001,6 +1000,8 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struc if (bucket_valid(ca, b) && ptr_stale(ca, ptr)) prt_printf(out, " stale"); } + rcu_read_unlock(); + --out->atomic; } void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, @@ -1079,7 +1080,7 @@ static int extent_ptr_invalid(struct bch_fs *c, struct bch_dev *ca; int ret = 0; - if (!bch2_dev_exists2(c, ptr->dev)) { + if (!bch2_dev_exists(c, ptr->dev)) { /* * If we're in the write path this key might have already been * overwritten, and we could be seeing a device that doesn't @@ -1092,7 +1093,7 @@ static int extent_ptr_invalid(struct bch_fs *c, "pointer to invalid device (%u)", ptr->dev); } - ca = bch_dev_bkey_exists(c, ptr->dev); + ca = bch2_dev_bkey_exists(c, ptr->dev); bkey_for_each_ptr(ptrs, ptr2) bkey_fsck_err_on(ptr != ptr2 && ptr->dev == ptr2->dev, c, err, ptr_to_duplicate_device, diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 475ab6c02dd1..af9bd7c45f21 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -541,7 +541,7 @@ static void __bch2_read_endio(struct work_struct *work) struct bch_read_bio *rbio = container_of(work, struct bch_read_bio, work); struct bch_fs *c = rbio->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, rbio->pick.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, rbio->pick.ptr.dev); struct bio *src = &rbio->bio; struct bio *dst = &bch2_rbio_parent(rbio)->bio; struct bvec_iter dst_iter = rbio->bvec_iter; @@ -675,7 +675,7 @@ static void bch2_read_endio(struct bio *bio) struct bch_read_bio *rbio = container_of(bio, struct bch_read_bio, bio); struct bch_fs *c = rbio->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, rbio->pick.ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, rbio->pick.ptr.dev); struct workqueue_struct *wq = NULL; enum rbio_context context = RBIO_CONTEXT_NULL; @@ -762,7 +762,7 @@ static noinline void read_from_stale_dirty_pointer(struct btree_trans *trans, struct bch_extent_ptr ptr) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr.dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr.dev); struct btree_iter iter; struct printbuf buf = PRINTBUF; int ret; @@ -831,7 +831,7 @@ retry_pick: goto err; } - ca = bch_dev_bkey_exists(c, pick.ptr.dev); + ca = bch2_dev_bkey_exists(c, pick.ptr.dev); /* * Stale dirty pointers are treated as IO errors, but @failed isn't diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 9826772c0eb4..217cb98ed707 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -407,9 +407,9 @@ void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, BUG_ON(c->opts.nochanges); bkey_for_each_ptr(ptrs, ptr) { - BUG_ON(!bch2_dev_exists2(c, ptr->dev)); + BUG_ON(!bch2_dev_exists(c, ptr->dev)); - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); if (to_entry(ptr + 1) < ptrs.end) { n = to_wbio(bio_alloc_clone(NULL, &wbio->bio, @@ -650,7 +650,7 @@ static void bch2_write_endio(struct bio *bio) struct bch_write_bio *wbio = to_wbio(bio); struct bch_write_bio *parent = wbio->split ? wbio->parent : NULL; struct bch_fs *c = wbio->c; - struct bch_dev *ca = bch_dev_bkey_exists(c, wbio->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, wbio->dev); if (bch2_dev_inum_io_err_on(bio->bi_status, ca, BCH_MEMBER_ERROR_write, op->pos.inode, @@ -1272,7 +1272,7 @@ retry: bucket_nocow_lock(&c->nocow_locks, bucket_to_u64(b)); prefetch(l); - if (unlikely(!bch2_dev_get_ioref(bch_dev_bkey_exists(c, ptr->dev), WRITE))) + if (unlikely(!bch2_dev_get_ioref(bch2_dev_bkey_exists(c, ptr->dev), WRITE))) goto err_get_ioref; /* XXX allocating memory with btree locks held - rare */ @@ -1293,7 +1293,7 @@ retry: bch2_cut_back(POS(op->pos.inode, op->pos.offset + bio_sectors(bio)), op->insert_keys.top); darray_for_each(buckets, i) { - struct bch_dev *ca = bch_dev_bkey_exists(c, i->b.inode); + struct bch_dev *ca = bch2_dev_bkey_exists(c, i->b.inode); __bch2_bucket_nocow_lock(&c->nocow_locks, i->l, bucket_to_u64(i->b), @@ -1370,7 +1370,7 @@ err: return; err_get_ioref: darray_for_each(buckets, i) - percpu_ref_put(&bch_dev_bkey_exists(c, i->b.inode)->io_ref); + percpu_ref_put(&bch2_dev_bkey_exists(c, i->b.inode)->io_ref); /* Fall back to COW path: */ goto out; diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index cd14636bc2ab..1d37eda8d7f0 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -21,7 +21,7 @@ void bch2_journal_ptrs_to_text(struct printbuf *out, struct bch_fs *c, struct journal_replay *j) { darray_for_each(j->ptrs, i) { - struct bch_dev *ca = bch_dev_bkey_exists(c, i->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, i->dev); u64 offset; div64_u64_rem(i->sector, ca->mi.bucket_size, &offset); @@ -677,7 +677,7 @@ static int journal_entry_dev_usage_validate(struct bch_fs *c, dev = le32_to_cpu(u->dev); - if (journal_entry_err_on(!bch2_dev_exists2(c, dev), + if (journal_entry_err_on(!bch2_dev_exists(c, dev), c, version, jset, entry, journal_entry_dev_usage_bad_dev, "bad dev")) { @@ -1390,7 +1390,7 @@ int bch2_journal_read(struct bch_fs *c, continue; darray_for_each(i->ptrs, ptr) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); if (!ptr->csum_good) bch_err_dev_offset(ca, ptr->sector, @@ -1400,7 +1400,7 @@ int bch2_journal_read(struct bch_fs *c, } ret = jset_validate(c, - bch_dev_bkey_exists(c, i->ptrs.data[0].dev), + bch2_dev_bkey_exists(c, i->ptrs.data[0].dev), &i->j, i->ptrs.data[0].sector, READ); @@ -1731,7 +1731,7 @@ static CLOSURE_CALLBACK(journal_write_submit) unsigned sectors = vstruct_sectors(w->data, c->block_bits); extent_for_each_ptr(bkey_i_to_s_extent(&w->key), ptr) { - struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); struct journal_device *ja = &ca->journal; if (!percpu_ref_tryget(&ca->io_ref)) { diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index db383dc239bb..12bc577354ff 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -711,7 +711,7 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, a = bch2_alloc_to_v4(k, &a_convert); dirty_sectors = bch2_bucket_sectors_dirty(*a); - bucket_size = bch_dev_bkey_exists(c, bucket.inode)->mi.bucket_size; + bucket_size = bch2_dev_bkey_exists(c, bucket.inode)->mi.bucket_size; fragmentation = a->fragmentation_lru; ret = bch2_btree_write_buffer_tryflush(trans); diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index e997217374c6..cc603276f917 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -372,7 +372,7 @@ static int journal_replay_entry_early(struct bch_fs *c, case BCH_JSET_ENTRY_dev_usage: { struct jset_entry_dev_usage *u = container_of(entry, struct jset_entry_dev_usage, entry); - struct bch_dev *ca = bch_dev_bkey_exists(c, le32_to_cpu(u->dev)); + struct bch_dev *ca = bch2_dev_bkey_exists(c, le32_to_cpu(u->dev)); unsigned i, nr_types = jset_entry_dev_usage_nr_types(u); for (i = 0; i < min_t(unsigned, nr_types, BCH_DATA_NR); i++) { diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c index 678b9c20e251..f2c99703c730 100644 --- a/fs/bcachefs/replicas.c +++ b/fs/bcachefs/replicas.c @@ -84,7 +84,7 @@ int bch2_replicas_entry_validate(struct bch_replicas_entry_v1 *r, } for (unsigned i = 0; i < r->nr_devs; i++) - if (!bch2_dev_exists(sb, r->devs[i])) { + if (!bch2_member_exists(sb, r->devs[i])) { prt_printf(err, "invalid device %u in entry ", r->devs[i]); goto bad; } @@ -200,7 +200,7 @@ cpu_replicas_add_entry(struct bch_fs *c, }; for (i = 0; i < new_entry->nr_devs; i++) - BUG_ON(!bch2_dev_exists2(c, new_entry->devs[i])); + BUG_ON(!bch2_dev_exists(c, new_entry->devs[i])); BUG_ON(!new_entry->data_type); verify_replicas_entry(new_entry); @@ -954,7 +954,7 @@ bool bch2_have_enough_devs(struct bch_fs *c, struct bch_devs_mask devs, continue; for (i = 0; i < e->nr_devs; i++) { - struct bch_dev *ca = bch_dev_bkey_exists(c, e->devs[i]); + struct bch_dev *ca = bch2_dev_bkey_exists(c, e->devs[i]); nr_online += test_bit(e->devs[i], devs.d); nr_failed += ca->mi.state == BCH_MEMBER_STATE_failed; diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 0007fcad5baa..80bf09cfc021 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -164,7 +164,7 @@ static void member_to_text(struct printbuf *out, u64 bucket_size = le16_to_cpu(m.bucket_size); u64 device_size = le64_to_cpu(m.nbuckets) * bucket_size; - if (!bch2_member_exists(&m)) + if (!bch2_member_alive(&m)) return; prt_printf(out, "Device:\t%u\n", i); @@ -390,7 +390,7 @@ void bch2_dev_errors_reset(struct bch_dev *ca) bool bch2_dev_btree_bitmap_marked(struct bch_fs *c, struct bkey_s_c k) { bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) - if (!bch2_dev_btree_bitmap_marked_sectors(bch_dev_bkey_exists(c, ptr->dev), + if (!bch2_dev_btree_bitmap_marked_sectors(bch2_dev_bkey_exists(c, ptr->dev), ptr->offset, btree_sectors(c))) return false; return true; diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index fa55b9666241..53034cea5843 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -158,26 +158,38 @@ static inline struct bch_dev *bch2_get_next_online_dev(struct bch_fs *c, #define for_each_readable_member(c, ca) \ __for_each_online_member(c, ca, BIT( BCH_MEMBER_STATE_rw)|BIT(BCH_MEMBER_STATE_ro)) +static inline bool bch2_dev_exists(const struct bch_fs *c, unsigned dev) +{ + return dev < c->sb.nr_devices && c->devs[dev]; +} + /* * If a key exists that references a device, the device won't be going away and * we can omit rcu_read_lock(): */ -static inline struct bch_dev *bch_dev_bkey_exists(const struct bch_fs *c, unsigned idx) +static inline struct bch_dev *bch2_dev_bkey_exists(const struct bch_fs *c, unsigned dev) { - EBUG_ON(idx >= c->sb.nr_devices || !c->devs[idx]); + EBUG_ON(!bch2_dev_exists(c, dev)); - return rcu_dereference_check(c->devs[idx], 1); + return rcu_dereference_check(c->devs[dev], 1); } -static inline struct bch_dev *bch_dev_locked(struct bch_fs *c, unsigned idx) +static inline struct bch_dev *bch2_dev_locked(struct bch_fs *c, unsigned dev) { - EBUG_ON(idx >= c->sb.nr_devices || !c->devs[idx]); + EBUG_ON(!bch2_dev_exists(c, dev)); - return rcu_dereference_protected(c->devs[idx], + return rcu_dereference_protected(c->devs[dev], lockdep_is_held(&c->sb_lock) || lockdep_is_held(&c->state_lock)); } +static inline struct bch_dev *bch2_dev_safe(struct bch_fs *c, unsigned dev) +{ + return c && dev < c->sb.nr_devices + ? rcu_dereference(c->devs[dev]) + : NULL; +} + /* XXX kill, move to struct bch_fs */ static inline struct bch_devs_mask bch2_online_devs(struct bch_fs *c) { @@ -192,16 +204,16 @@ static inline struct bch_devs_mask bch2_online_devs(struct bch_fs *c) extern const struct bch_sb_field_ops bch_sb_field_ops_members_v1; extern const struct bch_sb_field_ops bch_sb_field_ops_members_v2; -static inline bool bch2_member_exists(struct bch_member *m) +static inline bool bch2_member_alive(struct bch_member *m) { return !bch2_is_zero(&m->uuid, sizeof(m->uuid)); } -static inline bool bch2_dev_exists(struct bch_sb *sb, unsigned dev) +static inline bool bch2_member_exists(struct bch_sb *sb, unsigned dev) { if (dev < sb->nr_devices) { struct bch_member m = bch2_sb_member_get(sb, dev); - return bch2_member_exists(&m); + return bch2_member_alive(&m); } return false; } @@ -222,7 +234,7 @@ static inline struct bch_member_cpu bch2_mi_to_cpu(struct bch_member *mi) ? BCH_MEMBER_DURABILITY(mi) - 1 : 1, .freespace_initialized = BCH_MEMBER_FREESPACE_INITIALIZED(mi), - .valid = bch2_member_exists(mi), + .valid = bch2_member_alive(mi), .btree_bitmap_shift = mi->btree_bitmap_shift, .btree_allocated_bitmap = le64_to_cpu(mi->btree_allocated_bitmap), }; diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index 52f0ffd4a0d2..fb4574958681 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -1300,7 +1300,7 @@ void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb, printbuf_tabstop_push(out, 44); for (int i = 0; i < sb->nr_devices; i++) - nr_devices += bch2_dev_exists(sb, i); + nr_devices += bch2_member_exists(sb, i); prt_printf(out, "External UUID:\t"); pr_uuid(out, sb->user_uuid.b); diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index c16131391300..64fa76788ed1 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -941,7 +941,7 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) goto err; for (i = 0; i < c->sb.nr_devices; i++) - if (bch2_dev_exists(c->disk_sb.sb, i) && + if (bch2_member_exists(c->disk_sb.sb, i) && bch2_dev_alloc(c, i)) { ret = -EEXIST; goto err; @@ -1102,7 +1102,7 @@ static int bch2_dev_in_fs(struct bch_sb_handle *fs, if (!uuid_equal(&fs->sb->uuid, &sb->sb->uuid)) return -BCH_ERR_device_not_a_member_of_filesystem; - if (!bch2_dev_exists(fs->sb, sb->sb->dev_idx)) + if (!bch2_member_exists(fs->sb, sb->sb->dev_idx)) return -BCH_ERR_device_has_been_removed; if (fs->sb->block_size != sb->sb->block_size) @@ -1412,10 +1412,9 @@ static int bch2_dev_attach_bdev(struct bch_fs *c, struct bch_sb_handle *sb) le64_to_cpu(c->disk_sb.sb->seq)) bch2_sb_to_fs(c, sb->sb); - BUG_ON(sb->sb->dev_idx >= c->sb.nr_devices || - !c->devs[sb->sb->dev_idx]); + BUG_ON(!bch2_dev_exists(c, sb->sb->dev_idx)); - ca = bch_dev_locked(c, sb->sb->dev_idx); + ca = bch2_dev_locked(c, sb->sb->dev_idx); ret = __bch2_dev_attach_bdev(ca, sb); if (ret) @@ -1507,10 +1506,10 @@ static bool bch2_fs_may_start(struct bch_fs *c) mutex_lock(&c->sb_lock); for (i = 0; i < c->disk_sb.sb->nr_devices; i++) { - if (!bch2_dev_exists(c->disk_sb.sb, i)) + if (!bch2_member_exists(c->disk_sb.sb, i)) continue; - ca = bch_dev_locked(c, i); + ca = bch2_dev_locked(c, i); if (!bch2_dev_is_online(ca) && (ca->mi.state == BCH_MEMBER_STATE_rw || @@ -1779,7 +1778,7 @@ int bch2_dev_add(struct bch_fs *c, const char *path) goto no_slot; for (dev_idx = 0; dev_idx < BCH_SB_MEMBERS_MAX; dev_idx++) - if (!bch2_dev_exists(c->disk_sb.sb, dev_idx)) + if (!bch2_member_exists(c->disk_sb.sb, dev_idx)) goto have_slot; no_slot: ret = -BCH_ERR_ENOSPC_sb_members; @@ -1885,7 +1884,7 @@ int bch2_dev_online(struct bch_fs *c, const char *path) if (ret) goto err; - ca = bch_dev_locked(c, dev_idx); + ca = bch2_dev_locked(c, dev_idx); ret = bch2_trans_mark_dev_sb(c, ca, BTREE_TRIGGER_transactional); bch_err_msg(c, ret, "bringing %s online: error from bch2_trans_mark_dev_sb", path); -- cgit v1.2.3-59-g8ed1b From af3b39b4c64198a08271485c6c2214ff637c3cbd Mon Sep 17 00:00:00 2001 From: "Ricardo B. Marliere" Date: Fri, 8 Mar 2024 09:12:47 -0300 Subject: bcachefs: chardev: make bch_chardev_class constant Since commit 43a7206b0963 ("driver core: class: make class_register() take a const *"), the driver core allows for struct class to be in read-only memory, so move the bch_chardev_class structure to be declared at build time placing it into read-only memory, instead of having to be dynamically allocated at boot time. Also, correctly clean up after failing paths in bch2_chardev_init(). Cc: Hongbo Li Cc: Greg Kroah-Hartman Suggested-by: Greg Kroah-Hartman Signed-off-by: Ricardo B. Marliere Signed-off-by: Kent Overstreet --- fs/bcachefs/chardev.c | 35 ++++++++++++++++++++++------------- 1 file changed, 22 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index 4d14f19f5185..dbcd433db004 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -961,7 +961,9 @@ static const struct file_operations bch_chardev_fops = { }; static int bch_chardev_major; -static struct class *bch_chardev_class; +static const struct class bch_chardev_class = { + .name = "bcachefs", +}; static struct device *bch_chardev; void bch2_fs_chardev_exit(struct bch_fs *c) @@ -978,7 +980,7 @@ int bch2_fs_chardev_init(struct bch_fs *c) if (c->minor < 0) return c->minor; - c->chardev = device_create(bch_chardev_class, NULL, + c->chardev = device_create(&bch_chardev_class, NULL, MKDEV(bch_chardev_major, c->minor), c, "bcachefs%u-ctl", c->minor); if (IS_ERR(c->chardev)) @@ -989,32 +991,39 @@ int bch2_fs_chardev_init(struct bch_fs *c) void bch2_chardev_exit(void) { - if (!IS_ERR_OR_NULL(bch_chardev_class)) - device_destroy(bch_chardev_class, - MKDEV(bch_chardev_major, U8_MAX)); - if (!IS_ERR_OR_NULL(bch_chardev_class)) - class_destroy(bch_chardev_class); + device_destroy(&bch_chardev_class, MKDEV(bch_chardev_major, U8_MAX)); + class_unregister(&bch_chardev_class); if (bch_chardev_major > 0) unregister_chrdev(bch_chardev_major, "bcachefs"); } int __init bch2_chardev_init(void) { + int ret; + bch_chardev_major = register_chrdev(0, "bcachefs-ctl", &bch_chardev_fops); if (bch_chardev_major < 0) return bch_chardev_major; - bch_chardev_class = class_create("bcachefs"); - if (IS_ERR(bch_chardev_class)) - return PTR_ERR(bch_chardev_class); + ret = class_register(&bch_chardev_class); + if (ret) + goto major_out; - bch_chardev = device_create(bch_chardev_class, NULL, + bch_chardev = device_create(&bch_chardev_class, NULL, MKDEV(bch_chardev_major, U8_MAX), NULL, "bcachefs-ctl"); - if (IS_ERR(bch_chardev)) - return PTR_ERR(bch_chardev); + if (IS_ERR(bch_chardev)) { + ret = PTR_ERR(bch_chardev); + goto class_out; + } return 0; + +class_out: + class_unregister(&bch_chardev_class); +major_out: + unregister_chrdev(bch_chardev_major, "bcachefs-ctl"); + return ret; } #endif /* NO_BCACHEFS_CHARDEV */ -- cgit v1.2.3-59-g8ed1b From d434c2398fe30c8e7d8784410c0a9273af0f2f79 Mon Sep 17 00:00:00 2001 From: Lukas Bulwahn Date: Thu, 11 Apr 2024 10:29:31 +0200 Subject: bcachefs: fix typo in reference to BCACHEFS_DEBUG Commit ec9cc18fc2e6 ("bcachefs: Add checks for invalid snapshot IDs") intends to check the sanity of a snapshot and panic when BCACHEFS_DEBUG is set, but that conditional has a typo. Fix the typo to refer to the actual existing Kconfig symbol. This was found with ./scripts/checkkconfigsymbols.py. Signed-off-by: Lukas Bulwahn Signed-off-by: Kent Overstreet --- fs/bcachefs/snapshot.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index b7d2fed37c4f..3fdb41b33d2d 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -77,7 +77,7 @@ static inline u32 __bch2_snapshot_parent(struct bch_fs *c, u32 id) return 0; u32 parent = s->parent; - if (IS_ENABLED(CONFIG_BCACHEFS_DEBU) && + if (IS_ENABLED(CONFIG_BCACHEFS_DEBUG) && parent && s->depth != snapshot_t(c, parent)->depth + 1) panic("id %u depth=%u parent %u depth=%u\n", -- cgit v1.2.3-59-g8ed1b From 91b5d97fdf3ca307b7aec1b3c0dcbaa0473cf43d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 12 Apr 2024 15:44:09 -0400 Subject: bcachefs: get_unlocked_mut_path -> bch2_path_get_unlocked_mut Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 16 ++++++++++++++++ fs/bcachefs/btree_iter.h | 3 +++ fs/bcachefs/btree_update_interior.c | 30 +++++++----------------------- 3 files changed, 26 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index db47063e5844..e08167e3559d 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1729,6 +1729,22 @@ btree_path_idx_t bch2_path_get(struct btree_trans *trans, return path_idx; } +btree_path_idx_t bch2_path_get_unlocked_mut(struct btree_trans *trans, + enum btree_id btree_id, + unsigned level, + struct bpos pos) +{ + btree_path_idx_t path_idx = bch2_path_get(trans, btree_id, pos, level + 1, level, + BTREE_ITER_nopreserve| + BTREE_ITER_intent, _RET_IP_); + path_idx = bch2_btree_path_make_mut(trans, path_idx, true, _RET_IP_); + + struct btree_path *path = trans->paths + path_idx; + bch2_btree_path_downgrade(trans, path); + __bch2_btree_path_unlock(trans, path); + return path_idx; +} + struct bkey_s_c bch2_btree_path_peek_slot(struct btree_path *path, struct bkey *u) { diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index abbf7bc8d9c5..3bef5b1af16d 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -227,6 +227,9 @@ static inline int __must_check bch2_btree_path_traverse(struct btree_trans *tran btree_path_idx_t bch2_path_get(struct btree_trans *, enum btree_id, struct bpos, unsigned, unsigned, unsigned, unsigned long); +btree_path_idx_t bch2_path_get_unlocked_mut(struct btree_trans *, enum btree_id, + unsigned, struct bpos); + struct bkey_s_c bch2_btree_path_peek_slot(struct btree_path *, struct bkey *); /* diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 0b6645c2266f..6427328bcf17 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -38,22 +38,6 @@ static int bch2_btree_insert_node(struct btree_update *, struct btree_trans *, btree_path_idx_t, struct btree *, struct keylist *); static void bch2_btree_update_add_new_node(struct btree_update *, struct btree *); -static btree_path_idx_t get_unlocked_mut_path(struct btree_trans *trans, - enum btree_id btree_id, - unsigned level, - struct bpos pos) -{ - btree_path_idx_t path_idx = bch2_path_get(trans, btree_id, pos, level + 1, level, - BTREE_ITER_nopreserve| - BTREE_ITER_intent, _RET_IP_); - path_idx = bch2_btree_path_make_mut(trans, path_idx, true, _RET_IP_); - - struct btree_path *path = trans->paths + path_idx; - bch2_btree_path_downgrade(trans, path); - __bch2_btree_path_unlock(trans, path); - return path_idx; -} - /* * Verify that child nodes correctly span parent node's range: */ @@ -753,7 +737,7 @@ err: */ b = READ_ONCE(as->b); if (b) { - btree_path_idx_t path_idx = get_unlocked_mut_path(trans, + btree_path_idx_t path_idx = bch2_path_get_unlocked_mut(trans, as->btree_id, b->c.level, b->key.k.p); struct btree_path *path = trans->paths + path_idx; /* @@ -773,7 +757,7 @@ err: * btree_node_lock_nopath() (the use of which is always suspect, * we need to work on removing this in the future) * - * It should be, but get_unlocked_mut_path() -> bch2_path_get() + * It should be, but bch2_path_get_unlocked_mut() -> bch2_path_get() * calls bch2_path_upgrade(), before we call path_make_mut(), so * we may rarely end up with a locked path besides the one we * have here: @@ -1637,12 +1621,12 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans, six_unlock_write(&n2->c.lock); six_unlock_write(&n1->c.lock); - path1 = get_unlocked_mut_path(trans, as->btree_id, n1->c.level, n1->key.k.p); + path1 = bch2_path_get_unlocked_mut(trans, as->btree_id, n1->c.level, n1->key.k.p); six_lock_increment(&n1->c.lock, SIX_LOCK_intent); mark_btree_node_locked(trans, trans->paths + path1, n1->c.level, BTREE_NODE_INTENT_LOCKED); bch2_btree_path_level_init(trans, trans->paths + path1, n1); - path2 = get_unlocked_mut_path(trans, as->btree_id, n2->c.level, n2->key.k.p); + path2 = bch2_path_get_unlocked_mut(trans, as->btree_id, n2->c.level, n2->key.k.p); six_lock_increment(&n2->c.lock, SIX_LOCK_intent); mark_btree_node_locked(trans, trans->paths + path2, n2->c.level, BTREE_NODE_INTENT_LOCKED); bch2_btree_path_level_init(trans, trans->paths + path2, n2); @@ -1687,7 +1671,7 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans, bch2_btree_update_add_new_node(as, n1); six_unlock_write(&n1->c.lock); - path1 = get_unlocked_mut_path(trans, as->btree_id, n1->c.level, n1->key.k.p); + path1 = bch2_path_get_unlocked_mut(trans, as->btree_id, n1->c.level, n1->key.k.p); six_lock_increment(&n1->c.lock, SIX_LOCK_intent); mark_btree_node_locked(trans, trans->paths + path1, n1->c.level, BTREE_NODE_INTENT_LOCKED); bch2_btree_path_level_init(trans, trans->paths + path1, n1); @@ -2090,7 +2074,7 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, bch2_btree_update_add_new_node(as, n); six_unlock_write(&n->c.lock); - new_path = get_unlocked_mut_path(trans, btree, n->c.level, n->key.k.p); + new_path = bch2_path_get_unlocked_mut(trans, btree, n->c.level, n->key.k.p); six_lock_increment(&n->c.lock, SIX_LOCK_intent); mark_btree_node_locked(trans, trans->paths + new_path, n->c.level, BTREE_NODE_INTENT_LOCKED); bch2_btree_path_level_init(trans, trans->paths + new_path, n); @@ -2168,7 +2152,7 @@ int bch2_btree_node_rewrite(struct btree_trans *trans, bch2_btree_update_add_new_node(as, n); six_unlock_write(&n->c.lock); - new_path = get_unlocked_mut_path(trans, iter->btree_id, n->c.level, n->key.k.p); + new_path = bch2_path_get_unlocked_mut(trans, iter->btree_id, n->c.level, n->key.k.p); six_lock_increment(&n->c.lock, SIX_LOCK_intent); mark_btree_node_locked(trans, trans->paths + new_path, n->c.level, BTREE_NODE_INTENT_LOCKED); bch2_btree_path_level_init(trans, trans->paths + new_path, n); -- cgit v1.2.3-59-g8ed1b From b7f10636d51a8f07fdf569c2de5cbfabae549c91 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 17:40:06 -0400 Subject: bcachefs: prefer drop_locks_do() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_update_interior.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 6427328bcf17..ce9712d7c764 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1160,9 +1160,8 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, if (flags & BCH_TRANS_COMMIT_journal_reclaim) return ERR_PTR(-BCH_ERR_journal_reclaim_would_deadlock); - bch2_trans_unlock(trans); - wait_event(c->journal.wait, !test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)); - ret = bch2_trans_relock(trans); + ret = drop_locks_do(trans, + ({ wait_event(c->journal.wait, !test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)); 0; })); if (ret) return ERR_PTR(ret); } -- cgit v1.2.3-59-g8ed1b From 449ceafb49e43012a76ac26a2c12aa94bdb50176 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 13 Apr 2024 17:49:23 -0400 Subject: bcachefs: bch2_trans_commit_flags_to_text() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_trans_commit.c | 21 +++++++++++++++++++++ fs/bcachefs/btree_update.h | 2 ++ fs/bcachefs/btree_update_interior.c | 9 +++++---- fs/bcachefs/btree_update_interior.h | 2 +- 4 files changed, 29 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index c5317e74ee06..04831c9e603c 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" +#include "alloc_foreground.h" #include "btree_gc.h" #include "btree_io.h" #include "btree_iter.h" @@ -19,6 +20,26 @@ #include +static const char * const trans_commit_flags_strs[] = { +#define x(n, ...) #n, + BCH_TRANS_COMMIT_FLAGS() +#undef x + NULL +}; + +void bch2_trans_commit_flags_to_text(struct printbuf *out, enum bch_trans_commit_flags flags) +{ + enum bch_watermark watermark = flags & BCH_WATERMARK_MASK; + + prt_printf(out, "watermark=%s", bch2_watermarks[watermark]); + + flags >>= BCH_WATERMARK_BITS; + if (flags) { + prt_char(out, ' '); + bch2_prt_bitflags(out, trans_commit_flags_strs, flags); + } +} + static void verify_update_old_key(struct btree_trans *trans, struct btree_insert_entry *i) { #ifdef CONFIG_BCACHEFS_DEBUG diff --git a/fs/bcachefs/btree_update.h b/fs/bcachefs/btree_update.h index 3a04ede35475..b4894e4d5447 100644 --- a/fs/bcachefs/btree_update.h +++ b/fs/bcachefs/btree_update.h @@ -44,6 +44,8 @@ enum bch_trans_commit_flags { #undef x }; +void bch2_trans_commit_flags_to_text(struct printbuf *, enum bch_trans_commit_flags); + int bch2_btree_delete_extent_at(struct btree_trans *, struct btree_iter *, unsigned, unsigned); int bch2_btree_delete_at(struct btree_trans *, struct btree_iter *, unsigned); diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index ce9712d7c764..4aa62a74a59b 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1207,7 +1207,7 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, as->start_time = start_time; as->ip_started = _RET_IP_; as->mode = BTREE_UPDATE_none; - as->watermark = watermark; + as->flags = flags; as->took_gc_lock = true; as->btree_id = path->btree_id; as->update_level_start = level_start; @@ -2565,12 +2565,13 @@ void bch2_btree_root_alloc_fake(struct bch_fs *c, enum btree_id id, unsigned lev static void bch2_btree_update_to_text(struct printbuf *out, struct btree_update *as) { - prt_printf(out, "%ps: btree=%s l=%u-%u watermark=%s mode=%s nodes_written=%u cl.remaining=%u journal_seq=%llu\n", - (void *) as->ip_started, + prt_printf(out, "%ps: ", (void *) as->ip_started); + bch2_trans_commit_flags_to_text(out, as->flags); + + prt_printf(out, " btree=%s l=%u-%u mode=%s nodes_written=%u cl.remaining=%u journal_seq=%llu\n", bch2_btree_id_str(as->btree_id), as->update_level_start, as->update_level_end, - bch2_watermarks[as->watermark], bch2_btree_update_modes[as->mode], as->nodes_written, closure_nr_remaining(&as->cl), diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index ca1a3be43af8..b2f7f8110c50 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -52,7 +52,7 @@ struct btree_update { struct list_head unwritten_list; enum btree_update_mode mode; - enum bch_watermark watermark; + enum bch_trans_commit_flags flags; unsigned nodes_written:1; unsigned took_gc_lock:1; -- cgit v1.2.3-59-g8ed1b From f2d9823f46b4aebbe2d6a7f1a76158536c6291a8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 11 Apr 2024 18:04:00 -0400 Subject: bcachefs: maintain lock invariants in btree_iter_next_node() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index e08167e3559d..bb037e4da299 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1902,6 +1902,9 @@ struct btree *bch2_btree_iter_next_node(struct btree_iter *iter) if (bpos_eq(iter->pos, b->key.k.p)) { __btree_path_set_level_up(trans, path, path->level++); } else { + if (btree_lock_want(path, path->level + 1) == BTREE_NODE_UNLOCKED) + btree_node_unlock(trans, path, path->level + 1); + /* * Haven't gotten to the end of the parent node: go back down to * the next child node -- cgit v1.2.3-59-g8ed1b From 5d8c9d94283ff170e74f13b501c3f264365621b5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 22:19:40 -0400 Subject: bcachefs: bch2_btree_path_upgrade() checks nodes_locked, not uptodate In the key cache fill path, we use path_upgrade() on a path that isn't uptodate yet but should be locked. This change makes bch2_btree_path_upgrade() slightly looser so we can use it in key cache upgrade, instead of the __ version. Also, make the related assert - that path->uptodate implies nodes_locked - slightly clearer. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_locking.c | 13 ++++++------- fs/bcachefs/btree_locking.h | 4 ++-- 2 files changed, 8 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index e6adb2df3a57..9a3ab56f6cfc 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -840,20 +840,19 @@ int __bch2_trans_mutex_lock(struct btree_trans *trans, void bch2_btree_path_verify_locks(struct btree_path *path) { - unsigned l; - /* * A path may be uptodate and yet have nothing locked if and only if * there is no node at path->level, which generally means we were * iterating over all nodes and got to the end of the btree */ - if (!path->nodes_locked) { - BUG_ON(path->uptodate == BTREE_ITER_UPTODATE && - btree_path_node(path, path->level)); + BUG_ON(path->uptodate == BTREE_ITER_UPTODATE && + btree_path_node(path, path->level) && + !path->nodes_locked); + + if (!path->nodes_locked) return; - } - for (l = 0; l < BTREE_MAX_DEPTH; l++) { + for (unsigned l = 0; l < BTREE_MAX_DEPTH; l++) { int want = btree_lock_want(path, l); int have = btree_node_locked_type(path, l); diff --git a/fs/bcachefs/btree_locking.h b/fs/bcachefs/btree_locking.h index 4bd72c855da1..7f41545b9147 100644 --- a/fs/bcachefs/btree_locking.h +++ b/fs/bcachefs/btree_locking.h @@ -364,14 +364,14 @@ static inline int bch2_btree_path_upgrade(struct btree_trans *trans, struct btree_path *path, unsigned new_locks_want) { - struct get_locks_fail f; + struct get_locks_fail f = {}; unsigned old_locks_want = path->locks_want; new_locks_want = min(new_locks_want, BTREE_MAX_DEPTH); if (path->locks_want < new_locks_want ? __bch2_btree_path_upgrade(trans, path, new_locks_want, &f) - : path->uptodate == BTREE_ITER_UPTODATE) + : path->nodes_locked) return 0; trace_and_count(trans->c, trans_restart_upgrade, trans, _THIS_IP_, path, -- cgit v1.2.3-59-g8ed1b From 4984faff5d42f8069ab9223dd80dabd4f73469f5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 00:35:24 -0400 Subject: bcachefs: Use bch2_btree_path_upgrade() in key cache traverse Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 19 +++---------------- 1 file changed, 3 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 547ba9329678..6645264a481b 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -516,22 +516,9 @@ fill: path->uptodate = BTREE_ITER_UPTODATE; if (!ck->valid && !(flags & BTREE_ITER_cached_nofill)) { - /* - * Using the underscore version because we haven't set - * path->uptodate yet: - */ - if (!path->locks_want && - !__bch2_btree_path_upgrade(trans, path, 1, NULL)) { - trace_and_count(trans->c, trans_restart_key_cache_upgrade, trans, _THIS_IP_); - ret = btree_trans_restart(trans, BCH_ERR_transaction_restart_key_cache_upgrade); - goto err; - } - - ret = btree_key_cache_fill(trans, path, ck); - if (ret) - goto err; - - ret = bch2_btree_path_relock(trans, path, _THIS_IP_); + ret = bch2_btree_path_upgrade(trans, path, 1) ?: + btree_key_cache_fill(trans, path, ck) ?: + bch2_btree_path_relock(trans, path, _THIS_IP_); if (ret) goto err; -- cgit v1.2.3-59-g8ed1b From ca563dccb2f0fb9b77971dcbead37b7e5f75723d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 23:23:08 -0400 Subject: bcachefs: bch2_trans_unlock() must always be followed by relock() or begin() We're about to add new asserts for btree_trans locking consistency, and part of that requires that aren't using the btree_trans while it's unlocked. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 3 +++ fs/bcachefs/alloc_foreground.c | 4 ++++ fs/bcachefs/btree_iter.h | 2 ++ fs/bcachefs/btree_update_interior.c | 7 ++++--- fs/bcachefs/data_update.c | 2 ++ fs/bcachefs/fs.c | 4 ++++ fs/bcachefs/io_write.c | 4 ++++ fs/bcachefs/movinggc.c | 2 ++ fs/bcachefs/rebalance.c | 2 ++ fs/bcachefs/recovery.c | 3 ++- 10 files changed, 29 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 2f91ff67453f..f07373b78174 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -2172,6 +2172,9 @@ int bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev, u64 now; int ret = 0; + if (bch2_trans_relock(trans)) + bch2_trans_begin(trans); + a = bch2_trans_start_alloc_update(trans, &iter, POS(dev, bucket_nr)); ret = PTR_ERR_OR_ZERO(a); if (ret) diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index df4439a38df5..fb8825c4e7ad 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -1342,6 +1342,10 @@ retry: *wp_ret = wp = writepoint_find(trans, write_point.v); + ret = bch2_trans_relock(trans); + if (ret) + goto err; + /* metadata may not allocate on cache devices: */ if (wp->data_type != BCH_DATA_user) have_cache = true; diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 3bef5b1af16d..0d44d613e4ac 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -729,6 +729,8 @@ transaction_restart: \ #define for_each_btree_key_upto(_trans, _iter, _btree_id, \ _start, _end, _flags, _k, _do) \ ({ \ + bch2_trans_begin(trans); \ + \ struct btree_iter _iter; \ bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ (_start), (_flags)); \ diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 4aa62a74a59b..35f80c97b973 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -737,9 +737,6 @@ err: */ b = READ_ONCE(as->b); if (b) { - btree_path_idx_t path_idx = bch2_path_get_unlocked_mut(trans, - as->btree_id, b->c.level, b->key.k.p); - struct btree_path *path = trans->paths + path_idx; /* * @b is the node we did the final insert into: * @@ -763,6 +760,10 @@ err: * have here: */ bch2_trans_unlock(trans); + bch2_trans_begin(trans); + btree_path_idx_t path_idx = bch2_path_get_unlocked_mut(trans, + as->btree_id, b->c.level, b->key.k.p); + struct btree_path *path = trans->paths + path_idx; btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_intent); mark_btree_node_locked(trans, path, b->c.level, BTREE_NODE_INTENT_LOCKED); path->l[b->c.level].lock_seq = six_lock_seq(&b->c.lock); diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index cae4e4eb5329..e8c7c5cec03f 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -386,6 +386,8 @@ static void bch2_update_unwritten_extent(struct btree_trans *trans, while (bio_sectors(bio)) { unsigned sectors = bio_sectors(bio); + bch2_trans_begin(trans); + bch2_trans_iter_init(trans, &iter, update->btree_id, update->op.pos, BTREE_ITER_slots); ret = lockrestart_do(trans, ({ diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 81cfc74828fd..841bb92e53df 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -1036,6 +1036,10 @@ retry: bch2_btree_iter_set_pos(&iter, POS(iter.pos.inode, iter.pos.offset + sectors)); + + ret = bch2_trans_relock(trans); + if (ret) + break; } start = iter.pos.offset; bch2_trans_iter_exit(trans, &iter); diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 217cb98ed707..73e25250de75 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1248,6 +1248,10 @@ retry: buckets.nr = 0; + ret = bch2_trans_relock(trans); + if (ret) + break; + k = bch2_btree_iter_peek_slot(&iter); ret = bkey_err(k); if (ret) diff --git a/fs/bcachefs/movinggc.c b/fs/bcachefs/movinggc.c index d0afd8bc0193..10bfb31c151b 100644 --- a/fs/bcachefs/movinggc.c +++ b/fs/bcachefs/movinggc.c @@ -158,6 +158,8 @@ static int bch2_copygc_get_buckets(struct moving_context *ctxt, if (bch2_fs_fatal_err_on(ret, c, "%s: from bch2_btree_write_buffer_tryflush()", bch2_err_str(ret))) return ret; + bch2_trans_begin(trans); + ret = for_each_btree_key_upto(trans, iter, BTREE_ID_lru, lru_pos(BCH_LRU_FRAGMENTATION_START, 0, 0), lru_pos(BCH_LRU_FRAGMENTATION_START, U64_MAX, LRU_TIME_MAX), diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c index c640c0f13976..cf81e5128c3a 100644 --- a/fs/bcachefs/rebalance.c +++ b/fs/bcachefs/rebalance.c @@ -323,6 +323,8 @@ static int do_rebalance(struct moving_context *ctxt) struct bkey_s_c k; int ret = 0; + bch2_trans_begin(trans); + bch2_move_stats_init(&r->work_stats, "rebalance_work"); bch2_move_stats_init(&r->scan_stats, "rebalance_scan"); diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index cc603276f917..27ee27b285bd 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -202,7 +202,7 @@ int bch2_journal_replay(struct bch_fs *c) struct journal *j = &c->journal; u64 start_seq = c->journal_replay_seq_start; u64 end_seq = c->journal_replay_seq_start; - struct btree_trans *trans = bch2_trans_get(c); + struct btree_trans *trans = NULL; bool immediate_flush = false; int ret = 0; @@ -216,6 +216,7 @@ int bch2_journal_replay(struct bch_fs *c) BUG_ON(!atomic_read(&keys->ref)); move_gap(keys, keys->nr); + trans = bch2_trans_get(c); /* * First, attempt to replay keys in sorted order. This is more -- cgit v1.2.3-59-g8ed1b From e2e568bd9775aa86196ea89b9bacc17c04c652f5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 14 Apr 2024 18:42:42 -0400 Subject: bcachefs: bch2_btree_root_alloc_fake_trans() We're starting to be more strict about transaction locked state, and multiple transactions in a task. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 6 +++--- fs/bcachefs/btree_update_interior.c | 4 ++-- fs/bcachefs/btree_update_interior.h | 2 ++ 3 files changed, 7 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 4c5bd256d251..7549e806a6f4 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -540,9 +540,9 @@ reconstruct_root: if (!bch2_btree_has_scanned_nodes(c, i)) { mustfix_fsck_err(c, btree_root_unreadable_and_scan_found_nothing, "no nodes found for btree %s, continue?", bch2_btree_id_str(i)); - bch2_btree_root_alloc_fake(c, i, 0); + bch2_btree_root_alloc_fake_trans(trans, i, 0); } else { - bch2_btree_root_alloc_fake(c, i, 1); + bch2_btree_root_alloc_fake_trans(trans, i, 1); bch2_shoot_down_journal_keys(c, i, 1, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); ret = bch2_get_scanned_nodes(c, i, 0, POS_MIN, SPOS_MAX); if (ret) @@ -570,7 +570,7 @@ reconstruct_root: goto reconstruct_root; bch_err(c, "empty btree root %s", bch2_btree_id_str(i)); - bch2_btree_root_alloc_fake(c, i, 0); + bch2_btree_root_alloc_fake_trans(trans, i, 0); r->alive = false; ret = 0; } diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 35f80c97b973..a87608dd74c6 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -2513,7 +2513,7 @@ void bch2_btree_set_root_for_read(struct bch_fs *c, struct btree *b) bch2_btree_set_root_inmem(c, b); } -static int __bch2_btree_root_alloc_fake(struct btree_trans *trans, enum btree_id id, unsigned level) +int bch2_btree_root_alloc_fake_trans(struct btree_trans *trans, enum btree_id id, unsigned level) { struct bch_fs *c = trans->c; struct closure cl; @@ -2561,7 +2561,7 @@ static int __bch2_btree_root_alloc_fake(struct btree_trans *trans, enum btree_id void bch2_btree_root_alloc_fake(struct bch_fs *c, enum btree_id id, unsigned level) { - bch2_trans_run(c, __bch2_btree_root_alloc_fake(trans, id, level)); + bch2_trans_run(c, bch2_btree_root_alloc_fake_trans(trans, id, level)); } static void bch2_btree_update_to_text(struct printbuf *out, struct btree_update *as) diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index b2f7f8110c50..b5b76ce01cfc 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -175,6 +175,8 @@ int bch2_btree_node_update_key_get_iter(struct btree_trans *, struct btree *, struct bkey_i *, unsigned, bool); void bch2_btree_set_root_for_read(struct bch_fs *, struct btree *); + +int bch2_btree_root_alloc_fake_trans(struct btree_trans *, enum btree_id, unsigned); void bch2_btree_root_alloc_fake(struct bch_fs *, enum btree_id, unsigned); static inline unsigned btree_update_reserve_required(struct bch_fs *c, -- cgit v1.2.3-59-g8ed1b From 650db8a87c343c856dcbaed2eb8d184df0308c42 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 19:57:08 -0400 Subject: bcachefs: trans->locked Add a field for tracking whether a transaction object holds btree locks, and assertions to verify state. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 9 ++++++-- fs/bcachefs/btree_iter.h | 1 - fs/bcachefs/btree_locking.c | 52 ++++++++++++++++++++++++++++----------------- fs/bcachefs/btree_types.h | 4 +++- 4 files changed, 43 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index bb037e4da299..224001d8470d 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -999,6 +999,7 @@ retry_all: bch2_trans_unlock(trans); cond_resched(); + trans->locked = true; if (unlikely(trans->memory_allocation_failure)) { struct closure cl; @@ -3023,7 +3024,8 @@ u32 bch2_trans_begin(struct btree_trans *trans) if (!trans->restarted && (need_resched() || time_after64(now, trans->last_begin_time + BTREE_TRANS_MAX_LOCK_HOLD_TIME_NS))) { - drop_locks_do(trans, (cond_resched(), 0)); + bch2_trans_unlock(trans); + cond_resched(); now = local_clock(); } trans->last_begin_time = now; @@ -3033,6 +3035,8 @@ u32 bch2_trans_begin(struct btree_trans *trans) bch2_trans_srcu_unlock(trans); trans->last_begin_ip = _RET_IP_; + trans->locked = true; + if (trans->restarted) { bch2_btree_path_traverse_all(trans); trans->notrace_relock_fail = false; @@ -3090,7 +3094,7 @@ struct btree_trans *__bch2_trans_get(struct bch_fs *c, unsigned fn_idx) */ BUG_ON(pos_task && pid == pos_task->pid && - bch2_trans_locked(pos)); + pos->locked); if (pos_task && pid < pos_task->pid) { list_add_tail(&trans->list, &pos->list); @@ -3106,6 +3110,7 @@ got_trans: trans->last_begin_time = local_clock(); trans->fn_idx = fn_idx; trans->locking_wait.task = current; + trans->locked = true; trans->journal_replay_not_finished = unlikely(!test_bit(JOURNAL_REPLAY_DONE, &c->journal.flags)) && atomic_inc_not_zero(&c->journal_keys.ref); diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 0d44d613e4ac..e47866105390 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -286,7 +286,6 @@ int bch2_trans_relock(struct btree_trans *); int bch2_trans_relock_notrace(struct btree_trans *); void bch2_trans_unlock(struct btree_trans *); void bch2_trans_unlock_long(struct btree_trans *); -bool bch2_trans_locked(struct btree_trans *); static inline int trans_was_restarted(struct btree_trans *trans, u32 restart_count) { diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index 9a3ab56f6cfc..7824a7a2e50f 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -490,8 +490,6 @@ static inline bool btree_path_get_locks(struct btree_trans *trans, if (path->uptodate == BTREE_ITER_NEED_RELOCK) path->uptodate = BTREE_ITER_UPTODATE; - bch2_trans_verify_locks(trans); - return path->uptodate < BTREE_ITER_NEED_RELOCK; } @@ -607,7 +605,9 @@ bool bch2_btree_path_relock_norestart(struct btree_trans *trans, struct btree_pa { struct get_locks_fail f; - return btree_path_get_locks(trans, path, false, &f); + bool ret = btree_path_get_locks(trans, path, false, &f); + bch2_trans_verify_locks(trans); + return ret; } int __bch2_btree_path_relock(struct btree_trans *trans, @@ -630,7 +630,9 @@ bool bch2_btree_path_upgrade_noupgrade_sibs(struct btree_trans *trans, path->locks_want = new_locks_want; - return btree_path_get_locks(trans, path, true, f); + bool ret = btree_path_get_locks(trans, path, true, f); + bch2_trans_verify_locks(trans); + return ret; } bool __bch2_btree_path_upgrade(struct btree_trans *trans, @@ -638,8 +640,9 @@ bool __bch2_btree_path_upgrade(struct btree_trans *trans, unsigned new_locks_want, struct get_locks_fail *f) { - if (bch2_btree_path_upgrade_noupgrade_sibs(trans, path, new_locks_want, f)) - return true; + bool ret = bch2_btree_path_upgrade_noupgrade_sibs(trans, path, new_locks_want, f); + if (ret) + goto out; /* * XXX: this is ugly - we'd prefer to not be mucking with other @@ -673,8 +676,9 @@ bool __bch2_btree_path_upgrade(struct btree_trans *trans, btree_path_get_locks(trans, linked, true, NULL); } } - - return false; +out: + bch2_trans_verify_locks(trans); + return ret; } void __bch2_btree_path_downgrade(struct btree_trans *trans, @@ -774,6 +778,8 @@ static inline int __bch2_trans_relock(struct btree_trans *trans, bool trace) if (unlikely(trans->restarted)) return -((int) trans->restarted); + if (unlikely(trans->locked)) + goto out; trans_for_each_path(trans, path, i) { struct get_locks_fail f; @@ -783,6 +789,8 @@ static inline int __bch2_trans_relock(struct btree_trans *trans, bool trace) return bch2_trans_relock_fail(trans, path, &f, trace); } + trans->locked = true; +out: bch2_trans_verify_locks(trans); return 0; } @@ -800,11 +808,17 @@ int bch2_trans_relock_notrace(struct btree_trans *trans) void bch2_trans_unlock_noassert(struct btree_trans *trans) { __bch2_trans_unlock(trans); + + trans->locked = false; + trans->last_unlock_ip = _RET_IP_; } void bch2_trans_unlock(struct btree_trans *trans) { __bch2_trans_unlock(trans); + + trans->locked = false; + trans->last_unlock_ip = _RET_IP_; } void bch2_trans_unlock_long(struct btree_trans *trans) @@ -813,17 +827,6 @@ void bch2_trans_unlock_long(struct btree_trans *trans) bch2_trans_srcu_unlock(trans); } -bool bch2_trans_locked(struct btree_trans *trans) -{ - struct btree_path *path; - unsigned i; - - trans_for_each_path(trans, path, i) - if (path->nodes_locked) - return true; - return false; -} - int __bch2_trans_mutex_lock(struct btree_trans *trans, struct mutex *lock) { @@ -865,6 +868,17 @@ void bch2_btree_path_verify_locks(struct btree_path *path) } } +static bool bch2_trans_locked(struct btree_trans *trans) +{ + struct btree_path *path; + unsigned i; + + trans_for_each_path(trans, path, i) + if (path->nodes_locked) + return true; + return false; +} + void bch2_trans_verify_locks(struct btree_trans *trans) { struct btree_path *path; diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index 123abeec4ce9..a973ba6264d3 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -469,6 +469,8 @@ struct btree_trans { u8 lock_must_abort; bool lock_may_not_fail:1; bool srcu_held:1; + bool locked:1; + bool write_locked:1; bool used_mempool:1; bool in_traverse_all:1; bool paths_sorted:1; @@ -476,13 +478,13 @@ struct btree_trans { bool journal_transaction_names:1; bool journal_replay_not_finished:1; bool notrace_relock_fail:1; - bool write_locked:1; enum bch_errcode restarted:16; u32 restart_count; u64 last_begin_time; unsigned long last_begin_ip; unsigned long last_restarted_ip; + unsigned long last_unlock_ip; unsigned long srcu_lock_time; const char *fn; -- cgit v1.2.3-59-g8ed1b From e590e4e2229409d0e5dc6575a75488d5ebc33b45 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 10 Apr 2024 21:51:37 -0400 Subject: bcachefs: bch2_btree_path_can_relock() With the new assertions, we shouldn't be holding locks when trans->locked is false, thus, we shouldn't use relock when we just want to check if we can relock. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 33 +++++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 224001d8470d..7718630625ed 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1346,6 +1346,26 @@ static inline void __bch2_path_free(struct btree_trans *trans, btree_path_idx_t __clear_bit(path, trans->paths_allocated); } +static bool bch2_btree_path_can_relock(struct btree_trans *trans, struct btree_path *path) +{ + unsigned l = path->level; + + do { + if (!btree_path_node(path, l)) + break; + + if (!is_btree_node(path, l)) + return false; + + if (path->l[l].lock_seq != path->l[l].b->c.lock.seq) + return false; + + l++; + } while (l < path->locks_want); + + return true; +} + void bch2_path_put(struct btree_trans *trans, btree_path_idx_t path_idx, bool intent) { struct btree_path *path = trans->paths + path_idx, *dup; @@ -1360,10 +1380,15 @@ void bch2_path_put(struct btree_trans *trans, btree_path_idx_t path_idx, bool in if (!dup && !(!path->preserve && !is_btree_node(path, path->level))) return; - if (path->should_be_locked && - !trans->restarted && - (!dup || !bch2_btree_path_relock_norestart(trans, dup))) - return; + if (path->should_be_locked && !trans->restarted) { + if (!dup) + return; + + if (!(trans->locked + ? bch2_btree_path_relock_norestart(trans, dup) + : bch2_btree_path_can_relock(trans, dup))) + return; + } if (dup) { dup->preserve |= path->preserve; -- cgit v1.2.3-59-g8ed1b From fd104e2967b766f1151b4c58daa67cbde620b376 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 9 Apr 2024 20:14:21 -0400 Subject: bcachefs: bch2_trans_verify_not_unlocked() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 16 ++++++++++++++++ fs/bcachefs/btree_iter.h | 12 ++++++++++++ fs/bcachefs/btree_locking.c | 11 +++++++++-- fs/bcachefs/btree_trans_commit.c | 7 +++++++ fs/bcachefs/btree_update_interior.c | 2 ++ 5 files changed, 46 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 7718630625ed..27f3935d67cb 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1421,6 +1421,12 @@ void __noreturn bch2_trans_in_restart_error(struct btree_trans *trans) (void *) trans->last_restarted_ip); } +void __noreturn bch2_trans_unlocked_error(struct btree_trans *trans) +{ + panic("trans should be locked, unlocked by %pS\n", + (void *) trans->last_unlock_ip); +} + noinline __cold void bch2_trans_updates_to_text(struct printbuf *buf, struct btree_trans *trans) { @@ -1690,6 +1696,7 @@ btree_path_idx_t bch2_path_get(struct btree_trans *trans, struct trans_for_each_path_inorder_iter iter; btree_path_idx_t path_pos = 0, path_idx; + bch2_trans_verify_not_unlocked(trans); bch2_trans_verify_not_in_restart(trans); bch2_trans_verify_locks(trans); @@ -1826,6 +1833,8 @@ bch2_btree_iter_traverse(struct btree_iter *iter) struct btree_trans *trans = iter->trans; int ret; + bch2_trans_verify_not_unlocked(trans); + iter->path = bch2_btree_path_set_pos(trans, iter->path, btree_iter_search_key(iter), iter->flags & BTREE_ITER_intent, @@ -2102,6 +2111,9 @@ struct bkey_s_c btree_trans_peek_key_cache(struct btree_iter *iter, struct bpos struct bkey_s_c k; int ret; + bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked(trans); + if ((iter->flags & BTREE_ITER_key_cache_fill) && bpos_eq(iter->pos, pos)) return bkey_s_c_null; @@ -2240,6 +2252,7 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e struct bpos iter_pos; int ret; + bch2_trans_verify_not_unlocked(trans); EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bkey_eq(end, POS_MAX)); if (iter->update_path) { @@ -2412,6 +2425,7 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) btree_path_idx_t saved_path = 0; int ret; + bch2_trans_verify_not_unlocked(trans); EBUG_ON(btree_iter_path(trans, iter)->cached || btree_iter_path(trans, iter)->level); @@ -2548,6 +2562,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) struct bkey_s_c k; int ret; + bch2_trans_verify_not_unlocked(trans); bch2_btree_iter_verify(iter); bch2_btree_iter_verify_entry_exit(iter); EBUG_ON(btree_iter_path(trans, iter)->level && (iter->flags & BTREE_ITER_with_key_cache)); @@ -3067,6 +3082,7 @@ u32 bch2_trans_begin(struct btree_trans *trans) trans->notrace_relock_fail = false; } + bch2_trans_verify_not_unlocked(trans); return trans->restart_count; } diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index e47866105390..430b0d484874 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -216,9 +216,13 @@ int __must_check bch2_btree_path_traverse_one(struct btree_trans *, btree_path_idx_t, unsigned, unsigned long); +static inline void bch2_trans_verify_not_unlocked(struct btree_trans *); + static inline int __must_check bch2_btree_path_traverse(struct btree_trans *trans, btree_path_idx_t path, unsigned flags) { + bch2_trans_verify_not_unlocked(trans); + if (trans->paths[path].uptodate < BTREE_ITER_NEED_RELOCK) return 0; @@ -311,6 +315,14 @@ static inline void bch2_trans_verify_not_in_restart(struct btree_trans *trans) bch2_trans_in_restart_error(trans); } +void __noreturn bch2_trans_unlocked_error(struct btree_trans *); + +static inline void bch2_trans_verify_not_unlocked(struct btree_trans *trans) +{ + if (!trans->locked) + bch2_trans_unlocked_error(trans); +} + __always_inline static int btree_trans_restart_nounlock(struct btree_trans *trans, int err) { diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index 7824a7a2e50f..c3e9b0cc7bbd 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -773,14 +773,16 @@ out: static inline int __bch2_trans_relock(struct btree_trans *trans, bool trace) { - struct btree_path *path; - unsigned i; + bch2_trans_verify_locks(trans); if (unlikely(trans->restarted)) return -((int) trans->restarted); if (unlikely(trans->locked)) goto out; + struct btree_path *path; + unsigned i; + trans_for_each_path(trans, path, i) { struct get_locks_fail f; @@ -881,6 +883,11 @@ static bool bch2_trans_locked(struct btree_trans *trans) void bch2_trans_verify_locks(struct btree_trans *trans) { + if (!trans->locked) { + BUG_ON(bch2_trans_locked(trans)); + return; + } + struct btree_path *path; unsigned i; diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 04831c9e603c..0bf7c70da417 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -630,6 +630,9 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, unsigned u64s = 0; int ret; + bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_in_restart(trans); + if (race_fault()) { trace_and_count(c, trans_restart_fault_inject, trans, trace_ip); return btree_trans_restart_nounlock(trans, BCH_ERR_transaction_restart_fault_inject); @@ -1007,6 +1010,9 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) struct bch_fs *c = trans->c; int ret = 0; + bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_in_restart(trans); + if (!trans->nr_updates && !trans->journal_entries_u64s) goto out_reset; @@ -1105,6 +1111,7 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) } retry: errored_at = NULL; + bch2_trans_verify_not_unlocked(trans); bch2_trans_verify_not_in_restart(trans); if (likely(!(flags & BCH_TRANS_COMMIT_no_journal_res))) memset(&trans->journal_res, 0, sizeof(trans->journal_res)); diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index a87608dd74c6..1febab152bfa 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1949,6 +1949,8 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, u64 start_time = local_clock(); int ret = 0; + bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked(trans); BUG_ON(!trans->paths[path].should_be_locked); BUG_ON(!btree_node_locked(&trans->paths[path], level)); -- cgit v1.2.3-59-g8ed1b From feb255537d1e67ec7e08104683658dc5106457fe Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 14 Apr 2024 19:43:12 -0400 Subject: bcachefs: assert that online_reserved == 0 on shutdown Signed-off-by: Kent Overstreet --- fs/bcachefs/super.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 64fa76788ed1..adad2a703629 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -573,6 +573,7 @@ static void __bch2_fs_free(struct bch_fs *c) BUG_ON(atomic_read(&c->journal_keys.ref)); bch2_fs_btree_write_buffer_exit(c); percpu_free_rwsem(&c->mark_lock); + EBUG_ON(percpu_u64_get(c->online_reserved)); free_percpu(c->online_reserved); darray_exit(&c->btree_roots_extra); -- cgit v1.2.3-59-g8ed1b From 427e1bb8381899084fc316cce92dbc7508d9353b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 16 Apr 2024 00:11:33 -0400 Subject: bcachefs: fs_alloc_debug_to_text() Signed-off-by: Kent Overstreet --- fs/bcachefs/sysfs.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 607c930831bf..24d4c8ef25a5 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -362,6 +362,37 @@ static void bch2_btree_wakeup_all(struct bch_fs *c) seqmutex_unlock(&c->btree_trans_lock); } +static void fs_alloc_debug_to_text(struct printbuf *out, struct bch_fs *c) +{ + unsigned nr[BCH_DATA_NR]; + + memset(nr, 0, sizeof(nr)); + + for (unsigned i = 0; i < ARRAY_SIZE(c->open_buckets); i++) + nr[c->open_buckets[i].data_type]++; + + printbuf_tabstop_push(out, 24); + + percpu_down_read(&c->mark_lock); + prt_printf(out, "hidden\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.hidden)); + prt_printf(out, "btree\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.btree)); + prt_printf(out, "data\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.data)); + prt_printf(out, "cached\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.cached)); + prt_printf(out, "reserved\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.reserved)); + prt_printf(out, "online_reserved\t%llu\n", percpu_u64_get(c->online_reserved)); + prt_printf(out, "nr_inodes\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.nr_inodes)); + percpu_up_read(&c->mark_lock); + + prt_newline(out); + prt_printf(out, "freelist_wait\t%s\n", c->freelist_wait.list.first ? "waiting" : "empty"); + prt_printf(out, "open buckets allocated\t%i\n", OPEN_BUCKETS_COUNT - c->open_buckets_nr_free); + prt_printf(out, "open buckets total\t%u\n", OPEN_BUCKETS_COUNT); + prt_printf(out, "open_buckets_wait\t%s\n", c->open_buckets_wait.list.first ? "waiting" : "empty"); + prt_printf(out, "open_buckets_btree\t%u\n", nr[BCH_DATA_btree]); + prt_printf(out, "open_buckets_user\t%u\n", nr[BCH_DATA_user]); + prt_printf(out, "btree reserve cache\t%u\n", c->btree_reserve_cache_nr); +} + SHOW(bch2_fs) { struct bch_fs *c = container_of(kobj, struct bch_fs, kobj); @@ -444,6 +475,9 @@ SHOW(bch2_fs) if (attr == &sysfs_disk_groups) bch2_disk_groups_to_text(out, c); + if (attr == &sysfs_alloc_debug) + fs_alloc_debug_to_text(out, c); + return 0; } @@ -650,6 +684,7 @@ struct attribute *bch2_fs_internal_files[] = { &sysfs_internal_uuid, &sysfs_disk_groups, + &sysfs_alloc_debug, NULL }; -- cgit v1.2.3-59-g8ed1b From 60f2b1bcf519416dbffee219132aa949d0c39d0e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 16 Apr 2024 19:33:51 -0400 Subject: bcachefs: Add asserts to bch2_dev_btree_bitmap_marked_sectors() Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-members.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 80bf09cfc021..50dc9f937c45 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -415,6 +415,9 @@ static void __bch2_dev_btree_bitmap_mark(struct bch_sb_field_members_v2 *mi, uns m->btree_bitmap_shift += resize; } + BUG_ON(m->btree_bitmap_shift > 57); + BUG_ON(end > 64ULL << m->btree_bitmap_shift); + for (unsigned bit = start >> m->btree_bitmap_shift; (u64) bit << m->btree_bitmap_shift < end; bit++) -- cgit v1.2.3-59-g8ed1b From 9de40d77f0a070814cc1107f6794b219f10d8e0b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 16 Apr 2024 19:33:56 -0400 Subject: bcachefs: Check for writing btree_ptr_v2.sectors_written == 0 Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 5 +++++ fs/bcachefs/sb-errors_types.h | 3 ++- 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index cb8d52436ab1..1143288d7940 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -201,6 +201,11 @@ int bch2_btree_ptr_v2_invalid(struct bch_fs *c, struct bkey_s_c k, c, err, btree_ptr_v2_min_key_bad, "min_key > key"); + if (flags & BKEY_INVALID_WRITE) + bkey_fsck_err_on(!bp.v->sectors_written, + c, err, btree_ptr_v2_written_0, + "sectors_written == 0"); + ret = bch2_bkey_ptrs_invalid(c, k, flags, err); fsck_err: return ret; diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 06c7a644f4a4..87324747351a 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -272,7 +272,8 @@ x(snapshot_node_missing, 264) \ x(dup_backpointer_to_bad_csum_extent, 265) \ x(btree_bitmap_not_marked, 266) \ - x(sb_clean_entry_overrun, 267) + x(sb_clean_entry_overrun, 267) \ + x(btree_ptr_v2_written_0, 268) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, -- cgit v1.2.3-59-g8ed1b From 5a2d15213d3187ed3b059a2ec8865aa9172fd3a2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 16 Apr 2024 23:08:39 -0400 Subject: bcachefs: Rip bch2_snapshot_equiv() out of fsck Originally, when deleting snapshots we didn't collapse redundant snapshot nodes; thus, the notion of a class of equivalent snapshot nodes leaked into fsck. Now we do, so snapshot ID equivalence classes are purely local to snapshot deletion. Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 120 +++++++++++++------------------------------------ fs/bcachefs/snapshot.h | 5 --- 2 files changed, 30 insertions(+), 95 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index d166c17109dd..4adb96965e11 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -486,14 +486,9 @@ static int reconstruct_reg_inode(struct btree_trans *trans, u32 snapshot, u64 in return reconstruct_inode(trans, snapshot, inum, k.k->p.offset << 9, S_IFREG); } -struct snapshots_seen_entry { - u32 id; - u32 equiv; -}; - struct snapshots_seen { struct bpos pos; - DARRAY(struct snapshots_seen_entry) ids; + snapshot_id_list ids; }; static inline void snapshots_seen_exit(struct snapshots_seen *s) @@ -508,20 +503,15 @@ static inline void snapshots_seen_init(struct snapshots_seen *s) static int snapshots_seen_add_inorder(struct bch_fs *c, struct snapshots_seen *s, u32 id) { - struct snapshots_seen_entry *i, n = { - .id = id, - .equiv = bch2_snapshot_equiv(c, id), - }; - int ret = 0; - + u32 *i; __darray_for_each(s->ids, i) { - if (i->id == id) + if (*i == id) return 0; - if (i->id > id) + if (*i > id) break; } - ret = darray_insert_item(&s->ids, i - s->ids.data, n); + int ret = darray_insert_item(&s->ids, i - s->ids.data, id); if (ret) bch_err(c, "error reallocating snapshots_seen table (size %zu)", s->ids.size); @@ -531,42 +521,11 @@ static int snapshots_seen_add_inorder(struct bch_fs *c, struct snapshots_seen *s static int snapshots_seen_update(struct bch_fs *c, struct snapshots_seen *s, enum btree_id btree_id, struct bpos pos) { - struct snapshots_seen_entry n = { - .id = pos.snapshot, - .equiv = bch2_snapshot_equiv(c, pos.snapshot), - }; - int ret = 0; - if (!bkey_eq(s->pos, pos)) s->ids.nr = 0; - s->pos = pos; - s->pos.snapshot = n.equiv; - - darray_for_each(s->ids, i) { - if (i->id == n.id) - return 0; - - /* - * We currently don't rigorously track for snapshot cleanup - * needing to be run, so it shouldn't be a fsck error yet: - */ - if (i->equiv == n.equiv) { - bch_err(c, "snapshot deletion did not finish:\n" - " duplicate keys in btree %s at %llu:%llu snapshots %u, %u (equiv %u)\n", - bch2_btree_id_str(btree_id), - pos.inode, pos.offset, - i->id, n.id, n.equiv); - set_bit(BCH_FS_need_delete_dead_snapshots, &c->flags); - return bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_delete_dead_snapshots); - } - } - ret = darray_push(&s->ids, n); - if (ret) - bch_err(c, "error reallocating snapshots_seen table (size %zu)", - s->ids.size); - return ret; + return snapshot_list_add_nodup(c, &s->ids, pos.snapshot); } /** @@ -586,12 +545,10 @@ static bool key_visible_in_snapshot(struct bch_fs *c, struct snapshots_seen *see ssize_t i; EBUG_ON(id > ancestor); - EBUG_ON(!bch2_snapshot_is_equiv(c, id)); - EBUG_ON(!bch2_snapshot_is_equiv(c, ancestor)); /* @ancestor should be the snapshot most recently added to @seen */ EBUG_ON(ancestor != seen->pos.snapshot); - EBUG_ON(ancestor != seen->ids.data[seen->ids.nr - 1].equiv); + EBUG_ON(ancestor != darray_last(seen->ids)); if (id == ancestor) return true; @@ -610,9 +567,9 @@ static bool key_visible_in_snapshot(struct bch_fs *c, struct snapshots_seen *see */ for (i = seen->ids.nr - 2; - i >= 0 && seen->ids.data[i].equiv >= id; + i >= 0 && seen->ids.data[i] >= id; --i) - if (bch2_snapshot_is_ancestor(c, id, seen->ids.data[i].equiv)) + if (bch2_snapshot_is_ancestor(c, id, seen->ids.data[i])) return false; return true; @@ -643,9 +600,6 @@ static int ref_visible2(struct bch_fs *c, u32 src, struct snapshots_seen *src_seen, u32 dst, struct snapshots_seen *dst_seen) { - src = bch2_snapshot_equiv(c, src); - dst = bch2_snapshot_equiv(c, dst); - if (dst > src) { swap(dst, src); swap(dst_seen, src_seen); @@ -692,7 +646,7 @@ static int add_inode(struct bch_fs *c, struct inode_walker *w, return darray_push(&w->inodes, ((struct inode_walker_entry) { .inode = u, - .snapshot = bch2_snapshot_equiv(c, inode.k->p.snapshot), + .snapshot = inode.k->p.snapshot, })); } @@ -728,21 +682,20 @@ static struct inode_walker_entry * lookup_inode_for_snapshot(struct bch_fs *c, struct inode_walker *w, struct bkey_s_c k) { bool is_whiteout = k.k->type == KEY_TYPE_whiteout; - u32 snapshot = bch2_snapshot_equiv(c, k.k->p.snapshot); struct inode_walker_entry *i; __darray_for_each(w->inodes, i) - if (bch2_snapshot_is_ancestor(c, snapshot, i->snapshot)) + if (bch2_snapshot_is_ancestor(c, k.k->p.snapshot, i->snapshot)) goto found; return NULL; found: - BUG_ON(snapshot > i->snapshot); + BUG_ON(k.k->p.snapshot > i->snapshot); - if (snapshot != i->snapshot && !is_whiteout) { + if (k.k->p.snapshot != i->snapshot && !is_whiteout) { struct inode_walker_entry new = *i; - new.snapshot = snapshot; + new.snapshot = k.k->p.snapshot; new.count = 0; struct printbuf buf = PRINTBUF; @@ -751,10 +704,10 @@ found: bch_info(c, "have key for inode %llu:%u but have inode in ancestor snapshot %u\n" "unexpected because we should always update the inode when we update a key in that inode\n" "%s", - w->last_pos.inode, snapshot, i->snapshot, buf.buf); + w->last_pos.inode, k.k->p.snapshot, i->snapshot, buf.buf); printbuf_exit(&buf); - while (i > w->inodes.data && i[-1].snapshot > snapshot) + while (i > w->inodes.data && i[-1].snapshot > k.k->p.snapshot) --i; size_t pos = i - w->inodes.data; @@ -786,10 +739,10 @@ static struct inode_walker_entry *walk_inode(struct btree_trans *trans, return lookup_inode_for_snapshot(trans->c, w, k); } -static int __get_visible_inodes(struct btree_trans *trans, - struct inode_walker *w, - struct snapshots_seen *s, - u64 inum) +static int get_visible_inodes(struct btree_trans *trans, + struct inode_walker *w, + struct snapshots_seen *s, + u64 inum) { struct bch_fs *c = trans->c; struct btree_iter iter; @@ -800,18 +753,16 @@ static int __get_visible_inodes(struct btree_trans *trans, for_each_btree_key_norestart(trans, iter, BTREE_ID_inodes, POS(0, inum), BTREE_ITER_all_snapshots, k, ret) { - u32 equiv = bch2_snapshot_equiv(c, k.k->p.snapshot); - if (k.k->p.offset != inum) break; - if (!ref_visible(c, s, s->pos.snapshot, equiv)) + if (!ref_visible(c, s, s->pos.snapshot, k.k->p.snapshot)) continue; if (bkey_is_inode(k.k)) add_inode(c, w, k); - if (equiv >= s->pos.snapshot) + if (k.k->p.snapshot >= s->pos.snapshot) break; } bch2_trans_iter_exit(trans, &iter); @@ -1466,7 +1417,6 @@ static int check_overlapping_extents(struct btree_trans *trans, struct snapshots_seen *seen, struct extent_ends *extent_ends, struct bkey_s_c k, - u32 equiv, struct btree_iter *iter, bool *fixed) { @@ -1535,11 +1485,8 @@ static int check_extent(struct btree_trans *trans, struct btree_iter *iter, struct bch_fs *c = trans->c; struct inode_walker_entry *i; struct printbuf buf = PRINTBUF; - struct bpos equiv = k.k->p; int ret = 0; - equiv.snapshot = bch2_snapshot_equiv(c, k.k->p.snapshot); - ret = check_key_has_snapshot(trans, iter, k); if (ret) { ret = ret < 0 ? ret : 0; @@ -1589,8 +1536,7 @@ static int check_extent(struct btree_trans *trans, struct btree_iter *iter, bch2_bkey_val_to_text(&buf, c, k), buf.buf))) goto delete; - ret = check_overlapping_extents(trans, s, extent_ends, k, - equiv.snapshot, iter, + ret = check_overlapping_extents(trans, s, extent_ends, k, iter, &inode->recalculate_sums); if (ret) goto err; @@ -1607,8 +1553,8 @@ static int check_extent(struct btree_trans *trans, struct btree_iter *iter, for (; inode->inodes.data && i >= inode->inodes.data; --i) { - if (i->snapshot > equiv.snapshot || - !key_visible_in_snapshot(c, s, i->snapshot, equiv.snapshot)) + if (i->snapshot > k.k->p.snapshot || + !key_visible_in_snapshot(c, s, i->snapshot, k.k->p.snapshot)) continue; if (k.k->type != KEY_TYPE_whiteout) { @@ -2052,7 +1998,6 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, struct bch_fs *c = trans->c; struct inode_walker_entry *i; struct printbuf buf = PRINTBUF; - struct bpos equiv; int ret = 0; ret = check_key_has_snapshot(trans, iter, k); @@ -2061,9 +2006,6 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, goto out; } - equiv = k.k->p; - equiv.snapshot = bch2_snapshot_equiv(c, k.k->p.snapshot); - ret = snapshots_seen_update(c, s, iter->btree_id, k.k->p); if (ret) goto err; @@ -2140,14 +2082,13 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, if (ret) goto err; } else { - ret = __get_visible_inodes(trans, target, s, le64_to_cpu(d.v->d_inum)); + ret = get_visible_inodes(trans, target, s, le64_to_cpu(d.v->d_inum)); if (ret) goto err; if (fsck_err_on(!target->inodes.nr, c, dirent_to_missing_inode, - "dirent points to missing inode: (equiv %u)\n%s", - equiv.snapshot, + "dirent points to missing inode:\n%s", (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { @@ -2164,7 +2105,7 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, } if (d.v->d_type == DT_DIR) - for_each_visible_inode(c, s, dir, equiv.snapshot, i) + for_each_visible_inode(c, s, dir, d.k->p.snapshot, i) i->count++; } out: @@ -2457,7 +2398,7 @@ static int check_path(struct btree_trans *trans, pathbuf *p, struct bkey_s_c ino struct btree_iter inode_iter = {}; struct bch_inode_unpacked inode; struct printbuf buf = PRINTBUF; - u32 snapshot = bch2_snapshot_equiv(c, inode_k.k->p.snapshot); + u32 snapshot = inode_k.k->p.snapshot; int ret = 0; p->nr = 0; @@ -2717,8 +2658,7 @@ static int check_nlinks_walk_dirents(struct bch_fs *c, struct nlink_table *links if (d.v->d_type != DT_DIR && d.v->d_type != DT_SUBVOL) inc_link(c, &s, links, range_start, range_end, - le64_to_cpu(d.v->d_inum), - bch2_snapshot_equiv(c, d.k->p.snapshot)); + le64_to_cpu(d.v->d_inum), d.k->p.snapshot); } 0; }))); diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index 3fdb41b33d2d..ccb1f309b8d1 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -135,11 +135,6 @@ static inline u32 bch2_snapshot_equiv(struct bch_fs *c, u32 id) return id; } -static inline bool bch2_snapshot_is_equiv(struct bch_fs *c, u32 id) -{ - return id == bch2_snapshot_equiv(c, id); -} - static inline int bch2_snapshot_is_internal_node(struct bch_fs *c, u32 id) { rcu_read_lock(); -- cgit v1.2.3-59-g8ed1b From f7643bc9749f270d487c32dc35b578575bf1adb0 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 17 Apr 2024 01:26:02 -0400 Subject: bcachefs: make btree read errors silent during scan Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_io.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 37f00d844edd..c111976fd718 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -555,6 +555,7 @@ static int __btree_err(int ret, const char *fmt, ...) { struct printbuf out = PRINTBUF; + bool silent = c->curr_recovery_pass == BCH_RECOVERY_PASS_scan_for_btree_nodes; va_list args; btree_err_msg(&out, c, ca, b, i, b->written, write); @@ -576,12 +577,14 @@ static int __btree_err(int ret, if (!have_retry && ret == -BCH_ERR_btree_node_read_err_must_retry) ret = -BCH_ERR_btree_node_read_err_bad_node; - if (ret != -BCH_ERR_btree_node_read_err_fixable) + if (!silent && ret != -BCH_ERR_btree_node_read_err_fixable) bch2_sb_error_count(c, err_type); switch (ret) { case -BCH_ERR_btree_node_read_err_fixable: - ret = bch2_fsck_err(c, FSCK_CAN_FIX, err_type, "%s", out.buf); + ret = !silent + ? bch2_fsck_err(c, FSCK_CAN_FIX, err_type, "%s", out.buf) + : -BCH_ERR_fsck_fix; if (ret != -BCH_ERR_fsck_fix && ret != -BCH_ERR_fsck_ignore) goto fsck_err; @@ -589,14 +592,17 @@ static int __btree_err(int ret, break; case -BCH_ERR_btree_node_read_err_want_retry: case -BCH_ERR_btree_node_read_err_must_retry: - bch2_print_string_as_lines(KERN_ERR, out.buf); + if (!silent) + bch2_print_string_as_lines(KERN_ERR, out.buf); break; case -BCH_ERR_btree_node_read_err_bad_node: - bch2_print_string_as_lines(KERN_ERR, out.buf); + if (!silent) + bch2_print_string_as_lines(KERN_ERR, out.buf); ret = bch2_topology_error(c); break; case -BCH_ERR_btree_node_read_err_incompatible: - bch2_print_string_as_lines(KERN_ERR, out.buf); + if (!silent) + bch2_print_string_as_lines(KERN_ERR, out.buf); ret = -BCH_ERR_fsck_errors_not_fixed; break; default: -- cgit v1.2.3-59-g8ed1b From aef7eecb5711f8bcbaf0709793ff950c290e3493 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 17 Apr 2024 02:03:22 -0400 Subject: bcachefs: Sync journal when we complete a recovery pass Make things easier when we're debugging long fsck runs - persist the work that successful recovery passes did. Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery_passes.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index c065c0d01212..4a9eb9582b6e 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -222,7 +222,8 @@ int bch2_run_recovery_passes(struct bch_fs *c) if (should_run_recovery_pass(c, c->curr_recovery_pass)) { unsigned pass = c->curr_recovery_pass; - ret = bch2_run_recovery_pass(c, c->curr_recovery_pass); + ret = bch2_run_recovery_pass(c, c->curr_recovery_pass) ?: + bch2_journal_flush(&c->journal); if (bch2_err_matches(ret, BCH_ERR_restart_recovery) || (ret && c->curr_recovery_pass < pass)) continue; -- cgit v1.2.3-59-g8ed1b From b25fd02ab4edb0193e60754626704adacbfbeefb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 19 Apr 2024 20:23:32 -0400 Subject: bcachefs: fix flag printing in journal_buf_to_text() Signed-off-by: Kent Overstreet --- fs/bcachefs/journal.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index 1b62034e197d..17ceddbd8858 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -77,9 +77,9 @@ static void bch2_journal_buf_to_text(struct printbuf *out, struct journal *j, u6 if (buf->write_started) prt_str(out, "write_started "); if (buf->write_allocated) - prt_str(out, "write allocated "); + prt_str(out, "write_allocated "); if (buf->write_done) - prt_str(out, "write done"); + prt_str(out, "write_done"); prt_newline(out); printbuf_indent_sub(out, 2); -- cgit v1.2.3-59-g8ed1b From 103304021e54bfb5cab9ba04cd5ef0dc2bf33888 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 19 Apr 2024 22:44:12 -0400 Subject: bcachefs: Move gc of bucket.oldest_gen to workqueue This is a nice cleanup - and we've also been having problems with kthread creation in the mount path. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 2 +- fs/bcachefs/alloc_foreground.c | 2 +- fs/bcachefs/bcachefs.h | 5 ++- fs/bcachefs/btree_gc.c | 82 ++++++++---------------------------------- fs/bcachefs/btree_gc.h | 12 ++----- fs/bcachefs/super.c | 10 +----- fs/bcachefs/sysfs.c | 11 ------ 7 files changed, 22 insertions(+), 102 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index f07373b78174..6fa51ee16cc3 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -874,7 +874,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, bch2_do_invalidates(c); if (statechange(a->data_type == BCH_DATA_need_gc_gens)) - bch2_do_gc_gens(c); + bch2_gc_gens_async(c); } if ((flags & BTREE_TRIGGER_gc) && diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index fb8825c4e7ad..6cb878f5e5eb 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -541,7 +541,7 @@ again: bch2_do_discards(c); if (usage->d[BCH_DATA_need_gc_gens].buckets > avail) - bch2_do_gc_gens(c); + bch2_gc_gens_async(c); if (should_invalidate_buckets(ca, *usage)) bch2_do_invalidates(c); diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 7562446f2d2a..6e324c1fe924 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -718,6 +718,7 @@ struct btree_trans_buf { x(discard_fast) \ x(invalidate) \ x(delete_dead_snapshots) \ + x(gc_gens) \ x(snapshot_delete_pagecache) \ x(sysfs) \ x(btree_write_buffer) @@ -960,8 +961,7 @@ struct bch_fs { struct work_struct discard_fast_work; /* GARBAGE COLLECTION */ - struct task_struct *gc_thread; - atomic_t kick_gc; + struct work_struct gc_gens_work; unsigned long gc_count; enum btree_id gc_gens_btree; @@ -1118,7 +1118,6 @@ struct bch_fs { u64 counters_on_mount[BCH_COUNTER_NR]; u64 __percpu *counters; - unsigned btree_gc_periodic:1; unsigned copy_gc_enabled:1; bool promote_whole_extents; diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 7549e806a6f4..919fa1b9fc2e 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -1669,6 +1669,9 @@ static int gc_btree_gens_key(struct btree_trans *trans, struct bkey_i *u; int ret; + if (unlikely(test_bit(BCH_FS_going_ro, &c->flags))) + return -EROFS; + percpu_down_read(&c->mark_lock); bkey_for_each_ptr(ptrs, ptr) { struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); @@ -1802,80 +1805,23 @@ err: return ret; } -static int bch2_gc_thread(void *arg) +static void bch2_gc_gens_work(struct work_struct *work) { - struct bch_fs *c = arg; - struct io_clock *clock = &c->io_clock[WRITE]; - unsigned long last = atomic64_read(&clock->now); - unsigned last_kick = atomic_read(&c->kick_gc); - - set_freezable(); - - while (1) { - while (1) { - set_current_state(TASK_INTERRUPTIBLE); - - if (kthread_should_stop()) { - __set_current_state(TASK_RUNNING); - return 0; - } - - if (atomic_read(&c->kick_gc) != last_kick) - break; - - if (c->btree_gc_periodic) { - unsigned long next = last + c->capacity / 16; - - if (atomic64_read(&clock->now) >= next) - break; - - bch2_io_clock_schedule_timeout(clock, next); - } else { - schedule(); - } - - try_to_freeze(); - } - __set_current_state(TASK_RUNNING); - - last = atomic64_read(&clock->now); - last_kick = atomic_read(&c->kick_gc); - - bch2_gc_gens(c); - debug_check_no_locks_held(); - } - - return 0; + struct bch_fs *c = container_of(work, struct bch_fs, gc_gens_work); + bch2_gc_gens(c); + bch2_write_ref_put(c, BCH_WRITE_REF_gc_gens); } -void bch2_gc_thread_stop(struct bch_fs *c) +void bch2_gc_gens_async(struct bch_fs *c) { - struct task_struct *p; - - p = c->gc_thread; - c->gc_thread = NULL; - - if (p) { - kthread_stop(p); - put_task_struct(p); - } + if (bch2_write_ref_tryget(c, BCH_WRITE_REF_gc_gens) && + !queue_work(c->write_ref_wq, &c->gc_gens_work)) + bch2_write_ref_put(c, BCH_WRITE_REF_gc_gens); } -int bch2_gc_thread_start(struct bch_fs *c) +void bch2_fs_gc_init(struct bch_fs *c) { - struct task_struct *p; - - if (c->gc_thread) - return 0; + seqcount_init(&c->gc_pos_lock); - p = kthread_create(bch2_gc_thread, c, "bch-gc/%s", c->name); - if (IS_ERR(p)) { - bch_err_fn(c, PTR_ERR(p)); - return PTR_ERR(p); - } - - get_task_struct(p); - c->gc_thread = p; - wake_up_process(p); - return 0; + INIT_WORK(&c->gc_gens_work, bch2_gc_gens_work); } diff --git a/fs/bcachefs/btree_gc.h b/fs/bcachefs/btree_gc.h index 0d6c0a2df613..15315aab93bd 100644 --- a/fs/bcachefs/btree_gc.h +++ b/fs/bcachefs/btree_gc.h @@ -7,9 +7,6 @@ int bch2_check_topology(struct bch_fs *); int bch2_check_allocations(struct bch_fs *); -int bch2_gc_gens(struct bch_fs *); -void bch2_gc_thread_stop(struct bch_fs *); -int bch2_gc_thread_start(struct bch_fs *); /* * For concurrent mark and sweep (with other index updates), we define a total @@ -104,11 +101,8 @@ static inline bool gc_visited(struct bch_fs *c, struct gc_pos pos) return ret; } -static inline void bch2_do_gc_gens(struct bch_fs *c) -{ - atomic_inc(&c->kick_gc); - if (c->gc_thread) - wake_up_process(c->gc_thread); -} +int bch2_gc_gens(struct bch_fs *); +void bch2_gc_gens_async(struct bch_fs *); +void bch2_fs_gc_init(struct bch_fs *); #endif /* _BCACHEFS_BTREE_GC_H */ diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index adad2a703629..36833f86d50d 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -264,7 +264,6 @@ static void __bch2_fs_read_only(struct bch_fs *c) bch2_open_buckets_stop(c, NULL, true); bch2_rebalance_stop(c); bch2_copygc_stop(c); - bch2_gc_thread_stop(c); bch2_fs_ec_flush(c); bch_verbose(c, "flushing journal and stopping allocators, journal seq %llu", @@ -486,12 +485,6 @@ static int __bch2_fs_read_write(struct bch_fs *c, bool early) } #endif - ret = bch2_gc_thread_start(c); - if (ret) { - bch_err(c, "error starting gc thread"); - return ret; - } - ret = bch2_journal_reclaim_start(&c->journal); if (ret) goto err; @@ -780,6 +773,7 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) for (i = 0; i < BCH_TIME_STAT_NR; i++) bch2_time_stats_init(&c->times[i]); + bch2_fs_gc_init(c); bch2_fs_copygc_init(c); bch2_fs_btree_key_cache_init_early(&c->btree_key_cache); bch2_fs_btree_iter_init_early(c); @@ -810,8 +804,6 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) INIT_LIST_HEAD(&c->fsck_error_msgs); mutex_init(&c->fsck_error_msgs_lock); - seqcount_init(&c->gc_pos_lock); - seqcount_init(&c->usage_lock); sema_init(&c->io_in_flight, 128); diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 24d4c8ef25a5..43edda74d3cb 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -142,7 +142,6 @@ write_attribute(trigger_invalidates); write_attribute(trigger_journal_flush); write_attribute(prune_cache); write_attribute(btree_wakeup); -rw_attribute(btree_gc_periodic); rw_attribute(gc_gens_pos); read_attribute(uuid); @@ -408,8 +407,6 @@ SHOW(bch2_fs) if (attr == &sysfs_btree_write_stats) bch2_btree_write_stats_to_text(out, c); - sysfs_printf(btree_gc_periodic, "%u", (int) c->btree_gc_periodic); - if (attr == &sysfs_gc_gens_pos) bch2_gc_gens_pos_to_text(out, c); @@ -485,14 +482,6 @@ STORE(bch2_fs) { struct bch_fs *c = container_of(kobj, struct bch_fs, kobj); - if (attr == &sysfs_btree_gc_periodic) { - ssize_t ret = strtoul_safe(buf, c->btree_gc_periodic) - ?: (ssize_t) size; - - wake_up_process(c->gc_thread); - return ret; - } - if (attr == &sysfs_copy_gc_enabled) { ssize_t ret = strtoul_safe(buf, c->copy_gc_enabled) ?: (ssize_t) size; -- cgit v1.2.3-59-g8ed1b From e4f2c4dfeeaeb70164967aa4abfa1190f7e61781 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:50:01 +0100 Subject: bcachefs: Remove calls to folio_set_error Common code doesn't test the error flag, so we don't need to set it in bcachefs. We can use folio_end_read() to combine the setting (or not) of the uptodate flag and clearing the lock flag. Cc: Kent Overstreet Cc: Brian Foster Cc: linux-bcachefs@vger.kernel.org Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Kent Overstreet --- fs/bcachefs/fs-io-buffered.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs-io-buffered.c b/fs/bcachefs/fs-io-buffered.c index 120430abdae7..b0a33fabadf8 100644 --- a/fs/bcachefs/fs-io-buffered.c +++ b/fs/bcachefs/fs-io-buffered.c @@ -30,15 +30,8 @@ static void bch2_readpages_end_io(struct bio *bio) { struct folio_iter fi; - bio_for_each_folio_all(fi, bio) { - if (!bio->bi_status) { - folio_mark_uptodate(fi.folio); - } else { - folio_clear_uptodate(fi.folio); - folio_set_error(fi.folio); - } - folio_unlock(fi.folio); - } + bio_for_each_folio_all(fi, bio) + folio_end_read(fi.folio, bio->bi_status == BLK_STS_OK); bio_put(bio); } @@ -408,7 +401,6 @@ static void bch2_writepage_io_done(struct bch_write_op *op) bio_for_each_folio_all(fi, bio) { struct bch_folio *s; - folio_set_error(fi.folio); mapping_set_error(fi.folio->mapping, -EIO); s = __bch2_folio(fi.folio); -- cgit v1.2.3-59-g8ed1b From 5147b9ae768758982b196d1b259e6372e328955e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 14:49:22 -0400 Subject: bcachefs: Btree key cache instrumentation It turns out the btree key cache shrinker wasn't actually reclaiming anything, prior to the previous patch. This adds instrumentation so that if we have further issues we can see what's going on. Specifically, sysfs internal/btree_key_cache is greatly expanded with new counters, and the SRCU sequence numbers of the first 10 entries on each pending freelist, and we also add trigger_btree_key_cache_shrink for testing without having to prune all the system caches. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 53 ++++++++++++++++++++++++++++++++++--- fs/bcachefs/btree_key_cache_types.h | 8 ++++++ fs/bcachefs/sysfs.c | 34 +++++++++--------------- 3 files changed, 69 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 6645264a481b..203fbb38e9d4 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -822,6 +822,8 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, int srcu_idx; mutex_lock(&bc->lock); + bc->requested_to_free += sc->nr_to_scan; + srcu_idx = srcu_read_lock(&c->btree_trans_barrier); flags = memalloc_nofs_save(); @@ -840,6 +842,7 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, atomic_long_dec(&bc->nr_freed); freed++; bc->nr_freed_nonpcpu--; + bc->freed++; } list_for_each_entry_safe(ck, t, &bc->freed_pcpu, list) { @@ -853,6 +856,7 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, atomic_long_dec(&bc->nr_freed); freed++; bc->nr_freed_pcpu--; + bc->freed++; } rcu_read_lock(); @@ -871,13 +875,18 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, ck = container_of(pos, struct bkey_cached, hash); if (test_bit(BKEY_CACHED_DIRTY, &ck->flags)) { + bc->skipped_dirty++; goto next; } else if (test_bit(BKEY_CACHED_ACCESSED, &ck->flags)) { clear_bit(BKEY_CACHED_ACCESSED, &ck->flags); + bc->skipped_accessed++; goto next; } else if (bkey_cached_lock_for_evict(ck)) { bkey_cached_evict(bc, ck); bkey_cached_free(bc, ck); + bc->moved_to_freelist++; + } else { + bc->skipped_lock_fail++; } scanned++; @@ -1024,11 +1033,47 @@ int bch2_fs_btree_key_cache_init(struct btree_key_cache *bc) return 0; } -void bch2_btree_key_cache_to_text(struct printbuf *out, struct btree_key_cache *c) +void bch2_btree_key_cache_to_text(struct printbuf *out, struct btree_key_cache *bc) { - prt_printf(out, "nr_freed:\t%lu\n", atomic_long_read(&c->nr_freed)); - prt_printf(out, "nr_keys:\t%lu\n", atomic_long_read(&c->nr_keys)); - prt_printf(out, "nr_dirty:\t%lu\n", atomic_long_read(&c->nr_dirty)); + struct bch_fs *c = container_of(bc, struct bch_fs, btree_key_cache); + + printbuf_tabstop_push(out, 24); + printbuf_tabstop_push(out, 12); + + unsigned flags = memalloc_nofs_save(); + mutex_lock(&bc->lock); + prt_printf(out, "keys:\t%lu\r\n", atomic_long_read(&bc->nr_keys)); + prt_printf(out, "dirty:\t%lu\r\n", atomic_long_read(&bc->nr_dirty)); + prt_printf(out, "freelist:\t%lu\r\n", atomic_long_read(&bc->nr_freed)); + prt_printf(out, "nonpcpu freelist:\t%lu\r\n", bc->nr_freed_nonpcpu); + prt_printf(out, "pcpu freelist:\t%lu\r\n", bc->nr_freed_pcpu); + + prt_printf(out, "\nshrinker:\n"); + prt_printf(out, "requested_to_free:\t%lu\r\n", bc->requested_to_free); + prt_printf(out, "freed:\t%lu\r\n", bc->freed); + prt_printf(out, "moved_to_freelist:\t%lu\r\n", bc->moved_to_freelist); + prt_printf(out, "skipped_dirty:\t%lu\r\n", bc->skipped_dirty); + prt_printf(out, "skipped_accessed:\t%lu\r\n", bc->skipped_accessed); + prt_printf(out, "skipped_lock_fail:\t%lu\r\n", bc->skipped_lock_fail); + + prt_printf(out, "srcu seq:\t%lu\r\n", get_state_synchronize_srcu(&c->btree_trans_barrier)); + + struct bkey_cached *ck; + unsigned iter = 0; + list_for_each_entry(ck, &bc->freed_nonpcpu, list) { + prt_printf(out, "freed_nonpcpu:\t%lu\r\n", ck->btree_trans_barrier_seq); + if (++iter > 10) + break; + } + + iter = 0; + list_for_each_entry(ck, &bc->freed_pcpu, list) { + prt_printf(out, "freed_pcpu:\t%lu\r\n", ck->btree_trans_barrier_seq); + if (++iter > 10) + break; + } + mutex_unlock(&bc->lock); + memalloc_flags_restore(flags); } void bch2_btree_key_cache_exit(void) diff --git a/fs/bcachefs/btree_key_cache_types.h b/fs/bcachefs/btree_key_cache_types.h index 290e4e57df5b..237e8bb3ac40 100644 --- a/fs/bcachefs/btree_key_cache_types.h +++ b/fs/bcachefs/btree_key_cache_types.h @@ -24,6 +24,14 @@ struct btree_key_cache { atomic_long_t nr_freed; atomic_long_t nr_keys; atomic_long_t nr_dirty; + + /* shrinker stats */ + unsigned long requested_to_free; + unsigned long freed; + unsigned long moved_to_freelist; + unsigned long skipped_dirty; + unsigned long skipped_accessed; + unsigned long skipped_lock_fail; }; struct bkey_cached_key { diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 43edda74d3cb..4627b0ba179e 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -140,8 +140,8 @@ write_attribute(trigger_gc); write_attribute(trigger_discards); write_attribute(trigger_invalidates); write_attribute(trigger_journal_flush); -write_attribute(prune_cache); -write_attribute(btree_wakeup); +write_attribute(trigger_btree_cache_shrink); +write_attribute(trigger_btree_key_cache_shrink); rw_attribute(gc_gens_pos); read_attribute(uuid); @@ -346,21 +346,6 @@ static void bch2_gc_gens_pos_to_text(struct printbuf *out, struct bch_fs *c) prt_printf(out, "\n"); } -static void bch2_btree_wakeup_all(struct bch_fs *c) -{ - struct btree_trans *trans; - - seqmutex_lock(&c->btree_trans_lock); - list_for_each_entry(trans, &c->btree_trans_list, list) { - struct btree_bkey_cached_common *b = READ_ONCE(trans->locking); - - if (b) - six_lock_wakeup_all(&b->lock); - - } - seqmutex_unlock(&c->btree_trans_lock); -} - static void fs_alloc_debug_to_text(struct printbuf *out, struct bch_fs *c) { unsigned nr[BCH_DATA_NR]; @@ -513,7 +498,7 @@ STORE(bch2_fs) if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_sysfs)) return -EROFS; - if (attr == &sysfs_prune_cache) { + if (attr == &sysfs_trigger_btree_cache_shrink) { struct shrink_control sc; sc.gfp_mask = GFP_KERNEL; @@ -521,8 +506,13 @@ STORE(bch2_fs) c->btree_cache.shrink->scan_objects(c->btree_cache.shrink, &sc); } - if (attr == &sysfs_btree_wakeup) - bch2_btree_wakeup_all(c); + if (attr == &sysfs_trigger_btree_key_cache_shrink) { + struct shrink_control sc; + + sc.gfp_mask = GFP_KERNEL; + sc.nr_to_scan = strtoul_or_return(buf); + c->btree_key_cache.shrink->scan_objects(c->btree_cache.shrink, &sc); + } if (attr == &sysfs_trigger_gc) bch2_gc_gens(c); @@ -656,8 +646,8 @@ struct attribute *bch2_fs_internal_files[] = { &sysfs_trigger_discards, &sysfs_trigger_invalidates, &sysfs_trigger_journal_flush, - &sysfs_prune_cache, - &sysfs_btree_wakeup, + &sysfs_trigger_btree_cache_shrink, + &sysfs_trigger_btree_key_cache_shrink, &sysfs_gc_gens_pos, -- cgit v1.2.3-59-g8ed1b From 018b32a63fabbc4c3a69c350f7ef922f1a3ac5f4 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 16:50:46 -0400 Subject: bcachefs: Add btree_allocated_bitmap to member_to_text() Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-members.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 50dc9f937c45..52054f26982f 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -234,6 +234,14 @@ static void member_to_text(struct printbuf *out, prt_printf(out, "(none)"); prt_newline(out); + prt_printf(out, "Btree allocated bitmap blocksize:\t"); + prt_units_u64(out, 1ULL << m.btree_bitmap_shift); + prt_newline(out); + + prt_printf(out, "Btree allocated bitmap:\t"); + bch2_prt_u64_base2_nbits(out, le64_to_cpu(m.btree_allocated_bitmap), 64); + prt_newline(out); + prt_printf(out, "Durability:\t%llu\n", BCH_MEMBER_DURABILITY(&m) ? BCH_MEMBER_DURABILITY(&m) - 1 : 1); prt_printf(out, "Discard:\t%llu\n", BCH_MEMBER_DISCARD(&m)); -- cgit v1.2.3-59-g8ed1b From e7f63c67fcb4a479651ed8c50306bb654749faab Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 16:10:40 -0400 Subject: bcachefs: plumb data_type into bch2_bucket_alloc_trans() prep work for making the allocator try to keep btree nodes within the existing member info btree allocated bitmap Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 12 ++++++++---- fs/bcachefs/alloc_foreground.h | 3 ++- fs/bcachefs/journal.c | 3 ++- 3 files changed, 12 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 6cb878f5e5eb..847000a7ead2 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -516,6 +516,7 @@ again: * @trans: transaction object * @ca: device to allocate from * @watermark: how important is this allocation? + * @data_type: BCH_DATA_journal, btree, user... * @cl: if not NULL, closure to be used to wait if buckets not available * @usage: for secondarily also returning the current device usage * @@ -524,6 +525,7 @@ again: static struct open_bucket *bch2_bucket_alloc_trans(struct btree_trans *trans, struct bch_dev *ca, enum bch_watermark watermark, + enum bch_data_type data_type, struct closure *cl, struct bch_dev_usage *usage) { @@ -577,6 +579,9 @@ err: if (!ob) ob = ERR_PTR(-BCH_ERR_no_buckets_found); + if (!IS_ERR(ob)) + ob->data_type = data_type; + if (!IS_ERR(ob)) trace_and_count(c, bucket_alloc, ca, bch2_watermarks[watermark], @@ -605,6 +610,7 @@ err: struct open_bucket *bch2_bucket_alloc(struct bch_fs *c, struct bch_dev *ca, enum bch_watermark watermark, + enum bch_data_type data_type, struct closure *cl) { struct bch_dev_usage usage; @@ -612,7 +618,7 @@ struct open_bucket *bch2_bucket_alloc(struct bch_fs *c, struct bch_dev *ca, bch2_trans_do(c, NULL, NULL, 0, PTR_ERR_OR_ZERO(ob = bch2_bucket_alloc_trans(trans, ca, watermark, - cl, &usage))); + data_type, cl, &usage))); return ob; } @@ -738,7 +744,7 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, continue; } - ob = bch2_bucket_alloc_trans(trans, ca, watermark, cl, &usage); + ob = bch2_bucket_alloc_trans(trans, ca, watermark, data_type, cl, &usage); if (!IS_ERR(ob)) bch2_dev_stripe_increment_inlined(ca, stripe, &usage); percpu_ref_put(&ca->ref); @@ -750,8 +756,6 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, continue; } - ob->data_type = data_type; - if (add_new_bucket(c, ptrs, devs_may_alloc, nr_replicas, nr_effective, have_cache, flags, ob)) { diff --git a/fs/bcachefs/alloc_foreground.h b/fs/bcachefs/alloc_foreground.h index c101b2c36743..f0339eb7e30e 100644 --- a/fs/bcachefs/alloc_foreground.h +++ b/fs/bcachefs/alloc_foreground.h @@ -31,7 +31,8 @@ void bch2_dev_stripe_increment(struct bch_dev *, struct dev_stripe_state *); long bch2_bucket_alloc_new_fs(struct bch_dev *); struct open_bucket *bch2_bucket_alloc(struct bch_fs *, struct bch_dev *, - enum bch_watermark, struct closure *); + enum bch_watermark, enum bch_data_type, + struct closure *); static inline void ob_push(struct bch_fs *c, struct open_buckets *obs, struct open_bucket *ob) diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index 17ceddbd8858..4d03a77f1e83 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -938,7 +938,8 @@ static int __bch2_set_nr_journal_buckets(struct bch_dev *ca, unsigned nr, break; } } else { - ob[nr_got] = bch2_bucket_alloc(c, ca, BCH_WATERMARK_normal, cl); + ob[nr_got] = bch2_bucket_alloc(c, ca, BCH_WATERMARK_normal, + BCH_DATA_journal, cl); ret = PTR_ERR_OR_ZERO(ob[nr_got]); if (ret) break; -- cgit v1.2.3-59-g8ed1b From f04158290d8bdd282899c6dc1539300df40b77d1 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 22:19:48 -0400 Subject: bcachefs: journal seq blacklist gc no longer has to walk btree Since btree_ptr_v2, we no longer require the journal seq blacklist table for skipping blacklisted bsets (btree node entries); the pointer to a given node indicates how much data is present. Therefore there's no longer any need for journal seq blacklist gc to walk the btree - we can prune entries older than journal last_seq. Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 1 - fs/bcachefs/journal_io.c | 4 +++ fs/bcachefs/journal_seq_blacklist.c | 72 +++++++++---------------------------- fs/bcachefs/journal_seq_blacklist.h | 2 +- fs/bcachefs/journal_types.h | 1 + fs/bcachefs/recovery.c | 7 ++-- fs/bcachefs/super.c | 5 --- 7 files changed, 26 insertions(+), 66 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 6e324c1fe924..d8eb5beb5977 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -930,7 +930,6 @@ struct bch_fs { /* JOURNAL SEQ BLACKLIST */ struct journal_seq_blacklist_table * journal_seq_blacklist_table; - struct work_struct journal_seq_blacklist_gc_work; /* ALLOCATOR */ spinlock_t freelist_lock; diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 1d37eda8d7f0..98cf9a65216f 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -122,6 +122,10 @@ static int journal_entry_add(struct bch_fs *c, struct bch_dev *ca, struct printbuf buf = PRINTBUF; int ret = JOURNAL_ENTRY_ADD_OK; + if (!c->journal.oldest_seq_found_ondisk || + le64_to_cpu(j->seq) < c->journal.oldest_seq_found_ondisk) + c->journal.oldest_seq_found_ondisk = le64_to_cpu(j->seq); + /* Is this entry older than the range we need? */ if (!c->opts.read_entire_journal && le64_to_cpu(j->seq) < jlist->last_seq) diff --git a/fs/bcachefs/journal_seq_blacklist.c b/fs/bcachefs/journal_seq_blacklist.c index c87c68ae27f4..d8d40d46a27a 100644 --- a/fs/bcachefs/journal_seq_blacklist.c +++ b/fs/bcachefs/journal_seq_blacklist.c @@ -1,8 +1,8 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" -#include "btree_iter.h" #include "eytzinger.h" +#include "journal.h" #include "journal_seq_blacklist.h" #include "super-io.h" @@ -217,78 +217,40 @@ const struct bch_sb_field_ops bch_sb_field_ops_journal_seq_blacklist = { .to_text = bch2_sb_journal_seq_blacklist_to_text }; -void bch2_blacklist_entries_gc(struct work_struct *work) +bool bch2_blacklist_entries_gc(struct bch_fs *c) { - struct bch_fs *c = container_of(work, struct bch_fs, - journal_seq_blacklist_gc_work); - struct journal_seq_blacklist_table *t; - struct bch_sb_field_journal_seq_blacklist *bl; struct journal_seq_blacklist_entry *src, *dst; - struct btree_trans *trans = bch2_trans_get(c); - unsigned i, nr, new_nr; - int ret; - - for (i = 0; i < BTREE_ID_NR; i++) { - struct btree_iter iter; - struct btree *b; - - bch2_trans_node_iter_init(trans, &iter, i, POS_MIN, - 0, 0, BTREE_ITER_prefetch); -retry: - bch2_trans_begin(trans); - - b = bch2_btree_iter_peek_node(&iter); - - while (!(ret = PTR_ERR_OR_ZERO(b)) && - b && - !test_bit(BCH_FS_stopping, &c->flags)) - b = bch2_btree_iter_next_node(&iter); - - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - bch2_trans_iter_exit(trans, &iter); - } - - bch2_trans_put(trans); - if (ret) - return; - - mutex_lock(&c->sb_lock); - bl = bch2_sb_field_get(c->disk_sb.sb, journal_seq_blacklist); + struct bch_sb_field_journal_seq_blacklist *bl = + bch2_sb_field_get(c->disk_sb.sb, journal_seq_blacklist); if (!bl) - goto out; + return false; - nr = blacklist_nr_entries(bl); + unsigned nr = blacklist_nr_entries(bl); dst = bl->start; - t = c->journal_seq_blacklist_table; + struct journal_seq_blacklist_table *t = c->journal_seq_blacklist_table; BUG_ON(nr != t->nr); + unsigned i; for (src = bl->start, i = eytzinger0_first(t->nr); src < bl->start + nr; src++, i = eytzinger0_next(i, nr)) { BUG_ON(t->entries[i].start != le64_to_cpu(src->start)); BUG_ON(t->entries[i].end != le64_to_cpu(src->end)); - if (t->entries[i].dirty) + if (t->entries[i].dirty || t->entries[i].end >= c->journal.oldest_seq_found_ondisk) *dst++ = *src; } - new_nr = dst - bl->start; - - bch_info(c, "nr blacklist entries was %u, now %u", nr, new_nr); - - if (new_nr != nr) { - bl = bch2_sb_field_resize(&c->disk_sb, journal_seq_blacklist, - new_nr ? sb_blacklist_u64s(new_nr) : 0); - BUG_ON(new_nr && !bl); + unsigned new_nr = dst - bl->start; + if (new_nr == nr) + return false; - if (!new_nr) - c->disk_sb.sb->features[0] &= cpu_to_le64(~(1ULL << BCH_FEATURE_journal_seq_blacklist_v3)); + bch_verbose(c, "nr blacklist entries was %u, now %u", nr, new_nr); - bch2_write_super(c); - } -out: - mutex_unlock(&c->sb_lock); + bl = bch2_sb_field_resize(&c->disk_sb, journal_seq_blacklist, + new_nr ? sb_blacklist_u64s(new_nr) : 0); + BUG_ON(new_nr && !bl); + return true; } diff --git a/fs/bcachefs/journal_seq_blacklist.h b/fs/bcachefs/journal_seq_blacklist.h index afb886ec8e25..d47636f96fdc 100644 --- a/fs/bcachefs/journal_seq_blacklist.h +++ b/fs/bcachefs/journal_seq_blacklist.h @@ -17,6 +17,6 @@ int bch2_blacklist_table_initialize(struct bch_fs *); extern const struct bch_sb_field_ops bch_sb_field_ops_journal_seq_blacklist; -void bch2_blacklist_entries_gc(struct work_struct *); +bool bch2_blacklist_entries_gc(struct bch_fs *); #endif /* _BCACHEFS_JOURNAL_SEQ_BLACKLIST_H */ diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index 08debe6bfeef..3d2135e1d7a1 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -229,6 +229,7 @@ struct journal { u64 last_seq_ondisk; u64 err_seq; u64 last_empty_seq; + u64 oldest_seq_found_ondisk; /* * FIFO of journal entries whose btree updates have not yet been diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 27ee27b285bd..c33e20aa56a2 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -878,6 +878,9 @@ use_clean: write_sb = true; } + if (bch2_blacklist_entries_gc(c)) + write_sb = true; + if (write_sb) bch2_write_super(c); mutex_unlock(&c->sb_lock); @@ -900,10 +903,6 @@ use_clean: bch_info(c, "scanning for old btree nodes done"); } - if (c->journal_seq_blacklist_table && - c->journal_seq_blacklist_table->nr > 128) - queue_work(system_long_wq, &c->journal_seq_blacklist_gc_work); - ret = 0; out: bch2_flush_fsck_errs(c); diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 36833f86d50d..72dde1f1a3e6 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -611,8 +611,6 @@ void __bch2_fs_stop(struct bch_fs *c) set_bit(BCH_FS_stopping, &c->flags); - cancel_work_sync(&c->journal_seq_blacklist_gc_work); - down_write(&c->state_lock); bch2_fs_read_only(c); up_write(&c->state_lock); @@ -796,9 +794,6 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) spin_lock_init(&c->btree_write_error_lock); - INIT_WORK(&c->journal_seq_blacklist_gc_work, - bch2_blacklist_entries_gc); - INIT_LIST_HEAD(&c->journal_iters); INIT_LIST_HEAD(&c->fsck_error_msgs); -- cgit v1.2.3-59-g8ed1b From b769590f33dcb04cb8efc4905030f4a41df08e2b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 22:03:09 -0400 Subject: bcachefs: Clean up inode alloc There's no need to be using new_inode(); we can skip all that indirection and make the code easier to follow. Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 51 +++++++++++++++++++++++++++++---------------------- 1 file changed, 29 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 841bb92e53df..5624c2ea8d55 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -213,19 +213,43 @@ static struct bch_inode_info *bch2_inode_insert(struct bch_fs *c, struct bch_ino _ret; \ }) +static struct inode *bch2_alloc_inode(struct super_block *sb) +{ + BUG(); +} + +static struct bch_inode_info *__bch2_new_inode(struct bch_fs *c) +{ + struct bch_inode_info *inode = kmem_cache_alloc(bch2_inode_cache, GFP_NOFS); + if (!inode) + return NULL; + + inode_init_once(&inode->v); + mutex_init(&inode->ei_update_lock); + two_state_lock_init(&inode->ei_pagecache_lock); + INIT_LIST_HEAD(&inode->ei_vfs_inode_list); + mutex_init(&inode->ei_quota_lock); + inode->v.i_state = 0; + + if (unlikely(inode_init_always(c->vfs_sb, &inode->v))) { + kmem_cache_free(bch2_inode_cache, inode); + return NULL; + } + + return inode; +} + /* * Allocate a new inode, dropping/retaking btree locks if necessary: */ static struct bch_inode_info *bch2_new_inode(struct btree_trans *trans) { - struct bch_fs *c = trans->c; - struct bch_inode_info *inode = memalloc_flags_do(PF_MEMALLOC_NORECLAIM|PF_MEMALLOC_NOWARN, - to_bch_ei(new_inode(c->vfs_sb))); + __bch2_new_inode(trans->c)); if (unlikely(!inode)) { - int ret = drop_locks_do(trans, (inode = to_bch_ei(new_inode(c->vfs_sb))) ? 0 : -ENOMEM); + int ret = drop_locks_do(trans, (inode = __bch2_new_inode(trans->c)) ? 0 : -ENOMEM); if (ret && inode) { __destroy_inode(&inode->v); kmem_cache_free(bch2_inode_cache, inode); @@ -290,7 +314,7 @@ __bch2_create(struct mnt_idmap *idmap, if (ret) return ERR_PTR(ret); #endif - inode = to_bch_ei(new_inode(c->vfs_sb)); + inode = __bch2_new_inode(c); if (unlikely(!inode)) { inode = ERR_PTR(-ENOMEM); goto err; @@ -1487,23 +1511,6 @@ static void bch2_vfs_inode_init(struct btree_trans *trans, subvol_inum inum, mapping_set_large_folios(inode->v.i_mapping); } -static struct inode *bch2_alloc_inode(struct super_block *sb) -{ - struct bch_inode_info *inode; - - inode = kmem_cache_alloc(bch2_inode_cache, GFP_NOFS); - if (!inode) - return NULL; - - inode_init_once(&inode->v); - mutex_init(&inode->ei_update_lock); - two_state_lock_init(&inode->ei_pagecache_lock); - INIT_LIST_HEAD(&inode->ei_vfs_inode_list); - mutex_init(&inode->ei_quota_lock); - - return &inode->v; -} - static void bch2_i_callback(struct rcu_head *head) { struct inode *vinode = container_of(head, struct inode, i_rcu); -- cgit v1.2.3-59-g8ed1b From c4e8db2b5d31fc488c644019b99bf41fd616895f Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 22 Apr 2024 17:21:35 -0400 Subject: bcachefs: bucket_data_type_mismatch() We're working on potentially unifying bch2_check_bucket_ref() and bch2_check_fix_ptrs() - or at least eliminating gratuitious differences. Most immediately, there's a bunch of cleanups to be done regarding BCH_DATA_stripe. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.h | 15 ++++++++++++++- fs/bcachefs/btree_gc.c | 6 ++---- fs/bcachefs/buckets.c | 11 +---------- fs/bcachefs/ec.c | 19 +++++++++---------- 4 files changed, 26 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 44db71ccc953..e94432ced6a6 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -67,7 +67,20 @@ static inline enum bch_data_type alloc_data_type(struct bch_alloc_v4 a, static inline enum bch_data_type bucket_data_type(enum bch_data_type data_type) { - return data_type == BCH_DATA_stripe ? BCH_DATA_user : data_type; + switch (data_type) { + case BCH_DATA_cached: + case BCH_DATA_stripe: + return BCH_DATA_user; + default: + return data_type; + } +} + +static inline bool bucket_data_type_mismatch(enum bch_data_type bucket, + enum bch_data_type ptr) +{ + return !data_type_is_empty(bucket) && + bucket_data_type(bucket) != bucket_data_type(ptr); } static inline unsigned bch2_bucket_sectors(struct bch_alloc_v4 a) diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 919fa1b9fc2e..f0adc060e4e2 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -664,10 +664,8 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id if (data_type != BCH_DATA_btree && p.ptr.gen != g->gen) continue; - if (fsck_err_on(bucket_data_type(g->data_type) && - bucket_data_type(g->data_type) != - bucket_data_type(data_type), c, - ptr_bucket_data_type_mismatch, + if (fsck_err_on(bucket_data_type_mismatch(g->data_type, data_type), + c, ptr_bucket_data_type_mismatch, "bucket %u:%zu different types of data in same bucket: %s, %s\n" "while marking %s", p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 74e2098dfde5..36d13819a27d 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -498,13 +498,6 @@ int bch2_check_bucket_ref(struct btree_trans *trans, struct printbuf buf = PRINTBUF; int ret = 0; - if (bucket_data_type == BCH_DATA_cached) - bucket_data_type = BCH_DATA_user; - - if ((bucket_data_type == BCH_DATA_stripe && ptr_data_type == BCH_DATA_user) || - (bucket_data_type == BCH_DATA_user && ptr_data_type == BCH_DATA_stripe)) - bucket_data_type = ptr_data_type = BCH_DATA_stripe; - if (gen_after(ptr->gen, b_gen)) { bch2_fsck_err(c, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, BCH_FSCK_ERR_ptr_gen_newer_than_bucket_gen, @@ -552,9 +545,7 @@ int bch2_check_bucket_ref(struct btree_trans *trans, goto out; } - if (!data_type_is_empty(bucket_data_type) && - ptr_data_type && - bucket_data_type != ptr_data_type) { + if (bucket_data_type_mismatch(bucket_data_type, ptr_data_type)) { bch2_fsck_err(c, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, BCH_FSCK_ERR_ptr_bucket_data_type_mismatch, "bucket %u:%zu gen %u different types of data in same bucket: %s, %s\n" diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 97ed598a55a5..4ccf71565961 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -165,15 +165,16 @@ void bch2_stripe_to_text(struct printbuf *out, struct bch_fs *c, static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, struct bkey_s_c_stripe s, - unsigned idx, bool deleting) + unsigned ptr_idx, bool deleting) { struct bch_fs *c = trans->c; - const struct bch_extent_ptr *ptr = &s.v->ptrs[idx]; + const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; struct btree_iter iter; struct bkey_i_alloc_v4 *a; - enum bch_data_type data_type = idx >= s.v->nr_blocks - s.v->nr_redundant - ? BCH_DATA_parity : 0; - s64 sectors = data_type ? le16_to_cpu(s.v->sectors) : 0; + unsigned nr_data = s.v->nr_blocks - s.v->nr_redundant; + bool parity = ptr_idx >= nr_data; + enum bch_data_type data_type = parity ? BCH_DATA_parity : BCH_DATA_stripe; + s64 sectors = parity ? le16_to_cpu(s.v->sectors) : 0; int ret = 0; if (deleting) @@ -201,8 +202,8 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, goto err; } - if (bch2_trans_inconsistent_on(data_type && a->v.dirty_sectors, trans, - "bucket %llu:%llu gen %u data type %s dirty_sectors %u: data already in stripe bucket %llu", + if (bch2_trans_inconsistent_on(parity && a->v.dirty_sectors, trans, + "bucket %llu:%llu gen %u data type %s dirty_sectors %u: data already in parity bucket %llu", iter.pos.inode, iter.pos.offset, a->v.gen, bch2_data_type_str(a->v.data_type), a->v.dirty_sectors, @@ -213,7 +214,7 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, a->v.stripe = s.k->p.offset; a->v.stripe_redundancy = s.v->nr_redundant; - a->v.data_type = BCH_DATA_stripe; + a->v.data_type = data_type; } else { if (bch2_trans_inconsistent_on(a->v.stripe != s.k->p.offset || a->v.stripe_redundancy != s.v->nr_redundant, trans, @@ -230,8 +231,6 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, } a->v.dirty_sectors += sectors; - if (data_type) - a->v.data_type = !deleting ? data_type : 0; ret = bch2_trans_update(trans, &iter, &a->k_i, 0); if (ret) -- cgit v1.2.3-59-g8ed1b From d9307646505e8e81a24ea5c07ae2bfc85e13f65d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 00:04:07 -0400 Subject: bcachefs: mark_stripe_bucket cleanup Start to work on unifying mark_stripe_bucket() and trans_mark_stripe_bucket(); first, clean up all the unnecessary and gratuitious differences. Signed-off-by: Kent Overstreet --- fs/bcachefs/ec.c | 100 ++++++++++++++++++++++++++++++++++++------------------- 1 file changed, 65 insertions(+), 35 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 4ccf71565961..1848335707c1 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -169,91 +169,118 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, { struct bch_fs *c = trans->c; const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; - struct btree_iter iter; - struct bkey_i_alloc_v4 *a; unsigned nr_data = s.v->nr_blocks - s.v->nr_redundant; bool parity = ptr_idx >= nr_data; enum bch_data_type data_type = parity ? BCH_DATA_parity : BCH_DATA_stripe; s64 sectors = parity ? le16_to_cpu(s.v->sectors) : 0; + struct printbuf buf = PRINTBUF; int ret = 0; if (deleting) sectors = -sectors; - a = bch2_trans_start_alloc_update(trans, &iter, PTR_BUCKET_POS(c, ptr)); - if (IS_ERR(a)) - return PTR_ERR(a); - - ret = bch2_check_bucket_ref(trans, s.s_c, ptr, sectors, data_type, - a->v.gen, a->v.data_type, - a->v.dirty_sectors); + struct btree_iter iter; + struct bkey_i_alloc_v4 *a = + bch2_trans_start_alloc_update(trans, &iter, PTR_BUCKET_POS(c, ptr)); + ret = PTR_ERR_OR_ZERO(a); if (ret) goto err; if (!deleting) { if (bch2_trans_inconsistent_on(a->v.stripe || a->v.stripe_redundancy, trans, - "bucket %llu:%llu gen %u data type %s dirty_sectors %u: multiple stripes using same bucket (%u, %llu)", + "bucket %llu:%llu gen %u data type %s dirty_sectors %u: multiple stripes using same bucket (%u, %llu)\n%s", iter.pos.inode, iter.pos.offset, a->v.gen, bch2_data_type_str(a->v.data_type), a->v.dirty_sectors, - a->v.stripe, s.k->p.offset)) { + a->v.stripe, s.k->p.offset, + (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } - if (bch2_trans_inconsistent_on(parity && a->v.dirty_sectors, trans, - "bucket %llu:%llu gen %u data type %s dirty_sectors %u: data already in parity bucket %llu", + if (bch2_trans_inconsistent_on(parity && (a->v.dirty_sectors || a->v.cached_sectors), trans, + "bucket %llu:%llu gen %u data type %s dirty_sectors %u cached_sectors %u: data already in parity bucket\n%s", iter.pos.inode, iter.pos.offset, a->v.gen, bch2_data_type_str(a->v.data_type), a->v.dirty_sectors, - s.k->p.offset)) { + a->v.cached_sectors, + (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } - - a->v.stripe = s.k->p.offset; - a->v.stripe_redundancy = s.v->nr_redundant; - a->v.data_type = data_type; } else { if (bch2_trans_inconsistent_on(a->v.stripe != s.k->p.offset || a->v.stripe_redundancy != s.v->nr_redundant, trans, - "bucket %llu:%llu gen %u: not marked as stripe when deleting stripe %llu (got %u)", + "bucket %llu:%llu gen %u: not marked as stripe when deleting stripe (got %u)\n%s", iter.pos.inode, iter.pos.offset, a->v.gen, - s.k->p.offset, a->v.stripe)) { + a->v.stripe, + (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } + if (bch2_trans_inconsistent_on(a->v.data_type != data_type, trans, + "bucket %llu:%llu gen %u data type %s: wrong data type when stripe, should be %s\n%s", + iter.pos.inode, iter.pos.offset, a->v.gen, + bch2_data_type_str(a->v.data_type), + bch2_data_type_str(data_type), + (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { + ret = -EIO; + goto err; + } + + if (bch2_trans_inconsistent_on(parity && + (a->v.dirty_sectors != -sectors || + a->v.cached_sectors), trans, + "bucket %llu:%llu gen %u dirty_sectors %u cached_sectors %u: wrong sectors when deleting parity block of stripe\n%s", + iter.pos.inode, iter.pos.offset, a->v.gen, + a->v.dirty_sectors, + a->v.cached_sectors, + (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { + ret = -EIO; + goto err; + } + } + + ret = bch2_check_bucket_ref(trans, s.s_c, ptr, sectors, data_type, + a->v.gen, a->v.data_type, + a->v.dirty_sectors); + if (ret) + goto err; + + a->v.dirty_sectors += sectors; + + if (!deleting) { + a->v.stripe = s.k->p.offset; + a->v.stripe_redundancy = s.v->nr_redundant; + a->v.data_type = data_type; + } else { a->v.stripe = 0; a->v.stripe_redundancy = 0; a->v.data_type = alloc_data_type(a->v, BCH_DATA_user); } - a->v.dirty_sectors += sectors; - ret = bch2_trans_update(trans, &iter, &a->k_i, 0); if (ret) goto err; err: bch2_trans_iter_exit(trans, &iter); + printbuf_exit(&buf); return ret; } static int mark_stripe_bucket(struct btree_trans *trans, - struct bkey_s_c k, + struct bkey_s_c_stripe s, unsigned ptr_idx, enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; - const struct bch_stripe *s = bkey_s_c_to_stripe(k).v; - unsigned nr_data = s->nr_blocks - s->nr_redundant; + const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; + unsigned nr_data = s.v->nr_blocks - s.v->nr_redundant; bool parity = ptr_idx >= nr_data; enum bch_data_type data_type = parity ? BCH_DATA_parity : BCH_DATA_stripe; - s64 sectors = parity ? le16_to_cpu(s->sectors) : 0; - const struct bch_extent_ptr *ptr = s->ptrs + ptr_idx; - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - struct bucket old, new, *g; + s64 sectors = parity ? le16_to_cpu(s.v->sectors) : 0; struct printbuf buf = PRINTBUF; int ret = 0; @@ -261,15 +288,18 @@ static int mark_stripe_bucket(struct btree_trans *trans, /* * XXX doesn't handle deletion */ + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bucket old, new, *g; + percpu_down_read(&c->mark_lock); g = PTR_GC_BUCKET(ca, ptr); if (g->dirty_sectors || - (g->stripe && g->stripe != k.k->p.offset)) { + (g->stripe && g->stripe != s.k->p.offset)) { bch2_fs_inconsistent(c, "bucket %u:%zu gen %u: multiple stripes using same bucket\n%s", ptr->dev, PTR_BUCKET_NR(ca, ptr), g->gen, - (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); + (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf)); ret = -EINVAL; goto err; } @@ -277,7 +307,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, bucket_lock(g); old = *g; - ret = bch2_check_bucket_ref(trans, k, ptr, sectors, data_type, + ret = bch2_check_bucket_ref(trans, s.s_c, ptr, sectors, data_type, g->gen, g->data_type, g->dirty_sectors); if (ret) @@ -286,8 +316,8 @@ static int mark_stripe_bucket(struct btree_trans *trans, g->data_type = data_type; g->dirty_sectors += sectors; - g->stripe = k.k->p.offset; - g->stripe_redundancy = s->nr_redundant; + g->stripe = s.k->p.offset; + g->stripe_redundancy = s.v->nr_redundant; new = *g; err: bucket_unlock(g); @@ -439,7 +469,7 @@ int bch2_trigger_stripe(struct btree_trans *trans, memset(m->block_sectors, 0, sizeof(m->block_sectors)); for (unsigned i = 0; i < new_s->nr_blocks; i++) { - int ret = mark_stripe_bucket(trans, new, i, flags); + int ret = mark_stripe_bucket(trans, bkey_s_c_to_stripe(new), i, flags); if (ret) return ret; } -- cgit v1.2.3-59-g8ed1b From 9cc455d1bcd3bd44d1f103bf3f2e58293976a006 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 22 Apr 2024 19:01:40 -0400 Subject: bcachefs: Consolidate mark_stripe_bucket() and trans_mark_stripe_bucket() This eliminates some duplicated logic, and the gc path now handles stripe updates and deletions - we need this since soon we're bringing back runtime gc. Signed-off-by: Kent Overstreet --- fs/bcachefs/ec.c | 251 ++++++++++++++++++++++++++++++------------------------- 1 file changed, 135 insertions(+), 116 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 1848335707c1..2d1b20856841 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -163,9 +163,16 @@ void bch2_stripe_to_text(struct printbuf *out, struct bch_fs *c, /* Triggers: */ -static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, - struct bkey_s_c_stripe s, - unsigned ptr_idx, bool deleting) +static int __mark_stripe_bucket(struct btree_trans *trans, + struct bkey_s_c_stripe s, + unsigned ptr_idx, bool deleting, + struct bpos bucket, + u8 bucket_gen, + u8 *bucket_data_type, + u32 *bucket_dirty_sectors, + u32 *bucket_cached_sectors, + u32 *bucket_stripe, + u8 *bucket_stripe_redundancy) { struct bch_fs *c = trans->c; const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; @@ -179,51 +186,44 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, if (deleting) sectors = -sectors; - struct btree_iter iter; - struct bkey_i_alloc_v4 *a = - bch2_trans_start_alloc_update(trans, &iter, PTR_BUCKET_POS(c, ptr)); - ret = PTR_ERR_OR_ZERO(a); - if (ret) - goto err; - if (!deleting) { - if (bch2_trans_inconsistent_on(a->v.stripe || - a->v.stripe_redundancy, trans, + if (bch2_trans_inconsistent_on(*bucket_stripe || + *bucket_stripe_redundancy, trans, "bucket %llu:%llu gen %u data type %s dirty_sectors %u: multiple stripes using same bucket (%u, %llu)\n%s", - iter.pos.inode, iter.pos.offset, a->v.gen, - bch2_data_type_str(a->v.data_type), - a->v.dirty_sectors, - a->v.stripe, s.k->p.offset, + bucket.inode, bucket.offset, bucket_gen, + bch2_data_type_str(*bucket_data_type), + *bucket_dirty_sectors, + *bucket_stripe, s.k->p.offset, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } - if (bch2_trans_inconsistent_on(parity && (a->v.dirty_sectors || a->v.cached_sectors), trans, + if (bch2_trans_inconsistent_on(parity && (*bucket_dirty_sectors || *bucket_cached_sectors), trans, "bucket %llu:%llu gen %u data type %s dirty_sectors %u cached_sectors %u: data already in parity bucket\n%s", - iter.pos.inode, iter.pos.offset, a->v.gen, - bch2_data_type_str(a->v.data_type), - a->v.dirty_sectors, - a->v.cached_sectors, + bucket.inode, bucket.offset, bucket_gen, + bch2_data_type_str(*bucket_data_type), + *bucket_dirty_sectors, + *bucket_cached_sectors, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } } else { - if (bch2_trans_inconsistent_on(a->v.stripe != s.k->p.offset || - a->v.stripe_redundancy != s.v->nr_redundant, trans, + if (bch2_trans_inconsistent_on(*bucket_stripe != s.k->p.offset || + *bucket_stripe_redundancy != s.v->nr_redundant, trans, "bucket %llu:%llu gen %u: not marked as stripe when deleting stripe (got %u)\n%s", - iter.pos.inode, iter.pos.offset, a->v.gen, - a->v.stripe, + bucket.inode, bucket.offset, bucket_gen, + *bucket_stripe, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } - if (bch2_trans_inconsistent_on(a->v.data_type != data_type, trans, + if (bch2_trans_inconsistent_on(*bucket_data_type != data_type, trans, "bucket %llu:%llu gen %u data type %s: wrong data type when stripe, should be %s\n%s", - iter.pos.inode, iter.pos.offset, a->v.gen, - bch2_data_type_str(a->v.data_type), + bucket.inode, bucket.offset, bucket_gen, + bch2_data_type_str(*bucket_data_type), bch2_data_type_str(data_type), (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; @@ -231,12 +231,12 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, } if (bch2_trans_inconsistent_on(parity && - (a->v.dirty_sectors != -sectors || - a->v.cached_sectors), trans, + (*bucket_dirty_sectors != -sectors || + *bucket_cached_sectors), trans, "bucket %llu:%llu gen %u dirty_sectors %u cached_sectors %u: wrong sectors when deleting parity block of stripe\n%s", - iter.pos.inode, iter.pos.offset, a->v.gen, - a->v.dirty_sectors, - a->v.cached_sectors, + bucket.inode, bucket.offset, bucket_gen, + *bucket_dirty_sectors, + *bucket_cached_sectors, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; @@ -244,88 +244,128 @@ static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans, } ret = bch2_check_bucket_ref(trans, s.s_c, ptr, sectors, data_type, - a->v.gen, a->v.data_type, - a->v.dirty_sectors); + bucket_gen, *bucket_data_type, + *bucket_dirty_sectors); if (ret) goto err; - a->v.dirty_sectors += sectors; + *bucket_dirty_sectors += sectors; if (!deleting) { - a->v.stripe = s.k->p.offset; - a->v.stripe_redundancy = s.v->nr_redundant; - a->v.data_type = data_type; + *bucket_stripe = s.k->p.offset; + *bucket_stripe_redundancy = s.v->nr_redundant; + *bucket_data_type = data_type; } else { - a->v.stripe = 0; - a->v.stripe_redundancy = 0; - a->v.data_type = alloc_data_type(a->v, BCH_DATA_user); + *bucket_stripe = 0; + *bucket_stripe_redundancy = 0; + *bucket_data_type = BCH_DATA_free; } - - ret = bch2_trans_update(trans, &iter, &a->k_i, 0); - if (ret) - goto err; err: - bch2_trans_iter_exit(trans, &iter); printbuf_exit(&buf); return ret; } static int mark_stripe_bucket(struct btree_trans *trans, struct bkey_s_c_stripe s, - unsigned ptr_idx, + unsigned ptr_idx, bool deleting, enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; - unsigned nr_data = s.v->nr_blocks - s.v->nr_redundant; - bool parity = ptr_idx >= nr_data; - enum bch_data_type data_type = parity ? BCH_DATA_parity : BCH_DATA_stripe; - s64 sectors = parity ? le16_to_cpu(s.v->sectors) : 0; - struct printbuf buf = PRINTBUF; - int ret = 0; + struct bpos bucket = PTR_BUCKET_POS(c, ptr); - BUG_ON(!(flags & BTREE_TRIGGER_gc)); - - /* * XXX doesn't handle deletion */ + if (flags & BTREE_TRIGGER_transactional) { + struct btree_iter iter; + struct bkey_i_alloc_v4 *a = + bch2_trans_start_alloc_update(trans, &iter, bucket); + int ret = PTR_ERR_OR_ZERO(a) ?: + __mark_stripe_bucket(trans, s, ptr_idx, deleting, iter.pos, + a->v.gen, + &a->v.data_type, + &a->v.dirty_sectors, + &a->v.cached_sectors, + &a->v.stripe, + &a->v.stripe_redundancy); + if (ret) + goto err; - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - struct bucket old, new, *g; + if (deleting) + a->v.data_type = alloc_data_type(a->v, BCH_DATA_user); - percpu_down_read(&c->mark_lock); - g = PTR_GC_BUCKET(ca, ptr); + ret = bch2_trans_update(trans, &iter, &a->k_i, 0); + if (ret) + goto err; +err: + bch2_trans_iter_exit(trans, &iter); + return ret; + } - if (g->dirty_sectors || - (g->stripe && g->stripe != s.k->p.offset)) { - bch2_fs_inconsistent(c, - "bucket %u:%zu gen %u: multiple stripes using same bucket\n%s", - ptr->dev, PTR_BUCKET_NR(ca, ptr), g->gen, - (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf)); - ret = -EINVAL; - goto err; + if (flags & BTREE_TRIGGER_gc) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + + percpu_down_read(&c->mark_lock); + struct bucket *g = gc_bucket(ca, bucket.offset); + + bucket_lock(g); + struct bucket old = *g; + u8 data_type = g->data_type; + + int ret = __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, + g->gen, + &data_type, + &g->dirty_sectors, + &g->cached_sectors, + &g->stripe, + &g->stripe_redundancy); + g->data_type = data_type; + struct bucket new = *g; + bucket_unlock(g); + if (!ret) + bch2_dev_usage_update_m(c, ca, &old, &new); + percpu_up_read(&c->mark_lock); + return ret; } - bucket_lock(g); - old = *g; + BUG(); + return 0; +} - ret = bch2_check_bucket_ref(trans, s.s_c, ptr, sectors, data_type, - g->gen, g->data_type, - g->dirty_sectors); - if (ret) - goto err; +static int mark_stripe_buckets(struct btree_trans *trans, + struct bkey_s_c old, struct bkey_s_c new, + enum btree_iter_update_trigger_flags flags) +{ + const struct bch_stripe *old_s = old.k->type == KEY_TYPE_stripe + ? bkey_s_c_to_stripe(old).v : NULL; + const struct bch_stripe *new_s = new.k->type == KEY_TYPE_stripe + ? bkey_s_c_to_stripe(new).v : NULL; - g->data_type = data_type; - g->dirty_sectors += sectors; + BUG_ON(old_s && new_s && old_s->nr_blocks != new_s->nr_blocks); - g->stripe = s.k->p.offset; - g->stripe_redundancy = s.v->nr_redundant; - new = *g; -err: - bucket_unlock(g); - if (!ret) - bch2_dev_usage_update_m(c, ca, &old, &new); - percpu_up_read(&c->mark_lock); - printbuf_exit(&buf); - return ret; + unsigned nr_blocks = new_s ? new_s->nr_blocks : old_s->nr_blocks; + + for (unsigned i = 0; i < nr_blocks; i++) { + if (new_s && old_s && + !memcmp(&new_s->ptrs[i], + &old_s->ptrs[i], + sizeof(new_s->ptrs[i]))) + continue; + + if (new_s) { + int ret = mark_stripe_bucket(trans, + bkey_s_c_to_stripe(new), i, false, flags); + if (ret) + return ret; + } + + if (old_s) { + int ret = mark_stripe_bucket(trans, + bkey_s_c_to_stripe(old), i, true, flags); + if (ret) + return ret; + } + } + + return 0; } int bch2_trigger_stripe(struct btree_trans *trans, @@ -376,28 +416,9 @@ int bch2_trigger_stripe(struct btree_trans *trans, return ret; } - unsigned nr_blocks = new_s ? new_s->nr_blocks : old_s->nr_blocks; - for (unsigned i = 0; i < nr_blocks; i++) { - if (new_s && old_s && - !memcmp(&new_s->ptrs[i], - &old_s->ptrs[i], - sizeof(new_s->ptrs[i]))) - continue; - - if (new_s) { - int ret = bch2_trans_mark_stripe_bucket(trans, - bkey_s_c_to_stripe(new), i, false); - if (ret) - return ret; - } - - if (old_s) { - int ret = bch2_trans_mark_stripe_bucket(trans, - bkey_s_c_to_stripe(old), i, true); - if (ret) - return ret; - } - } + int ret = mark_stripe_buckets(trans, old, new, flags); + if (ret) + return ret; } if (flags & BTREE_TRIGGER_atomic) { @@ -468,13 +489,11 @@ int bch2_trigger_stripe(struct btree_trans *trans, */ memset(m->block_sectors, 0, sizeof(m->block_sectors)); - for (unsigned i = 0; i < new_s->nr_blocks; i++) { - int ret = mark_stripe_bucket(trans, bkey_s_c_to_stripe(new), i, flags); - if (ret) - return ret; - } + int ret = mark_stripe_buckets(trans, old, new, flags); + if (ret) + return ret; - int ret = bch2_update_replicas(c, new, &m->r.e, + ret = bch2_update_replicas(c, new, &m->r.e, ((s64) m->sectors * m->nr_redundant), 0, true); if (ret) { -- cgit v1.2.3-59-g8ed1b From 70e3e039cf65f67fa3c41b51cb00a58f6cd48886 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 19 Apr 2024 19:03:58 -0400 Subject: bcachefs: bch2_bucket_ref_update() If we hit an inconsistency when updating allocation information, we don't want to fail the update if it's for a deletion - only if it's for a new key. Rename check_bucket_ref() -> bucket_ref_update() so we can centralize the logic to do this. Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 55 ++++++++++++++++++++++++++++++--------------------- fs/bcachefs/buckets.h | 6 +++--- fs/bcachefs/ec.c | 14 ++++++------- 3 files changed, 42 insertions(+), 33 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 36d13819a27d..adaa66f7764a 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -485,19 +485,22 @@ int bch2_update_cached_sectors_list(struct btree_trans *trans, unsigned dev, s64 return bch2_update_replicas_list(trans, &r.e, sectors); } -int bch2_check_bucket_ref(struct btree_trans *trans, +int bch2_bucket_ref_update(struct btree_trans *trans, struct bkey_s_c k, const struct bch_extent_ptr *ptr, s64 sectors, enum bch_data_type ptr_data_type, u8 b_gen, u8 bucket_data_type, - u32 bucket_sectors) + u32 *bucket_sectors) { struct bch_fs *c = trans->c; struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); size_t bucket_nr = PTR_BUCKET_NR(ca, ptr); struct printbuf buf = PRINTBUF; + bool inserting = sectors > 0; int ret = 0; + BUG_ON(!sectors); + if (gen_after(ptr->gen, b_gen)) { bch2_fsck_err(c, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, BCH_FSCK_ERR_ptr_gen_newer_than_bucket_gen, @@ -507,8 +510,9 @@ int bch2_check_bucket_ref(struct btree_trans *trans, bch2_data_type_str(bucket_data_type ?: ptr_data_type), ptr->gen, (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); - ret = -EIO; - goto err; + if (inserting) + goto err; + goto out; } if (gen_cmp(b_gen, ptr->gen) > BUCKET_GC_GEN_MAX) { @@ -521,11 +525,17 @@ int bch2_check_bucket_ref(struct btree_trans *trans, ptr->gen, (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf)); - ret = -EIO; - goto err; + if (inserting) + goto err; + goto out; } - if (b_gen != ptr->gen && !ptr->cached) { + if (b_gen != ptr->gen && ptr->cached) { + ret = 1; + goto out; + } + + if (b_gen != ptr->gen) { bch2_fsck_err(c, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, BCH_FSCK_ERR_stale_dirty_ptr, "bucket %u:%zu gen %u (mem gen %u) data type %s: stale dirty ptr (gen %u)\n" @@ -536,12 +546,8 @@ int bch2_check_bucket_ref(struct btree_trans *trans, ptr->gen, (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf)); - ret = -EIO; - goto err; - } - - if (b_gen != ptr->gen) { - ret = 1; + if (inserting) + goto err; goto out; } @@ -555,28 +561,33 @@ int bch2_check_bucket_ref(struct btree_trans *trans, bch2_data_type_str(ptr_data_type), (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf)); - ret = -EIO; - goto err; + if (inserting) + goto err; + goto out; } - if ((u64) bucket_sectors + sectors > U32_MAX) { + if ((u64) *bucket_sectors + sectors > U32_MAX) { bch2_fsck_err(c, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, BCH_FSCK_ERR_bucket_sector_count_overflow, "bucket %u:%zu gen %u data type %s sector count overflow: %u + %lli > U32_MAX\n" "while marking %s", ptr->dev, bucket_nr, b_gen, bch2_data_type_str(bucket_data_type ?: ptr_data_type), - bucket_sectors, sectors, + *bucket_sectors, sectors, (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, k), buf.buf)); - ret = -EIO; - goto err; + if (inserting) + goto err; + sectors = -*bucket_sectors; } + + *bucket_sectors += sectors; out: printbuf_exit(&buf); return ret; err: bch2_dump_trans_updates(trans); + ret = -EIO; goto out; } @@ -723,14 +734,12 @@ static int __mark_pointer(struct btree_trans *trans, u32 *dst_sectors = !ptr->cached ? dirty_sectors : cached_sectors; - int ret = bch2_check_bucket_ref(trans, k, ptr, sectors, ptr_data_type, - bucket_gen, *bucket_data_type, *dst_sectors); + int ret = bch2_bucket_ref_update(trans, k, ptr, sectors, ptr_data_type, + bucket_gen, *bucket_data_type, dst_sectors); if (ret) return ret; - *dst_sectors += sectors; - if (!*dirty_sectors && !*cached_sectors) *bucket_data_type = 0; else if (*bucket_data_type != BCH_DATA_stripe) diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index a88b9033349f..c2c2ac36bee5 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -337,9 +337,9 @@ int bch2_replicas_deltas_realloc(struct btree_trans *, unsigned); void bch2_fs_usage_initialize(struct bch_fs *); -int bch2_check_bucket_ref(struct btree_trans *, struct bkey_s_c, - const struct bch_extent_ptr *, - s64, enum bch_data_type, u8, u8, u32); +int bch2_bucket_ref_update(struct btree_trans *, struct bkey_s_c, + const struct bch_extent_ptr *, + s64, enum bch_data_type, u8, u8, u32 *); int bch2_trigger_extent(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 2d1b20856841..31db3c8fce50 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -243,13 +243,13 @@ static int __mark_stripe_bucket(struct btree_trans *trans, } } - ret = bch2_check_bucket_ref(trans, s.s_c, ptr, sectors, data_type, - bucket_gen, *bucket_data_type, - *bucket_dirty_sectors); - if (ret) - goto err; - - *bucket_dirty_sectors += sectors; + if (sectors) { + ret = bch2_bucket_ref_update(trans, s.s_c, ptr, sectors, data_type, + bucket_gen, *bucket_data_type, + bucket_dirty_sectors); + if (ret) + goto err; + } if (!deleting) { *bucket_stripe = s.k->p.offset; -- cgit v1.2.3-59-g8ed1b From 930e1a92d62d3f0dbf8d27002c539819738ef6bb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 16 Apr 2024 22:35:02 -0400 Subject: bcachefs: kill gc looping for bucket gens looping when we change a bucket gen is not ideal - it means we risk failing if we'd go into an infinite loop, and it's better to make forward progress even if fsck doesn't fix everything. Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 1 - fs/bcachefs/btree_gc.c | 103 +++++++++---------------------------------------- 2 files changed, 18 insertions(+), 86 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index d8eb5beb5977..2345e090d8a1 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -630,7 +630,6 @@ struct bch_dev { x(clean_shutdown) \ x(fsck_running) \ x(initial_gc_unfixed) \ - x(need_another_gc) \ x(need_delete_dead_snapshots) \ x(error) \ x(topology_error) \ diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index f0adc060e4e2..7fdaf236e8d5 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -627,13 +627,14 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id p.ptr.gen, g->gen, (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) { - if (!p.ptr.cached) { + if (!p.ptr.cached && + (g->data_type != BCH_DATA_btree || + data_type == BCH_DATA_btree)) { g->gen_valid = true; g->gen = p.ptr.gen; g->data_type = 0; g->dirty_sectors = 0; g->cached_sectors = 0; - set_bit(BCH_FS_need_another_gc, &c->flags); } else { do_update = true; } @@ -666,16 +667,19 @@ static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id if (fsck_err_on(bucket_data_type_mismatch(g->data_type, data_type), c, ptr_bucket_data_type_mismatch, - "bucket %u:%zu different types of data in same bucket: %s, %s\n" + "bucket %u:%zu gen %u different types of data in same bucket: %s, %s\n" "while marking %s", - p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), + p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen, bch2_data_type_str(g->data_type), bch2_data_type_str(data_type), (printbuf_reset(&buf), bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) { if (data_type == BCH_DATA_btree) { - g->data_type = data_type; - set_bit(BCH_FS_need_another_gc, &c->flags); + g->gen_valid = true; + g->gen = p.ptr.gen; + g->data_type = data_type; + g->dirty_sectors = 0; + g->cached_sectors = 0; } else { do_update = true; } @@ -749,7 +753,7 @@ restart_drop_ptrs: gen_cmp(p.ptr.gen, g->gen) < 0) || gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX || (g->data_type && - g->data_type != data_type)) { + bucket_data_type(g->data_type) != data_type)) { bch2_bkey_drop_ptr(bkey_i_to_s(new), &entry->ptr); goto restart_drop_ptrs; } @@ -1182,19 +1186,6 @@ static int bch2_gc_start(struct bch_fs *c) return 0; } -static int bch2_gc_reset(struct bch_fs *c) -{ - for_each_member_device(c, ca) { - free_percpu(ca->usage_gc); - ca->usage_gc = NULL; - } - - free_percpu(c->usage_gc); - c->usage_gc = NULL; - - return bch2_gc_start(c); -} - /* returns true if not equal */ static inline bool bch2_alloc_v4_cmp(struct bch_alloc_v4 l, struct bch_alloc_v4 r) @@ -1363,20 +1354,6 @@ static int bch2_gc_alloc_start(struct bch_fs *c) return ret; } -static void bch2_gc_alloc_reset(struct bch_fs *c) -{ - for_each_member_device(c, ca) { - struct bucket_array *buckets = gc_bucket_array(ca); - struct bucket *g; - - for_each_bucket(g, buckets) { - g->data_type = 0; - g->dirty_sectors = 0; - g->cached_sectors = 0; - } - } -} - static int bch2_gc_write_reflink_key(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k, @@ -1469,15 +1446,6 @@ static int bch2_gc_reflink_start(struct bch_fs *c) return ret; } -static void bch2_gc_reflink_reset(struct bch_fs *c) -{ - struct genradix_iter iter; - struct reflink_gc *r; - - genradix_for_each(&c->reflink_gc_table, iter, r) - r->refcount = 0; -} - static int bch2_gc_write_stripes_key(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k) @@ -1541,11 +1509,6 @@ static int bch2_gc_stripes_done(struct bch_fs *c) bch2_gc_write_stripes_key(trans, &iter, k))); } -static void bch2_gc_stripes_reset(struct bch_fs *c) -{ - genradix_free(&c->gc_stripes); -} - /** * bch2_gc - walk _all_ references to buckets, and recompute them: * @@ -1571,7 +1534,6 @@ static void bch2_gc_stripes_reset(struct bch_fs *c) */ static int bch2_gc(struct bch_fs *c, bool initial) { - unsigned iter = 0; int ret; lockdep_assert_held(&c->state_lock); @@ -1585,7 +1547,7 @@ static int bch2_gc(struct bch_fs *c, bool initial) bch2_gc_reflink_start(c); if (ret) goto out; -again: + gc_pos_set(c, gc_phase(GC_PHASE_START)); ret = bch2_mark_superblocks(c); @@ -1597,43 +1559,14 @@ again: c->gc_count++; - if (test_bit(BCH_FS_need_another_gc, &c->flags) || - (!iter && bch2_test_restart_gc)) { - if (iter++ > 2) { - bch_info(c, "Unable to fix bucket gens, looping"); - ret = -EINVAL; - goto out; - } - - /* - * XXX: make sure gens we fixed got saved - */ - bch_info(c, "Second GC pass needed, restarting:"); - clear_bit(BCH_FS_need_another_gc, &c->flags); - __gc_pos_set(c, gc_phase(GC_PHASE_NOT_RUNNING)); - - bch2_gc_stripes_reset(c); - bch2_gc_alloc_reset(c); - bch2_gc_reflink_reset(c); - ret = bch2_gc_reset(c); - if (ret) - goto out; - - /* flush fsck errors, reset counters */ - bch2_flush_fsck_errs(c); - goto again; - } + bch2_journal_block(&c->journal); out: - if (!ret) { - bch2_journal_block(&c->journal); - - ret = bch2_gc_alloc_done(c) ?: - bch2_gc_done(c) ?: - bch2_gc_stripes_done(c) ?: - bch2_gc_reflink_done(c); + ret = bch2_gc_alloc_done(c) ?: + bch2_gc_done(c) ?: + bch2_gc_stripes_done(c) ?: + bch2_gc_reflink_done(c); - bch2_journal_unblock(&c->journal); - } + bch2_journal_unblock(&c->journal); percpu_down_write(&c->mark_lock); /* Indicates that gc is no longer in progress: */ -- cgit v1.2.3-59-g8ed1b From f40d13f94df744185f99732e89bc8d085729ecc9 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 19:07:09 -0400 Subject: bcachefs: Run bch2_check_fix_ptrs() via triggers Currently, the reflink_p gc trigger does repair as well - turning a reflink_p key into an error key if the reflink_v it points to doesn't exist. This won't work with online check/repair, because the repair path once online will be subject to transaction restarts, but BTREE_TRIGGER_gc is not idempotant - we can't run it multiple times if we get a transaction restart. So we need to split these paths; to do so this patch calls check_fix_ptrs() by a new general path - a new trigger type, BTREE_TRIGGER_check_repair. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 29 ++++ fs/bcachefs/btree_cache.h | 3 + fs/bcachefs/btree_gc.c | 337 ++++++---------------------------------------- fs/bcachefs/btree_types.h | 2 + fs/bcachefs/buckets.c | 242 ++++++++++++++++++++++++++++++++- fs/bcachefs/buckets.h | 4 + fs/bcachefs/ec.c | 5 +- 7 files changed, 326 insertions(+), 296 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 02c70e813fac..debdd7dc0477 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -190,6 +190,35 @@ int bch2_btree_node_hash_insert(struct btree_cache *bc, struct btree *b, return ret; } +void bch2_btree_node_update_key_early(struct btree_trans *trans, + enum btree_id btree, unsigned level, + struct bkey_s_c old, struct bkey_i *new) +{ + struct bch_fs *c = trans->c; + struct btree *b; + struct bkey_buf tmp; + int ret; + + bch2_bkey_buf_init(&tmp); + bch2_bkey_buf_reassemble(&tmp, c, old); + + b = bch2_btree_node_get_noiter(trans, tmp.k, btree, level, true); + if (!IS_ERR_OR_NULL(b)) { + mutex_lock(&c->btree_cache.lock); + + bch2_btree_node_hash_remove(&c->btree_cache, b); + + bkey_copy(&b->key, new); + ret = __bch2_btree_node_hash_insert(&c->btree_cache, b); + BUG_ON(ret); + + mutex_unlock(&c->btree_cache.lock); + six_unlock_read(&b->c.lock); + } + + bch2_bkey_buf_exit(&tmp, c); +} + __flatten static inline struct btree *btree_cache_find(struct btree_cache *bc, const struct bkey_i *k) diff --git a/fs/bcachefs/btree_cache.h b/fs/bcachefs/btree_cache.h index 6d33885fdbde..6fe91d1c0fd4 100644 --- a/fs/bcachefs/btree_cache.h +++ b/fs/bcachefs/btree_cache.h @@ -17,6 +17,9 @@ int __bch2_btree_node_hash_insert(struct btree_cache *, struct btree *); int bch2_btree_node_hash_insert(struct btree_cache *, struct btree *, unsigned, enum btree_id); +void bch2_btree_node_update_key_early(struct btree_trans *, enum btree_id, unsigned, + struct bkey_s_c, struct bkey_i *); + void bch2_btree_cache_cannibalize_unlock(struct btree_trans *); int bch2_btree_cache_cannibalize_lock(struct btree_trans *, struct closure *); diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 7fdaf236e8d5..e30fa3c47553 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -91,35 +91,6 @@ static void btree_ptr_to_v2(struct btree *b, struct bkey_i_btree_ptr_v2 *dst) } } -static void bch2_btree_node_update_key_early(struct btree_trans *trans, - enum btree_id btree, unsigned level, - struct bkey_s_c old, struct bkey_i *new) -{ - struct bch_fs *c = trans->c; - struct btree *b; - struct bkey_buf tmp; - int ret; - - bch2_bkey_buf_init(&tmp); - bch2_bkey_buf_reassemble(&tmp, c, old); - - b = bch2_btree_node_get_noiter(trans, tmp.k, btree, level, true); - if (!IS_ERR_OR_NULL(b)) { - mutex_lock(&c->btree_cache.lock); - - bch2_btree_node_hash_remove(&c->btree_cache, b); - - bkey_copy(&b->key, new); - ret = __bch2_btree_node_hash_insert(&c->btree_cache, b); - BUG_ON(ret); - - mutex_unlock(&c->btree_cache.lock); - six_unlock_read(&b->c.lock); - } - - bch2_bkey_buf_exit(&tmp, c); -} - static int set_node_min(struct bch_fs *c, struct btree *b, struct bpos new_min) { struct bkey_i_btree_ptr_v2 *new; @@ -580,245 +551,11 @@ fsck_err: return ret; } -static int bch2_check_fix_ptrs(struct btree_trans *trans, enum btree_id btree_id, - unsigned level, bool is_root, - struct bkey_s_c *k) -{ - struct bch_fs *c = trans->c; - struct bkey_ptrs_c ptrs_c = bch2_bkey_ptrs_c(*k); - const union bch_extent_entry *entry_c; - struct extent_ptr_decoded p = { 0 }; - bool do_update = false; - struct printbuf buf = PRINTBUF; - int ret = 0; - - /* - * XXX - * use check_bucket_ref here - */ - bkey_for_each_ptr_decode(k->k, ptrs_c, p, entry_c) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); - struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); - enum bch_data_type data_type = bch2_bkey_ptr_data_type(*k, p, entry_c); - - if (fsck_err_on(!g->gen_valid, - c, ptr_to_missing_alloc_key, - "bucket %u:%zu data type %s ptr gen %u missing in alloc btree\n" - "while marking %s", - p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), - bch2_data_type_str(ptr_data_type(k->k, &p.ptr)), - p.ptr.gen, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) { - if (!p.ptr.cached) { - g->gen_valid = true; - g->gen = p.ptr.gen; - } else { - do_update = true; - } - } - - if (fsck_err_on(gen_cmp(p.ptr.gen, g->gen) > 0, - c, ptr_gen_newer_than_bucket_gen, - "bucket %u:%zu data type %s ptr gen in the future: %u > %u\n" - "while marking %s", - p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), - bch2_data_type_str(ptr_data_type(k->k, &p.ptr)), - p.ptr.gen, g->gen, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) { - if (!p.ptr.cached && - (g->data_type != BCH_DATA_btree || - data_type == BCH_DATA_btree)) { - g->gen_valid = true; - g->gen = p.ptr.gen; - g->data_type = 0; - g->dirty_sectors = 0; - g->cached_sectors = 0; - } else { - do_update = true; - } - } - - if (fsck_err_on(gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX, - c, ptr_gen_newer_than_bucket_gen, - "bucket %u:%zu gen %u data type %s: ptr gen %u too stale\n" - "while marking %s", - p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen, - bch2_data_type_str(ptr_data_type(k->k, &p.ptr)), - p.ptr.gen, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) - do_update = true; - - if (fsck_err_on(!p.ptr.cached && gen_cmp(p.ptr.gen, g->gen) < 0, - c, stale_dirty_ptr, - "bucket %u:%zu data type %s stale dirty ptr: %u < %u\n" - "while marking %s", - p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), - bch2_data_type_str(ptr_data_type(k->k, &p.ptr)), - p.ptr.gen, g->gen, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) - do_update = true; - - if (data_type != BCH_DATA_btree && p.ptr.gen != g->gen) - continue; - - if (fsck_err_on(bucket_data_type_mismatch(g->data_type, data_type), - c, ptr_bucket_data_type_mismatch, - "bucket %u:%zu gen %u different types of data in same bucket: %s, %s\n" - "while marking %s", - p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen, - bch2_data_type_str(g->data_type), - bch2_data_type_str(data_type), - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) { - if (data_type == BCH_DATA_btree) { - g->gen_valid = true; - g->gen = p.ptr.gen; - g->data_type = data_type; - g->dirty_sectors = 0; - g->cached_sectors = 0; - } else { - do_update = true; - } - } - - if (p.has_ec) { - struct gc_stripe *m = genradix_ptr(&c->gc_stripes, p.ec.idx); - - if (fsck_err_on(!m || !m->alive, c, - ptr_to_missing_stripe, - "pointer to nonexistent stripe %llu\n" - "while marking %s", - (u64) p.ec.idx, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) - do_update = true; - - if (fsck_err_on(m && m->alive && !bch2_ptr_matches_stripe_m(m, p), c, - ptr_to_incorrect_stripe, - "pointer does not match stripe %llu\n" - "while marking %s", - (u64) p.ec.idx, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, *k), buf.buf))) - do_update = true; - } - } - - if (do_update) { - if (is_root) { - bch_err(c, "cannot update btree roots yet"); - ret = -EINVAL; - goto err; - } - - struct bkey_i *new = kmalloc(bkey_bytes(k->k), GFP_KERNEL); - if (!new) { - ret = -BCH_ERR_ENOMEM_gc_repair_key; - bch_err_msg(c, ret, "allocating new key"); - goto err; - } - - bkey_reassemble(new, *k); - - if (level) { - /* - * We don't want to drop btree node pointers - if the - * btree node isn't there anymore, the read path will - * sort it out: - */ - struct bkey_ptrs ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); - bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - struct bucket *g = PTR_GC_BUCKET(ca, ptr); - - ptr->gen = g->gen; - } - } else { - struct bkey_ptrs ptrs; - union bch_extent_entry *entry; -restart_drop_ptrs: - ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); - bkey_for_each_ptr_decode(bkey_i_to_s(new).k, ptrs, p, entry) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); - struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); - enum bch_data_type data_type = bch2_bkey_ptr_data_type(bkey_i_to_s_c(new), p, entry); - - if ((p.ptr.cached && - (!g->gen_valid || gen_cmp(p.ptr.gen, g->gen) > 0)) || - (!p.ptr.cached && - gen_cmp(p.ptr.gen, g->gen) < 0) || - gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX || - (g->data_type && - bucket_data_type(g->data_type) != data_type)) { - bch2_bkey_drop_ptr(bkey_i_to_s(new), &entry->ptr); - goto restart_drop_ptrs; - } - } -again: - ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); - bkey_extent_entry_for_each(ptrs, entry) { - if (extent_entry_type(entry) == BCH_EXTENT_ENTRY_stripe_ptr) { - struct gc_stripe *m = genradix_ptr(&c->gc_stripes, - entry->stripe_ptr.idx); - union bch_extent_entry *next_ptr; - - bkey_extent_entry_for_each_from(ptrs, next_ptr, entry) - if (extent_entry_type(next_ptr) == BCH_EXTENT_ENTRY_ptr) - goto found; - next_ptr = NULL; -found: - if (!next_ptr) { - bch_err(c, "aieee, found stripe ptr with no data ptr"); - continue; - } - - if (!m || !m->alive || - !__bch2_ptr_matches_stripe(&m->ptrs[entry->stripe_ptr.block], - &next_ptr->ptr, - m->sectors)) { - bch2_bkey_extent_entry_drop(new, entry); - goto again; - } - } - } - } - - if (level) - bch2_btree_node_update_key_early(trans, btree_id, level - 1, *k, new); - - if (0) { - printbuf_reset(&buf); - bch2_bkey_val_to_text(&buf, c, *k); - bch_info(c, "updated %s", buf.buf); - - printbuf_reset(&buf); - bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(new)); - bch_info(c, "new key %s", buf.buf); - } - - ret = bch2_journal_key_insert_take(c, btree_id, level, new); - if (ret) { - kfree(new); - goto err; - } - - *k = bkey_i_to_s_c(new); - } -err: -fsck_err: - printbuf_exit(&buf); - return ret; -} - /* marking of btree keys/nodes: */ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, unsigned level, bool is_root, - struct bkey_s_c *k, + struct bkey_s_c k, bool initial) { struct bch_fs *c = trans->c; @@ -827,40 +564,53 @@ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, struct printbuf buf = PRINTBUF; int ret = 0; - deleted.p = k->k->p; + deleted.p = k.k->p; if (initial) { BUG_ON(bch2_journal_seq_verify && - k->k->version.lo > atomic64_read(&c->journal.seq)); + k.k->version.lo > atomic64_read(&c->journal.seq)); - if (fsck_err_on(k->k->version.lo > atomic64_read(&c->key_version), c, + if (fsck_err_on(k.k->version.lo > atomic64_read(&c->key_version), c, bkey_version_in_future, "key version number higher than recorded: %llu > %llu", - k->k->version.lo, + k.k->version.lo, atomic64_read(&c->key_version))) - atomic64_set(&c->key_version, k->k->version.lo); + atomic64_set(&c->key_version, k.k->version.lo); } - ret = bch2_check_fix_ptrs(trans, btree_id, level, is_root, k); - if (ret) - goto err; - - if (mustfix_fsck_err_on(level && !bch2_dev_btree_bitmap_marked(c, *k), + if (mustfix_fsck_err_on(level && !bch2_dev_btree_bitmap_marked(c, k), c, btree_bitmap_not_marked, "btree ptr not marked in member info btree allocated bitmap\n %s", - (bch2_bkey_val_to_text(&buf, c, *k), + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { mutex_lock(&c->sb_lock); - bch2_dev_btree_bitmap_mark(c, *k); + bch2_dev_btree_bitmap_mark(c, k); bch2_write_super(c); mutex_unlock(&c->sb_lock); } - ret = commit_do(trans, NULL, NULL, 0, - bch2_key_trigger(trans, btree_id, level, old, - unsafe_bkey_s_c_to_s(*k), BTREE_TRIGGER_gc)); + /* + * We require a commit before key_trigger() because + * key_trigger(BTREE_TRIGGER_GC) is not idempotant; we'll calculate the + * wrong result if we run it multiple times. + */ + unsigned flags = is_root ? BTREE_TRIGGER_is_root : 0; + + ret = bch2_key_trigger(trans, btree_id, level, old, unsafe_bkey_s_c_to_s(k), + BTREE_TRIGGER_check_repair|flags); + if (ret) + goto out; + + if (trans->nr_updates) { + ret = bch2_trans_commit(trans, NULL, NULL, 0) ?: + -BCH_ERR_transaction_restart_nested; + goto out; + } + + ret = bch2_key_trigger(trans, btree_id, level, old, unsafe_bkey_s_c_to_s(k), + BTREE_TRIGGER_gc|flags); +out: fsck_err: -err: printbuf_exit(&buf); bch_err_fn(c, ret); return ret; @@ -879,7 +629,12 @@ static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool i bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { - ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false, &k, initial); + bch2_trans_begin(trans); + ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false, k, initial); + if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) { + ret = 0; + continue; + } if (ret) break; @@ -918,12 +673,10 @@ static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id, mutex_lock(&c->btree_root_lock); b = bch2_btree_id_root(c, btree_id)->b; - if (!btree_node_fake(b)) { - struct bkey_s_c k = bkey_i_to_s_c(&b->key); - - ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, - true, &k, initial); - } + if (!btree_node_fake(b)) + ret = lockrestart_do(trans, + bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, + true, bkey_i_to_s_c(&b->key), initial)); gc_pos_set(c, gc_pos_btree_root(b->c.btree_id)); mutex_unlock(&c->btree_root_lock); @@ -991,12 +744,10 @@ static int bch2_gc_btree_init(struct btree_trans *trans, if (b->c.level >= target_depth) ret = bch2_gc_btree_init_recurse(trans, b, target_depth); - if (!ret) { - struct bkey_s_c k = bkey_i_to_s_c(&b->key); - - ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, true, - &k, true); - } + if (!ret) + ret = lockrestart_do(trans, + bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, true, + bkey_i_to_s_c(&b->key), true)); six_unlock_read(&b->c.lock); bch_err_fn(c, ret); diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index a973ba6264d3..4ff5213219a5 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -237,9 +237,11 @@ struct btree_node_iter { x(norun) \ x(transactional) \ x(atomic) \ + x(check_repair) \ x(gc) \ x(insert) \ x(overwrite) \ + x(is_root) \ x(bucket_invalidate) enum { diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index adaa66f7764a..e0e991043aa5 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -485,6 +485,241 @@ int bch2_update_cached_sectors_list(struct btree_trans *trans, unsigned dev, s64 return bch2_update_replicas_list(trans, &r.e, sectors); } +int bch2_check_fix_ptrs(struct btree_trans *trans, + enum btree_id btree, unsigned level, struct bkey_s_c k, + enum btree_iter_update_trigger_flags flags) +{ + struct bch_fs *c = trans->c; + struct bkey_ptrs_c ptrs_c = bch2_bkey_ptrs_c(k); + const union bch_extent_entry *entry_c; + struct extent_ptr_decoded p = { 0 }; + bool do_update = false; + struct printbuf buf = PRINTBUF; + int ret = 0; + + percpu_down_read(&c->mark_lock); + + bkey_for_each_ptr_decode(k.k, ptrs_c, p, entry_c) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); + enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry_c); + + if (fsck_err_on(!g->gen_valid, + c, ptr_to_missing_alloc_key, + "bucket %u:%zu data type %s ptr gen %u missing in alloc btree\n" + "while marking %s", + p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), + bch2_data_type_str(ptr_data_type(k.k, &p.ptr)), + p.ptr.gen, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { + if (!p.ptr.cached) { + g->gen_valid = true; + g->gen = p.ptr.gen; + } else { + do_update = true; + } + } + + if (fsck_err_on(gen_cmp(p.ptr.gen, g->gen) > 0, + c, ptr_gen_newer_than_bucket_gen, + "bucket %u:%zu data type %s ptr gen in the future: %u > %u\n" + "while marking %s", + p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), + bch2_data_type_str(ptr_data_type(k.k, &p.ptr)), + p.ptr.gen, g->gen, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { + if (!p.ptr.cached && + (g->data_type != BCH_DATA_btree || + data_type == BCH_DATA_btree)) { + g->gen_valid = true; + g->gen = p.ptr.gen; + g->data_type = 0; + g->dirty_sectors = 0; + g->cached_sectors = 0; + } else { + do_update = true; + } + } + + if (fsck_err_on(gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX, + c, ptr_gen_newer_than_bucket_gen, + "bucket %u:%zu gen %u data type %s: ptr gen %u too stale\n" + "while marking %s", + p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen, + bch2_data_type_str(ptr_data_type(k.k, &p.ptr)), + p.ptr.gen, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + do_update = true; + + if (fsck_err_on(!p.ptr.cached && gen_cmp(p.ptr.gen, g->gen) < 0, + c, stale_dirty_ptr, + "bucket %u:%zu data type %s stale dirty ptr: %u < %u\n" + "while marking %s", + p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), + bch2_data_type_str(ptr_data_type(k.k, &p.ptr)), + p.ptr.gen, g->gen, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + do_update = true; + + if (data_type != BCH_DATA_btree && p.ptr.gen != g->gen) + continue; + + if (fsck_err_on(bucket_data_type_mismatch(g->data_type, data_type), + c, ptr_bucket_data_type_mismatch, + "bucket %u:%zu gen %u different types of data in same bucket: %s, %s\n" + "while marking %s", + p.ptr.dev, PTR_BUCKET_NR(ca, &p.ptr), g->gen, + bch2_data_type_str(g->data_type), + bch2_data_type_str(data_type), + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { + if (data_type == BCH_DATA_btree) { + g->gen_valid = true; + g->gen = p.ptr.gen; + g->data_type = data_type; + g->dirty_sectors = 0; + g->cached_sectors = 0; + } else { + do_update = true; + } + } + + if (p.has_ec) { + struct gc_stripe *m = genradix_ptr(&c->gc_stripes, p.ec.idx); + + if (fsck_err_on(!m || !m->alive, c, + ptr_to_missing_stripe, + "pointer to nonexistent stripe %llu\n" + "while marking %s", + (u64) p.ec.idx, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + do_update = true; + + if (fsck_err_on(m && m->alive && !bch2_ptr_matches_stripe_m(m, p), c, + ptr_to_incorrect_stripe, + "pointer does not match stripe %llu\n" + "while marking %s", + (u64) p.ec.idx, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + do_update = true; + } + } + + if (do_update) { + if (flags & BTREE_TRIGGER_is_root) { + bch_err(c, "cannot update btree roots yet"); + ret = -EINVAL; + goto err; + } + + struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, k); + ret = PTR_ERR_OR_ZERO(new); + if (ret) + goto err; + + if (level) { + /* + * We don't want to drop btree node pointers - if the + * btree node isn't there anymore, the read path will + * sort it out: + */ + struct bkey_ptrs ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); + bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bucket *g = PTR_GC_BUCKET(ca, ptr); + + ptr->gen = g->gen; + } + } else { + struct bkey_ptrs ptrs; + union bch_extent_entry *entry; +restart_drop_ptrs: + ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); + bkey_for_each_ptr_decode(bkey_i_to_s(new).k, ptrs, p, entry) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); + enum bch_data_type data_type = bch2_bkey_ptr_data_type(bkey_i_to_s_c(new), p, entry); + + if ((p.ptr.cached && + (!g->gen_valid || gen_cmp(p.ptr.gen, g->gen) > 0)) || + (!p.ptr.cached && + gen_cmp(p.ptr.gen, g->gen) < 0) || + gen_cmp(g->gen, p.ptr.gen) > BUCKET_GC_GEN_MAX || + (g->data_type && + g->data_type != data_type)) { + bch2_bkey_drop_ptr(bkey_i_to_s(new), &entry->ptr); + goto restart_drop_ptrs; + } + } +again: + ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); + bkey_extent_entry_for_each(ptrs, entry) { + if (extent_entry_type(entry) == BCH_EXTENT_ENTRY_stripe_ptr) { + struct gc_stripe *m = genradix_ptr(&c->gc_stripes, + entry->stripe_ptr.idx); + union bch_extent_entry *next_ptr; + + bkey_extent_entry_for_each_from(ptrs, next_ptr, entry) + if (extent_entry_type(next_ptr) == BCH_EXTENT_ENTRY_ptr) + goto found; + next_ptr = NULL; +found: + if (!next_ptr) { + bch_err(c, "aieee, found stripe ptr with no data ptr"); + continue; + } + + if (!m || !m->alive || + !__bch2_ptr_matches_stripe(&m->ptrs[entry->stripe_ptr.block], + &next_ptr->ptr, + m->sectors)) { + bch2_bkey_extent_entry_drop(new, entry); + goto again; + } + } + } + } + + if (0) { + printbuf_reset(&buf); + bch2_bkey_val_to_text(&buf, c, k); + bch_info(c, "updated %s", buf.buf); + + printbuf_reset(&buf); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(new)); + bch_info(c, "new key %s", buf.buf); + } + + percpu_up_read(&c->mark_lock); + struct btree_iter iter; + bch2_trans_node_iter_init(trans, &iter, btree, new->k.p, 0, level, + BTREE_ITER_intent|BTREE_ITER_all_snapshots); + ret = bch2_btree_iter_traverse(&iter) ?: + bch2_trans_update(trans, &iter, new, + BTREE_UPDATE_internal_snapshot_node| + BTREE_TRIGGER_norun); + bch2_trans_iter_exit(trans, &iter); + percpu_down_read(&c->mark_lock); + + if (ret) + goto err; + + if (level) + bch2_btree_node_update_key_early(trans, btree, level - 1, k, new); + } +err: +fsck_err: + percpu_up_read(&c->mark_lock); + printbuf_exit(&buf); + return ret; +} + int bch2_bucket_ref_update(struct btree_trans *trans, struct bkey_s_c k, const struct bch_extent_ptr *ptr, @@ -970,7 +1205,7 @@ static int __trigger_extent(struct btree_trans *trans, } int bch2_trigger_extent(struct btree_trans *trans, - enum btree_id btree_id, unsigned level, + enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s new, enum btree_iter_update_trigger_flags flags) { @@ -979,6 +1214,9 @@ int bch2_trigger_extent(struct btree_trans *trans, unsigned new_ptrs_bytes = (void *) new_ptrs.end - (void *) new_ptrs.start; unsigned old_ptrs_bytes = (void *) old_ptrs.end - (void *) old_ptrs.start; + if (unlikely(flags & BTREE_TRIGGER_check_repair)) + return bch2_check_fix_ptrs(trans, btree, level, new.s_c, flags); + /* if pointers aren't changing - nothing to do: */ if (new_ptrs_bytes == old_ptrs_bytes && !memcmp(new_ptrs.start, @@ -1000,7 +1238,7 @@ int bch2_trigger_extent(struct btree_trans *trans, } if (flags & (BTREE_TRIGGER_transactional|BTREE_TRIGGER_gc)) - return trigger_run_overwrite_then_insert(__trigger_extent, trans, btree_id, level, old, new, flags); + return trigger_run_overwrite_then_insert(__trigger_extent, trans, btree, level, old, new, flags); return 0; } diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index c2c2ac36bee5..df10f390101a 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -341,6 +341,10 @@ int bch2_bucket_ref_update(struct btree_trans *, struct bkey_s_c, const struct bch_extent_ptr *, s64, enum bch_data_type, u8, u8, u32 *); +int bch2_check_fix_ptrs(struct btree_trans *, + enum btree_id, unsigned, struct bkey_s_c, + enum btree_iter_update_trigger_flags); + int bch2_trigger_extent(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, enum btree_iter_update_trigger_flags); diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 31db3c8fce50..d6fa7e0bb266 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -369,7 +369,7 @@ static int mark_stripe_buckets(struct btree_trans *trans, } int bch2_trigger_stripe(struct btree_trans *trans, - enum btree_id btree_id, unsigned level, + enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s _new, enum btree_iter_update_trigger_flags flags) { @@ -381,6 +381,9 @@ int bch2_trigger_stripe(struct btree_trans *trans, const struct bch_stripe *new_s = new.k->type == KEY_TYPE_stripe ? bkey_s_c_to_stripe(new).v : NULL; + if (unlikely(flags & BTREE_TRIGGER_check_repair)) + return bch2_check_fix_ptrs(trans, btree, level, _new.s_c, flags); + if (flags & BTREE_TRIGGER_transactional) { /* * If the pointers aren't changing, we don't need to do anything: -- cgit v1.2.3-59-g8ed1b From c451986bf4c64e1f21932117ec6ade269ca825db Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 7 Apr 2024 19:47:31 -0400 Subject: bcachefs: do reflink_p repair from BTREE_TRIGGER_check_repair Signed-off-by: Kent Overstreet --- fs/bcachefs/reflink.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index a0cf5d97513f..a8a761d59a5d 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -164,10 +164,13 @@ static s64 gc_trigger_reflink_p_segment(struct btree_trans *trans, BUG_ON((s64) r->refcount + add < 0); - r->refcount += add; + if (flags & BTREE_TRIGGER_gc) + r->refcount += add; *idx = r->offset; return 0; not_found: + BUG_ON(!(flags & BTREE_TRIGGER_check_repair)); + if (fsck_err(c, reflink_p_to_missing_reflink_v, "pointer to missing indirect extent\n" " %s\n" @@ -216,7 +219,7 @@ static int __trigger_reflink_p(struct btree_trans *trans, ret = trans_trigger_reflink_p_segment(trans, p, &idx, flags); } - if (flags & BTREE_TRIGGER_gc) { + if (flags & (BTREE_TRIGGER_check_repair|BTREE_TRIGGER_gc)) { size_t l = 0, r = c->reflink_gc_nr; while (l < r) { -- cgit v1.2.3-59-g8ed1b From 24b27975a9866f32abb46b74834e963700624fcd Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 23:58:01 -0400 Subject: bcachefs: Kill gc_init_recurse() This unifies the online and offline btree gc passes; we're not yet running it online. We now iterate over one level of the btree at a time - the same as check_extents_to_backpointers(); this ordering preserves order of keys regardless of btree splits and merges, which will be important when we re-enable online gc. Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 2 +- fs/bcachefs/btree_gc.c | 181 ++++++++++++------------------------------------- fs/bcachefs/btree_gc.h | 30 +++----- 3 files changed, 55 insertions(+), 158 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 2345e090d8a1..82544f826c58 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -519,8 +519,8 @@ enum gc_phase { struct gc_pos { enum gc_phase phase; + u16 level; struct bpos pos; - unsigned level; }; struct reflink_gc { diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index e30fa3c47553..9694ffb0b098 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -63,7 +63,7 @@ static inline void __gc_pos_set(struct bch_fs *c, struct gc_pos new_pos) static inline void gc_pos_set(struct bch_fs *c, struct gc_pos new_pos) { - BUG_ON(gc_pos_cmp(new_pos, c->gc_pos) <= 0); + BUG_ON(gc_pos_cmp(new_pos, c->gc_pos) < 0); __gc_pos_set(c, new_pos); } @@ -554,11 +554,24 @@ fsck_err: /* marking of btree keys/nodes: */ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, - unsigned level, bool is_root, - struct bkey_s_c k, + unsigned level, struct btree **prev, + struct btree_iter *iter, struct bkey_s_c k, bool initial) { struct bch_fs *c = trans->c; + + if (iter) { + struct btree_path *path = btree_iter_path(trans, iter); + struct btree *b = path_l(path)->b; + + if (*prev != b) { + int ret = bch2_btree_node_check_topology(trans, b); + if (ret) + return ret; + } + *prev = b; + } + struct bkey deleted = KEY(0, 0, 0); struct bkey_s_c old = (struct bkey_s_c) { &deleted, NULL }; struct printbuf buf = PRINTBUF; @@ -594,7 +607,7 @@ static int bch2_gc_mark_key(struct btree_trans *trans, enum btree_id btree_id, * key_trigger(BTREE_TRIGGER_GC) is not idempotant; we'll calculate the * wrong result if we run it multiple times. */ - unsigned flags = is_root ? BTREE_TRIGGER_is_root : 0; + unsigned flags = !iter ? BTREE_TRIGGER_is_root : 0; ret = bch2_key_trigger(trans, btree_id, level, old, unsafe_bkey_s_c_to_s(k), BTREE_TRIGGER_check_repair|flags); @@ -616,151 +629,55 @@ fsck_err: return ret; } -static int btree_gc_mark_node(struct btree_trans *trans, struct btree *b, bool initial) -{ - struct btree_and_journal_iter iter; - struct bkey_s_c k; - int ret = 0; - - ret = bch2_btree_node_check_topology(trans, b); - if (ret) - return ret; - - bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); - - while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { - bch2_trans_begin(trans); - ret = bch2_gc_mark_key(trans, b->c.btree_id, b->c.level, false, k, initial); - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) { - ret = 0; - continue; - } - if (ret) - break; - - bch2_btree_and_journal_iter_advance(&iter); - } - - bch2_btree_and_journal_iter_exit(&iter); - return ret; -} - -static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree_id, - bool initial) +static int bch2_gc_btree(struct btree_trans *trans, enum btree_id btree, bool initial) { struct bch_fs *c = trans->c; - struct btree_iter iter; - struct btree *b; - unsigned target_depth = btree_type_has_ptrs(btree_id) ? 0 : 1; + int level = 0, target_depth = btree_node_type_needs_gc(__btree_node_type(0, btree)) ? 0 : 1; int ret = 0; - gc_pos_set(c, gc_pos_btree(btree_id, POS_MIN, 0)); - - __for_each_btree_node(trans, iter, btree_id, POS_MIN, - 0, target_depth, BTREE_ITER_prefetch, b, ret) { - bch2_verify_btree_nr_keys(b); - - gc_pos_set(c, gc_pos_btree_node(b)); - - ret = btree_gc_mark_node(trans, b, initial); - if (ret) - break; - } - bch2_trans_iter_exit(trans, &iter); - - if (ret) - return ret; + /* We need to make sure every leaf node is readable before going RW */ + if (initial) + target_depth = 0; + /* root */ mutex_lock(&c->btree_root_lock); - b = bch2_btree_id_root(c, btree_id)->b; - if (!btree_node_fake(b)) + struct btree *b = bch2_btree_id_root(c, btree)->b; + if (!btree_node_fake(b)) { + gc_pos_set(c, gc_pos_btree(btree, b->c.level + 1, SPOS_MAX)); ret = lockrestart_do(trans, bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, - true, bkey_i_to_s_c(&b->key), initial)); - gc_pos_set(c, gc_pos_btree_root(b->c.btree_id)); + NULL, NULL, bkey_i_to_s_c(&b->key), initial)); + level = b->c.level; + } mutex_unlock(&c->btree_root_lock); - return ret; -} - -static int bch2_gc_btree_init_recurse(struct btree_trans *trans, struct btree *b, - unsigned target_depth) -{ - int ret = btree_gc_mark_node(trans, b, true); if (ret) return ret; - if (b->c.level > target_depth) { - struct bch_fs *c = trans->c; - struct btree_and_journal_iter iter; - struct bkey_s_c k; - struct bkey_buf cur; - - bch2_bkey_buf_init(&cur); - bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); - iter.prefetch = true; - - while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { - bch2_bkey_buf_reassemble(&cur, c, k); - bch2_btree_and_journal_iter_advance(&iter); - - struct btree *child = - bch2_btree_node_get_noiter(trans, cur.k, - b->c.btree_id, b->c.level - 1, - false); - ret = PTR_ERR_OR_ZERO(child); - bch_err_msg(c, ret, "getting btree node"); - if (ret) - break; - - ret = bch2_gc_btree_init_recurse(trans, child, target_depth); - six_unlock_read(&child->c.lock); + for (; level >= target_depth; --level) { + struct btree *prev = NULL; + struct btree_iter iter; + bch2_trans_node_iter_init(trans, &iter, btree, POS_MIN, 0, level, + BTREE_ITER_prefetch); - if (ret) - break; - } - - bch2_bkey_buf_exit(&cur, c); - bch2_btree_and_journal_iter_exit(&iter); + ret = for_each_btree_key_continue(trans, iter, 0, k, ({ + gc_pos_set(c, gc_pos_btree(btree, level, k.k->p)); + bch2_gc_mark_key(trans, btree, level, &prev, &iter, k, initial); + })); + if (ret) + break; } return ret; } -static int bch2_gc_btree_init(struct btree_trans *trans, - enum btree_id btree_id) -{ - struct bch_fs *c = trans->c; - /* - * We need to make sure every leaf node is readable before going RW - unsigned target_depth = btree_node_type_needs_gc(__btree_node_type(0, btree_id)) ? 0 : 1; - */ - unsigned target_depth = 0; - int ret = 0; - - struct btree *b = bch2_btree_id_root(c, btree_id)->b; - - six_lock_read(&b->c.lock, NULL, NULL); - if (b->c.level >= target_depth) - ret = bch2_gc_btree_init_recurse(trans, b, target_depth); - - if (!ret) - ret = lockrestart_do(trans, - bch2_gc_mark_key(trans, b->c.btree_id, b->c.level + 1, true, - bkey_i_to_s_c(&b->key), true)); - six_unlock_read(&b->c.lock); - - bch_err_fn(c, ret); - return ret; -} - static inline int btree_id_gc_phase_cmp(enum btree_id l, enum btree_id r) { return (int) btree_id_to_gc_phase(l) - (int) btree_id_to_gc_phase(r); } -static int bch2_gc_btrees(struct bch_fs *c, bool initial) +static int bch2_gc_btrees(struct bch_fs *c) { struct btree_trans *trans = bch2_trans_get(c); enum btree_id ids[BTREE_ID_NR]; @@ -777,9 +694,7 @@ static int bch2_gc_btrees(struct bch_fs *c, bool initial) if (IS_ERR_OR_NULL(bch2_btree_id_root(c, btree)->b)) continue; - ret = initial - ? bch2_gc_btree_init(trans, btree) - : bch2_gc_btree(trans, btree, initial); + ret = bch2_gc_btree(trans, btree, true); if (mustfix_fsck_err_on(bch2_err_matches(ret, EIO), c, btree_node_read_error, @@ -1261,10 +1176,9 @@ static int bch2_gc_stripes_done(struct bch_fs *c) } /** - * bch2_gc - walk _all_ references to buckets, and recompute them: + * bch2_check_allocations - walk all references to buckets, and recompute them: * * @c: filesystem object - * @initial: are we in recovery? * * Returns: 0 on success, or standard errcode on failure * @@ -1283,7 +1197,7 @@ static int bch2_gc_stripes_done(struct bch_fs *c) * move around - if references move backwards in the ordering GC * uses, GC could skip past them */ -static int bch2_gc(struct bch_fs *c, bool initial) +int bch2_check_allocations(struct bch_fs *c) { int ret; @@ -1304,7 +1218,7 @@ static int bch2_gc(struct bch_fs *c, bool initial) ret = bch2_mark_superblocks(c); BUG_ON(ret); - ret = bch2_gc_btrees(c, initial); + ret = bch2_gc_btrees(c); if (ret) goto out; @@ -1337,11 +1251,6 @@ out: return ret; } -int bch2_check_allocations(struct bch_fs *c) -{ - return bch2_gc(c, true); -} - static int gc_btree_gens_key(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k) diff --git a/fs/bcachefs/btree_gc.h b/fs/bcachefs/btree_gc.h index 15315aab93bd..1b6489d8e0f4 100644 --- a/fs/bcachefs/btree_gc.h +++ b/fs/bcachefs/btree_gc.h @@ -34,16 +34,16 @@ static inline struct gc_pos gc_phase(enum gc_phase phase) { return (struct gc_pos) { .phase = phase, - .pos = POS_MIN, .level = 0, + .pos = POS_MIN, }; } static inline int gc_pos_cmp(struct gc_pos l, struct gc_pos r) { - return cmp_int(l.phase, r.phase) ?: - bpos_cmp(l.pos, r.pos) ?: - cmp_int(l.level, r.level); + return cmp_int(l.phase, r.phase) ?: + -cmp_int(l.level, r.level) ?: + bpos_cmp(l.pos, r.pos); } static inline enum gc_phase btree_id_to_gc_phase(enum btree_id id) @@ -57,13 +57,13 @@ static inline enum gc_phase btree_id_to_gc_phase(enum btree_id id) } } -static inline struct gc_pos gc_pos_btree(enum btree_id id, - struct bpos pos, unsigned level) +static inline struct gc_pos gc_pos_btree(enum btree_id btree, unsigned level, + struct bpos pos) { return (struct gc_pos) { - .phase = btree_id_to_gc_phase(id), - .pos = pos, + .phase = btree_id_to_gc_phase(btree), .level = level, + .pos = pos, }; } @@ -73,19 +73,7 @@ static inline struct gc_pos gc_pos_btree(enum btree_id id, */ static inline struct gc_pos gc_pos_btree_node(struct btree *b) { - return gc_pos_btree(b->c.btree_id, b->key.k.p, b->c.level); -} - -/* - * GC position of the pointer to a btree root: we don't use - * gc_pos_pointer_to_btree_node() here to avoid a potential race with - * btree_split() increasing the tree depth - the new root will have level > the - * old root and thus have a greater gc position than the old root, but that - * would be incorrect since once gc has marked the root it's not coming back. - */ -static inline struct gc_pos gc_pos_btree_root(enum btree_id id) -{ - return gc_pos_btree(id, SPOS_MAX, BTREE_MAX_DEPTH); + return gc_pos_btree(b->c.btree_id, b->c.level, b->key.k.p); } static inline bool gc_visited(struct bch_fs *c, struct gc_pos pos) -- cgit v1.2.3-59-g8ed1b From 2d288745ebc202d7ac4905cba4a540549dc8044e Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 23 Apr 2024 11:58:09 -0700 Subject: bcachefs: Fix type of flags parameter for some ->trigger() implementations When building with clang's -Wincompatible-function-pointer-types-strict (a warning designed to catch potential kCFI failures at build time), there are several warnings along the lines of: fs/bcachefs/bkey_methods.c:118:2: error: incompatible function pointer types initializing 'int (*)(struct btree_trans *, enum btree_id, unsigned int, struct bkey_s_c, struct bkey_s, enum btree_iter_update_trigger_flags)' with an expression of type 'int (struct btree_trans *, enum btree_id, unsigned int, struct bkey_s_c, struct bkey_s, unsigned int)' [-Werror,-Wincompatible-function-pointer-types-strict] 118 | BCH_BKEY_TYPES() | ^~~~~~~~~~~~~~~~ fs/bcachefs/bcachefs_format.h:394:2: note: expanded from macro 'BCH_BKEY_TYPES' 394 | x(inode, 8) \ | ^~~~~~~~~~~~~~~~~~~~~~~~~~ fs/bcachefs/bkey_methods.c:117:41: note: expanded from macro 'x' 117 | #define x(name, nr) [KEY_TYPE_##name] = bch2_bkey_ops_##name, | ^~~~~~~~~~~~~~~~~~~~ :277:1: note: expanded from here 277 | bch2_bkey_ops_inode | ^~~~~~~~~~~~~~~~~~~ fs/bcachefs/inode.h:26:13: note: expanded from macro 'bch2_bkey_ops_inode' 26 | .trigger = bch2_trigger_inode, \ | ^~~~~~~~~~~~~~~~~~ There are several functions that did not have their flags parameter converted to 'enum btree_iter_update_trigger_flags' in the recent unification, which will cause kCFI failures at runtime because the types, while ABI compatible (hence no warning from the non-strict version of this warning), do not match exactly. Fix up these functions (as well as a few other obvious functions that should have it, even if there are no warnings currently) to resolve the warnings and potential kCFI runtime failures. Fixes: 31e4ef3280c8 ("bcachefs: iter/update/trigger/str_hash flag cleanup") Signed-off-by: Nathan Chancellor Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 2 +- fs/bcachefs/alloc_background.h | 3 ++- fs/bcachefs/inode.c | 2 +- fs/bcachefs/inode.h | 3 ++- fs/bcachefs/reflink.c | 6 ++++-- fs/bcachefs/reflink.h | 2 +- fs/bcachefs/snapshot.c | 4 ++-- fs/bcachefs/snapshot.h | 3 ++- fs/bcachefs/subvolume.c | 2 +- fs/bcachefs/subvolume.h | 3 ++- 10 files changed, 18 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 6fa51ee16cc3..63bcba995ca2 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -724,7 +724,7 @@ static noinline int bch2_bucket_gen_update(struct btree_trans *trans, int bch2_trigger_alloc(struct btree_trans *trans, enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; int ret = 0; diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index e94432ced6a6..e0e0d72db03b 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -245,7 +245,8 @@ static inline bool bkey_is_alloc(const struct bkey *k) int bch2_alloc_read(struct bch_fs *); int bch2_trigger_alloc(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); int bch2_check_alloc_info(struct bch_fs *); int bch2_check_alloc_to_lru_refs(struct bch_fs *); void bch2_do_discards(struct bch_fs *); diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index fc8b30441bbc..91572e05d5f0 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -594,7 +594,7 @@ int bch2_trigger_inode(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { s64 nr = (s64) bkey_is_inode(new.k) - (s64) bkey_is_inode(old.k); diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h index 08e8505dfa33..46477a40846c 100644 --- a/fs/bcachefs/inode.h +++ b/fs/bcachefs/inode.h @@ -18,7 +18,8 @@ int bch2_inode_v3_invalid(struct bch_fs *, struct bkey_s_c, void bch2_inode_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_inode(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); #define bch2_bkey_ops_inode ((struct bkey_ops) { \ .key_invalid = bch2_inode_invalid, \ diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index a8a761d59a5d..fb31a68148ff 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -285,7 +285,9 @@ bool bch2_reflink_v_merge(struct bch_fs *c, struct bkey_s _l, struct bkey_s_c _r } #endif -static inline void check_indirect_extent_deleting(struct bkey_s new, unsigned *flags) +static inline void +check_indirect_extent_deleting(struct bkey_s new, + enum btree_iter_update_trigger_flags *flags) { if ((*flags & BTREE_TRIGGER_insert) && !*bkey_refcount(new)) { new.k->type = KEY_TYPE_deleted; @@ -330,7 +332,7 @@ void bch2_indirect_inline_data_to_text(struct printbuf *out, int bch2_trigger_indirect_inline_data(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { check_indirect_extent_deleting(new, &flags); diff --git a/fs/bcachefs/reflink.h b/fs/bcachefs/reflink.h index e34c4bdf7a76..59d4dbae3fb1 100644 --- a/fs/bcachefs/reflink.h +++ b/fs/bcachefs/reflink.h @@ -44,7 +44,7 @@ void bch2_indirect_inline_data_to_text(struct printbuf *, int bch2_trigger_indirect_inline_data(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, - unsigned); + enum btree_iter_update_trigger_flags); #define bch2_bkey_ops_indirect_inline_data ((struct bkey_ops) { \ .key_invalid = bch2_indirect_inline_data_invalid, \ diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 0b26dee17a5a..2368218070d4 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -298,7 +298,7 @@ static void set_is_ancestor_bitmap(struct bch_fs *c, u32 id) static int __bch2_mark_snapshot(struct btree_trans *trans, enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s_c new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; struct snapshot_t *t; @@ -352,7 +352,7 @@ err: int bch2_mark_snapshot(struct btree_trans *trans, enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { return __bch2_mark_snapshot(trans, btree, level, old, new.s_c, flags); } diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index ccb1f309b8d1..b3c43d579540 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -22,7 +22,8 @@ void bch2_snapshot_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_snapshot_invalid(struct bch_fs *, struct bkey_s_c, enum bkey_invalid_flags, struct printbuf *); int bch2_mark_snapshot(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); #define bch2_bkey_ops_snapshot ((struct bkey_ops) { \ .key_invalid = bch2_snapshot_invalid, \ diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index 33292dae4d50..3341285e7531 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -245,7 +245,7 @@ static int subvolume_children_mod(struct btree_trans *trans, struct bpos pos, bo int bch2_subvolume_trigger(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { if (flags & BTREE_TRIGGER_transactional) { struct bpos children_pos_old = subvolume_children_pos(old); diff --git a/fs/bcachefs/subvolume.h b/fs/bcachefs/subvolume.h index d2015d549bd2..1530a10327ab 100644 --- a/fs/bcachefs/subvolume.h +++ b/fs/bcachefs/subvolume.h @@ -14,7 +14,8 @@ int bch2_subvolume_invalid(struct bch_fs *, struct bkey_s_c, enum bkey_invalid_flags, struct printbuf *); void bch2_subvolume_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_subvolume_trigger(struct btree_trans *, enum btree_id, unsigned, - struct bkey_s_c, struct bkey_s, unsigned); + struct bkey_s_c, struct bkey_s, + enum btree_iter_update_trigger_flags); #define bch2_bkey_ops_subvolume ((struct bkey_ops) { \ .key_invalid = bch2_subvolume_invalid, \ -- cgit v1.2.3-59-g8ed1b From 8bb0eddbbcc99be5551521a089662f3b5b5a3204 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 23 Apr 2024 13:12:27 -0700 Subject: bcachefs: Fix format specifiers in bch2_btree_key_cache_to_text() When building for a 32-bit target, for which 'size_t' is 'unsigned int', there are two warnings around mismatched format specifiers and argument types: In file included from fs/bcachefs/vstructs.h:5, from fs/bcachefs/bcachefs_format.h:79, from fs/bcachefs/bcachefs.h:207, from fs/bcachefs/btree_key_cache.c:3: fs/bcachefs/btree_key_cache.c: In function 'bch2_btree_key_cache_to_text': fs/bcachefs/btree_key_cache.c:1046:25: error: format '%lu' expects argument of type 'long unsigned int', but argument 3 has type 'size_t' {aka 'unsigned int'} [-Werror=format=] 1046 | prt_printf(out, "nonpcpu freelist:\t%lu\r\n", bc->nr_freed_nonpcpu); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~ | | | size_t {aka unsigned int} fs/bcachefs/util.h:192:63: note: in definition of macro 'prt_printf' 192 | #define prt_printf(_out, ...) bch2_prt_printf(_out, __VA_ARGS__) | ^~~~~~~~~~~ fs/bcachefs/btree_key_cache.c:1046:47: note: format string is defined here 1046 | prt_printf(out, "nonpcpu freelist:\t%lu\r\n", bc->nr_freed_nonpcpu); | ~~^ | | | long unsigned int | %u fs/bcachefs/btree_key_cache.c:1047:25: error: format '%lu' expects argument of type 'long unsigned int', but argument 3 has type 'size_t' {aka 'unsigned int'} [-Werror=format=] 1047 | prt_printf(out, "pcpu freelist:\t%lu\r\n", bc->nr_freed_pcpu); | ^~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~ | | | size_t {aka unsigned int} fs/bcachefs/util.h:192:63: note: in definition of macro 'prt_printf' 192 | #define prt_printf(_out, ...) bch2_prt_printf(_out, __VA_ARGS__) | ^~~~~~~~~~~ fs/bcachefs/btree_key_cache.c:1047:44: note: format string is defined here 1047 | prt_printf(out, "pcpu freelist:\t%lu\r\n", bc->nr_freed_pcpu); | ~~^ | | | long unsigned int | %u cc1: all warnings being treated as error Use the proper 'size_t' specifier, '%zu', to clear up the warnings for these platforms. Fixes: f2d47ec26af5 ("bcachefs: Btree key cache instrumentation") Signed-off-by: Nathan Chancellor Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 203fbb38e9d4..e2eca7dab70e 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -1045,8 +1045,8 @@ void bch2_btree_key_cache_to_text(struct printbuf *out, struct btree_key_cache * prt_printf(out, "keys:\t%lu\r\n", atomic_long_read(&bc->nr_keys)); prt_printf(out, "dirty:\t%lu\r\n", atomic_long_read(&bc->nr_dirty)); prt_printf(out, "freelist:\t%lu\r\n", atomic_long_read(&bc->nr_freed)); - prt_printf(out, "nonpcpu freelist:\t%lu\r\n", bc->nr_freed_nonpcpu); - prt_printf(out, "pcpu freelist:\t%lu\r\n", bc->nr_freed_pcpu); + prt_printf(out, "nonpcpu freelist:\t%zu\r\n", bc->nr_freed_nonpcpu); + prt_printf(out, "pcpu freelist:\t%zu\r\n", bc->nr_freed_pcpu); prt_printf(out, "\nshrinker:\n"); prt_printf(out, "requested_to_free:\t%lu\r\n", bc->requested_to_free); -- cgit v1.2.3-59-g8ed1b From 56522d7276746bdeeea095883ba7022f0cb4daf0 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 25 Apr 2024 02:15:08 -0400 Subject: bcachefs: fix btree_path_clone() ip_allocated Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 27f3935d67cb..ea70e9efb91c 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1233,11 +1233,14 @@ static inline void btree_path_copy(struct btree_trans *trans, struct btree_path } static btree_path_idx_t btree_path_clone(struct btree_trans *trans, btree_path_idx_t src, - bool intent) + bool intent, unsigned long ip) { btree_path_idx_t new = btree_path_alloc(trans, src); btree_path_copy(trans, trans->paths + new, trans->paths + src); __btree_path_get(trans->paths + new, intent); +#ifdef TRACK_PATH_ALLOCATED + trans->paths[new].ip_allocated = ip; +#endif return new; } @@ -1246,7 +1249,7 @@ btree_path_idx_t __bch2_btree_path_make_mut(struct btree_trans *trans, btree_path_idx_t path, bool intent, unsigned long ip) { __btree_path_put(trans->paths + path, intent); - path = btree_path_clone(trans, path, intent); + path = btree_path_clone(trans, path, intent, ip); trans->paths[path].preserve = false; return path; } @@ -2491,7 +2494,8 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) bch2_path_put_nokeep(trans, saved_path, iter->flags & BTREE_ITER_intent); saved_path = btree_path_clone(trans, iter->path, - iter->flags & BTREE_ITER_intent); + iter->flags & BTREE_ITER_intent, + _THIS_IP_); path = btree_iter_path(trans, iter); saved_k = *k.k; saved_v = k.v; -- cgit v1.2.3-59-g8ed1b From 0af0b963b52b45513dcdabae9b77575befea738d Mon Sep 17 00:00:00 2001 From: Hongbo Li Date: Fri, 26 Apr 2024 11:21:35 +0800 Subject: bcachefs: eliminate the uninitialized compilation warning in bch2_reconstruct_snapshots MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When compiling the bcachefs-tools, the following compilation warning is reported: libbcachefs/snapshot.c: In function ‘bch2_reconstruct_snapshots’: libbcachefs/snapshot.c:915:19: warning: ‘tree_id’ may be used uninitialized in this function [-Wmaybe-uninitialized] 915 | snapshot->v.tree = cpu_to_le32(tree_id); libbcachefs/snapshot.c:903:6: note: ‘tree_id’ was declared here 903 | u32 tree_id; | ^~~~~~~ This is a false alert, because @tree_id is changed in bch2_snapshot_tree_create after it returns 0. And if this function returns other value, @tree_id wouldn't be used. Thus there should be nothing wrong in logical. Although the report itself is a false alert, we can still make it more explicit by setting the initial value of @tree_id to 0 (an invalid tree ID). Fixes: a292be3b68f3 ("bcachefs: Reconstruct missing snapshot nodes") Signed-off-by: Hongbo Li Signed-off-by: Kent Overstreet --- fs/bcachefs/snapshot.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 2368218070d4..3aa4686cc71f 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -900,7 +900,8 @@ static int check_snapshot_exists(struct btree_trans *trans, u32 id) if (bch2_snapshot_equiv(c, id)) return 0; - u32 tree_id; + /* 0 is an invalid tree ID */ + u32 tree_id = 0; int ret = bch2_snapshot_tree_create(trans, id, 0, &tree_id); if (ret) return ret; -- cgit v1.2.3-59-g8ed1b From c74954135353fac7c206ec15d71c95f981fd0e9d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 25 Apr 2024 02:20:20 -0400 Subject: bcachefs: uninline set_btree_iter_dontneed() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 2 +- fs/bcachefs/alloc_foreground.c | 8 ++++---- fs/bcachefs/btree_iter.c | 13 +++++++++++++ fs/bcachefs/btree_iter.h | 13 +------------ fs/bcachefs/btree_key_cache.c | 2 +- 5 files changed, 20 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 63bcba995ca2..37e8446b3039 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1303,7 +1303,7 @@ static noinline_for_stack int bch2_check_discard_freespace_key(struct btree_tran goto delete; out: fsck_err: - set_btree_iter_dontneed(&alloc_iter); + bch2_set_btree_iter_dontneed(&alloc_iter); bch2_trans_iter_exit(trans, &alloc_iter); printbuf_exit(&buf); return ret; diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 847000a7ead2..438e00c8316b 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -363,10 +363,10 @@ static struct open_bucket *try_alloc_bucket(struct btree_trans *trans, struct bc ob = __try_alloc_bucket(c, ca, b, watermark, a, s, cl); if (!ob) - set_btree_iter_dontneed(&iter); + bch2_set_btree_iter_dontneed(&iter); err: if (iter.path) - set_btree_iter_dontneed(&iter); + bch2_set_btree_iter_dontneed(&iter); bch2_trans_iter_exit(trans, &iter); printbuf_exit(&buf); return ob; @@ -433,7 +433,7 @@ again: ob = __try_alloc_bucket(trans->c, ca, k.k->p.offset, watermark, a, s, cl); next: - set_btree_iter_dontneed(&citer); + bch2_set_btree_iter_dontneed(&citer); bch2_trans_iter_exit(trans, &citer); if (ob) break; @@ -488,7 +488,7 @@ again: ob = try_alloc_bucket(trans, ca, watermark, alloc_cursor, s, k, cl); if (ob) { - set_btree_iter_dontneed(&iter); + bch2_set_btree_iter_dontneed(&iter); break; } } diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index ea70e9efb91c..7f0eb5d396a2 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1822,6 +1822,19 @@ hole: return (struct bkey_s_c) { u, NULL }; } + +void bch2_set_btree_iter_dontneed(struct btree_iter *iter) +{ + struct btree_trans *trans = iter->trans; + + if (!iter->path || trans->restarted) + return; + + struct btree_path *path = btree_iter_path(trans, iter); + path->preserve = false; + if (path->ref == 1) + path->should_be_locked = false; +} /* Btree iterators: */ int __must_check diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 430b0d484874..eab2a25bdc7a 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -508,18 +508,7 @@ void bch2_trans_node_iter_init(struct btree_trans *, struct btree_iter *, unsigned, unsigned, unsigned); void bch2_trans_copy_iter(struct btree_iter *, struct btree_iter *); -static inline void set_btree_iter_dontneed(struct btree_iter *iter) -{ - struct btree_trans *trans = iter->trans; - - if (!iter->path || trans->restarted) - return; - - struct btree_path *path = btree_iter_path(trans, iter); - path->preserve = false; - if (path->ref == 1) - path->should_be_locked = false; -} +void bch2_set_btree_iter_dontneed(struct btree_iter *); void *__bch2_trans_kmalloc(struct btree_trans *, size_t); diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index e2eca7dab70e..69e5f08db374 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -456,7 +456,7 @@ static int btree_key_cache_fill(struct btree_trans *trans, bch2_btree_node_unlock_write(trans, ck_path, ck_path->l[0].b); /* We're not likely to need this iterator again: */ - set_btree_iter_dontneed(&iter); + bch2_set_btree_iter_dontneed(&iter); err: bch2_trans_iter_exit(trans, &iter); return ret; -- cgit v1.2.3-59-g8ed1b From 45150765d307c1043baa68aff53cb6fa5ba34603 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 26 Apr 2024 00:32:56 -0400 Subject: bcachefs: bch_member.last_journal_bucket On recovery from clean shutdown we don't typically read the journal, but we still want to avoid overwriting existing entries in the journal for list_journal debugging. Thus, add some fields to the member info section so we can remember where we left off. Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 7 +++++++ fs/bcachefs/journal_io.c | 28 ++++++++++++++++++++++++++++ fs/bcachefs/journal_io.h | 3 +++ fs/bcachefs/recovery.c | 2 ++ fs/bcachefs/sb-clean.c | 2 ++ 5 files changed, 42 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 2e8b1a489c20..d5b90439e581 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -589,6 +589,13 @@ struct bch_member { __le64 errors_reset_time; __le64 seq; __le64 btree_allocated_bitmap; + /* + * On recovery from a clean shutdown we don't normally read the journal, + * but we still want to resume writing from where we left off so we + * don't overwrite more than is necessary, for list journal debugging: + */ + __le32 last_journal_bucket; + __le32 last_journal_bucket_offset; }; /* diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 98cf9a65216f..f3c5e9a423fd 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -17,6 +17,34 @@ #include "sb-clean.h" #include "trace.h" +void bch2_journal_pos_from_member_info_set(struct bch_fs *c) +{ + lockdep_assert_held(&c->sb_lock); + + for_each_member_device(c, ca) { + struct bch_member *m = bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx); + + m->last_journal_bucket = cpu_to_le32(ca->journal.cur_idx); + m->last_journal_bucket_offset = cpu_to_le32(ca->mi.bucket_size - ca->journal.sectors_free); + } +} + +void bch2_journal_pos_from_member_info_resume(struct bch_fs *c) +{ + mutex_lock(&c->sb_lock); + for_each_member_device(c, ca) { + struct bch_member m = bch2_sb_member_get(c->disk_sb.sb, ca->dev_idx); + + unsigned idx = le32_to_cpu(m.last_journal_bucket); + if (idx < ca->journal.nr) + ca->journal.cur_idx = idx; + unsigned offset = le32_to_cpu(m.last_journal_bucket_offset); + if (offset <= ca->mi.bucket_size) + ca->journal.sectors_free = ca->mi.bucket_size - offset; + } + mutex_unlock(&c->sb_lock); +} + void bch2_journal_ptrs_to_text(struct printbuf *out, struct bch_fs *c, struct journal_replay *j) { diff --git a/fs/bcachefs/journal_io.h b/fs/bcachefs/journal_io.h index 4f1e763ab506..43cfed58ef33 100644 --- a/fs/bcachefs/journal_io.h +++ b/fs/bcachefs/journal_io.h @@ -4,6 +4,9 @@ #include "darray.h" +void bch2_journal_pos_from_member_info_set(struct bch_fs *); +void bch2_journal_pos_from_member_info_resume(struct bch_fs *); + struct journal_ptr { bool csum_good; u8 dev; diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index c33e20aa56a2..a6447ffd336e 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -670,6 +670,8 @@ int bch2_fs_recovery(struct bch_fs *c) goto err; } + bch2_journal_pos_from_member_info_resume(c); + if (!c->sb.clean || c->opts.retain_recovery_info) { struct genradix_iter iter; struct journal_replay **i; diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index c2d4be52c8c5..e2beb2308c70 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -390,6 +390,8 @@ void bch2_fs_mark_clean(struct bch_fs *c) goto out; } + bch2_journal_pos_from_member_info_set(c); + bch2_write_super(c); out: mutex_unlock(&c->sb_lock); -- cgit v1.2.3-59-g8ed1b From 4da1713a8db5b6d5e421b20687ecde9354ee85cf Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 25 Apr 2024 22:11:49 -0400 Subject: bcachefs: check for inodes that should have backpointers in fsck Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 10 ++++++++++ fs/bcachefs/inode.h | 8 ++++++++ 2 files changed, 18 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 4adb96965e11..c8f57465131c 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -1713,6 +1713,15 @@ static int check_dirent_inode_dirent(struct btree_trans *trans, if (inode_points_to_dirent(target, d)) return 0; + if (bch2_inode_should_have_bp(target) && + !fsck_err(c, inode_wrong_backpointer, + "dirent points to inode that does not point back:\n %s", + (bch2_bkey_val_to_text(&buf, c, d.s_c), + prt_printf(&buf, "\n "), + bch2_inode_unpacked_to_text(&buf, target), + buf.buf))) + goto out_noiter; + if (!target->bi_dir && !target->bi_dir_offset) { target->bi_dir = d.k->p.inode; @@ -1781,6 +1790,7 @@ out: err: fsck_err: bch2_trans_iter_exit(trans, &bp_iter); +out_noiter: printbuf_exit(&buf); bch_err_fn(c, ret); return ret; diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h index 46477a40846c..abdfb1f3e4d0 100644 --- a/fs/bcachefs/inode.h +++ b/fs/bcachefs/inode.h @@ -221,6 +221,14 @@ static inline void bch2_inode_nlink_set(struct bch_inode_unpacked *bi, int bch2_inode_nlink_inc(struct bch_inode_unpacked *); void bch2_inode_nlink_dec(struct btree_trans *, struct bch_inode_unpacked *); +static inline bool bch2_inode_should_have_bp(struct bch_inode_unpacked *inode) +{ + bool inode_has_bp = inode->bi_dir || inode->bi_dir_offset; + + return S_ISDIR(inode->bi_mode) || + (!inode->bi_nlink && inode_has_bp); +} + struct bch_opts bch2_inode_opts_to_opts(struct bch_inode_unpacked *); void bch2_inode_opts_get(struct bch_io_opts *, struct bch_fs *, struct bch_inode_unpacked *); -- cgit v1.2.3-59-g8ed1b From 83c38e3ef8214cb5b80c113ac39e855548f79a44 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 25 Apr 2024 19:54:03 -0400 Subject: bcachefs: check inode backpointer in bch2_lookup() Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 5624c2ea8d55..138d416a64d5 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -400,6 +400,7 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, struct bch_fs *c = trans->c; struct btree_iter dirent_iter = {}; subvol_inum inum = {}; + struct printbuf buf = PRINTBUF; struct bkey_s_c k = bch2_hash_lookup(trans, &dirent_iter, bch2_dirent_hash_desc, dir_hash_info, dir, name, 0); @@ -426,20 +427,31 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, ret = bch2_subvolume_get(trans, inum.subvol, true, 0, &subvol) ?: bch2_inode_find_by_inum_nowarn_trans(trans, inum, &inode_u) ?: PTR_ERR_OR_ZERO(inode = bch2_new_inode(trans)); - if (bch2_err_matches(ret, ENOENT)) { - struct printbuf buf = PRINTBUF; - bch2_bkey_val_to_text(&buf, c, k); - bch_err(c, "%s points to missing inode", buf.buf); - printbuf_exit(&buf); - } + bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), + c, "dirent to missing inode:\n %s", + (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); if (ret) goto err; + /* regular files may have hardlinks: */ + if (bch2_fs_inconsistent_on(bch2_inode_should_have_bp(&inode_u) && + !bkey_eq(k.k->p, POS(inode_u.bi_dir, inode_u.bi_dir_offset)), + c, + "dirent points to inode that does not point back:\n %s", + (bch2_bkey_val_to_text(&buf, c, k), + prt_printf(&buf, "\n "), + bch2_inode_unpacked_to_text(&buf, &inode_u), + buf.buf))) { + ret = -ENOENT; + goto err; + } + bch2_vfs_inode_init(trans, inum, inode, &inode_u, &subvol); inode = bch2_inode_insert(c, inode); out: bch2_trans_iter_exit(trans, &dirent_iter); + printbuf_exit(&buf); return inode; err: inode = ERR_PTR(ret); -- cgit v1.2.3-59-g8ed1b From c8bda9f20a01cef7e12fd230ada83d53f7cdc884 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 26 Apr 2024 00:31:14 -0400 Subject: bcachefs: Simplify resuming of journal position Signed-off-by: Kent Overstreet --- fs/bcachefs/journal_io.c | 50 +++++++-------------------------------------- fs/bcachefs/journal_types.h | 1 + 2 files changed, 8 insertions(+), 43 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index f3c5e9a423fd..dec50e7a8cb4 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1089,6 +1089,13 @@ reread: goto err; } + if (le64_to_cpu(j->seq) > ja->highest_seq_found) { + ja->highest_seq_found = le64_to_cpu(j->seq); + ja->cur_idx = bucket; + ja->sectors_free = ca->mi.bucket_size - + bucket_remainder(ca, offset) - sectors; + } + /* * This happens sometimes if we don't have discards on - * when we've partially overwritten a bucket with new @@ -1157,8 +1164,6 @@ static CLOSURE_CALLBACK(bch2_journal_read_device) struct bch_fs *c = ca->fs; struct journal_list *jlist = container_of(cl->parent, struct journal_list, cl); - struct journal_replay *r, **_r; - struct genradix_iter iter; struct journal_read_buf buf = { NULL, 0 }; unsigned i; int ret = 0; @@ -1178,47 +1183,6 @@ static CLOSURE_CALLBACK(bch2_journal_read_device) goto err; } - ja->sectors_free = ca->mi.bucket_size; - - mutex_lock(&jlist->lock); - genradix_for_each_reverse(&c->journal_entries, iter, _r) { - r = *_r; - - if (!r) - continue; - - darray_for_each(r->ptrs, i) - if (i->dev == ca->dev_idx) { - unsigned wrote = bucket_remainder(ca, i->sector) + - vstruct_sectors(&r->j, c->block_bits); - - ja->cur_idx = i->bucket; - ja->sectors_free = ca->mi.bucket_size - wrote; - goto found; - } - } -found: - mutex_unlock(&jlist->lock); - - if (ja->bucket_seq[ja->cur_idx] && - ja->sectors_free == ca->mi.bucket_size) { -#if 0 - /* - * Debug code for ZNS support, where we (probably) want to be - * correlated where we stopped in the journal to the zone write - * points: - */ - bch_err(c, "ja->sectors_free == ca->mi.bucket_size"); - bch_err(c, "cur_idx %u/%u", ja->cur_idx, ja->nr); - for (i = 0; i < 3; i++) { - unsigned idx = (ja->cur_idx + ja->nr - 1 + i) % ja->nr; - - bch_err(c, "bucket_seq[%u] = %llu", idx, ja->bucket_seq[idx]); - } -#endif - ja->sectors_free = 0; - } - /* * Set dirty_idx to indicate the entire journal is full and needs to be * reclaimed - journal reclaim will immediately reclaim whatever isn't diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index 3d2135e1d7a1..45c1c48f4917 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -327,6 +327,7 @@ struct journal_device { /* for bch_journal_read_device */ struct closure read; + u64 highest_seq_found; }; /* -- cgit v1.2.3-59-g8ed1b From 36aa49d33e8d59246bd338064d6a516f693954e5 Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Fri, 26 Apr 2024 15:29:56 +0800 Subject: bcachefs: Change destroy_inode to free_inode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The vfs[1] documentation describes free_inode as follows: ``` free_inode this method is called from RCU callback. If you use call_rcu() in ->destroy_inode to free ‘struct inode’ memory, then it’s better to release memory in this method. ``` free_inode will be called by the RCU callback, so it might be better to move the inode free operation to destroy_inode. Similar to commit ae6b47b5653e ("fs/ntfs3: Change destroy_inode to free_inode"). Link: [1]: https://www.kernel.org/doc/html/latest/filesystems/vfs.html Signed-off-by: Youling Tang Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 138d416a64d5..a8d71cec1c17 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -1523,17 +1523,9 @@ static void bch2_vfs_inode_init(struct btree_trans *trans, subvol_inum inum, mapping_set_large_folios(inode->v.i_mapping); } -static void bch2_i_callback(struct rcu_head *head) +static void bch2_free_inode(struct inode *vinode) { - struct inode *vinode = container_of(head, struct inode, i_rcu); - struct bch_inode_info *inode = to_bch_ei(vinode); - - kmem_cache_free(bch2_inode_cache, inode); -} - -static void bch2_destroy_inode(struct inode *vinode) -{ - call_rcu(&vinode->i_rcu, bch2_i_callback); + kmem_cache_free(bch2_inode_cache, to_bch_ei(vinode)); } static int inode_update_times_fn(struct btree_trans *trans, @@ -1841,7 +1833,7 @@ static int bch2_unfreeze(struct super_block *sb) static const struct super_operations bch_super_operations = { .alloc_inode = bch2_alloc_inode, - .destroy_inode = bch2_destroy_inode, + .free_inode = bch2_free_inode, .write_inode = bch2_vfs_write_inode, .evict_inode = bch2_evict_inode, .sync_fs = bch2_sync_fs, -- cgit v1.2.3-59-g8ed1b From 9862022d09f785088619fd3c14063592fddc2aa5 Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Tue, 30 Apr 2024 11:28:39 +0800 Subject: bcachefs: Fix error path of bch2_link_trans() In bch2_link_trans(), if bch2_inode_nlink_inc() fails, it needs to call bch2_trans_iter_exit() in the error path. Signed-off-by: Youling Tang Signed-off-by: Kent Overstreet --- fs/bcachefs/fs-common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs-common.c b/fs/bcachefs/fs-common.c index 6c737def357e..508d029ac53d 100644 --- a/fs/bcachefs/fs-common.c +++ b/fs/bcachefs/fs-common.c @@ -200,12 +200,12 @@ int bch2_link_trans(struct btree_trans *trans, ret = bch2_inode_peek(trans, &inode_iter, inode_u, inum, BTREE_ITER_intent); if (ret) - goto err; + return ret; inode_u->bi_ctime = now; ret = bch2_inode_nlink_inc(inode_u); if (ret) - return ret; + goto err; ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_intent); if (ret) -- cgit v1.2.3-59-g8ed1b From 75a53a0a230529425fa27bdfd5b7d9f95736488e Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Tue, 30 Apr 2024 11:27:44 +0800 Subject: bcachefs: Correct the FS_IOC_GETFLAGS to FS_IOC32_GETFLAGS in bch2_compat_fs_ioctl() It should be FS_IOC32_GETFLAGS instead of FS_IOC_GETFLAGS in compat ioctl. Signed-off-by: Youling Tang Signed-off-by: Kent Overstreet --- fs/bcachefs/fs-ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/fs-ioctl.c b/fs/bcachefs/fs-ioctl.c index 3dc8630ff9fe..205a323ffc6d 100644 --- a/fs/bcachefs/fs-ioctl.c +++ b/fs/bcachefs/fs-ioctl.c @@ -548,7 +548,7 @@ long bch2_compat_fs_ioctl(struct file *file, unsigned cmd, unsigned long arg) { /* These are just misnamed, they actually get/put from/to user an int */ switch (cmd) { - case FS_IOC_GETFLAGS: + case FS_IOC32_GETFLAGS: cmd = FS_IOC_GETFLAGS; break; case FS_IOC32_SETFLAGS: -- cgit v1.2.3-59-g8ed1b From d3c44cfd5e98605f707a8eaaa096c0d17b2211fb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 00:41:42 -0400 Subject: bcachefs: delete old gen check bch2_alloc_write_key() this was from metadata only gc - we don't need it anymore Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 9694ffb0b098..c9dfcf3cd199 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -907,9 +907,6 @@ static int bch2_alloc_write_key(struct btree_trans *trans, bch2_dev_usage_update_m(c, ca, &old_gc, &gc); percpu_up_read(&c->mark_lock); - if (gen_after(old->gen, gc.gen)) - return 0; - if (fsck_err_on(new.data_type != gc.data_type, c, alloc_key_data_type_wrong, "bucket %llu:%llu gen %u has wrong data_type" -- cgit v1.2.3-59-g8ed1b From 2685c67d128339db01f8af0798c8a9c3ad04c5ba Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 01:49:59 -0400 Subject: bcachefs: dirty_sectors -> replicas_sectors Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index e0e991043aa5..93ce0eb350fe 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -1144,7 +1144,7 @@ static int __trigger_extent(struct btree_trans *trans, enum bch_data_type data_type = bkey_is_btree_ptr(k.k) ? BCH_DATA_btree : BCH_DATA_user; - s64 dirty_sectors = 0; + s64 replicas_sectors = 0; int ret = 0; r.e.data_type = data_type; @@ -1170,7 +1170,7 @@ static int __trigger_extent(struct btree_trans *trans, return ret; } } else if (!p.has_ec) { - dirty_sectors += disk_sectors; + replicas_sectors += disk_sectors; r.e.devs[r.e.nr_devs++] = p.ptr.dev; } else { ret = bch2_trigger_stripe_ptr(trans, k, p, data_type, disk_sectors, flags); @@ -1188,8 +1188,8 @@ static int __trigger_extent(struct btree_trans *trans, if (r.e.nr_devs) { ret = !gc - ? bch2_update_replicas_list(trans, &r.e, dirty_sectors) - : bch2_update_replicas(c, k, &r.e, dirty_sectors, 0, true); + ? bch2_update_replicas_list(trans, &r.e, replicas_sectors) + : bch2_update_replicas(c, k, &r.e, replicas_sectors, 0, true); if (unlikely(ret && gc)) { struct printbuf buf = PRINTBUF; -- cgit v1.2.3-59-g8ed1b From fa9bb741fea1173f94f4fbdc7523744cde6dcdb8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 02:47:30 -0400 Subject: bcachefs: alloc_data_type_set() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 6 ++--- fs/bcachefs/alloc_background.h | 57 +++++++++++++++++++++++------------------- fs/bcachefs/btree_gc.c | 2 +- fs/bcachefs/ec.c | 2 +- 4 files changed, 36 insertions(+), 31 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 37e8446b3039..bbd85f1e5973 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -741,7 +741,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, if (flags & BTREE_TRIGGER_transactional) { struct bch_alloc_v4 *new_a = bkey_s_to_alloc_v4(new).v; - new_a->data_type = alloc_data_type(*new_a, new_a->data_type); + alloc_data_type_set(new_a, new_a->data_type); if (bch2_bucket_sectors(*new_a) > bch2_bucket_sectors(*old_a)) { new_a->io_time[READ] = max_t(u64, 1, atomic64_read(&c->io_clock[READ].now)); @@ -1761,7 +1761,7 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, } SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); - a->v.data_type = alloc_data_type(a->v, a->v.data_type); + alloc_data_type_set(&a->v, a->v.data_type); write: ret = bch2_trans_update(trans, &iter, &a->k_i, 0) ?: bch2_trans_commit(trans, NULL, NULL, @@ -1830,7 +1830,7 @@ static int bch2_clear_bucket_needs_discard(struct btree_trans *trans, struct bpo BUG_ON(a->v.dirty_sectors); SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); - a->v.data_type = alloc_data_type(a->v, a->v.data_type); + alloc_data_type_set(&a->v, a->v.data_type); ret = bch2_trans_update(trans, &iter, &a->k_i, 0); err: diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index e0e0d72db03b..98ba934c352f 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -39,32 +39,6 @@ static inline u8 alloc_gc_gen(struct bch_alloc_v4 a) return a.gen - a.oldest_gen; } -static inline enum bch_data_type __alloc_data_type(u32 dirty_sectors, - u32 cached_sectors, - u32 stripe, - struct bch_alloc_v4 a, - enum bch_data_type data_type) -{ - if (stripe) - return data_type == BCH_DATA_parity ? data_type : BCH_DATA_stripe; - if (dirty_sectors) - return data_type; - if (cached_sectors) - return BCH_DATA_cached; - if (BCH_ALLOC_V4_NEED_DISCARD(&a)) - return BCH_DATA_need_discard; - if (alloc_gc_gen(a) >= BUCKET_GC_GEN_MAX) - return BCH_DATA_need_gc_gens; - return BCH_DATA_free; -} - -static inline enum bch_data_type alloc_data_type(struct bch_alloc_v4 a, - enum bch_data_type data_type) -{ - return __alloc_data_type(a.dirty_sectors, a.cached_sectors, - a.stripe, a, data_type); -} - static inline enum bch_data_type bucket_data_type(enum bch_data_type data_type) { switch (data_type) { @@ -101,6 +75,37 @@ static inline unsigned bch2_bucket_sectors_fragmented(struct bch_dev *ca, return d ? max(0, ca->mi.bucket_size - d) : 0; } +static inline enum bch_data_type __alloc_data_type(u32 dirty_sectors, + u32 cached_sectors, + u32 stripe, + struct bch_alloc_v4 a, + enum bch_data_type data_type) +{ + if (stripe) + return data_type == BCH_DATA_parity ? data_type : BCH_DATA_stripe; + if (dirty_sectors) + return data_type; + if (cached_sectors) + return BCH_DATA_cached; + if (BCH_ALLOC_V4_NEED_DISCARD(&a)) + return BCH_DATA_need_discard; + if (alloc_gc_gen(a) >= BUCKET_GC_GEN_MAX) + return BCH_DATA_need_gc_gens; + return BCH_DATA_free; +} + +static inline enum bch_data_type alloc_data_type(struct bch_alloc_v4 a, + enum bch_data_type data_type) +{ + return __alloc_data_type(a.dirty_sectors, a.cached_sectors, + a.stripe, a, data_type); +} + +static inline void alloc_data_type_set(struct bch_alloc_v4 *a, enum bch_data_type data_type) +{ + a->data_type = alloc_data_type(*a, data_type); +} + static inline u64 alloc_lru_idx_read(struct bch_alloc_v4 a) { return a.data_type == BCH_DATA_cached ? a.io_time[READ] : 0; diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index c9dfcf3cd199..ef325ff71f89 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -1307,7 +1307,7 @@ static int bch2_alloc_write_oldest_gen(struct btree_trans *trans, struct btree_i return ret; a_mut->v.oldest_gen = ca->oldest_gen[iter->pos.offset]; - a_mut->v.data_type = alloc_data_type(a_mut->v, a_mut->v.data_type); + alloc_data_type_set(&a_mut->v, a_mut->v.data_type); return bch2_trans_update(trans, iter, &a_mut->k_i, 0); } diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index d6fa7e0bb266..a4ae39b10e98 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -290,7 +290,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, goto err; if (deleting) - a->v.data_type = alloc_data_type(a->v, BCH_DATA_user); + alloc_data_type_set(&a->v, BCH_DATA_user); ret = bch2_trans_update(trans, &iter, &a->k_i, 0); if (ret) -- cgit v1.2.3-59-g8ed1b From c02eb9e89184bd016d3c907d1287d05430b2016b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 00:29:24 -0400 Subject: bcachefs: kill bch2_dev_usage_update_m() by using bucket_m_to_alloc() more, we can get some nice code cleanup. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.h | 36 +++++++++++++++++++++--------------- fs/bcachefs/btree_gc.c | 25 ++++++++++--------------- fs/bcachefs/buckets.c | 35 +++++++---------------------------- fs/bcachefs/buckets.h | 2 -- fs/bcachefs/ec.c | 6 +++--- 5 files changed, 41 insertions(+), 63 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 98ba934c352f..7d747ad0de66 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -39,6 +39,22 @@ static inline u8 alloc_gc_gen(struct bch_alloc_v4 a) return a.gen - a.oldest_gen; } +static inline void __bucket_m_to_alloc(struct bch_alloc_v4 *dst, struct bucket b) +{ + dst->gen = b.gen; + dst->data_type = b.data_type; + dst->dirty_sectors = b.dirty_sectors; + dst->cached_sectors = b.cached_sectors; + dst->stripe = b.stripe; +} + +static inline struct bch_alloc_v4 bucket_m_to_alloc(struct bucket b) +{ + struct bch_alloc_v4 ret = {}; + __bucket_m_to_alloc(&ret, b); + return ret; +} + static inline enum bch_data_type bucket_data_type(enum bch_data_type data_type) { switch (data_type) { @@ -75,17 +91,14 @@ static inline unsigned bch2_bucket_sectors_fragmented(struct bch_dev *ca, return d ? max(0, ca->mi.bucket_size - d) : 0; } -static inline enum bch_data_type __alloc_data_type(u32 dirty_sectors, - u32 cached_sectors, - u32 stripe, - struct bch_alloc_v4 a, - enum bch_data_type data_type) +static inline enum bch_data_type alloc_data_type(struct bch_alloc_v4 a, + enum bch_data_type data_type) { - if (stripe) + if (a.stripe) return data_type == BCH_DATA_parity ? data_type : BCH_DATA_stripe; - if (dirty_sectors) + if (a.dirty_sectors) return data_type; - if (cached_sectors) + if (a.cached_sectors) return BCH_DATA_cached; if (BCH_ALLOC_V4_NEED_DISCARD(&a)) return BCH_DATA_need_discard; @@ -94,13 +107,6 @@ static inline enum bch_data_type __alloc_data_type(u32 dirty_sectors, return BCH_DATA_free; } -static inline enum bch_data_type alloc_data_type(struct bch_alloc_v4 a, - enum bch_data_type data_type) -{ - return __alloc_data_type(a.dirty_sectors, a.cached_sectors, - a.stripe, a, data_type); -} - static inline void alloc_data_type_set(struct bch_alloc_v4 *a, enum bch_data_type data_type) { a->data_type = alloc_data_type(*a, data_type); diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index ef325ff71f89..e5289b53d175 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -871,40 +871,35 @@ static int bch2_alloc_write_key(struct btree_trans *trans, { struct bch_fs *c = trans->c; struct bch_dev *ca = bch2_dev_bkey_exists(c, iter->pos.inode); - struct bucket old_gc, gc, *b; struct bkey_i_alloc_v4 *a; - struct bch_alloc_v4 old_convert, new; + struct bch_alloc_v4 old_gc, gc, old_convert, new; const struct bch_alloc_v4 *old; int ret; old = bch2_alloc_to_v4(k, &old_convert); - new = *old; + gc = new = *old; percpu_down_read(&c->mark_lock); - b = gc_bucket(ca, iter->pos.offset); - old_gc = *b; + __bucket_m_to_alloc(&gc, *gc_bucket(ca, iter->pos.offset)); + + old_gc = gc; if ((old->data_type == BCH_DATA_sb || old->data_type == BCH_DATA_journal) && !bch2_dev_is_online(ca)) { - b->data_type = old->data_type; - b->dirty_sectors = old->dirty_sectors; + gc.data_type = old->data_type; + gc.dirty_sectors = old->dirty_sectors; } /* - * b->data_type doesn't yet include need_discard & need_gc_gen states - + * gc.data_type doesn't yet include need_discard & need_gc_gen states - * fix that here: */ - b->data_type = __alloc_data_type(b->dirty_sectors, - b->cached_sectors, - b->stripe, - *old, - b->data_type); - gc = *b; + alloc_data_type_set(&gc, gc.data_type); if (gc.data_type != old_gc.data_type || gc.dirty_sectors != old_gc.dirty_sectors) - bch2_dev_usage_update_m(c, ca, &old_gc, &gc); + bch2_dev_usage_update(c, ca, &old_gc, &gc, 0, true); percpu_up_read(&c->mark_lock); if (fsck_err_on(new.data_type != gc.data_type, c, diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 93ce0eb350fe..3acd35b771fa 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -318,26 +318,6 @@ void bch2_dev_usage_update(struct bch_fs *c, struct bch_dev *ca, preempt_enable(); } -static inline struct bch_alloc_v4 bucket_m_to_alloc(struct bucket b) -{ - return (struct bch_alloc_v4) { - .gen = b.gen, - .data_type = b.data_type, - .dirty_sectors = b.dirty_sectors, - .cached_sectors = b.cached_sectors, - .stripe = b.stripe, - }; -} - -void bch2_dev_usage_update_m(struct bch_fs *c, struct bch_dev *ca, - struct bucket *old, struct bucket *new) -{ - struct bch_alloc_v4 old_a = bucket_m_to_alloc(*old); - struct bch_alloc_v4 new_a = bucket_m_to_alloc(*new); - - bch2_dev_usage_update(c, ca, &old_a, &new_a, 0, true); -} - static inline int __update_replicas(struct bch_fs *c, struct bch_fs_usage *fs_usage, struct bch_replicas_entry_v1 *r, @@ -1028,7 +1008,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, percpu_down_read(&c->mark_lock); struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); bucket_lock(g); - struct bucket old = *g; + struct bch_alloc_v4 old = bucket_m_to_alloc(*g); u8 bucket_data_type = g->data_type; int ret = __mark_pointer(trans, k, &p.ptr, *sectors, @@ -1043,9 +1023,9 @@ static int bch2_trigger_pointer(struct btree_trans *trans, } g->data_type = bucket_data_type; - struct bucket new = *g; + struct bch_alloc_v4 new = bucket_m_to_alloc(*g); bucket_unlock(g); - bch2_dev_usage_update_m(c, ca, &old, &new); + bch2_dev_usage_update(c, ca, &old, &new, 0, true); percpu_up_read(&c->mark_lock); } @@ -1336,14 +1316,13 @@ static int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca, u64 b, enum bch_data_type data_type, unsigned sectors, enum btree_iter_update_trigger_flags flags) { - struct bucket old, new, *g; int ret = 0; percpu_down_read(&c->mark_lock); - g = gc_bucket(ca, b); + struct bucket *g = gc_bucket(ca, b); bucket_lock(g); - old = *g; + struct bch_alloc_v4 old = bucket_m_to_alloc(*g); if (bch2_fs_inconsistent_on(g->data_type && g->data_type != data_type, c, @@ -1365,11 +1344,11 @@ static int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca, g->data_type = data_type; g->dirty_sectors += sectors; - new = *g; + struct bch_alloc_v4 new = bucket_m_to_alloc(*g); err: bucket_unlock(g); if (!ret) - bch2_dev_usage_update_m(c, ca, &old, &new); + bch2_dev_usage_update(c, ca, &old, &new, 0, true); percpu_up_read(&c->mark_lock); return ret; } diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index df10f390101a..f328ddb82dd9 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -310,8 +310,6 @@ bch2_fs_usage_read_short(struct bch_fs *); void bch2_dev_usage_update(struct bch_fs *, struct bch_dev *, const struct bch_alloc_v4 *, const struct bch_alloc_v4 *, u64, bool); -void bch2_dev_usage_update_m(struct bch_fs *, struct bch_dev *, - struct bucket *, struct bucket *); /* key/bucket marking: */ diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index a4ae39b10e98..ce82cef1267d 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -307,7 +307,7 @@ err: struct bucket *g = gc_bucket(ca, bucket.offset); bucket_lock(g); - struct bucket old = *g; + struct bch_alloc_v4 old = bucket_m_to_alloc(*g); u8 data_type = g->data_type; int ret = __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, @@ -318,10 +318,10 @@ err: &g->stripe, &g->stripe_redundancy); g->data_type = data_type; - struct bucket new = *g; + struct bch_alloc_v4 new = bucket_m_to_alloc(*g); bucket_unlock(g); if (!ret) - bch2_dev_usage_update_m(c, ca, &old, &new); + bch2_dev_usage_update(c, ca, &old, &new, 0, true); percpu_up_read(&c->mark_lock); return ret; } -- cgit v1.2.3-59-g8ed1b From be11ae16c4907fc9ede68cf5589d0bdd2b195d01 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 02:10:57 -0400 Subject: bcachefs: __mark_pointer now takes bch_alloc_v4 Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 6 +++--- fs/bcachefs/alloc_background.h | 23 ++++++++++++++------- fs/bcachefs/buckets.c | 45 +++++++++++++----------------------------- 3 files changed, 33 insertions(+), 41 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index bbd85f1e5973..9122e886bbf2 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -263,7 +263,7 @@ int bch2_alloc_v4_invalid(struct bch_fs *c, struct bkey_s_c k, case BCH_DATA_free: case BCH_DATA_need_gc_gens: case BCH_DATA_need_discard: - bkey_fsck_err_on(bch2_bucket_sectors(*a.v) || a.v->stripe, + bkey_fsck_err_on(bch2_bucket_sectors_total(*a.v) || a.v->stripe, c, err, alloc_key_empty_but_have_data, "empty data type free but have data"); break; @@ -743,7 +743,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, alloc_data_type_set(new_a, new_a->data_type); - if (bch2_bucket_sectors(*new_a) > bch2_bucket_sectors(*old_a)) { + if (bch2_bucket_sectors_total(*new_a) > bch2_bucket_sectors_total(*old_a)) { new_a->io_time[READ] = max_t(u64, 1, atomic64_read(&c->io_clock[READ].now)); new_a->io_time[WRITE]= max_t(u64, 1, atomic64_read(&c->io_clock[WRITE].now)); SET_BCH_ALLOC_V4_NEED_INC_GEN(new_a, true); @@ -1703,7 +1703,7 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, if (ret) goto out; - if (a->v.dirty_sectors) { + if (bch2_bucket_sectors_total(a->v)) { if (bch2_trans_inconsistent_on(c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info, trans, "attempting to discard bucket with dirty data\n%s", (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 7d747ad0de66..fe72e8a53bdb 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -39,13 +39,22 @@ static inline u8 alloc_gc_gen(struct bch_alloc_v4 a) return a.gen - a.oldest_gen; } -static inline void __bucket_m_to_alloc(struct bch_alloc_v4 *dst, struct bucket b) +static inline void alloc_to_bucket(struct bucket *dst, struct bch_alloc_v4 src) { - dst->gen = b.gen; - dst->data_type = b.data_type; - dst->dirty_sectors = b.dirty_sectors; - dst->cached_sectors = b.cached_sectors; - dst->stripe = b.stripe; + dst->gen = src.gen; + dst->data_type = src.data_type; + dst->dirty_sectors = src.dirty_sectors; + dst->cached_sectors = src.cached_sectors; + dst->stripe = src.stripe; +} + +static inline void __bucket_m_to_alloc(struct bch_alloc_v4 *dst, struct bucket src) +{ + dst->gen = src.gen; + dst->data_type = src.data_type; + dst->dirty_sectors = src.dirty_sectors; + dst->cached_sectors = src.cached_sectors; + dst->stripe = src.stripe; } static inline struct bch_alloc_v4 bucket_m_to_alloc(struct bucket b) @@ -73,7 +82,7 @@ static inline bool bucket_data_type_mismatch(enum bch_data_type bucket, bucket_data_type(bucket) != bucket_data_type(ptr); } -static inline unsigned bch2_bucket_sectors(struct bch_alloc_v4 a) +static inline unsigned bch2_bucket_sectors_total(struct bch_alloc_v4 a) { return a.dirty_sectors + a.cached_sectors; } diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 3acd35b771fa..bb54931fcd69 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -943,23 +943,18 @@ static int __mark_pointer(struct btree_trans *trans, struct bkey_s_c k, const struct bch_extent_ptr *ptr, s64 sectors, enum bch_data_type ptr_data_type, - u8 bucket_gen, u8 *bucket_data_type, - u32 *dirty_sectors, u32 *cached_sectors) + struct bch_alloc_v4 *a) { u32 *dst_sectors = !ptr->cached - ? dirty_sectors - : cached_sectors; + ? &a->dirty_sectors + : &a->cached_sectors; int ret = bch2_bucket_ref_update(trans, k, ptr, sectors, ptr_data_type, - bucket_gen, *bucket_data_type, dst_sectors); + a->gen, a->data_type, dst_sectors); if (ret) return ret; - if (!*dirty_sectors && !*cached_sectors) - *bucket_data_type = 0; - else if (*bucket_data_type != BCH_DATA_stripe) - *bucket_data_type = ptr_data_type; - + alloc_data_type_set(a, ptr_data_type); return 0; } @@ -984,9 +979,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, if (ret) return ret; - ret = __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, - a->v.gen, &a->v.data_type, - &a->v.dirty_sectors, &a->v.cached_sectors) ?: + ret = __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, &a->v) ?: bch2_trans_update(trans, &iter, &a->k_i, 0); bch2_trans_iter_exit(trans, &iter); @@ -1003,30 +996,20 @@ static int bch2_trigger_pointer(struct btree_trans *trans, if (flags & BTREE_TRIGGER_gc) { struct bch_fs *c = trans->c; struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); - enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry); percpu_down_read(&c->mark_lock); - struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); + struct bucket *g = gc_bucket(ca, bucket.offset); bucket_lock(g); - struct bch_alloc_v4 old = bucket_m_to_alloc(*g); - - u8 bucket_data_type = g->data_type; - int ret = __mark_pointer(trans, k, &p.ptr, *sectors, - data_type, g->gen, - &bucket_data_type, - &g->dirty_sectors, - &g->cached_sectors); - if (ret) { - bucket_unlock(g); - percpu_up_read(&c->mark_lock); - return ret; - } + struct bch_alloc_v4 old = bucket_m_to_alloc(*g), new = old; - g->data_type = bucket_data_type; - struct bch_alloc_v4 new = bucket_m_to_alloc(*g); + int ret = __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, &new); + if (!ret) { + alloc_to_bucket(g, new); + bch2_dev_usage_update(c, ca, &old, &new, 0, true); + } bucket_unlock(g); - bch2_dev_usage_update(c, ca, &old, &new, 0, true); percpu_up_read(&c->mark_lock); + return ret; } return 0; -- cgit v1.2.3-59-g8ed1b From 0acf2169a5e9b1ce286c44a740536e1cccb18db7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 02:41:16 -0400 Subject: bcachefs: __mark_stripe_bucket() now takes bch_alloc_v4 Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 1 - fs/bcachefs/ec.c | 100 +++++++++++++++++++------------------------------- 2 files changed, 38 insertions(+), 63 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index bb54931fcd69..9de96539817b 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -1001,7 +1001,6 @@ static int bch2_trigger_pointer(struct btree_trans *trans, struct bucket *g = gc_bucket(ca, bucket.offset); bucket_lock(g); struct bch_alloc_v4 old = bucket_m_to_alloc(*g), new = old; - int ret = __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, &new); if (!ret) { alloc_to_bucket(g, new); diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index ce82cef1267d..1a5608488441 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -167,12 +167,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, struct bkey_s_c_stripe s, unsigned ptr_idx, bool deleting, struct bpos bucket, - u8 bucket_gen, - u8 *bucket_data_type, - u32 *bucket_dirty_sectors, - u32 *bucket_cached_sectors, - u32 *bucket_stripe, - u8 *bucket_stripe_redundancy) + struct bch_alloc_v4 *a) { struct bch_fs *c = trans->c; const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; @@ -187,43 +182,43 @@ static int __mark_stripe_bucket(struct btree_trans *trans, sectors = -sectors; if (!deleting) { - if (bch2_trans_inconsistent_on(*bucket_stripe || - *bucket_stripe_redundancy, trans, + if (bch2_trans_inconsistent_on(a->stripe || + a->stripe_redundancy, trans, "bucket %llu:%llu gen %u data type %s dirty_sectors %u: multiple stripes using same bucket (%u, %llu)\n%s", - bucket.inode, bucket.offset, bucket_gen, - bch2_data_type_str(*bucket_data_type), - *bucket_dirty_sectors, - *bucket_stripe, s.k->p.offset, + bucket.inode, bucket.offset, a->gen, + bch2_data_type_str(a->data_type), + a->dirty_sectors, + a->stripe, s.k->p.offset, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } - if (bch2_trans_inconsistent_on(parity && (*bucket_dirty_sectors || *bucket_cached_sectors), trans, + if (bch2_trans_inconsistent_on(parity && bch2_bucket_sectors_total(*a), trans, "bucket %llu:%llu gen %u data type %s dirty_sectors %u cached_sectors %u: data already in parity bucket\n%s", - bucket.inode, bucket.offset, bucket_gen, - bch2_data_type_str(*bucket_data_type), - *bucket_dirty_sectors, - *bucket_cached_sectors, + bucket.inode, bucket.offset, a->gen, + bch2_data_type_str(a->data_type), + a->dirty_sectors, + a->cached_sectors, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } } else { - if (bch2_trans_inconsistent_on(*bucket_stripe != s.k->p.offset || - *bucket_stripe_redundancy != s.v->nr_redundant, trans, + if (bch2_trans_inconsistent_on(a->stripe != s.k->p.offset || + a->stripe_redundancy != s.v->nr_redundant, trans, "bucket %llu:%llu gen %u: not marked as stripe when deleting stripe (got %u)\n%s", - bucket.inode, bucket.offset, bucket_gen, - *bucket_stripe, + bucket.inode, bucket.offset, a->gen, + a->stripe, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; } - if (bch2_trans_inconsistent_on(*bucket_data_type != data_type, trans, + if (bch2_trans_inconsistent_on(a->data_type != data_type, trans, "bucket %llu:%llu gen %u data type %s: wrong data type when stripe, should be %s\n%s", - bucket.inode, bucket.offset, bucket_gen, - bch2_data_type_str(*bucket_data_type), + bucket.inode, bucket.offset, a->gen, + bch2_data_type_str(a->data_type), bch2_data_type_str(data_type), (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; @@ -231,12 +226,12 @@ static int __mark_stripe_bucket(struct btree_trans *trans, } if (bch2_trans_inconsistent_on(parity && - (*bucket_dirty_sectors != -sectors || - *bucket_cached_sectors), trans, + (a->dirty_sectors != -sectors || + a->cached_sectors), trans, "bucket %llu:%llu gen %u dirty_sectors %u cached_sectors %u: wrong sectors when deleting parity block of stripe\n%s", - bucket.inode, bucket.offset, bucket_gen, - *bucket_dirty_sectors, - *bucket_cached_sectors, + bucket.inode, bucket.offset, a->gen, + a->dirty_sectors, + a->cached_sectors, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -EIO; goto err; @@ -245,21 +240,20 @@ static int __mark_stripe_bucket(struct btree_trans *trans, if (sectors) { ret = bch2_bucket_ref_update(trans, s.s_c, ptr, sectors, data_type, - bucket_gen, *bucket_data_type, - bucket_dirty_sectors); + a->gen, a->data_type, &a->dirty_sectors); if (ret) goto err; } if (!deleting) { - *bucket_stripe = s.k->p.offset; - *bucket_stripe_redundancy = s.v->nr_redundant; - *bucket_data_type = data_type; + a->stripe = s.k->p.offset; + a->stripe_redundancy = s.v->nr_redundant; } else { - *bucket_stripe = 0; - *bucket_stripe_redundancy = 0; - *bucket_data_type = BCH_DATA_free; + a->stripe = 0; + a->stripe_redundancy = 0; } + + alloc_data_type_set(a, data_type); err: printbuf_exit(&buf); return ret; @@ -279,19 +273,10 @@ static int mark_stripe_bucket(struct btree_trans *trans, struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, &iter, bucket); int ret = PTR_ERR_OR_ZERO(a) ?: - __mark_stripe_bucket(trans, s, ptr_idx, deleting, iter.pos, - a->v.gen, - &a->v.data_type, - &a->v.dirty_sectors, - &a->v.cached_sectors, - &a->v.stripe, - &a->v.stripe_redundancy); + __mark_stripe_bucket(trans, s, ptr_idx, deleting, iter.pos, &a->v); if (ret) goto err; - if (deleting) - alloc_data_type_set(&a->v, BCH_DATA_user); - ret = bch2_trans_update(trans, &iter, &a->k_i, 0); if (ret) goto err; @@ -305,23 +290,14 @@ err: percpu_down_read(&c->mark_lock); struct bucket *g = gc_bucket(ca, bucket.offset); - bucket_lock(g); - struct bch_alloc_v4 old = bucket_m_to_alloc(*g); - u8 data_type = g->data_type; - - int ret = __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, - g->gen, - &data_type, - &g->dirty_sectors, - &g->cached_sectors, - &g->stripe, - &g->stripe_redundancy); - g->data_type = data_type; - struct bch_alloc_v4 new = bucket_m_to_alloc(*g); - bucket_unlock(g); - if (!ret) + struct bch_alloc_v4 old = bucket_m_to_alloc(*g), new = old; + int ret = __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &new); + if (!ret) { + alloc_to_bucket(g, new); bch2_dev_usage_update(c, ca, &old, &new, 0, true); + } + bucket_unlock(g); percpu_up_read(&c->mark_lock); return ret; } -- cgit v1.2.3-59-g8ed1b From abe2f470bc18a8ba50ca0a4c73327d993f6678c5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 03:27:30 -0400 Subject: bcachefs: simplify bch2_trans_start_alloc_update() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 48 ++++++++++++++++++++++++------------------ fs/bcachefs/alloc_background.h | 4 +++- fs/bcachefs/buckets.c | 15 ++++--------- fs/bcachefs/ec.c | 16 +++----------- 4 files changed, 37 insertions(+), 46 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 9122e886bbf2..6b3bad385812 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -429,22 +429,18 @@ struct bkey_i_alloc_v4 *bch2_alloc_to_v4_mut(struct btree_trans *trans, struct b } struct bkey_i_alloc_v4 * -bch2_trans_start_alloc_update(struct btree_trans *trans, struct btree_iter *iter, - struct bpos pos) +bch2_trans_start_alloc_update_noupdate(struct btree_trans *trans, struct btree_iter *iter, + struct bpos pos) { - struct bkey_s_c k; - struct bkey_i_alloc_v4 *a; - int ret; - - k = bch2_bkey_get_iter(trans, iter, BTREE_ID_alloc, pos, - BTREE_ITER_with_updates| - BTREE_ITER_cached| - BTREE_ITER_intent); - ret = bkey_err(k); + struct bkey_s_c k = bch2_bkey_get_iter(trans, iter, BTREE_ID_alloc, pos, + BTREE_ITER_with_updates| + BTREE_ITER_cached| + BTREE_ITER_intent); + int ret = bkey_err(k); if (unlikely(ret)) return ERR_PTR(ret); - a = bch2_alloc_to_v4_mut_inlined(trans, k); + struct bkey_i_alloc_v4 *a = bch2_alloc_to_v4_mut_inlined(trans, k); ret = PTR_ERR_OR_ZERO(a); if (unlikely(ret)) goto err; @@ -454,6 +450,20 @@ err: return ERR_PTR(ret); } +__flatten +struct bkey_i_alloc_v4 *bch2_trans_start_alloc_update(struct btree_trans *trans, struct bpos pos) +{ + struct btree_iter iter; + struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update_noupdate(trans, &iter, pos); + int ret = PTR_ERR_OR_ZERO(a); + if (ret) + return ERR_PTR(ret); + + ret = bch2_trans_update(trans, &iter, &a->k_i, 0); + bch2_trans_iter_exit(trans, &iter); + return unlikely(ret) ? ERR_PTR(ret) : a; +} + static struct bpos alloc_gens_pos(struct bpos pos, unsigned *offset) { *offset = pos.offset & KEY_TYPE_BUCKET_GENS_MASK; @@ -1908,7 +1918,6 @@ static int invalidate_one_bucket(struct btree_trans *trans, s64 *nr_to_invalidate) { struct bch_fs *c = trans->c; - struct btree_iter alloc_iter = { NULL }; struct bkey_i_alloc_v4 *a = NULL; struct printbuf buf = PRINTBUF; struct bpos bucket = u64_to_bucket(lru_k.k->p.offset); @@ -1926,7 +1935,7 @@ static int invalidate_one_bucket(struct btree_trans *trans, if (bch2_bucket_is_open_safe(c, bucket.inode, bucket.offset)) return 0; - a = bch2_trans_start_alloc_update(trans, &alloc_iter, bucket); + a = bch2_trans_start_alloc_update(trans, bucket); ret = PTR_ERR_OR_ZERO(a); if (ret) goto out; @@ -1951,18 +1960,15 @@ static int invalidate_one_bucket(struct btree_trans *trans, a->v.io_time[READ] = atomic64_read(&c->io_clock[READ].now); a->v.io_time[WRITE] = atomic64_read(&c->io_clock[WRITE].now); - ret = bch2_trans_update(trans, &alloc_iter, &a->k_i, - BTREE_TRIGGER_bucket_invalidate) ?: - bch2_trans_commit(trans, NULL, NULL, - BCH_WATERMARK_btree| - BCH_TRANS_COMMIT_no_enospc); + ret = bch2_trans_commit(trans, NULL, NULL, + BCH_WATERMARK_btree| + BCH_TRANS_COMMIT_no_enospc); if (ret) goto out; trace_and_count(c, bucket_invalidate, c, bucket.inode, bucket.offset, cached_sectors); --*nr_to_invalidate; out: - bch2_trans_iter_exit(trans, &alloc_iter); printbuf_exit(&buf); return ret; err: @@ -2175,7 +2181,7 @@ int bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev, if (bch2_trans_relock(trans)) bch2_trans_begin(trans); - a = bch2_trans_start_alloc_update(trans, &iter, POS(dev, bucket_nr)); + a = bch2_trans_start_alloc_update_noupdate(trans, &iter, POS(dev, bucket_nr)); ret = PTR_ERR_OR_ZERO(a); if (ret) return ret; diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index fe72e8a53bdb..a75eba6d9586 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -179,7 +179,9 @@ static inline void set_alloc_v4_u64s(struct bkey_i_alloc_v4 *a) } struct bkey_i_alloc_v4 * -bch2_trans_start_alloc_update(struct btree_trans *, struct btree_iter *, struct bpos); +bch2_trans_start_alloc_update_noupdate(struct btree_trans *, struct btree_iter *, struct bpos); +struct bkey_i_alloc_v4 * +bch2_trans_start_alloc_update(struct btree_trans *, struct bpos); void __bch2_alloc_to_v4(struct bkey_s_c, struct bch_alloc_v4 *); diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 9de96539817b..1f810cfb3541 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -973,16 +973,9 @@ static int bch2_trigger_pointer(struct btree_trans *trans, *sectors = insert ? bp.bucket_len : -((s64) bp.bucket_len); if (flags & BTREE_TRIGGER_transactional) { - struct btree_iter iter; - struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, &iter, bucket); - int ret = PTR_ERR_OR_ZERO(a); - if (ret) - return ret; - - ret = __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, &a->v) ?: - bch2_trans_update(trans, &iter, &a->k_i, 0); - bch2_trans_iter_exit(trans, &iter); - + struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, bucket); + int ret = PTR_ERR_OR_ZERO(a) ?: + __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, &a->v); if (ret) return ret; @@ -1266,7 +1259,7 @@ static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans, int ret = 0; struct bkey_i_alloc_v4 *a = - bch2_trans_start_alloc_update(trans, &iter, POS(ca->dev_idx, b)); + bch2_trans_start_alloc_update_noupdate(trans, &iter, POS(ca->dev_idx, b)); if (IS_ERR(a)) return PTR_ERR(a); diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 1a5608488441..64dcf852daed 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -269,20 +269,10 @@ static int mark_stripe_bucket(struct btree_trans *trans, struct bpos bucket = PTR_BUCKET_POS(c, ptr); if (flags & BTREE_TRIGGER_transactional) { - struct btree_iter iter; struct bkey_i_alloc_v4 *a = - bch2_trans_start_alloc_update(trans, &iter, bucket); - int ret = PTR_ERR_OR_ZERO(a) ?: - __mark_stripe_bucket(trans, s, ptr_idx, deleting, iter.pos, &a->v); - if (ret) - goto err; - - ret = bch2_trans_update(trans, &iter, &a->k_i, 0); - if (ret) - goto err; -err: - bch2_trans_iter_exit(trans, &iter); - return ret; + bch2_trans_start_alloc_update(trans, bucket); + return PTR_ERR_OR_ZERO(a) ?: + __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &a->v); } if (flags & BTREE_TRIGGER_gc) { -- cgit v1.2.3-59-g8ed1b From ffcbec607613bf8f83c06ae2d3cfe017de52f917 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 6 Apr 2024 00:07:46 -0400 Subject: bcachefs: Kill opts.buckets_nouse Now explicitly allocate and free the buckets_nouse bitmap - this is going to be used for online fsck. To go RW when we haven't check allocations, we'll do a much slimmed down version that just initializes the buckets_nouse bitmaps. Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 43 +++++++++++++++++++++++++++---------------- fs/bcachefs/buckets.h | 3 +++ fs/bcachefs/opts.h | 5 ----- fs/bcachefs/super.c | 5 ++--- 4 files changed, 32 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 1f810cfb3541..525c72c0277c 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -1519,6 +1519,31 @@ recalculate: /* Startup/shutdown: */ +void bch2_buckets_nouse_free(struct bch_fs *c) +{ + for_each_member_device(c, ca) { + kvfree_rcu_mightsleep(ca->buckets_nouse); + ca->buckets_nouse = NULL; + } +} + +int bch2_buckets_nouse_alloc(struct bch_fs *c) +{ + for_each_member_device(c, ca) { + BUG_ON(ca->buckets_nouse); + + ca->buckets_nouse = kvmalloc(BITS_TO_LONGS(ca->mi.nbuckets) * + sizeof(unsigned long), + GFP_KERNEL|__GFP_ZERO); + if (!ca->buckets_nouse) { + percpu_ref_put(&ca->ref); + return -BCH_ERR_ENOMEM_buckets_nouse; + } + } + + return 0; +} + static void bucket_gens_free_rcu(struct rcu_head *rcu) { struct bucket_gens *buckets = @@ -1530,24 +1555,17 @@ static void bucket_gens_free_rcu(struct rcu_head *rcu) int bch2_dev_buckets_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) { struct bucket_gens *bucket_gens = NULL, *old_bucket_gens = NULL; - unsigned long *buckets_nouse = NULL; bool resize = ca->bucket_gens != NULL; int ret; + BUG_ON(resize && ca->buckets_nouse); + if (!(bucket_gens = kvmalloc(sizeof(struct bucket_gens) + nbuckets, GFP_KERNEL|__GFP_ZERO))) { ret = -BCH_ERR_ENOMEM_bucket_gens; goto err; } - if ((c->opts.buckets_nouse && - !(buckets_nouse = kvmalloc(BITS_TO_LONGS(nbuckets) * - sizeof(unsigned long), - GFP_KERNEL|__GFP_ZERO)))) { - ret = -BCH_ERR_ENOMEM_buckets_nouse; - goto err; - } - bucket_gens->first_bucket = ca->mi.first_bucket; bucket_gens->nbuckets = nbuckets; @@ -1565,17 +1583,11 @@ int bch2_dev_buckets_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) memcpy(bucket_gens->b, old_bucket_gens->b, n); - if (buckets_nouse) - memcpy(buckets_nouse, - ca->buckets_nouse, - BITS_TO_LONGS(n) * sizeof(unsigned long)); } rcu_assign_pointer(ca->bucket_gens, bucket_gens); bucket_gens = old_bucket_gens; - swap(ca->buckets_nouse, buckets_nouse); - nbuckets = ca->mi.nbuckets; if (resize) { @@ -1586,7 +1598,6 @@ int bch2_dev_buckets_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) ret = 0; err: - kvfree(buckets_nouse); if (bucket_gens) call_rcu(&bucket_gens->rcu, bucket_gens_free_rcu); diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index f328ddb82dd9..f352d88d9b8e 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -472,6 +472,9 @@ static inline u64 avail_factor(u64 r) return div_u64(r << RESERVE_FACTOR, (1 << RESERVE_FACTOR) + 1); } +void bch2_buckets_nouse_free(struct bch_fs *); +int bch2_buckets_nouse_alloc(struct bch_fs *); + int bch2_dev_buckets_resize(struct bch_fs *, struct bch_dev *, u64); void bch2_dev_buckets_free(struct bch_dev *); int bch2_dev_buckets_alloc(struct bch_fs *, struct bch_dev *); diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index 74f4d92849bb..25530e0bb2f3 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -426,11 +426,6 @@ enum fsck_err_opts { BCH_SB_VERSION_UPGRADE, BCH_VERSION_UPGRADE_compatible, \ NULL, "Set superblock to latest version,\n" \ "allowing any new features to be used") \ - x(buckets_nouse, u8, \ - 0, \ - OPT_BOOL(), \ - BCH2_NO_SB_OPT, false, \ - NULL, "Allocate the buckets_nouse bitmap") \ x(stdio, u64, \ 0, \ OPT_UINT(0, S64_MAX), \ diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 72dde1f1a3e6..1cbb2b3f4740 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -531,9 +531,7 @@ int bch2_fs_read_write_early(struct bch_fs *c) static void __bch2_fs_free(struct bch_fs *c) { - unsigned i; - - for (i = 0; i < BCH_TIME_STAT_NR; i++) + for (unsigned i = 0; i < BCH_TIME_STAT_NR; i++) bch2_time_stats_exit(&c->times[i]); bch2_find_btree_nodes_exit(&c->found_btree_nodes); @@ -1189,6 +1187,7 @@ static void bch2_dev_free(struct bch_dev *ca) if (ca->kobj.state_in_sysfs) kobject_del(&ca->kobj); + kfree(ca->buckets_nouse); bch2_free_super(&ca->disk_sb); bch2_dev_journal_exit(ca); -- cgit v1.2.3-59-g8ed1b From 3a718c0647680965171a19c2524c137be8f287b9 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 04:24:58 -0400 Subject: bcachefs: On device add, prefer unused slots We can't strictly guarantee that no pointers refer to nonexistent devices - we attempt to, but we need to be safe when the filesystem is corrupt. Therefore, change device_add to try to pick a slot that's never been used, or the slot that's been unused the longest. Signed-off-by: Kent Overstreet --- fs/bcachefs/super.c | 25 ++++++++++++++++++++++--- 1 file changed, 22 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 1cbb2b3f4740..6c6087039781 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -1764,9 +1764,28 @@ int bch2_dev_add(struct bch_fs *c, const char *path) if (dynamic_fault("bcachefs:add:no_slot")) goto no_slot; - for (dev_idx = 0; dev_idx < BCH_SB_MEMBERS_MAX; dev_idx++) - if (!bch2_member_exists(c->disk_sb.sb, dev_idx)) - goto have_slot; + if (c->sb.nr_devices < BCH_SB_MEMBERS_MAX) { + dev_idx = c->sb.nr_devices; + goto have_slot; + } + + int best = -1; + u64 best_last_mount = 0; + for (dev_idx = 0; dev_idx < BCH_SB_MEMBERS_MAX; dev_idx++) { + struct bch_member m = bch2_sb_member_get(c->disk_sb.sb, dev_idx); + if (bch2_member_alive(&m)) + continue; + + u64 last_mount = le64_to_cpu(m.last_mount); + if (best < 0 || last_mount < best_last_mount) { + best = dev_idx; + best_last_mount = last_mount; + } + } + if (best >= 0) { + dev_idx = best; + goto have_slot; + } no_slot: ret = -BCH_ERR_ENOSPC_sb_members; bch_err_msg(c, ret, "setting up new superblock"); -- cgit v1.2.3-59-g8ed1b From b895c70326aa43b48d9b862e5e017de5fa83d844 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 06:20:37 -0400 Subject: bcachefs: x-macroize journal flags enums Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 2 +- fs/bcachefs/btree_key_cache.c | 2 +- fs/bcachefs/btree_trans_commit.c | 2 +- fs/bcachefs/btree_update.c | 2 +- fs/bcachefs/btree_update_interior.c | 6 +++--- fs/bcachefs/journal.c | 20 +++++++++++++------- fs/bcachefs/journal.h | 6 +++--- fs/bcachefs/journal_io.c | 6 +++--- fs/bcachefs/journal_reclaim.c | 10 +++++----- fs/bcachefs/journal_types.h | 15 ++++++++++----- fs/bcachefs/super.c | 6 +++--- 11 files changed, 44 insertions(+), 33 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 7f0eb5d396a2..5bf98cb8b15d 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -3170,7 +3170,7 @@ got_trans: trans->locking_wait.task = current; trans->locked = true; trans->journal_replay_not_finished = - unlikely(!test_bit(JOURNAL_REPLAY_DONE, &c->journal.flags)) && + unlikely(!test_bit(JOURNAL_replay_done, &c->journal.flags)) && atomic_inc_not_zero(&c->journal_keys.ref); trans->nr_paths = ARRAY_SIZE(trans->_paths); trans->paths_allocated = trans->_paths_allocated; diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 69e5f08db374..75f5e6fe4634 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -648,7 +648,7 @@ static int btree_key_cache_flush_pos(struct btree_trans *trans, commit_flags |= BCH_WATERMARK_reclaim; if (ck->journal.seq != journal_last_seq(j) || - !test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)) + !test_bit(JOURNAL_space_low, &c->journal.flags)) commit_flags |= BCH_TRANS_COMMIT_no_journal_res; ret = bch2_btree_iter_traverse(&b_iter) ?: diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 0bf7c70da417..a219a87e9ef0 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -337,7 +337,7 @@ static inline void btree_insert_entry_checks(struct btree_trans *trans, EBUG_ON(!i->level && btree_type_has_snapshots(i->btree_id) && !(i->flags & BTREE_UPDATE_internal_snapshot_node) && - test_bit(JOURNAL_REPLAY_DONE, &trans->c->journal.flags) && + test_bit(JOURNAL_replay_done, &trans->c->journal.flags) && i->k->k.p.snapshot && bch2_snapshot_is_internal_node(trans->c, i->k->k.p.snapshot) > 0); } diff --git a/fs/bcachefs/btree_update.c b/fs/bcachefs/btree_update.c index ee10c1f845b8..f3c645a43dcb 100644 --- a/fs/bcachefs/btree_update.c +++ b/fs/bcachefs/btree_update.c @@ -849,7 +849,7 @@ __bch2_fs_log_msg(struct bch_fs *c, unsigned commit_flags, const char *fmt, if (ret) goto err; - if (!test_bit(JOURNAL_RUNNING, &c->journal.flags)) { + if (!test_bit(JOURNAL_running, &c->journal.flags)) { ret = darray_make_room(&c->journal.early_journal_entries, jset_u64s(u64s)); if (ret) goto err; diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 1febab152bfa..2e8b092a1820 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -1157,12 +1157,12 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path, flags |= watermark; if (watermark < BCH_WATERMARK_reclaim && - test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)) { + test_bit(JOURNAL_space_low, &c->journal.flags)) { if (flags & BCH_TRANS_COMMIT_journal_reclaim) return ERR_PTR(-BCH_ERR_journal_reclaim_would_deadlock); ret = drop_locks_do(trans, - ({ wait_event(c->journal.wait, !test_bit(JOURNAL_SPACE_LOW, &c->journal.flags)); 0; })); + ({ wait_event(c->journal.wait, !test_bit(JOURNAL_space_low, &c->journal.flags)); 0; })); if (ret) return ERR_PTR(ret); } @@ -1362,7 +1362,7 @@ static void bch2_insert_fixup_btree_ptr(struct btree_update *as, BUG_ON(insert->k.type == KEY_TYPE_btree_ptr_v2 && !btree_ptr_sectors_written(insert)); - if (unlikely(!test_bit(JOURNAL_REPLAY_DONE, &c->journal.flags))) + if (unlikely(!test_bit(JOURNAL_replay_done, &c->journal.flags))) bch2_journal_key_overwritten(c, b->c.btree_id, b->c.level, insert->k.p); if (bch2_bkey_invalid(c, bkey_i_to_s_c(insert), diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index 4d03a77f1e83..c984cdc2795d 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -1182,11 +1182,11 @@ void bch2_fs_journal_stop(struct journal *j) cancel_delayed_work_sync(&j->write_work); BUG_ON(!bch2_journal_error(j) && - test_bit(JOURNAL_REPLAY_DONE, &j->flags) && + test_bit(JOURNAL_replay_done, &j->flags) && j->last_empty_seq != journal_cur_seq(j)); if (!bch2_journal_error(j)) - clear_bit(JOURNAL_RUNNING, &j->flags); + clear_bit(JOURNAL_running, &j->flags); } int bch2_fs_journal_start(struct journal *j, u64 cur_seq) @@ -1260,7 +1260,7 @@ int bch2_fs_journal_start(struct journal *j, u64 cur_seq) spin_lock(&j->lock); - set_bit(JOURNAL_RUNNING, &j->flags); + set_bit(JOURNAL_running, &j->flags); j->last_flush_write = jiffies; j->reservations.idx = j->reservations.unwritten_idx = journal_cur_seq(j); @@ -1401,6 +1401,13 @@ int bch2_fs_journal_init(struct journal *j) /* debug: */ +static const char * const bch2_journal_flags_strs[] = { +#define x(n) #n, + JOURNAL_FLAGS() +#undef x + NULL +}; + void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) { struct bch_fs *c = container_of(j, struct bch_fs, journal); @@ -1415,6 +1422,9 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) rcu_read_lock(); s = READ_ONCE(j->reservations); + prt_printf(out, "flags:\t"); + prt_bitflags(out, bch2_journal_flags_strs, j->flags); + prt_newline(out); prt_printf(out, "dirty journal entries:\t%llu/%llu\n", fifo_used(&j->pin), j->pin.size); prt_printf(out, "seq:\t%llu\n", journal_cur_seq(j)); prt_printf(out, "seq_ondisk:\t%llu\n", j->seq_ondisk); @@ -1453,10 +1463,6 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) prt_printf(out, "unwritten entries:\n"); bch2_journal_bufs_to_text(out, j); - prt_printf(out, - "replay done:\t%i\n", - test_bit(JOURNAL_REPLAY_DONE, &j->flags)); - prt_printf(out, "space:\n"); printbuf_indent_add(out, 2); prt_printf(out, "discarded\t%u:%u\n", diff --git a/fs/bcachefs/journal.h b/fs/bcachefs/journal.h index 4b8c709bc317..fd1f7cdaa8bc 100644 --- a/fs/bcachefs/journal.h +++ b/fs/bcachefs/journal.h @@ -372,7 +372,7 @@ static inline int bch2_journal_res_get(struct journal *j, struct journal_res *re int ret; EBUG_ON(res->ref); - EBUG_ON(!test_bit(JOURNAL_RUNNING, &j->flags)); + EBUG_ON(!test_bit(JOURNAL_running, &j->flags)); res->u64s = u64s; @@ -418,8 +418,8 @@ struct bch_dev; static inline void bch2_journal_set_replay_done(struct journal *j) { - BUG_ON(!test_bit(JOURNAL_RUNNING, &j->flags)); - set_bit(JOURNAL_REPLAY_DONE, &j->flags); + BUG_ON(!test_bit(JOURNAL_running, &j->flags)); + set_bit(JOURNAL_replay_done, &j->flags); } void bch2_journal_unblock(struct journal *); diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index dec50e7a8cb4..e76eb98af74a 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1954,14 +1954,14 @@ static int bch2_journal_write_pick_flush(struct journal *j, struct journal_buf * * So if we're in an error state, and we're still starting up, we don't * write anything at all. */ - if (error && test_bit(JOURNAL_NEED_FLUSH_WRITE, &j->flags)) + if (error && test_bit(JOURNAL_need_flush_write, &j->flags)) return -EIO; if (error || w->noflush || (!w->must_flush && (jiffies - j->last_flush_write) < msecs_to_jiffies(c->opts.journal_flush_delay) && - test_bit(JOURNAL_MAY_SKIP_FLUSH, &j->flags))) { + test_bit(JOURNAL_may_skip_flush, &j->flags))) { w->noflush = true; SET_JSET_NO_FLUSH(w->data, true); w->data->last_seq = 0; @@ -1972,7 +1972,7 @@ static int bch2_journal_write_pick_flush(struct journal *j, struct journal_buf * w->must_flush = true; j->last_flush_write = jiffies; j->nr_flush_writes++; - clear_bit(JOURNAL_NEED_FLUSH_WRITE, &j->flags); + clear_bit(JOURNAL_need_flush_write, &j->flags); } return 0; diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index a0c9f7ac611d..79be0eaddfa0 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -67,7 +67,7 @@ void bch2_journal_set_watermark(struct journal *j) track_event_change(&c->times[BCH_TIME_blocked_write_buffer_full], low_on_wb)) trace_and_count(c, journal_full, c); - mod_bit(JOURNAL_SPACE_LOW, &j->flags, low_on_space || low_on_pin); + mod_bit(JOURNAL_space_low, &j->flags, low_on_space || low_on_pin); swap(watermark, j->watermark); if (watermark > j->watermark) @@ -225,9 +225,9 @@ void bch2_journal_space_available(struct journal *j) j->space[journal_space_clean_ondisk].total) && (clean - clean_ondisk <= total / 8) && (clean_ondisk * 2 > clean)) - set_bit(JOURNAL_MAY_SKIP_FLUSH, &j->flags); + set_bit(JOURNAL_may_skip_flush, &j->flags); else - clear_bit(JOURNAL_MAY_SKIP_FLUSH, &j->flags); + clear_bit(JOURNAL_may_skip_flush, &j->flags); bch2_journal_set_watermark(j); out: @@ -818,7 +818,7 @@ static int journal_flush_done(struct journal *j, u64 seq_to_flush, * If journal replay hasn't completed, the unreplayed journal entries * hold refs on their corresponding sequence numbers */ - ret = !test_bit(JOURNAL_REPLAY_DONE, &j->flags) || + ret = !test_bit(JOURNAL_replay_done, &j->flags) || journal_last_seq(j) > seq_to_flush || !fifo_used(&j->pin); @@ -833,7 +833,7 @@ bool bch2_journal_flush_pins(struct journal *j, u64 seq_to_flush) /* time_stats this */ bool did_work = false; - if (!test_bit(JOURNAL_RUNNING, &j->flags)) + if (!test_bit(JOURNAL_running, &j->flags)) return false; closure_wait_event(&j->async_wait, diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index 45c1c48f4917..19183fcf7ad7 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -129,12 +129,17 @@ enum journal_space_from { journal_space_nr, }; +#define JOURNAL_FLAGS() \ + x(replay_done) \ + x(running) \ + x(may_skip_flush) \ + x(need_flush_write) \ + x(space_low) + enum journal_flags { - JOURNAL_REPLAY_DONE, - JOURNAL_RUNNING, - JOURNAL_MAY_SKIP_FLUSH, - JOURNAL_NEED_FLUSH_WRITE, - JOURNAL_SPACE_LOW, +#define x(n) JOURNAL_##n, + JOURNAL_FLAGS() +#undef x }; /* Reasons we may fail to get a journal reservation: */ diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 6c6087039781..5af8450098fd 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -284,7 +284,7 @@ static void __bch2_fs_read_only(struct bch_fs *c) bch_verbose(c, "flushing journal and stopping allocators complete, journal seq %llu", journal_cur_seq(&c->journal)); - if (test_bit(JOURNAL_REPLAY_DONE, &c->journal.flags) && + if (test_bit(JOURNAL_replay_done, &c->journal.flags) && !test_bit(BCH_FS_emergency_ro, &c->flags)) set_bit(BCH_FS_clean_shutdown, &c->flags); @@ -466,8 +466,8 @@ static int __bch2_fs_read_write(struct bch_fs *c, bool early) * overwriting whatever was there previously, and there must always be * at least one non-flush write in the journal or recovery will fail: */ - set_bit(JOURNAL_NEED_FLUSH_WRITE, &c->journal.flags); - set_bit(JOURNAL_RUNNING, &c->journal.flags); + set_bit(JOURNAL_need_flush_write, &c->journal.flags); + set_bit(JOURNAL_running, &c->journal.flags); for_each_rw_member(c, ca) bch2_dev_allocator_add(c, ca); -- cgit v1.2.3-59-g8ed1b From 9a768ab75bef37eac23022f296e9be63b5f85565 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 19:04:13 -0400 Subject: bcachefs: bch2_bkey_drop_ptrs() declares loop iter Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_io.c | 1 - fs/bcachefs/btree_update_interior.c | 1 - fs/bcachefs/data_update.c | 1 - fs/bcachefs/ec.c | 2 +- fs/bcachefs/extents.c | 4 ---- fs/bcachefs/extents.h | 2 +- fs/bcachefs/io_write.c | 1 - 7 files changed, 2 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index c111976fd718..9dcaccda7b93 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1865,7 +1865,6 @@ static void btree_node_write_work(struct work_struct *work) container_of(work, struct btree_write_bio, work); struct bch_fs *c = wbio->wbio.c; struct btree *b = wbio->wbio.bio.bi_private; - struct bch_extent_ptr *ptr; int ret = 0; btree_bounce_free(c, diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 2e8b092a1820..80eb862a8962 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -2491,7 +2491,6 @@ int bch2_btree_node_update_key_get_iter(struct btree_trans *trans, BUG_ON(!btree_node_hashed(b)); - struct bch_extent_ptr *ptr; bch2_bkey_drop_ptrs(bkey_i_to_s(new_key), ptr, !bch2_bkey_has_device(bkey_i_to_s(&b->key), ptr->dev)); diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index e8c7c5cec03f..88f49e7d2a16 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -467,7 +467,6 @@ int bch2_extent_drop_ptrs(struct btree_trans *trans, while (data_opts.kill_ptrs) { unsigned i = 0, drop = __fls(data_opts.kill_ptrs); - struct bch_extent_ptr *ptr; bch2_bkey_drop_ptrs(bkey_i_to_s(n), ptr, i++ == drop); data_opts.kill_ptrs ^= 1U << drop; diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 64dcf852daed..330c3da70b96 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -1200,7 +1200,7 @@ static int ec_stripe_update_extent(struct btree_trans *trans, struct btree_iter iter; struct bkey_s_c k; const struct bch_extent_ptr *ptr_c; - struct bch_extent_ptr *ptr, *ec_ptr = NULL; + struct bch_extent_ptr *ec_ptr = NULL; struct bch_extent_stripe_ptr stripe_ptr; struct bkey_i *n; int ret, dev, block; diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 1143288d7940..0d1473ba584a 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -838,8 +838,6 @@ union bch_extent_entry *bch2_bkey_drop_ptr(struct bkey_s k, void bch2_bkey_drop_device(struct bkey_s k, unsigned dev) { - struct bch_extent_ptr *ptr; - bch2_bkey_drop_ptrs(k, ptr, ptr->dev == dev); } @@ -974,8 +972,6 @@ void bch2_extent_ptr_set_cached(struct bkey_s k, struct bch_extent_ptr *ptr) */ bool bch2_extent_normalize(struct bch_fs *c, struct bkey_s k) { - struct bch_extent_ptr *ptr; - bch2_bkey_drop_ptrs(k, ptr, ptr->cached && ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr)); diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h index 528e817eacbd..10e1817e83e3 100644 --- a/fs/bcachefs/extents.h +++ b/fs/bcachefs/extents.h @@ -654,7 +654,7 @@ union bch_extent_entry *bch2_bkey_drop_ptr(struct bkey_s, do { \ struct bkey_ptrs _ptrs = bch2_bkey_ptrs(_k); \ \ - _ptr = &_ptrs.start->ptr; \ + struct bch_extent_ptr *_ptr = &_ptrs.start->ptr; \ \ while ((_ptr = bkey_ptr_next(_ptrs, _ptr))) { \ if (_cond) { \ diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 73e25250de75..e96dad0dd60d 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -481,7 +481,6 @@ static void bch2_write_done(struct closure *cl) static noinline int bch2_write_drop_io_error_ptrs(struct bch_write_op *op) { struct keylist *keys = &op->insert_keys; - struct bch_extent_ptr *ptr; struct bkey_i *src, *dst = keys->keys, *n; for (src = keys->keys; src != keys->top; src = n) { -- cgit v1.2.3-59-g8ed1b From e98786ea855cb28176e27ffce23fb163a36ed32e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 14:49:23 -0400 Subject: bcachefs: bch2_print_allocator_stuck() If we block on the allocator for more than 10 seconds, print out some useful debugging info. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 101 +++++++++++++++++++++++++++++++++++++++++ fs/bcachefs/alloc_foreground.h | 5 ++ fs/bcachefs/io_write.c | 6 ++- fs/bcachefs/journal.c | 2 +- fs/bcachefs/sysfs.c | 76 +------------------------------ 5 files changed, 114 insertions(+), 76 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 438e00c8316b..13460aab3cae 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -1630,3 +1630,104 @@ void bch2_write_points_to_text(struct printbuf *out, struct bch_fs *c) prt_str(out, "Btree write point\n"); bch2_write_point_to_text(out, c, &c->btree_write_point); } + +void bch2_fs_alloc_debug_to_text(struct printbuf *out, struct bch_fs *c) +{ + unsigned nr[BCH_DATA_NR]; + + memset(nr, 0, sizeof(nr)); + + for (unsigned i = 0; i < ARRAY_SIZE(c->open_buckets); i++) + nr[c->open_buckets[i].data_type]++; + + printbuf_tabstop_push(out, 24); + + percpu_down_read(&c->mark_lock); + prt_printf(out, "hidden\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.hidden)); + prt_printf(out, "btree\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.btree)); + prt_printf(out, "data\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.data)); + prt_printf(out, "cached\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.cached)); + prt_printf(out, "reserved\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.reserved)); + prt_printf(out, "online_reserved\t%llu\n", percpu_u64_get(c->online_reserved)); + prt_printf(out, "nr_inodes\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.nr_inodes)); + percpu_up_read(&c->mark_lock); + + prt_newline(out); + prt_printf(out, "freelist_wait\t%s\n", c->freelist_wait.list.first ? "waiting" : "empty"); + prt_printf(out, "open buckets allocated\t%i\n", OPEN_BUCKETS_COUNT - c->open_buckets_nr_free); + prt_printf(out, "open buckets total\t%u\n", OPEN_BUCKETS_COUNT); + prt_printf(out, "open_buckets_wait\t%s\n", c->open_buckets_wait.list.first ? "waiting" : "empty"); + prt_printf(out, "open_buckets_btree\t%u\n", nr[BCH_DATA_btree]); + prt_printf(out, "open_buckets_user\t%u\n", nr[BCH_DATA_user]); + prt_printf(out, "btree reserve cache\t%u\n", c->btree_reserve_cache_nr); +} + +void bch2_dev_alloc_debug_to_text(struct printbuf *out, struct bch_dev *ca) +{ + struct bch_fs *c = ca->fs; + struct bch_dev_usage stats = bch2_dev_usage_read(ca); + unsigned nr[BCH_DATA_NR]; + + memset(nr, 0, sizeof(nr)); + + for (unsigned i = 0; i < ARRAY_SIZE(c->open_buckets); i++) + nr[c->open_buckets[i].data_type]++; + + printbuf_tabstop_push(out, 12); + printbuf_tabstop_push(out, 16); + printbuf_tabstop_push(out, 16); + printbuf_tabstop_push(out, 16); + printbuf_tabstop_push(out, 16); + + bch2_dev_usage_to_text(out, &stats); + + prt_newline(out); + + prt_printf(out, "reserves:\n"); + for (unsigned i = 0; i < BCH_WATERMARK_NR; i++) + prt_printf(out, "%s\t%llu\r\n", bch2_watermarks[i], bch2_dev_buckets_reserved(ca, i)); + + prt_newline(out); + + printbuf_tabstops_reset(out); + printbuf_tabstop_push(out, 12); + printbuf_tabstop_push(out, 16); + + prt_printf(out, "open buckets\t%i\r\n", ca->nr_open_buckets); + prt_printf(out, "buckets to invalidate\t%llu\r\n", should_invalidate_buckets(ca, stats)); +} + +void bch2_print_allocator_stuck(struct bch_fs *c) +{ + struct printbuf buf = PRINTBUF; + + prt_printf(&buf, "Allocator stuck? Waited for 10 seconds\n"); + + prt_printf(&buf, "Allocator debug:\n"); + printbuf_indent_add(&buf, 2); + bch2_fs_alloc_debug_to_text(&buf, c); + printbuf_indent_sub(&buf, 2); + prt_newline(&buf); + + for_each_online_member(c, ca) { + prt_printf(&buf, "Dev %u:\n", ca->dev_idx); + printbuf_indent_add(&buf, 2); + bch2_dev_alloc_debug_to_text(&buf, ca); + printbuf_indent_sub(&buf, 2); + prt_newline(&buf); + } + + prt_printf(&buf, "Copygc debug:\n"); + printbuf_indent_add(&buf, 2); + bch2_copygc_wait_to_text(&buf, c); + printbuf_indent_sub(&buf, 2); + prt_newline(&buf); + + prt_printf(&buf, "Journal debug:\n"); + printbuf_indent_add(&buf, 2); + bch2_journal_debug_to_text(&buf, &c->journal); + printbuf_indent_sub(&buf, 2); + + bch2_print_string_as_lines(KERN_ERR, buf.buf); + printbuf_exit(&buf); +} diff --git a/fs/bcachefs/alloc_foreground.h b/fs/bcachefs/alloc_foreground.h index f0339eb7e30e..96f24b67d066 100644 --- a/fs/bcachefs/alloc_foreground.h +++ b/fs/bcachefs/alloc_foreground.h @@ -222,4 +222,9 @@ void bch2_open_buckets_partial_to_text(struct printbuf *, struct bch_fs *); void bch2_write_points_to_text(struct printbuf *, struct bch_fs *); +void bch2_fs_alloc_debug_to_text(struct printbuf *, struct bch_fs *); +void bch2_dev_alloc_debug_to_text(struct printbuf *, struct bch_dev *); + +void bch2_print_allocator_stuck(struct bch_fs *); + #endif /* _BCACHEFS_ALLOC_FOREGROUND_H */ diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index e96dad0dd60d..55e24c83fb19 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1494,7 +1494,11 @@ err: if ((op->flags & BCH_WRITE_SYNC) || (!(op->flags & BCH_WRITE_DONE) && !(op->flags & BCH_WRITE_IN_WORKER))) { - closure_sync(&op->cl); + if (closure_sync_timeout(&op->cl, HZ * 10)) { + bch2_print_allocator_stuck(c); + closure_sync(&op->cl); + } + __bch2_write_index(op); if (!(op->flags & BCH_WRITE_DONE)) diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index c984cdc2795d..adec8e1ea73e 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -1416,7 +1416,7 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) u64 nr_writes = j->nr_flush_writes + j->nr_noflush_writes; if (!out->nr_tabstops) - printbuf_tabstop_push(out, 24); + printbuf_tabstop_push(out, 28); out->atomic++; rcu_read_lock(); diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 4627b0ba179e..df3d28659bfd 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -346,37 +346,6 @@ static void bch2_gc_gens_pos_to_text(struct printbuf *out, struct bch_fs *c) prt_printf(out, "\n"); } -static void fs_alloc_debug_to_text(struct printbuf *out, struct bch_fs *c) -{ - unsigned nr[BCH_DATA_NR]; - - memset(nr, 0, sizeof(nr)); - - for (unsigned i = 0; i < ARRAY_SIZE(c->open_buckets); i++) - nr[c->open_buckets[i].data_type]++; - - printbuf_tabstop_push(out, 24); - - percpu_down_read(&c->mark_lock); - prt_printf(out, "hidden\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.hidden)); - prt_printf(out, "btree\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.btree)); - prt_printf(out, "data\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.data)); - prt_printf(out, "cached\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.cached)); - prt_printf(out, "reserved\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.reserved)); - prt_printf(out, "online_reserved\t%llu\n", percpu_u64_get(c->online_reserved)); - prt_printf(out, "nr_inodes\t%llu\n", bch2_fs_usage_read_one(c, &c->usage_base->b.nr_inodes)); - percpu_up_read(&c->mark_lock); - - prt_newline(out); - prt_printf(out, "freelist_wait\t%s\n", c->freelist_wait.list.first ? "waiting" : "empty"); - prt_printf(out, "open buckets allocated\t%i\n", OPEN_BUCKETS_COUNT - c->open_buckets_nr_free); - prt_printf(out, "open buckets total\t%u\n", OPEN_BUCKETS_COUNT); - prt_printf(out, "open_buckets_wait\t%s\n", c->open_buckets_wait.list.first ? "waiting" : "empty"); - prt_printf(out, "open_buckets_btree\t%u\n", nr[BCH_DATA_btree]); - prt_printf(out, "open_buckets_user\t%u\n", nr[BCH_DATA_user]); - prt_printf(out, "btree reserve cache\t%u\n", c->btree_reserve_cache_nr); -} - SHOW(bch2_fs) { struct bch_fs *c = container_of(kobj, struct bch_fs, kobj); @@ -458,7 +427,7 @@ SHOW(bch2_fs) bch2_disk_groups_to_text(out, c); if (attr == &sysfs_alloc_debug) - fs_alloc_debug_to_text(out, c); + bch2_fs_alloc_debug_to_text(out, c); return 0; } @@ -779,47 +748,6 @@ struct attribute *bch2_fs_time_stats_files[] = { NULL }; -static void dev_alloc_debug_to_text(struct printbuf *out, struct bch_dev *ca) -{ - struct bch_fs *c = ca->fs; - struct bch_dev_usage stats = bch2_dev_usage_read(ca); - unsigned nr[BCH_DATA_NR]; - - memset(nr, 0, sizeof(nr)); - - for (unsigned i = 0; i < ARRAY_SIZE(c->open_buckets); i++) - nr[c->open_buckets[i].data_type]++; - - printbuf_tabstop_push(out, 8); - printbuf_tabstop_push(out, 16); - printbuf_tabstop_push(out, 16); - printbuf_tabstop_push(out, 16); - printbuf_tabstop_push(out, 16); - - bch2_dev_usage_to_text(out, &stats); - - prt_newline(out); - - prt_printf(out, "reserves:\n"); - for (unsigned i = 0; i < BCH_WATERMARK_NR; i++) - prt_printf(out, "%s\t%llu\r\n", bch2_watermarks[i], bch2_dev_buckets_reserved(ca, i)); - - prt_newline(out); - - printbuf_tabstops_reset(out); - printbuf_tabstop_push(out, 24); - - prt_printf(out, "freelist_wait\t%s\n", c->freelist_wait.list.first ? "waiting" : "empty"); - prt_printf(out, "open buckets allocated\t%i\n", OPEN_BUCKETS_COUNT - c->open_buckets_nr_free); - prt_printf(out, "open buckets this dev\t%i\n", ca->nr_open_buckets); - prt_printf(out, "open buckets total\t%u\n", OPEN_BUCKETS_COUNT); - prt_printf(out, "open_buckets_wait\t%s\n", c->open_buckets_wait.list.first ? "waiting" : "empty"); - prt_printf(out, "open_buckets_btree\t%u\n", nr[BCH_DATA_btree]); - prt_printf(out, "open_buckets_user\t%u\n", nr[BCH_DATA_user]); - prt_printf(out, "buckets_to_invalidate\t%llu\n", should_invalidate_buckets(ca, stats)); - prt_printf(out, "btree reserve cache\t%u\n", c->btree_reserve_cache_nr); -} - static const char * const bch2_rw[] = { "read", "write", @@ -889,7 +817,7 @@ SHOW(bch2_dev) * 100 / CONGESTED_MAX); if (attr == &sysfs_alloc_debug) - dev_alloc_debug_to_text(out, ca); + bch2_dev_alloc_debug_to_text(out, ca); return 0; } -- cgit v1.2.3-59-g8ed1b From f295298b8c6413f0ed2a5a69dd7f32409cc54f1d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 17:39:16 -0400 Subject: bcachefs: New helpers for device refcounts This will be used in the next patch for adding some new debug mode asserts. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 9 ++++----- fs/bcachefs/alloc_foreground.c | 6 +++--- fs/bcachefs/btree_gc.c | 12 +++++------- fs/bcachefs/buckets.c | 4 ++-- fs/bcachefs/chardev.c | 26 ++++++++++++-------------- fs/bcachefs/data_update.c | 6 +++--- fs/bcachefs/disk_groups.c | 2 +- fs/bcachefs/sb-members.h | 22 ++++++++++++++++++---- fs/bcachefs/super.c | 4 ++-- 9 files changed, 50 insertions(+), 41 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 6b3bad385812..e24656f4b710 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1656,10 +1656,9 @@ static void discard_buckets_next_dev(struct bch_fs *c, struct discard_buckets_st bch2_dev_usage_read(s->ca).d[BCH_DATA_free].buckets) bch2_journal_flush_async(&c->journal, NULL); - if (s->ca) - percpu_ref_put(&s->ca->ref); + bch2_dev_put(s->ca); if (ca) - percpu_ref_get(&ca->ref); + bch2_dev_get(ca); s->ca = ca; s->need_journal_commit_this_dev = 0; } @@ -2014,7 +2013,7 @@ static void bch2_do_invalidates_work(struct work_struct *work) invalidate_one_bucket(trans, &iter, k, &nr_to_invalidate)); if (ret < 0) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); break; } } @@ -2151,7 +2150,7 @@ int bch2_fs_freespace_init(struct bch_fs *c) ret = bch2_dev_freespace_init(c, ca, 0, ca->mi.nbuckets); if (ret) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); bch_err_fn(c, ret); return ret; } diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 13460aab3cae..92cafde165c0 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -733,21 +733,21 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, rcu_read_lock(); ca = rcu_dereference(c->devs[dev]); if (ca) - percpu_ref_get(&ca->ref); + bch2_dev_get(ca); rcu_read_unlock(); if (!ca) continue; if (!ca->mi.durability && *have_cache) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); continue; } ob = bch2_bucket_alloc_trans(trans, ca, watermark, data_type, cl, &usage); if (!IS_ERR(ob)) bch2_dev_stripe_increment_inlined(ca, stripe, &usage); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); if (IS_ERR(ob)) { ret = PTR_ERR(ob); diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index e5289b53d175..133c48c4047d 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -815,10 +815,8 @@ static int bch2_gc_done(struct bch_fs *c) #undef copy_stripe_field #undef copy_field fsck_err: - if (ca) - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); bch_err_fn(c, ret); - percpu_up_write(&c->mark_lock); printbuf_exit(&buf); return ret; @@ -841,7 +839,7 @@ static int bch2_gc_start(struct bch_fs *c) ca->usage_gc = alloc_percpu(struct bch_dev_usage); if (!ca->usage_gc) { bch_err(c, "error allocating ca->usage_gc"); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return -BCH_ERR_ENOMEM_gc_start; } @@ -969,7 +967,7 @@ static int bch2_gc_alloc_done(struct bch_fs *c) NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, bch2_alloc_write_key(trans, &iter, k))); if (ret) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); break; } } @@ -985,7 +983,7 @@ static int bch2_gc_alloc_start(struct bch_fs *c) ca->mi.nbuckets * sizeof(struct bucket), GFP_KERNEL|__GFP_ZERO); if (!buckets) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); bch_err(c, "error allocating ca->buckets[gc]"); return -BCH_ERR_ENOMEM_gc_alloc_start; } @@ -1330,7 +1328,7 @@ int bch2_gc_gens(struct bch_fs *c) ca->oldest_gen = kvmalloc(gens->nbuckets, GFP_KERNEL); if (!ca->oldest_gen) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); ret = -BCH_ERR_ENOMEM_gc_gens; goto err; } diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 525c72c0277c..525a7548fab7 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -1438,7 +1438,7 @@ int bch2_trans_mark_dev_sbs_flags(struct bch_fs *c, for_each_online_member(c, ca) { int ret = bch2_trans_mark_dev_sb(c, ca, flags); if (ret) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return ret; } } @@ -1536,7 +1536,7 @@ int bch2_buckets_nouse_alloc(struct bch_fs *c) sizeof(unsigned long), GFP_KERNEL|__GFP_ZERO); if (!ca->buckets_nouse) { - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return -BCH_ERR_ENOMEM_buckets_nouse; } } diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index dbcd433db004..ed1fe771a426 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -35,7 +35,7 @@ static struct bch_dev *bch2_device_lookup(struct bch_fs *c, u64 dev, rcu_read_lock(); ca = rcu_dereference(c->devs[dev]); if (ca) - percpu_ref_get(&ca->ref); + bch2_dev_get(ca); rcu_read_unlock(); if (!ca) @@ -391,7 +391,7 @@ static long bch2_ioctl_disk_offline(struct bch_fs *c, struct bch_ioctl_disk arg) return PTR_ERR(ca); ret = bch2_dev_offline(c, ca, arg.flags); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return ret; } @@ -420,7 +420,7 @@ static long bch2_ioctl_disk_set_state(struct bch_fs *c, if (ret) bch_err(c, "Error setting device state: %s", bch2_err_str(ret)); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return ret; } @@ -615,7 +615,7 @@ static long bch2_ioctl_dev_usage(struct bch_fs *c, arg.d[i].fragmented = src.d[i].fragmented; } - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return copy_to_user_errcode(user_arg, &arg, sizeof(arg)); } @@ -667,7 +667,7 @@ static long bch2_ioctl_dev_usage_v2(struct bch_fs *c, goto err; } err: - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return ret; } @@ -689,11 +689,9 @@ static long bch2_ioctl_read_super(struct bch_fs *c, if (arg.flags & BCH_READ_DEV) { ca = bch2_device_lookup(c, arg.dev, arg.flags); - - if (IS_ERR(ca)) { - ret = PTR_ERR(ca); - goto err; - } + ret = PTR_ERR_OR_ZERO(ca); + if (ret) + goto err_unlock; sb = ca->disk_sb.sb; } else { @@ -708,8 +706,8 @@ static long bch2_ioctl_read_super(struct bch_fs *c, ret = copy_to_user_errcode((void __user *)(unsigned long)arg.sb, sb, vstruct_bytes(sb)); err: - if (!IS_ERR_OR_NULL(ca)) - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); +err_unlock: mutex_unlock(&c->sb_lock); return ret; } @@ -753,7 +751,7 @@ static long bch2_ioctl_disk_resize(struct bch_fs *c, ret = bch2_dev_resize(c, ca, arg.nbuckets); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return ret; } @@ -779,7 +777,7 @@ static long bch2_ioctl_disk_resize_journal(struct bch_fs *c, ret = bch2_set_nr_journal_buckets(c, ca, arg.nbuckets); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return ret; } diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 88f49e7d2a16..c252f0619b73 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -360,7 +360,7 @@ void bch2_data_update_exit(struct data_update *update) if (c->opts.nocow_enabled) bch2_bucket_nocow_unlock(&c->nocow_locks, PTR_BUCKET_POS(c, ptr), 0); - percpu_ref_put(&bch2_dev_bkey_exists(c, ptr->dev)->ref); + bch2_dev_put(bch2_dev_bkey_exists(c, ptr->dev)); } bch2_bkey_buf_exit(&update->k, c); @@ -541,7 +541,7 @@ int bch2_data_update_init(struct btree_trans *trans, m->op.watermark = m->data_opts.btree_insert_flags & BCH_WATERMARK_MASK; bkey_for_each_ptr(ptrs, ptr) - percpu_ref_get(&bch2_dev_bkey_exists(c, ptr->dev)->ref); + bch2_dev_get(bch2_dev_bkey_exists(c, ptr->dev)); unsigned durability_have = 0, durability_removing = 0; @@ -653,7 +653,7 @@ err: if ((1U << i) & ptrs_locked) bch2_bucket_nocow_unlock(&c->nocow_locks, PTR_BUCKET_POS(c, &p.ptr), 0); - percpu_ref_put(&bch2_dev_bkey_exists(c, p.ptr.dev)->ref); + bch2_dev_put(bch2_dev_bkey_exists(c, p.ptr.dev)); i++; } diff --git a/fs/bcachefs/disk_groups.c b/fs/bcachefs/disk_groups.c index 04b6aa414476..1cb563dfa017 100644 --- a/fs/bcachefs/disk_groups.c +++ b/fs/bcachefs/disk_groups.c @@ -523,7 +523,7 @@ int bch2_opt_target_parse(struct bch_fs *c, const char *val, u64 *res, ca = bch2_dev_lookup(c, val); if (!IS_ERR(ca)) { *res = dev_to_target(ca->dev_idx); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); return 0; } diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index 53034cea5843..0aeb7285dc8c 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -105,14 +105,28 @@ static inline struct bch_dev *__bch2_next_dev(struct bch_fs *c, struct bch_dev * for (struct bch_dev *_ca = NULL; \ (_ca = __bch2_next_dev((_c), _ca, (_mask)));) -static inline struct bch_dev *bch2_get_next_dev(struct bch_fs *c, struct bch_dev *ca) +static inline void bch2_dev_get(struct bch_dev *ca) +{ + percpu_ref_get(&ca->ref); +} + +static inline void __bch2_dev_put(struct bch_dev *ca) +{ + percpu_ref_put(&ca->ref); +} + +static inline void bch2_dev_put(struct bch_dev *ca) { - rcu_read_lock(); if (ca) - percpu_ref_put(&ca->ref); + __bch2_dev_put(ca); +} +static inline struct bch_dev *bch2_get_next_dev(struct bch_fs *c, struct bch_dev *ca) +{ + rcu_read_lock(); + bch2_dev_put(ca); if ((ca = __bch2_next_dev(c, ca, NULL))) - percpu_ref_get(&ca->ref); + bch2_dev_get(ca); rcu_read_unlock(); return ca; diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 5af8450098fd..8182cc971e7c 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -710,7 +710,7 @@ static int bch2_fs_online(struct bch_fs *c) ret = bch2_dev_sysfs_online(c, ca); if (ret) { bch_err(c, "error creating sysfs objects"); - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); goto err; } } @@ -1613,7 +1613,7 @@ int bch2_dev_remove(struct bch_fs *c, struct bch_dev *ca, int flags) * We consume a reference to ca->ref, regardless of whether we succeed * or fail: */ - percpu_ref_put(&ca->ref); + bch2_dev_put(ca); if (!bch2_dev_state_allowed(c, ca, BCH_MEMBER_STATE_failed, flags)) { bch_err(ca, "Cannot remove without losing data"); -- cgit v1.2.3-59-g8ed1b From 552aa5486579c18b4f8e7ca03ef88fa573c517b5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 18:07:40 -0400 Subject: bcachefs: Debug asserts for ca->ref Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 6 ++++++ fs/bcachefs/sb-members.h | 13 +++++++++++++ fs/bcachefs/super.c | 21 ++++++++++++++++++--- 3 files changed, 37 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 82544f826c58..f5ba8c6a34d7 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -537,7 +537,13 @@ struct io_count { struct bch_dev { struct kobject kobj; +#ifdef CONFIG_BCACHEFS_DEBUG + atomic_long_t ref; + bool dying; + unsigned long last_put; +#else struct percpu_ref ref; +#endif struct completion ref_completion; struct percpu_ref io_ref; struct completion io_ref_completion; diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index 0aeb7285dc8c..ecb8284af0de 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -107,12 +107,25 @@ static inline struct bch_dev *__bch2_next_dev(struct bch_fs *c, struct bch_dev * static inline void bch2_dev_get(struct bch_dev *ca) { +#ifdef CONFIG_BCACHEFS_DEBUG + BUG_ON(atomic_long_inc_return(&ca->ref) <= 1L); +#else percpu_ref_get(&ca->ref); +#endif } static inline void __bch2_dev_put(struct bch_dev *ca) { +#ifdef CONFIG_BCACHEFS_DEBUG + long r = atomic_long_dec_return(&ca->ref); + if (r < (long) !ca->dying) + panic("bch_dev->ref underflow, last put: %pS\n", (void *) ca->last_put); + ca->last_put = _THIS_IP_; + if (!r) + complete(&ca->ref_completion); +#else percpu_ref_put(&ca->ref); +#endif } static inline void bch2_dev_put(struct bch_dev *ca) diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 8182cc971e7c..ed135166ab86 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -656,6 +656,7 @@ void bch2_fs_free(struct bch_fs *c) struct bch_dev *ca = rcu_dereference_protected(c->devs[i], true); if (ca) { + EBUG_ON(atomic_long_read(&ca->ref) != 1); bch2_free_super(&ca->disk_sb); bch2_dev_free(ca); } @@ -1200,7 +1201,9 @@ static void bch2_dev_free(struct bch_dev *ca) bch2_time_stats_quantiles_exit(&ca->io_latency[READ]); percpu_ref_exit(&ca->io_ref); +#ifndef CONFIG_BCACHEFS_DEBUG percpu_ref_exit(&ca->ref); +#endif kobject_put(&ca->kobj); } @@ -1227,12 +1230,14 @@ static void __bch2_dev_offline(struct bch_fs *c, struct bch_dev *ca) bch2_dev_journal_exit(ca); } +#ifndef CONFIG_BCACHEFS_DEBUG static void bch2_dev_ref_complete(struct percpu_ref *ref) { struct bch_dev *ca = container_of(ref, struct bch_dev, ref); complete(&ca->ref_completion); } +#endif static void bch2_dev_io_ref_complete(struct percpu_ref *ref) { @@ -1301,9 +1306,14 @@ static struct bch_dev *__bch2_dev_alloc(struct bch_fs *c, ca->nr_btree_reserve = DIV_ROUND_UP(BTREE_NODE_RESERVE, ca->mi.bucket_size / btree_sectors(c)); - if (percpu_ref_init(&ca->ref, bch2_dev_ref_complete, - 0, GFP_KERNEL) || - percpu_ref_init(&ca->io_ref, bch2_dev_io_ref_complete, +#ifndef CONFIG_BCACHEFS_DEBUG + if (percpu_ref_init(&ca->ref, bch2_dev_ref_complete, 0, GFP_KERNEL)) + goto err; +#else + atomic_long_set(&ca->ref, 1); +#endif + + if (percpu_ref_init(&ca->io_ref, bch2_dev_io_ref_complete, PERCPU_REF_INIT_DEAD, GFP_KERNEL) || !(ca->sb_read_scratch = (void *) __get_free_page(GFP_KERNEL)) || bch2_dev_buckets_alloc(c, ca) || @@ -1665,7 +1675,12 @@ int bch2_dev_remove(struct bch_fs *c, struct bch_dev *ca, int flags) rcu_assign_pointer(c->devs[ca->dev_idx], NULL); mutex_unlock(&c->sb_lock); +#ifndef CONFIG_BCACHEFS_DEBUG percpu_ref_kill(&ca->ref); +#else + ca->dying = true; + bch2_dev_put(ca); +#endif wait_for_completion(&ca->ref_completion); bch2_dev_free(ca); -- cgit v1.2.3-59-g8ed1b From 23f308ae19d345c8fc022edc5aae9a0af172ad73 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:30:35 -0400 Subject: bcachefs: bch2_dev_safe() -> bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 2 +- fs/bcachefs/sb-members.c | 1 + fs/bcachefs/sb-members.h | 2 +- 3 files changed, 3 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 0d1473ba584a..6376440f6f30 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -983,7 +983,7 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struc { out->atomic++; rcu_read_lock(); - struct bch_dev *ca = bch2_dev_safe(c, ptr->dev); + struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); if (!ca) { prt_printf(out, "ptr: %u:%llu gen %u%s", ptr->dev, (u64) ptr->offset, ptr->gen, diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 52054f26982f..96a49ad0525c 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -3,6 +3,7 @@ #include "bcachefs.h" #include "btree_cache.h" #include "disk_groups.h" +#include "error.h" #include "opts.h" #include "replicas.h" #include "sb-members.h" diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index ecb8284af0de..bea4c2efbb6e 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -210,7 +210,7 @@ static inline struct bch_dev *bch2_dev_locked(struct bch_fs *c, unsigned dev) lockdep_is_held(&c->state_lock)); } -static inline struct bch_dev *bch2_dev_safe(struct bch_fs *c, unsigned dev) +static inline struct bch_dev *bch2_dev_rcu(struct bch_fs *c, unsigned dev) { return c && dev < c->sb.nr_devices ? rcu_dereference(c->devs[dev]) -- cgit v1.2.3-59-g8ed1b From f5faf43f853abb40255059d2b5133699c8528eaa Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:44:24 -0400 Subject: bcachefs: Pass device to bch2_alloc_write_key() More elimating bch2_dev_bkey_exists() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 133c48c4047d..07b343e7ddeb 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -865,10 +865,10 @@ static inline bool bch2_alloc_v4_cmp(struct bch_alloc_v4 l, static int bch2_alloc_write_key(struct btree_trans *trans, struct btree_iter *iter, + struct bch_dev *ca, struct bkey_s_c k) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, iter->pos.inode); struct bkey_i_alloc_v4 *a; struct bch_alloc_v4 old_gc, gc, old_convert, new; const struct bch_alloc_v4 *old; @@ -965,7 +965,7 @@ static int bch2_gc_alloc_done(struct bch_fs *c) POS(ca->dev_idx, ca->mi.nbuckets - 1), BTREE_ITER_slots|BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, - bch2_alloc_write_key(trans, &iter, k))); + bch2_alloc_write_key(trans, &iter, ca, k))); if (ret) { bch2_dev_put(ca); break; -- cgit v1.2.3-59-g8ed1b From 267039d0fc78d6f9689b7314ba4b9efce95b37f5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:41:48 -0400 Subject: bcachefs: Pass device to bch2_bucket_do_index() Eliminating bch2_dev_bkey_exists() uses and replacing them with proper checks; this one was unnecessary since the caller already has it. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index e24656f4b710..97d70def49b3 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -625,12 +625,12 @@ int bch2_alloc_read(struct bch_fs *c) /* Free space/discard btree: */ static int bch2_bucket_do_index(struct btree_trans *trans, + struct bch_dev *ca, struct bkey_s_c alloc_k, const struct bch_alloc_v4 *a, bool set) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, alloc_k.k->p.inode); struct btree_iter iter; struct bkey_s_c old; struct bkey_i *k; @@ -770,8 +770,8 @@ int bch2_trigger_alloc(struct btree_trans *trans, if (old_a->data_type != new_a->data_type || (new_a->data_type == BCH_DATA_free && alloc_freespace_genbits(*old_a) != alloc_freespace_genbits(*new_a))) { - ret = bch2_bucket_do_index(trans, old, old_a, false) ?: - bch2_bucket_do_index(trans, new.s_c, new_a, true); + ret = bch2_bucket_do_index(trans, ca, old, old_a, false) ?: + bch2_bucket_do_index(trans, ca, new.s_c, new_a, true); if (ret) return ret; } @@ -790,8 +790,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, return ret; } - new_a->fragmentation_lru = alloc_lru_idx_fragmentation(*new_a, - bch2_dev_bkey_exists(c, new.k->p.inode)); + new_a->fragmentation_lru = alloc_lru_idx_fragmentation(*new_a, ca); if (old_a->fragmentation_lru != new_a->fragmentation_lru) { ret = bch2_lru_change(trans, BCH_LRU_FRAGMENTATION_START, @@ -2078,7 +2077,7 @@ int bch2_dev_freespace_init(struct bch_fs *c, struct bch_dev *ca, struct bch_alloc_v4 a_convert; const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert); - ret = bch2_bucket_do_index(trans, k, a, true) ?: + ret = bch2_bucket_do_index(trans, ca, k, a, true) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); if (ret) -- cgit v1.2.3-59-g8ed1b From 13a16dabde3394350a877a6a11a90887ca899668 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 16:11:15 -0400 Subject: bcachefs: bch2_dev_btree_bitmap_marked() -> bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-members.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 96a49ad0525c..e4a9c858ca8e 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -398,11 +398,20 @@ void bch2_dev_errors_reset(struct bch_dev *ca) bool bch2_dev_btree_bitmap_marked(struct bch_fs *c, struct bkey_s_c k) { - bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) - if (!bch2_dev_btree_bitmap_marked_sectors(bch2_dev_bkey_exists(c, ptr->dev), - ptr->offset, btree_sectors(c))) - return false; - return true; + bool ret = true; + rcu_read_lock(); + bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) { + struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); + if (!ca) + continue; + + if (!bch2_dev_btree_bitmap_marked_sectors(ca, ptr->offset, btree_sectors(c))) { + ret = false; + break; + } + } + rcu_read_unlock(); + return ret; } static void __bch2_dev_btree_bitmap_mark(struct bch_sb_field_members_v2 *mi, unsigned dev, @@ -440,6 +449,10 @@ void bch2_dev_btree_bitmap_mark(struct bch_fs *c, struct bkey_s_c k) lockdep_assert_held(&c->sb_lock); struct bch_sb_field_members_v2 *mi = bch2_sb_field_get(c->disk_sb.sb, members_v2); - bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) + bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) { + if (!bch2_member_exists(c->disk_sb.sb, ptr->dev)) + continue; + __bch2_dev_btree_bitmap_mark(mi, ptr->dev, ptr->offset, btree_sectors(c)); + } } -- cgit v1.2.3-59-g8ed1b From adf81796ee9cf2dd46214962a18015bffddbdbd2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:45:26 -0400 Subject: bcachefs: journal_replay_entry_early() checks for nonexistent device Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index a6447ffd336e..1266916ac03f 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -373,14 +373,17 @@ static int journal_replay_entry_early(struct bch_fs *c, case BCH_JSET_ENTRY_dev_usage: { struct jset_entry_dev_usage *u = container_of(entry, struct jset_entry_dev_usage, entry); - struct bch_dev *ca = bch2_dev_bkey_exists(c, le32_to_cpu(u->dev)); - unsigned i, nr_types = jset_entry_dev_usage_nr_types(u); - - for (i = 0; i < min_t(unsigned, nr_types, BCH_DATA_NR); i++) { - ca->usage_base->d[i].buckets = le64_to_cpu(u->d[i].buckets); - ca->usage_base->d[i].sectors = le64_to_cpu(u->d[i].sectors); - ca->usage_base->d[i].fragmented = le64_to_cpu(u->d[i].fragmented); - } + unsigned nr_types = jset_entry_dev_usage_nr_types(u); + + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, le32_to_cpu(u->dev)); + if (ca) + for (unsigned i = 0; i < min_t(unsigned, nr_types, BCH_DATA_NR); i++) { + ca->usage_base->d[i].buckets = le64_to_cpu(u->d[i].buckets); + ca->usage_base->d[i].sectors = le64_to_cpu(u->d[i].sectors); + ca->usage_base->d[i].fragmented = le64_to_cpu(u->d[i].fragmented); + } + rcu_read_unlock(); break; } -- cgit v1.2.3-59-g8ed1b From 6349b07c25a03dbad3ff0e9a20a27a367ddd2997 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:45:05 -0400 Subject: bcachefs: bch2_have_enough_devs() checks for nonexistent device Signed-off-by: Kent Overstreet --- fs/bcachefs/replicas.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c index f2c99703c730..f8ff7c8bb05e 100644 --- a/fs/bcachefs/replicas.c +++ b/fs/bcachefs/replicas.c @@ -947,18 +947,20 @@ bool bch2_have_enough_devs(struct bch_fs *c, struct bch_devs_mask devs, percpu_down_read(&c->mark_lock); for_each_cpu_replicas_entry(&c->replicas, e) { - unsigned i, nr_online = 0, nr_failed = 0, dflags = 0; + unsigned nr_online = 0, nr_failed = 0, dflags = 0; bool metadata = e->data_type < BCH_DATA_user; if (e->data_type == BCH_DATA_cached) continue; - for (i = 0; i < e->nr_devs; i++) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, e->devs[i]); - + rcu_read_lock(); + for (unsigned i = 0; i < e->nr_devs; i++) { nr_online += test_bit(e->devs[i], devs.d); - nr_failed += ca->mi.state == BCH_MEMBER_STATE_failed; + + struct bch_dev *ca = bch2_dev_rcu(c, e->devs[i]); + nr_failed += ca && ca->mi.state == BCH_MEMBER_STATE_failed; } + rcu_read_unlock(); if (nr_failed == e->nr_devs) continue; -- cgit v1.2.3-59-g8ed1b From b07eb8252fb1455cc54bb306e253e193a7a39228 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:30:35 -0400 Subject: bcachefs: bch2_dev_tryget() Most uses of bch2_dev_bkey_exists() are going away, where we assume that because a key references a device the device most exist - instead, we'll be explicitly checking if the device exists and getting a reference to it. This adds the new helpers. Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-members.c | 5 +++++ fs/bcachefs/sb-members.h | 20 ++++++++++++++++++++ 2 files changed, 25 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index e4a9c858ca8e..2ca557bffc0e 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -9,6 +9,11 @@ #include "sb-members.h" #include "super-io.h" +void bch2_dev_missing(struct bch_fs *c, unsigned dev) +{ + bch2_fs_inconsistent(c, "pointer to nonexistent device %u", dev); +} + #define x(t, n, ...) [n] = #t, static const char * const bch2_iops_measurements[] = { BCH_IOPS_MEASUREMENTS() diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index bea4c2efbb6e..714559e2ef59 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -217,6 +217,26 @@ static inline struct bch_dev *bch2_dev_rcu(struct bch_fs *c, unsigned dev) : NULL; } +static inline struct bch_dev *bch2_dev_tryget_noerror(struct bch_fs *c, unsigned dev) +{ + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, dev); + if (ca) + bch2_dev_get(ca); + rcu_read_unlock(); + return ca; +} + +void bch2_dev_missing(struct bch_fs *, unsigned); + +static inline struct bch_dev *bch2_dev_tryget(struct bch_fs *c, unsigned dev) +{ + struct bch_dev *ca = bch2_dev_tryget_noerror(c, dev); + if (!ca) + bch2_dev_missing(c, dev); + return ca; +} + /* XXX kill, move to struct bch_fs */ static inline struct bch_devs_mask bch2_online_devs(struct bch_fs *c) { -- cgit v1.2.3-59-g8ed1b From 4cd91e2f87a6b1f28806fc0082a9b31ce23d28b8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 17:48:23 -0400 Subject: bcachefs: Convert to bch2_dev_tryget_noerror() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 15 +++------------ fs/bcachefs/chardev.c | 7 +------ 2 files changed, 4 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 92cafde165c0..f54d90868314 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -717,25 +717,16 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, struct bch_fs *c = trans->c; struct dev_alloc_list devs_sorted = bch2_dev_alloc_list(c, stripe, devs_may_alloc); - unsigned dev; - struct bch_dev *ca; int ret = -BCH_ERR_insufficient_devices; - unsigned i; BUG_ON(*nr_effective >= nr_replicas); - for (i = 0; i < devs_sorted.nr; i++) { + for (unsigned i = 0; i < devs_sorted.nr; i++) { struct bch_dev_usage usage; struct open_bucket *ob; - dev = devs_sorted.devs[i]; - - rcu_read_lock(); - ca = rcu_dereference(c->devs[dev]); - if (ca) - bch2_dev_get(ca); - rcu_read_unlock(); - + unsigned dev = devs_sorted.devs[i]; + struct bch_dev *ca = bch2_dev_tryget_noerror(c, dev); if (!ca) continue; diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index ed1fe771a426..9e54323f0f5f 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -32,12 +32,7 @@ static struct bch_dev *bch2_device_lookup(struct bch_fs *c, u64 dev, if (dev >= c->sb.nr_devices) return ERR_PTR(-EINVAL); - rcu_read_lock(); - ca = rcu_dereference(c->devs[dev]); - if (ca) - bch2_dev_get(ca); - rcu_read_unlock(); - + ca = bch2_dev_tryget_noerror(c, dev); if (!ca) return ERR_PTR(-EINVAL); } else { -- cgit v1.2.3-59-g8ed1b From 9b3059a1b3e9a71e345edf52f8487fc881d9f414 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:43:20 -0400 Subject: bcachefs: bch2_check_alloc_key() -> bch2_dev_tryget_noerror() More elimination of bch2_dev_bkey_exists() usage. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 17 ++++++++++------- fs/bcachefs/buckets.h | 5 ----- fs/bcachefs/sb-members.c | 5 +++++ fs/bcachefs/sb-members.h | 25 +++++++++++++++++++++++++ 4 files changed, 40 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 97d70def49b3..e61f00efe619 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1024,24 +1024,25 @@ int bch2_check_alloc_key(struct btree_trans *trans, struct btree_iter *bucket_gens_iter) { struct bch_fs *c = trans->c; - struct bch_dev *ca; struct bch_alloc_v4 a_convert; const struct bch_alloc_v4 *a; unsigned discard_key_type, freespace_key_type; unsigned gens_offset; struct bkey_s_c k; struct printbuf buf = PRINTBUF; - int ret; + int ret = 0; - if (fsck_err_on(!bch2_dev_bucket_exists(c, alloc_k.k->p), c, - alloc_key_to_missing_dev_bucket, + struct bch_dev *ca = bch2_dev_bucket_tryget_noerror(c, alloc_k.k->p); + if (fsck_err_on(!ca, + c, alloc_key_to_missing_dev_bucket, "alloc key for invalid device:bucket %llu:%llu", alloc_k.k->p.inode, alloc_k.k->p.offset)) - return bch2_btree_delete_at(trans, alloc_iter, 0); + ret = bch2_btree_delete_at(trans, alloc_iter, 0); + if (!ca) + return ret; - ca = bch2_dev_bkey_exists(c, alloc_k.k->p.inode); if (!ca->mi.freespace_initialized) - return 0; + goto out; a = bch2_alloc_to_v4(alloc_k, &a_convert); @@ -1140,8 +1141,10 @@ int bch2_check_alloc_key(struct btree_trans *trans, if (ret) goto err; } +out: err: fsck_err: + bch2_dev_put(ca); printbuf_exit(&buf); return ret; } diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index f352d88d9b8e..95cf3c290734 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -35,11 +35,6 @@ static inline u64 sector_to_bucket_and_offset(const struct bch_dev *ca, sector_t return div_u64_rem(s, ca->mi.bucket_size, offset); } -static inline bool bucket_valid(const struct bch_dev *ca, u64 b) -{ - return b - ca->mi.first_bucket < ca->mi.nbuckets_minus_first; -} - #define for_each_bucket(_b, _buckets) \ for (_b = (_buckets)->b + (_buckets)->first_bucket; \ _b < (_buckets)->b + (_buckets)->nbuckets; _b++) diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 2ca557bffc0e..8f197bb088a0 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -14,6 +14,11 @@ void bch2_dev_missing(struct bch_fs *c, unsigned dev) bch2_fs_inconsistent(c, "pointer to nonexistent device %u", dev); } +void bch2_dev_bucket_missing(struct bch_fs *c, struct bpos bucket) +{ + bch2_fs_inconsistent(c, "pointer to nonexistent bucket %llu:%llu", bucket.inode, bucket.offset); +} + #define x(t, n, ...) [n] = #t, static const char * const bch2_iops_measurements[] = { BCH_IOPS_MEASUREMENTS() diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index 714559e2ef59..503b56051d7e 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -190,6 +190,11 @@ static inline bool bch2_dev_exists(const struct bch_fs *c, unsigned dev) return dev < c->sb.nr_devices && c->devs[dev]; } +static inline bool bucket_valid(const struct bch_dev *ca, u64 b) +{ + return b - ca->mi.first_bucket < ca->mi.nbuckets_minus_first; +} + /* * If a key exists that references a device, the device won't be going away and * we can omit rcu_read_lock(): @@ -237,6 +242,26 @@ static inline struct bch_dev *bch2_dev_tryget(struct bch_fs *c, unsigned dev) return ca; } +static inline struct bch_dev *bch2_dev_bucket_tryget_noerror(struct bch_fs *c, struct bpos bucket) +{ + struct bch_dev *ca = bch2_dev_tryget_noerror(c, bucket.inode); + if (ca && !bucket_valid(ca, bucket.offset)) { + bch2_dev_put(ca); + ca = NULL; + } + return ca; +} + +void bch2_dev_bucket_missing(struct bch_fs *, struct bpos); + +static inline struct bch_dev *bch2_dev_bucket_tryget(struct bch_fs *c, struct bpos bucket) +{ + struct bch_dev *ca = bch2_dev_bucket_tryget_noerror(c, bucket); + if (!ca) + bch2_dev_bucket_missing(c, bucket); + return ca; +} + /* XXX kill, move to struct bch_fs */ static inline struct bch_devs_mask bch2_online_devs(struct bch_fs *c) { -- cgit v1.2.3-59-g8ed1b From a7f1c26f5907dde57f58c329ef0c532643092e71 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:53:03 -0400 Subject: bcachefs: bch2_trigger_alloc() -> bch2_dev_tryget() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index e61f00efe619..41f2e19aa8fe 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -739,12 +739,10 @@ int bch2_trigger_alloc(struct btree_trans *trans, struct bch_fs *c = trans->c; int ret = 0; - if (bch2_trans_inconsistent_on(!bch2_dev_bucket_exists(c, new.k->p), trans, - "alloc key for invalid device or bucket")) + struct bch_dev *ca = bch2_dev_bucket_tryget(c, new.k->p); + if (!ca) return -EIO; - struct bch_dev *ca = bch2_dev_bkey_exists(c, new.k->p.inode); - struct bch_alloc_v4 old_a_convert; const struct bch_alloc_v4 *old_a = bch2_alloc_to_v4(old, &old_a_convert); @@ -773,7 +771,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, ret = bch2_bucket_do_index(trans, ca, old, old_a, false) ?: bch2_bucket_do_index(trans, ca, new.s_c, new_a, true); if (ret) - return ret; + goto err; } if (new_a->data_type == BCH_DATA_cached && @@ -787,7 +785,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, bucket_to_u64(new.k->p), old_lru, new_lru); if (ret) - return ret; + goto err; } new_a->fragmentation_lru = alloc_lru_idx_fragmentation(*new_a, ca); @@ -797,13 +795,13 @@ int bch2_trigger_alloc(struct btree_trans *trans, bucket_to_u64(new.k->p), old_a->fragmentation_lru, new_a->fragmentation_lru); if (ret) - return ret; + goto err; } if (old_a->gen != new_a->gen) { ret = bch2_bucket_gen_update(trans, new.k->p, new_a->gen); if (ret) - return ret; + goto err; } /* @@ -816,7 +814,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, ret = bch2_update_cached_sectors_list(trans, new.k->p.inode, -((s64) old_a->cached_sectors)); if (ret) - return ret; + goto err; } } @@ -853,7 +851,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, if (ret) { bch2_fs_fatal_error(c, "setting bucket_needs_journal_commit: %s", bch2_err_str(ret)); - return ret; + goto err; } } @@ -907,8 +905,9 @@ int bch2_trigger_alloc(struct btree_trans *trans, bucket_unlock(g); percpu_up_read(&c->mark_lock); } - - return 0; +err: + bch2_dev_put(ca); + return ret; } /* -- cgit v1.2.3-59-g8ed1b From 07d7c4da7bd1faefd338f4073846632fbac54880 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 16:20:49 -0400 Subject: bcachefs: bch2_bucket_ref_update() now takes bch_dev Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 47 ++++++++++++++++++++++++++--------------------- fs/bcachefs/buckets.h | 4 ++-- fs/bcachefs/ec.c | 19 ++++++++++++++----- 3 files changed, 42 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 525a7548fab7..68f2d5952abd 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -700,15 +700,14 @@ fsck_err: return ret; } -int bch2_bucket_ref_update(struct btree_trans *trans, - struct bkey_s_c k, - const struct bch_extent_ptr *ptr, - s64 sectors, enum bch_data_type ptr_data_type, - u8 b_gen, u8 bucket_data_type, - u32 *bucket_sectors) +int bch2_bucket_ref_update(struct btree_trans *trans, struct bch_dev *ca, + struct bkey_s_c k, + const struct bch_extent_ptr *ptr, + s64 sectors, enum bch_data_type ptr_data_type, + u8 b_gen, u8 bucket_data_type, + u32 *bucket_sectors) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); size_t bucket_nr = PTR_BUCKET_NR(ca, ptr); struct printbuf buf = PRINTBUF; bool inserting = sectors > 0; @@ -939,7 +938,7 @@ need_mark: /* KEY_TYPE_extent: */ -static int __mark_pointer(struct btree_trans *trans, +static int __mark_pointer(struct btree_trans *trans, struct bch_dev *ca, struct bkey_s_c k, const struct bch_extent_ptr *ptr, s64 sectors, enum bch_data_type ptr_data_type, @@ -948,7 +947,7 @@ static int __mark_pointer(struct btree_trans *trans, u32 *dst_sectors = !ptr->cached ? &a->dirty_sectors : &a->cached_sectors; - int ret = bch2_bucket_ref_update(trans, k, ptr, sectors, ptr_data_type, + int ret = bch2_bucket_ref_update(trans, ca, k, ptr, sectors, ptr_data_type, a->gen, a->data_type, dst_sectors); if (ret) @@ -966,45 +965,51 @@ static int bch2_trigger_pointer(struct btree_trans *trans, enum btree_iter_update_trigger_flags flags) { bool insert = !(flags & BTREE_TRIGGER_overwrite); + int ret = 0; + + struct bch_fs *c = trans->c; + struct bch_dev *ca = bch2_dev_tryget(c, p.ptr.dev); + if (unlikely(!ca)) { + if (insert) + ret = -EIO; + goto err; + } + struct bpos bucket; struct bch_backpointer bp; - bch2_extent_ptr_to_bp(trans->c, btree_id, level, k, p, entry, &bucket, &bp); *sectors = insert ? bp.bucket_len : -((s64) bp.bucket_len); if (flags & BTREE_TRIGGER_transactional) { struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, bucket); - int ret = PTR_ERR_OR_ZERO(a) ?: - __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, &a->v); + ret = PTR_ERR_OR_ZERO(a) ?: + __mark_pointer(trans, ca, k, &p.ptr, *sectors, bp.data_type, &a->v); if (ret) - return ret; + goto err; if (!p.ptr.cached) { ret = bch2_bucket_backpointer_mod(trans, bucket, bp, k, insert); if (ret) - return ret; + goto err; } } if (flags & BTREE_TRIGGER_gc) { - struct bch_fs *c = trans->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); - percpu_down_read(&c->mark_lock); struct bucket *g = gc_bucket(ca, bucket.offset); bucket_lock(g); struct bch_alloc_v4 old = bucket_m_to_alloc(*g), new = old; - int ret = __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type, &new); + ret = __mark_pointer(trans, ca, k, &p.ptr, *sectors, bp.data_type, &new); if (!ret) { alloc_to_bucket(g, new); bch2_dev_usage_update(c, ca, &old, &new, 0, true); } bucket_unlock(g); percpu_up_read(&c->mark_lock); - return ret; } - - return 0; +err: + bch2_dev_put(ca); + return ret; } static int bch2_trigger_stripe_ptr(struct btree_trans *trans, diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index 95cf3c290734..46bf66d6b254 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -330,8 +330,8 @@ int bch2_replicas_deltas_realloc(struct btree_trans *, unsigned); void bch2_fs_usage_initialize(struct bch_fs *); -int bch2_bucket_ref_update(struct btree_trans *, struct bkey_s_c, - const struct bch_extent_ptr *, +int bch2_bucket_ref_update(struct btree_trans *, struct bch_dev *, + struct bkey_s_c, const struct bch_extent_ptr *, s64, enum bch_data_type, u8, u8, u32 *); int bch2_check_fix_ptrs(struct btree_trans *, diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 330c3da70b96..ea10f838d9cc 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -167,9 +167,9 @@ static int __mark_stripe_bucket(struct btree_trans *trans, struct bkey_s_c_stripe s, unsigned ptr_idx, bool deleting, struct bpos bucket, - struct bch_alloc_v4 *a) + struct bch_alloc_v4 *a, + enum btree_iter_update_trigger_flags flags) { - struct bch_fs *c = trans->c; const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; unsigned nr_data = s.v->nr_blocks - s.v->nr_redundant; bool parity = ptr_idx >= nr_data; @@ -178,6 +178,14 @@ static int __mark_stripe_bucket(struct btree_trans *trans, struct printbuf buf = PRINTBUF; int ret = 0; + struct bch_fs *c = trans->c; + struct bch_dev *ca = bch2_dev_tryget(c, ptr->dev); + if (unlikely(!ca)) { + if (!(flags & BTREE_TRIGGER_overwrite)) + ret = -EIO; + goto err; + } + if (deleting) sectors = -sectors; @@ -239,7 +247,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, } if (sectors) { - ret = bch2_bucket_ref_update(trans, s.s_c, ptr, sectors, data_type, + ret = bch2_bucket_ref_update(trans, ca, s.s_c, ptr, sectors, data_type, a->gen, a->data_type, &a->dirty_sectors); if (ret) goto err; @@ -255,6 +263,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, alloc_data_type_set(a, data_type); err: + bch2_dev_put(ca); printbuf_exit(&buf); return ret; } @@ -272,7 +281,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, bucket); return PTR_ERR_OR_ZERO(a) ?: - __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &a->v); + __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &a->v, flags); } if (flags & BTREE_TRIGGER_gc) { @@ -282,7 +291,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, struct bucket *g = gc_bucket(ca, bucket.offset); bucket_lock(g); struct bch_alloc_v4 old = bucket_m_to_alloc(*g), new = old; - int ret = __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &new); + int ret = __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &new, flags); if (!ret) { alloc_to_bucket(g, new); bch2_dev_usage_update(c, ca, &old, &new, 0, true); -- cgit v1.2.3-59-g8ed1b From cb4d340a10295fcea7e4363f1b95c1a8c6c9bed4 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:46:45 -0400 Subject: bcachefs: bch2_evacuate_bucket() -> bch2_dev_tryget() Signed-off-by: Kent Overstreet --- fs/bcachefs/move.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index 12bc577354ff..c441ce7c92ec 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -690,6 +690,10 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, struct bpos bp_pos = POS_MIN; int ret = 0; + struct bch_dev *ca = bch2_dev_tryget(c, bucket.inode); + if (!ca) + return 0; + trace_bucket_evacuate(c, &bucket); bch2_bkey_buf_init(&sk); @@ -711,7 +715,7 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, a = bch2_alloc_to_v4(k, &a_convert); dirty_sectors = bch2_bucket_sectors_dirty(*a); - bucket_size = bch2_dev_bkey_exists(c, bucket.inode)->mi.bucket_size; + bucket_size = ca->mi.bucket_size; fragmentation = a->fragmentation_lru; ret = bch2_btree_write_buffer_tryflush(trans); @@ -823,6 +827,7 @@ next: trace_evacuate_bucket(c, &bucket, dirty_sectors, bucket_size, fragmentation, ret); err: + bch2_dev_put(ca); bch2_bkey_buf_exit(&sk, c); return ret; } -- cgit v1.2.3-59-g8ed1b From fa6cce09f070990dfe384ef4ddfefdea73970abe Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:37:25 -0400 Subject: bcachefs: bch2_dev_iterate() New helper for getting refs to devices as we iterate. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 18 +++++++++++------- fs/bcachefs/sb-members.h | 8 ++++++++ 2 files changed, 19 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 41f2e19aa8fe..69a77d268edf 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -567,6 +567,7 @@ iter_err: int bch2_alloc_read(struct bch_fs *c) { struct btree_trans *trans = bch2_trans_get(c); + struct bch_dev *ca = NULL; int ret; down_read(&c->gc_lock); @@ -580,16 +581,17 @@ int bch2_alloc_read(struct bch_fs *c) if (k.k->type != KEY_TYPE_bucket_gens) continue; - const struct bch_bucket_gens *g = bkey_s_c_to_bucket_gens(k).v; - + ca = bch2_dev_iterate(c, ca, k.k->p.inode); /* * Not a fsck error because this is checked/repaired by * bch2_check_alloc_key() which runs later: */ - if (!bch2_dev_exists(c, k.k->p.inode)) + if (!ca) { + bch2_btree_iter_set_pos(&iter, POS(k.k->p.inode + 1, 0)); continue; + } - struct bch_dev *ca = bch2_dev_bkey_exists(c, k.k->p.inode); + const struct bch_bucket_gens *g = bkey_s_c_to_bucket_gens(k).v; for (u64 b = max_t(u64, ca->mi.first_bucket, start); b < min_t(u64, ca->mi.nbuckets, end); @@ -600,14 +602,15 @@ int bch2_alloc_read(struct bch_fs *c) } else { ret = for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, BTREE_ITER_prefetch, k, ({ + ca = bch2_dev_iterate(c, ca, k.k->p.inode); /* * Not a fsck error because this is checked/repaired by * bch2_check_alloc_key() which runs later: */ - if (!bch2_dev_bucket_exists(c, k.k->p)) + if (!ca) { + bch2_btree_iter_set_pos(&iter, POS(k.k->p.inode + 1, 0)); continue; - - struct bch_dev *ca = bch2_dev_bkey_exists(c, k.k->p.inode); + } struct bch_alloc_v4 a; *bucket_gen(ca, k.k->p.offset) = bch2_alloc_to_v4(k, &a)->gen; @@ -615,6 +618,7 @@ int bch2_alloc_read(struct bch_fs *c) })); } + bch2_dev_put(ca); bch2_trans_put(trans); up_read(&c->gc_lock); diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index 503b56051d7e..d6d391dc21ca 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -262,6 +262,14 @@ static inline struct bch_dev *bch2_dev_bucket_tryget(struct bch_fs *c, struct bp return ca; } +static inline struct bch_dev *bch2_dev_iterate(struct bch_fs *c, struct bch_dev *ca, unsigned dev_idx) +{ + if (ca && ca->dev_idx == dev_idx) + return ca; + bch2_dev_put(ca); + return bch2_dev_tryget(c, dev_idx); +} + /* XXX kill, move to struct bch_fs */ static inline struct bch_devs_mask bch2_online_devs(struct bch_fs *c) { -- cgit v1.2.3-59-g8ed1b From 1f2f92ec3f2e5802c2870d0d34ccfa66ae9216ad Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 19:34:28 -0400 Subject: bcachefs: PTR_BUCKET_POS() now takes bch_dev Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 7 +++++-- fs/bcachefs/backpointers.h | 4 ++-- fs/bcachefs/buckets.c | 2 +- fs/bcachefs/buckets.h | 10 +++------- fs/bcachefs/data_update.c | 22 +++++++++++---------- fs/bcachefs/ec.c | 48 +++++++++++++++++++++++++--------------------- fs/bcachefs/io_read.c | 2 +- fs/bcachefs/io_write.c | 11 +++++++---- 8 files changed, 57 insertions(+), 49 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 36e5e63ec3d5..282aee70a51f 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -30,7 +30,9 @@ static bool extent_matches_bp(struct bch_fs *c, if (p.ptr.cached) continue; - bch2_extent_ptr_to_bp(c, btree_id, level, k, p, entry, &bucket2, &bp2); + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + + bch2_extent_ptr_to_bp(c, ca, btree_id, level, k, p, entry, &bucket2, &bp2); if (bpos_eq(bucket, bucket2) && !memcmp(&bp, &bp2, sizeof(bp))) return true; @@ -666,7 +668,8 @@ static int check_extent_to_backpointers(struct btree_trans *trans, if (p.ptr.cached) continue; - bch2_extent_ptr_to_bp(c, btree, level, k, p, entry, &bucket_pos, &bp); + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + bch2_extent_ptr_to_bp(c, ca, btree, level, k, p, entry, &bucket_pos, &bp); ret = check_bp_exists(trans, s, bucket_pos, bp, k); if (ret) diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index e7f1eddbf4f4..1d270c40fce9 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -120,7 +120,7 @@ static inline enum bch_data_type bch2_bkey_ptr_data_type(struct bkey_s_c k, } } -static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, +static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, struct bch_dev *ca, enum btree_id btree_id, unsigned level, struct bkey_s_c k, struct extent_ptr_decoded p, const union bch_extent_entry *entry, @@ -130,7 +130,7 @@ static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, s64 sectors = level ? btree_sectors(c) : k.k->size; u32 bucket_offset; - *bucket_pos = PTR_BUCKET_POS_OFFSET(c, &p.ptr, &bucket_offset); + *bucket_pos = PTR_BUCKET_POS_OFFSET(ca, &p.ptr, &bucket_offset); *bp = (struct bch_backpointer) { .btree_id = btree_id, .level = level, diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 68f2d5952abd..4774f9c7f060 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -977,7 +977,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, struct bpos bucket; struct bch_backpointer bp; - bch2_extent_ptr_to_bp(trans->c, btree_id, level, k, p, entry, &bucket, &bp); + bch2_extent_ptr_to_bp(trans->c, ca, btree_id, level, k, p, entry, &bucket, &bp); *sectors = insert ? bp.bucket_len : -((s64) bp.bucket_len); if (flags & BTREE_TRIGGER_transactional) { diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index 46bf66d6b254..0a54faf7c50a 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -120,20 +120,16 @@ static inline size_t PTR_BUCKET_NR(const struct bch_dev *ca, return sector_to_bucket(ca, ptr->offset); } -static inline struct bpos PTR_BUCKET_POS(const struct bch_fs *c, - const struct bch_extent_ptr *ptr) +static inline struct bpos PTR_BUCKET_POS(const struct bch_dev *ca, + const struct bch_extent_ptr *ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - return POS(ptr->dev, PTR_BUCKET_NR(ca, ptr)); } -static inline struct bpos PTR_BUCKET_POS_OFFSET(const struct bch_fs *c, +static inline struct bpos PTR_BUCKET_POS_OFFSET(const struct bch_dev *ca, const struct bch_extent_ptr *ptr, u32 *bucket_offset) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - return POS(ptr->dev, sector_to_bucket_and_offset(ca, ptr->offset, bucket_offset)); } diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index c252f0619b73..1cf92bea7f9f 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -357,10 +357,11 @@ void bch2_data_update_exit(struct data_update *update) bch2_bkey_ptrs_c(bkey_i_to_s_c(update->k.k)); bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); if (c->opts.nocow_enabled) bch2_bucket_nocow_unlock(&c->nocow_locks, - PTR_BUCKET_POS(c, ptr), 0); - bch2_dev_put(bch2_dev_bkey_exists(c, ptr->dev)); + PTR_BUCKET_POS(ca, ptr), 0); + bch2_dev_put(ca); } bch2_bkey_buf_exit(&update->k, c); @@ -547,6 +548,8 @@ int bch2_data_update_init(struct btree_trans *trans, i = 0; bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bpos bucket = PTR_BUCKET_POS(ca, &p.ptr); bool locked; if (((1U << i) & m->data_opts.rewrite_ptrs)) { @@ -580,15 +583,13 @@ int bch2_data_update_init(struct btree_trans *trans, if (ctxt) { move_ctxt_wait_event(ctxt, (locked = bch2_bucket_nocow_trylock(&c->nocow_locks, - PTR_BUCKET_POS(c, &p.ptr), 0)) || + bucket, 0)) || list_empty(&ctxt->ios)); if (!locked) - bch2_bucket_nocow_lock(&c->nocow_locks, - PTR_BUCKET_POS(c, &p.ptr), 0); + bch2_bucket_nocow_lock(&c->nocow_locks, bucket, 0); } else { - if (!bch2_bucket_nocow_trylock(&c->nocow_locks, - PTR_BUCKET_POS(c, &p.ptr), 0)) { + if (!bch2_bucket_nocow_trylock(&c->nocow_locks, bucket, 0)) { ret = -BCH_ERR_nocow_lock_blocked; goto err; } @@ -650,10 +651,11 @@ int bch2_data_update_init(struct btree_trans *trans, err: i = 0; bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bpos bucket = PTR_BUCKET_POS(ca, &p.ptr); if ((1U << i) & ptrs_locked) - bch2_bucket_nocow_unlock(&c->nocow_locks, - PTR_BUCKET_POS(c, &p.ptr), 0); - bch2_dev_put(bch2_dev_bkey_exists(c, p.ptr.dev)); + bch2_bucket_nocow_unlock(&c->nocow_locks, bucket, 0); + bch2_dev_put(ca); i++; } diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index ea10f838d9cc..f09df7f9036a 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -164,6 +164,7 @@ void bch2_stripe_to_text(struct printbuf *out, struct bch_fs *c, /* Triggers: */ static int __mark_stripe_bucket(struct btree_trans *trans, + struct bch_dev *ca, struct bkey_s_c_stripe s, unsigned ptr_idx, bool deleting, struct bpos bucket, @@ -179,13 +180,6 @@ static int __mark_stripe_bucket(struct btree_trans *trans, int ret = 0; struct bch_fs *c = trans->c; - struct bch_dev *ca = bch2_dev_tryget(c, ptr->dev); - if (unlikely(!ca)) { - if (!(flags & BTREE_TRIGGER_overwrite)) - ret = -EIO; - goto err; - } - if (deleting) sectors = -sectors; @@ -263,7 +257,6 @@ static int __mark_stripe_bucket(struct btree_trans *trans, alloc_data_type_set(a, data_type); err: - bch2_dev_put(ca); printbuf_exit(&buf); return ret; } @@ -275,34 +268,40 @@ static int mark_stripe_bucket(struct btree_trans *trans, { struct bch_fs *c = trans->c; const struct bch_extent_ptr *ptr = s.v->ptrs + ptr_idx; - struct bpos bucket = PTR_BUCKET_POS(c, ptr); + int ret = 0; + + struct bch_dev *ca = bch2_dev_tryget(c, ptr->dev); + if (unlikely(!ca)) { + if (!(flags & BTREE_TRIGGER_overwrite)) + ret = -EIO; + goto err; + } + + struct bpos bucket = PTR_BUCKET_POS(ca, ptr); if (flags & BTREE_TRIGGER_transactional) { struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, bucket); - return PTR_ERR_OR_ZERO(a) ?: - __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &a->v, flags); + ret = PTR_ERR_OR_ZERO(a) ?: + __mark_stripe_bucket(trans, ca, s, ptr_idx, deleting, bucket, &a->v, flags); } if (flags & BTREE_TRIGGER_gc) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - percpu_down_read(&c->mark_lock); struct bucket *g = gc_bucket(ca, bucket.offset); bucket_lock(g); struct bch_alloc_v4 old = bucket_m_to_alloc(*g), new = old; - int ret = __mark_stripe_bucket(trans, s, ptr_idx, deleting, bucket, &new, flags); + ret = __mark_stripe_bucket(trans, ca, s, ptr_idx, deleting, bucket, &new, flags); if (!ret) { alloc_to_bucket(g, new); bch2_dev_usage_update(c, ca, &old, &new, 0, true); } bucket_unlock(g); percpu_up_read(&c->mark_lock); - return ret; } - - BUG(); - return 0; +err: + bch2_dev_put(ca); + return ret; } static int mark_stripe_buckets(struct btree_trans *trans, @@ -1298,17 +1297,21 @@ static int ec_stripe_update_bucket(struct btree_trans *trans, struct ec_stripe_b { struct bch_fs *c = trans->c; struct bch_stripe *v = &bkey_i_to_stripe(&s->key)->v; - struct bch_extent_ptr bucket = v->ptrs[block]; - struct bpos bucket_pos = PTR_BUCKET_POS(c, &bucket); + struct bch_extent_ptr ptr = v->ptrs[block]; struct bpos bp_pos = POS_MIN; int ret = 0; + struct bch_dev *ca = bch2_dev_tryget(c, ptr.dev); + if (!ca) + return -EIO; + + struct bpos bucket_pos = PTR_BUCKET_POS(ca, &ptr); + while (1) { ret = commit_do(trans, NULL, NULL, BCH_TRANS_COMMIT_no_check_rw| BCH_TRANS_COMMIT_no_enospc, - ec_stripe_update_extent(trans, bucket_pos, bucket.gen, - s, &bp_pos)); + ec_stripe_update_extent(trans, bucket_pos, ptr.gen, s, &bp_pos)); if (ret) break; if (bkey_eq(bp_pos, POS_MAX)) @@ -1317,6 +1320,7 @@ static int ec_stripe_update_bucket(struct btree_trans *trans, struct ec_stripe_b bp_pos = bpos_nosnap_successor(bp_pos); } + bch2_dev_put(ca); return ret; } diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index af9bd7c45f21..1091b066639e 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -768,7 +768,7 @@ static noinline void read_from_stale_dirty_pointer(struct btree_trans *trans, int ret; bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, - PTR_BUCKET_POS(c, &ptr), + PTR_BUCKET_POS(ca, &ptr), BTREE_ITER_cached); prt_printf(&buf, "Attempting to read from stale dirty pointer:\n"); diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 55e24c83fb19..40c8f8844224 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1117,10 +1117,12 @@ static inline void bch2_nocow_write_unlock(struct bch_write_op *op) for_each_keylist_key(&op->insert_keys, k) { struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(bkey_i_to_s_c(k)); - bkey_for_each_ptr(ptrs, ptr) + bkey_for_each_ptr(ptrs, ptr) { + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); bch2_bucket_nocow_unlock(&c->nocow_locks, - PTR_BUCKET_POS(c, ptr), + PTR_BUCKET_POS(ca, ptr), BUCKET_NOCOW_LOCK_UPDATE); + } } } @@ -1270,12 +1272,13 @@ retry: /* Get iorefs before dropping btree locks: */ struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); bkey_for_each_ptr(ptrs, ptr) { - struct bpos b = PTR_BUCKET_POS(c, ptr); + struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bpos b = PTR_BUCKET_POS(ca, ptr); struct nocow_lock_bucket *l = bucket_nocow_lock(&c->nocow_locks, bucket_to_u64(b)); prefetch(l); - if (unlikely(!bch2_dev_get_ioref(bch2_dev_bkey_exists(c, ptr->dev), WRITE))) + if (unlikely(!bch2_dev_get_ioref(ca, WRITE))) goto err_get_ioref; /* XXX allocating memory with btree locks held - rare */ -- cgit v1.2.3-59-g8ed1b From 633cf069445d95384d88e9ff40f395c001f8b2b7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 16:50:28 -0400 Subject: bcachefs: Kill bch2_dev_bkey_exists() in backpointer code Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 2 +- fs/bcachefs/backpointers.c | 106 +++++++++++++++++++++++++++-------------- fs/bcachefs/backpointers.h | 37 +++++++++----- fs/bcachefs/buckets.c | 2 +- fs/bcachefs/ec.c | 5 +- fs/bcachefs/move.c | 2 +- 6 files changed, 102 insertions(+), 52 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index f54d90868314..cc182b263066 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -342,7 +342,7 @@ static struct open_bucket *try_alloc_bucket(struct btree_trans *trans, struct bc struct bch_backpointer bp; struct bpos bp_pos = POS_MIN; - ret = bch2_get_next_backpointer(trans, POS(ca->dev_idx, b), -1, + ret = bch2_get_next_backpointer(trans, ca, POS(ca->dev_idx, b), -1, &bp_pos, &bp, BTREE_ITER_nopreserve); if (ret) { diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 282aee70a51f..3f5333b4c3de 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -23,6 +23,7 @@ static bool extent_matches_bp(struct bch_fs *c, const union bch_extent_entry *entry; struct extent_ptr_decoded p; + rcu_read_lock(); bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { struct bpos bucket2; struct bch_backpointer bp2; @@ -30,13 +31,18 @@ static bool extent_matches_bp(struct bch_fs *c, if (p.ptr.cached) continue; - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); + if (!ca) + continue; bch2_extent_ptr_to_bp(c, ca, btree_id, level, k, p, entry, &bucket2, &bp2); if (bpos_eq(bucket, bucket2) && - !memcmp(&bp, &bp2, sizeof(bp))) + !memcmp(&bp, &bp2, sizeof(bp))) { + rcu_read_unlock(); return true; + } } + rcu_read_unlock(); return false; } @@ -47,16 +53,21 @@ int bch2_backpointer_invalid(struct bch_fs *c, struct bkey_s_c k, { struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); - /* these will be caught by fsck */ - if (!bch2_dev_exists(c, bp.k->p.inode)) + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, bp.k->p.inode); + if (!ca) { + /* these will be caught by fsck */ + rcu_read_unlock(); return 0; + } - struct bch_dev *ca = bch2_dev_bkey_exists(c, bp.k->p.inode); - struct bpos bucket = bp_pos_to_bucket(c, bp.k->p); + struct bpos bucket = bp_pos_to_bucket(ca, bp.k->p); + struct bpos bp_pos = bucket_pos_to_bp_noerror(ca, bucket, bp.v->bucket_offset); + rcu_read_unlock(); int ret = 0; bkey_fsck_err_on((bp.v->bucket_offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT) >= ca->mi.bucket_size || - !bpos_eq(bp.k->p, bucket_pos_to_bp_noerror(ca, bucket, bp.v->bucket_offset)), + !bpos_eq(bp.k->p, bp_pos), c, err, backpointer_bucket_offset_wrong, "backpointer bucket_offset wrong"); @@ -77,10 +88,16 @@ void bch2_backpointer_to_text(struct printbuf *out, const struct bch_backpointer void bch2_backpointer_k_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) { - if (bch2_dev_exists(c, k.k->p.inode)) { + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, k.k->p.inode); + if (ca) { + struct bpos bucket = bp_pos_to_bucket(ca, k.k->p); + rcu_read_unlock(); prt_str(out, "bucket="); - bch2_bpos_to_text(out, bp_pos_to_bucket(c, k.k->p)); + bch2_bpos_to_text(out, bucket); prt_str(out, " "); + } else { + rcu_read_unlock(); } bch2_backpointer_to_text(out, bkey_s_c_to_backpointer(k).v); @@ -146,6 +163,7 @@ static noinline int backpointer_mod_err(struct btree_trans *trans, } int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *trans, + struct bch_dev *ca, struct bpos bucket, struct bch_backpointer bp, struct bkey_s_c orig_k, @@ -162,7 +180,7 @@ int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *trans, return ret; bkey_backpointer_init(&bp_k->k_i); - bp_k->k.p = bucket_pos_to_bp(trans->c, bucket, bp.bucket_offset); + bp_k->k.p = bucket_pos_to_bp(ca, bucket, bp.bucket_offset); bp_k->v = bp; if (!insert) { @@ -198,13 +216,13 @@ err: * Find the next backpointer >= *bp_offset: */ int bch2_get_next_backpointer(struct btree_trans *trans, + struct bch_dev *ca, struct bpos bucket, int gen, struct bpos *bp_pos, struct bch_backpointer *bp, unsigned iter_flags) { - struct bch_fs *c = trans->c; - struct bpos bp_end_pos = bucket_pos_to_bp(c, bpos_nosnap_successor(bucket), 0); + struct bpos bp_end_pos = bucket_pos_to_bp(ca, bpos_nosnap_successor(bucket), 0); struct btree_iter alloc_iter = { NULL }, bp_iter = { NULL }; struct bkey_s_c k; int ret = 0; @@ -224,7 +242,7 @@ int bch2_get_next_backpointer(struct btree_trans *trans, goto done; } - *bp_pos = bpos_max(*bp_pos, bucket_pos_to_bp(c, bucket, 0)); + *bp_pos = bpos_max(*bp_pos, bucket_pos_to_bp(ca, bucket, 0)); for_each_btree_key_norestart(trans, bp_iter, BTREE_ID_backpointers, *bp_pos, iter_flags, k, ret) { @@ -250,7 +268,6 @@ static void backpointer_not_found(struct btree_trans *trans, { struct bch_fs *c = trans->c; struct printbuf buf = PRINTBUF; - struct bpos bucket = bp_pos_to_bucket(c, bp_pos); /* * If we're using the btree write buffer, the backpointer we were @@ -260,6 +277,10 @@ static void backpointer_not_found(struct btree_trans *trans, if (likely(!bch2_backpointers_no_use_write_buffer)) return; + struct bpos bucket; + if (!bp_pos_to_bucket_nodev(c, bp_pos, &bucket)) + return; + prt_printf(&buf, "backpointer doesn't match %s it points to:\n ", bp.level ? "btree node" : "extent"); prt_printf(&buf, "bucket: "); @@ -289,15 +310,17 @@ struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *trans, { if (likely(!bp.level)) { struct bch_fs *c = trans->c; - struct bpos bucket = bp_pos_to_bucket(c, bp_pos); - struct bkey_s_c k; + + struct bpos bucket; + if (!bp_pos_to_bucket_nodev(c, bp_pos, &bucket)) + return bkey_s_c_err(-EIO); bch2_trans_node_iter_init(trans, iter, bp.btree_id, bp.pos, 0, 0, iter_flags); - k = bch2_btree_iter_peek_slot(iter); + struct bkey_s_c k = bch2_btree_iter_peek_slot(iter); if (bkey_err(k)) { bch2_trans_iter_exit(trans, iter); return k; @@ -326,18 +349,20 @@ struct btree *bch2_backpointer_get_node(struct btree_trans *trans, struct bch_backpointer bp) { struct bch_fs *c = trans->c; - struct bpos bucket = bp_pos_to_bucket(c, bp_pos); - struct btree *b; BUG_ON(!bp.level); + struct bpos bucket; + if (!bp_pos_to_bucket_nodev(c, bp_pos, &bucket)) + return ERR_PTR(-EIO); + bch2_trans_node_iter_init(trans, iter, bp.btree_id, bp.pos, 0, bp.level - 1, 0); - b = bch2_btree_iter_peek_node(iter); + struct btree *b = bch2_btree_iter_peek_node(iter); if (IS_ERR_OR_NULL(b)) goto err; @@ -368,16 +393,16 @@ static int bch2_check_btree_backpointer(struct btree_trans *trans, struct btree_ struct printbuf buf = PRINTBUF; int ret = 0; - if (fsck_err_on(!bch2_dev_exists(c, k.k->p.inode), c, - backpointer_to_missing_device, - "backpointer for missing device:\n%s", - (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { - ret = bch2_btree_delete_at(trans, bp_iter, 0); + struct bpos bucket; + if (!bp_pos_to_bucket_nodev_noerror(c, k.k->p, &bucket)) { + if (fsck_err(c, backpointer_to_missing_device, + "backpointer for missing device:\n%s", + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + ret = bch2_btree_delete_at(trans, bp_iter, 0); goto out; } - alloc_k = bch2_bkey_get_iter(trans, &alloc_iter, BTREE_ID_alloc, - bp_pos_to_bucket(c, k.k->p), 0); + alloc_k = bch2_bkey_get_iter(trans, &alloc_iter, BTREE_ID_alloc, bucket, 0); ret = bkey_err(alloc_k); if (ret) goto out; @@ -512,25 +537,27 @@ static int check_bp_exists(struct btree_trans *trans, struct printbuf buf = PRINTBUF; struct bkey_s_c bp_k; struct bkey_buf tmp; - int ret; + int ret = 0; bch2_bkey_buf_init(&tmp); - if (!bch2_dev_bucket_exists(c, bucket)) { + struct bch_dev *ca = bch2_dev_bucket_tryget(c, bucket); + if (!ca) { prt_str(&buf, "extent for nonexistent device:bucket "); bch2_bpos_to_text(&buf, bucket); prt_str(&buf, "\n "); bch2_bkey_val_to_text(&buf, c, orig_k); bch_err(c, "%s", buf.buf); - return -BCH_ERR_fsck_repair_unimplemented; + ret = -BCH_ERR_fsck_repair_unimplemented; + goto err; } if (bpos_lt(bucket, s->bucket_start) || bpos_gt(bucket, s->bucket_end)) - return 0; + goto out; bp_k = bch2_bkey_get_iter(trans, &bp_iter, BTREE_ID_backpointers, - bucket_pos_to_bp(c, bucket, bp.bucket_offset), + bucket_pos_to_bp(ca, bucket, bp.bucket_offset), 0); ret = bkey_err(bp_k); if (ret) @@ -563,6 +590,7 @@ fsck_err: bch2_trans_iter_exit(trans, &other_extent_iter); bch2_trans_iter_exit(trans, &bp_iter); bch2_bkey_buf_exit(&tmp, c); + bch2_dev_put(ca); printbuf_exit(&buf); return ret; check_existing_bp: @@ -638,13 +666,13 @@ missing: struct bkey_i_backpointer n_bp_k; bkey_backpointer_init(&n_bp_k.k_i); - n_bp_k.k.p = bucket_pos_to_bp(trans->c, bucket, bp.bucket_offset); + n_bp_k.k.p = bucket_pos_to_bp(ca, bucket, bp.bucket_offset); n_bp_k.v = bp; prt_printf(&buf, "\n want: "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&n_bp_k.k_i)); if (fsck_err(c, ptr_to_missing_backpointer, "%s", buf.buf)) - ret = bch2_bucket_backpointer_mod(trans, bucket, bp, orig_k, true); + ret = bch2_bucket_backpointer_mod(trans, ca, bucket, bp, orig_k, true); goto out; } @@ -668,8 +696,14 @@ static int check_extent_to_backpointers(struct btree_trans *trans, if (p.ptr.cached) continue; - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); - bch2_extent_ptr_to_bp(c, ca, btree, level, k, p, entry, &bucket_pos, &bp); + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); + if (ca) + bch2_extent_ptr_to_bp(c, ca, btree, level, k, p, entry, &bucket_pos, &bp); + rcu_read_unlock(); + + if (!ca) + continue; ret = check_bp_exists(trans, s, bucket_pos, bp, k); if (ret) diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index 1d270c40fce9..929f4a068a0a 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -6,6 +6,7 @@ #include "btree_iter.h" #include "btree_update.h" #include "buckets.h" +#include "error.h" #include "super.h" static inline u64 swab40(u64 x) @@ -36,15 +37,29 @@ void bch2_backpointer_swab(struct bkey_s); * Convert from pos in backpointer btree to pos of corresponding bucket in alloc * btree: */ -static inline struct bpos bp_pos_to_bucket(const struct bch_fs *c, - struct bpos bp_pos) +static inline struct bpos bp_pos_to_bucket(const struct bch_dev *ca, struct bpos bp_pos) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, bp_pos.inode); u64 bucket_sector = bp_pos.offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT; return POS(bp_pos.inode, sector_to_bucket(ca, bucket_sector)); } +static inline bool bp_pos_to_bucket_nodev_noerror(struct bch_fs *c, struct bpos bp_pos, struct bpos *bucket) +{ + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, bp_pos.inode); + if (ca) + *bucket = bp_pos_to_bucket(ca, bp_pos); + rcu_read_unlock(); + return ca != NULL; +} + +static inline bool bp_pos_to_bucket_nodev(struct bch_fs *c, struct bpos bp_pos, struct bpos *bucket) +{ + return !bch2_fs_inconsistent_on(!bp_pos_to_bucket_nodev_noerror(c, bp_pos, bucket), + c, "backpointer for missing device %llu", bp_pos.inode); +} + static inline struct bpos bucket_pos_to_bp_noerror(const struct bch_dev *ca, struct bpos bucket, u64 bucket_offset) @@ -57,32 +72,32 @@ static inline struct bpos bucket_pos_to_bp_noerror(const struct bch_dev *ca, /* * Convert from pos in alloc btree + bucket offset to pos in backpointer btree: */ -static inline struct bpos bucket_pos_to_bp(const struct bch_fs *c, +static inline struct bpos bucket_pos_to_bp(const struct bch_dev *ca, struct bpos bucket, u64 bucket_offset) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, bucket.inode); struct bpos ret = bucket_pos_to_bp_noerror(ca, bucket, bucket_offset); - EBUG_ON(!bkey_eq(bucket, bp_pos_to_bucket(c, ret))); + EBUG_ON(!bkey_eq(bucket, bp_pos_to_bucket(ca, ret))); return ret; } -int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *, struct bpos bucket, - struct bch_backpointer, struct bkey_s_c, bool); +int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *, struct bch_dev *, + struct bpos bucket, struct bch_backpointer, struct bkey_s_c, bool); static inline int bch2_bucket_backpointer_mod(struct btree_trans *trans, + struct bch_dev *ca, struct bpos bucket, struct bch_backpointer bp, struct bkey_s_c orig_k, bool insert) { if (unlikely(bch2_backpointers_no_use_write_buffer)) - return bch2_bucket_backpointer_mod_nowritebuffer(trans, bucket, bp, orig_k, insert); + return bch2_bucket_backpointer_mod_nowritebuffer(trans, ca, bucket, bp, orig_k, insert); struct bkey_i_backpointer bp_k; bkey_backpointer_init(&bp_k.k_i); - bp_k.k.p = bucket_pos_to_bp(trans->c, bucket, bp.bucket_offset); + bp_k.k.p = bucket_pos_to_bp(ca, bucket, bp.bucket_offset); bp_k.v = bp; if (!insert) { @@ -142,7 +157,7 @@ static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, struct bch_dev *ca, }; } -int bch2_get_next_backpointer(struct btree_trans *, struct bpos, int, +int bch2_get_next_backpointer(struct btree_trans *, struct bch_dev *ca, struct bpos, int, struct bpos *, struct bch_backpointer *, unsigned); struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *, struct btree_iter *, struct bpos, struct bch_backpointer, diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 4774f9c7f060..ae6d6879e2e0 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -988,7 +988,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, goto err; if (!p.ptr.cached) { - ret = bch2_bucket_backpointer_mod(trans, bucket, bp, k, insert); + ret = bch2_bucket_backpointer_mod(trans, ca, bucket, bp, k, insert); if (ret) goto err; } diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index f09df7f9036a..6d41530615cb 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -1198,6 +1198,7 @@ err: } static int ec_stripe_update_extent(struct btree_trans *trans, + struct bch_dev *ca, struct bpos bucket, u8 gen, struct ec_stripe_buf *s, struct bpos *bp_pos) @@ -1213,7 +1214,7 @@ static int ec_stripe_update_extent(struct btree_trans *trans, struct bkey_i *n; int ret, dev, block; - ret = bch2_get_next_backpointer(trans, bucket, gen, + ret = bch2_get_next_backpointer(trans, ca, bucket, gen, bp_pos, &bp, BTREE_ITER_cached); if (ret) return ret; @@ -1311,7 +1312,7 @@ static int ec_stripe_update_bucket(struct btree_trans *trans, struct ec_stripe_b ret = commit_do(trans, NULL, NULL, BCH_TRANS_COMMIT_no_check_rw| BCH_TRANS_COMMIT_no_enospc, - ec_stripe_update_extent(trans, bucket_pos, ptr.gen, s, &bp_pos)); + ec_stripe_update_extent(trans, ca, bucket_pos, ptr.gen, s, &bp_pos)); if (ret) break; if (bkey_eq(bp_pos, POS_MAX)) diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index c441ce7c92ec..0e3203de1cd6 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -729,7 +729,7 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, bch2_trans_begin(trans); - ret = bch2_get_next_backpointer(trans, bucket, gen, + ret = bch2_get_next_backpointer(trans, ca, bucket, gen, &bp_pos, &bp, BTREE_ITER_cached); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) -- cgit v1.2.3-59-g8ed1b From dbd0408087853c7842aa21f58f99d395eff02544 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 20:32:44 -0400 Subject: bcachefs: move replica_set from bch_dev to bch_fs This is needed for the next patch - the write submit path has to be able to allocate a replica bio even when we weren't able to get a ref on the device. Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 4 +--- fs/bcachefs/io_write.c | 8 ++++---- fs/bcachefs/super.c | 3 --- 3 files changed, 5 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index f5ba8c6a34d7..22d303813d31 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -569,9 +569,6 @@ struct bch_dev { struct bch_devs_mask self; - /* biosets used in cloned bios for writing multiple replicas */ - struct bio_set replica_set; - /* * Buckets: * Per-bucket arrays are protected by c->mark_lock, bucket_lock and @@ -995,6 +992,7 @@ struct bch_fs { struct bio_set bio_read; struct bio_set bio_read_split; struct bio_set bio_write; + struct bio_set replica_set; struct mutex bio_bounce_pages_lock; mempool_t bio_bounce_pages; struct bucket_nocow_lock_table diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 40c8f8844224..5d9f349a1504 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -412,8 +412,7 @@ void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); if (to_entry(ptr + 1) < ptrs.end) { - n = to_wbio(bio_alloc_clone(NULL, &wbio->bio, - GFP_NOFS, &ca->replica_set)); + n = to_wbio(bio_alloc_clone(NULL, &wbio->bio, GFP_NOFS, &c->replica_set)); n->bio.bi_end_io = wbio->bio.bi_end_io; n->bio.bi_private = wbio->bio.bi_private; @@ -1667,13 +1666,14 @@ void bch2_write_op_to_text(struct printbuf *out, struct bch_write_op *op) void bch2_fs_io_write_exit(struct bch_fs *c) { mempool_exit(&c->bio_bounce_pages); + bioset_exit(&c->replica_set); bioset_exit(&c->bio_write); } int bch2_fs_io_write_init(struct bch_fs *c) { - if (bioset_init(&c->bio_write, 1, offsetof(struct bch_write_bio, bio), - BIOSET_NEED_BVECS)) + if (bioset_init(&c->bio_write, 1, offsetof(struct bch_write_bio, bio), BIOSET_NEED_BVECS) || + bioset_init(&c->replica_set, 4, offsetof(struct bch_write_bio, bio), 0)) return -BCH_ERR_ENOMEM_bio_write_init; if (mempool_init_page_pool(&c->bio_bounce_pages, diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index ed135166ab86..294a9d35a9f2 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -1193,7 +1193,6 @@ static void bch2_dev_free(struct bch_dev *ca) bch2_dev_journal_exit(ca); free_percpu(ca->io_done); - bioset_exit(&ca->replica_set); bch2_dev_buckets_free(ca); free_page((unsigned long) ca->sb_read_scratch); @@ -1317,8 +1316,6 @@ static struct bch_dev *__bch2_dev_alloc(struct bch_fs *c, PERCPU_REF_INIT_DEAD, GFP_KERNEL) || !(ca->sb_read_scratch = (void *) __get_free_page(GFP_KERNEL)) || bch2_dev_buckets_alloc(c, ca) || - bioset_init(&ca->replica_set, 4, - offsetof(struct bch_write_bio, bio), 0) || !(ca->io_done = alloc_percpu(*ca->io_done))) goto err; -- cgit v1.2.3-59-g8ed1b From 8783856ab15e9ad6faea8df3b72658f4cd2325c3 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 20:38:05 -0400 Subject: bcachefs: ob_dev() Wrapper around bch2_dev_have_ref() for open_buckets; we do guarantee that the device an open_bucket points to exists. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 15 +++++++-------- fs/bcachefs/alloc_foreground.h | 7 ++++++- fs/bcachefs/ec.c | 7 ++----- fs/bcachefs/sb-members.h | 7 +++++++ 4 files changed, 22 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index cc182b263066..77f3723e0fc0 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -100,7 +100,7 @@ static void bch2_open_bucket_hash_remove(struct bch_fs *c, struct open_bucket *o void __bch2_open_bucket_put(struct bch_fs *c, struct open_bucket *ob) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = ob_dev(c, ob); if (ob->ec) { ec_stripe_new_put(c, ob->ec, STRIPE_REF_io); @@ -684,8 +684,7 @@ static int add_new_bucket(struct bch_fs *c, unsigned flags, struct open_bucket *ob) { - unsigned durability = - bch2_dev_bkey_exists(c, ob->dev)->mi.durability; + unsigned durability = ob_dev(c, ob)->mi.durability; BUG_ON(*nr_effective >= nr_replicas); @@ -831,7 +830,7 @@ static bool want_bucket(struct bch_fs *c, bool *have_cache, bool ec, struct open_bucket *ob) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = ob_dev(c, ob); if (!test_bit(ob->dev, devs_may_alloc->d)) return false; @@ -901,7 +900,7 @@ static int bucket_alloc_set_partial(struct bch_fs *c, struct open_bucket *ob = c->open_buckets + c->open_buckets_partial[i]; if (want_bucket(c, wp, devs_may_alloc, have_cache, ec, ob)) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = ob_dev(c, ob); struct bch_dev_usage usage; u64 avail; @@ -1286,7 +1285,7 @@ deallocate_extra_replicas(struct bch_fs *c, unsigned i; open_bucket_for_each(c, ptrs, ob, i) { - unsigned d = bch2_dev_bkey_exists(c, ob->dev)->mi.durability; + unsigned d = ob_dev(c, ob)->mi.durability; if (d && d <= extra_replicas) { extra_replicas -= d; @@ -1443,7 +1442,7 @@ err: struct bch_extent_ptr bch2_ob_ptr(struct bch_fs *c, struct open_bucket *ob) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = ob_dev(c, ob); return (struct bch_extent_ptr) { .type = 1 << BCH_EXTENT_ENTRY_ptr, @@ -1519,7 +1518,7 @@ void bch2_fs_allocator_foreground_init(struct bch_fs *c) static void bch2_open_bucket_to_text(struct printbuf *out, struct bch_fs *c, struct open_bucket *ob) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = ob_dev(c, ob); unsigned data_type = ob->data_type; barrier(); /* READ_ONCE() doesn't work on bitfields */ diff --git a/fs/bcachefs/alloc_foreground.h b/fs/bcachefs/alloc_foreground.h index 96f24b67d066..a42c9730d32a 100644 --- a/fs/bcachefs/alloc_foreground.h +++ b/fs/bcachefs/alloc_foreground.h @@ -30,6 +30,11 @@ void bch2_dev_stripe_increment(struct bch_dev *, struct dev_stripe_state *); long bch2_bucket_alloc_new_fs(struct bch_dev *); +static inline struct bch_dev *ob_dev(struct bch_fs *c, struct open_bucket *ob) +{ + return bch2_dev_have_ref(c, ob->dev); +} + struct open_bucket *bch2_bucket_alloc(struct bch_fs *, struct bch_dev *, enum bch_watermark, enum bch_data_type, struct closure *); @@ -185,7 +190,7 @@ bch2_alloc_sectors_append_ptrs_inlined(struct bch_fs *c, struct write_point *wp, wp->sectors_allocated += sectors; open_bucket_for_each(c, &wp->ptrs, ob, i) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); + struct bch_dev *ca = ob_dev(c, ob); struct bch_extent_ptr ptr = bch2_ob_ptr(c, ob); ptr.cached = cached || diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 6d41530615cb..fb5aa2b763ca 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -1550,16 +1550,13 @@ void bch2_ec_bucket_cancel(struct bch_fs *c, struct open_bucket *ob) void *bch2_writepoint_ec_buf(struct bch_fs *c, struct write_point *wp) { struct open_bucket *ob = ec_open_bucket(c, &wp->ptrs); - struct bch_dev *ca; - unsigned offset; - if (!ob) return NULL; BUG_ON(!ob->ec->new_stripe.data[ob->ec_idx]); - ca = bch2_dev_bkey_exists(c, ob->dev); - offset = ca->mi.bucket_size - ob->sectors_free; + struct bch_dev *ca = ob_dev(c, ob); + unsigned offset = ca->mi.bucket_size - ob->sectors_free; return ob->ec->new_stripe.data[ob->ec_idx] + (offset << 9); } diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index d6d391dc21ca..a3ba1ddebe4d 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -195,6 +195,13 @@ static inline bool bucket_valid(const struct bch_dev *ca, u64 b) return b - ca->mi.first_bucket < ca->mi.nbuckets_minus_first; } +static inline struct bch_dev *bch2_dev_have_ref(const struct bch_fs *c, unsigned dev) +{ + EBUG_ON(!bch2_dev_exists(c, dev)); + + return rcu_dereference_check(c->devs[dev], 1); +} + /* * If a key exists that references a device, the device won't be going away and * we can omit rcu_read_lock(): -- cgit v1.2.3-59-g8ed1b From c387d84413711388a1bc5bf551c38f4b02529cc4 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 20:42:50 -0400 Subject: bcachefs: ec_validate_checksums() -> bch2_dev_tryget() Signed-off-by: Kent Overstreet --- fs/bcachefs/ec.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index fb5aa2b763ca..be8bddb38957 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -633,19 +633,21 @@ static void ec_validate_checksums(struct bch_fs *c, struct ec_stripe_buf *buf) struct bch_csum got = ec_block_checksum(buf, i, offset); if (bch2_crc_cmp(want, got)) { - struct printbuf err = PRINTBUF; - struct bch_dev *ca = bch2_dev_bkey_exists(c, v->ptrs[i].dev); + struct bch_dev *ca = bch2_dev_tryget(c, v->ptrs[i].dev); + if (ca) { + struct printbuf err = PRINTBUF; - prt_str(&err, "stripe "); - bch2_csum_err_msg(&err, v->csum_type, want, got); - prt_printf(&err, " for %ps at %u of\n ", (void *) _RET_IP_, i); - bch2_bkey_val_to_text(&err, c, bkey_i_to_s_c(&buf->key)); - bch_err_ratelimited(ca, "%s", err.buf); - printbuf_exit(&err); + prt_str(&err, "stripe "); + bch2_csum_err_msg(&err, v->csum_type, want, got); + prt_printf(&err, " for %ps at %u of\n ", (void *) _RET_IP_, i); + bch2_bkey_val_to_text(&err, c, bkey_i_to_s_c(&buf->key)); + bch_err_ratelimited(ca, "%s", err.buf); + printbuf_exit(&err); - clear_bit(i, buf->valid); + bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); + } - bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); + clear_bit(i, buf->valid); break; } -- cgit v1.2.3-59-g8ed1b From 3793b3f91f88269ff8ad8ed2b95c6779415e7fd5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 20:49:22 -0400 Subject: bcachefs: bch2_extent_merge() -> bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 6376440f6f30..46f8164b1a49 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -252,7 +252,6 @@ bool bch2_extent_merge(struct bch_fs *c, struct bkey_s l, struct bkey_s_c r) const union bch_extent_entry *en_r; struct extent_ptr_decoded lp, rp; bool use_right_ptr; - struct bch_dev *ca; en_l = l_ptrs.start; en_r = r_ptrs.start; @@ -283,8 +282,12 @@ bool bch2_extent_merge(struct bch_fs *c, struct bkey_s l, struct bkey_s_c r) return false; /* Extents may not straddle buckets: */ - ca = bch2_dev_bkey_exists(c, lp.ptr.dev); - if (PTR_BUCKET_NR(ca, &lp.ptr) != PTR_BUCKET_NR(ca, &rp.ptr)) + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, lp.ptr.dev); + bool same_bucket = ca && PTR_BUCKET_NR(ca, &lp.ptr) == PTR_BUCKET_NR(ca, &rp.ptr); + rcu_read_unlock(); + + if (!same_bucket) return false; if (lp.has_ec != rp.has_ec || -- cgit v1.2.3-59-g8ed1b From 302c980a818763d53dfa8fcd8549b04a5c7b8b17 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 20:54:20 -0400 Subject: bcachefs: extent_ptr_durability() -> bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/data_update.c | 5 +++++ fs/bcachefs/extents.c | 12 ++++++++---- fs/bcachefs/io_write.c | 7 ++++++- fs/bcachefs/move.c | 2 ++ 4 files changed, 21 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 1cf92bea7f9f..7aad6085ef53 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -203,6 +203,8 @@ restart_drop_conflicting_replicas: /* Now, drop excess replicas: */ restart_drop_extra_replicas: + + rcu_read_lock(); bkey_for_each_ptr_decode(old.k, bch2_bkey_ptrs(bkey_i_to_s(insert)), p, entry) { unsigned ptr_durability = bch2_extent_ptr_durability(c, &p); @@ -214,6 +216,7 @@ restart_drop_extra_replicas: goto restart_drop_extra_replicas; } } + rcu_read_unlock(); /* Finally, add the pointers we just wrote: */ extent_for_each_ptr_decode(extent_i_to_s(new), p, entry) @@ -552,6 +555,7 @@ int bch2_data_update_init(struct btree_trans *trans, struct bpos bucket = PTR_BUCKET_POS(ca, &p.ptr); bool locked; + rcu_read_lock(); if (((1U << i) & m->data_opts.rewrite_ptrs)) { BUG_ON(p.ptr.cached); @@ -565,6 +569,7 @@ int bch2_data_update_init(struct btree_trans *trans, bch2_dev_list_add_dev(&m->op.devs_have, p.ptr.dev); durability_have += bch2_extent_ptr_durability(c, &p); } + rcu_read_unlock(); /* * op->csum_type is normally initialized from the fs/file's diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 46f8164b1a49..dc78ace56181 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -675,16 +675,16 @@ static inline unsigned __extent_ptr_durability(struct bch_dev *ca, struct extent unsigned bch2_extent_ptr_desired_durability(struct bch_fs *c, struct extent_ptr_decoded *p) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p->ptr.dev); + struct bch_dev *ca = bch2_dev_rcu(c, p->ptr.dev); - return __extent_ptr_durability(ca, p); + return ca ? __extent_ptr_durability(ca, p) : 0; } unsigned bch2_extent_ptr_durability(struct bch_fs *c, struct extent_ptr_decoded *p) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p->ptr.dev); + struct bch_dev *ca = bch2_dev_rcu(c, p->ptr.dev); - if (ca->mi.state == BCH_MEMBER_STATE_failed) + if (!ca || ca->mi.state == BCH_MEMBER_STATE_failed) return 0; return __extent_ptr_durability(ca, p); @@ -697,8 +697,10 @@ unsigned bch2_bkey_durability(struct bch_fs *c, struct bkey_s_c k) struct extent_ptr_decoded p; unsigned durability = 0; + rcu_read_lock(); bkey_for_each_ptr_decode(k.k, ptrs, p, entry) durability += bch2_extent_ptr_durability(c, &p); + rcu_read_unlock(); return durability; } @@ -710,9 +712,11 @@ static unsigned bch2_bkey_durability_safe(struct bch_fs *c, struct bkey_s_c k) struct extent_ptr_decoded p; unsigned durability = 0; + rcu_read_lock(); bkey_for_each_ptr_decode(k.k, ptrs, p, entry) if (p.ptr.dev < c->sb.nr_devices && c->devs[p.ptr.dev]) durability += bch2_extent_ptr_durability(c, &p); + rcu_read_unlock(); return durability; } diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 5d9f349a1504..285e1a23bde5 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1099,12 +1099,17 @@ static bool bch2_extent_is_writeable(struct bch_write_op *op, return false; e = bkey_s_c_to_extent(k); + + rcu_read_lock(); extent_for_each_ptr_decode(e, p, entry) { - if (crc_is_encoded(p.crc) || p.has_ec) + if (crc_is_encoded(p.crc) || p.has_ec) { + rcu_read_unlock(); return false; + } replicas += bch2_extent_ptr_durability(c, &p); } + rcu_read_unlock(); return replicas >= op->opts.data_replicas; } diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index 0e3203de1cd6..8171f947fac8 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -1033,6 +1033,7 @@ static bool drop_extra_replicas_pred(struct bch_fs *c, void *arg, struct extent_ptr_decoded p; unsigned i = 0; + rcu_read_lock(); bkey_for_each_ptr_decode(k.k, bch2_bkey_ptrs_c(k), p, entry) { unsigned d = bch2_extent_ptr_durability(c, &p); @@ -1043,6 +1044,7 @@ static bool drop_extra_replicas_pred(struct bch_fs *c, void *arg, i++; } + rcu_read_unlock(); return data_opts->kill_ptrs != 0; } -- cgit v1.2.3-59-g8ed1b From 3858aa4268b2f71827097f548879e4405b4cc995 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 20:56:54 -0400 Subject: bcachefs: ptr_stale() -> dev_ptr_stale() Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 2 +- fs/bcachefs/btree_gc.c | 2 +- fs/bcachefs/buckets.h | 14 ++++++++------ fs/bcachefs/ec.c | 4 ++-- fs/bcachefs/extents.c | 8 ++++---- fs/bcachefs/io_read.c | 4 ++-- 6 files changed, 18 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 22d303813d31..ab415f0dd14e 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -573,7 +573,7 @@ struct bch_dev { * Buckets: * Per-bucket arrays are protected by c->mark_lock, bucket_lock and * gc_lock, for device resize - holding any is sufficient for access: - * Or rcu_read_lock(), but only for ptr_stale(): + * Or rcu_read_lock(), but only for dev_ptr_stale(): */ struct bucket_array __rcu *buckets_gc; struct bucket_gens __rcu *bucket_gens; diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 07b343e7ddeb..05ebf1186be1 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -1257,7 +1257,7 @@ static int gc_btree_gens_key(struct btree_trans *trans, bkey_for_each_ptr(ptrs, ptr) { struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - if (ptr_stale(ca, ptr) > 16) { + if (dev_ptr_stale(ca, ptr) > 16) { percpu_up_read(&c->mark_lock); goto update; } diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index 0a54faf7c50a..617ffde2fb7a 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -170,17 +170,19 @@ static inline int gen_after(u8 a, u8 b) return r > 0 ? r : 0; } +static inline u8 dev_ptr_stale_rcu(struct bch_dev *ca, const struct bch_extent_ptr *ptr) +{ + return gen_after(*bucket_gen(ca, PTR_BUCKET_NR(ca, ptr)), ptr->gen); +} + /** - * ptr_stale() - check if a pointer points into a bucket that has been + * dev_ptr_stale() - check if a pointer points into a bucket that has been * invalidated. */ -static inline u8 ptr_stale(struct bch_dev *ca, - const struct bch_extent_ptr *ptr) +static inline u8 dev_ptr_stale(struct bch_dev *ca, const struct bch_extent_ptr *ptr) { - u8 ret; - rcu_read_lock(); - ret = gen_after(*bucket_gen(ca, PTR_BUCKET_NR(ca, ptr)), ptr->gen); + u8 ret = dev_ptr_stale_rcu(ca, ptr); rcu_read_unlock(); return ret; diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index be8bddb38957..99a8c8fe8e3b 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -714,7 +714,7 @@ static void ec_block_endio(struct bio *bio) bch2_blk_status_to_str(bio->bi_status))) clear_bit(ec_bio->idx, ec_bio->buf->valid); - if (ptr_stale(ca, ptr)) { + if (dev_ptr_stale(ca, ptr)) { bch_err_ratelimited(ca->fs, "error %s stripe: stale pointer after io", bio_data_dir(bio) == READ ? "reading from" : "writing to"); @@ -738,7 +738,7 @@ static void ec_block_io(struct bch_fs *c, struct ec_stripe_buf *buf, : BCH_DATA_parity; int rw = op_is_write(opf); - if (ptr_stale(ca, ptr)) { + if (dev_ptr_stale(ca, ptr)) { bch_err_ratelimited(c, "error %s stripe: stale pointer", rw == READ ? "reading from" : "writing to"); diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index dc78ace56181..a12fffb32f61 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -132,7 +132,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, if (!ret && !p.ptr.cached) ret = -EIO; - if (p.ptr.cached && ptr_stale(ca, &p.ptr)) + if (p.ptr.cached && dev_ptr_stale(ca, &p.ptr)) continue; f = failed ? dev_io_failures(failed, p.ptr.dev) : NULL; @@ -874,7 +874,7 @@ bool bch2_bkey_has_target(struct bch_fs *c, struct bkey_s_c k, unsigned target) bkey_for_each_ptr(ptrs, ptr) if (bch2_dev_in_target(c, ptr->dev, target) && (!ptr->cached || - !ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr))) + !dev_ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr))) return true; return false; @@ -981,7 +981,7 @@ bool bch2_extent_normalize(struct bch_fs *c, struct bkey_s k) { bch2_bkey_drop_ptrs(k, ptr, ptr->cached && - ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr)); + dev_ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr)); return bkey_deleted(k.k); } @@ -1005,7 +1005,7 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struc prt_str(out, " cached"); if (ptr->unwritten) prt_str(out, " unwritten"); - if (bucket_valid(ca, b) && ptr_stale(ca, ptr)) + if (bucket_valid(ca, b) && dev_ptr_stale_rcu(ca, ptr)) prt_printf(out, " stale"); } rcu_read_unlock(); diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 1091b066639e..0107b2fb4d33 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -697,7 +697,7 @@ static void bch2_read_endio(struct bio *bio) } if (((rbio->flags & BCH_READ_RETRY_IF_STALE) && race_fault()) || - ptr_stale(ca, &rbio->pick.ptr)) { + dev_ptr_stale(ca, &rbio->pick.ptr)) { trace_and_count(c, read_reuse_race, &rbio->bio); if (rbio->flags & BCH_READ_RETRY_IF_STALE) @@ -841,7 +841,7 @@ retry_pick: */ if ((flags & BCH_READ_IN_RETRY) && !pick.ptr.cached && - unlikely(ptr_stale(ca, &pick.ptr))) { + unlikely(dev_ptr_stale(ca, &pick.ptr))) { read_from_stale_dirty_pointer(trans, k, pick.ptr); bch2_mark_io_failure(failed, &pick); goto retry_pick; -- cgit v1.2.3-59-g8ed1b From 8feecbed241b96957cd286e68bbcc113d2e45750 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 21:09:45 -0400 Subject: bcachefs: extent_ptr_invalid() -> bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 30 +++++++++++++++++++----------- 1 file changed, 19 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index a12fffb32f61..5f5038f0ca7c 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -1083,9 +1083,6 @@ static int extent_ptr_invalid(struct bch_fs *c, struct printbuf *err) { struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - u64 bucket; - u32 bucket_offset; - struct bch_dev *ca; int ret = 0; if (!bch2_dev_exists(c, ptr->dev)) { @@ -1101,24 +1098,35 @@ static int extent_ptr_invalid(struct bch_fs *c, "pointer to invalid device (%u)", ptr->dev); } - ca = bch2_dev_bkey_exists(c, ptr->dev); + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); + if (!ca) { + rcu_read_unlock(); + return 0; + } + u32 bucket_offset; + u64 bucket = sector_to_bucket_and_offset(ca, ptr->offset, &bucket_offset); + unsigned first_bucket = ca->mi.first_bucket; + u64 nbuckets = ca->mi.nbuckets; + unsigned bucket_size = ca->mi.bucket_size; + rcu_read_unlock(); + bkey_for_each_ptr(ptrs, ptr2) bkey_fsck_err_on(ptr != ptr2 && ptr->dev == ptr2->dev, c, err, ptr_to_duplicate_device, "multiple pointers to same device (%u)", ptr->dev); - bucket = sector_to_bucket_and_offset(ca, ptr->offset, &bucket_offset); - bkey_fsck_err_on(bucket >= ca->mi.nbuckets, c, err, + bkey_fsck_err_on(bucket >= nbuckets, c, err, ptr_after_last_bucket, - "pointer past last bucket (%llu > %llu)", bucket, ca->mi.nbuckets); - bkey_fsck_err_on(ptr->offset < bucket_to_sector(ca, ca->mi.first_bucket), c, err, + "pointer past last bucket (%llu > %llu)", bucket, nbuckets); + bkey_fsck_err_on(bucket < first_bucket, c, err, ptr_before_first_bucket, - "pointer before first bucket (%llu < %u)", bucket, ca->mi.first_bucket); - bkey_fsck_err_on(bucket_offset + size_ondisk > ca->mi.bucket_size, c, err, + "pointer before first bucket (%llu < %u)", bucket, first_bucket); + bkey_fsck_err_on(bucket_offset + size_ondisk > bucket_size, c, err, ptr_spans_multiple_buckets, "pointer spans multiple buckets (%u + %u > %u)", - bucket_offset, size_ondisk, ca->mi.bucket_size); + bucket_offset, size_ondisk, bucket_size); fsck_err: return ret; } -- cgit v1.2.3-59-g8ed1b From 8e3cc2003fd55f32b82c2a06ae7f0bd28b938bd7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 21:12:31 -0400 Subject: bcachefs: bch2_bkey_has_target() -> bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 5f5038f0ca7c..52e79b87b485 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -870,14 +870,21 @@ const struct bch_extent_ptr *bch2_bkey_has_device_c(struct bkey_s_c k, unsigned bool bch2_bkey_has_target(struct bch_fs *c, struct bkey_s_c k, unsigned target) { struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + struct bch_dev *ca; + bool ret = false; + rcu_read_lock(); bkey_for_each_ptr(ptrs, ptr) if (bch2_dev_in_target(c, ptr->dev, target) && + (ca = bch2_dev_rcu(c, ptr->dev)) && (!ptr->cached || - !dev_ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr))) - return true; + !dev_ptr_stale_rcu(ca, ptr))) { + ret = true; + break; + } + rcu_read_unlock(); - return false; + return ret; } bool bch2_bkey_matches_ptr(struct bch_fs *c, struct bkey_s_c k, -- cgit v1.2.3-59-g8ed1b From 9cadb4ea56d4d961ba0efd47321905b3fda1c664 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 21:12:46 -0400 Subject: bcachefs: bch2_extent_normalize() -> bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 52e79b87b485..7b889f93878b 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -986,9 +986,14 @@ void bch2_extent_ptr_set_cached(struct bkey_s k, struct bch_extent_ptr *ptr) */ bool bch2_extent_normalize(struct bch_fs *c, struct bkey_s k) { + struct bch_dev *ca; + + rcu_read_lock(); bch2_bkey_drop_ptrs(k, ptr, ptr->cached && - dev_ptr_stale(bch2_dev_bkey_exists(c, ptr->dev), ptr)); + (ca = bch2_dev_rcu(c, ptr->dev)) && + dev_ptr_stale_rcu(ca, ptr)); + rcu_read_unlock(); return bkey_deleted(k.k); } -- cgit v1.2.3-59-g8ed1b From ad897d241b703a3e81fd4785a4b007168471e9e2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 22:54:50 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in btree_gc.c Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_gc.c | 41 +++++++++++++++++++++++++++++++---------- 1 file changed, 31 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 05ebf1186be1..8035c8b797ab 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -993,19 +993,25 @@ static int bch2_gc_alloc_start(struct bch_fs *c) rcu_assign_pointer(ca->buckets_gc, buckets); } + struct bch_dev *ca = NULL; int ret = bch2_trans_run(c, for_each_btree_key(trans, iter, BTREE_ID_alloc, POS_MIN, BTREE_ITER_prefetch, k, ({ - struct bch_dev *ca = bch2_dev_bkey_exists(c, k.k->p.inode); - struct bucket *g = gc_bucket(ca, k.k->p.offset); + ca = bch2_dev_iterate(c, ca, k.k->p.inode); + if (!ca) { + bch2_btree_iter_set_pos(&iter, POS(k.k->p.inode + 1, 0)); + continue; + } struct bch_alloc_v4 a_convert; const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert); + struct bucket *g = gc_bucket(ca, k.k->p.offset); g->gen_valid = 1; g->gen = a->gen; 0; }))); + bch2_dev_put(ca); bch_err_fn(c, ret); return ret; } @@ -1254,22 +1260,29 @@ static int gc_btree_gens_key(struct btree_trans *trans, return -EROFS; percpu_down_read(&c->mark_lock); + rcu_read_lock(); bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); + if (!ca) + continue; if (dev_ptr_stale(ca, ptr) > 16) { + rcu_read_unlock(); percpu_up_read(&c->mark_lock); goto update; } } bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - u8 *gen = &ca->oldest_gen[PTR_BUCKET_NR(ca, ptr)]; + struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); + if (!ca) + continue; + u8 *gen = &ca->oldest_gen[PTR_BUCKET_NR(ca, ptr)]; if (gen_after(*gen, ptr->gen)) *gen = ptr->gen; } + rcu_read_unlock(); percpu_up_read(&c->mark_lock); return 0; update: @@ -1282,10 +1295,9 @@ update: return 0; } -static int bch2_alloc_write_oldest_gen(struct btree_trans *trans, struct btree_iter *iter, - struct bkey_s_c k) +static int bch2_alloc_write_oldest_gen(struct btree_trans *trans, struct bch_dev *ca, + struct btree_iter *iter, struct bkey_s_c k) { - struct bch_dev *ca = bch2_dev_bkey_exists(trans->c, iter->pos.inode); struct bch_alloc_v4 a_convert; const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert); struct bkey_i_alloc_v4 *a_mut; @@ -1355,14 +1367,23 @@ int bch2_gc_gens(struct bch_fs *c) goto err; } + struct bch_dev *ca = NULL; ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_alloc, POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, - BCH_TRANS_COMMIT_no_enospc, - bch2_alloc_write_oldest_gen(trans, &iter, k))); + BCH_TRANS_COMMIT_no_enospc, ({ + ca = bch2_dev_iterate(c, ca, k.k->p.inode); + if (!ca) { + bch2_btree_iter_set_pos(&iter, POS(k.k->p.inode + 1, 0)); + continue; + } + bch2_alloc_write_oldest_gen(trans, ca, &iter, k); + }))); + bch2_dev_put(ca); + if (ret) goto err; -- cgit v1.2.3-59-g8ed1b From 78e9b548f37b92ef81e1bc8cb4dfd4165c5a97d4 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 03:26:37 -0400 Subject: bcachefs: bch2_dev_bucket_exists() uses bch2_dev_rcu() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.h | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index a75eba6d9586..5f42dc1e7fcc 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -15,13 +15,11 @@ enum bkey_invalid_flags; static inline bool bch2_dev_bucket_exists(struct bch_fs *c, struct bpos pos) { - struct bch_dev *ca; - - if (!bch2_dev_exists(c, pos.inode)) - return false; - - ca = bch2_dev_bkey_exists(c, pos.inode); - return bucket_valid(ca, pos.offset); + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, pos.inode); + bool ret = ca && bucket_valid(ca, pos.offset); + rcu_read_unlock(); + return ret; } static inline u64 bucket_to_u64(struct bpos bucket) -- cgit v1.2.3-59-g8ed1b From db39a35dde9dcf77cf599aab5f5c2be35abdcff5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 03:58:13 -0400 Subject: bcachefs: pass bch_dev to read_from_stale_dirty_pointer() Signed-off-by: Kent Overstreet --- fs/bcachefs/io_read.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 0107b2fb4d33..91260d3dbf50 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -758,11 +758,11 @@ err: } static noinline void read_from_stale_dirty_pointer(struct btree_trans *trans, + struct bch_dev *ca, struct bkey_s_c k, struct bch_extent_ptr ptr) { struct bch_fs *c = trans->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr.dev); struct btree_iter iter; struct printbuf buf = PRINTBUF; int ret; @@ -842,7 +842,7 @@ retry_pick: if ((flags & BCH_READ_IN_RETRY) && !pick.ptr.cached && unlikely(dev_ptr_stale(ca, &pick.ptr))) { - read_from_stale_dirty_pointer(trans, k, pick.ptr); + read_from_stale_dirty_pointer(trans, ca, k, pick.ptr); bch2_mark_io_failure(failed, &pick); goto retry_pick; } -- cgit v1.2.3-59-g8ed1b From a9422fd40410842b51496a00297ab1dabde57af0 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 20:49:13 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in bkey_pick_read_device() Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 37 ++++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 7b889f93878b..b785712ae9b2 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -71,6 +71,12 @@ void bch2_mark_io_failure(struct bch_io_failures *failed, } } +static inline u64 dev_latency(struct bch_fs *c, unsigned dev) +{ + struct bch_dev *ca = bch2_dev_rcu(c, dev); + return ca ? atomic64_read(&ca->cur_latency[READ]) : S64_MAX; +} + /* * returns true if p1 is better than p2: */ @@ -79,11 +85,8 @@ static inline bool ptr_better(struct bch_fs *c, const struct extent_ptr_decoded p2) { if (likely(!p1.idx && !p2.idx)) { - struct bch_dev *dev1 = bch2_dev_bkey_exists(c, p1.ptr.dev); - struct bch_dev *dev2 = bch2_dev_bkey_exists(c, p2.ptr.dev); - - u64 l1 = atomic64_read(&dev1->cur_latency[READ]); - u64 l2 = atomic64_read(&dev2->cur_latency[READ]); + u64 l1 = dev_latency(c, p1.ptr.dev); + u64 l2 = dev_latency(c, p2.ptr.dev); /* Pick at random, biased in favor of the faster device: */ @@ -109,21 +112,21 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, const union bch_extent_entry *entry; struct extent_ptr_decoded p; struct bch_dev_io_failures *f; - struct bch_dev *ca; int ret = 0; if (k.k->type == KEY_TYPE_error) return -EIO; + rcu_read_lock(); bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { /* * Unwritten extent: no need to actually read, treat it as a * hole and return 0s: */ - if (p.ptr.unwritten) - return 0; - - ca = bch2_dev_bkey_exists(c, p.ptr.dev); + if (p.ptr.unwritten) { + ret = 0; + break; + } /* * If there are any dirty pointers it's an error if we can't @@ -132,7 +135,9 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, if (!ret && !p.ptr.cached) ret = -EIO; - if (p.ptr.cached && dev_ptr_stale(ca, &p.ptr)) + struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); + + if (p.ptr.cached && (!ca || dev_ptr_stale(ca, &p.ptr))) continue; f = failed ? dev_io_failures(failed, p.ptr.dev) : NULL; @@ -141,12 +146,13 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, ? f->idx : f->idx + 1; - if (!p.idx && - !bch2_dev_is_readable(ca)) + if (!p.idx && !ca) p.idx++; - if (bch2_force_reconstruct_read && - !p.idx && p.has_ec) + if (!p.idx && p.has_ec && bch2_force_reconstruct_read) + p.idx++; + + if (!p.idx && !bch2_dev_is_readable(ca)) p.idx++; if (p.idx >= (unsigned) p.has_ec + 1) @@ -158,6 +164,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, *pick = p; ret = 1; } + rcu_read_unlock(); return ret; } -- cgit v1.2.3-59-g8ed1b From 222eacabc17f360ede4efddda50534f828228ed0 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 18:06:35 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in data_update_init() Signed-off-by: Kent Overstreet --- fs/bcachefs/data_update.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 7aad6085ef53..1dee93eee7ad 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -544,8 +544,16 @@ int bch2_data_update_init(struct btree_trans *trans, m->op.compression_opt = background_compression(io_opts); m->op.watermark = m->data_opts.btree_insert_flags & BCH_WATERMARK_MASK; - bkey_for_each_ptr(ptrs, ptr) - bch2_dev_get(bch2_dev_bkey_exists(c, ptr->dev)); + bkey_for_each_ptr(ptrs, ptr) { + if (!bch2_dev_tryget(c, ptr->dev)) { + bkey_for_each_ptr(ptrs, ptr2) { + if (ptr2 == ptr) + break; + bch2_dev_put(bch2_dev_have_ref(c, ptr2->dev)); + } + return -BCH_ERR_data_update_done; + } + } unsigned durability_have = 0, durability_removing = 0; -- cgit v1.2.3-59-g8ed1b From d8585a79bebcfb3315f23c955e36ab8d8a7749d5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 03:59:45 -0400 Subject: bcachefs: bch2_dev_have_ref() bch2_dev_bkey_exists() is going away; bch2_dev_have_ref() documents that we're looking up a device without checking if it's present because we have a reference to it already. Signed-off-by: Kent Overstreet --- fs/bcachefs/data_update.c | 6 +++--- fs/bcachefs/io_write.c | 4 ++-- fs/bcachefs/journal_io.c | 4 ++-- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 1dee93eee7ad..0d807c2ce9c6 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -360,7 +360,7 @@ void bch2_data_update_exit(struct data_update *update) bch2_bkey_ptrs_c(bkey_i_to_s_c(update->k.k)); bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); if (c->opts.nocow_enabled) bch2_bucket_nocow_unlock(&c->nocow_locks, PTR_BUCKET_POS(ca, ptr), 0); @@ -559,7 +559,7 @@ int bch2_data_update_init(struct btree_trans *trans, i = 0; bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_have_ref(c, p.ptr.dev); struct bpos bucket = PTR_BUCKET_POS(ca, &p.ptr); bool locked; @@ -664,7 +664,7 @@ int bch2_data_update_init(struct btree_trans *trans, err: i = 0; bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_have_ref(c, p.ptr.dev); struct bpos bucket = PTR_BUCKET_POS(ca, &p.ptr); if ((1U << i) & ptrs_locked) bch2_bucket_nocow_unlock(&c->nocow_locks, bucket, 0); diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 285e1a23bde5..33202b77f6cf 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1303,7 +1303,7 @@ retry: bch2_cut_back(POS(op->pos.inode, op->pos.offset + bio_sectors(bio)), op->insert_keys.top); darray_for_each(buckets, i) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, i->b.inode); + struct bch_dev *ca = bch2_dev_have_ref(c, i->b.inode); __bch2_bucket_nocow_lock(&c->nocow_locks, i->l, bucket_to_u64(i->b), @@ -1380,7 +1380,7 @@ err: return; err_get_ioref: darray_for_each(buckets, i) - percpu_ref_put(&bch2_dev_bkey_exists(c, i->b.inode)->io_ref); + percpu_ref_put(&bch2_dev_have_ref(c, i->b.inode)->io_ref); /* Fall back to COW path: */ goto out; diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index e76eb98af74a..9561f1c2ce1a 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1386,7 +1386,7 @@ int bch2_journal_read(struct bch_fs *c, continue; darray_for_each(i->ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_have_ref(c, ptr->dev); if (!ptr->csum_good) bch_err_dev_offset(ca, ptr->sector, @@ -1396,7 +1396,7 @@ int bch2_journal_read(struct bch_fs *c, } ret = jset_validate(c, - bch2_dev_bkey_exists(c, i->ptrs.data[0].dev), + bch2_dev_have_ref(c, i->ptrs.data[0].dev), &i->j, i->ptrs.data[0].sector, READ); -- cgit v1.2.3-59-g8ed1b From bc3204c80ab69ac64a03b070623b9e7b084e2b2c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 17:04:08 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in check_alloc_info() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 76 ++++++++++++++++++++---------------------- fs/bcachefs/sb-members.h | 8 +++++ 2 files changed, 44 insertions(+), 40 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 69a77d268edf..ef1d17fbd750 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -960,35 +960,34 @@ static struct bkey_s_c bch2_get_key_or_hole(struct btree_iter *iter, struct bpos } } -static bool next_bucket(struct bch_fs *c, struct bpos *bucket) +static bool next_bucket(struct bch_fs *c, struct bch_dev **ca, struct bpos *bucket) { - struct bch_dev *ca; - - if (bch2_dev_bucket_exists(c, *bucket)) - return true; + if (*ca) { + if (bucket->offset < (*ca)->mi.first_bucket) + bucket->offset = (*ca)->mi.first_bucket; - if (bch2_dev_exists(c, bucket->inode)) { - ca = bch2_dev_bkey_exists(c, bucket->inode); - - if (bucket->offset < ca->mi.first_bucket) { - bucket->offset = ca->mi.first_bucket; + if (bucket->offset < (*ca)->mi.nbuckets) return true; - } + bch2_dev_put(*ca); + *ca = NULL; bucket->inode++; bucket->offset = 0; } rcu_read_lock(); - ca = __bch2_next_dev_idx(c, bucket->inode, NULL); - if (ca) - *bucket = POS(ca->dev_idx, ca->mi.first_bucket); + *ca = __bch2_next_dev_idx(c, bucket->inode, NULL); + if (*ca) { + *bucket = POS((*ca)->dev_idx, (*ca)->mi.first_bucket); + bch2_dev_get(*ca); + } rcu_read_unlock(); - return ca != NULL; + return *ca != NULL; } -static struct bkey_s_c bch2_get_key_or_real_bucket_hole(struct btree_iter *iter, struct bkey *hole) +static struct bkey_s_c bch2_get_key_or_real_bucket_hole(struct btree_iter *iter, + struct bch_dev **ca, struct bkey *hole) { struct bch_fs *c = iter->trans->c; struct bkey_s_c k; @@ -997,22 +996,21 @@ again: if (bkey_err(k)) return k; + *ca = bch2_dev_iterate_noerror(c, *ca, k.k->p.inode); + if (!k.k->type) { - struct bpos bucket = bkey_start_pos(k.k); + struct bpos hole_start = bkey_start_pos(k.k); - if (!bch2_dev_bucket_exists(c, bucket)) { - if (!next_bucket(c, &bucket)) + if (!*ca || !bucket_valid(*ca, hole_start.offset)) { + if (!next_bucket(c, ca, &hole_start)) return bkey_s_c_null; - bch2_btree_iter_set_pos(iter, bucket); + bch2_btree_iter_set_pos(iter, hole_start); goto again; } - if (!bch2_dev_bucket_exists(c, k.k->p)) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, bucket.inode); - - bch2_key_resize(hole, ca->mi.nbuckets - bucket.offset); - } + if (k.k->p.offset > (*ca)->mi.nbuckets) + bch2_key_resize(hole, (*ca)->mi.nbuckets - hole_start.offset); } return k; @@ -1154,17 +1152,16 @@ fsck_err: static noinline_for_stack int bch2_check_alloc_hole_freespace(struct btree_trans *trans, + struct bch_dev *ca, struct bpos start, struct bpos *end, struct btree_iter *freespace_iter) { struct bch_fs *c = trans->c; - struct bch_dev *ca; struct bkey_s_c k; struct printbuf buf = PRINTBUF; int ret; - ca = bch2_dev_bkey_exists(c, start.inode); if (!ca->mi.freespace_initialized) return 0; @@ -1342,30 +1339,25 @@ int bch2_check_bucket_gens_key(struct btree_trans *trans, { struct bch_fs *c = trans->c; struct bkey_i_bucket_gens g; - struct bch_dev *ca; u64 start = bucket_gens_pos_to_alloc(k.k->p, 0).offset; u64 end = bucket_gens_pos_to_alloc(bpos_nosnap_successor(k.k->p), 0).offset; u64 b; - bool need_update = false, dev_exists; + bool need_update = false; struct printbuf buf = PRINTBUF; int ret = 0; BUG_ON(k.k->type != KEY_TYPE_bucket_gens); bkey_reassemble(&g.k_i, k); - /* if no bch_dev, skip out whether we repair or not */ - dev_exists = bch2_dev_exists(c, k.k->p.inode); - if (!dev_exists) { - if (fsck_err_on(!dev_exists, c, - bucket_gens_to_invalid_dev, - "bucket_gens key for invalid device:\n %s", - (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { + struct bch_dev *ca = bch2_dev_tryget_noerror(c, k.k->p.inode); + if (!ca) { + if (fsck_err(c, bucket_gens_to_invalid_dev, + "bucket_gens key for invalid device:\n %s", + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) ret = bch2_btree_delete_at(trans, iter, 0); - } goto out; } - ca = bch2_dev_bkey_exists(c, k.k->p.inode); if (fsck_err_on(end <= ca->mi.first_bucket || start >= ca->mi.nbuckets, c, bucket_gens_to_invalid_buckets, @@ -1403,6 +1395,7 @@ int bch2_check_bucket_gens_key(struct btree_trans *trans, } out: fsck_err: + bch2_dev_put(ca); printbuf_exit(&buf); return ret; } @@ -1411,6 +1404,7 @@ int bch2_check_alloc_info(struct bch_fs *c) { struct btree_trans *trans = bch2_trans_get(c); struct btree_iter iter, discard_iter, freespace_iter, bucket_gens_iter; + struct bch_dev *ca = NULL; struct bkey hole; struct bkey_s_c k; int ret = 0; @@ -1429,7 +1423,7 @@ int bch2_check_alloc_info(struct bch_fs *c) bch2_trans_begin(trans); - k = bch2_get_key_or_real_bucket_hole(&iter, &hole); + k = bch2_get_key_or_real_bucket_hole(&iter, &ca, &hole); ret = bkey_err(k); if (ret) goto bkey_err; @@ -1450,7 +1444,7 @@ int bch2_check_alloc_info(struct bch_fs *c) } else { next = k.k->p; - ret = bch2_check_alloc_hole_freespace(trans, + ret = bch2_check_alloc_hole_freespace(trans, ca, bkey_start_pos(k.k), &next, &freespace_iter) ?: @@ -1478,6 +1472,8 @@ bkey_err: bch2_trans_iter_exit(trans, &freespace_iter); bch2_trans_iter_exit(trans, &discard_iter); bch2_trans_iter_exit(trans, &iter); + bch2_dev_put(ca); + ca = NULL; if (ret < 0) goto err; diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index a3ba1ddebe4d..7cecac44ab80 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -269,6 +269,14 @@ static inline struct bch_dev *bch2_dev_bucket_tryget(struct bch_fs *c, struct bp return ca; } +static inline struct bch_dev *bch2_dev_iterate_noerror(struct bch_fs *c, struct bch_dev *ca, unsigned dev_idx) +{ + if (ca && ca->dev_idx == dev_idx) + return ca; + bch2_dev_put(ca); + return bch2_dev_tryget_noerror(c, dev_idx); +} + static inline struct bch_dev *bch2_dev_iterate(struct bch_fs *c, struct bch_dev *ca, unsigned dev_idx) { if (ca && ca->dev_idx == dev_idx) -- cgit v1.2.3-59-g8ed1b From f4301b635a2ee6588d61d03fe261a2bba9225eb7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 17:08:40 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in discard_one_bucket_fast() Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index ef1d17fbd750..71b727f5d432 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1903,9 +1903,12 @@ static void bch2_do_discards_fast_work(struct work_struct *work) static void bch2_discard_one_bucket_fast(struct bch_fs *c, struct bpos bucket) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, bucket.inode); + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, bucket.inode); + bool dead = !ca || percpu_ref_is_dying(&ca->io_ref); + rcu_read_unlock(); - if (!percpu_ref_is_dying(&ca->io_ref) && + if (!dead && !discard_in_flight_add(c, bucket) && bch2_write_ref_tryget(c, BCH_WRITE_REF_discard_fast) && !queue_work(c->write_ref_wq, &c->discard_fast_work)) -- cgit v1.2.3-59-g8ed1b From b6d29b586920daf9f1c5e59ffea91ed162842781 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 18:20:52 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in journal_ptrs_to_text() Signed-off-by: Kent Overstreet --- fs/bcachefs/journal_io.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 9561f1c2ce1a..c9744546d48d 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -49,11 +49,6 @@ void bch2_journal_ptrs_to_text(struct printbuf *out, struct bch_fs *c, struct journal_replay *j) { darray_for_each(j->ptrs, i) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, i->dev); - u64 offset; - - div64_u64_rem(i->sector, ca->mi.bucket_size, &offset); - if (i != j->ptrs.data) prt_printf(out, " "); prt_printf(out, "%u:%u:%u (sector %llu)", -- cgit v1.2.3-59-g8ed1b From 62025697778cb5ca6173b3e52d7e7270edc65a2c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 18:56:40 -0400 Subject: bcachefs: Move nocow unlock to bch2_write_endio() This fixes a lifetime issue; bch2_nocow_write_unlock() uses PTR_BUCKET_POS(), which needs the device - but we drop our ref to the device in bch2_write_endio(). Signed-off-by: Kent Overstreet --- fs/bcachefs/io_write.c | 26 +++++++------------------- fs/bcachefs/io_write_types.h | 1 + 2 files changed, 8 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 33202b77f6cf..fa3413d19b6a 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -434,6 +434,8 @@ void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, n->nocow = nocow; n->submit_time = local_clock(); n->inode_offset = bkey_start_offset(&k->k); + if (nocow) + n->nocow_bucket = PTR_BUCKET_NR(ca, ptr); n->bio.bi_iter.bi_sector = ptr->offset; if (likely(n->have_ioref)) { @@ -659,8 +661,12 @@ static void bch2_write_endio(struct bio *bio) op->flags |= BCH_WRITE_IO_ERROR; } - if (wbio->nocow) + if (wbio->nocow) { + bch2_bucket_nocow_unlock(&c->nocow_locks, + POS(ca->dev_idx, wbio->nocow_bucket), + BUCKET_NOCOW_LOCK_UPDATE); set_bit(wbio->dev, op->devs_need_flush->d); + } if (wbio->have_ioref) { bch2_latency_acct(ca, wbio->submit_time, WRITE); @@ -1114,22 +1120,6 @@ static bool bch2_extent_is_writeable(struct bch_write_op *op, return replicas >= op->opts.data_replicas; } -static inline void bch2_nocow_write_unlock(struct bch_write_op *op) -{ - struct bch_fs *c = op->c; - - for_each_keylist_key(&op->insert_keys, k) { - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(bkey_i_to_s_c(k)); - - bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - bch2_bucket_nocow_unlock(&c->nocow_locks, - PTR_BUCKET_POS(ca, ptr), - BUCKET_NOCOW_LOCK_UPDATE); - } - } -} - static int bch2_nocow_write_convert_one_unwritten(struct btree_trans *trans, struct btree_iter *iter, struct bkey_i *orig, @@ -1200,8 +1190,6 @@ static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op) static void __bch2_nocow_write_done(struct bch_write_op *op) { - bch2_nocow_write_unlock(op); - if (unlikely(op->flags & BCH_WRITE_IO_ERROR)) { op->error = -EIO; } else if (unlikely(op->flags & BCH_WRITE_CONVERT_UNWRITTEN)) diff --git a/fs/bcachefs/io_write_types.h b/fs/bcachefs/io_write_types.h index c7f97c2c4805..6e878a6f2f0b 100644 --- a/fs/bcachefs/io_write_types.h +++ b/fs/bcachefs/io_write_types.h @@ -20,6 +20,7 @@ struct bch_write_bio { u64 submit_time; u64 inode_offset; + u64 nocow_bucket; struct bch_devs_list failed; u8 dev; -- cgit v1.2.3-59-g8ed1b From 40574946b8a5179ed811bc9783d21bed302eb949 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 4 May 2024 16:46:29 -0400 Subject: bcachefs: Better bucket alloc tracepoints Tracepoints are garbage, and perf trace even cuts off some of our fields. Much nicer to just trace a string, and then we can build nicely formatted output with printbufs. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 65 +++++++++++++++++++--------- fs/bcachefs/trace.h | 97 +++--------------------------------------- 2 files changed, 51 insertions(+), 111 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 77f3723e0fc0..9fbd0d97f68c 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -511,6 +511,44 @@ again: return ob; } +static noinline void trace_bucket_alloc2(struct bch_fs *c, struct bch_dev *ca, + enum bch_watermark watermark, + enum bch_data_type data_type, + struct closure *cl, + struct bch_dev_usage *usage, + struct bucket_alloc_state *s, + struct open_bucket *ob) +{ + struct printbuf buf = PRINTBUF; + + printbuf_tabstop_push(&buf, 24); + + prt_printf(&buf, "dev\t%s (%u)\n", ca->name, ca->dev_idx); + prt_printf(&buf, "watermark\t%s\n", bch2_watermarks[watermark]); + prt_printf(&buf, "data type\t%s\n", __bch2_data_types[data_type]); + prt_printf(&buf, "blocking\t%u\n", cl != NULL); + prt_printf(&buf, "free\t%llu\n", usage->d[BCH_DATA_free].buckets); + prt_printf(&buf, "avail\t%llu\n", dev_buckets_free(ca, *usage, watermark)); + prt_printf(&buf, "copygc_wait\t%lu/%lli\n", + bch2_copygc_wait_amount(c), + c->copygc_wait - atomic64_read(&c->io_clock[WRITE].now)); + prt_printf(&buf, "seen\t%llu\n", s->buckets_seen); + prt_printf(&buf, "open\t%llu\n", s->skipped_open); + prt_printf(&buf, "need journal commit\t%llu\n", s->skipped_need_journal_commit); + prt_printf(&buf, "nocow\t%llu\n", s->skipped_nocow); + prt_printf(&buf, "nouse\t%llu\n", s->skipped_nouse); + + if (!IS_ERR(ob)) { + prt_printf(&buf, "allocated\t%llu\n", ob->bucket); + trace_bucket_alloc(c, buf.buf); + } else { + prt_printf(&buf, "err\t%s\n", bch2_err_str(PTR_ERR(ob))); + trace_bucket_alloc_fail(c, buf.buf); + } + + printbuf_exit(&buf); +} + /** * bch2_bucket_alloc_trans - allocate a single bucket from a specific device * @trans: transaction object @@ -583,27 +621,14 @@ err: ob->data_type = data_type; if (!IS_ERR(ob)) - trace_and_count(c, bucket_alloc, ca, - bch2_watermarks[watermark], - ob->bucket, - usage->d[BCH_DATA_free].buckets, - avail, - bch2_copygc_wait_amount(c), - c->copygc_wait - atomic64_read(&c->io_clock[WRITE].now), - &s, - cl == NULL, - ""); + count_event(c, bucket_alloc); else if (!bch2_err_matches(PTR_ERR(ob), BCH_ERR_transaction_restart)) - trace_and_count(c, bucket_alloc_fail, ca, - bch2_watermarks[watermark], - 0, - usage->d[BCH_DATA_free].buckets, - avail, - bch2_copygc_wait_amount(c), - c->copygc_wait - atomic64_read(&c->io_clock[WRITE].now), - &s, - cl == NULL, - bch2_err_str(PTR_ERR(ob))); + count_event(c, bucket_alloc_fail); + + if (!IS_ERR(ob) + ? trace_bucket_alloc_enabled() + : trace_bucket_alloc_fail_enabled()) + trace_bucket_alloc2(c, ca, watermark, data_type, cl, usage, &s, ob); return ob; } diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h index 6aa81d1e6d36..362e1fc7ef6a 100644 --- a/fs/bcachefs/trace.h +++ b/fs/bcachefs/trace.h @@ -638,99 +638,14 @@ DEFINE_EVENT(bch_fs, gc_gens_end, /* Allocator */ -DECLARE_EVENT_CLASS(bucket_alloc, - TP_PROTO(struct bch_dev *ca, const char *alloc_reserve, - u64 bucket, - u64 free, - u64 avail, - u64 copygc_wait_amount, - s64 copygc_waiting_for, - struct bucket_alloc_state *s, - bool nonblocking, - const char *err), - TP_ARGS(ca, alloc_reserve, bucket, free, avail, - copygc_wait_amount, copygc_waiting_for, - s, nonblocking, err), - - TP_STRUCT__entry( - __field(u8, dev ) - __array(char, reserve, 16 ) - __field(u64, bucket ) - __field(u64, free ) - __field(u64, avail ) - __field(u64, copygc_wait_amount ) - __field(s64, copygc_waiting_for ) - __field(u64, seen ) - __field(u64, open ) - __field(u64, need_journal_commit ) - __field(u64, nouse ) - __field(bool, nonblocking ) - __field(u64, nocow ) - __array(char, err, 32 ) - ), - - TP_fast_assign( - __entry->dev = ca->dev_idx; - strscpy(__entry->reserve, alloc_reserve, sizeof(__entry->reserve)); - __entry->bucket = bucket; - __entry->free = free; - __entry->avail = avail; - __entry->copygc_wait_amount = copygc_wait_amount; - __entry->copygc_waiting_for = copygc_waiting_for; - __entry->seen = s->buckets_seen; - __entry->open = s->skipped_open; - __entry->need_journal_commit = s->skipped_need_journal_commit; - __entry->nouse = s->skipped_nouse; - __entry->nonblocking = nonblocking; - __entry->nocow = s->skipped_nocow; - strscpy(__entry->err, err, sizeof(__entry->err)); - ), - - TP_printk("reserve %s bucket %u:%llu free %llu avail %llu copygc_wait %llu/%lli seen %llu open %llu need_journal_commit %llu nouse %llu nocow %llu nonblocking %u err %s", - __entry->reserve, - __entry->dev, - __entry->bucket, - __entry->free, - __entry->avail, - __entry->copygc_wait_amount, - __entry->copygc_waiting_for, - __entry->seen, - __entry->open, - __entry->need_journal_commit, - __entry->nouse, - __entry->nocow, - __entry->nonblocking, - __entry->err) +DEFINE_EVENT(fs_str, bucket_alloc, + TP_PROTO(struct bch_fs *c, const char *str), + TP_ARGS(c, str) ); -DEFINE_EVENT(bucket_alloc, bucket_alloc, - TP_PROTO(struct bch_dev *ca, const char *alloc_reserve, - u64 bucket, - u64 free, - u64 avail, - u64 copygc_wait_amount, - s64 copygc_waiting_for, - struct bucket_alloc_state *s, - bool nonblocking, - const char *err), - TP_ARGS(ca, alloc_reserve, bucket, free, avail, - copygc_wait_amount, copygc_waiting_for, - s, nonblocking, err) -); - -DEFINE_EVENT(bucket_alloc, bucket_alloc_fail, - TP_PROTO(struct bch_dev *ca, const char *alloc_reserve, - u64 bucket, - u64 free, - u64 avail, - u64 copygc_wait_amount, - s64 copygc_waiting_for, - struct bucket_alloc_state *s, - bool nonblocking, - const char *err), - TP_ARGS(ca, alloc_reserve, bucket, free, avail, - copygc_wait_amount, copygc_waiting_for, - s, nonblocking, err) +DEFINE_EVENT(fs_str, bucket_alloc_fail, + TP_PROTO(struct bch_fs *c, const char *str), + TP_ARGS(c, str) ); TRACE_EVENT(discard_buckets, -- cgit v1.2.3-59-g8ed1b From c6705091342c06c963015dac07ede417d2e0ad04 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 20 Apr 2024 16:25:34 -0400 Subject: bcachefs: Allocator prefers not to expand mi.btree_allocated bitmap We now have a small bitmap in the member info section of the superblock for "regions that have btree nodes", so that if we ever have to scan for btree nodes in repair we don't have to scan the whole device(s). This tweaks the allocator to prefer allocating from regions that are already marked in this bitmap. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 68 +++++++++++++++++++++++++++++++++++------- fs/bcachefs/alloc_types.h | 7 +++++ fs/bcachefs/bcachefs.h | 2 +- 3 files changed, 66 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 9fbd0d97f68c..927a5f300b30 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -71,7 +71,7 @@ void bch2_reset_alloc_cursors(struct bch_fs *c) { rcu_read_lock(); for_each_member_device_rcu(c, ca, NULL) - ca->alloc_cursor = 0; + memset(ca->alloc_cursor, 0, sizeof(ca->alloc_cursor)); rcu_read_unlock(); } @@ -389,7 +389,8 @@ bch2_bucket_alloc_early(struct btree_trans *trans, struct bkey_s_c k, ck; struct open_bucket *ob = NULL; u64 first_bucket = max_t(u64, ca->mi.first_bucket, ca->new_fs_bucket_idx); - u64 alloc_start = max(first_bucket, READ_ONCE(ca->alloc_cursor)); + u64 *dev_alloc_cursor = &ca->alloc_cursor[s->btree_bitmap]; + u64 alloc_start = max(first_bucket, *dev_alloc_cursor); u64 alloc_cursor = alloc_start; int ret; @@ -405,8 +406,7 @@ bch2_bucket_alloc_early(struct btree_trans *trans, again: for_each_btree_key_norestart(trans, iter, BTREE_ID_alloc, POS(ca->dev_idx, alloc_cursor), BTREE_ITER_slots, k, ret) { - struct bch_alloc_v4 a_convert; - const struct bch_alloc_v4 *a; + u64 bucket = k.k->p.offset; if (bkey_ge(k.k->p, POS(ca->dev_idx, ca->mi.nbuckets))) break; @@ -415,7 +415,24 @@ again: is_superblock_bucket(ca, k.k->p.offset)) continue; - a = bch2_alloc_to_v4(k, &a_convert); + if (s->btree_bitmap != BTREE_BITMAP_ANY && + s->btree_bitmap != bch2_dev_btree_bitmap_marked_sectors(ca, + bucket_to_sector(ca, bucket), ca->mi.bucket_size)) { + if (s->btree_bitmap == BTREE_BITMAP_YES && + bucket_to_sector(ca, bucket) > 64ULL << ca->mi.btree_bitmap_shift) + break; + + bucket = sector_to_bucket(ca, + round_up(bucket_to_sector(ca, bucket) + 1, + 1ULL << ca->mi.btree_bitmap_shift)); + bch2_btree_iter_set_pos(&iter, POS(ca->dev_idx, bucket)); + s->buckets_seen++; + s->skipped_mi_btree_bitmap++; + continue; + } + + struct bch_alloc_v4 a_convert; + const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k, &a_convert); if (a->data_type != BCH_DATA_free) continue; @@ -441,7 +458,6 @@ next: bch2_trans_iter_exit(trans, &iter); alloc_cursor = iter.pos.offset; - ca->alloc_cursor = alloc_cursor; if (!ob && ret) ob = ERR_PTR(ret); @@ -451,6 +467,8 @@ next: goto again; } + *dev_alloc_cursor = alloc_cursor; + return ob; } @@ -463,7 +481,8 @@ static struct open_bucket *bch2_bucket_alloc_freelist(struct btree_trans *trans, struct btree_iter iter; struct bkey_s_c k; struct open_bucket *ob = NULL; - u64 alloc_start = max_t(u64, ca->mi.first_bucket, READ_ONCE(ca->alloc_cursor)); + u64 *dev_alloc_cursor = &ca->alloc_cursor[s->btree_bitmap]; + u64 alloc_start = max_t(u64, ca->mi.first_bucket, READ_ONCE(*dev_alloc_cursor)); u64 alloc_cursor = alloc_start; int ret; @@ -485,6 +504,26 @@ again: s->buckets_seen++; + u64 bucket = alloc_cursor & ~(~0ULL << 56); + if (s->btree_bitmap != BTREE_BITMAP_ANY && + s->btree_bitmap != bch2_dev_btree_bitmap_marked_sectors(ca, + bucket_to_sector(ca, bucket), ca->mi.bucket_size)) { + if (s->btree_bitmap == BTREE_BITMAP_YES && + bucket_to_sector(ca, bucket) > 64ULL << ca->mi.btree_bitmap_shift) + goto fail; + + bucket = sector_to_bucket(ca, + round_up(bucket_to_sector(ca, bucket) + 1, + 1ULL << ca->mi.btree_bitmap_shift)); + u64 genbits = alloc_cursor >> 56; + alloc_cursor = bucket | (genbits << 56); + + if (alloc_cursor > k.k->p.offset) + bch2_btree_iter_set_pos(&iter, POS(ca->dev_idx, alloc_cursor)); + s->skipped_mi_btree_bitmap++; + continue; + } + ob = try_alloc_bucket(trans, ca, watermark, alloc_cursor, s, k, cl); if (ob) { @@ -496,10 +535,9 @@ again: if (ob || ret) break; } +fail: bch2_trans_iter_exit(trans, &iter); - ca->alloc_cursor = alloc_cursor; - if (!ob && ret) ob = ERR_PTR(ret); @@ -508,6 +546,8 @@ again: goto again; } + *dev_alloc_cursor = alloc_cursor; + return ob; } @@ -537,6 +577,7 @@ static noinline void trace_bucket_alloc2(struct bch_fs *c, struct bch_dev *ca, prt_printf(&buf, "need journal commit\t%llu\n", s->skipped_need_journal_commit); prt_printf(&buf, "nocow\t%llu\n", s->skipped_nocow); prt_printf(&buf, "nouse\t%llu\n", s->skipped_nouse); + prt_printf(&buf, "mi_btree_bitmap\t%llu\n", s->skipped_mi_btree_bitmap); if (!IS_ERR(ob)) { prt_printf(&buf, "allocated\t%llu\n", ob->bucket); @@ -571,7 +612,9 @@ static struct open_bucket *bch2_bucket_alloc_trans(struct btree_trans *trans, struct open_bucket *ob = NULL; bool freespace = READ_ONCE(ca->mi.freespace_initialized); u64 avail; - struct bucket_alloc_state s = { 0 }; + struct bucket_alloc_state s = { + .btree_bitmap = data_type == BCH_DATA_btree, + }; bool waiting = false; again: bch2_dev_usage_read_fast(ca, usage); @@ -609,6 +652,11 @@ alloc: if (s.skipped_need_journal_commit * 2 > avail) bch2_journal_flush_async(&c->journal, NULL); + if (!ob && s.btree_bitmap != BTREE_BITMAP_ANY) { + s.btree_bitmap = BTREE_BITMAP_ANY; + goto alloc; + } + if (!ob && freespace && c->curr_recovery_pass <= BCH_RECOVERY_PASS_check_alloc_info) { freespace = false; goto alloc; diff --git a/fs/bcachefs/alloc_types.h b/fs/bcachefs/alloc_types.h index c2226e947c41..9bbb28e90b93 100644 --- a/fs/bcachefs/alloc_types.h +++ b/fs/bcachefs/alloc_types.h @@ -9,11 +9,18 @@ #include "fifo.h" struct bucket_alloc_state { + enum { + BTREE_BITMAP_NO, + BTREE_BITMAP_YES, + BTREE_BITMAP_ANY, + } btree_bitmap; + u64 buckets_seen; u64 skipped_open; u64 skipped_need_journal_commit; u64 skipped_nocow; u64 skipped_nouse; + u64 skipped_mi_btree_bitmap; }; #define BCH_WATERMARKS() \ diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index ab415f0dd14e..ab2dc2c70da2 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -587,7 +587,7 @@ struct bch_dev { /* Allocator: */ u64 new_fs_bucket_idx; - u64 alloc_cursor; + u64 alloc_cursor[3]; unsigned nr_open_buckets; unsigned nr_btree_reserve; -- cgit v1.2.3-59-g8ed1b From e11ecc6133b345a6c5919ae35920c384c6ff6f95 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 5 May 2024 09:47:53 -0400 Subject: bcachefs: Improve sysfs internal/btree_cache Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 33 ++++++++++++++++++++++++++++----- fs/bcachefs/btree_types.h | 2 ++ 2 files changed, 30 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index debdd7dc0477..2c226e0cc263 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -162,6 +162,9 @@ void bch2_btree_node_hash_remove(struct btree_cache *bc, struct btree *b) /* Cause future lookups for this node to fail: */ b->hash_val = 0; + + if (b->c.btree_id < BTREE_ID_NR) + --bc->used_by_btree[b->c.btree_id]; } int __bch2_btree_node_hash_insert(struct btree_cache *bc, struct btree *b) @@ -169,8 +172,11 @@ int __bch2_btree_node_hash_insert(struct btree_cache *bc, struct btree *b) BUG_ON(b->hash_val); b->hash_val = btree_ptr_hash_val(&b->key); - return rhashtable_lookup_insert_fast(&bc->table, &b->hash, - bch_btree_cache_params); + int ret = rhashtable_lookup_insert_fast(&bc->table, &b->hash, + bch_btree_cache_params); + if (!ret && b->c.btree_id < BTREE_ID_NR) + bc->used_by_btree[b->c.btree_id]++; + return ret; } int bch2_btree_node_hash_insert(struct btree_cache *bc, struct btree *b, @@ -1269,9 +1275,26 @@ void bch2_btree_node_to_text(struct printbuf *out, struct bch_fs *c, const struc stats.failed); } +static void prt_btree_cache_line(struct printbuf *out, const struct bch_fs *c, + const char *label, unsigned nr) +{ + prt_printf(out, "%s\t", label); + prt_human_readable_u64(out, nr * c->opts.btree_node_size); + prt_printf(out, " (%u)\n", nr); +} + void bch2_btree_cache_to_text(struct printbuf *out, const struct bch_fs *c) { - prt_printf(out, "nr nodes:\t\t%u\n", c->btree_cache.used); - prt_printf(out, "nr dirty:\t\t%u\n", atomic_read(&c->btree_cache.dirty)); - prt_printf(out, "cannibalize lock:\t%p\n", c->btree_cache.alloc_lock); + const struct btree_cache *bc = &c->btree_cache; + + if (!out->nr_tabstops) + printbuf_tabstop_push(out, 24); + + prt_btree_cache_line(out, c, "total:", bc->used); + prt_btree_cache_line(out, c, "nr dirty:", atomic_read(&bc->dirty)); + prt_printf(out, "cannibalize lock:\t%p\n", bc->alloc_lock); + prt_newline(out); + + for (unsigned i = 0; i < ARRAY_SIZE(bc->used_by_btree); i++) + prt_btree_cache_line(out, c, bch2_btree_id_str(i), bc->used_by_btree[i]); } diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index 4ff5213219a5..76364bd4347e 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -166,6 +166,8 @@ struct btree_cache { atomic_t dirty; struct shrinker *shrink; + unsigned used_by_btree[BTREE_ID_NR]; + /* * If we need to allocate memory for a new btree node and that * allocation fails, we can cannibalize another node in the btree cache -- cgit v1.2.3-59-g8ed1b From e2f48c48090dea172c0c571101041de64634dae5 Mon Sep 17 00:00:00 2001 From: Petr Vorel Date: Tue, 7 May 2024 17:37:57 +0200 Subject: bcachefs: Move BCACHEFS_STATFS_MAGIC value to UAPI magic.h Move BCACHEFS_STATFS_MAGIC value to UAPI under BCACHEFS_SUPER_MAGIC definition (use common approach for name) and reuse the definition in bcachefs_format.h BCACHEFS_STATFS_MAGIC. There are other bcachefs magic definitions: BCACHE_MAGIC, BCHFS_MAGIC, which use UUID_INIT() and are used only in libbcachefs. Therefore move only BCACHEFS_STATFS_MAGIC value, which can be used outside of libbcachefs for f_type field in struct statfs in statfs() or fstatfs(). Suggested-by: Su Yue Signed-off-by: Petr Vorel Acked-by: Brian Foster Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 3 ++- include/uapi/linux/magic.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index d5b90439e581..1bebba881d89 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -76,6 +76,7 @@ #include #include #include +#include #include "vstructs.h" #ifdef __KERNEL__ @@ -1290,7 +1291,7 @@ enum bch_compression_opts { UUID_INIT(0xc68573f6, 0x66ce, 0x90a9, \ 0xd9, 0x6a, 0x60, 0xcf, 0x80, 0x3d, 0xf7, 0xef) -#define BCACHEFS_STATFS_MAGIC 0xca451a4e +#define BCACHEFS_STATFS_MAGIC BCACHEFS_SUPER_MAGIC #define JSET_MAGIC __cpu_to_le64(0x245235c1a3625032ULL) #define BSET_MAGIC __cpu_to_le64(0x90135c78b99e07f5ULL) diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h index 1b40a968ba91..bb575f3ab45e 100644 --- a/include/uapi/linux/magic.h +++ b/include/uapi/linux/magic.h @@ -37,6 +37,7 @@ #define HOSTFS_SUPER_MAGIC 0x00c0ffee #define OVERLAYFS_SUPER_MAGIC 0x794c7630 #define FUSE_SUPER_MAGIC 0x65735546 +#define BCACHEFS_SUPER_MAGIC 0xca451a4e #define MINIX_SUPER_MAGIC 0x137F /* minix v1 fs, 14 char names */ #define MINIX_SUPER_MAGIC2 0x138F /* minix v1 fs, 30 char names */ -- cgit v1.2.3-59-g8ed1b From 7c35de4df1056a5a1fb4de042197b8f5b1033b61 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Thu, 9 May 2024 07:44:53 +0800 Subject: erofs: Zstandard compression support Add Zstandard compression as the 4th supported algorithm since it becomes more popular now and some end users have asked this for quite a while [1][2]. Each EROFS physical cluster contains only one valid standard Zstandard frame as described in [3] so that decompression can be performed on a per-pcluster basis independently. Currently, it just leverages multi-call stream decompression APIs with internal sliding window buffers. One-shot or bufferless decompression could be implemented later for even better performance if needed. [1] https://github.com/erofs/erofs-utils/issues/6 [2] https://lore.kernel.org/r/Y08h+z6CZdnS1XBm@B-P7TQMD6M-0146.lan [3] https://www.rfc-editor.org/rfc/rfc8478.txt Acked-by: Chao Yu Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20240508234453.17896-1-xiang@kernel.org --- fs/erofs/Kconfig | 15 +++ fs/erofs/Makefile | 1 + fs/erofs/compress.h | 4 + fs/erofs/decompressor.c | 7 ++ fs/erofs/decompressor_zstd.c | 279 +++++++++++++++++++++++++++++++++++++++++++ fs/erofs/erofs_fs.h | 10 ++ fs/erofs/internal.h | 8 ++ fs/erofs/super.c | 7 ++ fs/erofs/zmap.c | 3 +- 9 files changed, 333 insertions(+), 1 deletion(-) create mode 100644 fs/erofs/decompressor_zstd.c (limited to 'fs') diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig index fffd3919343e..7dcdce660cac 100644 --- a/fs/erofs/Kconfig +++ b/fs/erofs/Kconfig @@ -112,6 +112,21 @@ config EROFS_FS_ZIP_DEFLATE If unsure, say N. +config EROFS_FS_ZIP_ZSTD + bool "EROFS Zstandard compressed data support" + depends on EROFS_FS_ZIP + select ZSTD_DECOMPRESS + help + Saying Y here includes support for reading EROFS file systems + containing Zstandard compressed data. It gives better compression + ratios than the default LZ4 format, while it costs more CPU + overhead. + + Zstandard support is an experimental feature for now and so most + file systems will be readable without selecting this option. + + If unsure, say N. + config EROFS_FS_ONDEMAND bool "EROFS fscache-based on-demand read support" depends on EROFS_FS diff --git a/fs/erofs/Makefile b/fs/erofs/Makefile index 20d1ec422443..097d672e6b14 100644 --- a/fs/erofs/Makefile +++ b/fs/erofs/Makefile @@ -6,4 +6,5 @@ erofs-$(CONFIG_EROFS_FS_XATTR) += xattr.o erofs-$(CONFIG_EROFS_FS_ZIP) += decompressor.o zmap.o zdata.o zutil.o erofs-$(CONFIG_EROFS_FS_ZIP_LZMA) += decompressor_lzma.o erofs-$(CONFIG_EROFS_FS_ZIP_DEFLATE) += decompressor_deflate.o +erofs-$(CONFIG_EROFS_FS_ZIP_ZSTD) += decompressor_zstd.o erofs-$(CONFIG_EROFS_FS_ONDEMAND) += fscache.o diff --git a/fs/erofs/compress.h b/fs/erofs/compress.h index 333587ba6183..19d53c30c8af 100644 --- a/fs/erofs/compress.h +++ b/fs/erofs/compress.h @@ -90,8 +90,12 @@ int z_erofs_load_lzma_config(struct super_block *sb, struct erofs_super_block *dsb, void *data, int size); int z_erofs_load_deflate_config(struct super_block *sb, struct erofs_super_block *dsb, void *data, int size); +int z_erofs_load_zstd_config(struct super_block *sb, + struct erofs_super_block *dsb, void *data, int size); int z_erofs_lzma_decompress(struct z_erofs_decompress_req *rq, struct page **pagepool); int z_erofs_deflate_decompress(struct z_erofs_decompress_req *rq, struct page **pagepool); +int z_erofs_zstd_decompress(struct z_erofs_decompress_req *rq, + struct page **pgpl); #endif diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c index d2fe8130819e..9d85b6c11c6b 100644 --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -399,6 +399,13 @@ const struct z_erofs_decompressor erofs_decompressors[] = { .name = "deflate" }, #endif +#ifdef CONFIG_EROFS_FS_ZIP_ZSTD + [Z_EROFS_COMPRESSION_ZSTD] = { + .config = z_erofs_load_zstd_config, + .decompress = z_erofs_zstd_decompress, + .name = "zstd" + }, +#endif }; int z_erofs_parse_cfgs(struct super_block *sb, struct erofs_super_block *dsb) diff --git a/fs/erofs/decompressor_zstd.c b/fs/erofs/decompressor_zstd.c new file mode 100644 index 000000000000..63a23cac3af4 --- /dev/null +++ b/fs/erofs/decompressor_zstd.c @@ -0,0 +1,279 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +#include +#include "compress.h" + +struct z_erofs_zstd { + struct z_erofs_zstd *next; + u8 bounce[PAGE_SIZE]; + void *wksp; + unsigned int wkspsz; +}; + +static DEFINE_SPINLOCK(z_erofs_zstd_lock); +static unsigned int z_erofs_zstd_max_dictsize; +static unsigned int z_erofs_zstd_nstrms, z_erofs_zstd_avail_strms; +static struct z_erofs_zstd *z_erofs_zstd_head; +static DECLARE_WAIT_QUEUE_HEAD(z_erofs_zstd_wq); + +module_param_named(zstd_streams, z_erofs_zstd_nstrms, uint, 0444); + +static struct z_erofs_zstd *z_erofs_isolate_strms(bool all) +{ + struct z_erofs_zstd *strm; + +again: + spin_lock(&z_erofs_zstd_lock); + strm = z_erofs_zstd_head; + if (!strm) { + spin_unlock(&z_erofs_zstd_lock); + wait_event(z_erofs_zstd_wq, READ_ONCE(z_erofs_zstd_head)); + goto again; + } + z_erofs_zstd_head = all ? NULL : strm->next; + spin_unlock(&z_erofs_zstd_lock); + return strm; +} + +void z_erofs_zstd_exit(void) +{ + while (z_erofs_zstd_avail_strms) { + struct z_erofs_zstd *strm, *n; + + for (strm = z_erofs_isolate_strms(true); strm; strm = n) { + n = strm->next; + + kvfree(strm->wksp); + kfree(strm); + --z_erofs_zstd_avail_strms; + } + } +} + +int __init z_erofs_zstd_init(void) +{ + /* by default, use # of possible CPUs instead */ + if (!z_erofs_zstd_nstrms) + z_erofs_zstd_nstrms = num_possible_cpus(); + + for (; z_erofs_zstd_avail_strms < z_erofs_zstd_nstrms; + ++z_erofs_zstd_avail_strms) { + struct z_erofs_zstd *strm; + + strm = kzalloc(sizeof(*strm), GFP_KERNEL); + if (!strm) { + z_erofs_zstd_exit(); + return -ENOMEM; + } + spin_lock(&z_erofs_zstd_lock); + strm->next = z_erofs_zstd_head; + z_erofs_zstd_head = strm; + spin_unlock(&z_erofs_zstd_lock); + } + return 0; +} + +int z_erofs_load_zstd_config(struct super_block *sb, + struct erofs_super_block *dsb, void *data, int size) +{ + static DEFINE_MUTEX(zstd_resize_mutex); + struct z_erofs_zstd_cfgs *zstd = data; + unsigned int dict_size, wkspsz; + struct z_erofs_zstd *strm, *head = NULL; + void *wksp; + + if (!zstd || size < sizeof(struct z_erofs_zstd_cfgs) || zstd->format) { + erofs_err(sb, "unsupported zstd format, size=%u", size); + return -EINVAL; + } + + if (zstd->windowlog > ilog2(Z_EROFS_ZSTD_MAX_DICT_SIZE) - 10) { + erofs_err(sb, "unsupported zstd window log %u", zstd->windowlog); + return -EINVAL; + } + dict_size = 1U << (zstd->windowlog + 10); + + /* in case 2 z_erofs_load_zstd_config() race to avoid deadlock */ + mutex_lock(&zstd_resize_mutex); + if (z_erofs_zstd_max_dictsize >= dict_size) { + mutex_unlock(&zstd_resize_mutex); + return 0; + } + + /* 1. collect/isolate all streams for the following check */ + while (z_erofs_zstd_avail_strms) { + struct z_erofs_zstd *n; + + for (strm = z_erofs_isolate_strms(true); strm; strm = n) { + n = strm->next; + strm->next = head; + head = strm; + --z_erofs_zstd_avail_strms; + } + } + + /* 2. walk each isolated stream and grow max dict_size if needed */ + wkspsz = zstd_dstream_workspace_bound(dict_size); + for (strm = head; strm; strm = strm->next) { + wksp = kvmalloc(wkspsz, GFP_KERNEL); + if (!wksp) + break; + kvfree(strm->wksp); + strm->wksp = wksp; + strm->wkspsz = wkspsz; + } + + /* 3. push back all to the global list and update max dict_size */ + spin_lock(&z_erofs_zstd_lock); + DBG_BUGON(z_erofs_zstd_head); + z_erofs_zstd_head = head; + spin_unlock(&z_erofs_zstd_lock); + z_erofs_zstd_avail_strms = z_erofs_zstd_nstrms; + wake_up_all(&z_erofs_zstd_wq); + if (!strm) + z_erofs_zstd_max_dictsize = dict_size; + mutex_unlock(&zstd_resize_mutex); + return strm ? -ENOMEM : 0; +} + +int z_erofs_zstd_decompress(struct z_erofs_decompress_req *rq, + struct page **pgpl) +{ + const unsigned int nrpages_out = + PAGE_ALIGN(rq->pageofs_out + rq->outputsize) >> PAGE_SHIFT; + const unsigned int nrpages_in = + PAGE_ALIGN(rq->inputsize) >> PAGE_SHIFT; + zstd_dstream *stream; + struct super_block *sb = rq->sb; + unsigned int insz, outsz, pofs; + struct z_erofs_zstd *strm; + zstd_in_buffer in_buf = { NULL, 0, 0 }; + zstd_out_buffer out_buf = { NULL, 0, 0 }; + u8 *kin, *kout = NULL; + bool bounced = false; + int no = -1, ni = 0, j = 0, zerr, err; + + /* 1. get the exact compressed size */ + kin = kmap_local_page(*rq->in); + err = z_erofs_fixup_insize(rq, kin + rq->pageofs_in, + min_t(unsigned int, rq->inputsize, + sb->s_blocksize - rq->pageofs_in)); + if (err) { + kunmap_local(kin); + return err; + } + + /* 2. get an available ZSTD context */ + strm = z_erofs_isolate_strms(false); + + /* 3. multi-call decompress */ + insz = rq->inputsize; + outsz = rq->outputsize; + stream = zstd_init_dstream(z_erofs_zstd_max_dictsize, strm->wksp, strm->wkspsz); + if (!stream) { + err = -EIO; + goto failed_zinit; + } + + pofs = rq->pageofs_out; + in_buf.size = min_t(u32, insz, PAGE_SIZE - rq->pageofs_in); + insz -= in_buf.size; + in_buf.src = kin + rq->pageofs_in; + do { + if (out_buf.size == out_buf.pos) { + if (++no >= nrpages_out || !outsz) { + erofs_err(sb, "insufficient space for decompressed data"); + err = -EFSCORRUPTED; + break; + } + + if (kout) + kunmap_local(kout); + out_buf.size = min_t(u32, outsz, PAGE_SIZE - pofs); + outsz -= out_buf.size; + if (!rq->out[no]) { + rq->out[no] = erofs_allocpage(pgpl, rq->gfp); + if (!rq->out[no]) { + kout = NULL; + err = -ENOMEM; + break; + } + set_page_private(rq->out[no], + Z_EROFS_SHORTLIVED_PAGE); + } + kout = kmap_local_page(rq->out[no]); + out_buf.dst = kout + pofs; + out_buf.pos = 0; + pofs = 0; + } + + if (in_buf.size == in_buf.pos && insz) { + if (++ni >= nrpages_in) { + erofs_err(sb, "invalid compressed data"); + err = -EFSCORRUPTED; + break; + } + + if (kout) /* unlike kmap(), take care of the orders */ + kunmap_local(kout); + kunmap_local(kin); + in_buf.size = min_t(u32, insz, PAGE_SIZE); + insz -= in_buf.size; + kin = kmap_local_page(rq->in[ni]); + in_buf.src = kin; + in_buf.pos = 0; + bounced = false; + if (kout) { + j = (u8 *)out_buf.dst - kout; + kout = kmap_local_page(rq->out[no]); + out_buf.dst = kout + j; + } + } + + /* + * Handle overlapping: Use bounced buffer if the compressed + * data is under processing; Or use short-lived pages from the + * on-stack pagepool where pages share among the same request + * and not _all_ inplace I/O pages are needed to be doubled. + */ + if (!bounced && rq->out[no] == rq->in[ni]) { + memcpy(strm->bounce, in_buf.src, in_buf.size); + in_buf.src = strm->bounce; + bounced = true; + } + + for (j = ni + 1; j < nrpages_in; ++j) { + struct page *tmppage; + + if (rq->out[no] != rq->in[j]) + continue; + tmppage = erofs_allocpage(pgpl, rq->gfp); + if (!tmppage) { + err = -ENOMEM; + goto failed; + } + set_page_private(tmppage, Z_EROFS_SHORTLIVED_PAGE); + copy_highpage(tmppage, rq->in[j]); + rq->in[j] = tmppage; + } + zerr = zstd_decompress_stream(stream, &out_buf, &in_buf); + if (zstd_is_error(zerr) || (!zerr && outsz)) { + erofs_err(sb, "failed to decompress in[%u] out[%u]: %s", + rq->inputsize, rq->outputsize, + zerr ? zstd_get_error_name(zerr) : "unexpected end of stream"); + err = -EFSCORRUPTED; + break; + } + } while (outsz || out_buf.pos < out_buf.size); +failed: + if (kout) + kunmap_local(kout); +failed_zinit: + kunmap_local(kin); + /* 4. push back ZSTD stream context to the global list */ + spin_lock(&z_erofs_zstd_lock); + strm->next = z_erofs_zstd_head; + z_erofs_zstd_head = strm; + spin_unlock(&z_erofs_zstd_lock); + wake_up(&z_erofs_zstd_wq); + return err; +} diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h index 550baf1729d4..6c0c270c42e1 100644 --- a/fs/erofs/erofs_fs.h +++ b/fs/erofs/erofs_fs.h @@ -296,6 +296,7 @@ enum { Z_EROFS_COMPRESSION_LZ4 = 0, Z_EROFS_COMPRESSION_LZMA = 1, Z_EROFS_COMPRESSION_DEFLATE = 2, + Z_EROFS_COMPRESSION_ZSTD = 3, Z_EROFS_COMPRESSION_MAX }; #define Z_EROFS_ALL_COMPR_ALGS ((1 << Z_EROFS_COMPRESSION_MAX) - 1) @@ -322,6 +323,15 @@ struct z_erofs_deflate_cfgs { u8 reserved[5]; } __packed; +/* 6 bytes (+ length field = 8 bytes) */ +struct z_erofs_zstd_cfgs { + u8 format; + u8 windowlog; /* windowLog - ZSTD_WINDOWLOG_ABSOLUTEMIN(10) */ + u8 reserved[4]; +} __packed; + +#define Z_EROFS_ZSTD_MAX_DICT_SIZE Z_EROFS_PCLUSTER_MAX_SIZE + /* * bit 0 : COMPACTED_2B indexes (0 - off; 1 - on) * e.g. for 4k logical cluster size, 4B if compacted 2B is off; diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 53ebba952a2f..21def866a482 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -502,6 +502,14 @@ static inline int z_erofs_deflate_init(void) { return 0; } static inline int z_erofs_deflate_exit(void) { return 0; } #endif /* !CONFIG_EROFS_FS_ZIP_DEFLATE */ +#ifdef CONFIG_EROFS_FS_ZIP_ZSTD +int __init z_erofs_zstd_init(void); +void z_erofs_zstd_exit(void); +#else +static inline int z_erofs_zstd_init(void) { return 0; } +static inline int z_erofs_zstd_exit(void) { return 0; } +#endif /* !CONFIG_EROFS_FS_ZIP_ZSTD */ + #ifdef CONFIG_EROFS_FS_ONDEMAND int erofs_fscache_register_fs(struct super_block *sb); void erofs_fscache_unregister_fs(struct super_block *sb); diff --git a/fs/erofs/super.c b/fs/erofs/super.c index e3438f1a7bac..044c79229a78 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -859,6 +859,10 @@ static int __init erofs_module_init(void) if (err) goto deflate_err; + err = z_erofs_zstd_init(); + if (err) + goto zstd_err; + err = z_erofs_gbuf_init(); if (err) goto gbuf_err; @@ -884,6 +888,8 @@ sysfs_err: zip_err: z_erofs_gbuf_exit(); gbuf_err: + z_erofs_zstd_exit(); +zstd_err: z_erofs_deflate_exit(); deflate_err: z_erofs_lzma_exit(); @@ -903,6 +909,7 @@ static void __exit erofs_module_exit(void) erofs_exit_sysfs(); z_erofs_exit_zip_subsystem(); + z_erofs_zstd_exit(); z_erofs_deflate_exit(); z_erofs_lzma_exit(); erofs_exit_shrinker(); diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index 26637a60eba5..0a2454d8bcc1 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -550,7 +550,8 @@ static int z_erofs_do_map_blocks(struct inode *inode, if ((flags & EROFS_GET_BLOCKS_FIEMAP) || ((flags & EROFS_GET_BLOCKS_READMORE) && (map->m_algorithmformat == Z_EROFS_COMPRESSION_LZMA || - map->m_algorithmformat == Z_EROFS_COMPRESSION_DEFLATE) && + map->m_algorithmformat == Z_EROFS_COMPRESSION_DEFLATE || + map->m_algorithmformat == Z_EROFS_COMPRESSION_ZSTD) && map->m_llen >= i_blocksize(inode))) { err = z_erofs_get_extent_decompressedlen(&m); if (!err) -- cgit v1.2.3-59-g8ed1b From aa4074e8fec4d2e686daee627fcafb3503efe365 Mon Sep 17 00:00:00 2001 From: Wu Bo Date: Sun, 28 Apr 2024 21:51:42 -0600 Subject: f2fs: fix block migration when section is not aligned to pow2 As for zoned-UFS, f2fs section size is forced to zone size. And zone size may not aligned to pow2. Fixes: 859fca6b706e ("f2fs: swap: support migrating swapfile in aligned write mode") Signed-off-by: Liao Yuanhong Signed-off-by: Wu Bo Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index a990236f137e..bdbd4195cadc 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -3917,15 +3917,14 @@ static int check_swap_activate(struct swap_info_struct *sis, struct address_space *mapping = swap_file->f_mapping; struct inode *inode = mapping->host; struct f2fs_sb_info *sbi = F2FS_I_SB(inode); - sector_t cur_lblock; - sector_t last_lblock; - sector_t pblock; - sector_t lowest_pblock = -1; - sector_t highest_pblock = 0; + block_t cur_lblock; + block_t last_lblock; + block_t pblock; + block_t lowest_pblock = -1; + block_t highest_pblock = 0; int nr_extents = 0; - unsigned long nr_pblocks; + unsigned int nr_pblocks; unsigned int blks_per_sec = BLKS_PER_SEC(sbi); - unsigned int sec_blks_mask = BLKS_PER_SEC(sbi) - 1; unsigned int not_aligned = 0; int ret = 0; @@ -3963,8 +3962,8 @@ retry: pblock = map.m_pblk; nr_pblocks = map.m_len; - if ((pblock - SM_I(sbi)->main_blkaddr) & sec_blks_mask || - nr_pblocks & sec_blks_mask || + if ((pblock - SM_I(sbi)->main_blkaddr) % blks_per_sec || + nr_pblocks % blks_per_sec || !f2fs_valid_pinned_area(sbi, pblock)) { bool last_extent = false; -- cgit v1.2.3-59-g8ed1b From a78118406d52dde495311c0c4917613868b53169 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 6 May 2024 18:47:42 +0800 Subject: f2fs: use f2fs_{err,info}_ratelimited() for cleanup Commit b1c9d3f833ba ("f2fs: support printk_ratelimited() in f2fs_printk()") missed some cases, cover all remains for cleanup. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/compress.c | 54 ++++++++++++++++++++++++------------------------------ fs/f2fs/segment.c | 5 ++--- 2 files changed, 26 insertions(+), 33 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index b12d3a49bfda..1ef82a546391 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -198,8 +198,8 @@ static int lzo_compress_pages(struct compress_ctx *cc) ret = lzo1x_1_compress(cc->rbuf, cc->rlen, cc->cbuf->cdata, &cc->clen, cc->private); if (ret != LZO_E_OK) { - printk_ratelimited("%sF2FS-fs (%s): lzo compress failed, ret:%d\n", - KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id, ret); + f2fs_err_ratelimited(F2FS_I_SB(cc->inode), + "lzo compress failed, ret:%d", ret); return -EIO; } return 0; @@ -212,17 +212,15 @@ static int lzo_decompress_pages(struct decompress_io_ctx *dic) ret = lzo1x_decompress_safe(dic->cbuf->cdata, dic->clen, dic->rbuf, &dic->rlen); if (ret != LZO_E_OK) { - printk_ratelimited("%sF2FS-fs (%s): lzo decompress failed, ret:%d\n", - KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id, ret); + f2fs_err_ratelimited(F2FS_I_SB(dic->inode), + "lzo decompress failed, ret:%d", ret); return -EIO; } if (dic->rlen != PAGE_SIZE << dic->log_cluster_size) { - printk_ratelimited("%sF2FS-fs (%s): lzo invalid rlen:%zu, " - "expected:%lu\n", KERN_ERR, - F2FS_I_SB(dic->inode)->sb->s_id, - dic->rlen, - PAGE_SIZE << dic->log_cluster_size); + f2fs_err_ratelimited(F2FS_I_SB(dic->inode), + "lzo invalid rlen:%zu, expected:%lu", + dic->rlen, PAGE_SIZE << dic->log_cluster_size); return -EIO; } return 0; @@ -294,16 +292,15 @@ static int lz4_decompress_pages(struct decompress_io_ctx *dic) ret = LZ4_decompress_safe(dic->cbuf->cdata, dic->rbuf, dic->clen, dic->rlen); if (ret < 0) { - printk_ratelimited("%sF2FS-fs (%s): lz4 decompress failed, ret:%d\n", - KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id, ret); + f2fs_err_ratelimited(F2FS_I_SB(dic->inode), + "lz4 decompress failed, ret:%d", ret); return -EIO; } if (ret != PAGE_SIZE << dic->log_cluster_size) { - printk_ratelimited("%sF2FS-fs (%s): lz4 invalid ret:%d, " - "expected:%lu\n", KERN_ERR, - F2FS_I_SB(dic->inode)->sb->s_id, ret, - PAGE_SIZE << dic->log_cluster_size); + f2fs_err_ratelimited(F2FS_I_SB(dic->inode), + "lz4 invalid ret:%d, expected:%lu", + ret, PAGE_SIZE << dic->log_cluster_size); return -EIO; } return 0; @@ -350,9 +347,8 @@ static int zstd_init_compress_ctx(struct compress_ctx *cc) stream = zstd_init_cstream(¶ms, 0, workspace, workspace_size); if (!stream) { - printk_ratelimited("%sF2FS-fs (%s): %s zstd_init_cstream failed\n", - KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id, - __func__); + f2fs_err_ratelimited(F2FS_I_SB(cc->inode), + "%s zstd_init_cstream failed", __func__); kvfree(workspace); return -EIO; } @@ -390,16 +386,16 @@ static int zstd_compress_pages(struct compress_ctx *cc) ret = zstd_compress_stream(stream, &outbuf, &inbuf); if (zstd_is_error(ret)) { - printk_ratelimited("%sF2FS-fs (%s): %s zstd_compress_stream failed, ret: %d\n", - KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id, + f2fs_err_ratelimited(F2FS_I_SB(cc->inode), + "%s zstd_compress_stream failed, ret: %d", __func__, zstd_get_error_code(ret)); return -EIO; } ret = zstd_end_stream(stream, &outbuf); if (zstd_is_error(ret)) { - printk_ratelimited("%sF2FS-fs (%s): %s zstd_end_stream returned %d\n", - KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id, + f2fs_err_ratelimited(F2FS_I_SB(cc->inode), + "%s zstd_end_stream returned %d", __func__, zstd_get_error_code(ret)); return -EIO; } @@ -432,9 +428,8 @@ static int zstd_init_decompress_ctx(struct decompress_io_ctx *dic) stream = zstd_init_dstream(max_window_size, workspace, workspace_size); if (!stream) { - printk_ratelimited("%sF2FS-fs (%s): %s zstd_init_dstream failed\n", - KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id, - __func__); + f2fs_err_ratelimited(F2FS_I_SB(dic->inode), + "%s zstd_init_dstream failed", __func__); kvfree(workspace); return -EIO; } @@ -469,16 +464,15 @@ static int zstd_decompress_pages(struct decompress_io_ctx *dic) ret = zstd_decompress_stream(stream, &outbuf, &inbuf); if (zstd_is_error(ret)) { - printk_ratelimited("%sF2FS-fs (%s): %s zstd_decompress_stream failed, ret: %d\n", - KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id, + f2fs_err_ratelimited(F2FS_I_SB(dic->inode), + "%s zstd_decompress_stream failed, ret: %d", __func__, zstd_get_error_code(ret)); return -EIO; } if (dic->rlen != outbuf.pos) { - printk_ratelimited("%sF2FS-fs (%s): %s ZSTD invalid rlen:%zu, " - "expected:%lu\n", KERN_ERR, - F2FS_I_SB(dic->inode)->sb->s_id, + f2fs_err_ratelimited(F2FS_I_SB(dic->inode), + "%s ZSTD invalid rlen:%zu, expected:%lu", __func__, dic->rlen, PAGE_SIZE << dic->log_cluster_size); return -EIO; diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 8ddd4f1c665a..ab3e9a6e9c96 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1109,9 +1109,8 @@ static void __remove_discard_cmd(struct f2fs_sb_info *sbi, dc->error = 0; if (dc->error) - printk_ratelimited( - "%sF2FS-fs (%s): Issue discard(%u, %u, %u) failed, ret: %d", - KERN_INFO, sbi->sb->s_id, + f2fs_info_ratelimited(sbi, + "Issue discard(%u, %u, %u) failed, ret: %d", dc->di.lstart, dc->di.start, dc->di.len, dc->error); __detach_discard_cmd(dcc, dc); } -- cgit v1.2.3-59-g8ed1b From 968c4f72b23c0c8f1e94e942eab89b8c5a3022e7 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 6 May 2024 18:45:37 +0800 Subject: f2fs: remove unused GC_FAILURE_PIN After commit 3db1de0e582c ("f2fs: change the current atomic write way"), we removed all GC_FAILURE_ATOMIC usage, let's change i_gc_failures[] array to i_pin_failure for cleanup. Meanwhile, let's define i_current_depth and i_gc_failures as union variable due to they won't be valid at the same time. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 14 +++++--------- fs/f2fs/file.c | 12 +++++------- fs/f2fs/inode.c | 6 ++---- fs/f2fs/recovery.c | 3 +-- 4 files changed, 13 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 3a95c8ccb9e0..a4ebf3e91b68 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -765,11 +765,6 @@ enum { #define DEF_DIR_LEVEL 0 -enum { - GC_FAILURE_PIN, - MAX_GC_FAILURE -}; - /* used for f2fs_inode_info->flags */ enum { FI_NEW_INODE, /* indicate newly allocated inode */ @@ -816,9 +811,10 @@ struct f2fs_inode_info { unsigned long i_flags; /* keep an inode flags for ioctl */ unsigned char i_advise; /* use to give file attribute hints */ unsigned char i_dir_level; /* use for dentry level for large dir */ - unsigned int i_current_depth; /* only for directory depth */ - /* for gc failure statistic */ - unsigned int i_gc_failures[MAX_GC_FAILURE]; + union { + unsigned int i_current_depth; /* only for directory depth */ + unsigned int i_gc_failures; /* for gc failure statistic */ + }; unsigned int i_pino; /* parent inode number */ umode_t i_acl_mode; /* keep file acl mode temporarily */ @@ -3133,7 +3129,7 @@ static inline void f2fs_i_depth_write(struct inode *inode, unsigned int depth) static inline void f2fs_i_gc_failures_write(struct inode *inode, unsigned int count) { - F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN] = count; + F2FS_I(inode)->i_gc_failures = count; f2fs_mark_inode_dirty_sync(inode, true); } diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index f1a6b4da5746..92aaa6986f99 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -3217,13 +3217,11 @@ int f2fs_pin_file_control(struct inode *inode, bool inc) /* Use i_gc_failures for normal file as a risk signal. */ if (inc) - f2fs_i_gc_failures_write(inode, - fi->i_gc_failures[GC_FAILURE_PIN] + 1); + f2fs_i_gc_failures_write(inode, fi->i_gc_failures + 1); - if (fi->i_gc_failures[GC_FAILURE_PIN] > sbi->gc_pin_file_threshold) { + if (fi->i_gc_failures > sbi->gc_pin_file_threshold) { f2fs_warn(sbi, "%s: Enable GC = ino %lx after %x GC trials", - __func__, inode->i_ino, - fi->i_gc_failures[GC_FAILURE_PIN]); + __func__, inode->i_ino, fi->i_gc_failures); clear_inode_flag(inode, FI_PIN_FILE); return -EAGAIN; } @@ -3287,7 +3285,7 @@ static int f2fs_ioc_set_pin_file(struct file *filp, unsigned long arg) } set_inode_flag(inode, FI_PIN_FILE); - ret = F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN]; + ret = F2FS_I(inode)->i_gc_failures; done: f2fs_update_time(sbi, REQ_TIME); out: @@ -3302,7 +3300,7 @@ static int f2fs_ioc_get_pin_file(struct file *filp, unsigned long arg) __u32 pin = 0; if (is_inode_flag_set(inode, FI_PIN_FILE)) - pin = F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN]; + pin = F2FS_I(inode)->i_gc_failures; return put_user(pin, (u32 __user *)arg); } diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 7968b14d49f4..005dde72aff3 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -415,8 +415,7 @@ static int do_read_inode(struct inode *inode) if (S_ISDIR(inode->i_mode)) fi->i_current_depth = le32_to_cpu(ri->i_current_depth); else if (S_ISREG(inode->i_mode)) - fi->i_gc_failures[GC_FAILURE_PIN] = - le16_to_cpu(ri->i_gc_failures); + fi->i_gc_failures = le16_to_cpu(ri->i_gc_failures); fi->i_xattr_nid = le32_to_cpu(ri->i_xattr_nid); fi->i_flags = le32_to_cpu(ri->i_flags); if (S_ISREG(inode->i_mode)) @@ -686,8 +685,7 @@ void f2fs_update_inode(struct inode *inode, struct page *node_page) ri->i_current_depth = cpu_to_le32(F2FS_I(inode)->i_current_depth); else if (S_ISREG(inode->i_mode)) - ri->i_gc_failures = - cpu_to_le16(F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN]); + ri->i_gc_failures = cpu_to_le16(F2FS_I(inode)->i_gc_failures); ri->i_xattr_nid = cpu_to_le32(F2FS_I(inode)->i_xattr_nid); ri->i_flags = cpu_to_le32(F2FS_I(inode)->i_flags); ri->i_pino = cpu_to_le32(F2FS_I(inode)->i_pino); diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c index e7bf15b8240a..496aee53c38a 100644 --- a/fs/f2fs/recovery.c +++ b/fs/f2fs/recovery.c @@ -330,8 +330,7 @@ static int recover_inode(struct inode *inode, struct page *page) F2FS_I(inode)->i_advise = raw->i_advise; F2FS_I(inode)->i_flags = le32_to_cpu(raw->i_flags); f2fs_set_inode_flags(inode); - F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN] = - le16_to_cpu(raw->i_gc_failures); + F2FS_I(inode)->i_gc_failures = le16_to_cpu(raw->i_gc_failures); recover_inline_flags(inode, raw); -- cgit v1.2.3-59-g8ed1b From c521a6ab4ad75fe0755f4cd923a84e1289153aa2 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 6 May 2024 18:45:38 +0800 Subject: f2fs: fix to limit gc_pin_file_threshold type of f2fs_inode.i_gc_failures, f2fs_inode_info.i_gc_failures, and f2fs_sb_info.gc_pin_file_threshold is __le16, unsigned int, and u64, so it will cause truncation during comparison and persistence. Unifying variable of these three variables to unsigned short, and add an upper boundary limitation for gc_pin_file_threshold. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- Documentation/ABI/testing/sysfs-fs-f2fs | 2 +- fs/f2fs/f2fs.h | 4 ++-- fs/f2fs/file.c | 11 ++++++----- fs/f2fs/gc.h | 1 + fs/f2fs/sysfs.c | 7 +++++++ 5 files changed, 17 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs index 1a4d83953379..cad6c3dc1f9c 100644 --- a/Documentation/ABI/testing/sysfs-fs-f2fs +++ b/Documentation/ABI/testing/sysfs-fs-f2fs @@ -331,7 +331,7 @@ Date: January 2018 Contact: Jaegeuk Kim Description: This indicates how many GC can be failed for the pinned file. If it exceeds this, F2FS doesn't guarantee its pinning - state. 2048 trials is set by default. + state. 2048 trials is set by default, and 65535 as maximum. What: /sys/fs/f2fs//extension_list Date: February 2018 diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index a4ebf3e91b68..1352da2dfaa3 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -813,7 +813,7 @@ struct f2fs_inode_info { unsigned char i_dir_level; /* use for dentry level for large dir */ union { unsigned int i_current_depth; /* only for directory depth */ - unsigned int i_gc_failures; /* for gc failure statistic */ + unsigned short i_gc_failures; /* for gc failure statistic */ }; unsigned int i_pino; /* parent inode number */ umode_t i_acl_mode; /* keep file acl mode temporarily */ @@ -1673,7 +1673,7 @@ struct f2fs_sb_info { unsigned long long skipped_gc_rwsem; /* FG_GC only */ /* threshold for gc trials on pinned files */ - u64 gc_pin_file_threshold; + unsigned short gc_pin_file_threshold; struct f2fs_rwsem pin_sem; /* maximum # of trials to find a victim segment for SSR and GC */ diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 92aaa6986f99..72ce1a522fb2 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -3215,16 +3215,17 @@ int f2fs_pin_file_control(struct inode *inode, bool inc) struct f2fs_inode_info *fi = F2FS_I(inode); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); - /* Use i_gc_failures for normal file as a risk signal. */ - if (inc) - f2fs_i_gc_failures_write(inode, fi->i_gc_failures + 1); - - if (fi->i_gc_failures > sbi->gc_pin_file_threshold) { + if (fi->i_gc_failures >= sbi->gc_pin_file_threshold) { f2fs_warn(sbi, "%s: Enable GC = ino %lx after %x GC trials", __func__, inode->i_ino, fi->i_gc_failures); clear_inode_flag(inode, FI_PIN_FILE); return -EAGAIN; } + + /* Use i_gc_failures for normal file as a risk signal. */ + if (inc) + f2fs_i_gc_failures_write(inode, fi->i_gc_failures + 1); + return 0; } diff --git a/fs/f2fs/gc.h b/fs/f2fs/gc.h index 9c0d06c4d19a..a8ea3301b815 100644 --- a/fs/f2fs/gc.h +++ b/fs/f2fs/gc.h @@ -26,6 +26,7 @@ #define LIMIT_FREE_BLOCK 40 /* percentage over invalid + free space */ #define DEF_GC_FAILED_PINNED_FILES 2048 +#define MAX_GC_FAILED_PINNED_FILES USHRT_MAX /* Search max. number of dirty segments to select a victim segment */ #define DEF_MAX_VICTIM_SEARCH 4096 /* covers 8GB */ diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index a568ce96cf56..531fc5800109 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -675,6 +675,13 @@ out: return count; } + if (!strcmp(a->attr.name, "gc_pin_file_threshold")) { + if (t > MAX_GC_FAILED_PINNED_FILES) + return -EINVAL; + sbi->gc_pin_file_threshold = t; + return count; + } + if (!strcmp(a->attr.name, "gc_reclaimed_segments")) { if (t != 0) return -EINVAL; -- cgit v1.2.3-59-g8ed1b From 4ed886b187f47447ad559619c48c086f432d2b77 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 7 May 2024 11:38:47 +0800 Subject: f2fs: check validation of fault attrs in f2fs_build_fault_attr() - It missed to check validation of fault attrs in parse_options(), let's fix to add check condition in f2fs_build_fault_attr(). - Use f2fs_build_fault_attr() in __sbi_store() to clean up code. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 12 ++++++++---- fs/f2fs/super.c | 27 ++++++++++++++++++++------- fs/f2fs/sysfs.c | 14 ++++++++++---- 3 files changed, 38 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 1352da2dfaa3..30058e16a5d0 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -72,7 +72,7 @@ enum { struct f2fs_fault_info { atomic_t inject_ops; - unsigned int inject_rate; + int inject_rate; unsigned int inject_type; }; @@ -4598,10 +4598,14 @@ static inline bool f2fs_need_verity(const struct inode *inode, pgoff_t idx) } #ifdef CONFIG_F2FS_FAULT_INJECTION -extern void f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned int rate, - unsigned int type); +extern int f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned long rate, + unsigned long type); #else -#define f2fs_build_fault_attr(sbi, rate, type) do { } while (0) +static int f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned long rate, + unsigned long type) +{ + return 0; +} #endif static inline bool is_journalled_quota(struct f2fs_sb_info *sbi) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index ef673f853366..daf2c4dbe150 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -66,21 +66,31 @@ const char *f2fs_fault_name[FAULT_MAX] = { [FAULT_NO_SEGMENT] = "no free segment", }; -void f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned int rate, - unsigned int type) +int f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned long rate, + unsigned long type) { struct f2fs_fault_info *ffi = &F2FS_OPTION(sbi).fault_info; if (rate) { + if (rate > INT_MAX) + return -EINVAL; atomic_set(&ffi->inject_ops, 0); - ffi->inject_rate = rate; + ffi->inject_rate = (int)rate; } - if (type) - ffi->inject_type = type; + if (type) { + if (type >= BIT(FAULT_MAX)) + return -EINVAL; + ffi->inject_type = (unsigned int)type; + } if (!rate && !type) memset(ffi, 0, sizeof(struct f2fs_fault_info)); + else + f2fs_info(sbi, + "build fault injection attr: rate: %lu, type: 0x%lx", + rate, type); + return 0; } #endif @@ -886,14 +896,17 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount) case Opt_fault_injection: if (args->from && match_int(args, &arg)) return -EINVAL; - f2fs_build_fault_attr(sbi, arg, F2FS_ALL_FAULT_TYPE); + if (f2fs_build_fault_attr(sbi, arg, + F2FS_ALL_FAULT_TYPE)) + return -EINVAL; set_opt(sbi, FAULT_INJECTION); break; case Opt_fault_type: if (args->from && match_int(args, &arg)) return -EINVAL; - f2fs_build_fault_attr(sbi, 0, arg); + if (f2fs_build_fault_attr(sbi, 0, arg)) + return -EINVAL; set_opt(sbi, FAULT_INJECTION); break; #else diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 531fc5800109..09d3ecfaa4f1 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -484,10 +484,16 @@ out: if (ret < 0) return ret; #ifdef CONFIG_F2FS_FAULT_INJECTION - if (a->struct_type == FAULT_INFO_TYPE && t >= BIT(FAULT_MAX)) - return -EINVAL; - if (a->struct_type == FAULT_INFO_RATE && t >= UINT_MAX) - return -EINVAL; + if (a->struct_type == FAULT_INFO_TYPE) { + if (f2fs_build_fault_attr(sbi, 0, t)) + return -EINVAL; + return count; + } + if (a->struct_type == FAULT_INFO_RATE) { + if (f2fs_build_fault_attr(sbi, t, 0)) + return -EINVAL; + return count; + } #endif if (a->struct_type == RESERVED_BLOCKS) { spin_lock(&sbi->stat_lock); -- cgit v1.2.3-59-g8ed1b From ea4fd933ab4310822e244af28d22ff63785dea0e Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Sat, 20 Apr 2024 03:50:05 +0100 Subject: ext4: remove calls to to set/clear the folio error flag Nobody checks this flag on ext4 folios, stop setting and clearing it. Cc: Theodore Ts'o Cc: Andreas Dilger Cc: linux-ext4@vger.kernel.org Signed-off-by: Matthew Wilcox (Oracle) Link: https://lore.kernel.org/r/20240420025029.2166544-11-willy@infradead.org Signed-off-by: Theodore Ts'o --- fs/ext4/move_extent.c | 4 +--- fs/ext4/page-io.c | 3 --- fs/ext4/readpage.c | 1 - 3 files changed, 1 insertion(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c index 7cd4afa4de1d..204f53b23622 100644 --- a/fs/ext4/move_extent.c +++ b/fs/ext4/move_extent.c @@ -199,10 +199,8 @@ mext_page_mkuptodate(struct folio *folio, unsigned from, unsigned to) continue; if (!buffer_mapped(bh)) { err = ext4_get_block(inode, block, bh, 0); - if (err) { - folio_set_error(folio); + if (err) return err; - } if (!buffer_mapped(bh)) { folio_zero_range(folio, block_start, blocksize); set_buffer_uptodate(bh); diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 312bc6813357..ad5543866d21 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -117,7 +117,6 @@ static void ext4_finish_bio(struct bio *bio) if (bio->bi_status) { int err = blk_status_to_errno(bio->bi_status); - folio_set_error(folio); mapping_set_error(folio->mapping, err); } bh = head = folio_buffers(folio); @@ -441,8 +440,6 @@ int ext4_bio_write_folio(struct ext4_io_submit *io, struct folio *folio, BUG_ON(!folio_test_locked(folio)); BUG_ON(folio_test_writeback(folio)); - folio_clear_error(folio); - /* * Comments copied from block_write_full_folio: * diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c index 21e8f0aebb3c..8494492582ab 100644 --- a/fs/ext4/readpage.c +++ b/fs/ext4/readpage.c @@ -289,7 +289,6 @@ int ext4_mpage_readpages(struct inode *inode, if (ext4_map_blocks(NULL, inode, &map, 0) < 0) { set_error_page: - folio_set_error(folio); folio_zero_segment(folio, 0, folio_size(folio)); folio_unlock(folio); -- cgit v1.2.3-59-g8ed1b From 8d915bbf39266bb66082c1e4980e123883f19830 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 7 May 2024 09:37:14 -0400 Subject: NFSD: Force all NFSv4.2 COPY requests to be synchronous We've discovered that delivering a CB_OFFLOAD operation can be unreliable in some pretty unremarkable situations. Examples include: - The server dropped the connection because it lost a forechannel NFSv4 request and wishes to force the client to retransmit - The GSS sequence number window under-flowed - A network partition occurred When that happens, all pending callback operations, including CB_OFFLOAD, are lost. NFSD does not retransmit them. Moreover, the Linux NFS client does not yet support sending an OFFLOAD_STATUS operation to probe whether an asynchronous COPY operation has finished. Thus, on Linux NFS clients, when a CB_OFFLOAD is lost, asynchronous COPY can hang until manually interrupted. I've tried a couple of remedies, but so far the side-effects are worse than the disease and they have had to be reverted. So temporarily force COPY operations to be synchronous so that the use of CB_OFFLOAD is avoided entirely. This is a fix that can easily be backported to LTS kernels. I am working on client patches that introduce an implementation of OFFLOAD_STATUS. Note that NFSD arbitrarily limits the size of a copy_file_range to 4MB to avoid indefinitely blocking an nfsd thread. A short COPY result is returned in that case, and the client can present a fresh COPY request for the remainder. Signed-off-by: Chuck Lever --- fs/nfsd/nfs4proc.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'fs') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index ea3cc3e870a7..46bd20fe5c0f 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1807,6 +1807,13 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, __be32 status; struct nfsd4_copy *async_copy = NULL; + /* + * Currently, async COPY is not reliable. Force all COPY + * requests to be synchronous to avoid client application + * hangs waiting for COPY completion. + */ + nfsd4_copy_set_sync(copy, true); + copy->cp_clp = cstate->clp; if (nfsd4_ssc_is_inter(copy)) { trace_nfsd_copy_inter(copy); -- cgit v1.2.3-59-g8ed1b From b4b4fda34e535756f9e774fb2d09c4537b7dfd1c Mon Sep 17 00:00:00 2001 From: Baokun Li Date: Tue, 2 Jan 2024 21:37:30 +0800 Subject: ext4: fix uninitialized ratelimit_state->lock access in __ext4_fill_super() In the following concurrency we will access the uninitialized rs->lock: ext4_fill_super ext4_register_sysfs // sysfs registered msg_ratelimit_interval_ms // Other processes modify rs->interval to // non-zero via msg_ratelimit_interval_ms ext4_orphan_cleanup ext4_msg(sb, KERN_INFO, "Errors on filesystem, " __ext4_msg ___ratelimit(&(EXT4_SB(sb)->s_msg_ratelimit_state) if (!rs->interval) // do nothing if interval is 0 return 1; raw_spin_trylock_irqsave(&rs->lock, flags) raw_spin_trylock(lock) _raw_spin_trylock __raw_spin_trylock spin_acquire(&lock->dep_map, 0, 1, _RET_IP_) lock_acquire __lock_acquire register_lock_class assign_lock_key dump_stack(); ratelimit_state_init(&sbi->s_msg_ratelimit_state, 5 * HZ, 10); raw_spin_lock_init(&rs->lock); // init rs->lock here and get the following dump_stack: ========================================================= INFO: trying to register non-static key. The code is fine but needs lockdep annotation, or maybe you didn't initialize this object before use? turning off the locking correctness validator. CPU: 12 PID: 753 Comm: mount Tainted: G E 6.7.0-rc6-next-20231222 #504 [...] Call Trace: dump_stack_lvl+0xc5/0x170 dump_stack+0x18/0x30 register_lock_class+0x740/0x7c0 __lock_acquire+0x69/0x13a0 lock_acquire+0x120/0x450 _raw_spin_trylock+0x98/0xd0 ___ratelimit+0xf6/0x220 __ext4_msg+0x7f/0x160 [ext4] ext4_orphan_cleanup+0x665/0x740 [ext4] __ext4_fill_super+0x21ea/0x2b10 [ext4] ext4_fill_super+0x14d/0x360 [ext4] [...] ========================================================= Normally interval is 0 until s_msg_ratelimit_state is initialized, so ___ratelimit() does nothing. But registering sysfs precedes initializing rs->lock, so it is possible to change rs->interval to a non-zero value via the msg_ratelimit_interval_ms interface of sysfs while rs->lock is uninitialized, and then a call to ext4_msg triggers the problem by accessing an uninitialized rs->lock. Therefore register sysfs after all initializations are complete to avoid such problems. Signed-off-by: Baokun Li Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20240102133730.1098120-1-libaokun1@huawei.com Signed-off-by: Theodore Ts'o --- fs/ext4/super.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f9b2b18374f3..f9a4a4e89dac 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -5550,19 +5550,15 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) if (err) goto failed_mount6; - err = ext4_register_sysfs(sb); - if (err) - goto failed_mount7; - err = ext4_init_orphan_info(sb); if (err) - goto failed_mount8; + goto failed_mount7; #ifdef CONFIG_QUOTA /* Enable quota usage during mount. */ if (ext4_has_feature_quota(sb) && !sb_rdonly(sb)) { err = ext4_enable_quotas(sb); if (err) - goto failed_mount9; + goto failed_mount8; } #endif /* CONFIG_QUOTA */ @@ -5588,7 +5584,7 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) ext4_msg(sb, KERN_INFO, "recovery complete"); err = ext4_mark_recovery_complete(sb, es); if (err) - goto failed_mount10; + goto failed_mount9; } if (test_opt(sb, DISCARD) && !bdev_max_discard_sectors(sb->s_bdev)) @@ -5605,15 +5601,17 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) atomic_set(&sbi->s_warning_count, 0); atomic_set(&sbi->s_msg_count, 0); + /* Register sysfs after all initializations are complete. */ + err = ext4_register_sysfs(sb); + if (err) + goto failed_mount9; + return 0; -failed_mount10: +failed_mount9: ext4_quotas_off(sb, EXT4_MAXQUOTAS); -failed_mount9: __maybe_unused +failed_mount8: __maybe_unused ext4_release_orphan_info(sb); -failed_mount8: - ext4_unregister_sysfs(sb); - kobject_put(&sbi->s_kobj); failed_mount7: ext4_unregister_li_request(sb); failed_mount6: -- cgit v1.2.3-59-g8ed1b From 078760d950016f5982751f5512e69f26ad8feb31 Mon Sep 17 00:00:00 2001 From: Ye Bin Date: Sun, 7 Apr 2024 14:53:54 +0800 Subject: jbd2: use shrink_type type instead of bool type for __jbd2_journal_clean_checkpoint_list() "enum shrink_type" can clearly express the meaning of the parameter of __jbd2_journal_clean_checkpoint_list(), and there is no need to use the bool type. Signed-off-by: Ye Bin Reviewed-by: Jan Kara Reviewed-by: Zhang Yi Link: https://lore.kernel.org/r/20240407065355.1528580-2-yebin10@huawei.com Signed-off-by: Theodore Ts'o --- fs/jbd2/checkpoint.c | 16 +++++++++------- fs/jbd2/commit.c | 2 +- include/linux/jbd2.h | 4 +++- 3 files changed, 13 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 1c97e64c4784..80c0ab98bc63 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -337,8 +337,6 @@ int jbd2_cleanup_journal_tail(journal_t *journal) /* Checkpoint list management */ -enum shrink_type {SHRINK_DESTROY, SHRINK_BUSY_STOP, SHRINK_BUSY_SKIP}; - /* * journal_shrink_one_cp_list * @@ -472,21 +470,25 @@ out: * journal_clean_checkpoint_list * * Find all the written-back checkpoint buffers in the journal and release them. - * If 'destroy' is set, release all buffers unconditionally. + * If 'type' is SHRINK_DESTROY, release all buffers unconditionally. If 'type' + * is SHRINK_BUSY_STOP, will stop release buffers if encounters a busy buffer. + * To avoid wasting CPU cycles scanning the buffer list in some cases, don't + * pass SHRINK_BUSY_SKIP 'type' for this function. * * Called with j_list_lock held. */ -void __jbd2_journal_clean_checkpoint_list(journal_t *journal, bool destroy) +void __jbd2_journal_clean_checkpoint_list(journal_t *journal, + enum shrink_type type) { transaction_t *transaction, *last_transaction, *next_transaction; - enum shrink_type type; bool released; + WARN_ON_ONCE(type == SHRINK_BUSY_SKIP); + transaction = journal->j_checkpoint_transactions; if (!transaction) return; - type = destroy ? SHRINK_DESTROY : SHRINK_BUSY_STOP; last_transaction = transaction->t_cpprev; next_transaction = transaction; do { @@ -527,7 +529,7 @@ void jbd2_journal_destroy_checkpoint(journal_t *journal) spin_unlock(&journal->j_list_lock); break; } - __jbd2_journal_clean_checkpoint_list(journal, true); + __jbd2_journal_clean_checkpoint_list(journal, SHRINK_DESTROY); spin_unlock(&journal->j_list_lock); cond_resched(); } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 78a9d08ae9f8..b341c396311b 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -501,7 +501,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) * frees some memory */ spin_lock(&journal->j_list_lock); - __jbd2_journal_clean_checkpoint_list(journal, false); + __jbd2_journal_clean_checkpoint_list(journal, SHRINK_BUSY_STOP); spin_unlock(&journal->j_list_lock); jbd2_debug(3, "JBD2: commit phase 1\n"); diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 971f3e826e15..58a961999d70 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1434,7 +1434,9 @@ void jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block); extern void jbd2_journal_commit_transaction(journal_t *); /* Checkpoint list management */ -void __jbd2_journal_clean_checkpoint_list(journal_t *journal, bool destroy); +enum shrink_type {SHRINK_DESTROY, SHRINK_BUSY_STOP, SHRINK_BUSY_SKIP}; + +void __jbd2_journal_clean_checkpoint_list(journal_t *journal, enum shrink_type type); unsigned long jbd2_journal_shrink_checkpoint_list(journal_t *journal, unsigned long *nr_to_scan); int __jbd2_journal_remove_checkpoint(struct journal_head *); int jbd2_journal_try_remove_checkpoint(struct journal_head *jh); -- cgit v1.2.3-59-g8ed1b From 26770a717cac57041d9414725e3e01dd19b08dd2 Mon Sep 17 00:00:00 2001 From: Ye Bin Date: Sun, 7 Apr 2024 14:53:55 +0800 Subject: jbd2: add prefix 'jbd2' for 'shrink_type' As 'shrink_type' is exported. The module prefix 'jbd2' is added to distinguish from memory reclamation. Signed-off-by: Ye Bin Reviewed-by: Jan Kara Reviewed-by: Zhang Yi Link: https://lore.kernel.org/r/20240407065355.1528580-3-yebin10@huawei.com Signed-off-by: Theodore Ts'o --- fs/jbd2/checkpoint.c | 22 +++++++++++----------- fs/jbd2/commit.c | 2 +- include/linux/jbd2.h | 4 ++-- 3 files changed, 14 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 80c0ab98bc63..951f78634adf 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -348,7 +348,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal) * Called with j_list_lock held. */ static unsigned long journal_shrink_one_cp_list(struct journal_head *jh, - enum shrink_type type, + enum jbd2_shrink_type type, bool *released) { struct journal_head *last_jh; @@ -365,12 +365,12 @@ static unsigned long journal_shrink_one_cp_list(struct journal_head *jh, jh = next_jh; next_jh = jh->b_cpnext; - if (type == SHRINK_DESTROY) { + if (type == JBD2_SHRINK_DESTROY) { ret = __jbd2_journal_remove_checkpoint(jh); } else { ret = jbd2_journal_try_remove_checkpoint(jh); if (ret < 0) { - if (type == SHRINK_BUSY_SKIP) + if (type == JBD2_SHRINK_BUSY_SKIP) continue; break; } @@ -437,7 +437,7 @@ again: tid = transaction->t_tid; freed = journal_shrink_one_cp_list(transaction->t_checkpoint_list, - SHRINK_BUSY_SKIP, &released); + JBD2_SHRINK_BUSY_SKIP, &released); nr_freed += freed; (*nr_to_scan) -= min(*nr_to_scan, freed); if (*nr_to_scan == 0) @@ -470,20 +470,20 @@ out: * journal_clean_checkpoint_list * * Find all the written-back checkpoint buffers in the journal and release them. - * If 'type' is SHRINK_DESTROY, release all buffers unconditionally. If 'type' - * is SHRINK_BUSY_STOP, will stop release buffers if encounters a busy buffer. - * To avoid wasting CPU cycles scanning the buffer list in some cases, don't - * pass SHRINK_BUSY_SKIP 'type' for this function. + * If 'type' is JBD2_SHRINK_DESTROY, release all buffers unconditionally. If + * 'type' is JBD2_SHRINK_BUSY_STOP, will stop release buffers if encounters a + * busy buffer. To avoid wasting CPU cycles scanning the buffer list in some + * cases, don't pass JBD2_SHRINK_BUSY_SKIP 'type' for this function. * * Called with j_list_lock held. */ void __jbd2_journal_clean_checkpoint_list(journal_t *journal, - enum shrink_type type) + enum jbd2_shrink_type type) { transaction_t *transaction, *last_transaction, *next_transaction; bool released; - WARN_ON_ONCE(type == SHRINK_BUSY_SKIP); + WARN_ON_ONCE(type == JBD2_SHRINK_BUSY_SKIP); transaction = journal->j_checkpoint_transactions; if (!transaction) @@ -529,7 +529,7 @@ void jbd2_journal_destroy_checkpoint(journal_t *journal) spin_unlock(&journal->j_list_lock); break; } - __jbd2_journal_clean_checkpoint_list(journal, SHRINK_DESTROY); + __jbd2_journal_clean_checkpoint_list(journal, JBD2_SHRINK_DESTROY); spin_unlock(&journal->j_list_lock); cond_resched(); } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index b341c396311b..75ea4e9a5cab 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -501,7 +501,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) * frees some memory */ spin_lock(&journal->j_list_lock); - __jbd2_journal_clean_checkpoint_list(journal, SHRINK_BUSY_STOP); + __jbd2_journal_clean_checkpoint_list(journal, JBD2_SHRINK_BUSY_STOP); spin_unlock(&journal->j_list_lock); jbd2_debug(3, "JBD2: commit phase 1\n"); diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 58a961999d70..7479f64c0939 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1434,9 +1434,9 @@ void jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block); extern void jbd2_journal_commit_transaction(journal_t *); /* Checkpoint list management */ -enum shrink_type {SHRINK_DESTROY, SHRINK_BUSY_STOP, SHRINK_BUSY_SKIP}; +enum jbd2_shrink_type {JBD2_SHRINK_DESTROY, JBD2_SHRINK_BUSY_STOP, JBD2_SHRINK_BUSY_SKIP}; -void __jbd2_journal_clean_checkpoint_list(journal_t *journal, enum shrink_type type); +void __jbd2_journal_clean_checkpoint_list(journal_t *journal, enum jbd2_shrink_type type); unsigned long jbd2_journal_shrink_checkpoint_list(journal_t *journal, unsigned long *nr_to_scan); int __jbd2_journal_remove_checkpoint(struct journal_head *); int jbd2_journal_try_remove_checkpoint(struct journal_head *jh); -- cgit v1.2.3-59-g8ed1b From 1dd719a959794467c2a75b3813df86cc6f55f5e3 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Wed, 8 May 2024 18:21:11 +0300 Subject: isofs: Use *-y instead of *-objs in Makefile *-objs suffix is reserved rather for (user-space) host programs while usually *-y suffix is used for kernel drivers (although *-objs works for that purpose for now). Let's correct the old usages of *-objs in Makefiles. Signed-off-by: Andy Shevchenko Signed-off-by: Jan Kara Message-Id: <20240508152129.1445372-1-andriy.shevchenko@linux.intel.com> --- fs/isofs/Makefile | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/isofs/Makefile b/fs/isofs/Makefile index 6498fd2b0f60..b25bc542a22b 100644 --- a/fs/isofs/Makefile +++ b/fs/isofs/Makefile @@ -5,7 +5,6 @@ obj-$(CONFIG_ISO9660_FS) += isofs.o -isofs-objs-y := namei.o inode.o dir.o util.o rock.o export.o -isofs-objs-$(CONFIG_JOLIET) += joliet.o -isofs-objs-$(CONFIG_ZISOFS) += compress.o -isofs-objs := $(isofs-objs-y) +isofs-y := namei.o inode.o dir.o util.o rock.o export.o +isofs-$(CONFIG_JOLIET) += joliet.o +isofs-$(CONFIG_ZISOFS) += compress.o -- cgit v1.2.3-59-g8ed1b From b1992c3772e69a6fd0e3fc81cd4d2820c8b6eca0 Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Sat, 27 Apr 2024 23:55:02 +0900 Subject: kbuild: use $(src) instead of $(srctree)/$(src) for source directory Kbuild conventionally uses $(obj)/ for generated files, and $(src)/ for checked-in source files. It is merely a convention without any functional difference. In fact, $(obj) and $(src) are exactly the same, as defined in scripts/Makefile.build: src := $(obj) When the kernel is built in a separate output directory, $(src) does not accurately reflect the source directory location. While Kbuild resolves this discrepancy by specifying VPATH=$(srctree) to search for source files, it does not cover all cases. For example, when adding a header search path for local headers, -I$(srctree)/$(src) is typically passed to the compiler. This introduces inconsistency between upstream and downstream Makefiles because $(src) is used instead of $(srctree)/$(src) for the latter. To address this inconsistency, this commit changes the semantics of $(src) so that it always points to the directory in the source tree. Going forward, the variables used in Makefiles will have the following meanings: $(obj) - directory in the object tree $(src) - directory in the source tree (changed by this commit) $(objtree) - the top of the kernel object tree $(srctree) - the top of the kernel source tree Consequently, $(srctree)/$(src) in upstream Makefiles need to be replaced with $(src). Signed-off-by: Masahiro Yamada Reviewed-by: Nicolas Schier --- Documentation/Makefile | 8 ++++---- Documentation/devicetree/bindings/Makefile | 6 +++--- Documentation/kbuild/makefiles.rst | 12 ++++++------ Makefile | 7 +++++++ arch/arc/boot/dts/Makefile | 3 +-- arch/arm/Kbuild | 2 +- arch/arm/boot/Makefile | 3 +-- arch/arm/mach-s3c/Makefile | 2 +- arch/arm/plat-orion/Makefile | 2 +- arch/arm/tools/Makefile | 2 +- arch/arm64/kernel/vdso/Makefile | 2 +- arch/arm64/kvm/Makefile | 4 ++-- arch/arm64/kvm/hyp/Makefile | 2 +- arch/csky/boot/dts/Makefile | 4 +--- arch/csky/kernel/vdso/Makefile | 2 +- arch/loongarch/kvm/Makefile | 2 +- arch/loongarch/vdso/Makefile | 2 +- arch/mips/kernel/syscalls/Makefile | 2 +- arch/mips/vdso/Makefile | 4 ++-- arch/nios2/boot/dts/Makefile | 3 +-- arch/parisc/kernel/vdso32/Makefile | 2 +- arch/parisc/kernel/vdso64/Makefile | 2 +- arch/powerpc/boot/Makefile | 6 +++--- arch/powerpc/boot/dts/Makefile | 3 +-- arch/powerpc/boot/dts/fsl/Makefile | 3 +-- arch/powerpc/kernel/vdso/Makefile | 4 ++-- arch/riscv/kernel/compat_vdso/Makefile | 2 +- arch/riscv/kernel/vdso/Makefile | 2 +- arch/riscv/kvm/Makefile | 2 +- arch/s390/kernel/syscalls/Makefile | 4 ++-- arch/s390/kernel/vdso32/Makefile | 2 +- arch/s390/kernel/vdso64/Makefile | 2 +- arch/sparc/vdso/Makefile | 2 +- arch/um/kernel/Makefile | 2 +- arch/x86/boot/Makefile | 2 +- arch/x86/entry/vdso/Makefile | 2 +- arch/x86/kernel/Makefile | 2 +- arch/x86/kernel/cpu/Makefile | 2 +- arch/x86/mm/Makefile | 2 +- arch/x86/um/vdso/Makefile | 2 +- arch/xtensa/boot/dts/Makefile | 3 +-- certs/Makefile | 4 ++-- drivers/Makefile | 5 ----- drivers/crypto/intel/qat/qat_420xx/Makefile | 2 +- drivers/crypto/intel/qat/qat_4xxx/Makefile | 2 +- drivers/crypto/intel/qat/qat_c3xxx/Makefile | 2 +- drivers/crypto/intel/qat/qat_c3xxxvf/Makefile | 2 +- drivers/crypto/intel/qat/qat_c62x/Makefile | 2 +- drivers/crypto/intel/qat/qat_c62xvf/Makefile | 2 +- drivers/crypto/intel/qat/qat_dh895xcc/Makefile | 2 +- drivers/crypto/intel/qat/qat_dh895xccvf/Makefile | 2 +- drivers/gpu/drm/amd/amdgpu/Makefile | 2 +- drivers/gpu/drm/arm/display/komeda/Makefile | 4 ++-- drivers/gpu/drm/i915/Makefile | 4 ++-- drivers/gpu/drm/imagination/Makefile | 2 +- drivers/gpu/drm/msm/Makefile | 8 ++++---- drivers/gpu/drm/nouveau/Kbuild | 10 ++++------ drivers/gpu/drm/xe/Makefile | 10 +++++----- drivers/hid/amd-sfh-hid/Makefile | 2 +- drivers/hid/intel-ish-hid/Makefile | 2 +- drivers/md/dm-vdo/Makefile | 2 +- drivers/net/ethernet/aquantia/atlantic/Makefile | 2 +- drivers/net/ethernet/chelsio/libcxgb/Makefile | 2 +- drivers/net/ethernet/fungible/funeth/Makefile | 2 +- drivers/net/ethernet/hisilicon/hns3/Makefile | 2 +- drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile | 4 ++-- .../net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile | 6 +++--- .../net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile | 6 +++--- .../net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile | 6 +++--- drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile | 6 +++--- drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/dvm/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/mei/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/mvm/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/tests/Makefile | 2 +- drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile | 2 +- drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile | 2 +- drivers/scsi/aic7xxx/Makefile | 12 ++++++------ drivers/staging/rtl8723bs/Makefile | 2 +- fs/iomap/Makefile | 2 +- fs/unicode/Makefile | 14 +++++++------- fs/xfs/Makefile | 4 ++-- init/Makefile | 2 +- lib/Makefile | 6 +++--- lib/raid6/Makefile | 2 +- net/wireless/Makefile | 2 +- rust/Makefile | 6 +++--- samples/bpf/Makefile | 2 +- samples/hid/Makefile | 2 +- scripts/Kbuild.include | 3 +-- scripts/Makefile.asm-generic | 6 +++--- scripts/Makefile.build | 2 +- scripts/Makefile.clean | 2 +- scripts/Makefile.lib | 10 +++++----- scripts/Makefile.modpost | 2 +- scripts/dtc/Makefile | 6 +++--- scripts/gdb/linux/Makefile | 2 +- scripts/genksyms/Makefile | 4 ++-- scripts/kconfig/Makefile | 8 ++++---- security/tomoyo/Makefile | 2 +- usr/Makefile | 2 +- usr/include/Makefile | 2 +- 102 files changed, 173 insertions(+), 182 deletions(-) (limited to 'fs') diff --git a/Documentation/Makefile b/Documentation/Makefile index b68f8c816897..66d15fff2089 100644 --- a/Documentation/Makefile +++ b/Documentation/Makefile @@ -76,22 +76,22 @@ loop_cmd = $(echo-cmd) $(cmd_$(1)) || exit; # * dest folder relative to $(BUILDDIR) and # * cache folder relative to $(BUILDDIR)/.doctrees # $4 dest subfolder e.g. "man" for man pages at userspace-api/media/man -# $5 reST source folder relative to $(srctree)/$(src), +# $5 reST source folder relative to $(src), # e.g. "userspace-api/media" for the linux-tv book-set at ./Documentation/userspace-api/media quiet_cmd_sphinx = SPHINX $@ --> file://$(abspath $(BUILDDIR)/$3/$4) cmd_sphinx = $(MAKE) BUILDDIR=$(abspath $(BUILDDIR)) $(build)=Documentation/userspace-api/media $2 && \ PYTHONDONTWRITEBYTECODE=1 \ - BUILDDIR=$(abspath $(BUILDDIR)) SPHINX_CONF=$(abspath $(srctree)/$(src)/$5/$(SPHINX_CONF)) \ + BUILDDIR=$(abspath $(BUILDDIR)) SPHINX_CONF=$(abspath $(src)/$5/$(SPHINX_CONF)) \ $(PYTHON3) $(srctree)/scripts/jobserver-exec \ $(CONFIG_SHELL) $(srctree)/Documentation/sphinx/parallel-wrapper.sh \ $(SPHINXBUILD) \ -b $2 \ - -c $(abspath $(srctree)/$(src)) \ + -c $(abspath $(src)) \ -d $(abspath $(BUILDDIR)/.doctrees/$3) \ -D version=$(KERNELVERSION) -D release=$(KERNELRELEASE) \ $(ALLSPHINXOPTS) \ - $(abspath $(srctree)/$(src)/$5) \ + $(abspath $(src)/$5) \ $(abspath $(BUILDDIR)/$3/$4) && \ if [ "x$(DOCS_CSS)" != "x" ]; then \ cp $(if $(patsubst /%,,$(DOCS_CSS)),$(abspath $(srctree)/$(DOCS_CSS)),$(DOCS_CSS)) $(BUILDDIR)/$3/_static/; \ diff --git a/Documentation/devicetree/bindings/Makefile b/Documentation/devicetree/bindings/Makefile index 8cdda477987f..bf7d64632e20 100644 --- a/Documentation/devicetree/bindings/Makefile +++ b/Documentation/devicetree/bindings/Makefile @@ -25,7 +25,7 @@ quiet_cmd_extract_ex = DTEX $@ $(obj)/%.example.dts: $(src)/%.yaml check_dtschema_version FORCE $(call if_changed,extract_ex) -find_all_cmd = find $(srctree)/$(src) \( -name '*.yaml' ! \ +find_all_cmd = find $(src) \( -name '*.yaml' ! \ -name 'processed-schema*' \) find_cmd = $(find_all_cmd) | \ @@ -37,12 +37,12 @@ CHK_DT_EXAMPLES := $(patsubst $(srctree)/%.yaml,%.example.dtb, $(shell $(find_cm quiet_cmd_yamllint = LINT $(src) cmd_yamllint = ($(find_cmd) | \ xargs -n200 -P$$(nproc) \ - $(DT_SCHEMA_LINT) -f parsable -c $(srctree)/$(src)/.yamllint >&2) \ + $(DT_SCHEMA_LINT) -f parsable -c $(src)/.yamllint >&2) \ && touch $@ || true quiet_cmd_chk_bindings = CHKDT $(src) cmd_chk_bindings = ($(find_cmd) | \ - xargs -n200 -P$$(nproc) $(DT_DOC_CHECKER) -u $(srctree)/$(src)) \ + xargs -n200 -P$$(nproc) $(DT_DOC_CHECKER) -u $(src)) \ && touch $@ || true quiet_cmd_mk_schema = SCHEMA $@ diff --git a/Documentation/kbuild/makefiles.rst b/Documentation/kbuild/makefiles.rst index ad118b7a1806..991ce6081e35 100644 --- a/Documentation/kbuild/makefiles.rst +++ b/Documentation/kbuild/makefiles.rst @@ -346,7 +346,7 @@ ccflags-y, asflags-y and ldflags-y Example:: #arch/cris/boot/compressed/Makefile - ldflags-y += -T $(srctree)/$(src)/decompress_$(arch-y).lds + ldflags-y += -T $(src)/decompress_$(arch-y).lds subdir-ccflags-y, subdir-asflags-y The two flags listed above are similar to ccflags-y and asflags-y. @@ -426,14 +426,14 @@ path to prerequisite files and target files. Two variables are used when defining custom rules: $(src) - $(src) is a relative path which points to the directory - where the Makefile is located. Always use $(src) when + $(src) is the directory where the Makefile is located. Always use $(src) when referring to files located in the src tree. $(obj) - $(obj) is a relative path which points to the directory - where the target is saved. Always use $(obj) when - referring to generated files. + $(obj) is the directory where the target is saved. Always use $(obj) when + referring to generated files. Use $(obj) for pattern rules that need to work + for both generated files and real sources (VPATH will help to find the + prerequisites not only in the object tree but also in the source tree). Example:: diff --git a/Makefile b/Makefile index 12e1a792b8de..c97c8d96061c 100644 --- a/Makefile +++ b/Makefile @@ -263,7 +263,14 @@ srctree := $(abs_srctree) endif objtree := . + +VPATH := + +ifeq ($(KBUILD_EXTMOD),) +ifdef building_out_of_srctree VPATH := $(srctree) +endif +endif export building_out_of_srctree srctree objtree VPATH diff --git a/arch/arc/boot/dts/Makefile b/arch/arc/boot/dts/Makefile index 4237aa5de3a3..48704dfdf75c 100644 --- a/arch/arc/boot/dts/Makefile +++ b/arch/arc/boot/dts/Makefile @@ -10,8 +10,7 @@ obj-y += $(builtindtb-y).dtb.o dtb-y := $(builtindtb-y).dtb # for CONFIG_OF_ALL_DTBS test -dtstree := $(srctree)/$(src) -dtb- := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb- := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) # board-specific dtc flags DTC_FLAGS_hsdk += --pad 20 diff --git a/arch/arm/Kbuild b/arch/arm/Kbuild index b506622e7e23..69de6b6243c7 100644 --- a/arch/arm/Kbuild +++ b/arch/arm/Kbuild @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only obj-$(CONFIG_FPE_NWFPE) += nwfpe/ # Put arch/arm/fastfpe/ to use this. -obj-$(CONFIG_FPE_FASTFPE) += $(patsubst $(srctree)/$(src)/%,%,$(wildcard $(srctree)/$(src)/fastfpe/)) +obj-$(CONFIG_FPE_FASTFPE) += $(patsubst $(src)/%,%,$(wildcard $(src)/fastfpe/)) obj-$(CONFIG_VFP) += vfp/ obj-$(CONFIG_XEN) += xen/ obj-$(CONFIG_VDSO) += vdso/ diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile index abd6a2889fd0..ba9b9a802469 100644 --- a/arch/arm/boot/Makefile +++ b/arch/arm/boot/Makefile @@ -25,8 +25,7 @@ targets := Image zImage xipImage bootpImage uImage ifeq ($(CONFIG_XIP_KERNEL),y) -cmd_deflate_xip_data = $(CONFIG_SHELL) -c \ - '$(srctree)/$(src)/deflate_xip_data.sh $< $@' +cmd_deflate_xip_data = $(CONFIG_SHELL) -c '$(src)/deflate_xip_data.sh $< $@' ifeq ($(CONFIG_XIP_DEFLATED_DATA),y) quiet_cmd_mkxip = XIPZ $@ diff --git a/arch/arm/mach-s3c/Makefile b/arch/arm/mach-s3c/Makefile index 713827bef831..988c49672715 100644 --- a/arch/arm/mach-s3c/Makefile +++ b/arch/arm/mach-s3c/Makefile @@ -2,7 +2,7 @@ # # Copyright 2009 Simtec Electronics -include $(srctree)/$(src)/Makefile.s3c64xx +include $(src)/Makefile.s3c64xx # Objects we always build independent of SoC choice diff --git a/arch/arm/plat-orion/Makefile b/arch/arm/plat-orion/Makefile index 830b0be038c6..e8c7580df8ca 100644 --- a/arch/arm/plat-orion/Makefile +++ b/arch/arm/plat-orion/Makefile @@ -2,7 +2,7 @@ # # Makefile for the linux kernel. # -ccflags-y := -I$(srctree)/$(src)/include +ccflags-y := -I$(src)/include orion-gpio-$(CONFIG_GPIOLIB) += gpio.o obj-$(CONFIG_PLAT_ORION_LEGACY) += irq.o pcie.o time.o common.o mpp.o diff --git a/arch/arm/tools/Makefile b/arch/arm/tools/Makefile index 81f13bdf32f2..28b6da8ac5f6 100644 --- a/arch/arm/tools/Makefile +++ b/arch/arm/tools/Makefile @@ -9,7 +9,7 @@ gen := arch/$(ARCH)/include/generated kapi := $(gen)/asm uapi := $(gen)/uapi/asm syshdr := $(srctree)/scripts/syscallhdr.sh -sysnr := $(srctree)/$(src)/syscallnr.sh +sysnr := $(src)/syscallnr.sh systbl := $(srctree)/scripts/syscalltbl.sh syscall := $(src)/syscall.tbl diff --git a/arch/arm64/kernel/vdso/Makefile b/arch/arm64/kernel/vdso/Makefile index 8818287f1095..53e86d3bc159 100644 --- a/arch/arm64/kernel/vdso/Makefile +++ b/arch/arm64/kernel/vdso/Makefile @@ -68,7 +68,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile index c0c050e53157..ab474d444cef 100644 --- a/arch/arm64/kvm/Makefile +++ b/arch/arm64/kvm/Makefile @@ -3,7 +3,7 @@ # Makefile for Kernel-based Virtual Machine module # -ccflags-y += -I $(srctree)/$(src) +ccflags-y += -I $(src) include $(srctree)/virt/kvm/Makefile.kvm @@ -30,7 +30,7 @@ define rule_gen_hyp_constants $(call filechk,offsets,__HYP_CONSTANTS_H__) endef -CFLAGS_hyp-constants.o = -I $(srctree)/$(src)/hyp/include +CFLAGS_hyp-constants.o = -I $(src)/hyp/include $(obj)/hyp-constants.s: $(src)/hyp/hyp-constants.c FORCE $(call if_changed_dep,cc_s_c) diff --git a/arch/arm64/kvm/hyp/Makefile b/arch/arm64/kvm/hyp/Makefile index a38dea6186c9..d61e44642f98 100644 --- a/arch/arm64/kvm/hyp/Makefile +++ b/arch/arm64/kvm/hyp/Makefile @@ -3,7 +3,7 @@ # Makefile for Kernel-based Virtual Machine module, HYP part # -incdir := $(srctree)/$(src)/include +incdir := $(src)/include subdir-asflags-y := -I$(incdir) subdir-ccflags-y := -I$(incdir) diff --git a/arch/csky/boot/dts/Makefile b/arch/csky/boot/dts/Makefile index 5f1f55e911ad..aabffc1912e8 100644 --- a/arch/csky/boot/dts/Makefile +++ b/arch/csky/boot/dts/Makefile @@ -1,4 +1,2 @@ # SPDX-License-Identifier: GPL-2.0-only -dtstree := $(srctree)/$(src) - -dtb-y := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-y := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/csky/kernel/vdso/Makefile b/arch/csky/kernel/vdso/Makefile index 24a01c22293e..e79a725f5075 100644 --- a/arch/csky/kernel/vdso/Makefile +++ b/arch/csky/kernel/vdso/Makefile @@ -57,4 +57,4 @@ quiet_cmd_vdsold = VDSOLD $@ # Extracts symbol offsets from the VDSO, converting them into an assembly file # that contains the same symbols at the same offsets. quiet_cmd_so2s = SO2S $@ - cmd_so2s = $(NM) -D $< | $(srctree)/$(src)/so2s.sh > $@ + cmd_so2s = $(NM) -D $< | $(src)/so2s.sh > $@ diff --git a/arch/loongarch/kvm/Makefile b/arch/loongarch/kvm/Makefile index 244467d7792a..b2f4cbe01ae8 100644 --- a/arch/loongarch/kvm/Makefile +++ b/arch/loongarch/kvm/Makefile @@ -3,7 +3,7 @@ # Makefile for LoongArch KVM support # -ccflags-y += -I $(srctree)/$(src) +ccflags-y += -I $(src) include $(srctree)/virt/kvm/Makefile.kvm diff --git a/arch/loongarch/vdso/Makefile b/arch/loongarch/vdso/Makefile index 75c6726382c3..cdfc4c793e2c 100644 --- a/arch/loongarch/vdso/Makefile +++ b/arch/loongarch/vdso/Makefile @@ -52,7 +52,7 @@ quiet_cmd_vdsoas_o_S = AS $@ cmd_vdsoas_o_S = $(CC) $(a_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/mips/kernel/syscalls/Makefile b/arch/mips/kernel/syscalls/Makefile index e6b21de65cca..56f6f093bb88 100644 --- a/arch/mips/kernel/syscalls/Makefile +++ b/arch/mips/kernel/syscalls/Makefile @@ -5,7 +5,7 @@ uapi := arch/$(SRCARCH)/include/generated/uapi/asm $(shell mkdir -p $(uapi) $(kapi)) syshdr := $(srctree)/scripts/syscallhdr.sh -sysnr := $(srctree)/$(src)/syscallnr.sh +sysnr := $(src)/syscallnr.sh systbl := $(srctree)/scripts/syscalltbl.sh quiet_cmd_syshdr = SYSHDR $@ diff --git a/arch/mips/vdso/Makefile b/arch/mips/vdso/Makefile index eb56581f6d73..40b839e91806 100644 --- a/arch/mips/vdso/Makefile +++ b/arch/mips/vdso/Makefile @@ -43,8 +43,8 @@ CFLAGS_vgettimeofday.o = -include $(c-gettimeofday-y) # config-n32-o32-env.c prepares the environment to build a 32bit vDSO # library on a 64bit kernel. # Note: Needs to be included before than the generic library. -CFLAGS_vgettimeofday-o32.o = -include $(srctree)/$(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) -CFLAGS_vgettimeofday-n32.o = -include $(srctree)/$(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) +CFLAGS_vgettimeofday-o32.o = -include $(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) +CFLAGS_vgettimeofday-n32.o = -include $(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) endif CFLAGS_REMOVE_vgettimeofday.o = $(CC_FLAGS_FTRACE) diff --git a/arch/nios2/boot/dts/Makefile b/arch/nios2/boot/dts/Makefile index e9e31bb40df8..1a2e8996bec7 100644 --- a/arch/nios2/boot/dts/Makefile +++ b/arch/nios2/boot/dts/Makefile @@ -2,5 +2,4 @@ obj-y := $(patsubst %.dts,%.dtb.o,$(CONFIG_NIOS2_DTB_SOURCE)) -dtstree := $(srctree)/$(src) -dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/parisc/kernel/vdso32/Makefile b/arch/parisc/kernel/vdso32/Makefile index 60dc708a0f80..1350d50c6306 100644 --- a/arch/parisc/kernel/vdso32/Makefile +++ b/arch/parisc/kernel/vdso32/Makefile @@ -40,7 +40,7 @@ quiet_cmd_vdso32as = VDSO32A $@ cmd_vdso32as = $(CROSS32CC) $(a_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/parisc/kernel/vdso64/Makefile b/arch/parisc/kernel/vdso64/Makefile index 0982f3099ae2..0b1c1cc4c2c7 100644 --- a/arch/parisc/kernel/vdso64/Makefile +++ b/arch/parisc/kernel/vdso64/Makefile @@ -40,7 +40,7 @@ quiet_cmd_vdso64as = VDSO64A $@ cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 968aee2025b8..a0b3b5cc14cd 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -218,7 +218,7 @@ $(addprefix $(obj)/,$(libfdt) $(libfdtheader)): $(obj)/%: $(srctree)/scripts/dtc $(obj)/empty.c: $(Q)touch $@ -$(obj)/zImage.coff.lds $(obj)/zImage.ps3.lds : $(obj)/%: $(srctree)/$(src)/%.S +$(obj)/zImage.coff.lds $(obj)/zImage.ps3.lds : $(obj)/%: $(src)/%.S $(Q)cp $< $@ clean-files := $(zlib-) $(zlibheader-) $(zliblinuxheader-) \ @@ -252,9 +252,9 @@ targets += $(patsubst $(obj)/%,%,$(obj-boot) wrapper.a) zImage.lds extra-y := $(obj)/wrapper.a $(obj-plat) $(obj)/empty.o \ $(obj)/zImage.lds $(obj)/zImage.coff.lds $(obj)/zImage.ps3.lds -dtstree := $(srctree)/$(src)/dts +dtstree := $(src)/dts -wrapper :=$(srctree)/$(src)/wrapper +wrapper := $(src)/wrapper wrapperbits := $(extra-y) $(addprefix $(obj)/,addnote hack-coff mktree) \ $(wrapper) FORCE diff --git a/arch/powerpc/boot/dts/Makefile b/arch/powerpc/boot/dts/Makefile index fb335d05aae8..0cd0d8558b47 100644 --- a/arch/powerpc/boot/dts/Makefile +++ b/arch/powerpc/boot/dts/Makefile @@ -2,5 +2,4 @@ subdir-y += fsl -dtstree := $(srctree)/$(src) -dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/powerpc/boot/dts/fsl/Makefile b/arch/powerpc/boot/dts/fsl/Makefile index 3bae982641e9..d3ecdf14bc42 100644 --- a/arch/powerpc/boot/dts/fsl/Makefile +++ b/arch/powerpc/boot/dts/fsl/Makefile @@ -1,4 +1,3 @@ # SPDX-License-Identifier: GPL-2.0 -dtstree := $(srctree)/$(src) -dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/powerpc/kernel/vdso/Makefile b/arch/powerpc/kernel/vdso/Makefile index a14eab366993..7d66b6e07993 100644 --- a/arch/powerpc/kernel/vdso/Makefile +++ b/arch/powerpc/kernel/vdso/Makefile @@ -90,10 +90,10 @@ $(obj)/vgettimeofday-64.o: %-64.o: %.c FORCE $(call if_changed_dep,cc_o_c) # Generate VDSO offsets using helper script -gen-vdso32sym := $(srctree)/$(src)/gen_vdso32_offsets.sh +gen-vdso32sym := $(src)/gen_vdso32_offsets.sh quiet_cmd_vdso32sym = VDSO32SYM $@ cmd_vdso32sym = $(NM) $< | $(gen-vdso32sym) | LC_ALL=C sort > $@ -gen-vdso64sym := $(srctree)/$(src)/gen_vdso64_offsets.sh +gen-vdso64sym := $(src)/gen_vdso64_offsets.sh quiet_cmd_vdso64sym = VDSO64SYM $@ cmd_vdso64sym = $(NM) $< | $(gen-vdso64sym) | LC_ALL=C sort > $@ diff --git a/arch/riscv/kernel/compat_vdso/Makefile b/arch/riscv/kernel/compat_vdso/Makefile index 3df4cb788c1f..362208dfa7ee 100644 --- a/arch/riscv/kernel/compat_vdso/Makefile +++ b/arch/riscv/kernel/compat_vdso/Makefile @@ -58,7 +58,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # Generate VDSO offsets using helper script -gen-compat_vdsosym := $(srctree)/$(src)/gen_compat_vdso_offsets.sh +gen-compat_vdsosym := $(src)/gen_compat_vdso_offsets.sh quiet_cmd_compat_vdsosym = VDSOSYM $@ cmd_compat_vdsosym = $(NM) $< | $(gen-compat_vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makefile index 272c431ac5b9..25f0ee629971 100644 --- a/arch/riscv/kernel/vdso/Makefile +++ b/arch/riscv/kernel/vdso/Makefile @@ -60,7 +60,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/riscv/kvm/Makefile b/arch/riscv/kvm/Makefile index c9646521f113..c2cacfbc06a0 100644 --- a/arch/riscv/kvm/Makefile +++ b/arch/riscv/kvm/Makefile @@ -3,7 +3,7 @@ # Makefile for RISC-V KVM support # -ccflags-y += -I $(srctree)/$(src) +ccflags-y += -I $(src) include $(srctree)/virt/kvm/Makefile.kvm diff --git a/arch/s390/kernel/syscalls/Makefile b/arch/s390/kernel/syscalls/Makefile index fb85e797946d..1bb78b9468e8 100644 --- a/arch/s390/kernel/syscalls/Makefile +++ b/arch/s390/kernel/syscalls/Makefile @@ -4,8 +4,8 @@ gen := arch/$(ARCH)/include/generated kapi := $(gen)/asm uapi := $(gen)/uapi/asm -syscall := $(srctree)/$(src)/syscall.tbl -systbl := $(srctree)/$(src)/syscalltbl +syscall := $(src)/syscall.tbl +systbl := $(src)/syscalltbl gen-y := $(kapi)/syscall_table.h kapi-hdrs-y := $(kapi)/unistd_nr.h diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile index 70e9949c2612..c263b91adfb1 100644 --- a/arch/s390/kernel/vdso32/Makefile +++ b/arch/s390/kernel/vdso32/Makefile @@ -62,7 +62,7 @@ quiet_cmd_vdso32cc = VDSO32C $@ cmd_vdso32cc = $(CC) $(c_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile index 2b3617b6d162..9566bed7d5b2 100644 --- a/arch/s390/kernel/vdso64/Makefile +++ b/arch/s390/kernel/vdso64/Makefile @@ -72,7 +72,7 @@ quiet_cmd_vdso64cc = VDSO64C $@ cmd_vdso64cc = $(CC) $(c_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/sparc/vdso/Makefile b/arch/sparc/vdso/Makefile index e8aef2c8ae99..0fe134abbcf1 100644 --- a/arch/sparc/vdso/Makefile +++ b/arch/sparc/vdso/Makefile @@ -103,7 +103,7 @@ quiet_cmd_vdso = VDSO $@ cmd_vdso = $(LD) -nostdlib -o $@ \ $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -T $(filter %.lds,$^) $(filter %.o,$^) && \ - sh $(srctree)/$(src)/checkundef.sh '$(OBJDUMP)' '$@' + sh $(src)/checkundef.sh '$(OBJDUMP)' '$@' VDSO_LDFLAGS = -shared --hash-style=both --build-id=sha1 -Bsymbolic GCOV_PROFILE := n diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 811188be954c..f8567b933ffa 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -47,7 +47,7 @@ $(obj)/config.c: $(src)/config.c.in $(obj)/config.tmp FORCE $(call if_changed,quote2) quiet_cmd_mkcapflags = MKCAP $@ - cmd_mkcapflags = $(CONFIG_SHELL) $(srctree)/$(src)/../../x86/kernel/cpu/mkcapflags.sh $@ $^ + cmd_mkcapflags = $(CONFIG_SHELL) $(src)/../../x86/kernel/cpu/mkcapflags.sh $@ $^ cpufeature = $(src)/../../x86/include/asm/cpufeatures.h vmxfeature = $(src)/../../x86/include/asm/vmxfeatures.h diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile index 3cece19b7473..29cda98c65f8 100644 --- a/arch/x86/boot/Makefile +++ b/arch/x86/boot/Makefile @@ -129,7 +129,7 @@ targets += mtools.conf # genimage.sh requires bash, but it also has a bunch of other # external dependencies. quiet_cmd_genimage = GENIMAGE $3 -cmd_genimage = $(BASH) $(srctree)/$(src)/genimage.sh $2 $3 $(obj)/bzImage \ + cmd_genimage = $(BASH) $(src)/genimage.sh $2 $3 $(obj)/bzImage \ $(obj)/mtools.conf '$(FDARGS)' $(FDINITRD) PHONY += bzdisk fdimage fdimage144 fdimage288 hdimage isoimage diff --git a/arch/x86/entry/vdso/Makefile b/arch/x86/entry/vdso/Makefile index 3d64bcc403cf..c003452dba8c 100644 --- a/arch/x86/entry/vdso/Makefile +++ b/arch/x86/entry/vdso/Makefile @@ -176,7 +176,7 @@ quiet_cmd_vdso = VDSO $@ cmd_vdso = $(LD) -o $@ \ $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -T $(filter %.lds,$^) $(filter %.o,$^) && \ - sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@' + sh $(src)/checkundef.sh '$(NM)' '$@' VDSO_LDFLAGS = -shared --hash-style=both --build-id=sha1 \ $(call ld-option, --eh-frame-hdr) -Bsymbolic -z noexecstack diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 74077694da7d..8a9fc9ec69d3 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -40,7 +40,7 @@ KMSAN_SANITIZE_sev.o := n KCOV_INSTRUMENT_head$(BITS).o := n KCOV_INSTRUMENT_sev.o := n -CFLAGS_irq.o := -I $(srctree)/$(src)/../include/asm/trace +CFLAGS_irq.o := -I $(src)/../include/asm/trace obj-y += head_$(BITS).o obj-y += head$(BITS).o diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index eb4dbcdf41f1..a02bba0ed6b9 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -60,7 +60,7 @@ obj-$(CONFIG_ACRN_GUEST) += acrn.o obj-$(CONFIG_DEBUG_FS) += debugfs.o quiet_cmd_mkcapflags = MKCAP $@ - cmd_mkcapflags = $(CONFIG_SHELL) $(srctree)/$(src)/mkcapflags.sh $@ $^ + cmd_mkcapflags = $(CONFIG_SHELL) $(src)/mkcapflags.sh $@ $^ cpufeature = $(src)/../../include/asm/cpufeatures.h vmxfeature = $(src)/../../include/asm/vmxfeatures.h diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 428048e73bd2..8d3a00e5c528 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -34,7 +34,7 @@ obj-y += pat/ CFLAGS_physaddr.o := -fno-stack-protector CFLAGS_mem_encrypt_identity.o := -fno-stack-protector -CFLAGS_fault.o := -I $(srctree)/$(src)/../include/asm/trace +CFLAGS_fault.o := -I $(src)/../include/asm/trace obj-$(CONFIG_X86_32) += pgtable_32.o iomap_32.o diff --git a/arch/x86/um/vdso/Makefile b/arch/x86/um/vdso/Makefile index b86d634730b2..2303fa59971c 100644 --- a/arch/x86/um/vdso/Makefile +++ b/arch/x86/um/vdso/Makefile @@ -63,7 +63,7 @@ quiet_cmd_vdso = VDSO $@ cmd_vdso = $(CC) -nostdlib -o $@ \ $(CC_FLAGS_LTO) $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) && \ - sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@' + sh $(src)/checkundef.sh '$(NM)' '$@' VDSO_LDFLAGS = -fPIC -shared -Wl,--hash-style=sysv -z noexecstack GCOV_PROFILE := n diff --git a/arch/xtensa/boot/dts/Makefile b/arch/xtensa/boot/dts/Makefile index 720628c0d8b9..d6408c16d74e 100644 --- a/arch/xtensa/boot/dts/Makefile +++ b/arch/xtensa/boot/dts/Makefile @@ -10,5 +10,4 @@ obj-$(CONFIG_OF) += $(addsuffix .dtb.o, $(CONFIG_BUILTIN_DTB_SOURCE)) # for CONFIG_OF_ALL_DTBS test -dtstree := $(srctree)/$(src) -dtb- := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb- := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/certs/Makefile b/certs/Makefile index 799ad7b9e68a..1094e3860c2a 100644 --- a/certs/Makefile +++ b/certs/Makefile @@ -13,7 +13,7 @@ CFLAGS_blacklist_hashes.o := -I $(obj) quiet_cmd_check_and_copy_blacklist_hash_list = GEN $@ cmd_check_and_copy_blacklist_hash_list = \ $(if $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST), \ - $(AWK) -f $(srctree)/$(src)/check-blacklist-hashes.awk $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST) >&2; \ + $(AWK) -f $(src)/check-blacklist-hashes.awk $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST) >&2; \ { cat $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST); echo $(comma) NULL; } > $@, \ echo NULL > $@) @@ -55,7 +55,7 @@ $(obj)/signing_key.pem: $(obj)/x509.genkey FORCE targets += signing_key.pem quiet_cmd_copy_x509_config = COPY $@ - cmd_copy_x509_config = cat $(srctree)/$(src)/default_x509.genkey > $@ + cmd_copy_x509_config = cat $(src)/default_x509.genkey > $@ # You can provide your own config file. If not present, copy the default one. $(obj)/x509.genkey: diff --git a/drivers/Makefile b/drivers/Makefile index 3bf5cab4b451..fe9ceb0d2288 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -6,11 +6,6 @@ # Rewritten to use lists instead of if-statements. # -# Some driver Makefiles miss $(srctree)/ for include directive. -ifdef building_out_of_srctree -MAKEFLAGS += --include-dir=$(srctree) -endif - obj-y += cache/ obj-y += irqchip/ obj-y += bus/ diff --git a/drivers/crypto/intel/qat/qat_420xx/Makefile b/drivers/crypto/intel/qat/qat_420xx/Makefile index a90fbe00b3c8..45728659fbc4 100644 --- a/drivers/crypto/intel/qat/qat_420xx/Makefile +++ b/drivers/crypto/intel/qat/qat_420xx/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_420XX) += qat_420xx.o qat_420xx-objs := adf_drv.o adf_420xx_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_4xxx/Makefile b/drivers/crypto/intel/qat/qat_4xxx/Makefile index ff9c8b5897ea..9ba202079a22 100644 --- a/drivers/crypto/intel/qat/qat_4xxx/Makefile +++ b/drivers/crypto/intel/qat/qat_4xxx/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: (BSD-3-Clause OR GPL-2.0-only) -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_4XXX) += qat_4xxx.o qat_4xxx-objs := adf_drv.o adf_4xxx_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c3xxx/Makefile b/drivers/crypto/intel/qat/qat_c3xxx/Makefile index 92ef416ccc78..7a06ad519bfc 100644 --- a/drivers/crypto/intel/qat/qat_c3xxx/Makefile +++ b/drivers/crypto/intel/qat/qat_c3xxx/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C3XXX) += qat_c3xxx.o qat_c3xxx-objs := adf_drv.o adf_c3xxx_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile b/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile index b6d76825a92c..7ef633058c4f 100644 --- a/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile +++ b/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C3XXXVF) += qat_c3xxxvf.o qat_c3xxxvf-objs := adf_drv.o adf_c3xxxvf_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c62x/Makefile b/drivers/crypto/intel/qat/qat_c62x/Makefile index d581f7c87d6c..cc9255b3b198 100644 --- a/drivers/crypto/intel/qat/qat_c62x/Makefile +++ b/drivers/crypto/intel/qat/qat_c62x/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C62X) += qat_c62x.o qat_c62x-objs := adf_drv.o adf_c62x_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c62xvf/Makefile b/drivers/crypto/intel/qat/qat_c62xvf/Makefile index 446c3d638605..256786662d60 100644 --- a/drivers/crypto/intel/qat/qat_c62xvf/Makefile +++ b/drivers/crypto/intel/qat/qat_c62xvf/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C62XVF) += qat_c62xvf.o qat_c62xvf-objs := adf_drv.o adf_c62xvf_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_dh895xcc/Makefile b/drivers/crypto/intel/qat/qat_dh895xcc/Makefile index 38d6f8e1624a..cfd3bd757715 100644 --- a/drivers/crypto/intel/qat/qat_dh895xcc/Makefile +++ b/drivers/crypto/intel/qat/qat_dh895xcc/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_DH895xCC) += qat_dh895xcc.o qat_dh895xcc-objs := adf_drv.o adf_dh895xcc_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile b/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile index 0153c85ce743..64b54e92b2b4 100644 --- a/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile +++ b/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_DH895xCCVF) += qat_dh895xccvf.o qat_dh895xccvf-objs := adf_drv.o adf_dh895xccvf_hw_data.o diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 4536c8ad0e11..e568e89283c0 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -23,7 +23,7 @@ # Makefile for the drm device driver. This driver provides support for the # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher. -FULL_AMD_PATH=$(srctree)/$(src)/.. +FULL_AMD_PATH=$(src)/.. DISPLAY_FOLDER_NAME=display FULL_AMD_DISPLAY_PATH = $(FULL_AMD_PATH)/$(DISPLAY_FOLDER_NAME) diff --git a/drivers/gpu/drm/arm/display/komeda/Makefile b/drivers/gpu/drm/arm/display/komeda/Makefile index 1931a7fa1a14..cf5287fcbbc2 100644 --- a/drivers/gpu/drm/arm/display/komeda/Makefile +++ b/drivers/gpu/drm/arm/display/komeda/Makefile @@ -1,8 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 ccflags-y := \ - -I $(srctree)/$(src)/../include \ - -I $(srctree)/$(src) + -I $(src)/../include \ + -I $(src) komeda-y := \ komeda_drv.o \ diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index fba73c38e235..2220448ed07f 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -41,7 +41,7 @@ CFLAGS_display/intel_fbdev.o = -Wno-override-init # drivers. Define I915 when building i915. subdir-ccflags-y += -DI915 -subdir-ccflags-y += -I$(srctree)/$(src) +subdir-ccflags-y += -I$(src) # Please keep these build lists sorted! @@ -434,7 +434,7 @@ no-header-test := \ always-$(CONFIG_DRM_I915_WERROR) += \ $(patsubst %.h,%.hdrtest, $(filter-out $(no-header-test), \ - $(shell cd $(srctree)/$(src) && find * -name '*.h'))) + $(shell cd $(src) && find * -name '*.h'))) quiet_cmd_hdrtest = HDRTEST $(patsubst %.hdrtest,%.h,$@) cmd_hdrtest = $(CC) $(filter-out $(CFLAGS_GCOV), $(c_flags)) -S -o /dev/null -x c /dev/null -include $<; \ diff --git a/drivers/gpu/drm/imagination/Makefile b/drivers/gpu/drm/imagination/Makefile index ec6db8e9b403..9bc6a3884c22 100644 --- a/drivers/gpu/drm/imagination/Makefile +++ b/drivers/gpu/drm/imagination/Makefile @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only OR MIT # Copyright (c) 2023 Imagination Technologies Ltd. -subdir-ccflags-y := -I$(srctree)/$(src) +subdir-ccflags-y := -I$(src) powervr-y := \ pvr_ccb.o \ diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile index b21ae2880c71..b8cc007fc1b9 100644 --- a/drivers/gpu/drm/msm/Makefile +++ b/drivers/gpu/drm/msm/Makefile @@ -1,8 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 -ccflags-y := -I $(srctree)/$(src) -ccflags-y += -I $(srctree)/$(src)/disp/dpu1 -ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(srctree)/$(src)/dsi -ccflags-$(CONFIG_DRM_MSM_DP) += -I $(srctree)/$(src)/dp +ccflags-y := -I $(src) +ccflags-y += -I $(src)/disp/dpu1 +ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(src)/dsi +ccflags-$(CONFIG_DRM_MSM_DP) += -I $(src)/dp msm-y := \ adreno/adreno_device.o \ diff --git a/drivers/gpu/drm/nouveau/Kbuild b/drivers/gpu/drm/nouveau/Kbuild index cf6b3a80c0c8..c32c01827c1d 100644 --- a/drivers/gpu/drm/nouveau/Kbuild +++ b/drivers/gpu/drm/nouveau/Kbuild @@ -1,10 +1,8 @@ -NOUVEAU_PATH ?= $(srctree) - # SPDX-License-Identifier: MIT -ccflags-y += -I $(NOUVEAU_PATH)/$(src)/include -ccflags-y += -I $(NOUVEAU_PATH)/$(src)/include/nvkm -ccflags-y += -I $(NOUVEAU_PATH)/$(src)/nvkm -ccflags-y += -I $(NOUVEAU_PATH)/$(src) +ccflags-y += -I $(src)/include +ccflags-y += -I $(src)/include/nvkm +ccflags-y += -I $(src)/nvkm +ccflags-y += -I $(src) # NVKM - HW resource manager #- code also used by various userspace tools/tests diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index c29a850859ad..8162a522222b 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -32,7 +32,7 @@ endif # Enable -Werror in CI and development subdir-ccflags-$(CONFIG_DRM_XE_WERROR) += -Werror -subdir-ccflags-y += -I$(obj) -I$(srctree)/$(src) +subdir-ccflags-y += -I$(obj) -I$(src) # generated sources hostprogs := xe_gen_wa_oob @@ -43,7 +43,7 @@ quiet_cmd_wa_oob = GEN $(notdir $(generated_oob)) cmd_wa_oob = mkdir -p $(@D); $^ $(generated_oob) $(obj)/generated/%_wa_oob.c $(obj)/generated/%_wa_oob.h: $(obj)/xe_gen_wa_oob \ - $(srctree)/$(src)/xe_wa_oob.rules + $(src)/xe_wa_oob.rules $(call cmd,wa_oob) uses_generated_oob := \ @@ -166,8 +166,8 @@ endif # i915 Display compat #defines and #includes subdir-ccflags-$(CONFIG_DRM_XE_DISPLAY) += \ - -I$(srctree)/$(src)/display/ext \ - -I$(srctree)/$(src)/compat-i915-headers \ + -I$(src)/display/ext \ + -I$(src)/compat-i915-headers \ -I$(srctree)/drivers/gpu/drm/i915/display/ \ -Ddrm_i915_gem_object=xe_bo \ -Ddrm_i915_private=xe_device @@ -310,7 +310,7 @@ ifneq ($(CONFIG_DRM_XE_DISPLAY),y) endif always-$(CONFIG_DRM_XE_WERROR) += \ - $(patsubst %.h,%.hdrtest, $(shell cd $(srctree)/$(src) && find * -name '*.h' $(hdrtest_find_args))) + $(patsubst %.h,%.hdrtest, $(shell cd $(src) && find * -name '*.h' $(hdrtest_find_args))) quiet_cmd_hdrtest = HDRTEST $(patsubst %.hdrtest,%.h,$@) cmd_hdrtest = $(CC) -DHDRTEST $(filter-out $(CFLAGS_GCOV), $(c_flags)) -S -o /dev/null -x c /dev/null -include $<; touch $@ diff --git a/drivers/hid/amd-sfh-hid/Makefile b/drivers/hid/amd-sfh-hid/Makefile index 0222170ab7ad..106514b54d16 100644 --- a/drivers/hid/amd-sfh-hid/Makefile +++ b/drivers/hid/amd-sfh-hid/Makefile @@ -13,4 +13,4 @@ amd_sfh-objs += sfh1_1/amd_sfh_init.o amd_sfh-objs += sfh1_1/amd_sfh_interface.o amd_sfh-objs += sfh1_1/amd_sfh_desc.o -ccflags-y += -I $(srctree)/$(src)/ +ccflags-y += -I $(src)/ diff --git a/drivers/hid/intel-ish-hid/Makefile b/drivers/hid/intel-ish-hid/Makefile index f0a82b1c7cb9..db4974c43431 100644 --- a/drivers/hid/intel-ish-hid/Makefile +++ b/drivers/hid/intel-ish-hid/Makefile @@ -23,4 +23,4 @@ intel-ishtp-hid-objs += ishtp-hid-client.o obj-$(CONFIG_INTEL_ISH_FIRMWARE_DOWNLOADER) += intel-ishtp-loader.o intel-ishtp-loader-objs += ishtp-fw-loader.o -ccflags-y += -I $(srctree)/$(src)/ishtp +ccflags-y += -I $(src)/ishtp diff --git a/drivers/md/dm-vdo/Makefile b/drivers/md/dm-vdo/Makefile index 33e09abc6acd..9476957bfbf4 100644 --- a/drivers/md/dm-vdo/Makefile +++ b/drivers/md/dm-vdo/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I$(srctree)/$(src) -I$(srctree)/$(src)/indexer +ccflags-y := -I$(src) -I$(src)/indexer obj-$(CONFIG_DM_VDO) += dm-vdo.o diff --git a/drivers/net/ethernet/aquantia/atlantic/Makefile b/drivers/net/ethernet/aquantia/atlantic/Makefile index 8ebcc68e807f..f6a96931c89a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/Makefile +++ b/drivers/net/ethernet/aquantia/atlantic/Makefile @@ -8,7 +8,7 @@ obj-$(CONFIG_AQTION) += atlantic.o -ccflags-y += -I$(srctree)/$(src) +ccflags-y += -I$(src) atlantic-objs := aq_main.o \ aq_nic.o \ diff --git a/drivers/net/ethernet/chelsio/libcxgb/Makefile b/drivers/net/ethernet/chelsio/libcxgb/Makefile index aa79264e72ba..fbedc31674b3 100644 --- a/drivers/net/ethernet/chelsio/libcxgb/Makefile +++ b/drivers/net/ethernet/chelsio/libcxgb/Makefile @@ -1,5 +1,5 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../cxgb4 +ccflags-y := -I $(src)/../cxgb4 obj-$(CONFIG_CHELSIO_LIB) += libcxgb.o diff --git a/drivers/net/ethernet/fungible/funeth/Makefile b/drivers/net/ethernet/fungible/funeth/Makefile index 646d69595b4f..d51e4c2b4a1a 100644 --- a/drivers/net/ethernet/fungible/funeth/Makefile +++ b/drivers/net/ethernet/fungible/funeth/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) -ccflags-y += -I$(srctree)/$(src)/../funcore -I$(srctree)/$(src) +ccflags-y += -I$(src)/../funcore -I$(src) obj-$(CONFIG_FUN_ETH) += funeth.o diff --git a/drivers/net/ethernet/hisilicon/hns3/Makefile b/drivers/net/ethernet/hisilicon/hns3/Makefile index e214bfaece1f..8e9293e57bfd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/Makefile +++ b/drivers/net/ethernet/hisilicon/hns3/Makefile @@ -3,7 +3,7 @@ # Makefile for the HISILICON network device drivers. # -ccflags-y += -I$(srctree)/$(src) +ccflags-y += -I$(src) ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3pf ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3vf ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3_common diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile index dc6d27a36faa..e5ca0f511822 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile @@ -6,8 +6,8 @@ # ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/../include + -I $(src) \ + -I $(src)/../include obj-$(CONFIG_BRCMFMAC) += brcmfmac.o brcmfmac-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile index 46098705e236..5e37c638f966 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile @@ -3,9 +3,9 @@ # Copyright (c) 2022 Broadcom Corporation ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/.. \ - -I $(srctree)/$(src)/../../include + -I $(src) \ + -I $(src)/.. \ + -I $(src)/../../include obj-m += brcmfmac-bca.o brcmfmac-bca-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile index 5e1fddaff79e..33e86724ba14 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile @@ -3,9 +3,9 @@ # Copyright (c) 2022 Broadcom Corporation ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/.. \ - -I $(srctree)/$(src)/../../include + -I $(src) \ + -I $(src)/.. \ + -I $(src)/../../include obj-m += brcmfmac-cyw.o brcmfmac-cyw-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile index 7f455a19a2b1..3db4cce44f42 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile @@ -3,9 +3,9 @@ # Copyright (c) 2022 Broadcom Corporation ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/.. \ - -I $(srctree)/$(src)/../../include + -I $(src) \ + -I $(src)/.. \ + -I $(src)/../../include obj-m += brcmfmac-wcc.o brcmfmac-wcc-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile index 090757730ba6..ffdd17eabe1d 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile @@ -16,9 +16,9 @@ # CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. ccflags-y := \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/phy \ - -I $(srctree)/$(src)/../include + -I $(src) \ + -I $(src)/phy \ + -I $(src)/../include brcmsmac-y := \ mac80211_if.o \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile index 7a82d615ba2a..f9b40cab766c 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile @@ -4,7 +4,7 @@ # # Copyright (c) 2011 Broadcom Corporation # -ccflags-y := -I $(srctree)/$(src)/../include +ccflags-y := -I $(src)/../include obj-$(CONFIG_BRCMUTIL) += brcmutil.o brcmutil-objs = utils.o d11.o diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/Makefile b/drivers/net/wireless/intel/iwlwifi/dvm/Makefile index 0486b17d7c41..6109d64006db 100644 --- a/drivers/net/wireless/intel/iwlwifi/dvm/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/dvm/Makefile @@ -11,4 +11,4 @@ iwldvm-objs += rxon.o devices.o iwldvm-$(CONFIG_IWLWIFI_LEDS) += led.o iwldvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o -ccflags-y += -I $(srctree)/$(src)/../ +ccflags-y += -I $(src)/../ diff --git a/drivers/net/wireless/intel/iwlwifi/mei/Makefile b/drivers/net/wireless/intel/iwlwifi/mei/Makefile index 8e3ef0347db7..e05f9421be4a 100644 --- a/drivers/net/wireless/intel/iwlwifi/mei/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/mei/Makefile @@ -5,4 +5,4 @@ iwlmei-y += net.o iwlmei-$(CONFIG_IWLWIFI_DEVICE_TRACING) += trace.o CFLAGS_trace.o := -I$(src) -ccflags-y += -I $(srctree)/$(src)/../ +ccflags-y += -I $(src)/../ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile index 593fe28d89cf..764ba73cde1e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile @@ -15,4 +15,4 @@ iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o iwlmvm-$(CONFIG_PM) += d3.o iwlmvm-$(CONFIG_IWLMEI) += vendor-cmd.o -ccflags-y += -I $(srctree)/$(src)/../ +ccflags-y += -I $(src)/../ diff --git a/drivers/net/wireless/intel/iwlwifi/tests/Makefile b/drivers/net/wireless/intel/iwlwifi/tests/Makefile index 5658471bdf0a..84491488f589 100644 --- a/drivers/net/wireless/intel/iwlwifi/tests/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/tests/Makefile @@ -2,6 +2,6 @@ iwlwifi-tests-y += module.o devinfo.o -ccflags-y += -I$(srctree)/$(src)/../ +ccflags-y += -I$(src)/../ obj-$(CONFIG_IWLWIFI_KUNIT_TESTS) += iwlwifi-tests.o diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile b/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile index 565a9a114134..1044105c2557 100644 --- a/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile +++ b/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile @@ -3,4 +3,4 @@ rtl818x_pci-objs := dev.o rtl8225.o sa2400.o max2820.o grf5101.o rtl8225se.o obj-$(CONFIG_RTL8180) += rtl818x_pci.o -ccflags-y += -I $(srctree)/$(src)/.. +ccflags-y += -I $(src)/.. diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile b/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile index 0bf64dfb233a..3deded2c05b6 100644 --- a/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile +++ b/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile @@ -3,4 +3,4 @@ rtl8187-objs := dev.o rtl8225.o leds.o rfkill.o obj-$(CONFIG_RTL8187) += rtl8187.o -ccflags-y += -I $(srctree)/$(src)/.. +ccflags-y += -I $(src)/.. diff --git a/drivers/scsi/aic7xxx/Makefile b/drivers/scsi/aic7xxx/Makefile index e0188ecd85b2..853c72a81ae0 100644 --- a/drivers/scsi/aic7xxx/Makefile +++ b/drivers/scsi/aic7xxx/Makefile @@ -55,9 +55,9 @@ aicasm-7xxx-opts-$(CONFIG_AIC7XXX_REG_PRETTY_PRINT) := \ ifeq ($(CONFIG_AIC7XXX_BUILD_FIRMWARE),y) $(obj)/aic7xxx_seq.h: $(src)/aic7xxx.seq $(src)/aic7xxx.reg $(obj)/aicasm/aicasm - $(obj)/aicasm/aicasm -I$(srctree)/$(src) -r $(obj)/aic7xxx_reg.h \ + $(obj)/aicasm/aicasm -I $(src) -r $(obj)/aic7xxx_reg.h \ $(aicasm-7xxx-opts-y) -o $(obj)/aic7xxx_seq.h \ - $(srctree)/$(src)/aic7xxx.seq + $(src)/aic7xxx.seq $(aic7xxx-gen-y): $(objtree)/$(obj)/aic7xxx_seq.h @true @@ -73,9 +73,9 @@ aicasm-79xx-opts-$(CONFIG_AIC79XX_REG_PRETTY_PRINT) := \ ifeq ($(CONFIG_AIC79XX_BUILD_FIRMWARE),y) $(obj)/aic79xx_seq.h: $(src)/aic79xx.seq $(src)/aic79xx.reg $(obj)/aicasm/aicasm - $(obj)/aicasm/aicasm -I$(srctree)/$(src) -r $(obj)/aic79xx_reg.h \ + $(obj)/aicasm/aicasm -I $(src) -r $(obj)/aic79xx_reg.h \ $(aicasm-79xx-opts-y) -o $(obj)/aic79xx_seq.h \ - $(srctree)/$(src)/aic79xx.seq + $(src)/aic79xx.seq $(aic79xx-gen-y): $(objtree)/$(obj)/aic79xx_seq.h @true @@ -83,5 +83,5 @@ else $(obj)/aic79xx_reg_print.c: $(src)/aic79xx_reg_print.c_shipped endif -$(obj)/aicasm/aicasm: $(srctree)/$(src)/aicasm/*.[chyl] - $(MAKE) -C $(srctree)/$(src)/aicasm OUTDIR=$(shell pwd)/$(obj)/aicasm/ +$(obj)/aicasm/aicasm: $(src)/aicasm/*.[chyl] + $(MAKE) -C $(src)/aicasm OUTDIR=$(shell pwd)/$(obj)/aicasm/ diff --git a/drivers/staging/rtl8723bs/Makefile b/drivers/staging/rtl8723bs/Makefile index 590bde02058c..7f5067e89295 100644 --- a/drivers/staging/rtl8723bs/Makefile +++ b/drivers/staging/rtl8723bs/Makefile @@ -62,4 +62,4 @@ r8723bs-y = \ obj-$(CONFIG_RTL8723BS) := r8723bs.o -ccflags-y += -I$(srctree)/$(src)/include -I$(srctree)/$(src)/hal +ccflags-y += -I$(src)/include -I$(src)/hal diff --git a/fs/iomap/Makefile b/fs/iomap/Makefile index fc070184b7fa..381d76c5c232 100644 --- a/fs/iomap/Makefile +++ b/fs/iomap/Makefile @@ -4,7 +4,7 @@ # All Rights Reserved. # -ccflags-y += -I $(srctree)/$(src) # needed for trace events +ccflags-y += -I $(src) # needed for trace events obj-$(CONFIG_FS_IOMAP) += iomap.o diff --git a/fs/unicode/Makefile b/fs/unicode/Makefile index 0e51c0025a16..e309afe2b2bb 100644 --- a/fs/unicode/Makefile +++ b/fs/unicode/Makefile @@ -18,13 +18,13 @@ ifdef REGENERATE_UTF8DATA quiet_cmd_utf8data = GEN $@ cmd_utf8data = $< \ - -a $(srctree)/$(src)/DerivedAge.txt \ - -c $(srctree)/$(src)/DerivedCombiningClass.txt \ - -p $(srctree)/$(src)/DerivedCoreProperties.txt \ - -d $(srctree)/$(src)/UnicodeData.txt \ - -f $(srctree)/$(src)/CaseFolding.txt \ - -n $(srctree)/$(src)/NormalizationCorrections.txt \ - -t $(srctree)/$(src)/NormalizationTest.txt \ + -a $(src)/DerivedAge.txt \ + -c $(src)/DerivedCombiningClass.txt \ + -p $(src)/DerivedCoreProperties.txt \ + -d $(src)/UnicodeData.txt \ + -f $(src)/CaseFolding.txt \ + -n $(src)/NormalizationCorrections.txt \ + -t $(src)/NormalizationTest.txt \ -o $@ $(obj)/utf8data.c: $(obj)/mkutf8data $(filter %.txt, $(cmd_utf8data)) FORCE diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 76674ad5833e..c5a35e32adf0 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -4,8 +4,8 @@ # All Rights Reserved. # -ccflags-y += -I $(srctree)/$(src) # needed for trace events -ccflags-y += -I $(srctree)/$(src)/libxfs +ccflags-y += -I $(src) # needed for trace events +ccflags-y += -I $(src)/libxfs obj-$(CONFIG_XFS_FS) += xfs.o diff --git a/init/Makefile b/init/Makefile index cbac576c57d6..3c48d97538c1 100644 --- a/init/Makefile +++ b/init/Makefile @@ -52,7 +52,7 @@ CFLAGS_version.o := -include $(obj)/utsversion-tmp.h # Build version-timestamp.c with final UTS_VERSION # -include/generated/utsversion.h: build-version-auto = $(shell $(srctree)/$(src)/build-version) +include/generated/utsversion.h: build-version-auto = $(shell $(src)/build-version) include/generated/utsversion.h: build-timestamp-auto = $(shell LC_ALL=C date) include/generated/utsversion.h: FORCE $(call filechk,uts_version) diff --git a/lib/Makefile b/lib/Makefile index ffc6b2341b45..27baa6e6daa8 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -352,7 +352,7 @@ $(obj)/oid_registry_data.c: $(srctree)/include/linux/oid_registry.h \ $(call cmd,build_OID_registry) quiet_cmd_build_OID_registry = GEN $@ - cmd_build_OID_registry = perl $(srctree)/$(src)/build_OID_registry $< $@ + cmd_build_OID_registry = perl $(src)/build_OID_registry $< $@ clean-files += oid_registry_data.c @@ -412,8 +412,8 @@ obj-$(CONFIG_GENERIC_LIB_DEVMEM_IS_ALLOWED) += devmem_is_allowed.o obj-$(CONFIG_FIRMWARE_TABLE) += fw_table.o # FORTIFY_SOURCE compile-time behavior tests -TEST_FORTIFY_SRCS = $(wildcard $(srctree)/$(src)/test_fortify/*-*.c) -TEST_FORTIFY_LOGS = $(patsubst $(srctree)/$(src)/%.c, %.log, $(TEST_FORTIFY_SRCS)) +TEST_FORTIFY_SRCS = $(wildcard $(src)/test_fortify/*-*.c) +TEST_FORTIFY_LOGS = $(patsubst $(src)/%.c, %.log, $(TEST_FORTIFY_SRCS)) TEST_FORTIFY_LOG = test_fortify.log quiet_cmd_test_fortify = TEST $@ diff --git a/lib/raid6/Makefile b/lib/raid6/Makefile index 385a94aa0b99..8785353c6140 100644 --- a/lib/raid6/Makefile +++ b/lib/raid6/Makefile @@ -53,7 +53,7 @@ endif endif quiet_cmd_unroll = UNROLL $@ - cmd_unroll = $(AWK) -v N=$* -f $(srctree)/$(src)/unroll.awk < $< > $@ + cmd_unroll = $(AWK) -v N=$* -f $(src)/unroll.awk < $< > $@ targets += int1.c int2.c int4.c int8.c $(obj)/int%.c: $(src)/int.uc $(src)/unroll.awk FORCE diff --git a/net/wireless/Makefile b/net/wireless/Makefile index 72074fd36df4..1d49cc8b6da1 100644 --- a/net/wireless/Makefile +++ b/net/wireless/Makefile @@ -25,7 +25,7 @@ ifneq ($(CONFIG_CFG80211_EXTRA_REGDB_KEYDIR),) cfg80211-y += extra-certs.o endif -$(obj)/shipped-certs.c: $(sort $(wildcard $(srctree)/$(src)/certs/*.hex)) +$(obj)/shipped-certs.c: $(sort $(wildcard $(src)/certs/*.hex)) @$(kecho) " GEN $@" $(Q)(echo '#include "reg.h"'; \ echo 'const u8 shipped_regdb_certs[] = {'; \ diff --git a/rust/Makefile b/rust/Makefile index 846e6ab9d5a9..85e46dde2a5b 100644 --- a/rust/Makefile +++ b/rust/Makefile @@ -239,7 +239,7 @@ quiet_cmd_rustsysroot = RUSTSYSROOT rm -rf $(objtree)/$(obj)/test; \ mkdir -p $(objtree)/$(obj)/test; \ cp -a $(rustc_sysroot) $(objtree)/$(obj)/test/sysroot; \ - cp -r $(srctree)/$(src)/alloc/* \ + cp -r $(src)/alloc/* \ $(objtree)/$(obj)/test/sysroot/lib/rustlib/src/rust/library/alloc/src; \ echo '\#!/bin/sh' > $(objtree)/$(obj)/test/rustc_sysroot; \ echo "$(RUSTC) --sysroot=$(abspath $(objtree)/$(obj)/test/sysroot) \"\$$@\"" \ @@ -340,7 +340,7 @@ quiet_cmd_bindgen = BINDGEN $@ $(bindgen_target_cflags) $(bindgen_target_extra) $(obj)/bindings/bindings_generated.rs: private bindgen_target_flags = \ - $(shell grep -Ev '^#|^$$' $(srctree)/$(src)/bindgen_parameters) + $(shell grep -Ev '^#|^$$' $(src)/bindgen_parameters) $(obj)/bindings/bindings_generated.rs: private bindgen_target_extra = ; \ sed -Ei 's/pub const RUST_CONST_HELPER_([a-zA-Z0-9_]*)/pub const \1/g' $@ $(obj)/bindings/bindings_generated.rs: $(src)/bindings/bindings_helper.h \ @@ -348,7 +348,7 @@ $(obj)/bindings/bindings_generated.rs: $(src)/bindings/bindings_helper.h \ $(call if_changed_dep,bindgen) $(obj)/uapi/uapi_generated.rs: private bindgen_target_flags = \ - $(shell grep -Ev '^#|^$$' $(srctree)/$(src)/bindgen_parameters) + $(shell grep -Ev '^#|^$$' $(src)/bindgen_parameters) $(obj)/uapi/uapi_generated.rs: $(src)/uapi/uapi_helper.h \ $(src)/bindgen_parameters FORCE $(call if_changed_dep,bindgen) diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile index 933f6c3fe6b0..f054e0828f13 100644 --- a/samples/bpf/Makefile +++ b/samples/bpf/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 -BPF_SAMPLES_PATH ?= $(abspath $(srctree)/$(src)) +BPF_SAMPLES_PATH ?= $(abspath $(src)) TOOLS_PATH := $(BPF_SAMPLES_PATH)/../../tools pound := \# diff --git a/samples/hid/Makefile b/samples/hid/Makefile index 9f7fe29dd749..c128ccd49974 100644 --- a/samples/hid/Makefile +++ b/samples/hid/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 -HID_SAMPLES_PATH ?= $(abspath $(srctree)/$(src)) +HID_SAMPLES_PATH ?= $(abspath $(src)) TOOLS_PATH := $(HID_SAMPLES_PATH)/../../tools pound := \# diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include index 2f331879816b..faf37bafa3f8 100644 --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -62,8 +62,7 @@ stringify = $(squote)$(quote)$1$(quote)$(squote) ### # The path to Kbuild or Makefile. Kbuild has precedence over Makefile. -kbuild-dir = $(if $(filter /%,$(src)),$(src),$(srctree)/$(src)) -kbuild-file = $(or $(wildcard $(kbuild-dir)/Kbuild),$(kbuild-dir)/Makefile) +kbuild-file = $(or $(wildcard $(src)/Kbuild),$(src)/Makefile) ### # Read a file, replacing newlines with spaces diff --git a/scripts/Makefile.asm-generic b/scripts/Makefile.asm-generic index 8d01b37b7677..1486abf34c7c 100644 --- a/scripts/Makefile.asm-generic +++ b/scripts/Makefile.asm-generic @@ -9,7 +9,7 @@ PHONY := all all: -src := $(subst /generated,,$(obj)) +src := $(srctree)/$(subst /generated,,$(obj)) include $(srctree)/scripts/Kbuild.include -include $(kbuild-file) @@ -20,14 +20,14 @@ include $(srctree)/$(generic)/Kbuild endif redundant := $(filter $(mandatory-y) $(generated-y), $(generic-y)) -redundant += $(foreach f, $(generic-y), $(if $(wildcard $(srctree)/$(src)/$(f)),$(f))) +redundant += $(foreach f, $(generic-y), $(if $(wildcard $(src)/$(f)),$(f))) redundant := $(sort $(redundant)) $(if $(redundant),\ $(warning redundant generic-y found in $(src)/Kbuild: $(redundant))) # If arch does not implement mandatory headers, fallback to asm-generic ones. mandatory-y := $(filter-out $(generated-y), $(mandatory-y)) -generic-y += $(foreach f, $(mandatory-y), $(if $(wildcard $(srctree)/$(src)/$(f)),,$(f))) +generic-y += $(foreach f, $(mandatory-y), $(if $(wildcard $(src)/$(f)),,$(f))) generic-y := $(addprefix $(obj)/, $(generic-y)) generated-y := $(addprefix $(obj)/, $(generated-y)) diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 41138346afb0..43b71c3d0ff6 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -3,7 +3,7 @@ # Building # ========================================================================== -src := $(obj) +src := $(if $(VPATH),$(VPATH)/)$(obj) PHONY := $(obj)/ $(obj)/: diff --git a/scripts/Makefile.clean b/scripts/Makefile.clean index f2cb4d7ffd96..4fcfab40ed61 100644 --- a/scripts/Makefile.clean +++ b/scripts/Makefile.clean @@ -3,7 +3,7 @@ # Cleaning up # ========================================================================== -src := $(obj) +src := $(if $(VPATH),$(VPATH)/)$(obj) PHONY := __clean __clean: diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index 42d3c772c8e0..08d42e93bea0 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -209,13 +209,13 @@ _c_flags += $(if $(patsubst n%,, \ -D__KCSAN_INSTRUMENT_BARRIERS__) endif -# $(srctree)/$(src) for including checkin headers from generated source files -# $(objtree)/$(obj) for including generated headers from checkin source files +# $(src) for including checkin headers from generated source files +# $(obj) for including generated headers from checkin source files ifeq ($(KBUILD_EXTMOD),) ifdef building_out_of_srctree -_c_flags += $(addprefix -I $(srctree)/,$(src)) $(addprefix -I $(objtree)/,$(obj)) -_a_flags += $(addprefix -I $(srctree)/,$(src)) $(addprefix -I $(objtree)/,$(obj)) -_cpp_flags += $(addprefix -I $(srctree)/,$(src)) $(addprefix -I $(objtree)/,$(obj)) +_c_flags += $(addprefix -I, $(src) $(obj)) +_a_flags += $(addprefix -I, $(src) $(obj)) +_cpp_flags += $(addprefix -I, $(src) $(obj)) endif endif diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost index 739402f45509..68c4f9ee0c90 100644 --- a/scripts/Makefile.modpost +++ b/scripts/Makefile.modpost @@ -112,7 +112,7 @@ else # set src + obj - they may be used in the modules's Makefile obj := $(KBUILD_EXTMOD) -src := $(obj) +src := $(if $(VPATH),$(VPATH)/)$(obj) # Include the module's Makefile to find KBUILD_EXTRA_SYMBOLS include $(kbuild-file) diff --git a/scripts/dtc/Makefile b/scripts/dtc/Makefile index 4d32b9497da9..a18657072541 100644 --- a/scripts/dtc/Makefile +++ b/scripts/dtc/Makefile @@ -16,12 +16,12 @@ libfdt = $(addprefix libfdt/,$(libfdt-objs)) fdtoverlay-objs := $(libfdt) fdtoverlay.o util.o # Source files need to get at the userspace version of libfdt_env.h to compile -HOST_EXTRACFLAGS += -I $(srctree)/$(src)/libfdt +HOST_EXTRACFLAGS += -I $(src)/libfdt HOST_EXTRACFLAGS += -DNO_YAML # Generated files need one more search path to include headers in source tree -HOSTCFLAGS_dtc-lexer.lex.o := -I $(srctree)/$(src) -HOSTCFLAGS_dtc-parser.tab.o := -I $(srctree)/$(src) +HOSTCFLAGS_dtc-lexer.lex.o := -I $(src) +HOSTCFLAGS_dtc-parser.tab.o := -I $(src) # dependencies on generated files need to be listed explicitly $(obj)/dtc-lexer.lex.o: $(obj)/dtc-parser.tab.h diff --git a/scripts/gdb/linux/Makefile b/scripts/gdb/linux/Makefile index 48941faa6ea6..d77ad9079d0f 100644 --- a/scripts/gdb/linux/Makefile +++ b/scripts/gdb/linux/Makefile @@ -2,7 +2,7 @@ ifdef building_out_of_srctree -symlinks := $(patsubst $(srctree)/$(src)/%,%,$(wildcard $(srctree)/$(src)/*.py)) +symlinks := $(patsubst $(src)/%,%,$(wildcard $(src)/*.py)) quiet_cmd_symlink = SYMLINK $@ cmd_symlink = ln -fsn $(patsubst $(obj)/%,$(abspath $(srctree))/$(src)/%,$@) $@ diff --git a/scripts/genksyms/Makefile b/scripts/genksyms/Makefile index d6a422a63b6a..312edccda736 100644 --- a/scripts/genksyms/Makefile +++ b/scripts/genksyms/Makefile @@ -23,8 +23,8 @@ $(obj)/pars%.tab.c $(obj)/pars%.tab.h: $(src)/pars%.y FORCE endif # -I needed for generated C source to include headers in source tree -HOSTCFLAGS_parse.tab.o := -I $(srctree)/$(src) -HOSTCFLAGS_lex.lex.o := -I $(srctree)/$(src) +HOSTCFLAGS_parse.tab.o := -I $(src) +HOSTCFLAGS_lex.lex.o := -I $(src) # dependencies on generated files need to be listed explicitly $(obj)/lex.lex.o: $(obj)/parse.tab.h diff --git a/scripts/kconfig/Makefile b/scripts/kconfig/Makefile index ea1bf3b3dbde..a0a0be38cbdc 100644 --- a/scripts/kconfig/Makefile +++ b/scripts/kconfig/Makefile @@ -57,7 +57,7 @@ $(foreach c, config menuconfig nconfig gconfig xconfig, $(eval $(call config_rul PHONY += localmodconfig localyesconfig localyesconfig localmodconfig: $(obj)/conf - $(Q)$(PERL) $(srctree)/$(src)/streamline_config.pl --$@ $(srctree) $(Kconfig) > .tmp.config + $(Q)$(PERL) $(src)/streamline_config.pl --$@ $(srctree) $(Kconfig) > .tmp.config $(Q)if [ -f .config ]; then \ cmp -s .tmp.config .config || \ (mv -f .config .config.old.1; \ @@ -118,7 +118,7 @@ tinyconfig: # CHECK: -o cache_dir= working? PHONY += testconfig testconfig: $(obj)/conf - $(Q)$(PYTHON3) -B -m pytest $(srctree)/$(src)/tests \ + $(Q)$(PYTHON3) -B -m pytest $(src)/tests \ -o cache_dir=$(abspath $(obj)/tests/.cache) \ $(if $(findstring 1,$(KBUILD_VERBOSE)),--capture=no) clean-files += tests/.cache @@ -165,8 +165,8 @@ common-objs := confdata.o expr.o lexer.lex.o menu.o parser.tab.o \ preprocess.o symbol.o util.o $(obj)/lexer.lex.o: $(obj)/parser.tab.h -HOSTCFLAGS_lexer.lex.o := -I $(srctree)/$(src) -HOSTCFLAGS_parser.tab.o := -I $(srctree)/$(src) +HOSTCFLAGS_lexer.lex.o := -I $(src) +HOSTCFLAGS_parser.tab.o := -I $(src) # conf: Used for defconfig, oldconfig and related targets hostprogs += conf diff --git a/security/tomoyo/Makefile b/security/tomoyo/Makefile index 884ff155edc3..55c67b9846a9 100644 --- a/security/tomoyo/Makefile +++ b/security/tomoyo/Makefile @@ -11,7 +11,7 @@ quiet_cmd_policy = POLICY $@ printf '\t"";\n';) \ } > $@ -$(obj)/builtin-policy.h: $(wildcard $(obj)/policy/*.conf $(srctree)/$(src)/policy/*.conf.default) FORCE +$(obj)/builtin-policy.h: $(wildcard $(obj)/policy/*.conf $(src)/policy/*.conf.default) FORCE $(call if_changed,policy) ifndef CONFIG_SECURITY_TOMOYO_INSECURE_BUILTIN_SETTING diff --git a/usr/Makefile b/usr/Makefile index f8e1ad19e05c..132ef7e96e6d 100644 --- a/usr/Makefile +++ b/usr/Makefile @@ -22,7 +22,7 @@ cpio-data := # If CONFIG_INITRAMFS_SOURCE is empty, generate a small initramfs with the # default contents. ifeq ($(ramfs-input),) -ramfs-input := $(srctree)/$(src)/default_cpio_list +ramfs-input := $(src)/default_cpio_list endif ifeq ($(words $(ramfs-input)),1) diff --git a/usr/include/Makefile b/usr/include/Makefile index 338c81f1fcf3..771e32872b2a 100644 --- a/usr/include/Makefile +++ b/usr/include/Makefile @@ -78,7 +78,7 @@ quiet_cmd_hdrtest = HDRTEST $< cmd_hdrtest = \ $(CC) $(c_flags) -fsyntax-only -x c /dev/null \ $(if $(filter-out $(no-header-test), $*.h), -include $< -include $<); \ - $(PERL) $(srctree)/$(src)/headers_check.pl $(obj) $(SRCARCH) $<; \ + $(PERL) $(src)/headers_check.pl $(obj) $(SRCARCH) $<; \ touch $@ $(obj)/%.hdrtest: $(obj)/%.h FORCE -- cgit v1.2.3-59-g8ed1b From b6fb4269e707bb410e202fb160c8d7a188ef60f8 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 01:42:10 -0400 Subject: bcachefs: for_each_bset() declares loop iter Signed-off-by: Kent Overstreet --- fs/bcachefs/bset.c | 29 ++++++++++------------------- fs/bcachefs/bset.h | 6 ++++-- fs/bcachefs/btree_cache.c | 3 --- fs/bcachefs/btree_io.c | 7 ------- fs/bcachefs/btree_io.h | 2 -- fs/bcachefs/btree_update_interior.c | 1 - 6 files changed, 14 insertions(+), 34 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/bset.c b/fs/bcachefs/bset.c index 3bb477840eab..575e1d0b6eeb 100644 --- a/fs/bcachefs/bset.c +++ b/fs/bcachefs/bset.c @@ -103,8 +103,6 @@ void bch2_dump_bset(struct bch_fs *c, struct btree *b, void bch2_dump_btree_node(struct bch_fs *c, struct btree *b) { - struct bset_tree *t; - console_lock(); for_each_bset(b, t) bch2_dump_bset(c, b, bset(b, t), t - b->set); @@ -136,7 +134,6 @@ void bch2_dump_btree_node_iter(struct btree *b, struct btree_nr_keys bch2_btree_node_count_keys(struct btree *b) { - struct bset_tree *t; struct bkey_packed *k; struct btree_nr_keys nr = {}; @@ -198,7 +195,6 @@ void bch2_btree_node_iter_verify(struct btree_node_iter *iter, { struct btree_node_iter_set *set, *s2; struct bkey_packed *k, *p; - struct bset_tree *t; if (bch2_btree_node_iter_end(iter)) return; @@ -213,12 +209,14 @@ void bch2_btree_node_iter_verify(struct btree_node_iter *iter, /* Verify that set->end is correct: */ btree_node_iter_for_each(iter, set) { for_each_bset(b, t) - if (set->end == t->end_offset) + if (set->end == t->end_offset) { + BUG_ON(set->k < btree_bkey_first_offset(t) || + set->k >= t->end_offset); goto found; + } BUG(); found: - BUG_ON(set->k < btree_bkey_first_offset(t) || - set->k >= t->end_offset); + do {} while (0); } /* Verify iterator is sorted: */ @@ -377,11 +375,9 @@ static struct bkey_float *bkey_float(const struct btree *b, return ro_aux_tree_base(b, t)->f + idx; } -static void bset_aux_tree_verify(const struct btree *b) +static void bset_aux_tree_verify(struct btree *b) { #ifdef CONFIG_BCACHEFS_DEBUG - const struct bset_tree *t; - for_each_bset(b, t) { if (t->aux_data_offset == U16_MAX) continue; @@ -685,20 +681,20 @@ static __always_inline void make_bfloat(struct btree *b, struct bset_tree *t, } /* bytes remaining - only valid for last bset: */ -static unsigned __bset_tree_capacity(const struct btree *b, const struct bset_tree *t) +static unsigned __bset_tree_capacity(struct btree *b, const struct bset_tree *t) { bset_aux_tree_verify(b); return btree_aux_data_bytes(b) - t->aux_data_offset * sizeof(u64); } -static unsigned bset_ro_tree_capacity(const struct btree *b, const struct bset_tree *t) +static unsigned bset_ro_tree_capacity(struct btree *b, const struct bset_tree *t) { return __bset_tree_capacity(b, t) / (sizeof(struct bkey_float) + sizeof(u8)); } -static unsigned bset_rw_tree_capacity(const struct btree *b, const struct bset_tree *t) +static unsigned bset_rw_tree_capacity(struct btree *b, const struct bset_tree *t) { return __bset_tree_capacity(b, t) / sizeof(struct rw_aux_tree); } @@ -1374,8 +1370,6 @@ void bch2_btree_node_iter_init(struct btree_node_iter *iter, void bch2_btree_node_iter_init_from_start(struct btree_node_iter *iter, struct btree *b) { - struct bset_tree *t; - memset(iter, 0, sizeof(*iter)); for_each_bset(b, t) @@ -1481,7 +1475,6 @@ struct bkey_packed *bch2_btree_node_iter_prev_all(struct btree_node_iter *iter, { struct bkey_packed *k, *prev = NULL; struct btree_node_iter_set *set; - struct bset_tree *t; unsigned end = 0; if (bch2_expensive_debug_checks) @@ -1550,9 +1543,7 @@ struct bkey_s_c bch2_btree_node_iter_peek_unpack(struct btree_node_iter *iter, void bch2_btree_keys_stats(const struct btree *b, struct bset_stats *stats) { - const struct bset_tree *t; - - for_each_bset(b, t) { + for_each_bset_c(b, t) { enum bset_aux_tree_type type = bset_aux_tree_type(t); size_t j; diff --git a/fs/bcachefs/bset.h b/fs/bcachefs/bset.h index 120a79fd456b..5c6c7a14fa0f 100644 --- a/fs/bcachefs/bset.h +++ b/fs/bcachefs/bset.h @@ -206,7 +206,10 @@ static inline size_t btree_aux_data_u64s(const struct btree *b) } #define for_each_bset(_b, _t) \ - for (_t = (_b)->set; _t < (_b)->set + (_b)->nsets; _t++) + for (struct bset_tree *_t = (_b)->set; _t < (_b)->set + (_b)->nsets; _t++) + +#define for_each_bset_c(_b, _t) \ + for (const struct bset_tree *_t = (_b)->set; _t < (_b)->set + (_b)->nsets; _t++) #define bset_tree_for_each_key(_b, _t, _k) \ for (_k = btree_bkey_first(_b, _t); \ @@ -294,7 +297,6 @@ static inline struct bset_tree * bch2_bkey_to_bset_inlined(struct btree *b, struct bkey_packed *k) { unsigned offset = __btree_node_key_to_offset(b, k); - struct bset_tree *t; for_each_bset(b, t) if (offset <= t->end_offset) { diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 2c226e0cc263..7dd35d6224d9 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -881,7 +881,6 @@ static struct btree *__bch2_btree_node_get(struct btree_trans *trans, struct btr struct bch_fs *c = trans->c; struct btree_cache *bc = &c->btree_cache; struct btree *b; - struct bset_tree *t; bool need_relock = false; int ret; @@ -1001,7 +1000,6 @@ struct btree *bch2_btree_node_get(struct btree_trans *trans, struct btree_path * { struct bch_fs *c = trans->c; struct btree *b; - struct bset_tree *t; int ret; EBUG_ON(level >= BTREE_MAX_DEPTH); @@ -1078,7 +1076,6 @@ struct btree *bch2_btree_node_get_noiter(struct btree_trans *trans, struct bch_fs *c = trans->c; struct btree_cache *bc = &c->btree_cache; struct btree *b; - struct bset_tree *t; int ret; EBUG_ON(level >= BTREE_MAX_DEPTH); diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 9dcaccda7b93..e14c404d25c1 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -229,7 +229,6 @@ static bool should_compact_bset(struct btree *b, struct bset_tree *t, static bool bch2_drop_whiteouts(struct btree *b, enum compact_mode mode) { - struct bset_tree *t; bool ret = false; for_each_bset(b, t) { @@ -451,8 +450,6 @@ static bool btree_node_compact(struct bch_fs *c, struct btree *b) void bch2_btree_build_aux_trees(struct btree *b) { - struct bset_tree *t; - for_each_bset(b, t) bch2_bset_build_aux_tree(b, t, !bset_written(b, bset(b, t)) && @@ -637,8 +634,6 @@ fsck_err: __cold void bch2_btree_node_drop_keys_outside_node(struct btree *b) { - struct bset_tree *t; - for_each_bset(b, t) { struct bset *i = bset(b, t); struct bkey_packed *k; @@ -1987,7 +1982,6 @@ static void btree_write_submit(struct work_struct *work) void __bch2_btree_node_write(struct bch_fs *c, struct btree *b, unsigned flags) { struct btree_write_bio *wbio; - struct bset_tree *t; struct bset *i; struct btree_node *bn = NULL; struct btree_node_entry *bne = NULL; @@ -2244,7 +2238,6 @@ bool bch2_btree_post_write_cleanup(struct bch_fs *c, struct btree *b) { bool invalidated_iter = false; struct btree_node_entry *bne; - struct bset_tree *t; if (!btree_node_just_written(b)) return false; diff --git a/fs/bcachefs/btree_io.h b/fs/bcachefs/btree_io.h index e251cb6b965f..2b8b564fc560 100644 --- a/fs/bcachefs/btree_io.h +++ b/fs/bcachefs/btree_io.h @@ -81,8 +81,6 @@ static inline bool should_compact_bset_lazy(struct btree *b, static inline bool bch2_maybe_compact_whiteouts(struct bch_fs *c, struct btree *b) { - struct bset_tree *t; - for_each_bset(b, t) if (should_compact_bset_lazy(b, t)) return bch2_compact_whiteouts(c, b, COMPACT_LAZY); diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 80eb862a8962..60b8544cea48 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -160,7 +160,6 @@ topology_repair: static void __bch2_btree_calc_format(struct bkey_format_state *s, struct btree *b) { struct bkey_packed *k; - struct bset_tree *t; struct bkey uk; for_each_bset(b, t) -- cgit v1.2.3-59-g8ed1b From 0e57996c6964027e36a854247bfd266d207314c7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 12:43:31 -0400 Subject: bcachefs: bch2_dev_get_ioref2(); alloc_background.c Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 19 ++++++++----------- fs/bcachefs/sb-members.h | 18 ++++++++++++++++++ 2 files changed, 26 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 71b727f5d432..b9ded84d2d94 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1657,9 +1657,8 @@ static void discard_buckets_next_dev(struct bch_fs *c, struct discard_buckets_st bch2_dev_usage_read(s->ca).d[BCH_DATA_free].buckets) bch2_journal_flush_async(&c->journal, NULL); - bch2_dev_put(s->ca); - if (ca) - bch2_dev_get(ca); + if (s->ca) + percpu_ref_put(&s->ca->io_ref); s->ca = ca; s->need_journal_commit_this_dev = 0; } @@ -1673,15 +1672,15 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, struct bpos pos = need_discard_iter->pos; struct btree_iter iter = { NULL }; struct bkey_s_c k; - struct bch_dev *ca; struct bkey_i_alloc_v4 *a; struct printbuf buf = PRINTBUF; bool discard_locked = false; int ret = 0; - ca = bch2_dev_bkey_exists(c, pos.inode); - - if (!percpu_ref_tryget(&ca->io_ref)) { + struct bch_dev *ca = s->ca && s->ca->dev_idx == pos.inode + ? s->ca + : bch2_dev_get_ioref2(c, pos.inode, WRITE); + if (!ca) { bch2_btree_iter_set_pos(need_discard_iter, POS(pos.inode + 1, 0)); return 0; } @@ -1787,7 +1786,6 @@ out: discard_in_flight_remove(c, iter.pos); s->seen++; bch2_trans_iter_exit(trans, &iter); - percpu_ref_put(&ca->io_ref); printbuf_exit(&buf); return ret; } @@ -1862,9 +1860,8 @@ static void bch2_do_discards_fast_work(struct work_struct *work) if (i->snapshot) continue; - ca = bch2_dev_bkey_exists(c, i->inode); - - if (!percpu_ref_tryget(&ca->io_ref)) { + ca = bch2_dev_get_ioref2(c, i->inode, WRITE); + if (!ca) { darray_remove_item(&c->discard_buckets_in_flight, i); continue; } diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index 7cecac44ab80..f2a7794b0b50 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -285,6 +285,24 @@ static inline struct bch_dev *bch2_dev_iterate(struct bch_fs *c, struct bch_dev return bch2_dev_tryget(c, dev_idx); } +static inline struct bch_dev *bch2_dev_get_ioref2(struct bch_fs *c, unsigned dev, int rw) +{ + rcu_read_lock(); + struct bch_dev *ca = bch2_dev_rcu(c, dev); + if (ca && !percpu_ref_tryget(&ca->io_ref)) + ca = NULL; + rcu_read_unlock(); + + if (ca && + (ca->mi.state == BCH_MEMBER_STATE_rw || + (ca->mi.state == BCH_MEMBER_STATE_ro && rw == READ))) + return ca; + + if (ca) + percpu_ref_put(&ca->io_ref); + return NULL; +} + /* XXX kill, move to struct bch_fs */ static inline struct bch_devs_mask bch2_online_devs(struct bch_fs *c) { -- cgit v1.2.3-59-g8ed1b From 466298e2f6df31b5d1f7eebb1479ed4e154c75bf Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 12:44:07 -0400 Subject: bcachefs: bch2_dev_get_ioref2(); backpointers.c Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 3f5333b4c3de..9ac46f868bb4 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -486,8 +486,8 @@ found: bytes = p.crc.compressed_size << 9; - struct bch_dev *ca = bch2_dev_bkey_exists(c, dev); - if (!bch2_dev_get_ioref(ca, READ)) + struct bch_dev *ca = bch2_dev_get_ioref2(c, dev, READ); + if (!ca) return false; data_buf = kvmalloc(bytes, GFP_KERNEL); -- cgit v1.2.3-59-g8ed1b From 690f7cdf732a032388e26c5daed6ad46c9e48256 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 12:45:16 -0400 Subject: bcachefs: bch2_dev_get_ioref2(); btree_io.c Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_io.c | 33 ++++++++++++++++++--------------- 1 file changed, 18 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index e14c404d25c1..876a26ae0497 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1109,7 +1109,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, nonce = btree_nonce(i, b->written << 9); struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, bne); csum_bad = bch2_crc_cmp(bne->csum, csum); - if (csum_bad) + if (ca && csum_bad) bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); btree_err_on(csum_bad, @@ -1263,12 +1263,14 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, btree_node_reset_sib_u64s(b); + rcu_read_lock(); bkey_for_each_ptr(bch2_bkey_ptrs(bkey_i_to_s(&b->key)), ptr) { - struct bch_dev *ca2 = bch2_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca2 = bch2_dev_rcu(c, ptr->dev); - if (ca2->mi.state != BCH_MEMBER_STATE_rw) + if (!ca2 || ca2->mi.state != BCH_MEMBER_STATE_rw) set_btree_node_need_rewrite(b); } + rcu_read_unlock(); if (!ptr_written) set_btree_node_need_rewrite(b); @@ -1293,8 +1295,8 @@ static void btree_node_read_work(struct work_struct *work) struct btree_read_bio *rb = container_of(work, struct btree_read_bio, work); struct bch_fs *c = rb->c; + struct bch_dev *ca = rb->have_ioref ? bch2_dev_have_ref(c, rb->pick.ptr.dev) : NULL; struct btree *b = rb->b; - struct bch_dev *ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); struct bio *bio = &rb->bio; struct bch_io_failures failed = { .nr = 0 }; struct printbuf buf = PRINTBUF; @@ -1306,8 +1308,8 @@ static void btree_node_read_work(struct work_struct *work) while (1) { retry = true; bch_info(c, "retrying read"); - ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); - rb->have_ioref = bch2_dev_get_ioref(ca, READ); + ca = bch2_dev_get_ioref2(c, rb->pick.ptr.dev, READ); + rb->have_ioref = ca != NULL; bio_reset(bio, NULL, REQ_OP_READ|REQ_SYNC|REQ_META); bio->bi_iter.bi_sector = rb->pick.ptr.offset; bio->bi_iter.bi_size = btree_buf_bytes(b); @@ -1321,7 +1323,7 @@ static void btree_node_read_work(struct work_struct *work) start: printbuf_reset(&buf); bch2_btree_pos_to_text(&buf, c, b); - bch2_dev_io_err_on(bio->bi_status, ca, BCH_MEMBER_ERROR_read, + bch2_dev_io_err_on(ca && bio->bi_status, ca, BCH_MEMBER_ERROR_read, "btree read error %s for %s", bch2_blk_status_to_str(bio->bi_status), buf.buf); if (rb->have_ioref) @@ -1377,7 +1379,7 @@ static void btree_node_read_endio(struct bio *bio) struct bch_fs *c = rb->c; if (rb->have_ioref) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); + struct bch_dev *ca = bch2_dev_have_ref(c, rb->pick.ptr.dev); bch2_latency_acct(ca, rb->start_time, READ); } @@ -1574,7 +1576,7 @@ static void btree_node_read_all_replicas_endio(struct bio *bio) struct btree_node_read_all *ra = rb->ra; if (rb->have_ioref) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, rb->pick.ptr.dev); + struct bch_dev *ca = bch2_dev_have_ref(c, rb->pick.ptr.dev); bch2_latency_acct(ca, rb->start_time, READ); } @@ -1616,14 +1618,14 @@ static int btree_node_read_all_replicas(struct bch_fs *c, struct btree *b, bool i = 0; bkey_for_each_ptr_decode(k.k, ptrs, pick, entry) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, pick.ptr.dev); + struct bch_dev *ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); struct btree_read_bio *rb = container_of(ra->bio[i], struct btree_read_bio, bio); rb->c = c; rb->b = b; rb->ra = ra; rb->start_time = local_clock(); - rb->have_ioref = bch2_dev_get_ioref(ca, READ); + rb->have_ioref = ca != NULL; rb->idx = i; rb->pick = pick; rb->bio.bi_iter.bi_sector = pick.ptr.offset; @@ -1693,7 +1695,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, return; } - ca = bch2_dev_bkey_exists(c, pick.ptr.dev); + ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); bio = bio_alloc_bioset(NULL, buf_pages(b->data, btree_buf_bytes(b)), @@ -1705,7 +1707,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, rb->b = b; rb->ra = NULL; rb->start_time = local_clock(); - rb->have_ioref = bch2_dev_get_ioref(ca, READ); + rb->have_ioref = ca != NULL; rb->pick = pick; INIT_WORK(&rb->work, btree_node_read_work); bio->bi_iter.bi_sector = pick.ptr.offset; @@ -1909,13 +1911,14 @@ static void btree_node_write_endio(struct bio *bio) struct btree_write_bio *wb = container_of(orig, struct btree_write_bio, wbio); struct bch_fs *c = wbio->c; struct btree *b = wbio->bio.bi_private; - struct bch_dev *ca = bch2_dev_bkey_exists(c, wbio->dev); + struct bch_dev *ca = wbio->have_ioref ? bch2_dev_have_ref(c, wbio->dev) : NULL; unsigned long flags; if (wbio->have_ioref) bch2_latency_acct(ca, wbio->submit_time, WRITE); - if (bch2_dev_io_err_on(bio->bi_status, ca, BCH_MEMBER_ERROR_write, + if (!ca || + bch2_dev_io_err_on(bio->bi_status, ca, BCH_MEMBER_ERROR_write, "btree write error: %s", bch2_blk_status_to_str(bio->bi_status)) || bch2_meta_write_fault("btree")) { -- cgit v1.2.3-59-g8ed1b From 48af853925998c04a4be1473b32101ed4ba367e6 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 12:53:27 -0400 Subject: bcachefs: bch2_dev_get_ioref2(); io_write.c Signed-off-by: Kent Overstreet --- fs/bcachefs/io_write.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index fa3413d19b6a..d87c0d2f9902 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -407,9 +407,9 @@ void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, BUG_ON(c->opts.nochanges); bkey_for_each_ptr(ptrs, ptr) { - BUG_ON(!bch2_dev_exists(c, ptr->dev)); - - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = nocow + ? bch2_dev_have_ref(c, ptr->dev) + : bch2_dev_get_ioref2(c, ptr->dev, type == BCH_DATA_btree ? READ : WRITE); if (to_entry(ptr + 1) < ptrs.end) { n = to_wbio(bio_alloc_clone(NULL, &wbio->bio, GFP_NOFS, &c->replica_set)); @@ -429,8 +429,7 @@ void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, n->c = c; n->dev = ptr->dev; - n->have_ioref = nocow || bch2_dev_get_ioref(ca, - type == BCH_DATA_btree ? READ : WRITE); + n->have_ioref = ca != NULL; n->nocow = nocow; n->submit_time = local_clock(); n->inode_offset = bkey_start_offset(&k->k); @@ -650,7 +649,9 @@ static void bch2_write_endio(struct bio *bio) struct bch_write_bio *wbio = to_wbio(bio); struct bch_write_bio *parent = wbio->split ? wbio->parent : NULL; struct bch_fs *c = wbio->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, wbio->dev); + struct bch_dev *ca = wbio->have_ioref + ? bch2_dev_have_ref(c, wbio->dev) + : NULL; if (bch2_dev_inum_io_err_on(bio->bi_status, ca, BCH_MEMBER_ERROR_write, op->pos.inode, @@ -1264,15 +1265,15 @@ retry: /* Get iorefs before dropping btree locks: */ struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_get_ioref2(c, ptr->dev, WRITE); + if (unlikely(!ca)) + goto err_get_ioref; + struct bpos b = PTR_BUCKET_POS(ca, ptr); struct nocow_lock_bucket *l = bucket_nocow_lock(&c->nocow_locks, bucket_to_u64(b)); prefetch(l); - if (unlikely(!bch2_dev_get_ioref(ca, WRITE))) - goto err_get_ioref; - /* XXX allocating memory with btree locks held - rare */ darray_push_gfp(&buckets, ((struct bucket_to_lock) { .b = b, .gen = ptr->gen, .l = l, -- cgit v1.2.3-59-g8ed1b From 6212ea24975faf47f7c0f402b46b3a5c7ecc8053 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 12:54:25 -0400 Subject: bcachefs: bch2_dev_get_ioref2(); journal_io.c Signed-off-by: Kent Overstreet --- fs/bcachefs/journal_io.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index c9744546d48d..f6f9d03bb8a3 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1722,10 +1722,8 @@ static CLOSURE_CALLBACK(journal_write_submit) unsigned sectors = vstruct_sectors(w->data, c->block_bits); extent_for_each_ptr(bkey_i_to_s_extent(&w->key), ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); - struct journal_device *ja = &ca->journal; - - if (!percpu_ref_tryget(&ca->io_ref)) { + struct bch_dev *ca = bch2_dev_get_ioref2(c, ptr->dev, WRITE); + if (!ca) { /* XXX: fix this */ bch_err(c, "missing device for journal write\n"); continue; @@ -1734,6 +1732,7 @@ static CLOSURE_CALLBACK(journal_write_submit) this_cpu_add(ca->io_done->sectors[WRITE][BCH_DATA_journal], sectors); + struct journal_device *ja = &ca->journal; struct bio *bio = &ja->bio[w->idx]->bio; bio_reset(bio, ca->disk_sb.bdev, REQ_OP_WRITE|REQ_SYNC|REQ_META); bio->bi_iter.bi_sector = ptr->offset; -- cgit v1.2.3-59-g8ed1b From 91ffdecfc796e79ce9d6ae1cf052afebbda8b01e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 12:55:55 -0400 Subject: bcachefs: bch2_dev_get_ioref2(); debug.c Signed-off-by: Kent Overstreet --- fs/bcachefs/debug.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index b5aa4e5b224b..d75735aad27e 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -37,11 +37,11 @@ static bool bch2_btree_verify_replica(struct bch_fs *c, struct btree *b, struct btree_node *n_ondisk = c->verify_ondisk; struct btree_node *n_sorted = c->verify_data->data; struct bset *sorted, *inmemory = &b->data->keys; - struct bch_dev *ca = bch2_dev_bkey_exists(c, pick.ptr.dev); struct bio *bio; bool failed = false, saw_error = false; - if (!bch2_dev_get_ioref(ca, READ)) + struct bch_dev *ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); + if (!ca) return false; bio = bio_alloc_bioset(ca->disk_sb.bdev, @@ -194,8 +194,8 @@ void bch2_btree_node_ondisk_to_text(struct printbuf *out, struct bch_fs *c, return; } - ca = bch2_dev_bkey_exists(c, pick.ptr.dev); - if (!bch2_dev_get_ioref(ca, READ)) { + ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); + if (!ca) { prt_printf(out, "error getting device to read from: not online\n"); return; } -- cgit v1.2.3-59-g8ed1b From 465bf6f42aac1474c6d58258501e55e8760c0a34 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 3 May 2024 12:50:22 -0400 Subject: bcachefs: bch2_dev_get_ioref2(); io_read.c Signed-off-by: Kent Overstreet --- fs/bcachefs/io_read.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 91260d3dbf50..4aa7c6d8d693 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -800,7 +800,6 @@ int __bch2_read_extent(struct btree_trans *trans, struct bch_read_bio *orig, struct bch_fs *c = trans->c; struct extent_ptr_decoded pick; struct bch_read_bio *rbio = NULL; - struct bch_dev *ca = NULL; struct promote_op *promote = NULL; bool bounce = false, read_full = false, narrow_crcs = false; struct bpos data_pos = bkey_start_pos(k.k); @@ -831,7 +830,7 @@ retry_pick: goto err; } - ca = bch2_dev_bkey_exists(c, pick.ptr.dev); + struct bch_dev *ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); /* * Stale dirty pointers are treated as IO errors, but @failed isn't @@ -841,9 +840,11 @@ retry_pick: */ if ((flags & BCH_READ_IN_RETRY) && !pick.ptr.cached && + ca && unlikely(dev_ptr_stale(ca, &pick.ptr))) { read_from_stale_dirty_pointer(trans, ca, k, pick.ptr); bch2_mark_io_failure(failed, &pick); + percpu_ref_put(&ca->io_ref); goto retry_pick; } @@ -858,8 +859,11 @@ retry_pick: * can happen if we retry, and the extent we were going to read * has been merged in the meantime: */ - if (pick.crc.compressed_size > orig->bio.bi_vcnt * PAGE_SECTORS) + if (pick.crc.compressed_size > orig->bio.bi_vcnt * PAGE_SECTORS) { + if (ca) + percpu_ref_put(&ca->io_ref); goto hole; + } iter.bi_size = pick.crc.compressed_size << 9; goto get_bio; @@ -964,7 +968,7 @@ get_bio: rbio->bvec_iter = iter; rbio->offset_into_extent= offset_into_extent; rbio->flags = flags; - rbio->have_ioref = pick_ret > 0 && bch2_dev_get_ioref(ca, READ); + rbio->have_ioref = ca != NULL; rbio->narrow_crcs = narrow_crcs; rbio->hole = 0; rbio->retry = 0; @@ -994,7 +998,7 @@ get_bio: * If it's being moved internally, we don't want to flag it as a cache * hit: */ - if (pick.ptr.cached && !(flags & BCH_READ_NODECODE)) + if (ca && pick.ptr.cached && !(flags & BCH_READ_NODECODE)) bch2_bucket_io_time_reset(trans, pick.ptr.dev, PTR_BUCKET_NR(ca, &pick.ptr), READ); -- cgit v1.2.3-59-g8ed1b From 2c91ab7262e6b08bbc76b83ed1981602fa6ef835 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 30 Apr 2024 15:37:51 -0400 Subject: bcachefs: bch2_dev_get_ioref() checks for device not present Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 4 ++-- fs/bcachefs/backpointers.c | 2 +- fs/bcachefs/btree_io.c | 6 +++--- fs/bcachefs/debug.c | 4 ++-- fs/bcachefs/ec.c | 21 ++++++++++----------- fs/bcachefs/io_read.c | 2 +- fs/bcachefs/io_write.c | 4 ++-- fs/bcachefs/journal_io.c | 2 +- fs/bcachefs/sb-members.h | 15 +-------------- 9 files changed, 23 insertions(+), 37 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index b9ded84d2d94..66a1d57db25d 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1679,7 +1679,7 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, struct bch_dev *ca = s->ca && s->ca->dev_idx == pos.inode ? s->ca - : bch2_dev_get_ioref2(c, pos.inode, WRITE); + : bch2_dev_get_ioref(c, pos.inode, WRITE); if (!ca) { bch2_btree_iter_set_pos(need_discard_iter, POS(pos.inode + 1, 0)); return 0; @@ -1860,7 +1860,7 @@ static void bch2_do_discards_fast_work(struct work_struct *work) if (i->snapshot) continue; - ca = bch2_dev_get_ioref2(c, i->inode, WRITE); + ca = bch2_dev_get_ioref(c, i->inode, WRITE); if (!ca) { darray_remove_item(&c->discard_buckets_in_flight, i); continue; diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 9ac46f868bb4..921d95c46cf7 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -486,7 +486,7 @@ found: bytes = p.crc.compressed_size << 9; - struct bch_dev *ca = bch2_dev_get_ioref2(c, dev, READ); + struct bch_dev *ca = bch2_dev_get_ioref(c, dev, READ); if (!ca) return false; diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 876a26ae0497..cbf8f5d90602 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1308,7 +1308,7 @@ static void btree_node_read_work(struct work_struct *work) while (1) { retry = true; bch_info(c, "retrying read"); - ca = bch2_dev_get_ioref2(c, rb->pick.ptr.dev, READ); + ca = bch2_dev_get_ioref(c, rb->pick.ptr.dev, READ); rb->have_ioref = ca != NULL; bio_reset(bio, NULL, REQ_OP_READ|REQ_SYNC|REQ_META); bio->bi_iter.bi_sector = rb->pick.ptr.offset; @@ -1618,7 +1618,7 @@ static int btree_node_read_all_replicas(struct bch_fs *c, struct btree *b, bool i = 0; bkey_for_each_ptr_decode(k.k, ptrs, pick, entry) { - struct bch_dev *ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); + struct bch_dev *ca = bch2_dev_get_ioref(c, pick.ptr.dev, READ); struct btree_read_bio *rb = container_of(ra->bio[i], struct btree_read_bio, bio); rb->c = c; @@ -1695,7 +1695,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, return; } - ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); + ca = bch2_dev_get_ioref(c, pick.ptr.dev, READ); bio = bio_alloc_bioset(NULL, buf_pages(b->data, btree_buf_bytes(b)), diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index d75735aad27e..51cbf3928361 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -40,7 +40,7 @@ static bool bch2_btree_verify_replica(struct bch_fs *c, struct btree *b, struct bio *bio; bool failed = false, saw_error = false; - struct bch_dev *ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); + struct bch_dev *ca = bch2_dev_get_ioref(c, pick.ptr.dev, READ); if (!ca) return false; @@ -194,7 +194,7 @@ void bch2_btree_node_ondisk_to_text(struct printbuf *out, struct bch_fs *c, return; } - ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); + ca = bch2_dev_get_ioref(c, pick.ptr.dev, READ); if (!ca) { prt_printf(out, "error getting device to read from: not online\n"); return; diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 99a8c8fe8e3b..055be0ca9250 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -732,12 +732,17 @@ static void ec_block_io(struct bch_fs *c, struct ec_stripe_buf *buf, struct bch_stripe *v = &bkey_i_to_stripe(&buf->key)->v; unsigned offset = 0, bytes = buf->size << 9; struct bch_extent_ptr *ptr = &v->ptrs[idx]; - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); enum bch_data_type data_type = idx < v->nr_blocks - v->nr_redundant ? BCH_DATA_user : BCH_DATA_parity; int rw = op_is_write(opf); + struct bch_dev *ca = bch2_dev_get_ioref(c, ptr->dev, rw); + if (!ca) { + clear_bit(idx, buf->valid); + return; + } + if (dev_ptr_stale(ca, ptr)) { bch_err_ratelimited(c, "error %s stripe: stale pointer", @@ -746,10 +751,6 @@ static void ec_block_io(struct bch_fs *c, struct ec_stripe_buf *buf, return; } - if (!bch2_dev_get_ioref(ca, rw)) { - clear_bit(idx, buf->valid); - return; - } this_cpu_add(ca->io_done->sectors[rw][data_type], buf->size); @@ -1354,20 +1355,18 @@ static void zero_out_rest_of_ec_bucket(struct bch_fs *c, unsigned block, struct open_bucket *ob) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ob->dev); - unsigned offset = ca->mi.bucket_size - ob->sectors_free; - int ret; - - if (!bch2_dev_get_ioref(ca, WRITE)) { + struct bch_dev *ca = bch2_dev_get_ioref(c, ob->dev, WRITE); + if (!ca) { s->err = -BCH_ERR_erofs_no_writes; return; } + unsigned offset = ca->mi.bucket_size - ob->sectors_free; memset(s->new_stripe.data[block] + (offset << 9), 0, ob->sectors_free << 9); - ret = blkdev_issue_zeroout(ca->disk_sb.bdev, + int ret = blkdev_issue_zeroout(ca->disk_sb.bdev, ob->bucket * ca->mi.bucket_size + offset, ob->sectors_free, GFP_KERNEL, 0); diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 4aa7c6d8d693..3e7dceef4cb5 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -830,7 +830,7 @@ retry_pick: goto err; } - struct bch_dev *ca = bch2_dev_get_ioref2(c, pick.ptr.dev, READ); + struct bch_dev *ca = bch2_dev_get_ioref(c, pick.ptr.dev, READ); /* * Stale dirty pointers are treated as IO errors, but @failed isn't diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index d87c0d2f9902..9401d13e31bb 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -409,7 +409,7 @@ void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, bkey_for_each_ptr(ptrs, ptr) { struct bch_dev *ca = nocow ? bch2_dev_have_ref(c, ptr->dev) - : bch2_dev_get_ioref2(c, ptr->dev, type == BCH_DATA_btree ? READ : WRITE); + : bch2_dev_get_ioref(c, ptr->dev, type == BCH_DATA_btree ? READ : WRITE); if (to_entry(ptr + 1) < ptrs.end) { n = to_wbio(bio_alloc_clone(NULL, &wbio->bio, GFP_NOFS, &c->replica_set)); @@ -1265,7 +1265,7 @@ retry: /* Get iorefs before dropping btree locks: */ struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_get_ioref2(c, ptr->dev, WRITE); + struct bch_dev *ca = bch2_dev_get_ioref(c, ptr->dev, WRITE); if (unlikely(!ca)) goto err_get_ioref; diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index f6f9d03bb8a3..d3bc5df0f3de 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1722,7 +1722,7 @@ static CLOSURE_CALLBACK(journal_write_submit) unsigned sectors = vstruct_sectors(w->data, c->block_bits); extent_for_each_ptr(bkey_i_to_s_extent(&w->key), ptr) { - struct bch_dev *ca = bch2_dev_get_ioref2(c, ptr->dev, WRITE); + struct bch_dev *ca = bch2_dev_get_ioref(c, ptr->dev, WRITE); if (!ca) { /* XXX: fix this */ bch_err(c, "missing device for journal write\n"); diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index f2a7794b0b50..ae28b5d401e3 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -29,19 +29,6 @@ static inline bool bch2_dev_is_readable(struct bch_dev *ca) ca->mi.state != BCH_MEMBER_STATE_failed; } -static inline bool bch2_dev_get_ioref(struct bch_dev *ca, int rw) -{ - if (!percpu_ref_tryget(&ca->io_ref)) - return false; - - if (ca->mi.state == BCH_MEMBER_STATE_rw || - (ca->mi.state == BCH_MEMBER_STATE_ro && rw == READ)) - return true; - - percpu_ref_put(&ca->io_ref); - return false; -} - static inline unsigned dev_mask_nr(const struct bch_devs_mask *devs) { return bitmap_weight(devs->d, BCH_SB_MEMBERS_MAX); @@ -285,7 +272,7 @@ static inline struct bch_dev *bch2_dev_iterate(struct bch_fs *c, struct bch_dev return bch2_dev_tryget(c, dev_idx); } -static inline struct bch_dev *bch2_dev_get_ioref2(struct bch_fs *c, unsigned dev, int rw) +static inline struct bch_dev *bch2_dev_get_ioref(struct bch_fs *c, unsigned dev, int rw) { rcu_read_lock(); struct bch_dev *ca = bch2_dev_rcu(c, dev); -- cgit v1.2.3-59-g8ed1b From 02b7fa4fe5307b1de74e734e28192623b7af7020 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 03:59:52 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in bch2_read_endio() Signed-off-by: Kent Overstreet --- fs/bcachefs/io_read.c | 33 +++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 3e7dceef4cb5..f57486794484 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -541,7 +541,6 @@ static void __bch2_read_endio(struct work_struct *work) struct bch_read_bio *rbio = container_of(work, struct bch_read_bio, work); struct bch_fs *c = rbio->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, rbio->pick.ptr.dev); struct bio *src = &rbio->bio; struct bio *dst = &bch2_rbio_parent(rbio)->bio; struct bvec_iter dst_iter = rbio->bvec_iter; @@ -647,13 +646,15 @@ csum_err: prt_str(&buf, "data "); bch2_csum_err_msg(&buf, crc.csum_type, rbio->pick.crc.csum, csum); - bch_err_inum_offset_ratelimited(ca, - rbio->read_pos.inode, - rbio->read_pos.offset << 9, - "data %s", buf.buf); + struct bch_dev *ca = rbio->have_ioref ? bch2_dev_have_ref(c, rbio->pick.ptr.dev) : NULL; + if (ca) { + bch_err_inum_offset_ratelimited(ca, + rbio->read_pos.inode, + rbio->read_pos.offset << 9, + "data %s", buf.buf); + bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); + } printbuf_exit(&buf); - - bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); bch2_rbio_error(rbio, READ_RETRY_AVOID, BLK_STS_IOERR); goto out; decompression_err: @@ -675,7 +676,7 @@ static void bch2_read_endio(struct bio *bio) struct bch_read_bio *rbio = container_of(bio, struct bch_read_bio, bio); struct bch_fs *c = rbio->c; - struct bch_dev *ca = bch2_dev_bkey_exists(c, rbio->pick.ptr.dev); + struct bch_dev *ca = rbio->have_ioref ? bch2_dev_have_ref(c, rbio->pick.ptr.dev) : NULL; struct workqueue_struct *wq = NULL; enum rbio_context context = RBIO_CONTEXT_NULL; @@ -687,17 +688,21 @@ static void bch2_read_endio(struct bio *bio) if (!rbio->split) rbio->bio.bi_end_io = rbio->end_io; - if (bch2_dev_inum_io_err_on(bio->bi_status, ca, BCH_MEMBER_ERROR_read, - rbio->read_pos.inode, - rbio->read_pos.offset, - "data read error: %s", - bch2_blk_status_to_str(bio->bi_status))) { + if (bio->bi_status) { + if (ca) { + bch_err_inum_offset_ratelimited(ca, + rbio->read_pos.inode, + rbio->read_pos.offset, + "data read error: %s", + bch2_blk_status_to_str(bio->bi_status)); + bch2_io_error(ca, BCH_MEMBER_ERROR_read); + } bch2_rbio_error(rbio, READ_RETRY_AVOID, bio->bi_status); return; } if (((rbio->flags & BCH_READ_RETRY_IF_STALE) && race_fault()) || - dev_ptr_stale(ca, &rbio->pick.ptr)) { + (ca && dev_ptr_stale(ca, &rbio->pick.ptr))) { trace_and_count(c, read_reuse_race, &rbio->bio); if (rbio->flags & BCH_READ_RETRY_IF_STALE) -- cgit v1.2.3-59-g8ed1b From 2f4b4a3b44755d8c86eff64b9c54e3242a9b659c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 19:10:17 -0400 Subject: bcachefs: kill bch2_dev_bkey_exists() in bch2_check_fix_ptrs() Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 27 ++++++++++++++++++++++++--- fs/bcachefs/sb-members.h | 11 ----------- 2 files changed, 24 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index ae6d6879e2e0..e28d28ac2a13 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -479,8 +479,20 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, percpu_down_read(&c->mark_lock); + rcu_read_lock(); bkey_for_each_ptr_decode(k.k, ptrs_c, p, entry_c) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); + if (!ca) { + if (fsck_err(c, ptr_to_invalid_device, + "pointer to missing device %u\n" + "while marking %s", + p.ptr.dev, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + do_update = true; + continue; + } + struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry_c); @@ -590,6 +602,7 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, do_update = true; } } + rcu_read_unlock(); if (do_update) { if (flags & BTREE_TRIGGER_is_root) { @@ -603,6 +616,10 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, if (ret) goto err; + rcu_read_lock(); + bch2_bkey_drop_ptrs(bkey_i_to_s(new), ptr, !bch2_dev_rcu(c, ptr->dev)); + rcu_read_unlock(); + if (level) { /* * We don't want to drop btree node pointers - if the @@ -610,19 +627,22 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, * sort it out: */ struct bkey_ptrs ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); + rcu_read_lock(); bkey_for_each_ptr(ptrs, ptr) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, ptr->dev); + struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); struct bucket *g = PTR_GC_BUCKET(ca, ptr); ptr->gen = g->gen; } + rcu_read_unlock(); } else { struct bkey_ptrs ptrs; union bch_extent_entry *entry; restart_drop_ptrs: ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); + rcu_read_lock(); bkey_for_each_ptr_decode(bkey_i_to_s(new).k, ptrs, p, entry) { - struct bch_dev *ca = bch2_dev_bkey_exists(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); enum bch_data_type data_type = bch2_bkey_ptr_data_type(bkey_i_to_s_c(new), p, entry); @@ -637,6 +657,7 @@ restart_drop_ptrs: goto restart_drop_ptrs; } } + rcu_read_unlock(); again: ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); bkey_extent_entry_for_each(ptrs, entry) { diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index ae28b5d401e3..dd93192ec065 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -189,17 +189,6 @@ static inline struct bch_dev *bch2_dev_have_ref(const struct bch_fs *c, unsigned return rcu_dereference_check(c->devs[dev], 1); } -/* - * If a key exists that references a device, the device won't be going away and - * we can omit rcu_read_lock(): - */ -static inline struct bch_dev *bch2_dev_bkey_exists(const struct bch_fs *c, unsigned dev) -{ - EBUG_ON(!bch2_dev_exists(c, dev)); - - return rcu_dereference_check(c->devs[dev], 1); -} - static inline struct bch_dev *bch2_dev_locked(struct bch_fs *c, unsigned dev) { EBUG_ON(!bch2_dev_exists(c, dev)); -- cgit v1.2.3-59-g8ed1b From 99179fb89847f312229bf816200e1e6e5038e3a3 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 1 May 2024 19:15:29 -0400 Subject: bcachefs: Invalid devices are now checked for by fsck, not .invalid methods Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 15 +-------------- fs/bcachefs/journal_io.c | 11 ----------- 2 files changed, 1 insertion(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index b785712ae9b2..094caf1cdb35 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -1101,22 +1101,8 @@ static int extent_ptr_invalid(struct bch_fs *c, bool metadata, struct printbuf *err) { - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); int ret = 0; - if (!bch2_dev_exists(c, ptr->dev)) { - /* - * If we're in the write path this key might have already been - * overwritten, and we could be seeing a device that doesn't - * exist anymore due to racing with device removal: - */ - if (flags & BKEY_INVALID_WRITE) - return 0; - - bkey_fsck_err(c, err, ptr_to_invalid_device, - "pointer to invalid device (%u)", ptr->dev); - } - rcu_read_lock(); struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); if (!ca) { @@ -1130,6 +1116,7 @@ static int extent_ptr_invalid(struct bch_fs *c, unsigned bucket_size = ca->mi.bucket_size; rcu_read_unlock(); + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); bkey_for_each_ptr(ptrs, ptr2) bkey_fsck_err_on(ptr != ptr2 && ptr->dev == ptr2->dev, c, err, ptr_to_duplicate_device, diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index d3bc5df0f3de..56f4d46ef094 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -690,7 +690,6 @@ static int journal_entry_dev_usage_validate(struct bch_fs *c, container_of(entry, struct jset_entry_dev_usage, entry); unsigned bytes = jset_u64s(le16_to_cpu(entry->u64s)) * sizeof(u64); unsigned expected = sizeof(*u); - unsigned dev; int ret = 0; if (journal_entry_err_on(bytes < expected, @@ -702,16 +701,6 @@ static int journal_entry_dev_usage_validate(struct bch_fs *c, return ret; } - dev = le32_to_cpu(u->dev); - - if (journal_entry_err_on(!bch2_dev_exists(c, dev), - c, version, jset, entry, - journal_entry_dev_usage_bad_dev, - "bad dev")) { - journal_entry_null_range(entry, vstruct_next(entry)); - return ret; - } - if (journal_entry_err_on(u->pad, c, version, jset, entry, journal_entry_dev_usage_bad_pad, -- cgit v1.2.3-59-g8ed1b From d09a8468d915850709ae5f34c23e2b24cb5c3c62 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 17:33:29 -0400 Subject: bcachefs: fsync() should not return -EROFS fsync has a slightly odd usage of -EROFS, where it means "does not support fsync". I didn't choose it... Signed-off-by: Kent Overstreet --- fs/bcachefs/fs-io.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c index 442bcb0793c4..ef20b64033e0 100644 --- a/fs/bcachefs/fs-io.c +++ b/fs/bcachefs/fs-io.c @@ -202,7 +202,10 @@ int bch2_fsync(struct file *file, loff_t start, loff_t end, int datasync) goto out; ret = bch2_flush_inode(c, inode); out: - return bch2_err_class(ret); + ret = bch2_err_class(ret); + if (ret == -EROFS) + ret = -EIO; + return ret; } /* truncate: */ -- cgit v1.2.3-59-g8ed1b From 65eaf4e24ab6b4809491248e1fed36b8d49c1ea9 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 18:40:42 -0400 Subject: bcachefs: s/bkey_invalid_flags/bch_validate_flags We're about to start using bch_validate_flags for superblock section validation - it's no longer bkey specific. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 10 ++++----- fs/bcachefs/alloc_background.h | 12 +++++------ fs/bcachefs/backpointers.c | 2 +- fs/bcachefs/backpointers.h | 2 +- fs/bcachefs/bkey.c | 2 +- fs/bcachefs/bkey.h | 10 ++++----- fs/bcachefs/bkey_methods.c | 16 +++++++------- fs/bcachefs/bkey_methods.h | 8 +++---- fs/bcachefs/btree_trans_commit.c | 10 ++++----- fs/bcachefs/dirent.c | 4 ++-- fs/bcachefs/dirent.h | 4 ++-- fs/bcachefs/ec.c | 2 +- fs/bcachefs/ec.h | 4 ++-- fs/bcachefs/extents.c | 14 ++++++------ fs/bcachefs/extents.h | 10 ++++----- fs/bcachefs/inode.c | 8 +++---- fs/bcachefs/inode.h | 10 ++++----- fs/bcachefs/journal_io.c | 46 ++++++++++++++++++++-------------------- fs/bcachefs/journal_io.h | 2 +- fs/bcachefs/lru.c | 2 +- fs/bcachefs/lru.h | 2 +- fs/bcachefs/quota.c | 2 +- fs/bcachefs/quota.h | 4 ++-- fs/bcachefs/reflink.c | 6 +++--- fs/bcachefs/reflink.h | 8 +++---- fs/bcachefs/snapshot.c | 4 ++-- fs/bcachefs/snapshot.h | 6 +++--- fs/bcachefs/subvolume.c | 2 +- fs/bcachefs/subvolume.h | 4 ++-- fs/bcachefs/xattr.c | 2 +- fs/bcachefs/xattr.h | 2 +- 31 files changed, 110 insertions(+), 110 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index 66a1d57db25d..346cd91f91f9 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -195,7 +195,7 @@ static unsigned bch_alloc_v1_val_u64s(const struct bch_alloc *a) } int bch2_alloc_v1_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_alloc a = bkey_s_c_to_alloc(k); @@ -211,7 +211,7 @@ fsck_err: } int bch2_alloc_v2_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_alloc_unpacked u; @@ -225,7 +225,7 @@ fsck_err: } int bch2_alloc_v3_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_alloc_unpacked u; @@ -239,7 +239,7 @@ fsck_err: } int bch2_alloc_v4_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_alloc_v4 a = bkey_s_c_to_alloc_v4(k); int ret = 0; @@ -487,7 +487,7 @@ static unsigned alloc_gen(struct bkey_s_c k, unsigned offset) } int bch2_bucket_gens_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 5f42dc1e7fcc..ae31a94be6f9 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -8,7 +8,7 @@ #include "debug.h" #include "super.h" -enum bkey_invalid_flags; +enum bch_validate_flags; /* How out of date a pointer gen is allowed to be: */ #define BUCKET_GC_GEN_MAX 96U @@ -205,13 +205,13 @@ struct bkey_i_alloc_v4 *bch2_alloc_to_v4_mut(struct btree_trans *, struct bkey_s int bch2_bucket_io_time_reset(struct btree_trans *, unsigned, size_t, int); int bch2_alloc_v1_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_alloc_v2_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_alloc_v3_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_alloc_v4_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_alloc_v4_swab(struct bkey_s); void bch2_alloc_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); @@ -245,7 +245,7 @@ void bch2_alloc_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); }) int bch2_bucket_gens_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_bucket_gens_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_bucket_gens ((struct bkey_ops) { \ diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 921d95c46cf7..692b1c7d5018 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -48,7 +48,7 @@ static bool extent_matches_bp(struct bch_fs *c, } int bch2_backpointer_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index 929f4a068a0a..6021de1c5e98 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -19,7 +19,7 @@ static inline u64 swab40(u64 x) } int bch2_backpointer_invalid(struct bch_fs *, struct bkey_s_c k, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_backpointer_to_text(struct printbuf *, const struct bch_backpointer *); void bch2_backpointer_k_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); void bch2_backpointer_swab(struct bkey_s); diff --git a/fs/bcachefs/bkey.c b/fs/bcachefs/bkey.c index 952299a2e416..f46978e5cb7c 100644 --- a/fs/bcachefs/bkey.c +++ b/fs/bcachefs/bkey.c @@ -640,7 +640,7 @@ struct bkey_format bch2_bkey_format_done(struct bkey_format_state *s) int bch2_bkey_format_invalid(struct bch_fs *c, struct bkey_format *f, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { unsigned i, bits = KEY_PACKED_BITS_START; diff --git a/fs/bcachefs/bkey.h b/fs/bcachefs/bkey.h index 1bb0bd4371b0..fcd43915df07 100644 --- a/fs/bcachefs/bkey.h +++ b/fs/bcachefs/bkey.h @@ -9,10 +9,10 @@ #include "util.h" #include "vstructs.h" -enum bkey_invalid_flags { - BKEY_INVALID_WRITE = (1U << 0), - BKEY_INVALID_COMMIT = (1U << 1), - BKEY_INVALID_JOURNAL = (1U << 2), +enum bch_validate_flags { + BCH_VALIDATE_write = (1U << 0), + BCH_VALIDATE_commit = (1U << 1), + BCH_VALIDATE_journal = (1U << 2), }; #if 0 @@ -598,7 +598,7 @@ static inline bool bch2_bkey_format_field_overflows(struct bkey_format *f, unsig } int bch2_bkey_format_invalid(struct bch_fs *, struct bkey_format *, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_bkey_format_to_text(struct printbuf *, const struct bkey_format *); #endif /* _BCACHEFS_BKEY_H */ diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c index a275a9e8e341..f692f9d8f00d 100644 --- a/fs/bcachefs/bkey_methods.c +++ b/fs/bcachefs/bkey_methods.c @@ -27,7 +27,7 @@ const char * const bch2_bkey_types[] = { }; static int deleted_key_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { return 0; } @@ -41,7 +41,7 @@ static int deleted_key_invalid(struct bch_fs *c, struct bkey_s_c k, }) static int empty_val_key_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; @@ -58,7 +58,7 @@ fsck_err: }) static int key_type_cookie_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { return 0; } @@ -82,7 +82,7 @@ static void key_type_cookie_to_text(struct printbuf *out, struct bch_fs *c, }) static int key_type_inline_data_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { return 0; } @@ -123,7 +123,7 @@ const struct bkey_ops bch2_bkey_null_ops = { }; int bch2_bkey_val_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { const struct bkey_ops *ops = bch2_bkey_type_ops(k.k->type); @@ -159,7 +159,7 @@ const char *bch2_btree_node_type_str(enum btree_node_type type) int __bch2_bkey_invalid(struct bch_fs *c, struct bkey_s_c k, enum btree_node_type type, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; @@ -172,7 +172,7 @@ int __bch2_bkey_invalid(struct bch_fs *c, struct bkey_s_c k, return 0; bkey_fsck_err_on(k.k->type < KEY_TYPE_MAX && - (type == BKEY_TYPE_btree || (flags & BKEY_INVALID_COMMIT)) && + (type == BKEY_TYPE_btree || (flags & BCH_VALIDATE_commit)) && !(bch2_key_types_allowed[type] & BIT_ULL(k.k->type)), c, err, bkey_invalid_type_for_btree, "invalid key type for btree %s (%s)", @@ -224,7 +224,7 @@ fsck_err: int bch2_bkey_invalid(struct bch_fs *c, struct bkey_s_c k, enum btree_node_type type, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { return __bch2_bkey_invalid(c, k, type, flags, err) ?: diff --git a/fs/bcachefs/bkey_methods.h b/fs/bcachefs/bkey_methods.h index 7351c0d38534..726ef7483763 100644 --- a/fs/bcachefs/bkey_methods.h +++ b/fs/bcachefs/bkey_methods.h @@ -22,7 +22,7 @@ extern const struct bkey_ops bch2_bkey_null_ops; */ struct bkey_ops { int (*key_invalid)(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, struct printbuf *err); + enum bch_validate_flags flags, struct printbuf *err); void (*val_to_text)(struct printbuf *, struct bch_fs *, struct bkey_s_c); void (*swab)(struct bkey_s); @@ -49,11 +49,11 @@ static inline const struct bkey_ops *bch2_bkey_type_ops(enum bch_bkey_type type) } int bch2_bkey_val_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int __bch2_bkey_invalid(struct bch_fs *, struct bkey_s_c, enum btree_node_type, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_bkey_invalid(struct bch_fs *, struct bkey_s_c, enum btree_node_type, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_bkey_in_btree_node(struct bch_fs *, struct btree *, struct bkey_s_c, struct printbuf *); diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index a219a87e9ef0..74e1ff225674 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -790,7 +790,7 @@ static noinline void bch2_drop_overwrites_from_journal(struct btree_trans *trans } static noinline int bch2_trans_commit_bkey_invalid(struct btree_trans *trans, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct btree_insert_entry *i, struct printbuf *err) { @@ -1025,10 +1025,10 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) trans_for_each_update(trans, i) { struct printbuf buf = PRINTBUF; - enum bkey_invalid_flags invalid_flags = 0; + enum bch_validate_flags invalid_flags = 0; if (!(flags & BCH_TRANS_COMMIT_no_journal_res)) - invalid_flags |= BKEY_INVALID_WRITE|BKEY_INVALID_COMMIT; + invalid_flags |= BCH_VALIDATE_write|BCH_VALIDATE_commit; if (unlikely(bch2_bkey_invalid(c, bkey_i_to_s_c(i->k), i->bkey_type, invalid_flags, &buf))) @@ -1043,10 +1043,10 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) for (struct jset_entry *i = trans->journal_entries; i != (void *) ((u64 *) trans->journal_entries + trans->journal_entries_u64s); i = vstruct_next(i)) { - enum bkey_invalid_flags invalid_flags = 0; + enum bch_validate_flags invalid_flags = 0; if (!(flags & BCH_TRANS_COMMIT_no_journal_res)) - invalid_flags |= BKEY_INVALID_WRITE|BKEY_INVALID_COMMIT; + invalid_flags |= BCH_VALIDATE_write|BCH_VALIDATE_commit; if (unlikely(bch2_journal_entry_validate(c, NULL, i, bcachefs_metadata_version_current, diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c index dd9de8e5a28a..6bbf9a7d9e4d 100644 --- a/fs/bcachefs/dirent.c +++ b/fs/bcachefs/dirent.c @@ -98,7 +98,7 @@ const struct bch_hash_desc bch2_dirent_hash_desc = { }; int bch2_dirent_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_dirent d = bkey_s_c_to_dirent(k); @@ -118,7 +118,7 @@ int bch2_dirent_invalid(struct bch_fs *c, struct bkey_s_c k, * Check new keys don't exceed the max length * (older keys may be larger.) */ - bkey_fsck_err_on((flags & BKEY_INVALID_COMMIT) && d_name.len > BCH_NAME_MAX, c, err, + bkey_fsck_err_on((flags & BCH_VALIDATE_commit) && d_name.len > BCH_NAME_MAX, c, err, dirent_name_too_long, "dirent name too big (%u > %u)", d_name.len, BCH_NAME_MAX); diff --git a/fs/bcachefs/dirent.h b/fs/bcachefs/dirent.h index 635ceb3c07fb..24037e6e0a09 100644 --- a/fs/bcachefs/dirent.h +++ b/fs/bcachefs/dirent.h @@ -4,11 +4,11 @@ #include "str_hash.h" -enum bkey_invalid_flags; +enum bch_validate_flags; extern const struct bch_hash_desc bch2_dirent_hash_desc; int bch2_dirent_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_dirent_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_dirent ((struct bkey_ops) { \ diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 055be0ca9250..b26dc7424662 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -107,7 +107,7 @@ struct ec_bio { /* Stripes btree keys: */ int bch2_stripe_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { const struct bch_stripe *s = bkey_s_c_to_stripe(k).v; diff --git a/fs/bcachefs/ec.h b/fs/bcachefs/ec.h index a1c2bc74ef45..84a23eeb6249 100644 --- a/fs/bcachefs/ec.h +++ b/fs/bcachefs/ec.h @@ -6,10 +6,10 @@ #include "buckets_types.h" #include "extents_types.h" -enum bkey_invalid_flags; +enum bch_validate_flags; int bch2_stripe_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_stripe_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_stripe(struct btree_trans *, enum btree_id, unsigned, diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 094caf1cdb35..469037929685 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -172,7 +172,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, /* KEY_TYPE_btree_ptr: */ int bch2_btree_ptr_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; @@ -193,7 +193,7 @@ void bch2_btree_ptr_to_text(struct printbuf *out, struct bch_fs *c, } int bch2_btree_ptr_v2_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_btree_ptr_v2 bp = bkey_s_c_to_btree_ptr_v2(k); @@ -208,7 +208,7 @@ int bch2_btree_ptr_v2_invalid(struct bch_fs *c, struct bkey_s_c k, c, err, btree_ptr_v2_min_key_bad, "min_key > key"); - if (flags & BKEY_INVALID_WRITE) + if (flags & BCH_VALIDATE_write) bkey_fsck_err_on(!bp.v->sectors_written, c, err, btree_ptr_v2_written_0, "sectors_written == 0"); @@ -400,7 +400,7 @@ bool bch2_extent_merge(struct bch_fs *c, struct bkey_s l, struct bkey_s_c r) /* KEY_TYPE_reservation: */ int bch2_reservation_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_reservation r = bkey_s_c_to_reservation(k); @@ -1095,7 +1095,7 @@ void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, static int extent_ptr_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, const struct bch_extent_ptr *ptr, unsigned size_ondisk, bool metadata, @@ -1138,7 +1138,7 @@ fsck_err: } int bch2_bkey_ptrs_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); @@ -1214,7 +1214,7 @@ int bch2_bkey_ptrs_invalid(struct bch_fs *c, struct bkey_s_c k, bkey_fsck_err_on(crc_is_encoded(crc) && (crc.uncompressed_size > c->opts.encoded_extent_max >> 9) && - (flags & (BKEY_INVALID_WRITE|BKEY_INVALID_COMMIT)), c, err, + (flags & (BCH_VALIDATE_write|BCH_VALIDATE_commit)), c, err, ptr_crc_uncompressed_size_too_big, "too large encoded extent"); diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h index 10e1817e83e3..1ade959652b2 100644 --- a/fs/bcachefs/extents.h +++ b/fs/bcachefs/extents.h @@ -8,7 +8,7 @@ struct bch_fs; struct btree_trans; -enum bkey_invalid_flags; +enum bch_validate_flags; /* extent entries: */ @@ -406,12 +406,12 @@ int bch2_bkey_pick_read_device(struct bch_fs *, struct bkey_s_c, /* KEY_TYPE_btree_ptr: */ int bch2_btree_ptr_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_btree_ptr_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_btree_ptr_v2_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_btree_ptr_v2_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); void bch2_btree_ptr_v2_compat(enum btree_id, unsigned, unsigned, int, struct bkey_s); @@ -448,7 +448,7 @@ bool bch2_extent_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); /* KEY_TYPE_reservation: */ int bch2_reservation_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_reservation_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); bool bch2_reservation_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); @@ -680,7 +680,7 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *, const struct void bch2_bkey_ptrs_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_bkey_ptrs_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_ptr_swab(struct bkey_s); diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index 91572e05d5f0..aafa79fa6351 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -473,7 +473,7 @@ fsck_err: } int bch2_inode_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_inode inode = bkey_s_c_to_inode(k); @@ -490,7 +490,7 @@ fsck_err: } int bch2_inode_v2_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_inode_v2 inode = bkey_s_c_to_inode_v2(k); @@ -507,7 +507,7 @@ fsck_err: } int bch2_inode_v3_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_inode_v3 inode = bkey_s_c_to_inode_v3(k); @@ -635,7 +635,7 @@ int bch2_trigger_inode(struct btree_trans *trans, } int bch2_inode_generation_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h index abdfb1f3e4d0..679f5f5e5d15 100644 --- a/fs/bcachefs/inode.h +++ b/fs/bcachefs/inode.h @@ -6,15 +6,15 @@ #include "bkey_methods.h" #include "opts.h" -enum bkey_invalid_flags; +enum bch_validate_flags; extern const char * const bch2_inode_opts[]; int bch2_inode_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_inode_v2_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_inode_v3_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_inode_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_inode(struct btree_trans *, enum btree_id, unsigned, @@ -50,7 +50,7 @@ static inline bool bkey_is_inode(const struct bkey *k) } int bch2_inode_generation_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_inode_generation_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_inode_generation ((struct bkey_ops) { \ diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 56f4d46ef094..cdcb1ad49af4 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -299,7 +299,7 @@ static void journal_entry_err_msg(struct printbuf *out, journal_entry_err_msg(&_buf, version, jset, entry); \ prt_printf(&_buf, msg, ##__VA_ARGS__); \ \ - switch (flags & BKEY_INVALID_WRITE) { \ + switch (flags & BCH_VALIDATE_write) { \ case READ: \ mustfix_fsck_err(c, _err, "%s", _buf.buf); \ break; \ @@ -328,9 +328,9 @@ static int journal_validate_key(struct bch_fs *c, unsigned level, enum btree_id btree_id, struct bkey_i *k, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { - int write = flags & BKEY_INVALID_WRITE; + int write = flags & BCH_VALIDATE_write; void *next = vstruct_next(entry); struct printbuf buf = PRINTBUF; int ret = 0; @@ -403,7 +403,7 @@ static int journal_entry_btree_keys_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { struct bkey_i *k = entry->start; @@ -412,7 +412,7 @@ static int journal_entry_btree_keys_validate(struct bch_fs *c, entry->level, entry->btree_id, k, version, big_endian, - flags|BKEY_INVALID_JOURNAL); + flags|BCH_VALIDATE_journal); if (ret == FSCK_DELETED_KEY) continue; @@ -443,7 +443,7 @@ static int journal_entry_btree_root_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { struct bkey_i *k = entry->start; int ret = 0; @@ -482,7 +482,7 @@ static int journal_entry_prio_ptrs_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { /* obsolete, don't care: */ return 0; @@ -497,7 +497,7 @@ static int journal_entry_blacklist_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { int ret = 0; @@ -524,7 +524,7 @@ static int journal_entry_blacklist_v2_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { struct jset_entry_blacklist_v2 *bl_entry; int ret = 0; @@ -566,7 +566,7 @@ static int journal_entry_usage_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { struct jset_entry_usage *u = container_of(entry, struct jset_entry_usage, entry); @@ -600,7 +600,7 @@ static int journal_entry_data_usage_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { struct jset_entry_data_usage *u = container_of(entry, struct jset_entry_data_usage, entry); @@ -644,7 +644,7 @@ static int journal_entry_clock_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { struct jset_entry_clock *clock = container_of(entry, struct jset_entry_clock, entry); @@ -684,7 +684,7 @@ static int journal_entry_dev_usage_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { struct jset_entry_dev_usage *u = container_of(entry, struct jset_entry_dev_usage, entry); @@ -735,7 +735,7 @@ static int journal_entry_log_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { return 0; } @@ -753,7 +753,7 @@ static int journal_entry_overwrite_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { return journal_entry_btree_keys_validate(c, jset, entry, version, big_endian, READ); @@ -769,7 +769,7 @@ static int journal_entry_write_buffer_keys_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { return journal_entry_btree_keys_validate(c, jset, entry, version, big_endian, READ); @@ -785,7 +785,7 @@ static int journal_entry_datetime_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { unsigned bytes = vstruct_bytes(entry); unsigned expected = 16; @@ -815,7 +815,7 @@ static void journal_entry_datetime_to_text(struct printbuf *out, struct bch_fs * struct jset_entry_ops { int (*validate)(struct bch_fs *, struct jset *, struct jset_entry *, unsigned, int, - enum bkey_invalid_flags); + enum bch_validate_flags); void (*to_text)(struct printbuf *, struct bch_fs *, struct jset_entry *); }; @@ -833,7 +833,7 @@ int bch2_journal_entry_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { return entry->type < BCH_JSET_ENTRY_NR ? bch2_jset_entry_ops[entry->type].validate(c, jset, entry, @@ -853,7 +853,7 @@ void bch2_journal_entry_to_text(struct printbuf *out, struct bch_fs *c, } static int jset_validate_entries(struct bch_fs *c, struct jset *jset, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { unsigned version = le32_to_cpu(jset->version); int ret = 0; @@ -879,7 +879,7 @@ fsck_err: static int jset_validate(struct bch_fs *c, struct bch_dev *ca, struct jset *jset, u64 sector, - enum bkey_invalid_flags flags) + enum bch_validate_flags flags) { unsigned version; int ret = 0; @@ -934,7 +934,7 @@ static int jset_validate_early(struct bch_fs *c, { size_t bytes = vstruct_bytes(jset); unsigned version; - enum bkey_invalid_flags flags = BKEY_INVALID_JOURNAL; + enum bch_validate_flags flags = BCH_VALIDATE_journal; int ret = 0; if (le64_to_cpu(jset->magic) != jset_magic(c)) @@ -1235,7 +1235,7 @@ int bch2_journal_read(struct bch_fs *c, * those entries will be blacklisted: */ genradix_for_each_reverse(&c->journal_entries, radix_iter, _i) { - enum bkey_invalid_flags flags = BKEY_INVALID_JOURNAL; + enum bch_validate_flags flags = BCH_VALIDATE_journal; i = *_i; diff --git a/fs/bcachefs/journal_io.h b/fs/bcachefs/journal_io.h index 43cfed58ef33..2ca9cde30ea8 100644 --- a/fs/bcachefs/journal_io.h +++ b/fs/bcachefs/journal_io.h @@ -63,7 +63,7 @@ static inline struct jset_entry *__jset_entry_type_next(struct jset *jset, int bch2_journal_entry_validate(struct bch_fs *, struct jset *, struct jset_entry *, unsigned, int, - enum bkey_invalid_flags); + enum bch_validate_flags); void bch2_journal_entry_to_text(struct printbuf *, struct bch_fs *, struct jset_entry *); diff --git a/fs/bcachefs/lru.c b/fs/bcachefs/lru.c index 19285fcd252c..a40d116224ed 100644 --- a/fs/bcachefs/lru.c +++ b/fs/bcachefs/lru.c @@ -11,7 +11,7 @@ /* KEY_TYPE_lru is obsolete: */ int bch2_lru_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; diff --git a/fs/bcachefs/lru.h b/fs/bcachefs/lru.h index 429dca816df5..fb11ab0dd00e 100644 --- a/fs/bcachefs/lru.h +++ b/fs/bcachefs/lru.h @@ -49,7 +49,7 @@ static inline enum bch_lru_type lru_type(struct bkey_s_c l) } int bch2_lru_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_lru_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); void bch2_lru_pos_to_text(struct printbuf *, struct bpos); diff --git a/fs/bcachefs/quota.c b/fs/bcachefs/quota.c index 9d7498e774e2..d86381e9ac86 100644 --- a/fs/bcachefs/quota.c +++ b/fs/bcachefs/quota.c @@ -60,7 +60,7 @@ const struct bch_sb_field_ops bch_sb_field_ops_quota = { }; int bch2_quota_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; diff --git a/fs/bcachefs/quota.h b/fs/bcachefs/quota.h index 884f601f41c4..02d37a332218 100644 --- a/fs/bcachefs/quota.h +++ b/fs/bcachefs/quota.h @@ -5,11 +5,11 @@ #include "inode.h" #include "quota_types.h" -enum bkey_invalid_flags; +enum bch_validate_flags; extern const struct bch_sb_field_ops bch_sb_field_ops_quota; int bch2_quota_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_quota_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_quota ((struct bkey_ops) { \ diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index fb31a68148ff..9ac6cf21cfbf 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -30,7 +30,7 @@ static inline unsigned bkey_type_to_indirect(const struct bkey *k) /* reflink pointers */ int bch2_reflink_p_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k); @@ -257,7 +257,7 @@ int bch2_trigger_reflink_p(struct btree_trans *trans, /* indirect extents */ int bch2_reflink_v_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { return bch2_bkey_ptrs_invalid(c, k, flags, err); @@ -312,7 +312,7 @@ int bch2_trigger_reflink_v(struct btree_trans *trans, /* indirect inline data */ int bch2_indirect_inline_data_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { return 0; diff --git a/fs/bcachefs/reflink.h b/fs/bcachefs/reflink.h index 59d4dbae3fb1..e894f3a2c67a 100644 --- a/fs/bcachefs/reflink.h +++ b/fs/bcachefs/reflink.h @@ -2,10 +2,10 @@ #ifndef _BCACHEFS_REFLINK_H #define _BCACHEFS_REFLINK_H -enum bkey_invalid_flags; +enum bch_validate_flags; int bch2_reflink_p_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_reflink_p_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); bool bch2_reflink_p_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); @@ -22,7 +22,7 @@ int bch2_trigger_reflink_p(struct btree_trans *, enum btree_id, unsigned, }) int bch2_reflink_v_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_reflink_v_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_reflink_v(struct btree_trans *, enum btree_id, unsigned, @@ -38,7 +38,7 @@ int bch2_trigger_reflink_v(struct btree_trans *, enum btree_id, unsigned, }) int bch2_indirect_inline_data_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_indirect_inline_data_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_indirect_inline_data(struct btree_trans *, diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index 3aa4686cc71f..629900a5e641 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -32,7 +32,7 @@ void bch2_snapshot_tree_to_text(struct printbuf *out, struct bch_fs *c, } int bch2_snapshot_tree_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; @@ -223,7 +223,7 @@ void bch2_snapshot_to_text(struct printbuf *out, struct bch_fs *c, } int bch2_snapshot_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_snapshot s; diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index b3c43d579540..bd5d74269d15 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -2,11 +2,11 @@ #ifndef _BCACHEFS_SNAPSHOT_H #define _BCACHEFS_SNAPSHOT_H -enum bkey_invalid_flags; +enum bch_validate_flags; void bch2_snapshot_tree_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_snapshot_tree_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); #define bch2_bkey_ops_snapshot_tree ((struct bkey_ops) { \ .key_invalid = bch2_snapshot_tree_invalid, \ @@ -20,7 +20,7 @@ int bch2_snapshot_tree_lookup(struct btree_trans *, u32, struct bch_snapshot_tre void bch2_snapshot_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_snapshot_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); int bch2_mark_snapshot(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, enum btree_iter_update_trigger_flags); diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index 3341285e7531..132213761ef6 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -208,7 +208,7 @@ int bch2_check_subvol_children(struct bch_fs *c) /* Subvolumes: */ int bch2_subvolume_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; diff --git a/fs/bcachefs/subvolume.h b/fs/bcachefs/subvolume.h index 1530a10327ab..afa5e871efb2 100644 --- a/fs/bcachefs/subvolume.h +++ b/fs/bcachefs/subvolume.h @@ -5,13 +5,13 @@ #include "darray.h" #include "subvolume_types.h" -enum bkey_invalid_flags; +enum bch_validate_flags; int bch2_check_subvols(struct bch_fs *); int bch2_check_subvol_children(struct bch_fs *); int bch2_subvolume_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_subvolume_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_subvolume_trigger(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, diff --git a/fs/bcachefs/xattr.c b/fs/bcachefs/xattr.c index 786ccd7a7662..c11bf6dacc2c 100644 --- a/fs/bcachefs/xattr.c +++ b/fs/bcachefs/xattr.c @@ -71,7 +71,7 @@ const struct bch_hash_desc bch2_xattr_hash_desc = { }; int bch2_xattr_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bkey_invalid_flags flags, + enum bch_validate_flags flags, struct printbuf *err) { struct bkey_s_c_xattr xattr = bkey_s_c_to_xattr(k); diff --git a/fs/bcachefs/xattr.h b/fs/bcachefs/xattr.h index 1337f31a5c49..1574b9eb4c85 100644 --- a/fs/bcachefs/xattr.h +++ b/fs/bcachefs/xattr.h @@ -7,7 +7,7 @@ extern const struct bch_hash_desc bch2_xattr_hash_desc; int bch2_xattr_invalid(struct bch_fs *, struct bkey_s_c, - enum bkey_invalid_flags, struct printbuf *); + enum bch_validate_flags, struct printbuf *); void bch2_xattr_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_xattr ((struct bkey_ops) { \ -- cgit v1.2.3-59-g8ed1b From a5c3e265d3b61ab661df4f259a97b57840cb041e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 8 May 2024 18:49:14 -0400 Subject: bcachefs: Plumb bch_validate_flags to sb_field_ops.validate() Signed-off-by: Kent Overstreet --- fs/bcachefs/checksum.c | 5 ++--- fs/bcachefs/disk_groups.c | 5 ++--- fs/bcachefs/journal_sb.c | 10 ++++------ fs/bcachefs/journal_seq_blacklist.c | 5 ++--- fs/bcachefs/quota.c | 5 ++--- fs/bcachefs/replicas.c | 4 ++-- fs/bcachefs/sb-clean.c | 5 ++--- fs/bcachefs/sb-counters.c | 5 ++--- fs/bcachefs/sb-downgrade.c | 2 +- fs/bcachefs/sb-errors.c | 2 +- fs/bcachefs/sb-members.c | 10 ++++------ fs/bcachefs/super-io.c | 24 ++++++++++++------------ fs/bcachefs/super-io.h | 3 ++- 13 files changed, 38 insertions(+), 47 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index 2b3677ae0942..85198f391e9c 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -469,9 +469,8 @@ int bch2_rechecksum_bio(struct bch_fs *c, struct bio *bio, /* BCH_SB_FIELD_crypt: */ -static int bch2_sb_crypt_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_crypt_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_crypt *crypt = field_to_type(f, crypt); diff --git a/fs/bcachefs/disk_groups.c b/fs/bcachefs/disk_groups.c index 1cb563dfa017..521a86df5e52 100644 --- a/fs/bcachefs/disk_groups.c +++ b/fs/bcachefs/disk_groups.c @@ -18,9 +18,8 @@ static int group_cmp(const void *_l, const void *_r) strncmp(l->label, r->label, sizeof(l->label)); } -static int bch2_sb_disk_groups_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_disk_groups_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_disk_groups *groups = field_to_type(f, disk_groups); diff --git a/fs/bcachefs/journal_sb.c b/fs/bcachefs/journal_sb.c index ae4fb8c3a2bc..db80e506e3ab 100644 --- a/fs/bcachefs/journal_sb.c +++ b/fs/bcachefs/journal_sb.c @@ -16,9 +16,8 @@ static int u64_cmp(const void *_l, const void *_r) return cmp_int(*l, *r); } -static int bch2_sb_journal_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_journal_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_journal *journal = field_to_type(f, journal); struct bch_member m = bch2_sb_member_get(sb, sb->dev_idx); @@ -99,9 +98,8 @@ static int u64_range_cmp(const void *_l, const void *_r) return cmp_int(l->start, r->start); } -static int bch2_sb_journal_v2_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_journal_v2_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_journal_v2 *journal = field_to_type(f, journal_v2); struct bch_member m = bch2_sb_member_get(sb, sb->dev_idx); diff --git a/fs/bcachefs/journal_seq_blacklist.c b/fs/bcachefs/journal_seq_blacklist.c index d8d40d46a27a..ed4846709611 100644 --- a/fs/bcachefs/journal_seq_blacklist.c +++ b/fs/bcachefs/journal_seq_blacklist.c @@ -162,9 +162,8 @@ int bch2_blacklist_table_initialize(struct bch_fs *c) return 0; } -static int bch2_sb_journal_seq_blacklist_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_journal_seq_blacklist_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_journal_seq_blacklist *bl = field_to_type(f, journal_seq_blacklist); diff --git a/fs/bcachefs/quota.c b/fs/bcachefs/quota.c index d86381e9ac86..a0cca8b70e0a 100644 --- a/fs/bcachefs/quota.c +++ b/fs/bcachefs/quota.c @@ -20,7 +20,7 @@ static const char * const bch2_quota_counters[] = { }; static int bch2_sb_quota_validate(struct bch_sb *sb, struct bch_sb_field *f, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_quota *q = field_to_type(f, quota); @@ -60,8 +60,7 @@ const struct bch_sb_field_ops bch_sb_field_ops_quota = { }; int bch2_quota_invalid(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { int ret = 0; diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c index f8ff7c8bb05e..bd1d5d085e23 100644 --- a/fs/bcachefs/replicas.c +++ b/fs/bcachefs/replicas.c @@ -860,7 +860,7 @@ static int bch2_cpu_replicas_validate(struct bch_replicas_cpu *cpu_r, } static int bch2_sb_replicas_validate(struct bch_sb *sb, struct bch_sb_field *f, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_replicas *sb_r = field_to_type(f, replicas); struct bch_replicas_cpu cpu_r; @@ -899,7 +899,7 @@ const struct bch_sb_field_ops bch_sb_field_ops_replicas = { }; static int bch2_sb_replicas_v0_validate(struct bch_sb *sb, struct bch_sb_field *f, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_replicas_v0 *sb_r = field_to_type(f, replicas_v0); struct bch_replicas_cpu cpu_r; diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index e2beb2308c70..47f10ab57f40 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -266,9 +266,8 @@ void bch2_journal_super_entries_add_common(struct bch_fs *c, } } -static int bch2_sb_clean_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_clean_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_clean *clean = field_to_type(f, clean); diff --git a/fs/bcachefs/sb-counters.c b/fs/bcachefs/sb-counters.c index 7282f7886192..6992e7469112 100644 --- a/fs/bcachefs/sb-counters.c +++ b/fs/bcachefs/sb-counters.c @@ -20,9 +20,8 @@ static size_t bch2_sb_counter_nr_entries(struct bch_sb_field_counters *ctrs) return (__le64 *) vstruct_end(&ctrs->field) - &ctrs->d[0]; }; -static int bch2_sb_counters_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_counters_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { return 0; }; diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c index 90b06f8aa1df..7d2abadbd6e3 100644 --- a/fs/bcachefs/sb-downgrade.c +++ b/fs/bcachefs/sb-downgrade.c @@ -138,7 +138,7 @@ downgrade_entry_next_c(const struct bch_sb_field_downgrade_entry *e) _i = downgrade_entry_next_c(_i)) static int bch2_sb_downgrade_validate(struct bch_sb *sb, struct bch_sb_field *f, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_downgrade *e = field_to_type(f, downgrade); diff --git a/fs/bcachefs/sb-errors.c b/fs/bcachefs/sb-errors.c index 5f5bcae391fb..bda33e59e226 100644 --- a/fs/bcachefs/sb-errors.c +++ b/fs/bcachefs/sb-errors.c @@ -30,7 +30,7 @@ static inline unsigned bch2_sb_field_errors_u64s(unsigned nr) } static int bch2_sb_errors_validate(struct bch_sb *sb, struct bch_sb_field *f, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_errors *e = field_to_type(f, errors); unsigned i, nr = bch2_sb_field_errors_nr_entries(e); diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 8f197bb088a0..39196f2a4197 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -261,9 +261,8 @@ static void member_to_text(struct printbuf *out, printbuf_indent_sub(out, 2); } -static int bch2_sb_members_v1_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_members_v1_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_members_v1 *mi = field_to_type(f, members_v1); unsigned i; @@ -311,9 +310,8 @@ static void bch2_sb_members_v2_to_text(struct printbuf *out, struct bch_sb *sb, member_to_text(out, members_v2_get(mi, i), gi, sb, i); } -static int bch2_sb_members_v2_validate(struct bch_sb *sb, - struct bch_sb_field *f, - struct printbuf *err) +static int bch2_sb_members_v2_validate(struct bch_sb *sb, struct bch_sb_field *f, + enum bch_validate_flags flags, struct printbuf *err) { struct bch_sb_field_members_v2 *mi = field_to_type(f, members_v2); size_t mi_bytes = (void *) __bch2_members_v2_get_mut(mi, sb->nr_devices) - diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index fb4574958681..f1bee6c5222d 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -76,7 +76,7 @@ const char * const bch2_sb_fields[] = { }; static int bch2_sb_field_validate(struct bch_sb *, struct bch_sb_field *, - struct printbuf *); + enum bch_validate_flags, struct printbuf *); struct bch_sb_field *bch2_sb_field_get_id(struct bch_sb *sb, enum bch_sb_field_type type) @@ -344,8 +344,8 @@ static int bch2_sb_compatible(struct bch_sb *sb, struct printbuf *out) return 0; } -static int bch2_sb_validate(struct bch_sb_handle *disk_sb, struct printbuf *out, - int rw) +static int bch2_sb_validate(struct bch_sb_handle *disk_sb, + enum bch_validate_flags flags, struct printbuf *out) { struct bch_sb *sb = disk_sb->sb; struct bch_sb_field_members_v1 *mi; @@ -401,7 +401,7 @@ static int bch2_sb_validate(struct bch_sb_handle *disk_sb, struct printbuf *out, return -BCH_ERR_invalid_sb_time_precision; } - if (rw == READ) { + if (!flags) { /* * Been seeing a bug where these are getting inexplicably * zeroed, so we're now validating them, but we have to be @@ -457,7 +457,7 @@ static int bch2_sb_validate(struct bch_sb_handle *disk_sb, struct printbuf *out, return -BCH_ERR_invalid_sb_members_missing; } - ret = bch2_sb_field_validate(sb, &mi->field, out); + ret = bch2_sb_field_validate(sb, &mi->field, flags, out); if (ret) return ret; @@ -465,12 +465,12 @@ static int bch2_sb_validate(struct bch_sb_handle *disk_sb, struct printbuf *out, if (le32_to_cpu(f->type) == BCH_SB_FIELD_members_v1) continue; - ret = bch2_sb_field_validate(sb, f, out); + ret = bch2_sb_field_validate(sb, f, flags, out); if (ret) return ret; } - if (rw == WRITE && + if ((flags & BCH_VALIDATE_write) && bch2_sb_member_get(sb, sb->dev_idx).seq != sb->seq) { prt_printf(out, "Invalid superblock: member seq %llu != sb seq %llu", le64_to_cpu(bch2_sb_member_get(sb, sb->dev_idx).seq), @@ -819,7 +819,7 @@ got_super: sb->have_layout = true; - ret = bch2_sb_validate(sb, &err, READ); + ret = bch2_sb_validate(sb, 0, &err); if (ret) { bch2_print_opts(opts, KERN_ERR "bcachefs (%s): error validating superblock: %s\n", path, err.buf); @@ -975,7 +975,7 @@ int bch2_write_super(struct bch_fs *c) darray_for_each(online_devices, ca) { printbuf_reset(&err); - ret = bch2_sb_validate(&(*ca)->disk_sb, &err, WRITE); + ret = bch2_sb_validate(&(*ca)->disk_sb, BCH_VALIDATE_write, &err); if (ret) { bch2_fs_inconsistent(c, "sb invalid before write: %s", err.buf); goto out; @@ -1161,7 +1161,7 @@ void bch2_sb_upgrade(struct bch_fs *c, unsigned new_version) } static int bch2_sb_ext_validate(struct bch_sb *sb, struct bch_sb_field *f, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { if (vstruct_bytes(f) < 88) { prt_printf(err, "field too small (%zu < %u)", vstruct_bytes(f), 88); @@ -1219,14 +1219,14 @@ static const struct bch_sb_field_ops *bch2_sb_field_type_ops(unsigned type) } static int bch2_sb_field_validate(struct bch_sb *sb, struct bch_sb_field *f, - struct printbuf *err) + enum bch_validate_flags flags, struct printbuf *err) { unsigned type = le32_to_cpu(f->type); struct printbuf field_err = PRINTBUF; const struct bch_sb_field_ops *ops = bch2_sb_field_type_ops(type); int ret; - ret = ops->validate ? ops->validate(sb, f, &field_err) : 0; + ret = ops->validate ? ops->validate(sb, f, flags, &field_err) : 0; if (ret) { prt_printf(err, "Invalid superblock section %s: %s", bch2_sb_fields[type], field_err.buf); diff --git a/fs/bcachefs/super-io.h b/fs/bcachefs/super-io.h index 95e80e06316b..fadd364e2802 100644 --- a/fs/bcachefs/super-io.h +++ b/fs/bcachefs/super-io.h @@ -51,7 +51,8 @@ void bch2_sb_field_delete(struct bch_sb_handle *, enum bch_sb_field_type); extern const char * const bch2_sb_fields[]; struct bch_sb_field_ops { - int (*validate)(struct bch_sb *, struct bch_sb_field *, struct printbuf *); + int (*validate)(struct bch_sb *, struct bch_sb_field *, + enum bch_validate_flags, struct printbuf *); void (*to_text)(struct printbuf *, struct bch_sb *, struct bch_sb_field *); }; -- cgit v1.2.3-59-g8ed1b From 692aa7a54b2b28d59f24b3bf8250837805484b99 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 6 May 2024 09:16:33 -0400 Subject: bcachefs: Fix sb_field_downgrade validation - bch2_sb_downgrade_validate() wasn't checking for a downgrade entry extending past the end of the superblock section - for_each_downgrade_entry() is used in to_text() and needs to work on malformed input; it also was missing a check for a field extending past the end of the section Reported-by: syzbot+e49ccab73449180bc9be@syzkaller.appspotmail.com Fixes: 84f1638795da ("bcachefs: bch_sb_field_downgrade") Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-downgrade.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c index 7d2abadbd6e3..390a1bbd2567 100644 --- a/fs/bcachefs/sb-downgrade.c +++ b/fs/bcachefs/sb-downgrade.c @@ -134,7 +134,8 @@ downgrade_entry_next_c(const struct bch_sb_field_downgrade_entry *e) #define for_each_downgrade_entry(_d, _i) \ for (const struct bch_sb_field_downgrade_entry *_i = (_d)->entries; \ (void *) _i < vstruct_end(&(_d)->field) && \ - (void *) &_i->errors[0] < vstruct_end(&(_d)->field); \ + (void *) &_i->errors[0] <= vstruct_end(&(_d)->field) && \ + (void *) downgrade_entry_next_c(_i) <= vstruct_end(&(_d)->field); \ _i = downgrade_entry_next_c(_i)) static int bch2_sb_downgrade_validate(struct bch_sb *sb, struct bch_sb_field *f, @@ -142,7 +143,16 @@ static int bch2_sb_downgrade_validate(struct bch_sb *sb, struct bch_sb_field *f, { struct bch_sb_field_downgrade *e = field_to_type(f, downgrade); - for_each_downgrade_entry(e, i) { + for (const struct bch_sb_field_downgrade_entry *i = e->entries; + (void *) i < vstruct_end(&e->field); + i = downgrade_entry_next_c(i)) { + if (flags & BCH_VALIDATE_write && + ((void *) &i->errors[0] > vstruct_end(&e->field) || + (void *) downgrade_entry_next_c(i) > vstruct_end(&e->field))) { + prt_printf(err, "downgrade entry overruns end of superblock section)"); + return -BCH_ERR_invalid_sb_downgrade; + } + if (BCH_VERSION_MAJOR(le16_to_cpu(i->version)) != BCH_VERSION_MAJOR(le16_to_cpu(sb->version))) { prt_printf(err, "downgrade entry with mismatched major version (%u != %u)", -- cgit v1.2.3-59-g8ed1b From bceacfa97ec8b67a76efad2f95899434230b317c Mon Sep 17 00:00:00 2001 From: Daniel Hill Date: Fri, 30 Sep 2022 16:37:15 +1300 Subject: bcachefs: add counters for failed shrinker reclaim This adds distinct counters for every reason the btree node shrinker can fail to free an object - if our shrinker isn't making progress, this will tell us why. Signed-off-by: Daniel Hill Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 88 ++++++++++++++++++++++++++++++++++++----------- fs/bcachefs/btree_cache.h | 2 +- fs/bcachefs/btree_types.h | 10 ++++++ fs/bcachefs/sysfs.c | 2 +- 4 files changed, 80 insertions(+), 22 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 7dd35d6224d9..9e4ed75d3675 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -16,6 +16,12 @@ #include #include +#define BTREE_CACHE_NOT_FREED_INCREMENT(counter) \ +do { \ + if (shrinker_counter) \ + bc->not_freed_##counter++; \ +} while (0) + const char * const bch2_btree_node_flags[] = { #define x(f) #f, BTREE_FLAGS() @@ -238,7 +244,7 @@ static inline struct btree *btree_cache_find(struct btree_cache *bc, * this version is for btree nodes that have already been freed (we're not * reaping a real btree node) */ -static int __btree_node_reclaim(struct bch_fs *c, struct btree *b, bool flush) +static int __btree_node_reclaim(struct bch_fs *c, struct btree *b, bool flush, bool shrinker_counter) { struct btree_cache *bc = &c->btree_cache; int ret = 0; @@ -260,38 +266,64 @@ wait_on_io: if (b->flags & ((1U << BTREE_NODE_dirty)| (1U << BTREE_NODE_read_in_flight)| (1U << BTREE_NODE_write_in_flight))) { - if (!flush) + if (!flush) { + if (btree_node_dirty(b)) + BTREE_CACHE_NOT_FREED_INCREMENT(dirty); + else if (btree_node_read_in_flight(b)) + BTREE_CACHE_NOT_FREED_INCREMENT(read_in_flight); + else if (btree_node_write_in_flight(b)) + BTREE_CACHE_NOT_FREED_INCREMENT(write_in_flight); return -BCH_ERR_ENOMEM_btree_node_reclaim; + } /* XXX: waiting on IO with btree cache lock held */ bch2_btree_node_wait_on_read(b); bch2_btree_node_wait_on_write(b); } - if (!six_trylock_intent(&b->c.lock)) + if (!six_trylock_intent(&b->c.lock)) { + BTREE_CACHE_NOT_FREED_INCREMENT(lock_intent); return -BCH_ERR_ENOMEM_btree_node_reclaim; + } - if (!six_trylock_write(&b->c.lock)) + if (!six_trylock_write(&b->c.lock)) { + BTREE_CACHE_NOT_FREED_INCREMENT(lock_write); goto out_unlock_intent; + } /* recheck under lock */ if (b->flags & ((1U << BTREE_NODE_read_in_flight)| (1U << BTREE_NODE_write_in_flight))) { - if (!flush) + if (!flush) { + if (btree_node_read_in_flight(b)) + BTREE_CACHE_NOT_FREED_INCREMENT(read_in_flight); + else if (btree_node_write_in_flight(b)) + BTREE_CACHE_NOT_FREED_INCREMENT(write_in_flight); goto out_unlock; + } six_unlock_write(&b->c.lock); six_unlock_intent(&b->c.lock); goto wait_on_io; } - if (btree_node_noevict(b) || - btree_node_write_blocked(b) || - btree_node_will_make_reachable(b)) + if (btree_node_noevict(b)) { + BTREE_CACHE_NOT_FREED_INCREMENT(noevict); + goto out_unlock; + } + if (btree_node_write_blocked(b)) { + BTREE_CACHE_NOT_FREED_INCREMENT(write_blocked); goto out_unlock; + } + if (btree_node_will_make_reachable(b)) { + BTREE_CACHE_NOT_FREED_INCREMENT(will_make_reachable); + goto out_unlock; + } if (btree_node_dirty(b)) { - if (!flush) + if (!flush) { + BTREE_CACHE_NOT_FREED_INCREMENT(dirty); goto out_unlock; + } /* * Using the underscore version because we don't want to compact * bsets after the write, since this node is about to be evicted @@ -321,14 +353,14 @@ out_unlock_intent: goto out; } -static int btree_node_reclaim(struct bch_fs *c, struct btree *b) +static int btree_node_reclaim(struct bch_fs *c, struct btree *b, bool shrinker_counter) { - return __btree_node_reclaim(c, b, false); + return __btree_node_reclaim(c, b, false, shrinker_counter); } static int btree_node_write_and_reclaim(struct bch_fs *c, struct btree *b) { - return __btree_node_reclaim(c, b, true); + return __btree_node_reclaim(c, b, true, false); } static unsigned long bch2_btree_cache_scan(struct shrinker *shrink, @@ -376,11 +408,12 @@ static unsigned long bch2_btree_cache_scan(struct shrinker *shrink, if (touched >= nr) goto out; - if (!btree_node_reclaim(c, b)) { + if (!btree_node_reclaim(c, b, true)) { btree_node_data_free(c, b); six_unlock_write(&b->c.lock); six_unlock_intent(&b->c.lock); freed++; + bc->freed++; } } restart: @@ -389,9 +422,11 @@ restart: if (btree_node_accessed(b)) { clear_btree_node_accessed(b); - } else if (!btree_node_reclaim(c, b)) { + bc->not_freed_access_bit++; + } else if (!btree_node_reclaim(c, b, true)) { freed++; btree_node_data_free(c, b); + bc->freed++; bch2_btree_node_hash_remove(bc, b); six_unlock_write(&b->c.lock); @@ -599,7 +634,7 @@ static struct btree *btree_node_cannibalize(struct bch_fs *c) struct btree *b; list_for_each_entry_reverse(b, &bc->live, list) - if (!btree_node_reclaim(c, b)) + if (!btree_node_reclaim(c, b, false)) return b; while (1) { @@ -635,7 +670,7 @@ struct btree *bch2_btree_node_mem_alloc(struct btree_trans *trans, bool pcpu_rea * disk node. Check the freed list before allocating a new one: */ list_for_each_entry(b, freed, list) - if (!btree_node_reclaim(c, b)) { + if (!btree_node_reclaim(c, b, false)) { list_del_init(&b->list); goto got_node; } @@ -661,7 +696,7 @@ got_node: * the list. Check if there's any freed nodes there: */ list_for_each_entry(b2, &bc->freeable, list) - if (!btree_node_reclaim(c, b2)) { + if (!btree_node_reclaim(c, b2, false)) { swap(b->data, b2->data); swap(b->aux_data, b2->aux_data); btree_node_to_freedlist(bc, b2); @@ -1280,12 +1315,12 @@ static void prt_btree_cache_line(struct printbuf *out, const struct bch_fs *c, prt_printf(out, " (%u)\n", nr); } -void bch2_btree_cache_to_text(struct printbuf *out, const struct bch_fs *c) +void bch2_btree_cache_to_text(struct printbuf *out, const struct btree_cache *bc) { - const struct btree_cache *bc = &c->btree_cache; + struct bch_fs *c = container_of(bc, struct bch_fs, btree_cache); if (!out->nr_tabstops) - printbuf_tabstop_push(out, 24); + printbuf_tabstop_push(out, 32); prt_btree_cache_line(out, c, "total:", bc->used); prt_btree_cache_line(out, c, "nr dirty:", atomic_read(&bc->dirty)); @@ -1294,4 +1329,17 @@ void bch2_btree_cache_to_text(struct printbuf *out, const struct bch_fs *c) for (unsigned i = 0; i < ARRAY_SIZE(bc->used_by_btree); i++) prt_btree_cache_line(out, c, bch2_btree_id_str(i), bc->used_by_btree[i]); + + prt_newline(out); + prt_printf(out, "freed:\t%u\n", bc->freed); + prt_printf(out, "not freed:\n"); + prt_printf(out, " dirty\t%u\n", bc->not_freed_dirty); + prt_printf(out, " write in flight\t%u\n", bc->not_freed_write_in_flight); + prt_printf(out, " read in flight\t%u\n", bc->not_freed_read_in_flight); + prt_printf(out, " lock intent failed\t%u\n", bc->not_freed_lock_intent); + prt_printf(out, " lock write failed\t%u\n", bc->not_freed_lock_write); + prt_printf(out, " access bit\t%u\n", bc->not_freed_access_bit); + prt_printf(out, " no evict failed\t%u\n", bc->not_freed_noevict); + prt_printf(out, " write blocked\t%u\n", bc->not_freed_write_blocked); + prt_printf(out, " will make reachable\t%u\n", bc->not_freed_will_make_reachable); } diff --git a/fs/bcachefs/btree_cache.h b/fs/bcachefs/btree_cache.h index 6fe91d1c0fd4..fed35de3e4de 100644 --- a/fs/bcachefs/btree_cache.h +++ b/fs/bcachefs/btree_cache.h @@ -134,6 +134,6 @@ static inline struct btree *btree_node_root(struct bch_fs *c, struct btree *b) const char *bch2_btree_id_str(enum btree_id); void bch2_btree_pos_to_text(struct printbuf *, struct bch_fs *, const struct btree *); void bch2_btree_node_to_text(struct printbuf *, struct bch_fs *, const struct btree *); -void bch2_btree_cache_to_text(struct printbuf *, const struct bch_fs *); +void bch2_btree_cache_to_text(struct printbuf *, const struct btree_cache *); #endif /* _BCACHEFS_BTREE_CACHE_H */ diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index 76364bd4347e..d63db4fefe73 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -163,6 +163,16 @@ struct btree_cache { /* Number of elements in live + freeable lists */ unsigned used; unsigned reserve; + unsigned freed; + unsigned not_freed_lock_intent; + unsigned not_freed_lock_write; + unsigned not_freed_dirty; + unsigned not_freed_read_in_flight; + unsigned not_freed_write_in_flight; + unsigned not_freed_noevict; + unsigned not_freed_write_blocked; + unsigned not_freed_will_make_reachable; + unsigned not_freed_access_bit; atomic_t dirty; struct shrinker *shrink; diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index df3d28659bfd..93ca74d108b1 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -383,7 +383,7 @@ SHOW(bch2_fs) bch2_journal_debug_to_text(out, &c->journal); if (attr == &sysfs_btree_cache) - bch2_btree_cache_to_text(out, c); + bch2_btree_cache_to_text(out, &c->btree_cache); if (attr == &sysfs_btree_key_cache) bch2_btree_key_cache_to_text(out, &c->btree_key_cache); -- cgit v1.2.3-59-g8ed1b From 07f9a27f1969764d11374942961d51fee0ab628f Mon Sep 17 00:00:00 2001 From: Thomas Bertschinger Date: Thu, 9 May 2024 12:37:24 -0600 Subject: bcachefs: add no_invalid_checks flag Setting this flag on a filesystem results in validity checks being skipped when writing bkeys. This flag will be used by tooling that deliberately injects corruption into a filesystem in order to exercise fsck. It shouldn't be set outside of testing/debugging code. Signed-off-by: Thomas Bertschinger Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs.h | 3 ++- fs/bcachefs/bkey_methods.c | 6 ++++++ 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index ab2dc2c70da2..bc0ea2c4efef 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -637,7 +637,8 @@ struct bch_dev { x(error) \ x(topology_error) \ x(errors_fixed) \ - x(errors_not_fixed) + x(errors_not_fixed) \ + x(no_invalid_checks) enum bch_fs_flags { #define x(n) BCH_FS_##n, diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c index f692f9d8f00d..c2c3dae52186 100644 --- a/fs/bcachefs/bkey_methods.c +++ b/fs/bcachefs/bkey_methods.c @@ -126,6 +126,9 @@ int bch2_bkey_val_invalid(struct bch_fs *c, struct bkey_s_c k, enum bch_validate_flags flags, struct printbuf *err) { + if (test_bit(BCH_FS_no_invalid_checks, &c->flags)) + return 0; + const struct bkey_ops *ops = bch2_bkey_type_ops(k.k->type); int ret = 0; @@ -162,6 +165,9 @@ int __bch2_bkey_invalid(struct bch_fs *c, struct bkey_s_c k, enum bch_validate_flags flags, struct printbuf *err) { + if (test_bit(BCH_FS_no_invalid_checks, &c->flags)) + return 0; + int ret = 0; bkey_fsck_err_on(k.k->u64s < BKEY_U64s, c, err, -- cgit v1.2.3-59-g8ed1b From 186e7d71534df4589405925caca5597af7626c12 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 6 May 2024 18:41:36 +0800 Subject: f2fs: compress: fix to update i_compr_blocks correctly Previously, we account reserved blocks and compressed blocks into @compr_blocks, then, f2fs_i_compr_blocks_update(,compr_blocks) will update i_compr_blocks incorrectly, fix it. Meanwhile, for the case all blocks in cluster were reserved, fix to update dn->ofs_in_node correctly. Fixes: eb8fbaa53374 ("f2fs: compress: fix to check unreleased compressed cluster") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 72ce1a522fb2..13b430b54299 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -3666,7 +3666,8 @@ static int reserve_compress_blocks(struct dnode_of_data *dn, pgoff_t count, while (count) { int compr_blocks = 0; - blkcnt_t reserved; + blkcnt_t reserved = 0; + blkcnt_t to_reserved; int ret; for (i = 0; i < cluster_size; i++) { @@ -3686,20 +3687,26 @@ static int reserve_compress_blocks(struct dnode_of_data *dn, pgoff_t count, * fails in release_compress_blocks(), so NEW_ADDR * is a possible case. */ - if (blkaddr == NEW_ADDR || - __is_valid_data_blkaddr(blkaddr)) { + if (blkaddr == NEW_ADDR) { + reserved++; + continue; + } + if (__is_valid_data_blkaddr(blkaddr)) { compr_blocks++; continue; } } - reserved = cluster_size - compr_blocks; + to_reserved = cluster_size - compr_blocks - reserved; /* for the case all blocks in cluster were reserved */ - if (reserved == 1) + if (to_reserved == 1) { + dn->ofs_in_node += cluster_size; goto next; + } - ret = inc_valid_block_count(sbi, dn->inode, &reserved, false); + ret = inc_valid_block_count(sbi, dn->inode, + &to_reserved, false); if (unlikely(ret)) return ret; @@ -3710,7 +3717,7 @@ static int reserve_compress_blocks(struct dnode_of_data *dn, pgoff_t count, f2fs_i_compr_blocks_update(dn->inode, compr_blocks, true); - *reserved_blocks += reserved; + *reserved_blocks += to_reserved; next: count -= cluster_size; } -- cgit v1.2.3-59-g8ed1b From a3a0bc6c223908a2a06736bcd43db962e0657c67 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 6 May 2024 18:41:38 +0800 Subject: f2fs: compress: fix typo in f2fs_reserve_compress_blocks() s/released/reserved. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 13b430b54299..f03ae3d952cc 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -3810,7 +3810,7 @@ unlock_inode: } else if (reserved_blocks && atomic_read(&F2FS_I(inode)->i_compr_blocks)) { set_sbi_flag(sbi, SBI_NEED_FSCK); - f2fs_warn(sbi, "%s: partial blocks were released i_ino=%lx " + f2fs_warn(sbi, "%s: partial blocks were reserved i_ino=%lx " "iblocks=%llu, reserved=%u, compr_blocks=%u, " "run fsck to fix.", __func__, inode->i_ino, inode->i_blocks, -- cgit v1.2.3-59-g8ed1b From 043c832371cd9023fbd725138ddc6c7f288dc469 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 6 May 2024 18:41:37 +0800 Subject: f2fs: compress: fix error path of inc_valid_block_count() If inc_valid_block_count() can not allocate all requested blocks, it needs to release block count in .total_valid_block_count and resevation blocks in inode. Fixes: 54607494875e ("f2fs: compress: fix to avoid inconsistence bewteen i_blocks and dnode") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 30058e16a5d0..ef7de97be647 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -2306,7 +2306,7 @@ static inline void f2fs_i_blocks_write(struct inode *, block_t, bool, bool); static inline int inc_valid_block_count(struct f2fs_sb_info *sbi, struct inode *inode, blkcnt_t *count, bool partial) { - blkcnt_t diff = 0, release = 0; + long long diff = 0, release = 0; block_t avail_user_block_count; int ret; @@ -2326,26 +2326,27 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi, percpu_counter_add(&sbi->alloc_valid_block_count, (*count)); spin_lock(&sbi->stat_lock); - sbi->total_valid_block_count += (block_t)(*count); - avail_user_block_count = get_available_block_count(sbi, inode, true); - if (unlikely(sbi->total_valid_block_count > avail_user_block_count)) { + avail_user_block_count = get_available_block_count(sbi, inode, true); + diff = (long long)sbi->total_valid_block_count + *count - + avail_user_block_count; + if (unlikely(diff > 0)) { if (!partial) { spin_unlock(&sbi->stat_lock); + release = *count; goto enospc; } - - diff = sbi->total_valid_block_count - avail_user_block_count; if (diff > *count) diff = *count; *count -= diff; release = diff; - sbi->total_valid_block_count -= diff; if (!*count) { spin_unlock(&sbi->stat_lock); goto enospc; } } + sbi->total_valid_block_count += (block_t)(*count); + spin_unlock(&sbi->stat_lock); if (unlikely(release)) { -- cgit v1.2.3-59-g8ed1b From 0a4ed2d97cb6d044196cc3e726b6699222b41019 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 6 May 2024 18:41:39 +0800 Subject: f2fs: compress: fix to cover {reserve,release}_compress_blocks() w/ cp_rwsem lock It needs to cover {reserve,release}_compress_blocks() w/ cp_rwsem lock to avoid racing with checkpoint, otherwise, filesystem metadata including blkaddr in dnode, inode fields and .total_valid_block_count may be corrupted after SPO case. Fixes: ef8d563f184e ("f2fs: introduce F2FS_IOC_RELEASE_COMPRESS_BLOCKS") Fixes: c75488fb4d82 ("f2fs: introduce F2FS_IOC_RESERVE_COMPRESS_BLOCKS") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index f03ae3d952cc..2a9de84d2a71 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -3593,9 +3593,12 @@ static int f2fs_release_compress_blocks(struct file *filp, unsigned long arg) struct dnode_of_data dn; pgoff_t end_offset, count; + f2fs_lock_op(sbi); + set_new_dnode(&dn, inode, NULL, NULL, 0); ret = f2fs_get_dnode_of_data(&dn, page_idx, LOOKUP_NODE); if (ret) { + f2fs_unlock_op(sbi); if (ret == -ENOENT) { page_idx = f2fs_get_next_page_offset(&dn, page_idx); @@ -3613,6 +3616,8 @@ static int f2fs_release_compress_blocks(struct file *filp, unsigned long arg) f2fs_put_dnode(&dn); + f2fs_unlock_op(sbi); + if (ret < 0) break; @@ -3765,9 +3770,12 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg) struct dnode_of_data dn; pgoff_t end_offset, count; + f2fs_lock_op(sbi); + set_new_dnode(&dn, inode, NULL, NULL, 0); ret = f2fs_get_dnode_of_data(&dn, page_idx, LOOKUP_NODE); if (ret) { + f2fs_unlock_op(sbi); if (ret == -ENOENT) { page_idx = f2fs_get_next_page_offset(&dn, page_idx); @@ -3785,6 +3793,8 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg) f2fs_put_dnode(&dn); + f2fs_unlock_op(sbi); + if (ret < 0) break; -- cgit v1.2.3-59-g8ed1b From 0fa4e57c1db263effd72d2149d4e21da0055c316 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 7 May 2024 11:31:00 +0800 Subject: f2fs: fix to release node block count in error path of f2fs_new_node_page() It missed to call dec_valid_node_count() to release node block count in error path, fix it. Fixes: 141170b759e0 ("f2fs: fix to avoid use f2fs_bug_on() in f2fs_new_node_page()") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/node.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 95cecf08cb37..b72ef96f7e33 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -1329,6 +1329,7 @@ struct page *f2fs_new_node_page(struct dnode_of_data *dn, unsigned int ofs) } if (unlikely(new_ni.blk_addr != NULL_ADDR)) { err = -EFSCORRUPTED; + dec_valid_node_count(sbi, dn->inode, !ofs); set_sbi_flag(sbi, SBI_NEED_FSCK); f2fs_handle_error(sbi, ERROR_INVALID_BLKADDR); goto fail; @@ -1355,7 +1356,6 @@ struct page *f2fs_new_node_page(struct dnode_of_data *dn, unsigned int ofs) if (ofs == 0) inc_valid_inode_count(sbi); return page; - fail: clear_node_page_dirty(page); f2fs_put_page(page, 1); -- cgit v1.2.3-59-g8ed1b From 29ed2b5dd521ce7c5d8466cd70bf0cc9d07afeee Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 10 May 2024 11:33:39 +0800 Subject: f2fs: compress: don't allow unaligned truncation on released compress inode f2fs image may be corrupted after below testcase: - mkfs.f2fs -O extra_attr,compression -f /dev/vdb - mount /dev/vdb /mnt/f2fs - touch /mnt/f2fs/file - f2fs_io setflags compression /mnt/f2fs/file - dd if=/dev/zero of=/mnt/f2fs/file bs=4k count=4 - f2fs_io release_cblocks /mnt/f2fs/file - truncate -s 8192 /mnt/f2fs/file - umount /mnt/f2fs - fsck.f2fs /dev/vdb [ASSERT] (fsck_chk_inode_blk:1256) --> ino: 0x5 has i_blocks: 0x00000002, but has 0x3 blocks [FSCK] valid_block_count matching with CP [Fail] [0x4, 0x5] [FSCK] other corrupted bugs [Fail] The reason is: partial truncation assume compressed inode has reserved blocks, after partial truncation, valid block count may change w/o .i_blocks and .total_valid_block_count update, result in corruption. This patch only allow cluster size aligned truncation on released compress inode for fixing. Fixes: c61404153eb6 ("f2fs: introduce FI_COMPRESS_RELEASED instead of using IMMUTABLE bit") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 2a9de84d2a71..ef4cfb4436ef 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -952,9 +952,14 @@ int f2fs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, ATTR_GID | ATTR_TIMES_SET)))) return -EPERM; - if ((attr->ia_valid & ATTR_SIZE) && - !f2fs_is_compress_backend_ready(inode)) - return -EOPNOTSUPP; + if ((attr->ia_valid & ATTR_SIZE)) { + if (!f2fs_is_compress_backend_ready(inode)) + return -EOPNOTSUPP; + if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED) && + !IS_ALIGNED(attr->ia_size, + F2FS_BLK_TO_BYTES(F2FS_I(inode)->i_cluster_size))) + return -EINVAL; + } err = setattr_prepare(idmap, dentry, attr); if (err) -- cgit v1.2.3-59-g8ed1b From c62b758bae6af16fee94f556091fa74883a96b1e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 9 May 2024 13:04:24 +0200 Subject: fcntl: add F_DUPFD_QUERY fcntl() Often userspace needs to know whether two file descriptors refer to the same struct file. For example, systemd uses this to filter out duplicate file descriptors in it's file descriptor store (cf. [1]) and vulkan uses it to compare dma-buf fds (cf. [2]). The only api we provided for this was kcmp() but that's not generally available or might be disallowed because it is way more powerful (allows ordering of file pointers, operates on non-current task) etc. So give userspace a simple way of comparing two file descriptors for sameness adding a new fcntl() F_DUDFD_QUERY. Link: https://github.com/systemd/systemd/blob/a4f0e0da3573a10bc5404142be8799418760b1d1/src/basic/fd-util.c#L517 [1] Link: https://gitlab.freedesktop.org/wlroots/wlroots/-/blob/master/render/vulkan/texture.c#L490 [2] Signed-off-by: Linus Torvalds [brauner: commit message] Signed-off-by: Christian Brauner --- fs/fcntl.c | 20 ++++++++++++++++++++ include/uapi/linux/fcntl.h | 14 ++++++++------ 2 files changed, 28 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/fcntl.c b/fs/fcntl.c index 54cc85d3338e..300e5d9ad913 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -327,6 +327,22 @@ static long fcntl_set_rw_hint(struct file *file, unsigned int cmd, return 0; } +/* Is the file descriptor a dup of the file? */ +static long f_dupfd_query(int fd, struct file *filp) +{ + CLASS(fd_raw, f)(fd); + + /* + * We can do the 'fdput()' immediately, as the only thing that + * matters is the pointer value which isn't changed by the fdput. + * + * Technically we didn't need a ref at all, and 'fdget()' was + * overkill, but given our lockless file pointer lookup, the + * alternatives are complicated. + */ + return f.file == filp; +} + static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, struct file *filp) { @@ -342,6 +358,9 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, case F_DUPFD_CLOEXEC: err = f_dupfd(argi, filp, O_CLOEXEC); break; + case F_DUPFD_QUERY: + err = f_dupfd_query(argi, filp); + break; case F_GETFD: err = get_close_on_exec(fd) ? FD_CLOEXEC : 0; break; @@ -446,6 +465,7 @@ static int check_fcntl_cmd(unsigned cmd) switch (cmd) { case F_DUPFD: case F_DUPFD_CLOEXEC: + case F_DUPFD_QUERY: case F_GETFD: case F_SETFD: case F_GETFL: diff --git a/include/uapi/linux/fcntl.h b/include/uapi/linux/fcntl.h index 282e90aeb163..c0bcc185fa48 100644 --- a/include/uapi/linux/fcntl.h +++ b/include/uapi/linux/fcntl.h @@ -8,6 +8,14 @@ #define F_SETLEASE (F_LINUX_SPECIFIC_BASE + 0) #define F_GETLEASE (F_LINUX_SPECIFIC_BASE + 1) +/* + * Request nofications on a directory. + * See below for events that may be notified. + */ +#define F_NOTIFY (F_LINUX_SPECIFIC_BASE + 2) + +#define F_DUPFD_QUERY (F_LINUX_SPECIFIC_BASE + 3) + /* * Cancel a blocking posix lock; internal use only until we expose an * asynchronous lock api to userspace: @@ -17,12 +25,6 @@ /* Create a file descriptor with FD_CLOEXEC set. */ #define F_DUPFD_CLOEXEC (F_LINUX_SPECIFIC_BASE + 6) -/* - * Request nofications on a directory. - * See below for events that may be notified. - */ -#define F_NOTIFY (F_LINUX_SPECIFIC_BASE+2) - /* * Set and get of pipe page size array */ -- cgit v1.2.3-59-g8ed1b From da0e01cc7079124cb1e86a2c35dd90ba12897e1a Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 3 May 2024 15:22:32 +0100 Subject: afs: Fix fileserver rotation getting stuck Fix the fileserver rotation code in a couple of ways: (1) op->server_states is an array, not a pointer to a single record, so fix the places that access it to index it. (2) In the places that go through an address list to work out which one has the best priority, fix the loops to skip known failed addresses. Without this, the rotation algorithm may get stuck on addresses that are inaccessible or don't respond. This can be triggered manually by finding a server that advertises a non-routable address and giving it a higher priority, eg.: echo "add udp 192.168.0.0/16 3000" >/proc/fs/afs/addr_prefs if the server, say, includes the address 192.168.7.7 in its address list, and then attempting to access a volume on that server. Fixes: 495f2ae9e355 ("afs: Fix fileserver rotation") Signed-off-by: David Howells cc: Marc Dionne cc: linux-afs@lists.infradead.org Link: https://lore.kernel.org/r/4005300.1712309731@warthog.procyon.org.uk/ # v1 Link: https://lore.kernel.org/r/998836.1714746152@warthog.procyon.org.uk Signed-off-by: Christian Brauner --- fs/afs/rotate.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/afs/rotate.c b/fs/afs/rotate.c index ed04bd1eeae8..ed09d4d4c211 100644 --- a/fs/afs/rotate.c +++ b/fs/afs/rotate.c @@ -541,11 +541,13 @@ pick_server: test_bit(AFS_SE_EXCLUDED, &se->flags) || !test_bit(AFS_SERVER_FL_RESPONDING, &s->flags)) continue; - es = op->server_states->endpoint_state; + es = op->server_states[i].endpoint_state; sal = es->addresses; afs_get_address_preferences_rcu(op->net, sal); for (j = 0; j < sal->nr_addrs; j++) { + if (es->failed_set & (1 << j)) + continue; if (!sal->addrs[j].peer) continue; if (sal->addrs[j].prio > best_prio) { @@ -605,6 +607,8 @@ iterate_address: best_prio = -1; addr_index = 0; for (i = 0; i < alist->nr_addrs; i++) { + if (!(set & (1 << i))) + continue; if (alist->addrs[i].prio > best_prio) { addr_index = i; best_prio = alist->addrs[i].prio; @@ -674,7 +678,7 @@ no_more_servers: for (i = 0; i < op->server_list->nr_servers; i++) { struct afs_endpoint_state *estate; - estate = op->server_states->endpoint_state; + estate = op->server_states[i].endpoint_state; error = READ_ONCE(estate->error); if (error < 0) afs_op_accumulate_error(op, error, estate->abort_code); -- cgit v1.2.3-59-g8ed1b From 42815f8ac54c5113bf450ec4b7ccc5b62af0f6a7 Mon Sep 17 00:00:00 2001 From: Hou Tao Date: Thu, 9 May 2024 20:21:53 +0800 Subject: fuse: set FR_PENDING atomically in fuse_resend() When fuse_resend() moves the requests from processing lists to pending list, it uses __set_bit() to set FR_PENDING bit in req->flags. Using __set_bit() is not safe, because other functions may update req->flags concurrently (e.g., request_wait_answer() may call set_bit(FR_INTERRUPTED, &flags)). Fix it by using set_bit() instead. Fixes: 760eac73f9f6 ("fuse: Introduce a new notification type for resend pending requests") Signed-off-by: Hou Tao Signed-off-by: Miklos Szeredi --- fs/fuse/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 3ec8bb5e68ff..8eb2ce7c0b01 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1813,7 +1813,7 @@ static void fuse_resend(struct fuse_conn *fc) spin_unlock(&fc->lock); list_for_each_entry_safe(req, next, &to_queue, list) { - __set_bit(FR_PENDING, &req->flags); + set_bit(FR_PENDING, &req->flags); /* mark the request as resend request */ req->in.h.unique |= FUSE_UNIQUE_RESEND; } -- cgit v1.2.3-59-g8ed1b From 246014876d782bbf2e652267482cd2e799fb5fcd Mon Sep 17 00:00:00 2001 From: Hou Tao Date: Thu, 9 May 2024 20:21:54 +0800 Subject: fuse: clear FR_SENT when re-adding requests into pending list The following warning was reported by lee bruce: ------------[ cut here ]------------ WARNING: CPU: 0 PID: 8264 at fs/fuse/dev.c:300 fuse_request_end+0x685/0x7e0 fs/fuse/dev.c:300 Modules linked in: CPU: 0 PID: 8264 Comm: ab2 Not tainted 6.9.0-rc7 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) RIP: 0010:fuse_request_end+0x685/0x7e0 fs/fuse/dev.c:300 ...... Call Trace: fuse_dev_do_read.constprop.0+0xd36/0x1dd0 fs/fuse/dev.c:1334 fuse_dev_read+0x166/0x200 fs/fuse/dev.c:1367 call_read_iter include/linux/fs.h:2104 [inline] new_sync_read fs/read_write.c:395 [inline] vfs_read+0x85b/0xba0 fs/read_write.c:476 ksys_read+0x12f/0x260 fs/read_write.c:619 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xce/0x260 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f ...... The warning is due to the FUSE_NOTIFY_RESEND notify sent by the write() syscall in the reproducer program and it happens as follows: (1) calls fuse_dev_read() to read the INIT request The read succeeds. During the read, bit FR_SENT will be set on the request. (2) calls fuse_dev_write() to send an USE_NOTIFY_RESEND notify The resend notify will resend all processing requests, so the INIT request is moved from processing list to pending list again. (3) calls fuse_dev_read() with an invalid output address fuse_dev_read() will try to copy the same INIT request to the output address, but it will fail due to the invalid address, so the INIT request is ended and triggers the warning in fuse_request_end(). Fix it by clearing FR_SENT when re-adding requests into pending list. Acked-by: Miklos Szeredi Reported-by: xingwei lee Reported-by: yue sun Closes: https://lore.kernel.org/linux-fsdevel/58f13e47-4765-fce4-daf4-dffcc5ae2330@huaweicloud.com/T/#m091614e5ea2af403b259e7cea6a49e51b9ee07a7 Fixes: 760eac73f9f6 ("fuse: Introduce a new notification type for resend pending requests") Signed-off-by: Hou Tao Signed-off-by: Miklos Szeredi --- fs/fuse/dev.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 8eb2ce7c0b01..9eb191b5c4de 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1814,6 +1814,7 @@ static void fuse_resend(struct fuse_conn *fc) list_for_each_entry_safe(req, next, &to_queue, list) { set_bit(FR_PENDING, &req->flags); + clear_bit(FR_SENT, &req->flags); /* mark the request as resend request */ req->in.h.unique |= FUSE_UNIQUE_RESEND; } -- cgit v1.2.3-59-g8ed1b From e9229c18dae3b3c2556cea8413edd1f76c78d767 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Thu, 2 May 2024 23:26:30 +0200 Subject: ovl: remove duplicate included header Remove duplicate included header file linux/posix_acl.h Signed-off-by: Thorsten Blum Signed-off-by: Miklos Szeredi --- fs/overlayfs/inode.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c index c63b31a460be..35fd3e3e1778 100644 --- a/fs/overlayfs/inode.c +++ b/fs/overlayfs/inode.c @@ -8,7 +8,6 @@ #include #include #include -#include #include #include #include -- cgit v1.2.3-59-g8ed1b From 103c2de111bf32f7c36a0ce8f638b114a37e0b76 Mon Sep 17 00:00:00 2001 From: Peter-Jan Gootzen Date: Wed, 1 May 2024 17:38:16 +0200 Subject: virtio-fs: limit number of request queues Virtio-fs devices might allocate significant resources to virtio queues such as CPU cores that busy poll on the queue. The device indicates how many request queues it can support and the driver should initialize the number of queues that they want to utilize. In this patch we limit the number of initialized request queues to the number of CPUs, to limit the resource consumption on the device-side and to prepare for the upcoming multi-queue patch. Signed-off-by: Peter-Jan Gootzen Signed-off-by: Yoray Zack Suggested-by: Max Gurtovoy Reviewed-by: Max Gurtovoy Reviewed-by: Stefan Hajnoczi Acked-by: Michael S. Tsirkin Signed-off-by: Miklos Szeredi --- fs/fuse/virtio_fs.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index bb3e941b9503..fa4c16abfe10 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -751,6 +751,9 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev, if (fs->num_request_queues == 0) return -EINVAL; + /* Truncate nr of request queues to nr_cpu_id */ + fs->num_request_queues = min_t(unsigned int, fs->num_request_queues, + nr_cpu_ids); fs->nvqs = VQ_REQUEST + fs->num_request_queues; fs->vqs = kcalloc(fs->nvqs, sizeof(fs->vqs[VQ_HIPRIO]), GFP_KERNEL); if (!fs->vqs) -- cgit v1.2.3-59-g8ed1b From 529395d2ae6456c556405016ea0c43081fe607f3 Mon Sep 17 00:00:00 2001 From: Peter-Jan Gootzen Date: Wed, 1 May 2024 17:38:17 +0200 Subject: virtio-fs: add multi-queue support This commit creates a multi-queue mapping at device bring-up. The driver first attempts to use the existing MSI-X interrupt affinities (previously disabled), and if not present, will distribute the request queues evenly over the CPUs. If the latter fails as well, all CPUs are mapped to request queue zero. When a request is handed from FUSE to the virtio-fs device driver, the driver will use the current CPU to index into the multi-queue mapping and determine the optimal request queue to use. We measured the performance of this patch with the fio benchmarking tool, increasing the number of queues results in a significant speedup for both read and write operations, demonstrating the effectiveness of multi-queue support. Host: - Dell PowerEdge R760 - CPU: Intel(R) Xeon(R) Gold 6438M, 128 cores - VM: KVM with 32 cores Virtio-fs device: - BlueField-3 DPU - CPU: ARM Cortex-A78AE, 16 cores - One thread per queue, each busy polling on one request queue - Each queue is 1024 descriptors deep Workload: - fio, sequential read or write, ioengine=libaio, numjobs=32, 4GiB file per job, iodepth=8, bs=256KiB, runtime=30s Performance Results: +===========================+==========+===========+ | Number of queues | Fio read | Fio write | +===========================+==========+===========+ | 1 request queue (GiB/s) | 6.1 | 4.6 | +---------------------------+----------+-----------+ | 8 request queues (GiB/s) | 25.8 | 10.3 | +---------------------------+----------+-----------+ | 16 request queues (GiB/s) | 30.9 | 19.5 | +---------------------------+----------+-----------+ | 32 request queue (GiB/s) | 33.2 | 22.6 | +---------------------------+----------+-----------+ | Speedup | 5.5x | 5x | +---------------=-----------+----------+-----------+ Signed-off-by: Peter-Jan Gootzen Signed-off-by: Yoray Zack Signed-off-by: Max Gurtovoy Reviewed-by: Stefan Hajnoczi Acked-by: Michael S. Tsirkin Signed-off-by: Miklos Szeredi --- fs/fuse/virtio_fs.c | 70 +++++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 62 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index fa4c16abfe10..8ffa4f063a37 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -7,6 +7,8 @@ #include #include #include +#include +#include #include #include #include @@ -67,6 +69,8 @@ struct virtio_fs { unsigned int num_request_queues; /* number of request queues */ struct dax_device *dax_dev; + unsigned int *mq_map; /* index = cpu id, value = request vq id */ + /* DAX memory window where file contents are mapped */ void *window_kaddr; phys_addr_t window_phys_addr; @@ -185,6 +189,7 @@ static void virtio_fs_ktype_release(struct kobject *kobj) { struct virtio_fs *vfs = container_of(kobj, struct virtio_fs, kobj); + kfree(vfs->mq_map); kfree(vfs->vqs); kfree(vfs); } @@ -706,6 +711,44 @@ static void virtio_fs_requests_done_work(struct work_struct *work) } } +static void virtio_fs_map_queues(struct virtio_device *vdev, struct virtio_fs *fs) +{ + const struct cpumask *mask, *masks; + unsigned int q, cpu; + + /* First attempt to map using existing transport layer affinities + * e.g. PCIe MSI-X + */ + if (!vdev->config->get_vq_affinity) + goto fallback; + + for (q = 0; q < fs->num_request_queues; q++) { + mask = vdev->config->get_vq_affinity(vdev, VQ_REQUEST + q); + if (!mask) + goto fallback; + + for_each_cpu(cpu, mask) + fs->mq_map[cpu] = q; + } + + return; +fallback: + /* Attempt to map evenly in groups over the CPUs */ + masks = group_cpus_evenly(fs->num_request_queues); + /* If even this fails we default to all CPUs use queue zero */ + if (!masks) { + for_each_possible_cpu(cpu) + fs->mq_map[cpu] = 0; + return; + } + + for (q = 0; q < fs->num_request_queues; q++) { + for_each_cpu(cpu, &masks[q]) + fs->mq_map[cpu] = q; + } + kfree(masks); +} + /* Virtqueue interrupt handler */ static void virtio_fs_vq_done(struct virtqueue *vq) { @@ -742,6 +785,11 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev, { struct virtqueue **vqs; vq_callback_t **callbacks; + /* Specify pre_vectors to ensure that the queues before the + * request queues (e.g. hiprio) don't claim any of the CPUs in + * the multi-queue mapping and interrupt affinities + */ + struct irq_affinity desc = { .pre_vectors = VQ_REQUEST }; const char **names; unsigned int i; int ret = 0; @@ -763,7 +811,9 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev, callbacks = kmalloc_array(fs->nvqs, sizeof(callbacks[VQ_HIPRIO]), GFP_KERNEL); names = kmalloc_array(fs->nvqs, sizeof(names[VQ_HIPRIO]), GFP_KERNEL); - if (!vqs || !callbacks || !names) { + fs->mq_map = kcalloc_node(nr_cpu_ids, sizeof(*fs->mq_map), GFP_KERNEL, + dev_to_node(&vdev->dev)); + if (!vqs || !callbacks || !names || !fs->mq_map) { ret = -ENOMEM; goto out; } @@ -783,7 +833,7 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev, names[i] = fs->vqs[i].name; } - ret = virtio_find_vqs(vdev, fs->nvqs, vqs, callbacks, names, NULL); + ret = virtio_find_vqs(vdev, fs->nvqs, vqs, callbacks, names, &desc); if (ret < 0) goto out; @@ -795,8 +845,10 @@ out: kfree(names); kfree(callbacks); kfree(vqs); - if (ret) + if (ret) { kfree(fs->vqs); + kfree(fs->mq_map); + } return ret; } @@ -942,7 +994,7 @@ static int virtio_fs_probe(struct virtio_device *vdev) if (ret < 0) goto out; - /* TODO vq affinity */ + virtio_fs_map_queues(vdev, fs); ret = virtio_fs_setup_dax(vdev, fs); if (ret < 0) @@ -1291,7 +1343,7 @@ out: static void virtio_fs_wake_pending_and_unlock(struct fuse_iqueue *fiq) __releases(fiq->lock) { - unsigned int queue_id = VQ_REQUEST; /* TODO multiqueue */ + unsigned int queue_id; struct virtio_fs *fs; struct fuse_req *req; struct virtio_fs_vq *fsvq; @@ -1305,11 +1357,13 @@ __releases(fiq->lock) spin_unlock(&fiq->lock); fs = fiq->priv; + queue_id = VQ_REQUEST + fs->mq_map[raw_smp_processor_id()]; - pr_debug("%s: opcode %u unique %#llx nodeid %#llx in.len %u out.len %u\n", - __func__, req->in.h.opcode, req->in.h.unique, + pr_debug("%s: opcode %u unique %#llx nodeid %#llx in.len %u out.len %u queue_id %u\n", + __func__, req->in.h.opcode, req->in.h.unique, req->in.h.nodeid, req->in.h.len, - fuse_len_args(req->args->out_numargs, req->args->out_args)); + fuse_len_args(req->args->out_numargs, req->args->out_args), + queue_id); fsvq = &fs->vqs[queue_id]; ret = virtio_fs_enqueue_req(fsvq, req, false); -- cgit v1.2.3-59-g8ed1b From f2526c5cf1d94359467d9472387363d57c6b3e6d Mon Sep 17 00:00:00 2001 From: Daeho Jeong Date: Fri, 10 May 2024 06:49:08 -0700 Subject: f2fs: allow dirty sections with zero valid block for checkpoint disabled Following the semantic for dirty segments in checkpoint disabled mode, apply the same rule to dirty sections. Signed-off-by: Daeho Jeong Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index ab3e9a6e9c96..a0ce3d080f80 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -771,8 +771,10 @@ static void __locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno, block_t valid_blocks = get_valid_blocks(sbi, segno, true); - f2fs_bug_on(sbi, unlikely(!valid_blocks || - valid_blocks == CAP_BLKS_PER_SEC(sbi))); + f2fs_bug_on(sbi, + (!is_sbi_flag_set(sbi, SBI_CP_DISABLED) && + !valid_blocks) || + valid_blocks == CAP_BLKS_PER_SEC(sbi)); if (!IS_CURSEC(sbi, secno)) set_bit(secno, dirty_i->dirty_secmap); -- cgit v1.2.3-59-g8ed1b From a798ff17cd2dabe47d5d4ed3d509631793c36e19 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 10 May 2024 11:43:33 +0800 Subject: f2fs: fix to add missing iput() in gc_data_segment() During gc_data_segment(), if inode state is abnormal, it missed to call iput(), fix it. Fixes: b73e52824c89 ("f2fs: reposition unlock_new_inode to prevent accessing invalid inode") Fixes: 9056d6489f5a ("f2fs: fix to do sanity check on inode type during garbage collection") Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/gc.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index ac4cbbe50c2f..6066c6eecf41 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -1554,10 +1554,15 @@ next_step: int err; inode = f2fs_iget(sb, dni.ino); - if (IS_ERR(inode) || is_bad_inode(inode) || - special_file(inode->i_mode)) + if (IS_ERR(inode)) continue; + if (is_bad_inode(inode) || + special_file(inode->i_mode)) { + iput(inode); + continue; + } + err = f2fs_gc_pinned_control(inode, gc_type, segno); if (err == -EAGAIN) { iput(inode); -- cgit v1.2.3-59-g8ed1b From a7ac59f4f23660473e6350306f9b88f24fcc38f1 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 30 Apr 2024 14:09:01 +0900 Subject: nilfs2: remove calls to folio_set_error() and folio_clear_error() Nobody checks this flag on nilfs2 folios, stop setting and clearing it. That lets us simplify nilfs_end_folio_io() slightly. Link: https://lkml.kernel.org/r/20240420025029.2166544-17-willy@infradead.org Link: https://lkml.kernel.org/r/20240430050901.3239-1-konishi.ryusuke@gmail.com Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Ryusuke Konishi Cc: kernel test robot Cc: Peter Zijlstra Cc: Song Liu Signed-off-by: Andrew Morton --- fs/nilfs2/dir.c | 1 - fs/nilfs2/segment.c | 8 +------- 2 files changed, 1 insertion(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c index bc846b904b68..0900fcad2d0c 100644 --- a/fs/nilfs2/dir.c +++ b/fs/nilfs2/dir.c @@ -174,7 +174,6 @@ Eend: dir->i_ino, (folio->index << PAGE_SHIFT) + offs, (unsigned long)le64_to_cpu(p->inode)); fail: - folio_set_error(folio); return false; } diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index aa5290cb7467..8654ab8ad534 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -1725,14 +1725,8 @@ static void nilfs_end_folio_io(struct folio *folio, int err) return; } - if (!err) { - if (!nilfs_folio_buffers_clean(folio)) - filemap_dirty_folio(folio->mapping, folio); - folio_clear_error(folio); - } else { + if (err || !nilfs_folio_buffers_clean(folio)) filemap_dirty_folio(folio->mapping, folio); - folio_set_error(folio); - } folio_end_writeback(folio); } -- cgit v1.2.3-59-g8ed1b From 0a73eac1ed10097d1799c10dff2172605fd40c75 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Thu, 9 May 2024 07:14:29 +0900 Subject: nilfs2: convert BUG_ON() in nilfs_finish_roll_forward() to WARN_ON() The BUG_ON check performed on the return value of __getblk() in nilfs_finish_roll_forward() assumes that a buffer that has been successfully read once is retrieved with the same parameters and does not fail (__getblk() does not return an error due to memory allocation failure). Also, nilfs_finish_roll_forward() is called at most once during mount. Taking these into consideration, rewrite the check to use WARN_ON() to avoid using BUG_ON(). Link: https://lkml.kernel.org/r/20240508221429.7559-1-konishi.ryusuke@gmail.com Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton --- fs/nilfs2/recovery.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c index e48372618ac4..020f304c600e 100644 --- a/fs/nilfs2/recovery.c +++ b/fs/nilfs2/recovery.c @@ -699,7 +699,9 @@ static void nilfs_finish_roll_forward(struct the_nilfs *nilfs, return; bh = __getblk(nilfs->ns_bdev, ri->ri_lsegs_start, nilfs->ns_blocksize); - BUG_ON(!bh); + if (WARN_ON(!bh)) + return; /* should never happen */ + memset(bh->b_data, 0, bh->b_size); set_buffer_dirty(bh); err = sync_dirty_buffer(bh); -- cgit v1.2.3-59-g8ed1b From 5cbcb62dddf5346077feb82b7b0c9254222d3445 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Tue, 7 May 2024 09:18:58 -0400 Subject: fs/proc: fix softlockup in __read_vmcore While taking a kernel core dump with makedumpfile on a larger system, softlockup messages often appear. While softlockup warnings can be harmless, they can also interfere with things like RCU freeing memory, which can be problematic when the kdump kexec image is configured with as little memory as possible. Avoid the softlockup, and give things like work items and RCU a chance to do their thing during __read_vmcore by adding a cond_resched. Link: https://lkml.kernel.org/r/20240507091858.36ff767f@imladris.surriel.com Signed-off-by: Rik van Riel Acked-by: Baoquan He Cc: Dave Young Cc: Vivek Goyal Cc: Signed-off-by: Andrew Morton --- fs/proc/vmcore.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 5d08d4d159d3..b52d85f8ad59 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -383,6 +383,8 @@ static ssize_t __read_vmcore(struct iov_iter *iter, loff_t *fpos) /* leave now if filled buffer already */ if (!iov_iter_count(iter)) return acc; + + cond_resched(); } list_for_each_entry(m, &vmcore_list, list) { -- cgit v1.2.3-59-g8ed1b From 0162a70d8e25ed06db99c7abb8630f9b71aaba98 Mon Sep 17 00:00:00 2001 From: Christian Heusel Date: Sun, 11 Feb 2024 01:39:04 +0100 Subject: jffs2: print symbolic error name instead of error code Utilize the %pe print specifier to get the symbolic error name as a string (i.e "-ENOMEM") in the log message instead of the error code to increase its readablility. This change was suggested in https://lore.kernel.org/all/92972476-0b1f-4d0a-9951-af3fc8bc6e65@suswa.mountain/ Signed-off-by: Christian Heusel Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/jffs2/background.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/background.c b/fs/jffs2/background.c index 6da92ecaf66d..bb0ee1a59e71 100644 --- a/fs/jffs2/background.c +++ b/fs/jffs2/background.c @@ -44,8 +44,8 @@ int jffs2_start_garbage_collect_thread(struct jffs2_sb_info *c) tsk = kthread_run(jffs2_garbage_collect_thread, c, "jffs2_gcd_mtd%d", c->mtd->index); if (IS_ERR(tsk)) { - pr_warn("fork failed for JFFS2 garbage collect thread: %ld\n", - -PTR_ERR(tsk)); + pr_warn("fork failed for JFFS2 garbage collect thread: %pe\n", + tsk); complete(&c->gc_thread_exit); ret = PTR_ERR(tsk); } else { -- cgit v1.2.3-59-g8ed1b From 2e0a808224028457040caa52fe883740013adac8 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 5 Dec 2023 09:32:22 -0800 Subject: jffs2: nodemgmt: fix kernel-doc comments Update the end of one sentence where a comment was truncated. (dwmw2) Fix a bunch of kernel-doc warnings: nodemgmt.c:72: warning: Function parameter or member 'sumsize' not described in 'jffs2_do_reserve_space' nodemgmt.c:72: warning: expecting prototype for jffs2_reserve_space(). Prototype was for jffs2_do_reserve_space() instead nodemgmt.c:76: warning: Function parameter or member 'sumsize' not described in 'jffs2_reserve_space' nodemgmt.c:76: warning: No description found for return value of 'jffs2_reserve_space' nodemgmt.c:503: warning: Function parameter or member 'ofs' not described in 'jffs2_add_physical_node_ref' nodemgmt.c:503: warning: Function parameter or member 'ic' not described in 'jffs2_add_physical_node_ref' nodemgmt.c:503: warning: Excess function parameter 'new' description in 'jffs2_add_physical_node_ref' nodemgmt.c:503: warning: No description found for return value of 'jffs2_add_physical_node_ref' Signed-off-by: Randy Dunlap Cc: David Woodhouse Cc: Richard Weinberger Cc: linux-mtd@lists.infradead.org Reviewed-by: Zhihao Cheng Reviewed-by: Jeff Johnson Signed-off-by: Richard Weinberger --- fs/jffs2/nodemgmt.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/nodemgmt.c b/fs/jffs2/nodemgmt.c index a7bbe879cfc3..bbab2bdc71b6 100644 --- a/fs/jffs2/nodemgmt.c +++ b/fs/jffs2/nodemgmt.c @@ -49,28 +49,31 @@ static int jffs2_rp_can_write(struct jffs2_sb_info *c) return 0; } +static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, + uint32_t *len, uint32_t sumsize); + /** * jffs2_reserve_space - request physical space to write nodes to flash * @c: superblock info * @minsize: Minimum acceptable size of allocation * @len: Returned value of allocation length * @prio: Allocation type - ALLOC_{NORMAL,DELETION} + * @sumsize: summary size requested or JFFS2_SUMMARY_NOSUM_SIZE for no summary + * + * Requests a block of physical space on the flash. * - * Requests a block of physical space on the flash. Returns zero for success - * and puts 'len' into the appropriate place, or returns -ENOSPC or other - * error if appropriate. Doesn't return len since that's + * Returns: %0 for success and puts 'len' into the appropriate place, + * or returns -ENOSPC or other error if appropriate. + * Doesn't return len since that's already returned in @len. * - * If it returns zero, jffs2_reserve_space() also downs the per-filesystem + * If it returns %0, jffs2_reserve_space() also downs the per-filesystem * allocation semaphore, to prevent more than one allocation from being - * active at any time. The semaphore is later released by jffs2_commit_allocation() + * active at any time. The semaphore is later released by jffs2_commit_allocation(). * * jffs2_reserve_space() may trigger garbage collection in order to make room * for the requested allocation. */ -static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, - uint32_t *len, uint32_t sumsize); - int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *len, int prio, uint32_t sumsize) { @@ -488,13 +491,16 @@ static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, /** * jffs2_add_physical_node_ref - add a physical node reference to the list * @c: superblock info - * @new: new node reference to add + * @ofs: offset in the block * @len: length of this physical node + * @ic: inode cache pointer * * Should only be used to report nodes for which space has been allocated * by jffs2_reserve_space. * * Must be called with the alloc_sem held. + * + * Returns: pointer to new node on success or -errno code on error */ struct jffs2_raw_node_ref *jffs2_add_physical_node_ref(struct jffs2_sb_info *c, -- cgit v1.2.3-59-g8ed1b From 7096fae56f82d00bf9f1217e6267811cc553350c Mon Sep 17 00:00:00 2001 From: Kunwu Chan Date: Mon, 5 Feb 2024 15:51:44 +0800 Subject: jffs2: Simplify the allocation of slab caches Use the new KMEM_CACHE() macro instead of direct kmem_cache_create to simplify the creation of SLAB caches. And change cache name from 'jffs2_tmp_dnode' to 'jffs2_tmp_dnode_info'. Signed-off-by: Kunwu Chan Reviewed-by: Zhihao Cheng Signed-off-by: Richard Weinberger --- fs/jffs2/malloc.c | 32 ++++++++------------------------ 1 file changed, 8 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/jffs2/malloc.c b/fs/jffs2/malloc.c index ce1189793288..411de8b361b2 100644 --- a/fs/jffs2/malloc.c +++ b/fs/jffs2/malloc.c @@ -33,27 +33,19 @@ static struct kmem_cache *xattr_ref_cache; int __init jffs2_create_slab_caches(void) { - full_dnode_slab = kmem_cache_create("jffs2_full_dnode", - sizeof(struct jffs2_full_dnode), - 0, 0, NULL); + full_dnode_slab = KMEM_CACHE(jffs2_full_dnode, 0); if (!full_dnode_slab) goto err; - raw_dirent_slab = kmem_cache_create("jffs2_raw_dirent", - sizeof(struct jffs2_raw_dirent), - 0, SLAB_HWCACHE_ALIGN, NULL); + raw_dirent_slab = KMEM_CACHE(jffs2_raw_dirent, SLAB_HWCACHE_ALIGN); if (!raw_dirent_slab) goto err; - raw_inode_slab = kmem_cache_create("jffs2_raw_inode", - sizeof(struct jffs2_raw_inode), - 0, SLAB_HWCACHE_ALIGN, NULL); + raw_inode_slab = KMEM_CACHE(jffs2_raw_inode, SLAB_HWCACHE_ALIGN); if (!raw_inode_slab) goto err; - tmp_dnode_info_slab = kmem_cache_create("jffs2_tmp_dnode", - sizeof(struct jffs2_tmp_dnode_info), - 0, 0, NULL); + tmp_dnode_info_slab = KMEM_CACHE(jffs2_tmp_dnode_info, 0); if (!tmp_dnode_info_slab) goto err; @@ -63,28 +55,20 @@ int __init jffs2_create_slab_caches(void) if (!raw_node_ref_slab) goto err; - node_frag_slab = kmem_cache_create("jffs2_node_frag", - sizeof(struct jffs2_node_frag), - 0, 0, NULL); + node_frag_slab = KMEM_CACHE(jffs2_node_frag, 0); if (!node_frag_slab) goto err; - inode_cache_slab = kmem_cache_create("jffs2_inode_cache", - sizeof(struct jffs2_inode_cache), - 0, 0, NULL); + inode_cache_slab = KMEM_CACHE(jffs2_inode_cache, 0); if (!inode_cache_slab) goto err; #ifdef CONFIG_JFFS2_FS_XATTR - xattr_datum_cache = kmem_cache_create("jffs2_xattr_datum", - sizeof(struct jffs2_xattr_datum), - 0, 0, NULL); + xattr_datum_cache = KMEM_CACHE(jffs2_xattr_datum, 0); if (!xattr_datum_cache) goto err; - xattr_ref_cache = kmem_cache_create("jffs2_xattr_ref", - sizeof(struct jffs2_xattr_ref), - 0, 0, NULL); + xattr_ref_cache = KMEM_CACHE(jffs2_xattr_ref, 0); if (!xattr_ref_cache) goto err; #endif -- cgit v1.2.3-59-g8ed1b From af9a8730ddb6a4b2edd779ccc0aceb994d616830 Mon Sep 17 00:00:00 2001 From: Wang Yong Date: Tue, 7 May 2024 15:00:46 +0800 Subject: jffs2: Fix potential illegal address access in jffs2_free_inode During the stress testing of the jffs2 file system,the following abnormal printouts were found: [ 2430.649000] Unable to handle kernel paging request at virtual address 0069696969696948 [ 2430.649622] Mem abort info: [ 2430.649829] ESR = 0x96000004 [ 2430.650115] EC = 0x25: DABT (current EL), IL = 32 bits [ 2430.650564] SET = 0, FnV = 0 [ 2430.650795] EA = 0, S1PTW = 0 [ 2430.651032] FSC = 0x04: level 0 translation fault [ 2430.651446] Data abort info: [ 2430.651683] ISV = 0, ISS = 0x00000004 [ 2430.652001] CM = 0, WnR = 0 [ 2430.652558] [0069696969696948] address between user and kernel address ranges [ 2430.653265] Internal error: Oops: 96000004 [#1] PREEMPT SMP [ 2430.654512] CPU: 2 PID: 20919 Comm: cat Not tainted 5.15.25-g512f31242bf6 #33 [ 2430.655008] Hardware name: linux,dummy-virt (DT) [ 2430.655517] pstate: 20000005 (nzCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 2430.656142] pc : kfree+0x78/0x348 [ 2430.656630] lr : jffs2_free_inode+0x24/0x48 [ 2430.657051] sp : ffff800009eebd10 [ 2430.657355] x29: ffff800009eebd10 x28: 0000000000000001 x27: 0000000000000000 [ 2430.658327] x26: ffff000038f09d80 x25: 0080000000000000 x24: ffff800009d38000 [ 2430.658919] x23: 5a5a5a5a5a5a5a5a x22: ffff000038f09d80 x21: ffff8000084f0d14 [ 2430.659434] x20: ffff0000bf9a6ac0 x19: 0169696969696940 x18: 0000000000000000 [ 2430.659969] x17: ffff8000b6506000 x16: ffff800009eec000 x15: 0000000000004000 [ 2430.660637] x14: 0000000000000000 x13: 00000001000820a1 x12: 00000000000d1b19 [ 2430.661345] x11: 0004000800000000 x10: 0000000000000001 x9 : ffff8000084f0d14 [ 2430.662025] x8 : ffff0000bf9a6b40 x7 : ffff0000bf9a6b48 x6 : 0000000003470302 [ 2430.662695] x5 : ffff00002e41dcc0 x4 : ffff0000bf9aa3b0 x3 : 0000000003470342 [ 2430.663486] x2 : 0000000000000000 x1 : ffff8000084f0d14 x0 : fffffc0000000000 [ 2430.664217] Call trace: [ 2430.664528] kfree+0x78/0x348 [ 2430.664855] jffs2_free_inode+0x24/0x48 [ 2430.665233] i_callback+0x24/0x50 [ 2430.665528] rcu_do_batch+0x1ac/0x448 [ 2430.665892] rcu_core+0x28c/0x3c8 [ 2430.666151] rcu_core_si+0x18/0x28 [ 2430.666473] __do_softirq+0x138/0x3cc [ 2430.666781] irq_exit+0xf0/0x110 [ 2430.667065] handle_domain_irq+0x6c/0x98 [ 2430.667447] gic_handle_irq+0xac/0xe8 [ 2430.667739] call_on_irq_stack+0x28/0x54 The parameter passed to kfree was 5a5a5a5a, which corresponds to the target field of the jffs_inode_info structure. It was found that all variables in the jffs_inode_info structure were 5a5a5a5a, except for the first member sem. It is suspected that these variables are not initialized because they were set to 5a5a5a5a during memory testing, which is meant to detect uninitialized memory.The sem variable is initialized in the function jffs2_i_init_once, while other members are initialized in the function jffs2_init_inode_info. The function jffs2_init_inode_info is called after iget_locked, but in the iget_locked function, the destroy_inode process is triggered, which releases the inode and consequently, the target member of the inode is not initialized.In concurrent high pressure scenarios, iget_locked may enter the destroy_inode branch as described in the code. Since the destroy_inode functionality of jffs2 only releases the target, the fix method is to set target to NULL in jffs2_i_init_once. Signed-off-by: Wang Yong Reviewed-by: Lu Zhongjun Reviewed-by: Yang Tao Cc: Xu Xin Cc: Yang Yang Signed-off-by: Richard Weinberger --- fs/jffs2/super.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index aede1be4dc0c..4545f885c41e 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -58,6 +58,7 @@ static void jffs2_i_init_once(void *foo) struct jffs2_inode_info *f = foo; mutex_init(&f->sem); + f->target = NULL; inode_init_once(&f->vfs_inode); } -- cgit v1.2.3-59-g8ed1b From c91ecba9e421e4f2c9219cf5042fa63a12025310 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Thu, 9 May 2024 15:25:39 +0900 Subject: ksmbd: avoid to send duplicate oplock break notifications This patch fixes generic/011 when oplocks is enable. Avoid to send duplicate oplock break notifications like smb2 leases case. Fixes: 97c2ec64667b ("ksmbd: avoid to send duplicate lease break notifications") Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/oplock.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index b9d9116fc2b3..a8f52c4ebbda 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -610,19 +610,24 @@ static int oplock_break_pending(struct oplock_info *opinfo, int req_op_level) if (opinfo->op_state == OPLOCK_CLOSING) return -ENOENT; else if (opinfo->level <= req_op_level) { - if (opinfo->is_lease && - opinfo->o_lease->state != - (SMB2_LEASE_HANDLE_CACHING_LE | - SMB2_LEASE_READ_CACHING_LE)) + if (opinfo->is_lease == false) + return 1; + + if (opinfo->o_lease->state != + (SMB2_LEASE_HANDLE_CACHING_LE | + SMB2_LEASE_READ_CACHING_LE)) return 1; } } if (opinfo->level <= req_op_level) { - if (opinfo->is_lease && - opinfo->o_lease->state != - (SMB2_LEASE_HANDLE_CACHING_LE | - SMB2_LEASE_READ_CACHING_LE)) { + if (opinfo->is_lease == false) { + wake_up_oplock_break(opinfo); + return 1; + } + if (opinfo->o_lease->state != + (SMB2_LEASE_HANDLE_CACHING_LE | + SMB2_LEASE_READ_CACHING_LE)) { wake_up_oplock_break(opinfo); return 1; } -- cgit v1.2.3-59-g8ed1b From d1654fd98be731b08d2b7b857ef2026df6767e3d Mon Sep 17 00:00:00 2001 From: Günther Noack Date: Fri, 19 Apr 2024 16:11:22 +0000 Subject: fs/ioctl: Add a comment to keep the logic in sync with LSM policies MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Landlock's IOCTL support needs to partially replicate the list of IOCTLs from do_vfs_ioctl(). The list of commands implemented in do_vfs_ioctl() should be kept in sync with Landlock's IOCTL policies. Suggested-by: Paul Moore Suggested-by: Mickaël Salaün Signed-off-by: Günther Noack Link: https://lore.kernel.org/r/20240419161122.2023765-12-gnoack@google.com Signed-off-by: Mickaël Salaün --- fs/ioctl.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/ioctl.c b/fs/ioctl.c index fb0628e680c4..64776891120c 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -796,6 +796,9 @@ static int ioctl_get_fs_sysfs_path(struct file *file, void __user *argp) * * When you add any new common ioctls to the switches above and below, * please ensure they have compatible arguments in compat mode. + * + * The LSM mailing list should also be notified of any command additions or + * changes, as specific LSMs may be affected. */ static int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, unsigned long arg) -- cgit v1.2.3-59-g8ed1b From 9f9bef9bc5c68a6ca9bbffcd29cd188167c29ae1 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Thu, 11 Apr 2024 09:35:42 -0600 Subject: smb: smb2pdu.h: Avoid -Wflex-array-member-not-at-end warnings -Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally. So, in order to avoid ending up with a flexible-array member in the middle of multiple other structs, we use the `__struct_group()` helper to separate the flexible array from the rest of the members in the flexible structure, and use the tagged `struct create_context_hdr` instead of `struct create_context`. So, with these changes, fix 51 of the following warnings[1]: fs/smb/client/../common/smb2pdu.h:1225:31: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Link: https://gist.github.com/GustavoARSilva/772526a39be3dd4db39e71497f0a9893 [1] Link: https://github.com/KSPP/linux/issues/202 Signed-off-by: Gustavo A. R. Silva Signed-off-by: Steve French --- fs/smb/client/smb2pdu.h | 12 ++++++------ fs/smb/common/smb2pdu.h | 33 ++++++++++++++++++--------------- fs/smb/server/smb2pdu.h | 18 +++++++++--------- 3 files changed, 33 insertions(+), 30 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/smb2pdu.h b/fs/smb/client/smb2pdu.h index 2fccf0d4f53d..5c458ab3b05a 100644 --- a/fs/smb/client/smb2pdu.h +++ b/fs/smb/client/smb2pdu.h @@ -145,7 +145,7 @@ struct durable_context_v2 { } __packed; struct create_durable_v2 { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct durable_context_v2 dcontext; } __packed; @@ -167,7 +167,7 @@ struct durable_reconnect_context_v2_rsp { } __packed; struct create_durable_handle_reconnect_v2 { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct durable_reconnect_context_v2 dcontext; __u8 Pad[4]; @@ -175,7 +175,7 @@ struct create_durable_handle_reconnect_v2 { /* See MS-SMB2 2.2.13.2.5 */ struct crt_twarp_ctxt { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; __le64 Timestamp; @@ -183,12 +183,12 @@ struct crt_twarp_ctxt { /* See MS-SMB2 2.2.13.2.9 */ struct crt_query_id_ctxt { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; } __packed; struct crt_sd_ctxt { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct smb3_sd sd; } __packed; @@ -415,7 +415,7 @@ struct smb2_posix_info_parsed { }; struct smb2_create_ea_ctx { - struct create_context ctx; + struct create_context_hdr ctx; __u8 name[8]; struct smb2_file_full_ea_info ea; } __packed; diff --git a/fs/smb/common/smb2pdu.h b/fs/smb/common/smb2pdu.h index 202ff9128156..8d10be1fe18a 100644 --- a/fs/smb/common/smb2pdu.h +++ b/fs/smb/common/smb2pdu.h @@ -1171,12 +1171,15 @@ struct smb2_server_client_notification { #define SMB2_CREATE_FLAG_REPARSEPOINT 0x01 struct create_context { - __le32 Next; - __le16 NameOffset; - __le16 NameLength; - __le16 Reserved; - __le16 DataOffset; - __le32 DataLength; + /* New members must be added within the struct_group() macro below. */ + __struct_group(create_context_hdr, hdr, __packed, + __le32 Next; + __le16 NameOffset; + __le16 NameLength; + __le16 Reserved; + __le16 DataOffset; + __le32 DataLength; + ); __u8 Buffer[]; } __packed; @@ -1222,7 +1225,7 @@ struct smb2_create_rsp { } __packed; struct create_posix { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[16]; __le32 Mode; __u32 Reserved; @@ -1230,7 +1233,7 @@ struct create_posix { /* See MS-SMB2 2.2.13.2.3 and MS-SMB2 2.2.13.2.4 */ struct create_durable { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; union { __u8 Reserved[16]; @@ -1243,14 +1246,14 @@ struct create_durable { /* See MS-SMB2 2.2.13.2.5 */ struct create_mxac_req { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; __le64 Timestamp; } __packed; /* See MS-SMB2 2.2.14.2.5 */ struct create_mxac_rsp { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; __le32 QueryStatus; __le32 MaximalAccess; @@ -1286,13 +1289,13 @@ struct lease_context_v2 { } __packed; struct create_lease { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct lease_context lcontext; } __packed; struct create_lease_v2 { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct lease_context_v2 lcontext; __u8 Pad[4]; @@ -1300,7 +1303,7 @@ struct create_lease_v2 { /* See MS-SMB2 2.2.14.2.9 */ struct create_disk_id_rsp { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; __le64 DiskFileId; __le64 VolumeId; @@ -1309,7 +1312,7 @@ struct create_disk_id_rsp { /* See MS-SMB2 2.2.13.2.13 */ struct create_app_inst_id { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[16]; __le32 StructureSize; /* Must be 20 */ __u16 Reserved; @@ -1318,7 +1321,7 @@ struct create_app_inst_id { /* See MS-SMB2 2.2.13.2.15 */ struct create_app_inst_id_vers { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[16]; __le32 StructureSize; /* Must be 24 */ __u16 Reserved; diff --git a/fs/smb/server/smb2pdu.h b/fs/smb/server/smb2pdu.h index bd1d2a0e9203..643f5e1cfe35 100644 --- a/fs/smb/server/smb2pdu.h +++ b/fs/smb/server/smb2pdu.h @@ -64,7 +64,7 @@ struct preauth_integrity_info { #define SMB2_SESSION_TIMEOUT (10 * HZ) struct create_durable_req_v2 { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; __le32 Timeout; __le32 Flags; @@ -73,7 +73,7 @@ struct create_durable_req_v2 { } __packed; struct create_durable_reconn_req { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; union { __u8 Reserved[16]; @@ -85,7 +85,7 @@ struct create_durable_reconn_req { } __packed; struct create_durable_reconn_v2_req { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct { __u64 PersistentFileId; @@ -96,13 +96,13 @@ struct create_durable_reconn_v2_req { } __packed; struct create_alloc_size_req { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; __le64 AllocationSize; } __packed; struct create_durable_rsp { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; union { __u8 Reserved[8]; @@ -114,7 +114,7 @@ struct create_durable_rsp { /* Flags */ #define SMB2_DHANDLE_FLAG_PERSISTENT 0x00000002 struct create_durable_v2_rsp { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; __le32 Timeout; __le32 Flags; @@ -122,7 +122,7 @@ struct create_durable_v2_rsp { /* equivalent of the contents of SMB3.1.1 POSIX open context response */ struct create_posix_rsp { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[16]; __le32 nlink; __le32 reparse_tag; @@ -381,13 +381,13 @@ struct smb2_ea_info { } __packed; /* level 15 Query */ struct create_ea_buf_req { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct smb2_ea_info ea; } __packed; struct create_sd_buf_req { - struct create_context ccontext; + struct create_context_hdr ccontext; __u8 Name[8]; struct smb_ntsd ntsd; } __packed; -- cgit v1.2.3-59-g8ed1b From 29b4c7bb8565118e2c7e08709fce0dbe8bf61011 Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 13 May 2024 12:49:22 -0500 Subject: cifs: Change from mempool_destroy to mempool_exit for request pools insmod followed by rmmod was oopsing with the new mempools cifs request patch Fixes: edea94a69730 ("cifs: Add mempools for cifs_io_request and cifs_io_subrequest structs") Suggested-by: David Howells Reviewed-by: Enzo Matsumiya Signed-off-by: Steve French --- fs/smb/client/cifsfs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 6e1698614745..ec5b639f421a 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -1790,9 +1790,9 @@ nomem_req: static void cifs_destroy_netfs(void) { - mempool_destroy(&cifs_io_subrequest_pool); + mempool_exit(&cifs_io_subrequest_pool); kmem_cache_destroy(cifs_io_subrequest_cachep); - mempool_destroy(&cifs_io_request_pool); + mempool_exit(&cifs_io_request_pool); kmem_cache_destroy(cifs_io_request_cachep); } -- cgit v1.2.3-59-g8ed1b From 14b1cd25346b1d615616a9c2dfdad9b4e6581e0d Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 13 May 2024 17:02:05 -0500 Subject: cifs: Fix locking in cifs_strict_readv() Fix to take the i_rwsem (through the netfs locking wrappers) before taking cinode->lock_sem. Fixes: 3ee1a1fc3981 ("cifs: Cut over to using netfslib") Reported-by: Enzo Matsumiya Signed-off-by: David Howells Signed-off-by: Steve French --- fs/netfs/direct_read.c | 3 ++- fs/smb/client/cifsglob.h | 1 + fs/smb/client/file.c | 34 +++++++++++++++++++++++++--------- include/linux/netfs.h | 1 + 4 files changed, 29 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/netfs/direct_read.c b/fs/netfs/direct_read.c index ad4370b3935d..10a1e4da6bda 100644 --- a/fs/netfs/direct_read.c +++ b/fs/netfs/direct_read.c @@ -26,7 +26,7 @@ * * The caller must hold any appropriate locks. */ -static ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *iter) +ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *iter) { struct netfs_io_request *rreq; ssize_t ret; @@ -98,6 +98,7 @@ out: iov_iter_revert(iter, orig_count - iov_iter_count(iter)); return ret; } +EXPORT_SYMBOL(netfs_unbuffered_read_iter_locked); /** * netfs_unbuffered_read_iter - Perform an unbuffered or direct I/O read diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 65574e69ba4f..73482734a8d8 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -1995,6 +1995,7 @@ require use of the stronger protocol */ * ->chans_need_reconnect * ->chans_in_reconnect * cifs_tcon->tc_lock (anything that is not protected by another lock and can change) + * inode->i_rwsem, taken by fs/netfs/locking.c e.g. should be taken before cifsInodeInfo locks * cifsInodeInfo->open_file_lock cifsInodeInfo->openFileList cifs_alloc_inode * cifsInodeInfo->writers_lock cifsInodeInfo->writers cifsInodeInfo_alloc * cifsInodeInfo->lock_sem cifsInodeInfo->llist cifs_init_once diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 4c981ce89f8a..9d38294a7e68 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -2916,16 +2916,32 @@ cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) * We need to hold the sem to be sure nobody modifies lock list * with a brlock that prevents reading. */ - down_read(&cinode->lock_sem); - if (!cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(to), - tcon->ses->server->vals->shared_lock_type, - 0, NULL, CIFS_READ_OP)) { - if (iocb->ki_flags & IOCB_DIRECT) - rc = netfs_unbuffered_read_iter(iocb, to); - else - rc = netfs_buffered_read_iter(iocb, to); + if (iocb->ki_flags & IOCB_DIRECT) { + rc = netfs_start_io_direct(inode); + if (rc < 0) + goto out; + down_read(&cinode->lock_sem); + if (!cifs_find_lock_conflict( + cfile, iocb->ki_pos, iov_iter_count(to), + tcon->ses->server->vals->shared_lock_type, + 0, NULL, CIFS_READ_OP)) + rc = netfs_unbuffered_read_iter_locked(iocb, to); + up_read(&cinode->lock_sem); + netfs_end_io_direct(inode); + } else { + rc = netfs_start_io_read(inode); + if (rc < 0) + goto out; + down_read(&cinode->lock_sem); + if (!cifs_find_lock_conflict( + cfile, iocb->ki_pos, iov_iter_count(to), + tcon->ses->server->vals->shared_lock_type, + 0, NULL, CIFS_READ_OP)) + rc = filemap_read(iocb, to, 0); + up_read(&cinode->lock_sem); + netfs_end_io_read(inode); } - up_read(&cinode->lock_sem); +out: return rc; } diff --git a/include/linux/netfs.h b/include/linux/netfs.h index f45d06284f2f..ca56a4428043 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -389,6 +389,7 @@ struct netfs_cache_ops { }; /* High-level read API. */ +ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *iter); ssize_t netfs_unbuffered_read_iter(struct kiocb *iocb, struct iov_iter *iter); ssize_t netfs_buffered_read_iter(struct kiocb *iocb, struct iov_iter *iter); ssize_t netfs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter); -- cgit v1.2.3-59-g8ed1b From 92ef0fd55ac80dfc2e4654edfe5d1ddfa6e070fe Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 9 May 2024 09:20:08 -0600 Subject: net: change proto and proto_ops accept type Rather than pass in flags, error pointer, and whether this is a kernel invocation or not, add a struct proto_accept_arg struct as the argument. This then holds all of these arguments, and prepares accept for being able to pass back more information. No functional changes in this patch. Acked-by: Jakub Kicinski Signed-off-by: Jens Axboe --- crypto/af_alg.c | 11 ++++++----- crypto/algif_hash.c | 10 +++++----- drivers/xen/pvcalls-back.c | 6 +++++- fs/ocfs2/cluster/tcp.c | 5 ++++- include/crypto/if_alg.h | 3 ++- include/linux/net.h | 4 +++- include/net/inet_common.h | 4 ++-- include/net/inet_connection_sock.h | 2 +- include/net/sock.h | 12 +++++++++--- net/atm/svc.c | 8 ++++---- net/ax25/af_ax25.c | 6 +++--- net/bluetooth/iso.c | 4 ++-- net/bluetooth/l2cap_sock.c | 4 ++-- net/bluetooth/rfcomm/sock.c | 6 +++--- net/bluetooth/sco.c | 4 ++-- net/core/sock.c | 4 ++-- net/ipv4/af_inet.c | 10 +++++----- net/ipv4/inet_connection_sock.c | 6 +++--- net/iucv/af_iucv.c | 4 ++-- net/llc/af_llc.c | 7 +++---- net/mptcp/protocol.c | 11 +++++------ net/netrom/af_netrom.c | 6 +++--- net/nfc/llcp_sock.c | 4 ++-- net/phonet/pep.c | 12 ++++++------ net/phonet/socket.c | 7 +++---- net/rds/tcp_listen.c | 6 +++++- net/rose/af_rose.c | 6 +++--- net/sctp/socket.c | 8 ++++---- net/smc/af_smc.c | 6 +++--- net/socket.c | 13 ++++++++++--- net/tipc/socket.c | 13 +++++-------- net/unix/af_unix.c | 21 ++++++++++----------- net/vmw_vsock/af_vsock.c | 6 +++--- net/x25/af_x25.c | 4 ++-- 34 files changed, 132 insertions(+), 111 deletions(-) (limited to 'fs') diff --git a/crypto/af_alg.c b/crypto/af_alg.c index 5bc6d0fa7498..18cfead0081d 100644 --- a/crypto/af_alg.c +++ b/crypto/af_alg.c @@ -407,7 +407,8 @@ unlock: return err; } -int af_alg_accept(struct sock *sk, struct socket *newsock, bool kern) +int af_alg_accept(struct sock *sk, struct socket *newsock, + struct proto_accept_arg *arg) { struct alg_sock *ask = alg_sk(sk); const struct af_alg_type *type; @@ -422,7 +423,7 @@ int af_alg_accept(struct sock *sk, struct socket *newsock, bool kern) if (!type) goto unlock; - sk2 = sk_alloc(sock_net(sk), PF_ALG, GFP_KERNEL, &alg_proto, kern); + sk2 = sk_alloc(sock_net(sk), PF_ALG, GFP_KERNEL, &alg_proto, arg->kern); err = -ENOMEM; if (!sk2) goto unlock; @@ -468,10 +469,10 @@ unlock: } EXPORT_SYMBOL_GPL(af_alg_accept); -static int alg_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int alg_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { - return af_alg_accept(sock->sk, newsock, kern); + return af_alg_accept(sock->sk, newsock, arg); } static const struct proto_ops alg_proto_ops = { diff --git a/crypto/algif_hash.c b/crypto/algif_hash.c index e24c829d7a01..7c7394d46a23 100644 --- a/crypto/algif_hash.c +++ b/crypto/algif_hash.c @@ -223,8 +223,8 @@ unlock: return err ?: len; } -static int hash_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int hash_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct alg_sock *ask = alg_sk(sk); @@ -252,7 +252,7 @@ static int hash_accept(struct socket *sock, struct socket *newsock, int flags, if (err) goto out_free_state; - err = af_alg_accept(ask->parent, newsock, kern); + err = af_alg_accept(ask->parent, newsock, arg); if (err) goto out_free_state; @@ -355,7 +355,7 @@ static int hash_recvmsg_nokey(struct socket *sock, struct msghdr *msg, } static int hash_accept_nokey(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { int err; @@ -363,7 +363,7 @@ static int hash_accept_nokey(struct socket *sock, struct socket *newsock, if (err) return err; - return hash_accept(sock, newsock, flags, kern); + return hash_accept(sock, newsock, arg); } static struct proto_ops algif_hash_ops_nokey = { diff --git a/drivers/xen/pvcalls-back.c b/drivers/xen/pvcalls-back.c index d52593466a79..fd7ed65e0197 100644 --- a/drivers/xen/pvcalls-back.c +++ b/drivers/xen/pvcalls-back.c @@ -517,6 +517,10 @@ static void __pvcalls_back_accept(struct work_struct *work) { struct sockpass_mapping *mappass = container_of( work, struct sockpass_mapping, register_work); + struct proto_accept_arg arg = { + .flags = O_NONBLOCK, + .kern = true, + }; struct sock_mapping *map; struct pvcalls_ioworker *iow; struct pvcalls_fedata *fedata; @@ -548,7 +552,7 @@ static void __pvcalls_back_accept(struct work_struct *work) sock->type = mappass->sock->type; sock->ops = mappass->sock->ops; - ret = inet_accept(mappass->sock, sock, O_NONBLOCK, true); + ret = inet_accept(mappass->sock, sock, &arg); if (ret == -EAGAIN) { sock_release(sock); return; diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index 960080753d3b..2b8fa3e782fb 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -1784,6 +1784,9 @@ static int o2net_accept_one(struct socket *sock, int *more) struct o2nm_node *node = NULL; struct o2nm_node *local_node = NULL; struct o2net_sock_container *sc = NULL; + struct proto_accept_arg arg = { + .flags = O_NONBLOCK, + }; struct o2net_node *nn; unsigned int nofs_flag; @@ -1802,7 +1805,7 @@ static int o2net_accept_one(struct socket *sock, int *more) new_sock->type = sock->type; new_sock->ops = sock->ops; - ret = sock->ops->accept(sock, new_sock, O_NONBLOCK, false); + ret = sock->ops->accept(sock, new_sock, &arg); if (ret < 0) goto out; diff --git a/include/crypto/if_alg.h b/include/crypto/if_alg.h index 78ecaf5db04c..f7b3b93f3a49 100644 --- a/include/crypto/if_alg.h +++ b/include/crypto/if_alg.h @@ -166,7 +166,8 @@ int af_alg_unregister_type(const struct af_alg_type *type); int af_alg_release(struct socket *sock); void af_alg_release_parent(struct sock *sk); -int af_alg_accept(struct sock *sk, struct socket *newsock, bool kern); +int af_alg_accept(struct sock *sk, struct socket *newsock, + struct proto_accept_arg *arg); void af_alg_free_sg(struct af_alg_sgl *sgl); diff --git a/include/linux/net.h b/include/linux/net.h index 15df6d5f27a7..688320b79fcc 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -153,6 +153,7 @@ struct sockaddr; struct msghdr; struct module; struct sk_buff; +struct proto_accept_arg; typedef int (*sk_read_actor_t)(read_descriptor_t *, struct sk_buff *, unsigned int, size_t); typedef int (*skb_read_actor_t)(struct sock *, struct sk_buff *); @@ -171,7 +172,8 @@ struct proto_ops { int (*socketpair)(struct socket *sock1, struct socket *sock2); int (*accept) (struct socket *sock, - struct socket *newsock, int flags, bool kern); + struct socket *newsock, + struct proto_accept_arg *arg); int (*getname) (struct socket *sock, struct sockaddr *addr, int peer); diff --git a/include/net/inet_common.h b/include/net/inet_common.h index f50a644d87a9..c17a6585d0b0 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -29,8 +29,8 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags, int is_sendmsg); int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags); -int inet_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern); +int inet_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg); void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *newsk); int inet_send_prepare(struct sock *sk); diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 20e7b0c0b3d1..7d6b1254c92d 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -250,7 +250,7 @@ inet_csk_rto_backoff(const struct inet_connection_sock *icsk, return (unsigned long)min_t(u64, when, max_when); } -struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern); +struct sock *inet_csk_accept(struct sock *sk, struct proto_accept_arg *arg); int inet_csk_get_port(struct sock *sk, unsigned short snum); diff --git a/include/net/sock.h b/include/net/sock.h index 0450494a1766..217079b3e3e8 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1194,6 +1194,12 @@ static inline void sk_prot_clear_nulls(struct sock *sk, int size) size - offsetof(struct sock, sk_node.pprev)); } +struct proto_accept_arg { + int flags; + int err; + bool kern; +}; + /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface */ @@ -1208,8 +1214,8 @@ struct proto { int addr_len); int (*disconnect)(struct sock *sk, int flags); - struct sock * (*accept)(struct sock *sk, int flags, int *err, - bool kern); + struct sock * (*accept)(struct sock *sk, + struct proto_accept_arg *arg); int (*ioctl)(struct sock *sk, int cmd, int *karg); @@ -1804,7 +1810,7 @@ int sock_cmsg_send(struct sock *sk, struct msghdr *msg, int sock_no_bind(struct socket *, struct sockaddr *, int); int sock_no_connect(struct socket *, struct sockaddr *, int, int); int sock_no_socketpair(struct socket *, struct socket *); -int sock_no_accept(struct socket *, struct socket *, int, bool); +int sock_no_accept(struct socket *, struct socket *, struct proto_accept_arg *); int sock_no_getname(struct socket *, struct sockaddr *, int); int sock_no_ioctl(struct socket *, unsigned int, unsigned long); int sock_no_listen(struct socket *, int); diff --git a/net/atm/svc.c b/net/atm/svc.c index 36a814f1fbd1..f8137ae693b0 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -324,8 +324,8 @@ out: return error; } -static int svc_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int svc_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sk_buff *skb; @@ -336,7 +336,7 @@ static int svc_accept(struct socket *sock, struct socket *newsock, int flags, lock_sock(sk); - error = svc_create(sock_net(sk), newsock, 0, kern); + error = svc_create(sock_net(sk), newsock, 0, arg->kern); if (error) goto out; @@ -355,7 +355,7 @@ static int svc_accept(struct socket *sock, struct socket *newsock, int flags, error = -sk->sk_err; break; } - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { error = -EAGAIN; break; } diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index 9169efb2f43a..8077cf2ee448 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -1373,8 +1373,8 @@ out_release: return err; } -static int ax25_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int ax25_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sk_buff *skb; struct sock *newsk; @@ -1409,7 +1409,7 @@ static int ax25_accept(struct socket *sock, struct socket *newsock, int flags, if (skb) break; - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { err = -EWOULDBLOCK; break; } diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index ef0cc80b4c0c..2a075119d65d 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1186,7 +1186,7 @@ done: } static int iso_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *ch; @@ -1195,7 +1195,7 @@ static int iso_sock_accept(struct socket *sock, struct socket *newsock, lock_sock(sk); - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 5cc83f906c12..125dddc77452 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -327,7 +327,7 @@ done: } static int l2cap_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *nsk; @@ -336,7 +336,7 @@ static int l2cap_sock_accept(struct socket *sock, struct socket *newsock, lock_sock_nested(sk, L2CAP_NESTING_PARENT); - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 29aa07e9db9d..37d63d768afb 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -468,8 +468,8 @@ done: return err; } -static int rfcomm_sock_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int rfcomm_sock_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *nsk; @@ -483,7 +483,7 @@ static int rfcomm_sock_accept(struct socket *sock, struct socket *newsock, int f goto done; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index e0ad30862ee4..94c6f2b46279 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -647,7 +647,7 @@ done: } static int sco_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *ch; @@ -656,7 +656,7 @@ static int sco_sock_accept(struct socket *sock, struct socket *newsock, lock_sock(sk); - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/core/sock.c b/net/core/sock.c index 8d6e638b5426..8629f9aecf91 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3241,8 +3241,8 @@ int sock_no_socketpair(struct socket *sock1, struct socket *sock2) } EXPORT_SYMBOL(sock_no_socketpair); -int sock_no_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +int sock_no_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { return -EOPNOTSUPP; } diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a7bad18bc8b5..de3449e16b89 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -771,16 +771,16 @@ void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *new * Accept a pending connection. The TCP layer now gives BSD semantics. */ -int inet_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +int inet_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk1 = sock->sk, *sk2; - int err = -EINVAL; /* IPV6_ADDRFORM can change sk->sk_prot under us. */ - sk2 = READ_ONCE(sk1->sk_prot)->accept(sk1, flags, &err, kern); + arg->err = -EINVAL; + sk2 = READ_ONCE(sk1->sk_prot)->accept(sk1, arg); if (!sk2) - return err; + return arg->err; lock_sock(sk2); __inet_accept(sock, newsock, sk2); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 3b38610958ee..7734d189c66b 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -661,7 +661,7 @@ static int inet_csk_wait_for_connect(struct sock *sk, long timeo) /* * This will accept the next outstanding connection. */ -struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern) +struct sock *inet_csk_accept(struct sock *sk, struct proto_accept_arg *arg) { struct inet_connection_sock *icsk = inet_csk(sk); struct request_sock_queue *queue = &icsk->icsk_accept_queue; @@ -680,7 +680,7 @@ struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern) /* Find already established connection */ if (reqsk_queue_empty(queue)) { - long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + long timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); /* If this is a non blocking socket don't sleep */ error = -EAGAIN; @@ -745,7 +745,7 @@ out: out_err: newsk = NULL; req = NULL; - *err = error; + arg->err = error; goto out; } EXPORT_SYMBOL(inet_csk_accept); diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index c951bb9cc2e0..c3b0b610b0aa 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -795,7 +795,7 @@ done: /* Accept a pending connection */ static int iucv_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DECLARE_WAITQUEUE(wait, current); struct sock *sk = sock->sk, *nsk; @@ -809,7 +809,7 @@ static int iucv_sock_accept(struct socket *sock, struct socket *newsock, goto done; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); /* Wait for an incoming connection */ add_wait_queue_exclusive(sk_sleep(sk), &wait); diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index fde1140d899e..4eb52add7103 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -688,14 +688,13 @@ static void llc_cmsg_rcv(struct msghdr *msg, struct sk_buff *skb) * llc_ui_accept - accept a new incoming connection. * @sock: Socket which connections arrive on. * @newsock: Socket to move incoming connection to. - * @flags: User specified operational flags. - * @kern: If the socket is kernel internal + * @arg: User specified arguments * * Accept a new incoming connection. * Returns 0 upon success, negative otherwise. */ -static int llc_ui_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int llc_ui_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk, *newsk; struct llc_sock *llc, *newllc; diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index bb8f96f2b86f..815ce439183c 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3881,11 +3881,10 @@ unlock: } static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { struct mptcp_sock *msk = mptcp_sk(sock->sk); struct sock *ssk, *newsk; - int err; pr_debug("msk=%p", msk); @@ -3897,9 +3896,9 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, return -EINVAL; pr_debug("ssk=%p, listener=%p", ssk, mptcp_subflow_ctx(ssk)); - newsk = inet_csk_accept(ssk, flags, &err, kern); + newsk = inet_csk_accept(ssk, arg); if (!newsk) - return err; + return arg->err; pr_debug("newsk=%p, subflow is mptcp=%d", newsk, sk_is_mptcp(newsk)); if (sk_is_mptcp(newsk)) { @@ -3920,7 +3919,7 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, newsk = new_mptcp_sock; MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_MPCAPABLEPASSIVEACK); - newsk->sk_kern_sock = kern; + newsk->sk_kern_sock = arg->kern; lock_sock(newsk); __inet_accept(sock, newsock, newsk); @@ -3949,7 +3948,7 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, } } else { tcpfallback: - newsk->sk_kern_sock = kern; + newsk->sk_kern_sock = arg->kern; lock_sock(newsk); __inet_accept(sock, newsock, newsk); /* we are being invoked after accepting a non-mp-capable diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index 104a80b75477..6ee148f0e6d0 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -772,8 +772,8 @@ out_release: return err; } -static int nr_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int nr_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sk_buff *skb; struct sock *newsk; @@ -805,7 +805,7 @@ static int nr_accept(struct socket *sock, struct socket *newsock, int flags, if (skb) break; - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { err = -EWOULDBLOCK; break; } diff --git a/net/nfc/llcp_sock.c b/net/nfc/llcp_sock.c index d5344563e525..57a2f97004e1 100644 --- a/net/nfc/llcp_sock.c +++ b/net/nfc/llcp_sock.c @@ -447,7 +447,7 @@ struct sock *nfc_llcp_accept_dequeue(struct sock *parent, } static int llcp_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DECLARE_WAITQUEUE(wait, current); struct sock *sk = sock->sk, *new_sk; @@ -463,7 +463,7 @@ static int llcp_sock_accept(struct socket *sock, struct socket *newsock, goto error; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); /* Wait for an incoming connection. */ add_wait_queue_exclusive(sk_sleep(sk), &wait); diff --git a/net/phonet/pep.c b/net/phonet/pep.c index 3dd5f52bc1b5..53a858478e22 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -759,8 +759,8 @@ static void pep_sock_close(struct sock *sk, long timeout) sock_put(sk); } -static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, - bool kern) +static struct sock *pep_sock_accept(struct sock *sk, + struct proto_accept_arg *arg) { struct pep_sock *pn = pep_sk(sk), *newpn; struct sock *newsk = NULL; @@ -772,8 +772,8 @@ static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, u8 pipe_handle, enabled, n_sb; u8 aligned = 0; - skb = skb_recv_datagram(sk, (flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, - errp); + skb = skb_recv_datagram(sk, (arg->flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, + &arg->err); if (!skb) return NULL; @@ -836,7 +836,7 @@ static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, /* Create a new to-be-accepted sock */ newsk = sk_alloc(sock_net(sk), PF_PHONET, GFP_KERNEL, sk->sk_prot, - kern); + arg->kern); if (!newsk) { pep_reject_conn(sk, skb, PN_PIPE_ERR_OVERLOAD, GFP_KERNEL); err = -ENOBUFS; @@ -878,7 +878,7 @@ static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, drop: release_sock(sk); kfree_skb(skb); - *errp = err; + arg->err = err; return newsk; } diff --git a/net/phonet/socket.c b/net/phonet/socket.c index 1018340d89a7..5ce0b3ee5def 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -292,18 +292,17 @@ out: } static int pn_socket_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sock *newsk; - int err; if (unlikely(sk->sk_state != TCP_LISTEN)) return -EINVAL; - newsk = sk->sk_prot->accept(sk, flags, &err, kern); + newsk = sk->sk_prot->accept(sk, arg); if (!newsk) - return err; + return arg->err; lock_sock(newsk); sock_graft(newsk, newsock); diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 05008ce5c421..d89bd8d0c354 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -105,6 +105,10 @@ int rds_tcp_accept_one(struct socket *sock) int conn_state; struct rds_conn_path *cp; struct in6_addr *my_addr, *peer_addr; + struct proto_accept_arg arg = { + .flags = O_NONBLOCK, + .kern = true, + }; #if !IS_ENABLED(CONFIG_IPV6) struct in6_addr saddr, daddr; #endif @@ -119,7 +123,7 @@ int rds_tcp_accept_one(struct socket *sock) if (ret) goto out; - ret = sock->ops->accept(sock, new_sock, O_NONBLOCK, true); + ret = sock->ops->accept(sock, new_sock, &arg); if (ret < 0) goto out; diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index ef81d019b20f..59050caab65c 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -919,8 +919,8 @@ out_release: return err; } -static int rose_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int rose_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sk_buff *skb; struct sock *newsk; @@ -953,7 +953,7 @@ static int rose_accept(struct socket *sock, struct socket *newsock, int flags, if (skb) break; - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { err = -EWOULDBLOCK; break; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 64196b1dce1d..c009383369b2 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4847,7 +4847,7 @@ static int sctp_disconnect(struct sock *sk, int flags) * descriptor will be returned from accept() to represent the newly * formed association. */ -static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) +static struct sock *sctp_accept(struct sock *sk, struct proto_accept_arg *arg) { struct sctp_sock *sp; struct sctp_endpoint *ep; @@ -4871,7 +4871,7 @@ static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) goto out; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); error = sctp_wait_for_accept(sk, timeo); if (error) @@ -4882,7 +4882,7 @@ static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) */ asoc = list_entry(ep->asocs.next, struct sctp_association, asocs); - newsk = sp->pf->create_accept_sk(sk, asoc, kern); + newsk = sp->pf->create_accept_sk(sk, asoc, arg->kern); if (!newsk) { error = -ENOMEM; goto out; @@ -4899,7 +4899,7 @@ static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) out: release_sock(sk); - *err = error; + arg->err = error; return newsk; } diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 9389f0cfa374..e50a286fd0fb 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -2689,7 +2689,7 @@ out: } static int smc_accept(struct socket *sock, struct socket *new_sock, - int flags, bool kern) + struct proto_accept_arg *arg) { struct sock *sk = sock->sk, *nsk; DECLARE_WAITQUEUE(wait, current); @@ -2708,7 +2708,7 @@ static int smc_accept(struct socket *sock, struct socket *new_sock, } /* Wait for an incoming connection */ - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); add_wait_queue_exclusive(sk_sleep(sk), &wait); while (!(nsk = smc_accept_dequeue(sk, new_sock))) { set_current_state(TASK_INTERRUPTIBLE); @@ -2735,7 +2735,7 @@ static int smc_accept(struct socket *sock, struct socket *new_sock, if (rc) goto out; - if (lsmc->sockopt_defer_accept && !(flags & O_NONBLOCK)) { + if (lsmc->sockopt_defer_accept && !(arg->flags & O_NONBLOCK)) { /* wait till data arrives on the socket */ timeo = msecs_to_jiffies(lsmc->sockopt_defer_accept * MSEC_PER_SEC); diff --git a/net/socket.c b/net/socket.c index 01a71ae10c35..6ff5f21d9633 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1898,6 +1898,9 @@ struct file *do_accept(struct file *file, unsigned file_flags, struct file *newfile; int err, len; struct sockaddr_storage address; + struct proto_accept_arg arg = { + .flags = file_flags, + }; const struct proto_ops *ops; sock = sock_from_file(file); @@ -1926,8 +1929,8 @@ struct file *do_accept(struct file *file, unsigned file_flags, if (err) goto out_fd; - err = ops->accept(sock, newsock, sock->file->f_flags | file_flags, - false); + arg.flags |= sock->file->f_flags; + err = ops->accept(sock, newsock, &arg); if (err < 0) goto out_fd; @@ -3580,6 +3583,10 @@ int kernel_accept(struct socket *sock, struct socket **newsock, int flags) { struct sock *sk = sock->sk; const struct proto_ops *ops = READ_ONCE(sock->ops); + struct proto_accept_arg arg = { + .flags = flags, + .kern = true, + }; int err; err = sock_create_lite(sk->sk_family, sk->sk_type, sk->sk_protocol, @@ -3587,7 +3594,7 @@ int kernel_accept(struct socket *sock, struct socket **newsock, int flags) if (err < 0) goto done; - err = ops->accept(sock, *newsock, flags, true); + err = ops->accept(sock, *newsock, &arg); if (err < 0) { sock_release(*newsock); *newsock = NULL; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 798397b6811e..2d58ecae4e21 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -146,8 +146,6 @@ static void tipc_data_ready(struct sock *sk); static void tipc_write_space(struct sock *sk); static void tipc_sock_destruct(struct sock *sk); static int tipc_release(struct socket *sock); -static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, - bool kern); static void tipc_sk_timeout(struct timer_list *t); static int tipc_sk_publish(struct tipc_sock *tsk, struct tipc_uaddr *ua); static int tipc_sk_withdraw(struct tipc_sock *tsk, struct tipc_uaddr *ua); @@ -2711,13 +2709,12 @@ static int tipc_wait_for_accept(struct socket *sock, long timeo) * tipc_accept - wait for connection request * @sock: listening socket * @new_sock: new socket that is to be connected - * @flags: file-related flags associated with socket - * @kern: caused by kernel or by userspace? + * @arg: arguments for accept * * Return: 0 on success, errno otherwise */ -static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, - bool kern) +static int tipc_accept(struct socket *sock, struct socket *new_sock, + struct proto_accept_arg *arg) { struct sock *new_sk, *sk = sock->sk; struct tipc_sock *new_tsock; @@ -2733,14 +2730,14 @@ static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, res = -EINVAL; goto exit; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); res = tipc_wait_for_accept(sock, timeo); if (res) goto exit; buf = skb_peek(&sk->sk_receive_queue); - res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, kern); + res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, arg->kern); if (res) goto exit; security_sk_clone(sock->sk, new_sock->sk); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index dc1651541723..26e3b5f1ee46 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -755,7 +755,7 @@ static int unix_bind(struct socket *, struct sockaddr *, int); static int unix_stream_connect(struct socket *, struct sockaddr *, int addr_len, int flags); static int unix_socketpair(struct socket *, struct socket *); -static int unix_accept(struct socket *, struct socket *, int, bool); +static int unix_accept(struct socket *, struct socket *, struct proto_accept_arg *arg); static int unix_getname(struct socket *, struct sockaddr *, int); static __poll_t unix_poll(struct file *, struct socket *, poll_table *); static __poll_t unix_dgram_poll(struct file *, struct socket *, @@ -1689,19 +1689,18 @@ static void unix_sock_inherit_flags(const struct socket *old, set_bit(SOCK_PASSSEC, &new->flags); } -static int unix_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int unix_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sk_buff *skb; struct sock *tsk; - int err; - err = -EOPNOTSUPP; + arg->err = -EOPNOTSUPP; if (sock->type != SOCK_STREAM && sock->type != SOCK_SEQPACKET) goto out; - err = -EINVAL; + arg->err = -EINVAL; if (sk->sk_state != TCP_LISTEN) goto out; @@ -1709,12 +1708,12 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, * so that no locks are necessary. */ - skb = skb_recv_datagram(sk, (flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, - &err); + skb = skb_recv_datagram(sk, (arg->flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, + &arg->err); if (!skb) { /* This means receive shutdown. */ - if (err == 0) - err = -EINVAL; + if (arg->err == 0) + arg->err = -EINVAL; goto out; } @@ -1732,7 +1731,7 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, return 0; out: - return err; + return arg->err; } diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 54ba7316f808..4b040285aa78 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1500,8 +1500,8 @@ out: return err; } -static int vsock_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int vsock_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *listener; int err; @@ -1528,7 +1528,7 @@ static int vsock_accept(struct socket *sock, struct socket *newsock, int flags, /* Wait for children sockets to appear; these are the new sockets * created upon connection establishment. */ - timeout = sock_rcvtimeo(listener, flags & O_NONBLOCK); + timeout = sock_rcvtimeo(listener, arg->flags & O_NONBLOCK); prepare_to_wait(sk_sleep(listener), &wait, TASK_INTERRUPTIBLE); while ((connected = vsock_dequeue_accept(listener)) == NULL && diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index d18d51412cc0..8dda4178497c 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -871,8 +871,8 @@ static int x25_wait_for_data(struct sock *sk, long timeout) return rc; } -static int x25_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int x25_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sock *newsk; -- cgit v1.2.3-59-g8ed1b From d4e9a968738bf66d3bb852dd5588d4c7afd6d7f4 Mon Sep 17 00:00:00 2001 From: Hao Ge Date: Mon, 13 May 2024 13:33:38 +0800 Subject: eventfs: Fix a possible null pointer dereference in eventfs_find_events() In function eventfs_find_events,there is a potential null pointer that may be caused by calling update_events_attr which will perform some operations on the members of the ei struct when ei is NULL. Hence,When ei->is_freed is set,return NULL directly. Link: https://lore.kernel.org/linux-trace-kernel/20240513053338.63017-1-hao.ge@linux.dev Cc: stable@vger.kernel.org Fixes: 8186fff7ab64 ("tracefs/eventfs: Use root and instance inodes as default ownership") Signed-off-by: Hao Ge Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index a878cea70f4c..0256afdd4acf 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -345,10 +345,9 @@ static struct eventfs_inode *eventfs_find_events(struct dentry *dentry) * If the ei is being freed, the ownership of the children * doesn't matter. */ - if (ei->is_freed) { - ei = NULL; - break; - } + if (ei->is_freed) + return NULL; + // Walk upwards until you find the events inode } while (!ei->is_events); -- cgit v1.2.3-59-g8ed1b From edfc6481faf896301cab940da776229fe39e9fc9 Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 14 May 2024 02:27:19 -0500 Subject: smb3: fix perf regression with cached writes with netfs conversion Write through mode is for cache=none, not for default (when caching is allowed if we have a lease). Some tests were running much, much more slowly as a result of disabling caching of writes by default. Reviewed-by: David Howells Reviewed-by: Enzo Matsumiya Fixes: 3ee1a1fc3981 ("cifs: Cut over to using netfslib") Signed-off-by: Steve French --- fs/smb/client/inode.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index e8bfeea23660..5d4b0fd3a59e 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -34,11 +34,8 @@ static void cifs_set_netfs_context(struct inode *inode) { struct cifsInodeInfo *cifs_i = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); netfs_inode_init(&cifs_i->netfs, &cifs_req_ops, true); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) - __set_bit(NETFS_ICTX_WRITETHROUGH, &cifs_i->netfs.flags); } static void cifs_set_ops(struct inode *inode) -- cgit v1.2.3-59-g8ed1b From 0d8968287a1cf7b03d07387dc871de3861b9f6b9 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Mon, 13 May 2024 08:40:27 -0700 Subject: f2fs: Add inline to f2fs_build_fault_attr() stub When building without CONFIG_F2FS_FAULT_INJECTION, there is a warning from each file that includes f2fs.h because the stub for f2fs_build_fault_attr() is missing inline: In file included from fs/f2fs/segment.c:21: fs/f2fs/f2fs.h:4605:12: warning: 'f2fs_build_fault_attr' defined but not used [-Wunused-function] 4605 | static int f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned long rate, | ^~~~~~~~~~~~~~~~~~~~~ Add the missing inline to resolve all of the warnings for this configuration. Fixes: 4ed886b187f4 ("f2fs: check validation of fault attrs in f2fs_build_fault_attr()") Signed-off-by: Nathan Chancellor Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/f2fs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index ef7de97be647..1974b6aff397 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -4602,8 +4602,8 @@ static inline bool f2fs_need_verity(const struct inode *inode, pgoff_t idx) extern int f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned long rate, unsigned long type); #else -static int f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned long rate, - unsigned long type) +static inline int f2fs_build_fault_attr(struct f2fs_sb_info *sbi, + unsigned long rate, unsigned long type) { return 0; } -- cgit v1.2.3-59-g8ed1b From 16409fdbb8828d7ae829bc4ac4e09e7ff02f8878 Mon Sep 17 00:00:00 2001 From: Wu Bo Date: Tue, 14 May 2024 05:35:29 -0600 Subject: f2fs: initialize last_block_in_bio variable Initialize last_block_in_bio of struct f2fs_bio_info and clean up code. Signed-off-by: Wu Bo Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/data.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index bdbd4195cadc..50ceb25b341d 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -595,17 +595,20 @@ int f2fs_init_write_merge_io(struct f2fs_sb_info *sbi) return -ENOMEM; for (j = HOT; j < n; j++) { - init_f2fs_rwsem(&sbi->write_io[i][j].io_rwsem); - sbi->write_io[i][j].sbi = sbi; - sbi->write_io[i][j].bio = NULL; - spin_lock_init(&sbi->write_io[i][j].io_lock); - INIT_LIST_HEAD(&sbi->write_io[i][j].io_list); - INIT_LIST_HEAD(&sbi->write_io[i][j].bio_list); - init_f2fs_rwsem(&sbi->write_io[i][j].bio_list_lock); + struct f2fs_bio_info *io = &sbi->write_io[i][j]; + + init_f2fs_rwsem(&io->io_rwsem); + io->sbi = sbi; + io->bio = NULL; + io->last_block_in_bio = 0; + spin_lock_init(&io->io_lock); + INIT_LIST_HEAD(&io->io_list); + INIT_LIST_HEAD(&io->bio_list); + init_f2fs_rwsem(&io->bio_list_lock); #ifdef CONFIG_BLK_DEV_ZONED - init_completion(&sbi->write_io[i][j].zone_wait); - sbi->write_io[i][j].zone_pending_bio = NULL; - sbi->write_io[i][j].bi_private = NULL; + init_completion(&io->zone_wait); + io->zone_pending_bio = NULL; + io->bi_private = NULL; #endif } } -- cgit v1.2.3-59-g8ed1b From 1fa7603d569b9e738e9581937ba8725cd7d39b48 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Thu, 2 May 2024 22:45:58 +0200 Subject: btrfs: qgroup: update rescan message levels and error codes On filesystems without enabled quotas there's still a warning message in the logs when rescan is called. In that case it's not a problem that should be reported, rescan can be called unconditionally. Change the error code to ENOTCONN which is used for 'quotas not enabled' elsewhere. Remove message (also a warning) when rescan is called during an ongoing rescan, this brings no useful information and the error code is sufficient. Change message levels to debug for now, they can be removed eventually. CC: stable@vger.kernel.org # 6.6+ Reviewed-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index eb28141d5c37..f93354a96909 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3820,14 +3820,14 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, /* we're resuming qgroup rescan at mount time */ if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN)) { - btrfs_warn(fs_info, + btrfs_debug(fs_info, "qgroup rescan init failed, qgroup rescan is not queued"); ret = -EINVAL; } else if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON)) { - btrfs_warn(fs_info, + btrfs_debug(fs_info, "qgroup rescan init failed, qgroup is not enabled"); - ret = -EINVAL; + ret = -ENOTCONN; } if (ret) @@ -3838,14 +3838,12 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, if (init_flags) { if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) { - btrfs_warn(fs_info, - "qgroup rescan is already in progress"); ret = -EINPROGRESS; } else if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON)) { - btrfs_warn(fs_info, + btrfs_debug(fs_info, "qgroup rescan init failed, qgroup is not enabled"); - ret = -EINVAL; + ret = -ENOTCONN; } else if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_DISABLED) { /* Quota disable is in progress */ ret = -EBUSY; -- cgit v1.2.3-59-g8ed1b From 2b8aa78cf1279ec5e418baa26bfed5df682568d8 Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Thu, 9 May 2024 15:34:40 -0700 Subject: btrfs: qgroup: fix qgroup id collision across mounts If we delete subvolumes whose ID is the largest in the filesystem, then unmount and mount again, then btrfs_init_root_free_objectid on the tree_root will select a subvolid smaller than that one and thus allow reusing it. If we are also using qgroups (and particularly squotas) it is possible to delete the subvol without deleting the qgroup. In that case, we will be able to create a new subvol whose id already has a level 0 qgroup. This will result in re-using that qgroup which would then lead to incorrect accounting. Fixes: 6ed05643ddb1 ("btrfs: create qgroup earlier in snapshot creation") CC: stable@vger.kernel.org # 6.7+ Reviewed-by: Qu Wenruo Signed-off-by: Boris Burkov Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index f93354a96909..fc2a7ea26354 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -468,6 +468,7 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info) } if (!qgroup) { struct btrfs_qgroup *prealloc; + struct btrfs_root *tree_root = fs_info->tree_root; prealloc = kzalloc(sizeof(*prealloc), GFP_KERNEL); if (!prealloc) { @@ -475,6 +476,25 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info) goto out; } qgroup = add_qgroup_rb(fs_info, prealloc, found_key.offset); + /* + * If a qgroup exists for a subvolume ID, it is possible + * that subvolume has been deleted, in which case + * re-using that ID would lead to incorrect accounting. + * + * Ensure that we skip any such subvol ids. + * + * We don't need to lock because this is only called + * during mount before we start doing things like creating + * subvolumes. + */ + if (is_fstree(qgroup->qgroupid) && + qgroup->qgroupid > tree_root->free_objectid) + /* + * Don't need to check against BTRFS_LAST_FREE_OBJECTID, + * as it will get checked on the next call to + * btrfs_get_free_objectid. + */ + tree_root->free_objectid = qgroup->qgroupid + 1; } ret = btrfs_sysfs_add_one_qgroup(fs_info, qgroup); if (ret < 0) -- cgit v1.2.3-59-g8ed1b From 0090d6e1b210551e63cf43958dc7a1ec942cdde9 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 8 May 2024 11:51:07 +0100 Subject: btrfs: zoned: fix use-after-free due to race with dev replace While loading a zone's info during creation of a block group, we can race with a device replace operation and then trigger a use-after-free on the device that was just replaced (source device of the replace operation). This happens because at btrfs_load_zone_info() we extract a device from the chunk map into a local variable and then use the device while not under the protection of the device replace rwsem. So if there's a device replace operation happening when we extract the device and that device is the source of the replace operation, we will trigger a use-after-free if before we finish using the device the replace operation finishes and frees the device. Fix this by enlarging the critical section under the protection of the device replace rwsem so that all uses of the device are done inside the critical section. CC: stable@vger.kernel.org # 6.1.x: 15c12fcc50a1: btrfs: zoned: introduce a zone_info struct in btrfs_load_block_group_zone_info CC: stable@vger.kernel.org # 6.1.x: 09a46725cc84: btrfs: zoned: factor out per-zone logic from btrfs_load_block_group_zone_info CC: stable@vger.kernel.org # 6.1.x: 9e0e3e74dc69: btrfs: zoned: factor out single bg handling from btrfs_load_block_group_zone_info CC: stable@vger.kernel.org # 6.1.x: 87463f7e0250: btrfs: zoned: factor out DUP bg handling from btrfs_load_block_group_zone_info CC: stable@vger.kernel.org # 6.1.x Reviewed-by: Johannes Thumshirn Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/zoned.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 4cba80b34387..459514da1660 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -1290,7 +1290,7 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx, struct btrfs_chunk_map *map) { struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; - struct btrfs_device *device = map->stripes[zone_idx].dev; + struct btrfs_device *device; int dev_replace_is_ongoing = 0; unsigned int nofs_flag; struct blk_zone zone; @@ -1298,7 +1298,11 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx, info->physical = map->stripes[zone_idx].physical; + down_read(&dev_replace->rwsem); + device = map->stripes[zone_idx].dev; + if (!device->bdev) { + up_read(&dev_replace->rwsem); info->alloc_offset = WP_MISSING_DEV; return 0; } @@ -1308,6 +1312,7 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx, __set_bit(zone_idx, active); if (!btrfs_dev_is_sequential(device, info->physical)) { + up_read(&dev_replace->rwsem); info->alloc_offset = WP_CONVENTIONAL; return 0; } @@ -1315,11 +1320,9 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx, /* This zone will be used for allocation, so mark this zone non-empty. */ btrfs_dev_clear_zone_empty(device, info->physical); - down_read(&dev_replace->rwsem); dev_replace_is_ongoing = btrfs_dev_replace_is_ongoing(dev_replace); if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL) btrfs_dev_clear_zone_empty(dev_replace->tgtdev, info->physical); - up_read(&dev_replace->rwsem); /* * The group is mapped to a sequential zone. Get the zone write pointer @@ -1330,6 +1333,7 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx, ret = btrfs_get_dev_zone(device, info->physical, &zone); memalloc_nofs_restore(nofs_flag); if (ret) { + up_read(&dev_replace->rwsem); if (ret != -EIO && ret != -EOPNOTSUPP) return ret; info->alloc_offset = WP_MISSING_DEV; @@ -1341,6 +1345,7 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx, "zoned: unexpected conventional zone %llu on device %s (devid %llu)", zone.start << SECTOR_SHIFT, rcu_str_deref(device->name), device->devid); + up_read(&dev_replace->rwsem); return -EIO; } @@ -1368,6 +1373,8 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx, break; } + up_read(&dev_replace->rwsem); + return 0; } -- cgit v1.2.3-59-g8ed1b From b4e585fffc1cf877112ed231a91f089e85688c2a Mon Sep 17 00:00:00 2001 From: Lu Yao Date: Tue, 7 May 2024 10:34:17 +0800 Subject: btrfs: scrub: initialize ret in scrub_simple_mirror() to fix compilation warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The following error message is displayed: ../fs/btrfs/scrub.c:2152:9: error: ‘ret’ may be used uninitialized in this function [-Werror=maybe-uninitialized]" Compiler version: gcc version: (Debian 10.2.1-6) 10.2.1 20210110 Reviewed-by: Boris Burkov Signed-off-by: Lu Yao Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/scrub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 4b22cfe9a98c..afd6932f5e89 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -2100,7 +2100,7 @@ static int scrub_simple_mirror(struct scrub_ctx *sctx, struct btrfs_fs_info *fs_info = sctx->fs_info; const u64 logical_end = logical_start + logical_length; u64 cur_logical = logical_start; - int ret; + int ret = 0; /* The range must be inside the bg */ ASSERT(logical_start >= bg->start && logical_end <= bg->start + bg->length); -- cgit v1.2.3-59-g8ed1b From dddff821b660e08eb8418cfd3909e61204804534 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 14 May 2024 15:36:35 +0100 Subject: btrfs: fix end of tree detection when searching for data extent ref At lookup_extent_data_ref() we are incorrectly checking if we are at the last slot of the last leaf in the extent tree. We are returning -ENOENT if btrfs_next_leaf() returns a value greater than 1, but btrfs_next_leaf() never returns anything greater than 1: 1) It returns < 0 on error; 2) 0 if there is a next leaf (or a new item was added to the end of the current leaf after releasing the path); 3) 1 if there are no more leaves (and no new items were added to the last leaf after releasing the path). So fix this by checking if the return value is greater than zero instead of being greater than one. Fixes: 1618aa3c2e01 ("btrfs: simplify return variables in lookup_extent_data_ref()") Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 47d48233b592..3774c191e36d 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -477,7 +477,7 @@ again: if (path->slots[0] >= nritems) { ret = btrfs_next_leaf(root, path); if (ret) { - if (ret > 1) + if (ret > 0) return -ENOENT; return ret; } -- cgit v1.2.3-59-g8ed1b From a395726cf823fe8f62f1b8c3829010e5652ce98c Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 15 May 2024 16:59:01 -0500 Subject: cifs: fix data corruption in read after invalidate When invalidating a file as part of breaking a lease, the folios holding the file data are disposed of, and truncate calls ->invalidate_folio() to get rid of them rather than calling ->release_folio(). This means that the netfs_inode::zero_point value didn't get updated in current upstream code to reflect the point after which we can assume that the server will only return zeroes, and future reads will then return blocks of zeroes if the file got extended for any region beyond the old zero point. Fix this by updating zero_point before invalidating the inode in cifs_revalidate_mapping(). Suggested-by: David Howells Fixes: 3ee1a1fc3981 ("cifs: Cut over to using netfslib") Reviewed-by: David Howells Signed-off-by: Steve French --- fs/smb/client/inode.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index 5d4b0fd3a59e..262576573eb5 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -2465,7 +2465,8 @@ int cifs_revalidate_mapping(struct inode *inode) { int rc; - unsigned long *flags = &CIFS_I(inode)->flags; + struct cifsInodeInfo *cifs_inode = CIFS_I(inode); + unsigned long *flags = &cifs_inode->flags; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); /* swapfiles are not supposed to be shared */ @@ -2482,6 +2483,7 @@ cifs_revalidate_mapping(struct inode *inode) if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE) goto skip_invalidate; + cifs_inode->netfs.zero_point = cifs_inode->netfs.remote_i_size; rc = filemap_invalidate_inode(inode, true, 0, LLONG_MAX); if (rc) { cifs_dbg(VFS, "%s: invalidate inode %p failed with rc %d\n", -- cgit v1.2.3-59-g8ed1b From c6a6c9694aadc4c3ab8d89bdd44aed3eab1e43c6 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 10 May 2024 18:22:53 +0300 Subject: ext4: fix error pointer dereference in ext4_mb_load_buddy_gfp() This code calls folio_put() on an error pointer which will lead to a crash. Check for both error pointers and NULL pointers before calling folio_put(). Fixes: 5eea586b47f0 ("ext4: convert bd_buddy_page to bd_buddy_folio") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/eaafa1d9-a61c-4af4-9f97-d3ad72c60200@moroto.mountain Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 648989c125f2..9dda9cd68ab2 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -1717,7 +1717,7 @@ ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group, return 0; err: - if (folio) + if (!IS_ERR_OR_NULL(folio)) folio_put(folio); if (e4b->bd_bitmap_folio) folio_put(e4b->bd_bitmap_folio); -- cgit v1.2.3-59-g8ed1b From c34110e0fdfddc22b7fd606ca81303d20330bacb Mon Sep 17 00:00:00 2001 From: Hongzhen Luo Date: Fri, 17 May 2024 17:56:52 +0800 Subject: erofs: clean up erofs_show_options() Avoid unnecessary #ifdefs and simplify the code a bit. Signed-off-by: Hongzhen Luo Link: https://lore.kernel.org/r/20240517095652.2282972-1-hongzhen@linux.alibaba.com Reviewed-by: Gao Xiang Signed-off-by: Gao Xiang --- fs/erofs/internal.h | 3 --- fs/erofs/super.c | 28 ++++++++-------------------- 2 files changed, 8 insertions(+), 23 deletions(-) (limited to 'fs') diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 00bbb288c08c..c5254dc1f965 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -64,15 +64,12 @@ enum { }; struct erofs_mount_opts { -#ifdef CONFIG_EROFS_FS_ZIP /* current strategy of how to use managed cache */ unsigned char cache_strategy; /* strategy of sync decompression (0 - auto, 1 - force on, 2 - force off) */ unsigned int sync_decompress; - /* threshold for decompression synchronously */ unsigned int max_sync_decompress_pages; -#endif unsigned int mount_opt; }; diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 348ef1660e50..3013bfb2e9ed 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -943,26 +943,14 @@ static int erofs_show_options(struct seq_file *seq, struct dentry *root) struct erofs_sb_info *sbi = EROFS_SB(root->d_sb); struct erofs_mount_opts *opt = &sbi->opt; -#ifdef CONFIG_EROFS_FS_XATTR - if (test_opt(opt, XATTR_USER)) - seq_puts(seq, ",user_xattr"); - else - seq_puts(seq, ",nouser_xattr"); -#endif -#ifdef CONFIG_EROFS_FS_POSIX_ACL - if (test_opt(opt, POSIX_ACL)) - seq_puts(seq, ",acl"); - else - seq_puts(seq, ",noacl"); -#endif -#ifdef CONFIG_EROFS_FS_ZIP - if (opt->cache_strategy == EROFS_ZIP_CACHE_DISABLED) - seq_puts(seq, ",cache_strategy=disabled"); - else if (opt->cache_strategy == EROFS_ZIP_CACHE_READAHEAD) - seq_puts(seq, ",cache_strategy=readahead"); - else if (opt->cache_strategy == EROFS_ZIP_CACHE_READAROUND) - seq_puts(seq, ",cache_strategy=readaround"); -#endif + if (IS_ENABLED(CONFIG_EROFS_FS_XATTR)) + seq_puts(seq, test_opt(opt, XATTR_USER) ? + ",user_xattr" : ",nouser_xattr"); + if (IS_ENABLED(CONFIG_EROFS_FS_POSIX_ACL)) + seq_puts(seq, test_opt(opt, POSIX_ACL) ? ",acl" : ",noacl"); + if (IS_ENABLED(CONFIG_EROFS_FS_ZIP)) + seq_printf(seq, ",cache_strategy=%s", + erofs_param_cache_strategy[opt->cache_strategy].name); if (test_opt(opt, DAX_ALWAYS)) seq_puts(seq, ",dax=always"); if (test_opt(opt, DAX_NEVER)) -- cgit v1.2.3-59-g8ed1b From e09815446d6944fc5590a6e5f15dd51697202441 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 25 Apr 2024 20:58:46 +0100 Subject: erofs: mechanically convert erofs_read_metabuf() to offsets just lift the call of erofs_pos() into the callers; it will collapse in most of them, but that's better done caller-by-caller. Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240425195846.GC1031757@ZenIV Signed-off-by: Gao Xiang --- fs/erofs/data.c | 8 ++++---- fs/erofs/fscache.c | 2 +- fs/erofs/inode.c | 4 ++-- fs/erofs/internal.h | 2 +- fs/erofs/super.c | 2 +- fs/erofs/zdata.c | 2 +- fs/erofs/zmap.c | 6 +++--- 7 files changed, 13 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/erofs/data.c b/fs/erofs/data.c index e1a170e45c70..82a196e02b5c 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -72,10 +72,10 @@ void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb) } void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb, - erofs_blk_t blkaddr, enum erofs_kmap_type type) + erofs_off_t offset, enum erofs_kmap_type type) { erofs_init_metabuf(buf, sb); - return erofs_bread(buf, erofs_pos(sb, blkaddr), type); + return erofs_bread(buf, offset, type); } static int erofs_map_blocks_flatmode(struct inode *inode, @@ -152,7 +152,7 @@ int erofs_map_blocks(struct inode *inode, struct erofs_map_blocks *map) pos = ALIGN(erofs_iloc(inode) + vi->inode_isize + vi->xattr_isize, unit) + unit * chunknr; - kaddr = erofs_read_metabuf(&buf, sb, erofs_blknr(sb, pos), EROFS_KMAP); + kaddr = erofs_read_metabuf(&buf, sb, erofs_pos(sb, erofs_blknr(sb, pos)), EROFS_KMAP); if (IS_ERR(kaddr)) { err = PTR_ERR(kaddr); goto out; @@ -295,7 +295,7 @@ static int erofs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, iomap->type = IOMAP_INLINE; ptr = erofs_read_metabuf(&buf, sb, - erofs_blknr(sb, mdev.m_pa), EROFS_KMAP); + erofs_pos(sb, erofs_blknr(sb, mdev.m_pa)), EROFS_KMAP); if (IS_ERR(ptr)) return PTR_ERR(ptr); iomap->inline_data = ptr + erofs_blkoff(sb, mdev.m_pa); diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c index 62da538d91cb..ac618b3484f1 100644 --- a/fs/erofs/fscache.c +++ b/fs/erofs/fscache.c @@ -282,7 +282,7 @@ static int erofs_fscache_data_read_slice(struct erofs_fscache_rq *req) blknr = erofs_blknr(sb, map.m_pa); size = map.m_llen; - src = erofs_read_metabuf(&buf, sb, blknr, EROFS_KMAP); + src = erofs_read_metabuf(&buf, sb, erofs_pos(sb, blknr), EROFS_KMAP); if (IS_ERR(src)) return PTR_ERR(src); diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c index 0eb0e6f933c3..5f6439a63af7 100644 --- a/fs/erofs/inode.c +++ b/fs/erofs/inode.c @@ -26,7 +26,7 @@ static void *erofs_read_inode(struct erofs_buf *buf, blkaddr = erofs_blknr(sb, inode_loc); *ofs = erofs_blkoff(sb, inode_loc); - kaddr = erofs_read_metabuf(buf, sb, blkaddr, EROFS_KMAP); + kaddr = erofs_read_metabuf(buf, sb, erofs_pos(sb, blkaddr), EROFS_KMAP); if (IS_ERR(kaddr)) { erofs_err(sb, "failed to get inode (nid: %llu) page, err %ld", vi->nid, PTR_ERR(kaddr)); @@ -66,7 +66,7 @@ static void *erofs_read_inode(struct erofs_buf *buf, goto err_out; } memcpy(copied, dic, gotten); - kaddr = erofs_read_metabuf(buf, sb, blkaddr + 1, + kaddr = erofs_read_metabuf(buf, sb, erofs_pos(sb, blkaddr + 1), EROFS_KMAP); if (IS_ERR(kaddr)) { erofs_err(sb, "failed to get inode payload block (nid: %llu), err %ld", diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index c5254dc1f965..0c1b44ac9524 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -403,7 +403,7 @@ void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset, enum erofs_kmap_type type); void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb); void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb, - erofs_blk_t blkaddr, enum erofs_kmap_type type); + erofs_off_t offset, enum erofs_kmap_type type); int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *dev); int erofs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len); diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 3013bfb2e9ed..95b05a9490a2 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -180,7 +180,7 @@ static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb, struct file *bdev_file; void *ptr; - ptr = erofs_read_metabuf(buf, sb, erofs_blknr(sb, *pos), EROFS_KMAP); + ptr = erofs_read_metabuf(buf, sb, erofs_pos(sb, erofs_blknr(sb, *pos)), EROFS_KMAP); if (IS_ERR(ptr)) return PTR_ERR(ptr); dis = ptr + erofs_blkoff(sb, *pos); diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 283c9c3a611d..d417e189f1a0 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -868,7 +868,7 @@ static int z_erofs_pcluster_begin(struct z_erofs_decompress_frontend *fe) } else { void *mptr; - mptr = erofs_read_metabuf(&map->buf, sb, blknr, EROFS_NO_KMAP); + mptr = erofs_read_metabuf(&map->buf, sb, erofs_pos(sb, blknr), EROFS_NO_KMAP); if (IS_ERR(mptr)) { ret = PTR_ERR(mptr); erofs_err(sb, "failed to get inline data %d", ret); diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index 0a2454d8bcc1..222fc66366e0 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -34,7 +34,7 @@ static int z_erofs_load_full_lcluster(struct z_erofs_maprecorder *m, unsigned int advise; m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb, - erofs_blknr(inode->i_sb, pos), EROFS_KMAP); + erofs_pos(inode->i_sb, erofs_blknr(inode->i_sb, pos)), EROFS_KMAP); if (IS_ERR(m->kaddr)) return PTR_ERR(m->kaddr); @@ -256,7 +256,7 @@ static int z_erofs_load_compact_lcluster(struct z_erofs_maprecorder *m, out: pos += lcn * (1 << amortizedshift); m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb, - erofs_blknr(inode->i_sb, pos), EROFS_KMAP); + erofs_pos(inode->i_sb, erofs_blknr(inode->i_sb, pos)), EROFS_KMAP); if (IS_ERR(m->kaddr)) return PTR_ERR(m->kaddr); return unpack_compacted_index(m, amortizedshift, pos, lookahead); @@ -590,7 +590,7 @@ static int z_erofs_fill_inode_lazy(struct inode *inode) goto out_unlock; pos = ALIGN(erofs_iloc(inode) + vi->inode_isize + vi->xattr_isize, 8); - kaddr = erofs_read_metabuf(&buf, sb, erofs_blknr(sb, pos), EROFS_KMAP); + kaddr = erofs_read_metabuf(&buf, sb, erofs_pos(sb, erofs_blknr(sb, pos)), EROFS_KMAP); if (IS_ERR(kaddr)) { err = PTR_ERR(kaddr); goto out_unlock; -- cgit v1.2.3-59-g8ed1b From 076d965eb812f2ad88daf693d745ea1f28bf8f80 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 25 Apr 2024 20:59:15 +0100 Subject: erofs: don't align offset for erofs_read_metabuf() (simple cases) Most of the callers of erofs_read_metabuf() have the following form: block = erofs_blknr(sb, offset); off = erofs_blkoff(sb, offset); p = erofs_read_metabuf(...., erofs_pos(sb, block), ...); if (IS_ERR(p)) return PTR_ERR(p); q = p + off; // no further uses of p, block or off. The value passed to erofs_read_metabuf() is offset rounded down to block size, i.e. offset - off. Passing offset as-is would increase the return value by off in case of success and keep the return value unchanged in in case of error. In other words, the same could be achieved by q = erofs_read_metabuf(...., offset, ...); if (IS_ERR(q)) return PTR_ERR(q); This commit convert these simple cases. Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240425195915.GD1031757@ZenIV Signed-off-by: Gao Xiang --- fs/erofs/data.c | 11 +++++------ fs/erofs/fscache.c | 12 +++--------- fs/erofs/super.c | 8 +++----- fs/erofs/zmap.c | 8 +++----- 4 files changed, 14 insertions(+), 25 deletions(-) (limited to 'fs') diff --git a/fs/erofs/data.c b/fs/erofs/data.c index 82a196e02b5c..604d0bc82a0e 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -152,7 +152,7 @@ int erofs_map_blocks(struct inode *inode, struct erofs_map_blocks *map) pos = ALIGN(erofs_iloc(inode) + vi->inode_isize + vi->xattr_isize, unit) + unit * chunknr; - kaddr = erofs_read_metabuf(&buf, sb, erofs_pos(sb, erofs_blknr(sb, pos)), EROFS_KMAP); + kaddr = erofs_read_metabuf(&buf, sb, pos, EROFS_KMAP); if (IS_ERR(kaddr)) { err = PTR_ERR(kaddr); goto out; @@ -163,7 +163,7 @@ int erofs_map_blocks(struct inode *inode, struct erofs_map_blocks *map) /* handle block map */ if (!(vi->chunkformat & EROFS_CHUNK_FORMAT_INDEXES)) { - __le32 *blkaddr = kaddr + erofs_blkoff(sb, pos); + __le32 *blkaddr = kaddr; if (le32_to_cpu(*blkaddr) == EROFS_NULL_ADDR) { map->m_flags = 0; @@ -174,7 +174,7 @@ int erofs_map_blocks(struct inode *inode, struct erofs_map_blocks *map) goto out_unlock; } /* parse chunk indexes */ - idx = kaddr + erofs_blkoff(sb, pos); + idx = kaddr; switch (le32_to_cpu(idx->blkaddr)) { case EROFS_NULL_ADDR: map->m_flags = 0; @@ -294,11 +294,10 @@ static int erofs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, struct erofs_buf buf = __EROFS_BUF_INITIALIZER; iomap->type = IOMAP_INLINE; - ptr = erofs_read_metabuf(&buf, sb, - erofs_pos(sb, erofs_blknr(sb, mdev.m_pa)), EROFS_KMAP); + ptr = erofs_read_metabuf(&buf, sb, mdev.m_pa, EROFS_KMAP); if (IS_ERR(ptr)) return PTR_ERR(ptr); - iomap->inline_data = ptr + erofs_blkoff(sb, mdev.m_pa); + iomap->inline_data = ptr; iomap->private = buf.base; } else { iomap->type = IOMAP_MAPPED; diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c index ac618b3484f1..fda16eedafb5 100644 --- a/fs/erofs/fscache.c +++ b/fs/erofs/fscache.c @@ -273,21 +273,15 @@ static int erofs_fscache_data_read_slice(struct erofs_fscache_rq *req) if (map.m_flags & EROFS_MAP_META) { struct erofs_buf buf = __EROFS_BUF_INITIALIZER; struct iov_iter iter; - erofs_blk_t blknr; - size_t offset, size; + size_t size = map.m_llen; void *src; - /* For tail packing layout, the offset may be non-zero. */ - offset = erofs_blkoff(sb, map.m_pa); - blknr = erofs_blknr(sb, map.m_pa); - size = map.m_llen; - - src = erofs_read_metabuf(&buf, sb, erofs_pos(sb, blknr), EROFS_KMAP); + src = erofs_read_metabuf(&buf, sb, map.m_pa, EROFS_KMAP); if (IS_ERR(src)) return PTR_ERR(src); iov_iter_xarray(&iter, ITER_DEST, &mapping->i_pages, pos, PAGE_SIZE); - if (copy_to_iter(src + offset, size, &iter) != size) { + if (copy_to_iter(src, size, &iter) != size) { erofs_put_metabuf(&buf); return -EFAULT; } diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 95b05a9490a2..c93bd24d2771 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -178,12 +178,10 @@ static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb, struct erofs_fscache *fscache; struct erofs_deviceslot *dis; struct file *bdev_file; - void *ptr; - ptr = erofs_read_metabuf(buf, sb, erofs_pos(sb, erofs_blknr(sb, *pos)), EROFS_KMAP); - if (IS_ERR(ptr)) - return PTR_ERR(ptr); - dis = ptr + erofs_blkoff(sb, *pos); + dis = erofs_read_metabuf(buf, sb, *pos, EROFS_KMAP); + if (IS_ERR(dis)) + return PTR_ERR(dis); if (!sbi->devs->flatdev && !dif->path) { if (!dis->tag[0]) { diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index 222fc66366e0..8979def17b58 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -570,7 +570,6 @@ static int z_erofs_fill_inode_lazy(struct inode *inode) int err, headnr; erofs_off_t pos; struct erofs_buf buf = __EROFS_BUF_INITIALIZER; - void *kaddr; struct z_erofs_map_header *h; if (test_bit(EROFS_I_Z_INITED_BIT, &vi->flags)) { @@ -590,13 +589,12 @@ static int z_erofs_fill_inode_lazy(struct inode *inode) goto out_unlock; pos = ALIGN(erofs_iloc(inode) + vi->inode_isize + vi->xattr_isize, 8); - kaddr = erofs_read_metabuf(&buf, sb, erofs_pos(sb, erofs_blknr(sb, pos)), EROFS_KMAP); - if (IS_ERR(kaddr)) { - err = PTR_ERR(kaddr); + h = erofs_read_metabuf(&buf, sb, pos, EROFS_KMAP); + if (IS_ERR(h)) { + err = PTR_ERR(h); goto out_unlock; } - h = kaddr + erofs_blkoff(sb, pos); /* * if the highest bit of the 8-byte map header is set, the whole file * is stored in the packed inode. The rest bits keeps z_fragmentoff. -- cgit v1.2.3-59-g8ed1b From 4afe6b8d21e5ff644fedd7db5673fe5a48b177b7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 25 Apr 2024 20:59:44 +0100 Subject: erofs: don't round offset down for erofs_read_metabuf() There's only one place where struct z_erofs_maprecorder ->kaddr is used not in the same function that has assigned it - the value read in unpack_compacted_index() gets calculated in z_erofs_load_compact_lcluster(). With minor massage we can switch to storing it with offset in block already added. Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240425195944.GE1031757@ZenIV Signed-off-by: Gao Xiang --- fs/erofs/zmap.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index 8979def17b58..9b248ee5fef2 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -34,13 +34,13 @@ static int z_erofs_load_full_lcluster(struct z_erofs_maprecorder *m, unsigned int advise; m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb, - erofs_pos(inode->i_sb, erofs_blknr(inode->i_sb, pos)), EROFS_KMAP); + pos, EROFS_KMAP); if (IS_ERR(m->kaddr)) return PTR_ERR(m->kaddr); m->nextpackoff = pos + sizeof(struct z_erofs_lcluster_index); m->lcn = lcn; - di = m->kaddr + erofs_blkoff(inode->i_sb, pos); + di = m->kaddr; advise = le16_to_cpu(di->di_advise); m->type = advise & Z_EROFS_LI_LCLUSTER_TYPE_MASK; @@ -109,7 +109,7 @@ static int unpack_compacted_index(struct z_erofs_maprecorder *m, { struct erofs_inode *const vi = EROFS_I(m->inode); const unsigned int lclusterbits = vi->z_logical_clusterbits; - unsigned int vcnt, base, lo, lobits, encodebits, nblk, eofs; + unsigned int vcnt, lo, lobits, encodebits, nblk, bytes; int i; u8 *in, type; bool big_pcluster; @@ -127,11 +127,11 @@ static int unpack_compacted_index(struct z_erofs_maprecorder *m, big_pcluster = vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_1; lobits = max(lclusterbits, ilog2(Z_EROFS_LI_D0_CBLKCNT) + 1U); encodebits = ((vcnt << amortizedshift) - sizeof(__le32)) * 8 / vcnt; - eofs = erofs_blkoff(m->inode->i_sb, pos); - base = round_down(eofs, vcnt << amortizedshift); - in = m->kaddr + base; + bytes = pos & ((vcnt << amortizedshift) - 1); - i = (eofs - base) >> amortizedshift; + in = m->kaddr - bytes; + + i = bytes >> amortizedshift; lo = decode_compactedbits(lobits, in, encodebits * i, &type); m->type = type; @@ -256,7 +256,7 @@ static int z_erofs_load_compact_lcluster(struct z_erofs_maprecorder *m, out: pos += lcn * (1 << amortizedshift); m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb, - erofs_pos(inode->i_sb, erofs_blknr(inode->i_sb, pos)), EROFS_KMAP); + pos, EROFS_KMAP); if (IS_ERR(m->kaddr)) return PTR_ERR(m->kaddr); return unpack_compacted_index(m, amortizedshift, pos, lookahead); -- cgit v1.2.3-59-g8ed1b From 5587a8172eb6040e388c3fc9fa6553b99510da9e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 25 Apr 2024 21:00:17 +0100 Subject: z_erofs_pcluster_begin(): don't bother with rounding position down ... and be more idiomatic when calculating ->pageofs_in. Signed-off-by: Al Viro Link: https://lore.kernel.org/r/20240425200017.GF1031757@ZenIV [ Gao Xiang: don't use `offset_in_page(mptr)` due to EROFS_NO_KMAP. ] Signed-off-by: Gao Xiang --- fs/erofs/zdata.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index d417e189f1a0..d6fe002a4a71 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -868,7 +868,7 @@ static int z_erofs_pcluster_begin(struct z_erofs_decompress_frontend *fe) } else { void *mptr; - mptr = erofs_read_metabuf(&map->buf, sb, erofs_pos(sb, blknr), EROFS_NO_KMAP); + mptr = erofs_read_metabuf(&map->buf, sb, map->m_pa, EROFS_NO_KMAP); if (IS_ERR(mptr)) { ret = PTR_ERR(mptr); erofs_err(sb, "failed to get inline data %d", ret); -- cgit v1.2.3-59-g8ed1b From db3e24a02e29b507c24c0adb4d22914c65dab763 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Sat, 11 May 2024 09:29:42 +0900 Subject: nilfs2: make block erasure safe in nilfs_finish_roll_forward() The implementation of writing a zero-fill block in nilfs_finish_roll_forward() is not safe. The buffer is being cleared without acquiring a lock or setting the uptodate flag, so theoretically, between the time the buffer's data is cleared and the time it is written back to the block device using sync_dirty_buffer(), that zero data can be undone by concurrent block device reads. Since this buffer points to a location that has been read from disk once, the uptodate flag will most likely remain, but since it was obtained with __getblk(), that is not guaranteed. In other words, this is exceptional, and this function itself is not normally called (only once when mounting after a specific pattern of unclean shutdown), so it is highly unlikely that this will actually cause a problem. Anyway, eliminate this potential race issue by protecting the clearing of buffer data with a buffer lock and setting the buffer's uptodate flag within the protected section. Link: https://lkml.kernel.org/r/20240511002942.9608-1-konishi.ryusuke@gmail.com Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton --- fs/nilfs2/recovery.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs') diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c index 020f304c600e..b638dc06df2f 100644 --- a/fs/nilfs2/recovery.c +++ b/fs/nilfs2/recovery.c @@ -702,8 +702,12 @@ static void nilfs_finish_roll_forward(struct the_nilfs *nilfs, if (WARN_ON(!bh)) return; /* should never happen */ + lock_buffer(bh); memset(bh->b_data, 0, bh->b_size); + set_buffer_uptodate(bh); set_buffer_dirty(bh); + unlock_buffer(bh); + err = sync_dirty_buffer(bh); if (unlikely(err)) nilfs_warn(nilfs->ns_sb, -- cgit v1.2.3-59-g8ed1b From 427ba555036e85ed0805f45ca3896b68602e0a80 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 12:23:51 -0400 Subject: bcachefs: Fix rcu splat in check_fix_ptrs() Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index e28d28ac2a13..02bb02b70981 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -479,9 +479,8 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, percpu_down_read(&c->mark_lock); - rcu_read_lock(); bkey_for_each_ptr_decode(k.k, ptrs_c, p, entry_c) { - struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_tryget(c, p.ptr.dev); if (!ca) { if (fsck_err(c, ptr_to_invalid_device, "pointer to missing device %u\n" @@ -558,7 +557,7 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, do_update = true; if (data_type != BCH_DATA_btree && p.ptr.gen != g->gen) - continue; + goto next; if (fsck_err_on(bucket_data_type_mismatch(g->data_type, data_type), c, ptr_bucket_data_type_mismatch, @@ -601,8 +600,9 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, bch2_bkey_val_to_text(&buf, c, k), buf.buf))) do_update = true; } +next: + bch2_dev_put(ca); } - rcu_read_unlock(); if (do_update) { if (flags & BTREE_TRIGGER_is_root) { @@ -638,9 +638,10 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, } else { struct bkey_ptrs ptrs; union bch_extent_entry *entry; + + rcu_read_lock(); restart_drop_ptrs: ptrs = bch2_bkey_ptrs(bkey_i_to_s(new)); - rcu_read_lock(); bkey_for_each_ptr_decode(bkey_i_to_s(new).k, ptrs, p, entry) { struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr); -- cgit v1.2.3-59-g8ed1b From 54429c902af909448371f40af59d4437a57949c7 Mon Sep 17 00:00:00 2001 From: Youling Tang Date: Fri, 10 May 2024 14:10:58 +0800 Subject: bcachefs: set FMODE_CAN_ODIRECT instead of a dummy direct_IO method Since commit a2ad63daa88b ("VFS: add FMODE_CAN_ODIRECT file flag") file systems can just set the FMODE_CAN_ODIRECT flag at open time instead of wiring up a dummy direct_IO method to indicate support for direct I/O. Do that for bcachefs so that noop_direct_IO can eventually be removed. Similar to commit b29434999371 ("xfs: set FMODE_CAN_ODIRECT instead of a dummy direct_IO method"). Signed-off-by: Youling Tang Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index fd851f10d11c..9b41c6e9deef 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -1145,6 +1145,8 @@ static int bch2_open(struct inode *vinode, struct file *file) return ret; } + file->f_mode |= FMODE_CAN_ODIRECT; + return generic_file_open(vinode, file); } @@ -1237,7 +1239,6 @@ static const struct address_space_operations bch_address_space_operations = { .write_end = bch2_write_end, .invalidate_folio = bch2_invalidate_folio, .release_folio = bch2_release_folio, - .direct_IO = noop_direct_IO, #ifdef CONFIG_MIGRATION .migrate_folio = filemap_migrate_folio, #endif -- cgit v1.2.3-59-g8ed1b From 9667214b30ef19c1cbead7893257619bca4d0d4d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 17:07:32 -0400 Subject: bcachefs: Fix ref in trans_mark_dev_sbs() error path Reported-by: syzbot+5c7f715a7107a608a544@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/buckets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 02bb02b70981..b469586517a8 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -1465,7 +1465,7 @@ int bch2_trans_mark_dev_sbs_flags(struct bch_fs *c, for_each_online_member(c, ca) { int ret = bch2_trans_mark_dev_sb(c, ca, flags); if (ret) { - bch2_dev_put(ca); + percpu_ref_put(&ca->io_ref); return ret; } } -- cgit v1.2.3-59-g8ed1b From f108ddd467b5cad84a6d93cdd0a251ba47ceaccb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 17:08:15 -0400 Subject: bcachefs: Fix shift overflow in btree_lost_data() Reported-by: syzbot+29f65db1a5fe427b5c56@syzkaller.appspotmail.com Fixes: 55936afe1107 ("bcachefs: Flag btrees with missing data") Signed-off-by: Kent Overstreet --- fs/bcachefs/bcachefs_format.h | 6 ++++++ fs/bcachefs/recovery.c | 3 +++ 2 files changed, 9 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 1bebba881d89..d801e19cb489 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -1555,6 +1555,12 @@ enum btree_id { BTREE_ID_NR }; +/* + * Maximum number of btrees that we will _ever_ have under the current scheme, + * where we refer to them with bitfields + */ +#define BTREE_ID_NR_MAX 64 + static inline bool btree_id_is_alloc(enum btree_id id) { switch (id) { diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 1266916ac03f..63f990f5c63a 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -35,6 +35,9 @@ void bch2_btree_lost_data(struct bch_fs *c, enum btree_id btree) { + if (btree >= BTREE_ID_NR_MAX) + return; + u64 b = BIT_ULL(btree); if (!(c->sb.btrees_lost_data & b)) { -- cgit v1.2.3-59-g8ed1b From bcfbaea8e547881cd2cab6e05c3cd75449128590 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 17:28:44 -0400 Subject: bcachefs: Fix shift overflows in replicas.c We can't disallow unknown data_types in verify() - we have to preserve them unchanged for backwards compat; that means we have to add a few more guards. Reported-by: syzbot+249018ea545364f78d04@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/replicas.c | 29 +++++++++++++++++++++-------- 1 file changed, 21 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c index bd1d5d085e23..d45b10bc531c 100644 --- a/fs/bcachefs/replicas.c +++ b/fs/bcachefs/replicas.c @@ -524,13 +524,16 @@ int bch2_replicas_gc_start(struct bch_fs *c, unsigned typemask) c->replicas_gc.nr = 0; c->replicas_gc.entry_size = 0; - for_each_cpu_replicas_entry(&c->replicas, e) - if (!((1 << e->data_type) & typemask)) { + for_each_cpu_replicas_entry(&c->replicas, e) { + /* Preserve unknown data types */ + if (e->data_type >= BCH_DATA_NR || + !((1 << e->data_type) & typemask)) { c->replicas_gc.nr++; c->replicas_gc.entry_size = max_t(unsigned, c->replicas_gc.entry_size, replicas_entry_bytes(e)); } + } c->replicas_gc.entries = kcalloc(c->replicas_gc.nr, c->replicas_gc.entry_size, @@ -542,7 +545,8 @@ int bch2_replicas_gc_start(struct bch_fs *c, unsigned typemask) } for_each_cpu_replicas_entry(&c->replicas, e) - if (!((1 << e->data_type) & typemask)) + if (e->data_type >= BCH_DATA_NR || + !((1 << e->data_type) & typemask)) memcpy(cpu_replicas_entry(&c->replicas_gc, i++), e, c->replicas_gc.entry_size); @@ -998,7 +1002,7 @@ unsigned bch2_sb_dev_has_data(struct bch_sb *sb, unsigned dev) { struct bch_sb_field_replicas *replicas; struct bch_sb_field_replicas_v0 *replicas_v0; - unsigned i, data_has = 0; + unsigned data_has = 0; replicas = bch2_sb_field_get(sb, replicas); replicas_v0 = bch2_sb_field_get(sb, replicas_v0); @@ -1006,17 +1010,26 @@ unsigned bch2_sb_dev_has_data(struct bch_sb *sb, unsigned dev) if (replicas) { struct bch_replicas_entry_v1 *r; - for_each_replicas_entry(replicas, r) - for (i = 0; i < r->nr_devs; i++) + for_each_replicas_entry(replicas, r) { + if (r->data_type >= sizeof(data_has) * 8) + continue; + + for (unsigned i = 0; i < r->nr_devs; i++) if (r->devs[i] == dev) data_has |= 1 << r->data_type; + } + } else if (replicas_v0) { struct bch_replicas_entry_v0 *r; - for_each_replicas_entry_v0(replicas_v0, r) - for (i = 0; i < r->nr_devs; i++) + for_each_replicas_entry_v0(replicas_v0, r) { + if (r->data_type >= sizeof(data_has) * 8) + continue; + + for (unsigned i = 0; i < r->nr_devs; i++) if (r->devs[i] == dev) data_has |= 1 << r->data_type; + } } -- cgit v1.2.3-59-g8ed1b From 25989f4a9bf5f9bee7ba541de68f34cf406dff38 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 17:52:12 -0400 Subject: bcachefs: Improve bch2_assert_pos_locked() Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 5bf98cb8b15d..d3bcb4e4e230 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -332,6 +332,8 @@ out: void bch2_assert_pos_locked(struct btree_trans *trans, enum btree_id id, struct bpos pos, bool key_cache) { + bch2_trans_verify_not_unlocked(trans); + struct btree_path *path; struct trans_for_each_path_inorder_iter iter; struct printbuf buf = PRINTBUF; -- cgit v1.2.3-59-g8ed1b From 6ce26ad376bdc65c2bc6d4ef7b8a939bf1bd19dc Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 17:55:02 -0400 Subject: bcachefs: Fix missing parens in drop_locks_do() Reported-by: syzbot+95db43b0a06f157ee865@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index eab2a25bdc7a..798eb1c47966 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -838,7 +838,7 @@ __bch2_btree_iter_peek_and_restart(struct btree_trans *trans, #define drop_locks_do(_trans, _do) \ ({ \ bch2_trans_unlock(_trans); \ - _do ?: bch2_trans_relock(_trans); \ + (_do) ?: bch2_trans_relock(_trans); \ }) #define allocate_dropping_locks_errcode(_trans, _do) \ -- cgit v1.2.3-59-g8ed1b From 6d48e61364aedd774d688f28b5a6c9b5f7f0018c Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 19:16:22 -0400 Subject: bcachefs: Add missing guard in bch2_snapshot_has_children() We additionally need to be going inconsistent if passed an invalid snapshot ID; that patch will need more thorough testing. Reported-by: syzbot+1c9fca23fe478633b305@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/snapshot.h | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index bd5d74269d15..ab13d8f4b41e 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -176,12 +176,9 @@ static inline bool bch2_snapshot_is_ancestor(struct bch_fs *c, u32 id, u32 ances static inline bool bch2_snapshot_has_children(struct bch_fs *c, u32 id) { - const struct snapshot_t *t; - bool ret; - rcu_read_lock(); - t = snapshot_t(c, id); - ret = (t->children[0]|t->children[1]) != 0; + const struct snapshot_t *t = snapshot_t(c, id); + bool ret = t && (t->children[0]|t->children[1]) != 0; rcu_read_unlock(); return ret; -- cgit v1.2.3-59-g8ed1b From c06a8b75679d73473c0c382bcc676ee4b7657bee Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 19:58:44 -0400 Subject: bcachefs: Fix bch2_alloc_ciphers() Don't put error pointers in bch_fs, that's gross. This fixes (?) the check in bch2_checksum_type_valid() - depending on our error paths, or depending on what our error paths are doing it at least makes the code saner. Reported-by: syzbot+2e3cb81b5d1fe18a374b@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/checksum.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index 85198f391e9c..fc9a2b2a916f 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -648,26 +648,26 @@ err: static int bch2_alloc_ciphers(struct bch_fs *c) { - int ret; - - if (!c->chacha20) - c->chacha20 = crypto_alloc_sync_skcipher("chacha20", 0, 0); - ret = PTR_ERR_OR_ZERO(c->chacha20); + if (c->chacha20) + return 0; + struct crypto_sync_skcipher *chacha20 = crypto_alloc_sync_skcipher("chacha20", 0, 0); + int ret = PTR_ERR_OR_ZERO(chacha20); if (ret) { bch_err(c, "error requesting chacha20 module: %s", bch2_err_str(ret)); return ret; } - if (!c->poly1305) - c->poly1305 = crypto_alloc_shash("poly1305", 0, 0); - ret = PTR_ERR_OR_ZERO(c->poly1305); - + struct crypto_shash *poly1305 = crypto_alloc_shash("poly1305", 0, 0); + ret = PTR_ERR_OR_ZERO(poly1305); if (ret) { bch_err(c, "error requesting poly1305 module: %s", bch2_err_str(ret)); + crypto_free_sync_skcipher(chacha20); return ret; } + c->chacha20 = chacha20; + c->poly1305 = poly1305; return 0; } @@ -762,11 +762,11 @@ err: void bch2_fs_encryption_exit(struct bch_fs *c) { - if (!IS_ERR_OR_NULL(c->poly1305)) + if (c->poly1305) crypto_free_shash(c->poly1305); - if (!IS_ERR_OR_NULL(c->chacha20)) + if (c->chacha20) crypto_free_sync_skcipher(c->chacha20); - if (!IS_ERR_OR_NULL(c->sha256)) + if (c->sha256) crypto_free_shash(c->sha256); } @@ -779,6 +779,7 @@ int bch2_fs_encryption_init(struct bch_fs *c) c->sha256 = crypto_alloc_shash("sha256", 0, 0); ret = PTR_ERR_OR_ZERO(c->sha256); if (ret) { + c->sha256 = NULL; bch_err(c, "error requesting sha256 module: %s", bch2_err_str(ret)); goto out; } -- cgit v1.2.3-59-g8ed1b From 6b74fdcc8e25323cd256a573186f9c6c6c2410c0 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 19 May 2024 20:09:35 -0400 Subject: bcachefs: bch2_checksum() returns 0 for unknown checksum type This fixes missing guards on trying to calculate a checksum with an invalid/unknown checksum type; moving the guards up to e.g. btree_io.c might be "more correct", but doesn't buy us anything - an unknown checksum type will always be flagged as at least a checksum error so we aren't losing any safety doing it this way and it makes it less likely to accidentally pop an assert we don't want. Reported-by: syzbot+e951ad5349f3a34a715a@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/checksum.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index fc9a2b2a916f..4b3d6f0137f6 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -233,7 +233,7 @@ struct bch_csum bch2_checksum(struct bch_fs *c, unsigned type, return ret; } default: - BUG(); + return (struct bch_csum) {}; } } @@ -307,7 +307,7 @@ static struct bch_csum __bch2_checksum_bio(struct bch_fs *c, unsigned type, return ret; } default: - BUG(); + return (struct bch_csum) {}; } } -- cgit v1.2.3-59-g8ed1b From 765b8cb8acabdff9261a11b80c87496717f67824 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 20 May 2024 01:11:20 -0400 Subject: bcachefs: Check for subvolues with bogus snapshot/inode fields This fixes an assertion pop in btree_iter.c that checks for forgetting to pass a snapshot ID when iterating over snapshots btrees. Reported-by: syzbot+0dfe05235e38653e2aee@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/sb-errors_types.h | 4 +++- fs/bcachefs/subvolume.c | 9 +++++++++ 2 files changed, 12 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/sb-errors_types.h b/fs/bcachefs/sb-errors_types.h index 87324747351a..666599d3fb9d 100644 --- a/fs/bcachefs/sb-errors_types.h +++ b/fs/bcachefs/sb-errors_types.h @@ -273,7 +273,9 @@ x(dup_backpointer_to_bad_csum_extent, 265) \ x(btree_bitmap_not_marked, 266) \ x(sb_clean_entry_overrun, 267) \ - x(btree_ptr_v2_written_0, 268) + x(btree_ptr_v2_written_0, 268) \ + x(subvol_snapshot_bad, 269) \ + x(subvol_inode_bad, 270) enum bch_sb_error_id { #define x(t, n) BCH_FSCK_ERR_##t = n, diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index 132213761ef6..dfc9cf305756 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -210,12 +210,21 @@ int bch2_check_subvol_children(struct bch_fs *c) int bch2_subvolume_invalid(struct bch_fs *c, struct bkey_s_c k, enum bch_validate_flags flags, struct printbuf *err) { + struct bkey_s_c_subvolume subvol = bkey_s_c_to_subvolume(k); int ret = 0; bkey_fsck_err_on(bkey_lt(k.k->p, SUBVOL_POS_MIN) || bkey_gt(k.k->p, SUBVOL_POS_MAX), c, err, subvol_pos_bad, "invalid pos"); + + bkey_fsck_err_on(!subvol.v->snapshot, c, err, + subvol_snapshot_bad, + "invalid snapshot"); + + bkey_fsck_err_on(!subvol.v->inode, c, err, + subvol_inode_bad, + "invalid inode"); fsck_err: return ret; } -- cgit v1.2.3-59-g8ed1b From 9c75576e3bbf0153e92ab51ff161962d8632c290 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Wed, 20 Mar 2024 17:11:19 -0400 Subject: pNFS/filelayout: Remove the whole file layout requirement Layout segments have been supported in pNFS for years, so remove the requirement that the server always sends whole file layouts. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/filelayout/filelayout.c | 8 -------- 1 file changed, 8 deletions(-) (limited to 'fs') diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index ce8f8934bca5..3fb18b16a5b4 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -605,14 +605,6 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, dprintk("--> %s\n", __func__); - /* FIXME: remove this check when layout segment support is added */ - if (lgr->range.offset != 0 || - lgr->range.length != NFS4_MAX_UINT64) { - dprintk("%s Only whole file layouts supported. Use MDS i/o\n", - __func__); - goto out; - } - if (fl->pattern_offset > lgr->range.offset) { dprintk("%s pattern_offset %lld too large\n", __func__, fl->pattern_offset); -- cgit v1.2.3-59-g8ed1b From 464b424fb09b894f792a494f10539c190db503cf Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Wed, 20 Mar 2024 17:11:20 -0400 Subject: pNFS/filelayout: Specify the layout segment range in LAYOUTGET Move from only requesting full file layout segments to requesting layout segments that match our I/O size. This means the server is still free to return a full file layout if it wants, but partial layouts will no longer cause an error. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/filelayout/filelayout.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index 3fb18b16a5b4..cc2ed4b5a4fd 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -871,8 +871,8 @@ filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), - 0, - NFS4_MAX_UINT64, + req_offset(req), + req->wb_bytes, IOMODE_READ, false, GFP_KERNEL); @@ -895,8 +895,8 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), - 0, - NFS4_MAX_UINT64, + req_offset(req), + req->wb_bytes, IOMODE_RW, false, GFP_NOFS); -- cgit v1.2.3-59-g8ed1b From 7c6c5249f061b64fc6b5b90bc147169a048691bf Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 25 Mar 2024 16:36:05 +1100 Subject: NFS: add atomic_open for NFSv3 to handle O_TRUNC correctly. With two clients, each with NFSv3 mounts of the same directory, the sequence: client1 client2 ls -l afile echo hello there > afile echo HELLO > afile cat afile will show HELLO there because the O_TRUNC requested in the final 'echo' doesn't take effect. This is because the "Negative dentry, just create a file" section in lookup_open() assumes that the file *does* get created since the dentry was negative, so it sets FMODE_CREATED, and this causes do_open() to clear O_TRUNC and so the file doesn't get truncated. Even mounting with -o lookupcache=none does not help as nfs_neg_need_reval() always returns false if LOOKUP_CREATE is set. This patch fixes the problem by providing an atomic_open inode operation for NFSv3 (and v2). The code is largely the code from the branch in lookup_open() when atomic_open is not provided. The significant change is that the O_TRUNC flag is passed a new nfs_do_create() which add 'trunc' handling to nfs_create(). With this change we also optimise away an unnecessary LOOKUP before the file is created. Signed-off-by: NeilBrown Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 54 +++++++++++++++++++++++++++++++++++++++++++++++--- fs/nfs/nfs3proc.c | 1 + fs/nfs/proc.c | 1 + include/linux/nfs_fs.h | 3 +++ 4 files changed, 56 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index ac505671efbd..342930996226 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -56,6 +56,8 @@ static int nfs_readdir(struct file *, struct dir_context *); static int nfs_fsync_dir(struct file *, loff_t, loff_t, int); static loff_t nfs_llseek_dir(struct file *, loff_t, int); static void nfs_readdir_clear_array(struct folio *); +static int nfs_do_create(struct inode *dir, struct dentry *dentry, + umode_t mode, int open_flags); const struct file_operations nfs_dir_operations = { .llseek = nfs_llseek_dir, @@ -2243,6 +2245,41 @@ static int nfs4_lookup_revalidate(struct dentry *dentry, unsigned int flags) #endif /* CONFIG_NFSV4 */ +int nfs_atomic_open_v23(struct inode *dir, struct dentry *dentry, + struct file *file, unsigned int open_flags, + umode_t mode) +{ + + /* Same as look+open from lookup_open(), but with different O_TRUNC + * handling. + */ + int error = 0; + + if (open_flags & O_CREAT) { + file->f_mode |= FMODE_CREATED; + error = nfs_do_create(dir, dentry, mode, open_flags); + if (error) + return error; + return finish_open(file, dentry, NULL); + } else if (d_in_lookup(dentry)) { + /* The only flags nfs_lookup considers are + * LOOKUP_EXCL and LOOKUP_RENAME_TARGET, and + * we want those to be zero so the lookup isn't skipped. + */ + struct dentry *res = nfs_lookup(dir, dentry, 0); + + d_lookup_done(dentry); + if (unlikely(res)) { + if (IS_ERR(res)) + return PTR_ERR(res); + return finish_no_open(file, res); + } + } + return finish_no_open(file, NULL); + +} +EXPORT_SYMBOL_GPL(nfs_atomic_open_v23); + struct dentry * nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, struct nfs_fattr *fattr) @@ -2303,18 +2340,23 @@ EXPORT_SYMBOL_GPL(nfs_instantiate); * that the operation succeeded on the server, but an error in the * reply path made it appear to have failed. */ -int nfs_create(struct mnt_idmap *idmap, struct inode *dir, - struct dentry *dentry, umode_t mode, bool excl) +static int nfs_do_create(struct inode *dir, struct dentry *dentry, + umode_t mode, int open_flags) { struct iattr attr; - int open_flags = excl ? O_CREAT | O_EXCL : O_CREAT; int error; + open_flags |= O_CREAT; + dfprintk(VFS, "NFS: create(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); attr.ia_mode = mode; attr.ia_valid = ATTR_MODE; + if (open_flags & O_TRUNC) { + attr.ia_size = 0; + attr.ia_valid |= ATTR_SIZE; + } trace_nfs_create_enter(dir, dentry, open_flags); error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags); @@ -2326,6 +2368,12 @@ out_err: d_drop(dentry); return error; } + +int nfs_create(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) +{ + return nfs_do_create(dir, dentry, mode, excl ? O_EXCL : 0); +} EXPORT_SYMBOL_GPL(nfs_create); /* diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index cbbe3f0193b8..74bda639a7cf 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -986,6 +986,7 @@ static int nfs3_have_delegation(struct inode *inode, fmode_t flags) static const struct inode_operations nfs3_dir_inode_operations = { .create = nfs_create, + .atomic_open = nfs_atomic_open_v23, .lookup = nfs_lookup, .link = nfs_link, .unlink = nfs_unlink, diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index ad3a321ae997..d105e5b2659d 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -695,6 +695,7 @@ static int nfs_have_delegation(struct inode *inode, fmode_t flags) static const struct inode_operations nfs_dir_inode_operations = { .create = nfs_create, .lookup = nfs_lookup, + .atomic_open = nfs_atomic_open_v23, .link = nfs_link, .unlink = nfs_unlink, .symlink = nfs_symlink, diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index d59116ac8209..039898d70954 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -561,6 +561,9 @@ extern int nfs_may_open(struct inode *inode, const struct cred *cred, int openfl extern void nfs_access_zap_cache(struct inode *inode); extern int nfs_access_get_cached(struct inode *inode, const struct cred *cred, u32 *mask, bool may_block); +extern int nfs_atomic_open_v23(struct inode *dir, struct dentry *dentry, + struct file *file, unsigned int open_flags, + umode_t mode); /* * linux/fs/nfs/symlink.c -- cgit v1.2.3-59-g8ed1b From bf95f82e6a569f41bae1e37204b219a5e1e8b971 Mon Sep 17 00:00:00 2001 From: Chen Hanxiao Date: Tue, 2 Apr 2024 18:33:55 +0800 Subject: NFS: make sure lock/nolock overriding local_lock mount option Currently, mount option lock/nolock and local_lock option may override NFS_MOUNT_LOCAL_FLOCK NFS_MOUNT_LOCAL_FCNTL flags when passing in different order: mount -o vers=3,local_lock=all,lock: local_lock=none mount -o vers=3,lock,local_lock=all: local_lock=all This patch will let lock/nolock override local_lock option as nfs(5) suggested. Signed-off-by: Chen Hanxiao Signed-off-by: Trond Myklebust --- fs/nfs/fs_context.c | 2 ++ fs/nfs/internal.h | 7 +++++++ fs/nfs/super.c | 10 ++++++++++ 3 files changed, 19 insertions(+) (limited to 'fs') diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c index d0a0956f8a13..ec93306b7e79 100644 --- a/fs/nfs/fs_context.c +++ b/fs/nfs/fs_context.c @@ -600,9 +600,11 @@ static int nfs_fs_context_parse_param(struct fs_context *fc, break; case Opt_lock: if (result.negated) { + ctx->lock_status = NFS_LOCK_NOLOCK; ctx->flags |= NFS_MOUNT_NONLM; ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); } else { + ctx->lock_status = NFS_LOCK_LOCK; ctx->flags &= ~NFS_MOUNT_NONLM; ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); } diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 06253695fe53..dc0693b3d214 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -112,6 +112,7 @@ struct nfs_fs_context { unsigned short protofamily; unsigned short mountfamily; bool has_sec_mnt_opts; + int lock_status; struct { union { @@ -153,6 +154,12 @@ struct nfs_fs_context { } clone_data; }; +enum nfs_lock_status { + NFS_LOCK_NOT_SET = 0, + NFS_LOCK_LOCK = 1, + NFS_LOCK_NOLOCK = 2, +}; + #define nfs_errorf(fc, fmt, ...) ((fc)->log.log ? \ errorf(fc, fmt, ## __VA_ARGS__) : \ ({ dprintk(fmt "\n", ## __VA_ARGS__); })) diff --git a/fs/nfs/super.c b/fs/nfs/super.c index dc03f98f7616..cbbd4866b0b7 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -901,6 +901,16 @@ static struct nfs_server *nfs_try_mount_request(struct fs_context *fc) rpc_authflavor_t authlist[NFS_MAX_SECFLAVORS]; unsigned int authlist_len = ARRAY_SIZE(authlist); + /* make sure 'nolock'/'lock' override the 'local_lock' mount option */ + if (ctx->lock_status) { + if (ctx->lock_status == NFS_LOCK_NOLOCK) { + ctx->flags |= NFS_MOUNT_NONLM; + ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); + } else { + ctx->flags &= ~NFS_MOUNT_NONLM; + ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); + } + } status = nfs_request_mount(fc, ctx->mntfh, authlist, &authlist_len); if (status) return ERR_PTR(status); -- cgit v1.2.3-59-g8ed1b From 37ffe06537af3e3ec212e7cbe941046fce0a822f Mon Sep 17 00:00:00 2001 From: Benjamin Coddington Date: Wed, 17 Apr 2024 14:49:29 -0400 Subject: NFSv4: Fixup smatch warning for ambiguous return Dan Carpenter reports smatch warning for nfs4_try_migration() when a memory allocation failure results in a zero return value. In this case, a transient allocation failure error will likely be retried the next time the server responds with NFS4ERR_MOVED. We can fixup the smatch warning with a small refactor: attempt all three allocations before testing and returning on a failure. Reported-by: Dan Carpenter Fixes: c3ed222745d9 ("NFSv4: Fix free of uninitialized nfs4_label on referral lookup.") Signed-off-by: Benjamin Coddington Reviewed-by: Dan Carpenter Reviewed-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfs4state.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 662e86ea3a2d..5b452411e8fd 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -2116,6 +2116,7 @@ static int nfs4_try_migration(struct nfs_server *server, const struct cred *cred { struct nfs_client *clp = server->nfs_client; struct nfs4_fs_locations *locations = NULL; + struct nfs_fattr *fattr; struct inode *inode; struct page *page; int status, result; @@ -2125,19 +2126,16 @@ static int nfs4_try_migration(struct nfs_server *server, const struct cred *cred (unsigned long long)server->fsid.minor, clp->cl_hostname); - result = 0; page = alloc_page(GFP_KERNEL); locations = kmalloc(sizeof(struct nfs4_fs_locations), GFP_KERNEL); - if (page == NULL || locations == NULL) { - dprintk("<-- %s: no memory\n", __func__); - goto out; - } - locations->fattr = nfs_alloc_fattr(); - if (locations->fattr == NULL) { + fattr = nfs_alloc_fattr(); + if (page == NULL || locations == NULL || fattr == NULL) { dprintk("<-- %s: no memory\n", __func__); + result = 0; goto out; } + locations->fattr = fattr; inode = d_inode(server->super->s_root); result = nfs4_proc_get_locations(server, NFS_FH(inode), locations, page, cred); -- cgit v1.2.3-59-g8ed1b From b322bf9e983addedff0894c55e92d58f4d16d92a Mon Sep 17 00:00:00 2001 From: Martin Kaiser Date: Sun, 14 Apr 2024 19:01:09 +0200 Subject: nfs: keep server info for remounts With newer kernels that use fs_context for nfs mounts, remounts fail with -EINVAL. $ mount -t nfs -o nolock 10.0.0.1:/tmp/test /mnt/test/ $ mount -t nfs -o remount /mnt/test/ mount: mounting 10.0.0.1:/tmp/test on /mnt/test failed: Invalid argument For remounts, the nfs server address and port are populated by nfs_init_fs_context and later overwritten with 0x00 bytes by nfs23_parse_monolithic. The remount then fails as the server address is invalid. Fix this by not overwriting nfs server info in nfs23_parse_monolithic if we're doing a remount. Fixes: f2aedb713c28 ("NFS: Add fs_context support.") Signed-off-by: Martin Kaiser Signed-off-by: Trond Myklebust --- fs/nfs/fs_context.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c index ec93306b7e79..6c9f3f6645dd 100644 --- a/fs/nfs/fs_context.c +++ b/fs/nfs/fs_context.c @@ -1114,9 +1114,12 @@ static int nfs23_parse_monolithic(struct fs_context *fc, ctx->acdirmax = data->acdirmax; ctx->need_mount = false; - memcpy(sap, &data->addr, sizeof(data->addr)); - ctx->nfs_server.addrlen = sizeof(data->addr); - ctx->nfs_server.port = ntohs(data->addr.sin_port); + if (!is_remount_fc(fc)) { + memcpy(sap, &data->addr, sizeof(data->addr)); + ctx->nfs_server.addrlen = sizeof(data->addr); + ctx->nfs_server.port = ntohs(data->addr.sin_port); + } + if (sap->ss_family != AF_INET || !nfs_verify_server_address(sap)) goto out_no_address; -- cgit v1.2.3-59-g8ed1b From f06d1b10cb016d5aaecdb1804fefca025387bd10 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 25 Apr 2024 16:24:29 -0400 Subject: NFS: Fix READ_PLUS when server doesn't support OP_READ_PLUS Olga showed me a case where the client was sending multiple READ_PLUS calls to the server in parallel, and the server replied NFS4ERR_OPNOTSUPP to each. The client would fall back to READ for the first reply, but fail to retry the other calls. I fix this by removing the test for NFS_CAP_READ_PLUS in nfs4_read_plus_not_supported(). This allows us to reschedule any READ_PLUS call that has a NFS4ERR_OPNOTSUPP return value, even after the capability has been cleared. Reported-by: Olga Kornievskaia Fixes: c567552612ec ("NFS: Add READ_PLUS data segment support") Cc: stable@vger.kernel.org # v5.10+ Signed-off-by: Anna Schumaker Reviewed-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index ea390db94b62..c93c12063b3a 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -5456,7 +5456,7 @@ static bool nfs4_read_plus_not_supported(struct rpc_task *task, struct rpc_message *msg = &task->tk_msg; if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_READ_PLUS] && - server->caps & NFS_CAP_READ_PLUS && task->tk_status == -ENOTSUPP) { + task->tk_status == -ENOTSUPP) { server->caps &= ~NFS_CAP_READ_PLUS; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; rpc_restart_call_prepare(task); -- cgit v1.2.3-59-g8ed1b From d1404e46ae4688c74e7504195f4fe253bcce1522 Mon Sep 17 00:00:00 2001 From: Anna Schumaker Date: Thu, 25 Apr 2024 16:34:40 -0400 Subject: NFS: Don't enable NFS v2 by default This came up during one of the Bake-a-thon discussions. NFS v2 support was dropped from nfs-utils/mount.nfs in December 2021. Let's turn it off by default in the kernel too, since this means there isn't a way to mount and test it. Signed-off-by: Anna Schumaker Reviewed-by: Jeffrey Layton Signed-off-by: Trond Myklebust --- fs/nfs/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index f7e32d76e34d..57249f040dfc 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -33,12 +33,12 @@ config NFS_FS config NFS_V2 tristate "NFS client support for NFS version 2" depends on NFS_FS - default y + default n help This option enables support for version 2 of the NFS protocol (RFC 1094) in the kernel's NFS client. - If unsure, say Y. + If unsure, say N. config NFS_V3 tristate "NFS client support for NFS version 3" -- cgit v1.2.3-59-g8ed1b From 80eb4f62056d6ae709bdd0636ab96ce660f494b2 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Mon, 20 May 2024 17:01:06 +0800 Subject: erofs: avoid allocating DEFLATE streams before mounting Currently, each DEFLATE stream takes one 32 KiB permanent internal window buffer even if there is no running instance which uses DEFLATE algorithm. It's unexpected and wasteful on embedded devices with limited resources and servers with hundreds of CPU cores if DEFLATE is enabled but unused. Fixes: ffa09b3bd024 ("erofs: DEFLATE compression support") Cc: # 6.6+ Reviewed-by: Sandeep Dhavale Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20240520090106.2898681-1-hsiangkao@linux.alibaba.com --- fs/erofs/decompressor_deflate.c | 55 ++++++++++++++++++++++------------------- 1 file changed, 29 insertions(+), 26 deletions(-) (limited to 'fs') diff --git a/fs/erofs/decompressor_deflate.c b/fs/erofs/decompressor_deflate.c index 81e65c453ef0..3a3461561a3c 100644 --- a/fs/erofs/decompressor_deflate.c +++ b/fs/erofs/decompressor_deflate.c @@ -46,39 +46,15 @@ int __init z_erofs_deflate_init(void) /* by default, use # of possible CPUs instead */ if (!z_erofs_deflate_nstrms) z_erofs_deflate_nstrms = num_possible_cpus(); - - for (; z_erofs_deflate_avail_strms < z_erofs_deflate_nstrms; - ++z_erofs_deflate_avail_strms) { - struct z_erofs_deflate *strm; - - strm = kzalloc(sizeof(*strm), GFP_KERNEL); - if (!strm) - goto out_failed; - - /* XXX: in-kernel zlib cannot shrink windowbits currently */ - strm->z.workspace = vmalloc(zlib_inflate_workspacesize()); - if (!strm->z.workspace) { - kfree(strm); - goto out_failed; - } - - spin_lock(&z_erofs_deflate_lock); - strm->next = z_erofs_deflate_head; - z_erofs_deflate_head = strm; - spin_unlock(&z_erofs_deflate_lock); - } return 0; - -out_failed: - erofs_err(NULL, "failed to allocate zlib workspace"); - z_erofs_deflate_exit(); - return -ENOMEM; } int z_erofs_load_deflate_config(struct super_block *sb, struct erofs_super_block *dsb, void *data, int size) { struct z_erofs_deflate_cfgs *dfl = data; + static DEFINE_MUTEX(deflate_resize_mutex); + static bool inited; if (!dfl || size < sizeof(struct z_erofs_deflate_cfgs)) { erofs_err(sb, "invalid deflate cfgs, size=%u", size); @@ -89,9 +65,36 @@ int z_erofs_load_deflate_config(struct super_block *sb, erofs_err(sb, "unsupported windowbits %u", dfl->windowbits); return -EOPNOTSUPP; } + mutex_lock(&deflate_resize_mutex); + if (!inited) { + for (; z_erofs_deflate_avail_strms < z_erofs_deflate_nstrms; + ++z_erofs_deflate_avail_strms) { + struct z_erofs_deflate *strm; + + strm = kzalloc(sizeof(*strm), GFP_KERNEL); + if (!strm) + goto failed; + /* XXX: in-kernel zlib cannot customize windowbits */ + strm->z.workspace = vmalloc(zlib_inflate_workspacesize()); + if (!strm->z.workspace) { + kfree(strm); + goto failed; + } + spin_lock(&z_erofs_deflate_lock); + strm->next = z_erofs_deflate_head; + z_erofs_deflate_head = strm; + spin_unlock(&z_erofs_deflate_lock); + } + inited = true; + } + mutex_unlock(&deflate_resize_mutex); erofs_info(sb, "EXPERIMENTAL DEFLATE feature in use. Use at your own risk!"); return 0; +failed: + mutex_unlock(&deflate_resize_mutex); + z_erofs_deflate_exit(); + return -ENOMEM; } int z_erofs_deflate_decompress(struct z_erofs_decompress_req *rq, -- cgit v1.2.3-59-g8ed1b From d0e71e23ec5e71655e1a046c9be6f35f78b1d6bb Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 20 May 2024 12:43:58 -0700 Subject: Revert "fanotify: remove unneeded sub-zero check for unsigned value" This reverts commit e6595224464b692ddae193d783402130d1625147. These kinds of patches are only making the code worse. Compilers don't care about the unnecessary check, but removing it makes the code less obvious to a human. The declaration of 'len' is more than 80 lines earlier, so a human won't easily see that 'len' is of an unsigned type, so to a human the range check that checks against zero is much more explicit and obvious. Any tool that complains about a range check like this just because the variable is unsigned is actively detrimental, and should be ignored. Signed-off-by: Linus Torvalds --- fs/notify/fanotify/fanotify_user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 483a6a1255fb..9ec313e9f6e1 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -502,7 +502,7 @@ static int copy_fid_info_to_user(__kernel_fsid_t *fsid, struct fanotify_fh *fh, } /* Pad with 0's */ - WARN_ON_ONCE(len >= FANOTIFY_EVENT_ALIGN); + WARN_ON_ONCE(len < 0 || len >= FANOTIFY_EVENT_ALIGN); if (len > 0 && clear_user(buf, len)) return -EFAULT; -- cgit v1.2.3-59-g8ed1b From 3ebcb24646f8c5bfad2866892d3f3cff05514452 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Tue, 7 May 2024 11:15:45 -0400 Subject: pNFS/filelayout: fixup pNfs allocation modes Change left over allocation flags. Fixes: a245832aaa99 ("pNFS/files: Ensure pNFS allocation modes are consistent with nfsiod") Signed-off-by: Olga Kornievskaia Reviewed-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- fs/nfs/filelayout/filelayout.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index cc2ed4b5a4fd..85d2dc9bc212 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -875,7 +875,7 @@ filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, req->wb_bytes, IOMODE_READ, false, - GFP_KERNEL); + nfs_io_gfp_mask()); if (IS_ERR(pgio->pg_lseg)) { pgio->pg_error = PTR_ERR(pgio->pg_lseg); pgio->pg_lseg = NULL; @@ -899,7 +899,7 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, req->wb_bytes, IOMODE_RW, false, - GFP_NOFS); + nfs_io_gfp_mask()); if (IS_ERR(pgio->pg_lseg)) { pgio->pg_error = PTR_ERR(pgio->pg_lseg); pgio->pg_lseg = NULL; -- cgit v1.2.3-59-g8ed1b From 523412b904c4ad2e9649d536bde0e6c75c88638b Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Tue, 7 May 2024 15:59:33 -0400 Subject: pNFS/filelayout: check layout segment range Before doing the IO, check that we have the layout covering the range of IO. Signed-off-by: Olga Kornievskaia Reviewed-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- fs/nfs/filelayout/filelayout.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index 85d2dc9bc212..bf3ba2e98f33 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -868,6 +868,7 @@ filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { pnfs_generic_pg_check_layout(pgio); + pnfs_generic_pg_check_range(pgio, req); if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), @@ -892,6 +893,7 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { pnfs_generic_pg_check_layout(pgio); + pnfs_generic_pg_check_range(pgio, req); if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), -- cgit v1.2.3-59-g8ed1b From a01b077a8743b2ed329c587cf4bbe49682da243f Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Fri, 10 May 2024 15:07:43 -0400 Subject: pNFS: rework pnfs_generic_pg_check_layout to check IO range All callers of pnfs_generic_pg_check_layout() also want to do a call to check that the layout's range covers the IO range. Merge the functionality of the pnfs_generic_pg_check_range() into that of pnfs_generic_pg_check_layout(). Signed-off-by: Olga Kornievskaia Reviewed-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- fs/nfs/filelayout/filelayout.c | 6 ++---- fs/nfs/flexfilelayout/flexfilelayout.c | 12 ++---------- fs/nfs/pnfs.c | 29 +++++++++-------------------- fs/nfs/pnfs.h | 3 +-- 4 files changed, 14 insertions(+), 36 deletions(-) (limited to 'fs') diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index bf3ba2e98f33..29d84dc66ca3 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -867,8 +867,7 @@ static void filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { - pnfs_generic_pg_check_layout(pgio); - pnfs_generic_pg_check_range(pgio, req); + pnfs_generic_pg_check_layout(pgio, req); if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), @@ -892,8 +891,7 @@ static void filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { - pnfs_generic_pg_check_layout(pgio); - pnfs_generic_pg_check_range(pgio, req); + pnfs_generic_pg_check_layout(pgio, req); if (!pgio->pg_lseg) { pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 3e724cb7ef01..24188af56d5b 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -822,14 +822,6 @@ ff_layout_pg_get_read(struct nfs_pageio_descriptor *pgio, } } -static void -ff_layout_pg_check_layout(struct nfs_pageio_descriptor *pgio, - struct nfs_page *req) -{ - pnfs_generic_pg_check_layout(pgio); - pnfs_generic_pg_check_range(pgio, req); -} - static void ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) @@ -840,7 +832,7 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio, u32 ds_idx; retry: - ff_layout_pg_check_layout(pgio, req); + pnfs_generic_pg_check_layout(pgio, req); /* Use full layout for now */ if (!pgio->pg_lseg) { ff_layout_pg_get_read(pgio, req, false); @@ -895,7 +887,7 @@ ff_layout_pg_init_write(struct nfs_pageio_descriptor *pgio, u32 i; retry: - ff_layout_pg_check_layout(pgio, req); + pnfs_generic_pg_check_layout(pgio, req); if (!pgio->pg_lseg) { pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index a5cc6199127f..b5834728f31b 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -2705,38 +2705,28 @@ pnfs_layout_return_unused_byclid(struct nfs_client *clp, &range); } +/* Check if we have we have a valid layout but if there isn't an intersection + * between the request and the pgio->pg_lseg, put this pgio->pg_lseg away. + */ void -pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio) +pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) { if (pgio->pg_lseg == NULL || - test_bit(NFS_LSEG_VALID, &pgio->pg_lseg->pls_flags)) + (test_bit(NFS_LSEG_VALID, &pgio->pg_lseg->pls_flags) && + pnfs_lseg_request_intersecting(pgio->pg_lseg, req))) return; pnfs_put_lseg(pgio->pg_lseg); pgio->pg_lseg = NULL; } EXPORT_SYMBOL_GPL(pnfs_generic_pg_check_layout); -/* - * Check for any intersection between the request and the pgio->pg_lseg, - * and if none, put this pgio->pg_lseg away. - */ -void -pnfs_generic_pg_check_range(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) -{ - if (pgio->pg_lseg && !pnfs_lseg_request_intersecting(pgio->pg_lseg, req)) { - pnfs_put_lseg(pgio->pg_lseg); - pgio->pg_lseg = NULL; - } -} -EXPORT_SYMBOL_GPL(pnfs_generic_pg_check_range); - void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { u64 rd_size; - pnfs_generic_pg_check_layout(pgio); - pnfs_generic_pg_check_range(pgio, req); + pnfs_generic_pg_check_layout(pgio, req); if (pgio->pg_lseg == NULL) { if (pgio->pg_dreq == NULL) rd_size = i_size_read(pgio->pg_inode) - req_offset(req); @@ -2766,8 +2756,7 @@ void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req, u64 wb_size) { - pnfs_generic_pg_check_layout(pgio); - pnfs_generic_pg_check_range(pgio, req); + pnfs_generic_pg_check_layout(pgio, req); if (pgio->pg_lseg == NULL) { pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index db57a85500ee..fa5beeaaf5da 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -257,8 +257,7 @@ void pnfs_put_lseg(struct pnfs_layout_segment *lseg); void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, struct nfs_fsinfo *); void unset_pnfs_layoutdriver(struct nfs_server *); -void pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio); -void pnfs_generic_pg_check_range(struct nfs_pageio_descriptor *pgio, struct nfs_page *req); +void pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio, struct nfs_page *req); void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *); int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc); void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, -- cgit v1.2.3-59-g8ed1b From 3c0a2e0b0ae661457c8505fecc7be5501aa7a715 Mon Sep 17 00:00:00 2001 From: Sergey Shtylyov Date: Fri, 10 May 2024 23:24:04 +0300 Subject: nfs: fix undefined behavior in nfs_block_bits() Shifting *signed int* typed constant 1 left by 31 bits causes undefined behavior. Specify the correct *unsigned long* type by using 1UL instead. Found by Linux Verification Center (linuxtesting.org) with the Svace static analysis tool. Cc: stable@vger.kernel.org Signed-off-by: Sergey Shtylyov Reviewed-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- fs/nfs/internal.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index dc0693b3d214..9f0f4534744b 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -717,9 +717,9 @@ unsigned long nfs_block_bits(unsigned long bsize, unsigned char *nrbitsp) if ((bsize & (bsize - 1)) || nrbitsp) { unsigned char nrbits; - for (nrbits = 31; nrbits && !(bsize & (1 << nrbits)); nrbits--) + for (nrbits = 31; nrbits && !(bsize & (1UL << nrbits)); nrbits--) ; - bsize = 1 << nrbits; + bsize = 1UL << nrbits; if (nrbitsp) *nrbitsp = nrbits; } -- cgit v1.2.3-59-g8ed1b From 440861b1a03c72cc7be4a307e178dcaa6894479b Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Tue, 21 May 2024 19:27:31 +0930 Subject: btrfs: re-introduce 'norecovery' mount option Although 'norecovery' mount option was marked as deprecated for a long time and a warning message was printed during the deprecation window, it's still actively utilized by several projects that need a safer way to mount a btrfs without any writes. Furthermore this 'norecovery' mount option is supported by other major filesystems, which makes it less clear what's our motivation to remove it. Re-introduce the 'norecovery' mount option, and output a message to recommend 'rescue=nologreplay' option. Link: https://lore.kernel.org/linux-btrfs/ZkxZT0J-z0GYvfy8@gardel-login/#t Link: https://github.com/systemd/systemd/pull/32892 Link: https://bugzilla.suse.com/show_bug.cgi?id=1222429 Reported-by: Lennart Poettering Reported-by: Jiri Slaby Fixes: a1912f712188 ("btrfs: remove code for inode_cache and recovery mount options") CC: stable@vger.kernel.org # 6.8+ Reviewed-by: Johannes Thumshirn Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/super.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 2dbc930a20f7..f05cce7c8b8d 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -119,6 +119,7 @@ enum { Opt_thread_pool, Opt_treelog, Opt_user_subvol_rm_allowed, + Opt_norecovery, /* Rescue options */ Opt_rescue, @@ -245,6 +246,8 @@ static const struct fs_parameter_spec btrfs_fs_parameters[] = { __fsparam(NULL, "nologreplay", Opt_nologreplay, fs_param_deprecated, NULL), /* Deprecated, with alias rescue=usebackuproot */ __fsparam(NULL, "usebackuproot", Opt_usebackuproot, fs_param_deprecated, NULL), + /* For compatibility only, alias for "rescue=nologreplay". */ + fsparam_flag("norecovery", Opt_norecovery), /* Debugging options. */ fsparam_flag_no("enospc_debug", Opt_enospc_debug), @@ -438,6 +441,11 @@ static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param) "'nologreplay' is deprecated, use 'rescue=nologreplay' instead"); btrfs_set_opt(ctx->mount_opt, NOLOGREPLAY); break; + case Opt_norecovery: + btrfs_info(NULL, +"'norecovery' is for compatibility only, recommended to use 'rescue=nologreplay'"); + btrfs_set_opt(ctx->mount_opt, NOLOGREPLAY); + break; case Opt_flushoncommit: if (result.negated) btrfs_clear_opt(ctx->mount_opt, FLUSHONCOMMIT); -- cgit v1.2.3-59-g8ed1b From db3d841ac9edb0b98cc002e3b27c0b266ecfe5ba Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 May 2024 14:34:43 +0200 Subject: fs/pidfs: make 'lsof' happy with our inode changes pidfs started using much saner inodes in commit b28ddcc32d8f ("pidfs: convert to path_from_stashed() helper"), but that exposed the fact that lsof had some knowledge of just how odd our old anon_inode usage was. For example, legacy anon_inodes hadn't even initialized the inode type in the inode mode, so everything had a type of zero. So sane tools like 'stat' would report these files as "weird file", but 'lsof' instead used that (together with the name of the link in proc) to notice that it's an anonymous inode, and used it to detect pidfd files. Let's keep our internal new sane inode model, but mask the file type bits at 'stat()' time in the getattr() function we already have, and by making the dentry name match what lsof expects too. This keeps our internal models sane, but should make user space see the same old odd behavior. Reported-by: Jiri Slaby Link: https://lore.kernel.org/all/a15b1050-4b52-4740-a122-a4d055c17f11@kernel.org/ Link: https://github.com/lsof-org/lsof/issues/317 Cc: Alexander Viro Cc: Seth Forshee Cc: Tycho Andersen Signed-off-by: Christian Brauner Signed-off-by: Linus Torvalds --- fs/pidfs.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/pidfs.c b/fs/pidfs.c index a63d5d24aa02..dbb9d854d1c5 100644 --- a/fs/pidfs.c +++ b/fs/pidfs.c @@ -169,6 +169,24 @@ static int pidfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, return -EOPNOTSUPP; } + +/* + * User space expects pidfs inodes to have no file type in st_mode. + * + * In particular, 'lsof' has this legacy logic: + * + * type = s->st_mode & S_IFMT; + * switch (type) { + * ... + * case 0: + * if (!strcmp(p, "anon_inode")) + * Lf->ntype = Ntype = N_ANON_INODE; + * + * to detect our old anon_inode logic. + * + * Rather than mess with our internal sane inode data, just fix it + * up here in getattr() by masking off the format bits. + */ static int pidfs_getattr(struct mnt_idmap *idmap, const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) @@ -176,6 +194,7 @@ static int pidfs_getattr(struct mnt_idmap *idmap, const struct path *path, struct inode *inode = d_inode(path->dentry); generic_fillattr(&nop_mnt_idmap, request_mask, inode, stat); + stat->mode &= ~S_IFMT; return 0; } @@ -199,12 +218,13 @@ static const struct super_operations pidfs_sops = { .statfs = simple_statfs, }; +/* + * 'lsof' has knowledge of out historical anon_inode use, and expects + * the pidfs dentry name to start with 'anon_inode'. + */ static char *pidfs_dname(struct dentry *dentry, char *buffer, int buflen) { - struct inode *inode = d_inode(dentry); - struct pid *pid = inode->i_private; - - return dynamic_dname(buffer, buflen, "pidfd:[%llu]", pid->ino); + return dynamic_dname(buffer, buflen, "anon_inode:[pidfd]"); } static const struct dentry_operations pidfs_dentry_operations = { -- cgit v1.2.3-59-g8ed1b From 16e00683dc74cf1fcdf00046b90852bee05eb94a Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 15 May 2024 18:06:03 -0500 Subject: smb3: reenable swapfiles over SMB3 mounts With the changes to folios/netfs it is now easier to reenable swapfile support over SMB3 which fixes various xfstests Reviewed-by: David Howells Suggested-by: David Howells Fixes: e1209d3a7a67 ("mm: introduce ->swap_rw and use it for reads from SWP_FS_OPS swap-space") Signed-off-by: Steve French --- fs/netfs/direct_write.c | 3 ++- fs/smb/client/file.c | 23 +++++++++++++++++++++++ include/linux/netfs.h | 2 ++ 3 files changed, 27 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c index 608ba6416919..f516460e994e 100644 --- a/fs/netfs/direct_write.c +++ b/fs/netfs/direct_write.c @@ -27,7 +27,7 @@ static void netfs_cleanup_dio_write(struct netfs_io_request *wreq) * Perform an unbuffered write where we may have to do an RMW operation on an * encrypted file. This can also be used for direct I/O writes. */ -static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *iter, +ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *iter, struct netfs_group *netfs_group) { struct netfs_io_request *wreq; @@ -117,6 +117,7 @@ out: netfs_put_request(wreq, false, netfs_rreq_trace_put_return); return ret; } +EXPORT_SYMBOL(netfs_unbuffered_write_iter_locked); /** * netfs_unbuffered_write_iter - Unbuffered write to a file diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 9d38294a7e68..9d5c2440abfc 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -3189,6 +3189,28 @@ static void cifs_swap_deactivate(struct file *file) /* do we need to unpin (or unlock) the file */ } +/** + * cifs_swap_rw - SMB3 address space operation for swap I/O + * @iocb: target I/O control block + * @iter: I/O buffer + * + * Perform IO to the swap-file. This is much like direct IO. + */ +static int cifs_swap_rw(struct kiocb *iocb, struct iov_iter *iter) +{ + ssize_t ret; + + WARN_ON_ONCE(iov_iter_count(iter) != PAGE_SIZE); + + if (iov_iter_rw(iter) == READ) + ret = netfs_unbuffered_read_iter_locked(iocb, iter); + else + ret = netfs_unbuffered_write_iter_locked(iocb, iter, NULL); + if (ret < 0) + return ret; + return 0; +} + const struct address_space_operations cifs_addr_ops = { .read_folio = netfs_read_folio, .readahead = netfs_readahead, @@ -3204,6 +3226,7 @@ const struct address_space_operations cifs_addr_ops = { */ .swap_activate = cifs_swap_activate, .swap_deactivate = cifs_swap_deactivate, + .swap_rw = cifs_swap_rw, }; /* diff --git a/include/linux/netfs.h b/include/linux/netfs.h index ca56a4428043..d2d291a9cdad 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -400,6 +400,8 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, ssize_t netfs_buffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *from, struct netfs_group *netfs_group); ssize_t netfs_unbuffered_write_iter(struct kiocb *iocb, struct iov_iter *from); +ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *iter, + struct netfs_group *netfs_group); ssize_t netfs_file_write_iter(struct kiocb *iocb, struct iov_iter *from); /* Address operations API */ -- cgit v1.2.3-59-g8ed1b From 10c623a1956e673a9268493e0dacf373ddb2f9bf Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 15 May 2024 11:03:35 -0500 Subject: cifs: update internal version number to 2.49 Signed-off-by: Steve French --- fs/smb/client/cifsfs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.h b/fs/smb/client/cifsfs.h index 87310f05d397..62d5fee3e5eb 100644 --- a/fs/smb/client/cifsfs.h +++ b/fs/smb/client/cifsfs.h @@ -147,6 +147,6 @@ extern const struct export_operations cifs_export_ops; #endif /* CONFIG_CIFS_NFSD_EXPORT */ /* when changing internal version - update following two lines at same time */ -#define SMB3_PRODUCT_BUILD 48 -#define CIFS_VERSION "2.48" +#define SMB3_PRODUCT_BUILD 49 +#define CIFS_VERSION "2.49" #endif /* _CIFSFS_H */ -- cgit v1.2.3-59-g8ed1b From 240a1853b4d2bce51e5cac9ba65cd646152ab6d6 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Fri, 15 Mar 2024 19:47:07 -0500 Subject: kernel: Remove signal hacks for vhost_tasks This removes the signal/coredump hacks added for vhost_tasks in: Commit f9010dbdce91 ("fork, vhost: Use CLONE_THREAD to fix freezer/ps regression") When that patch was added vhost_tasks did not handle SIGKILL and would try to ignore/clear the signal and continue on until the device's close function was called. In the previous patches vhost_tasks and the vhost drivers were converted to support SIGKILL by cleaning themselves up and exiting. The hacks are no longer needed so this removes them. Signed-off-by: Mike Christie Message-Id: <20240316004707.45557-10-michael.christie@oracle.com> Signed-off-by: Michael S. Tsirkin --- fs/coredump.c | 4 +--- kernel/exit.c | 5 +---- kernel/signal.c | 4 +--- 3 files changed, 3 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/coredump.c b/fs/coredump.c index be6403b4b14b..8eae24afb3cb 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -371,9 +371,7 @@ static int zap_process(struct task_struct *start, int exit_code) if (t != current && !(t->flags & PF_POSTCOREDUMP)) { sigaddset(&t->pending.signal, SIGKILL); signal_wake_up(t, 1); - /* The vhost_worker does not particpate in coredumps */ - if ((t->flags & (PF_USER_WORKER | PF_IO_WORKER)) != PF_USER_WORKER) - nr++; + nr++; } } diff --git a/kernel/exit.c b/kernel/exit.c index 41a12630cbbc..fca3a3234954 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -414,10 +414,7 @@ static void coredump_task_exit(struct task_struct *tsk) tsk->flags |= PF_POSTCOREDUMP; core_state = tsk->signal->core_state; spin_unlock_irq(&tsk->sighand->siglock); - - /* The vhost_worker does not particpate in coredumps */ - if (core_state && - ((tsk->flags & (PF_IO_WORKER | PF_USER_WORKER)) != PF_USER_WORKER)) { + if (core_state) { struct core_thread self; self.task = current; diff --git a/kernel/signal.c b/kernel/signal.c index 7bdbcf1b78d0..41d5cbccab2a 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1375,9 +1375,7 @@ int zap_other_threads(struct task_struct *p) for_other_threads(p, t) { task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK); - /* Don't require de_thread to wait for the vhost_worker */ - if ((t->flags & (PF_IO_WORKER | PF_USER_WORKER)) != PF_USER_WORKER) - count++; + count++; /* Don't bother with already dead threads */ if (t->exit_state) -- cgit v1.2.3-59-g8ed1b From bc21020f61f5f93f39eeedc7ae56797e59b98816 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sun, 31 Mar 2024 10:44:11 +0200 Subject: fuse: virtio: drop owner assignment virtio core already sets the .owner, so driver does not need to. Signed-off-by: Krzysztof Kozlowski Message-Id: <20240331-module-owner-virtio-v2-24-98f04bfaf46a@linaro.org> Signed-off-by: Michael S. Tsirkin Reviewed-by: Stefan Hajnoczi --- fs/fuse/virtio_fs.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs') diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index 322af827a232..ca7b64f9c3c7 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -1023,7 +1023,6 @@ static const unsigned int feature_table[] = {}; static struct virtio_driver virtio_fs_driver = { .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, .id_table = id_table, .feature_table = feature_table, .feature_table_size = ARRAY_SIZE(feature_table), -- cgit v1.2.3-59-g8ed1b From 681ce8623567ba7e7333908e9826b77145312dda Mon Sep 17 00:00:00 2001 From: Yafang Shao Date: Wed, 15 May 2024 17:17:27 +0800 Subject: vfs: Delete the associated dentry when deleting a file Our applications, built on Elasticsearch[0], frequently create and delete files. These applications operate within containers, some with a memory limit exceeding 100GB. Over prolonged periods, the accumulation of negative dentries within these containers can amount to tens of gigabytes. Upon container exit, directories are deleted. However, due to the numerous associated dentries, this process can be time-consuming. Our users have expressed frustration with this prolonged exit duration, which constitutes our first issue. Simultaneously, other processes may attempt to access the parent directory of the Elasticsearch directories. Since the task responsible for deleting the dentries holds the inode lock, processes attempting directory lookup experience significant delays. This issue, our second problem, is easily demonstrated: - Task 1 generates negative dentries: $ pwd ~/test $ mkdir es && cd es/ && ./create_and_delete_files.sh [ After generating tens of GB dentries ] $ cd ~/test && rm -rf es [ It will take a long duration to finish ] - Task 2 attempts to lookup the 'test/' directory $ pwd ~/test $ ls The 'ls' command in Task 2 experiences prolonged execution as Task 1 is deleting the dentries. We've devised a solution to address both issues by deleting associated dentry when removing a file. Interestingly, we've noted that a similar patch was proposed years ago[1], although it was rejected citing the absence of tangible issues caused by negative dentries. Given our current challenges, we're resubmitting the proposal. All relevant stakeholders from previous discussions have been included for reference. Some alternative solutions are also under discussion[2][3], such as shrinking child dentries outside of the parent inode lock or even asynchronously shrinking child dentries. However, given the straightforward nature of the current solution, I believe this approach is still necessary. [ NOTE! This is a pretty fundamental change in how we deal with unlinking dentries, and it doesn't change the fact that you can have lots of negative dentries from just doing negative lookups. But the kernel test robot is at least initially happy with this from a performance angle, so I'm applying this ASAP just to get more testing and as a "known fix for an issue people hit in real life". Put another way: we should still look at the alternatives, and this patch may get reverted if somebody finds a performance regression on some other load. - Linus ] Suggested-by: Linus Torvalds Signed-off-by: Yafang Shao Link: https://github.com/elastic/elasticsearch [0] Link: https://patchwork.kernel.org/project/linux-fsdevel/patch/1502099673-31620-1-git-send-email-wangkai86@huawei.com [1] Link: https://lore.kernel.org/linux-fsdevel/20240511200240.6354-2-torvalds@linux-foundation.org/ [2] Link: https://lore.kernel.org/linux-fsdevel/CAHk-=wjEMf8Du4UFzxuToGDnF3yLaMcrYeyNAaH1NJWa6fwcNQ@mail.gmail.com/ [3] Cc: Al Viro Cc: Christian Brauner Cc: Jan Kara Cc: Waiman Long Cc: Matthew Wilcox Cc: Wangkai Cc: Colin Walters Tested-by: kernel test robot Link: https://lore.kernel.org/all/202405221518.ecea2810-oliver.sang@intel.com/ Signed-off-by: Linus Torvalds --- fs/dcache.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/dcache.c b/fs/dcache.c index 407095188f83..1ee6404b430b 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -2360,19 +2360,17 @@ EXPORT_SYMBOL(d_hash_and_lookup); * - unhash this dentry and free it. * * Usually, we want to just turn this into - * a negative dentry, but if anybody else is - * currently using the dentry or the inode - * we can't do that and we fall back on removing - * it from the hash queues and waiting for - * it to be deleted later when it has no users + * a negative dentry, but certain workloads can + * generate a large number of negative dentries. + * Therefore, it would be better to simply + * unhash it. */ /** * d_delete - delete a dentry * @dentry: The dentry to delete * - * Turn the dentry into a negative dentry if possible, otherwise - * remove it from the hash queues so it can be deleted later + * Remove the dentry from the hash queues so it can be deleted later. */ void d_delete(struct dentry * dentry) @@ -2381,6 +2379,8 @@ void d_delete(struct dentry * dentry) spin_lock(&inode->i_lock); spin_lock(&dentry->d_lock); + __d_drop(dentry); + /* * Are we the only user? */ @@ -2388,7 +2388,6 @@ void d_delete(struct dentry * dentry) dentry->d_flags &= ~DCACHE_CANT_MOUNT; dentry_unlink_inode(dentry); } else { - __d_drop(dentry); spin_unlock(&dentry->d_lock); spin_unlock(&inode->i_lock); } -- cgit v1.2.3-59-g8ed1b From 5fa421448d1f51b8991dd550a4b6347229116cc7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 20 May 2024 00:58:04 -0400 Subject: bcachefs: Fix bogus verify_replicas_entry() assert verify_replicas_entry() is only for newly created replicas entries - existing entries on disk may have unknown data types, and we have real verifiers for them. Reported-by: syzbot+73414091bd382684ee2b@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/replicas.c | 15 ++------------- 1 file changed, 2 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c index d45b10bc531c..57a1f09cca09 100644 --- a/fs/bcachefs/replicas.c +++ b/fs/bcachefs/replicas.c @@ -23,14 +23,12 @@ static int bch2_memcmp(const void *l, const void *r, const void *priv) static void verify_replicas_entry(struct bch_replicas_entry_v1 *e) { #ifdef CONFIG_BCACHEFS_DEBUG - unsigned i; - BUG_ON(e->data_type >= BCH_DATA_NR); BUG_ON(!e->nr_devs); BUG_ON(e->nr_required > 1 && e->nr_required >= e->nr_devs); - for (i = 0; i + 1 < e->nr_devs; i++) + for (unsigned i = 0; i + 1 < e->nr_devs; i++) BUG_ON(e->devs[i] >= e->devs[i + 1]); #endif } @@ -192,24 +190,17 @@ cpu_replicas_add_entry(struct bch_fs *c, struct bch_replicas_cpu *old, struct bch_replicas_entry_v1 *new_entry) { - unsigned i; struct bch_replicas_cpu new = { .nr = old->nr + 1, .entry_size = max_t(unsigned, old->entry_size, replicas_entry_bytes(new_entry)), }; - for (i = 0; i < new_entry->nr_devs; i++) - BUG_ON(!bch2_dev_exists(c, new_entry->devs[i])); - - BUG_ON(!new_entry->data_type); - verify_replicas_entry(new_entry); - new.entries = kcalloc(new.nr, new.entry_size, GFP_KERNEL); if (!new.entries) return new; - for (i = 0; i < old->nr; i++) + for (unsigned i = 0; i < old->nr; i++) memcpy(cpu_replicas_entry(&new, i), cpu_replicas_entry(old, i), old->entry_size); @@ -230,8 +221,6 @@ static inline int __replicas_entry_idx(struct bch_replicas_cpu *r, if (unlikely(entry_size > r->entry_size)) return -1; - verify_replicas_entry(search); - #define entry_cmp(_l, _r) memcmp(_l, _r, entry_size) idx = eytzinger0_find(r->entries, r->nr, r->entry_size, entry_cmp, search); -- cgit v1.2.3-59-g8ed1b From 70dd062e27e679247cd8828c23c0e12728de5465 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 20 May 2024 03:03:29 -0400 Subject: bcachefs: Fix btree_trans leak in bch2_readahead() Reported-by: syzbot+d797fe78808e968d6c84@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/fs-io-buffered.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs-io-buffered.c b/fs/bcachefs/fs-io-buffered.c index b0a33fabadf8..6b69e5cd68dd 100644 --- a/fs/bcachefs/fs-io-buffered.c +++ b/fs/bcachefs/fs-io-buffered.c @@ -257,7 +257,6 @@ void bch2_readahead(struct readahead_control *ractl) struct bch_inode_info *inode = to_bch_ei(ractl->mapping->host); struct bch_fs *c = inode->v.i_sb->s_fs_info; struct bch_io_opts opts; - struct btree_trans *trans = bch2_trans_get(c); struct folio *folio; struct readpages_iter readpages_iter; @@ -269,6 +268,7 @@ void bch2_readahead(struct readahead_control *ractl) bch2_pagecache_add_get(inode); + struct btree_trans *trans = bch2_trans_get(c); while ((folio = readpage_iter_peek(&readpages_iter))) { unsigned n = min_t(unsigned, readpages_iter.folios.nr - @@ -289,10 +289,10 @@ void bch2_readahead(struct readahead_control *ractl) &readpages_iter); bch2_trans_unlock(trans); } + bch2_trans_put(trans); bch2_pagecache_add_put(inode); - bch2_trans_put(trans); darray_exit(&readpages_iter.folios); } -- cgit v1.2.3-59-g8ed1b From 2ba24864d2f61b52210ba645ee5af1d0422a7ea9 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 20 May 2024 03:13:57 -0400 Subject: bcachefs: Fix stack oob in __bch2_encrypt_bio() Reported-by: syzbot+fff6b0fb00259873576a@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/checksum.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index 4b3d6f0137f6..3bd3aba90d8f 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -352,8 +352,12 @@ int __bch2_encrypt_bio(struct bch_fs *c, unsigned type, bytes += bv.bv_len; } - sg_mark_end(sg - 1); - return do_encrypt_sg(c->chacha20, nonce, sgl, bytes); + if (sg != sgl) { + sg_mark_end(sg - 1); + return do_encrypt_sg(c->chacha20, nonce, sgl, bytes); + } + + return ret; } struct bch_csum bch2_checksum_merge(unsigned type, struct bch_csum a, -- cgit v1.2.3-59-g8ed1b From 2195b755ebd23992c5758d2d02a8080eac2baeca Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 22 May 2024 19:14:08 -0400 Subject: bcachefs: Fix unsafety in bch2_dirent_name_bytes() Reported-by: syzbot+84fa6fb8c7f98b93cdea@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/dirent.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c index 6bbf9a7d9e4d..c67460d8205d 100644 --- a/fs/bcachefs/dirent.c +++ b/fs/bcachefs/dirent.c @@ -15,6 +15,9 @@ static unsigned bch2_dirent_name_bytes(struct bkey_s_c_dirent d) { + if (bkey_val_bytes(d.k) < offsetof(struct bch_dirent, d_name)) + return 0; + unsigned bkey_u64s = bkey_val_u64s(d.k); unsigned bkey_bytes = bkey_u64s * sizeof(u64); u64 last_u64 = ((u64*)d.v)[bkey_u64s - 1]; -- cgit v1.2.3-59-g8ed1b From 405ee4097c4bc3e70556520aed5ba52a511c2266 Mon Sep 17 00:00:00 2001 From: Nandor Kracser Date: Tue, 21 May 2024 22:56:17 +0900 Subject: ksmbd: ignore trailing slashes in share paths Trailing slashes in share paths (like: /home/me/Share/) caused permission issues with shares for clients on iOS and on Android TV for me, but otherwise they work fine with plain old Samba. Cc: stable@vger.kernel.org Signed-off-by: Nandor Kracser Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/mgmt/share_config.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smb/server/mgmt/share_config.c b/fs/smb/server/mgmt/share_config.c index a2f0a2edceb8..e0a6b758094f 100644 --- a/fs/smb/server/mgmt/share_config.c +++ b/fs/smb/server/mgmt/share_config.c @@ -165,8 +165,12 @@ static struct ksmbd_share_config *share_config_request(struct unicode_map *um, share->path = kstrndup(ksmbd_share_config_path(resp), path_len, GFP_KERNEL); - if (share->path) + if (share->path) { share->path_sz = strlen(share->path); + while (share->path_sz > 1 && + share->path[share->path_sz - 1] == '/') + share->path[--share->path_sz] = '\0'; + } share->create_mask = resp->create_mask; share->directory_mask = resp->directory_mask; share->force_create_mode = resp->force_create_mode; -- cgit v1.2.3-59-g8ed1b From d293ece108104967ec3465f253834c0511170f04 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 22 May 2024 19:53:03 -0400 Subject: bcachefs: Fix shutdown ordering the btree key cache uses the srcu struct created/destroyed by btree_iter.c; btree_iter needs to be exited last. Reported-by: syzbot+3af9daea347788b15213@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/printbuf.c | 7 +++++++ fs/bcachefs/super.c | 2 +- 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/printbuf.c b/fs/bcachefs/printbuf.c index 8b0369185f5c..9f529e4c1b16 100644 --- a/fs/bcachefs/printbuf.c +++ b/fs/bcachefs/printbuf.c @@ -45,6 +45,13 @@ int bch2_printbuf_make_room(struct printbuf *out, unsigned extra) unsigned new_size = roundup_pow_of_two(out->size + extra); + /* Sanity check... */ + if (new_size > PAGE_SIZE << MAX_PAGE_ORDER) { + out->allocation_failure = true; + out->overflow = true; + return -ENOMEM; + } + /* * Note: output buffer must be freeable with kfree(), it's not required * that the user use printbuf_exit(). diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 294a9d35a9f2..2206a8dee693 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -551,9 +551,9 @@ static void __bch2_fs_free(struct bch_fs *c) bch2_fs_io_read_exit(c); bch2_fs_buckets_waiting_for_journal_exit(c); bch2_fs_btree_interior_update_exit(c); - bch2_fs_btree_iter_exit(c); bch2_fs_btree_key_cache_exit(&c->btree_key_cache); bch2_fs_btree_cache_exit(c); + bch2_fs_btree_iter_exit(c); bch2_fs_replicas_exit(c); bch2_fs_journal_exit(&c->journal); bch2_io_clock_exit(&c->io_clock[WRITE]); -- cgit v1.2.3-59-g8ed1b From 2c92ca849fcc6ee7d0c358e9959abc9f58661aea Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 16 May 2024 13:34:54 -0400 Subject: tracing/treewide: Remove second parameter of __assign_str() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With the rework of how the __string() handles dynamic strings where it saves off the source string in field in the helper structure[1], the assignment of that value to the trace event field is stored in the helper value and does not need to be passed in again. This means that with: __string(field, mystring) Which use to be assigned with __assign_str(field, mystring), no longer needs the second parameter and it is unused. With this, __assign_str() will now only get a single parameter. There's over 700 users of __assign_str() and because coccinelle does not handle the TRACE_EVENT() macro I ended up using the following sed script: git grep -l __assign_str | while read a ; do sed -e 's/\(__assign_str([^,]*[^ ,]\) *,[^;]*/\1)/' $a > /tmp/test-file; mv /tmp/test-file $a; done I then searched for __assign_str() that did not end with ';' as those were multi line assignments that the sed script above would fail to catch. Note, the same updates will need to be done for: __assign_str_len() __assign_rel_str() __assign_rel_str_len() I tested this with both an allmodconfig and an allyesconfig (build only for both). [1] https://lore.kernel.org/linux-trace-kernel/20240222211442.634192653@goodmis.org/ Link: https://lore.kernel.org/linux-trace-kernel/20240516133454.681ba6a0@rorschach.local.home Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Cc: Linus Torvalds Cc: Julia Lawall Signed-off-by: Steven Rostedt (Google) Acked-by: Jani Nikula Acked-by: Christian König for the amdgpu parts. Acked-by: Thomas Hellström #for Acked-by: Rafael J. Wysocki # for thermal Acked-by: Takashi Iwai Acked-by: Darrick J. Wong # xfs Tested-by: Guenter Roeck --- arch/arm64/kernel/trace-events-emulation.h | 2 +- arch/powerpc/include/asm/trace.h | 4 +- arch/x86/kvm/trace.h | 2 +- drivers/base/regmap/trace.h | 18 ++-- drivers/base/trace.h | 2 +- drivers/block/rnbd/rnbd-srv-trace.h | 12 +-- drivers/bus/mhi/host/trace.h | 12 +-- drivers/cxl/core/trace.h | 32 +++--- drivers/dma-buf/sync_trace.h | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h | 16 +-- .../drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h | 2 +- drivers/gpu/drm/i915/display/intel_display_trace.h | 56 +++++----- drivers/gpu/drm/lima/lima_trace.h | 2 +- drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h | 12 +-- drivers/gpu/drm/scheduler/gpu_scheduler_trace.h | 4 +- drivers/gpu/drm/virtio/virtgpu_trace.h | 2 +- drivers/infiniband/core/cma_trace.h | 4 +- drivers/infiniband/hw/hfi1/hfi.h | 2 +- drivers/infiniband/hw/hfi1/trace_dbg.h | 2 +- drivers/infiniband/hw/hfi1/trace_rx.h | 2 +- drivers/infiniband/hw/hfi1/trace_tid.h | 4 +- drivers/infiniband/hw/hfi1/trace_tx.h | 4 +- drivers/infiniband/sw/rdmavt/trace.h | 2 +- drivers/infiniband/sw/rdmavt/trace_rvt.h | 2 +- drivers/interconnect/trace.h | 10 +- drivers/iommu/intel/trace.h | 14 +-- drivers/media/platform/nvidia/tegra-vde/trace.h | 2 +- drivers/misc/mei/mei-trace.h | 6 +- drivers/net/dsa/mv88e6xxx/trace.h | 4 +- .../net/ethernet/freescale/dpaa/dpaa_eth_trace.h | 2 +- .../net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h | 4 +- .../net/ethernet/fungible/funeth/funeth_trace.h | 6 +- drivers/net/ethernet/hisilicon/hns3/hns3_trace.h | 4 +- .../ethernet/hisilicon/hns3/hns3pf/hclge_trace.h | 12 +-- .../ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h | 10 +- drivers/net/ethernet/intel/i40e/i40e_trace.h | 10 +- drivers/net/ethernet/intel/iavf/iavf_trace.h | 6 +- drivers/net/ethernet/intel/ice/ice_trace.h | 12 +-- .../net/ethernet/marvell/octeontx2/af/rvu_trace.h | 12 +-- .../mellanox/mlx5/core/diag/cmd_tracepoint.h | 4 +- .../mellanox/mlx5/core/diag/en_rep_tracepoint.h | 2 +- .../mellanox/mlx5/core/diag/en_tc_tracepoint.h | 2 +- .../mellanox/mlx5/core/diag/fw_tracer_tracepoint.h | 5 +- .../mellanox/mlx5/core/esw/diag/qos_tracepoint.h | 8 +- .../mlx5/core/sf/dev/diag/dev_tracepoint.h | 2 +- .../mellanox/mlx5/core/sf/diag/sf_tracepoint.h | 14 +-- .../mellanox/mlx5/core/sf/diag/vhca_tracepoint.h | 2 +- drivers/net/fjes/fjes_trace.h | 10 +- drivers/net/hyperv/netvsc_trace.h | 8 +- drivers/net/wireless/ath/ath10k/trace.h | 64 +++++------ drivers/net/wireless/ath/ath11k/trace.h | 44 ++++---- drivers/net/wireless/ath/ath12k/trace.h | 16 +-- drivers/net/wireless/ath/ath6kl/trace.h | 4 +- drivers/net/wireless/ath/trace.h | 4 +- .../broadcom/brcm80211/brcmfmac/tracepoint.h | 4 +- .../brcm80211/brcmsmac/brcms_trace_brcmsmac.h | 2 +- .../brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h | 2 +- .../brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h | 6 +- .../net/wireless/intel/iwlwifi/iwl-devtrace-msg.h | 2 +- drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h | 2 +- drivers/soc/qcom/pmic_pdcharger_ulog.h | 2 +- drivers/soc/qcom/trace-aoss.h | 4 +- drivers/soc/qcom/trace-rpmh.h | 4 +- drivers/thermal/thermal_trace.h | 10 +- drivers/usb/cdns3/cdns3-trace.h | 26 ++--- drivers/usb/cdns3/cdnsp-trace.h | 10 +- drivers/usb/chipidea/trace.h | 4 +- drivers/usb/dwc3/trace.h | 8 +- drivers/usb/gadget/udc/cdns2/cdns2-trace.h | 22 ++-- drivers/usb/gadget/udc/trace.h | 4 +- drivers/usb/mtu3/mtu3_trace.h | 8 +- drivers/usb/musb/musb_trace.h | 12 +-- fs/bcachefs/trace.h | 6 +- fs/nfs/nfs4trace.h | 42 ++++---- fs/nfs/nfstrace.h | 41 ++++--- fs/nfsd/trace.h | 40 +++---- fs/ocfs2/ocfs2_trace.h | 60 +++++------ fs/smb/client/trace.h | 18 ++-- fs/xfs/scrub/trace.h | 10 +- fs/xfs/xfs_trace.h | 28 ++--- include/ras/ras_event.h | 12 +-- include/trace/events/asoc.h | 22 ++-- include/trace/events/avc.h | 6 +- include/trace/events/bridge.h | 16 +-- include/trace/events/btrfs.h | 6 +- include/trace/events/cgroup.h | 10 +- include/trace/events/clk.h | 18 ++-- include/trace/events/cma.h | 8 +- include/trace/events/devfreq.h | 4 +- include/trace/events/devlink.h | 50 ++++----- include/trace/events/dma_fence.h | 4 +- include/trace/events/erofs.h | 2 +- include/trace/events/f2fs.h | 20 ++-- include/trace/events/habanalabs.h | 10 +- include/trace/events/huge_memory.h | 4 +- include/trace/events/hwmon.h | 6 +- include/trace/events/initcall.h | 2 +- include/trace/events/intel_ish.h | 2 +- include/trace/events/io_uring.h | 14 +-- include/trace/events/iocost.h | 14 +-- include/trace/events/iommu.h | 8 +- include/trace/events/irq.h | 2 +- include/trace/events/iscsi.h | 2 +- include/trace/events/kmem.h | 2 +- include/trace/events/lock.h | 4 +- include/trace/events/mmap_lock.h | 4 +- include/trace/events/mmc.h | 4 +- include/trace/events/module.h | 8 +- include/trace/events/napi.h | 2 +- include/trace/events/neigh.h | 6 +- include/trace/events/net.h | 12 +-- include/trace/events/netlink.h | 2 +- include/trace/events/oom.h | 2 +- include/trace/events/osnoise.h | 2 +- include/trace/events/power.h | 23 ++-- include/trace/events/pwc.h | 4 +- include/trace/events/qdisc.h | 12 +-- include/trace/events/qla.h | 2 +- include/trace/events/qrtr.h | 2 +- include/trace/events/regulator.h | 6 +- include/trace/events/rpcgss.h | 20 ++-- include/trace/events/rpcrdma.h | 52 ++++----- include/trace/events/rpm.h | 6 +- include/trace/events/sched.h | 8 +- include/trace/events/sof.h | 12 +-- include/trace/events/sof_intel.h | 16 +-- include/trace/events/sunrpc.h | 118 ++++++++++----------- include/trace/events/swiotlb.h | 2 +- include/trace/events/target.h | 4 +- include/trace/events/tegra_apb_dma.h | 6 +- include/trace/events/ufs.h | 24 ++--- include/trace/events/workqueue.h | 2 +- include/trace/events/xdp.h | 2 +- include/trace/stages/stage6_event_callback.h | 4 +- kernel/trace/bpf_trace.h | 2 +- net/batman-adv/trace.h | 4 +- net/dsa/trace.h | 34 +++--- net/ieee802154/trace.h | 2 +- net/mac80211/trace.h | 2 +- net/openvswitch/openvswitch_trace.h | 8 +- net/smc/smc_tracepoint.h | 4 +- net/tipc/trace.h | 16 +-- net/wireless/trace.h | 2 +- samples/trace_events/trace-events-sample.h | 19 ++-- sound/core/pcm_trace.h | 2 +- sound/hda/trace.h | 6 +- sound/soc/intel/avs/trace.h | 4 +- 147 files changed, 794 insertions(+), 808 deletions(-) (limited to 'fs') diff --git a/arch/arm64/kernel/trace-events-emulation.h b/arch/arm64/kernel/trace-events-emulation.h index 6c40f58b844a..c51b547b583e 100644 --- a/arch/arm64/kernel/trace-events-emulation.h +++ b/arch/arm64/kernel/trace-events-emulation.h @@ -18,7 +18,7 @@ TRACE_EVENT(instruction_emulation, ), TP_fast_assign( - __assign_str(instr, instr); + __assign_str(instr); __entry->addr = addr; ), diff --git a/arch/powerpc/include/asm/trace.h b/arch/powerpc/include/asm/trace.h index d9ac3a4f46e1..a7b69b25296b 100644 --- a/arch/powerpc/include/asm/trace.h +++ b/arch/powerpc/include/asm/trace.h @@ -137,7 +137,7 @@ TRACE_EVENT(rtas_input, TP_fast_assign( __entry->nargs = be32_to_cpu(rtas_args->nargs); - __assign_str(name, name); + __assign_str(name); be32_to_cpu_array(__get_dynamic_array(inputs), rtas_args->args, __entry->nargs); ), @@ -162,7 +162,7 @@ TRACE_EVENT(rtas_output, TP_fast_assign( __entry->nr_other = be32_to_cpu(rtas_args->nret) - 1; __entry->status = be32_to_cpu(rtas_args->rets[0]); - __assign_str(name, name); + __assign_str(name); be32_to_cpu_array(__get_dynamic_array(other_outputs), &rtas_args->rets[1], __entry->nr_other); ), diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h index 9d0b02ef307e..e19fed438a67 100644 --- a/arch/x86/kvm/trace.h +++ b/arch/x86/kvm/trace.h @@ -1678,7 +1678,7 @@ TRACE_EVENT(kvm_nested_vmenter_failed, ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); __entry->err = err; ), diff --git a/drivers/base/regmap/trace.h b/drivers/base/regmap/trace.h index 704e106e5dbd..bcc5a8b226a6 100644 --- a/drivers/base/regmap/trace.h +++ b/drivers/base/regmap/trace.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(regmap_reg, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->val = val; ), @@ -74,7 +74,7 @@ DECLARE_EVENT_CLASS(regmap_bulk, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->val_len = val_len; memcpy(__get_dynamic_array(buf), val, val_len); @@ -113,7 +113,7 @@ DECLARE_EVENT_CLASS(regmap_block, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->count = count; ), @@ -163,9 +163,9 @@ TRACE_EVENT(regcache_sync, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); - __assign_str(status, status); - __assign_str(type, type); + __assign_str(name); + __assign_str(status); + __assign_str(type); ), TP_printk("%s type=%s status=%s", __get_str(name), @@ -184,7 +184,7 @@ DECLARE_EVENT_CLASS(regmap_bool, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->flag = flag; ), @@ -216,7 +216,7 @@ DECLARE_EVENT_CLASS(regmap_async, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -264,7 +264,7 @@ TRACE_EVENT(regcache_drop_region, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->from = from; __entry->to = to; ), diff --git a/drivers/base/trace.h b/drivers/base/trace.h index 3192e18f877e..e52b6eae060d 100644 --- a/drivers/base/trace.h +++ b/drivers/base/trace.h @@ -28,7 +28,7 @@ DECLARE_EVENT_CLASS(devres, __field(size_t, size) ), TP_fast_assign( - __assign_str(devname, dev_name(dev)); + __assign_str(devname); __entry->op = op; __entry->node = node; __entry->name = name; diff --git a/drivers/block/rnbd/rnbd-srv-trace.h b/drivers/block/rnbd/rnbd-srv-trace.h index 8dedf73bdd28..89d0bcb17195 100644 --- a/drivers/block/rnbd/rnbd-srv-trace.h +++ b/drivers/block/rnbd/rnbd-srv-trace.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(rnbd_srv_link_class, TP_fast_assign( __entry->qdepth = srv->queue_depth; - __assign_str(sessname, srv->sessname); + __assign_str(sessname); ), TP_printk("sessname: %s qdepth: %d", @@ -85,7 +85,7 @@ TRACE_EVENT(process_rdma, ), TP_fast_assign( - __assign_str(sessname, srv->sessname); + __assign_str(sessname); __entry->dir = id->dir; __entry->ver = srv->ver; __entry->device_id = le32_to_cpu(msg->device_id); @@ -130,7 +130,7 @@ TRACE_EVENT(process_msg_sess_info, __entry->proto_ver = srv->ver; __entry->clt_ver = msg->ver; __entry->srv_ver = RNBD_PROTO_VER_MAJOR; - __assign_str(sessname, srv->sessname); + __assign_str(sessname); ), TP_printk("Session %s using proto-ver %d (clt-ver: %d, srv-ver: %d)", @@ -165,8 +165,8 @@ TRACE_EVENT(process_msg_open, TP_fast_assign( __entry->access_mode = msg->access_mode; - __assign_str(sessname, srv->sessname); - __assign_str(dev_name, msg->dev_name); + __assign_str(sessname); + __assign_str(dev_name); ), TP_printk("Open message received: session='%s' path='%s' access_mode=%s", @@ -189,7 +189,7 @@ TRACE_EVENT(process_msg_close, TP_fast_assign( __entry->device_id = le32_to_cpu(msg->device_id); - __assign_str(sessname, srv->sessname); + __assign_str(sessname); ), TP_printk("Close message received: session='%s' device id='%d'", diff --git a/drivers/bus/mhi/host/trace.h b/drivers/bus/mhi/host/trace.h index 368515dcb22d..95613c8ebe06 100644 --- a/drivers/bus/mhi/host/trace.h +++ b/drivers/bus/mhi/host/trace.h @@ -103,7 +103,7 @@ TRACE_EVENT(mhi_gen_tre, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->ch_num = mhi_chan->chan; __entry->wp = mhi_tre; __entry->tre_ptr = mhi_tre->ptr; @@ -131,7 +131,7 @@ TRACE_EVENT(mhi_intvec_states, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->local_ee = mhi_cntrl->ee; __entry->state = mhi_cntrl->dev_state; __entry->dev_ee = dev_ee; @@ -158,7 +158,7 @@ TRACE_EVENT(mhi_tryset_pm_state, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); if (pm_state) pm_state = __fls(pm_state); __entry->pm_state = pm_state; @@ -184,7 +184,7 @@ DECLARE_EVENT_CLASS(mhi_process_event_ring, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->rp = rp; __entry->ptr = rp->ptr; __entry->dword0 = rp->dword[0]; @@ -226,7 +226,7 @@ DECLARE_EVENT_CLASS(mhi_update_channel_state, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->ch_num = mhi_chan->chan; __entry->state = state; __entry->reason = reason; @@ -265,7 +265,7 @@ TRACE_EVENT(mhi_pm_st_transition, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->state = state; ), diff --git a/drivers/cxl/core/trace.h b/drivers/cxl/core/trace.h index 07a0394b1d99..ee5cd4eb2f16 100644 --- a/drivers/cxl/core/trace.h +++ b/drivers/cxl/core/trace.h @@ -60,8 +60,8 @@ TRACE_EVENT(cxl_aer_uncorrectable_error, __array(u32, header_log, CXL_HEADERLOG_SIZE_U32) ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->status = status; __entry->first_error = fe; @@ -106,8 +106,8 @@ TRACE_EVENT(cxl_aer_correctable_error, __field(u32, status) ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->status = status; ), @@ -142,8 +142,8 @@ TRACE_EVENT(cxl_overflow, ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->log = log; __entry->count = le16_to_cpu(payload->overflow_err_count); @@ -200,8 +200,8 @@ TRACE_EVENT(cxl_overflow, __field(u8, hdr_maint_op_class) #define CXL_EVT_TP_fast_assign(cxlmd, l, hdr) \ - __assign_str(memdev, dev_name(&(cxlmd)->dev)); \ - __assign_str(host, dev_name((cxlmd)->dev.parent)); \ + __assign_str(memdev); \ + __assign_str(host); \ __entry->log = (l); \ __entry->serial = (cxlmd)->cxlds->serial; \ __entry->hdr_length = (hdr).length; \ @@ -359,10 +359,10 @@ TRACE_EVENT(cxl_general_media, __entry->validity_flags = get_unaligned_le16(&rec->validity_flags); __entry->hpa = hpa; if (cxlr) { - __assign_str(region_name, dev_name(&cxlr->dev)); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &cxlr->params.uuid); } else { - __assign_str(region_name, ""); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &uuid_null); } ), @@ -462,10 +462,10 @@ TRACE_EVENT(cxl_dram, CXL_EVENT_DER_CORRECTION_MASK_SIZE); __entry->hpa = hpa; if (cxlr) { - __assign_str(region_name, dev_name(&cxlr->dev)); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &cxlr->params.uuid); } else { - __assign_str(region_name, ""); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &uuid_null); } ), @@ -692,8 +692,8 @@ TRACE_EVENT(cxl_poison, ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->overflow_ts = cxl_poison_overflow(flags, overflow_ts); __entry->dpa = cxl_poison_record_dpa(record); @@ -702,12 +702,12 @@ TRACE_EVENT(cxl_poison, __entry->trace_type = trace_type; __entry->flags = flags; if (cxlr) { - __assign_str(region, dev_name(&cxlr->dev)); + __assign_str(region); memcpy(__entry->uuid, &cxlr->params.uuid, 16); __entry->hpa = cxl_trace_hpa(cxlr, cxlmd, __entry->dpa); } else { - __assign_str(region, ""); + __assign_str(region); memset(__entry->uuid, 0, 16); __entry->hpa = ULLONG_MAX; } diff --git a/drivers/dma-buf/sync_trace.h b/drivers/dma-buf/sync_trace.h index 06e468a218ff..d71dcf954b8d 100644 --- a/drivers/dma-buf/sync_trace.h +++ b/drivers/dma-buf/sync_trace.h @@ -20,7 +20,7 @@ TRACE_EVENT(sync_timeline, ), TP_fast_assign( - __assign_str(name, timeline->name); + __assign_str(name); __entry->value = timeline->value; ), diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h index f539b1d00234..7aafeb763e5d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h @@ -178,10 +178,10 @@ TRACE_EVENT(amdgpu_cs_ioctl, TP_fast_assign( __entry->sched_job_id = job->base.id; - __assign_str(timeline, AMDGPU_JOB_GET_TIMELINE_NAME(job)); + __assign_str(timeline); __entry->context = job->base.s_fence->finished.context; __entry->seqno = job->base.s_fence->finished.seqno; - __assign_str(ring, to_amdgpu_ring(job->base.sched)->name); + __assign_str(ring); __entry->num_ibs = job->num_ibs; ), TP_printk("sched_job=%llu, timeline=%s, context=%u, seqno=%u, ring_name=%s, num_ibs=%u", @@ -203,10 +203,10 @@ TRACE_EVENT(amdgpu_sched_run_job, TP_fast_assign( __entry->sched_job_id = job->base.id; - __assign_str(timeline, AMDGPU_JOB_GET_TIMELINE_NAME(job)); + __assign_str(timeline); __entry->context = job->base.s_fence->finished.context; __entry->seqno = job->base.s_fence->finished.seqno; - __assign_str(ring, to_amdgpu_ring(job->base.sched)->name); + __assign_str(ring); __entry->num_ibs = job->num_ibs; ), TP_printk("sched_job=%llu, timeline=%s, context=%u, seqno=%u, ring_name=%s, num_ibs=%u", @@ -231,7 +231,7 @@ TRACE_EVENT(amdgpu_vm_grab_id, TP_fast_assign( __entry->pasid = vm->pasid; - __assign_str(ring, ring->name); + __assign_str(ring); __entry->vmid = job->vmid; __entry->vm_hub = ring->vm_hub, __entry->pd_addr = job->vm_pd_addr; @@ -425,7 +425,7 @@ TRACE_EVENT(amdgpu_vm_flush, ), TP_fast_assign( - __assign_str(ring, ring->name); + __assign_str(ring); __entry->vmid = vmid; __entry->vm_hub = ring->vm_hub; __entry->pd_addr = pd_addr; @@ -526,7 +526,7 @@ TRACE_EVENT(amdgpu_ib_pipe_sync, ), TP_fast_assign( - __assign_str(ring, sched_job->base.sched->name); + __assign_str(ring); __entry->id = sched_job->base.id; __entry->fence = fence; __entry->ctx = fence->context; @@ -563,7 +563,7 @@ TRACE_EVENT(amdgpu_runpm_reference_dumps, ), TP_fast_assign( __entry->index = index; - __assign_str(func, func); + __assign_str(func); ), TP_printk("amdgpu runpm reference dump 0x%x: 0x%s\n", __entry->index, diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h index 133af994a08c..4686d4b0cbad 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h @@ -87,7 +87,7 @@ TRACE_EVENT(amdgpu_dc_performance, __entry->writes = write_count; __entry->read_delta = read_count - *last_read; __entry->write_delta = write_count - *last_write; - __assign_str(func, func); + __assign_str(func); __entry->line = line; *last_read = read_count; *last_write = write_count; diff --git a/drivers/gpu/drm/i915/display/intel_display_trace.h b/drivers/gpu/drm/i915/display/intel_display_trace.h index 7862e7cefe02..49a5e6d9dc0d 100644 --- a/drivers/gpu/drm/i915/display/intel_display_trace.h +++ b/drivers/gpu/drm/i915/display/intel_display_trace.h @@ -34,7 +34,7 @@ TRACE_EVENT(intel_pipe_enable, TP_fast_assign( struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_crtc *it__; - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); for_each_intel_crtc(&dev_priv->drm, it__) { __entry->frame[it__->pipe] = intel_crtc_get_vblank_counter(it__); __entry->scanline[it__->pipe] = intel_get_crtc_scanline(it__); @@ -63,7 +63,7 @@ TRACE_EVENT(intel_pipe_disable, TP_fast_assign( struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_crtc *it__; - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); for_each_intel_crtc(&dev_priv->drm, it__) { __entry->frame[it__->pipe] = intel_crtc_get_vblank_counter(it__); __entry->scanline[it__->pipe] = intel_get_crtc_scanline(it__); @@ -91,7 +91,7 @@ TRACE_EVENT(intel_pipe_crc, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -119,7 +119,7 @@ TRACE_EVENT(intel_cpu_fifo_underrun, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -144,7 +144,7 @@ TRACE_EVENT(intel_pch_fifo_underrun, TP_fast_assign( enum pipe pipe = pch_transcoder; struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); - __assign_str(dev, __dev_name_i915(dev_priv)); + __assign_str(dev); __entry->pipe = pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -169,7 +169,7 @@ TRACE_EVENT(intel_memory_cxsr, TP_fast_assign( struct intel_crtc *crtc; - __assign_str(dev, __dev_name_i915(dev_priv)); + __assign_str(dev); for_each_intel_crtc(&dev_priv->drm, crtc) { __entry->frame[crtc->pipe] = intel_crtc_get_vblank_counter(crtc); __entry->scanline[crtc->pipe] = intel_get_crtc_scanline(crtc); @@ -209,7 +209,7 @@ TRACE_EVENT(g4x_wm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -256,7 +256,7 @@ TRACE_EVENT(vlv_wm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -293,7 +293,7 @@ TRACE_EVENT(vlv_fifo_size, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -323,8 +323,8 @@ TRACE_EVENT(intel_plane_update_noarm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -354,8 +354,8 @@ TRACE_EVENT(intel_plane_update_arm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -383,8 +383,8 @@ TRACE_EVENT(intel_plane_disable_arm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -410,8 +410,8 @@ TRACE_EVENT(intel_fbc_activate, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(to_i915(plane->base.dev), plane->pipe); - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -437,8 +437,8 @@ TRACE_EVENT(intel_fbc_deactivate, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(to_i915(plane->base.dev), plane->pipe); - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -464,8 +464,8 @@ TRACE_EVENT(intel_fbc_nuke, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(to_i915(plane->base.dev), plane->pipe); - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -488,7 +488,7 @@ TRACE_EVENT(intel_crtc_vblank_work_start, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -511,7 +511,7 @@ TRACE_EVENT(intel_crtc_vblank_work_end, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -536,7 +536,7 @@ TRACE_EVENT(intel_pipe_update_start, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -564,7 +564,7 @@ TRACE_EVENT(intel_pipe_update_vblank_evaded, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = crtc->debug.start_vbl_count; __entry->scanline = crtc->debug.scanline_start; @@ -590,7 +590,7 @@ TRACE_EVENT(intel_pipe_update_end, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = frame; __entry->scanline = scanline_end; @@ -613,7 +613,7 @@ TRACE_EVENT(intel_frontbuffer_invalidate, ), TP_fast_assign( - __assign_str(dev, __dev_name_i915(i915)); + __assign_str(dev); __entry->frontbuffer_bits = frontbuffer_bits; __entry->origin = origin; ), @@ -634,7 +634,7 @@ TRACE_EVENT(intel_frontbuffer_flush, ), TP_fast_assign( - __assign_str(dev, __dev_name_i915(i915)); + __assign_str(dev); __entry->frontbuffer_bits = frontbuffer_bits; __entry->origin = origin; ), diff --git a/drivers/gpu/drm/lima/lima_trace.h b/drivers/gpu/drm/lima/lima_trace.h index 494b9790b1da..3a349d10304e 100644 --- a/drivers/gpu/drm/lima/lima_trace.h +++ b/drivers/gpu/drm/lima/lima_trace.h @@ -24,7 +24,7 @@ DECLARE_EVENT_CLASS(lima_task, __entry->task_id = task->base.id; __entry->context = task->base.s_fence->finished.context; __entry->seqno = task->base.s_fence->finished.seqno; - __assign_str(pipe, task->base.sched->name); + __assign_str(pipe); ), TP_printk("task=%llu, context=%u seqno=%u pipe=%s", diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h index bd92fb2979aa..0fdd41162e4b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h @@ -113,7 +113,7 @@ TRACE_EVENT(tracing_mark_write, ), TP_fast_assign( __entry->pid = pid; - __assign_str(trace_name, name); + __assign_str(trace_name); __entry->trace_begin = trace_begin; ), TP_printk("%s|%d|%s", __entry->trace_begin ? "B" : "E", @@ -130,7 +130,7 @@ TRACE_EVENT(dpu_trace_counter, ), TP_fast_assign( __entry->pid = current->tgid; - __assign_str(counter_name, name); + __assign_str(counter_name); __entry->value = value; ), TP_printk("%d|%s|%d", __entry->pid, @@ -379,7 +379,7 @@ TRACE_EVENT(dpu_enc_rc, __entry->sw_event = sw_event; __entry->idle_pc_supported = idle_pc_supported; __entry->rc_state = rc_state; - __assign_str(stage_str, stage); + __assign_str(stage_str); ), TP_printk("%s: id:%u, sw_event:%d, idle_pc_supported:%s, rc_state:%d", __get_str(stage_str), __entry->drm_id, __entry->sw_event, @@ -401,7 +401,7 @@ TRACE_EVENT(dpu_enc_frame_done_cb_not_busy, TP_fast_assign( __entry->drm_id = drm_id; __entry->event = event; - __assign_str(intf_mode_str, intf_mode); + __assign_str(intf_mode_str); __entry->intf_idx = intf_idx; __entry->wb_idx = wb_idx; ), @@ -446,7 +446,7 @@ TRACE_EVENT(dpu_enc_trigger_flush, ), TP_fast_assign( __entry->drm_id = drm_id; - __assign_str(intf_mode_str, intf_mode); + __assign_str(intf_mode_str); __entry->intf_idx = intf_idx; __entry->wb_idx = wb_idx; __entry->pending_kickoff_cnt = pending_kickoff_cnt; @@ -946,7 +946,7 @@ TRACE_EVENT(dpu_core_perf_update_clk, __field( u64, clk_rate ) ), TP_fast_assign( - __assign_str(dev_name, dev->unique); + __assign_str(dev_name); __entry->stop_req = stop_req; __entry->clk_rate = clk_rate; ), diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h index f8ed093b7356..c75302ca3427 100644 --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h @@ -48,7 +48,7 @@ DECLARE_EVENT_CLASS(drm_sched_job, __entry->entity = entity; __entry->id = sched_job->id; __entry->fence = &sched_job->s_fence->finished; - __assign_str(name, sched_job->sched->name); + __assign_str(name); __entry->job_count = spsc_queue_count(&entity->job_queue); __entry->hw_job_count = atomic_read( &sched_job->sched->credit_count); @@ -94,7 +94,7 @@ TRACE_EVENT(drm_sched_job_wait_dep, ), TP_fast_assign( - __assign_str(name, sched_job->sched->name); + __assign_str(name); __entry->id = sched_job->id; __entry->fence = fence; __entry->ctx = fence->context; diff --git a/drivers/gpu/drm/virtio/virtgpu_trace.h b/drivers/gpu/drm/virtio/virtgpu_trace.h index 031bc77689d5..227bf0ae7ed5 100644 --- a/drivers/gpu/drm/virtio/virtgpu_trace.h +++ b/drivers/gpu/drm/virtio/virtgpu_trace.h @@ -25,7 +25,7 @@ DECLARE_EVENT_CLASS(virtio_gpu_cmd, TP_fast_assign( __entry->dev = vq->vdev->index; __entry->vq = vq->index; - __assign_str(name, vq->name); + __assign_str(name); __entry->type = le32_to_cpu(hdr->type); __entry->flags = le32_to_cpu(hdr->flags); __entry->fence_id = le64_to_cpu(hdr->fence_id); diff --git a/drivers/infiniband/core/cma_trace.h b/drivers/infiniband/core/cma_trace.h index 47f3c6e4be89..dc622f3778be 100644 --- a/drivers/infiniband/core/cma_trace.h +++ b/drivers/infiniband/core/cma_trace.h @@ -84,7 +84,7 @@ TRACE_EVENT(cm_id_attach, sizeof(struct sockaddr_in6)); memcpy(__entry->dstaddr, &id_priv->id.route.addr.dst_addr, sizeof(struct sockaddr_in6)); - __assign_str(devname, device->name); + __assign_str(devname); ), TP_printk("cm.id=%u src=%pISpc dst=%pISpc device=%s", @@ -334,7 +334,7 @@ DECLARE_EVENT_CLASS(cma_client_class, ), TP_fast_assign( - __assign_str(name, device->name); + __assign_str(name); ), TP_printk("device name=%s", diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h index 4b3f1cb125fc..eb38f81aeeb1 100644 --- a/drivers/infiniband/hw/hfi1/hfi.h +++ b/drivers/infiniband/hw/hfi1/hfi.h @@ -2425,7 +2425,7 @@ static inline bool hfi1_need_drop(struct hfi1_devdata *dd) int hfi1_tempsense_rd(struct hfi1_devdata *dd, struct hfi1_temp *temp); #define DD_DEV_ENTRY(dd) __string(dev, dev_name(&(dd)->pcidev->dev)) -#define DD_DEV_ASSIGN(dd) __assign_str(dev, dev_name(&(dd)->pcidev->dev)) +#define DD_DEV_ASSIGN(dd) __assign_str(dev) static inline void hfi1_update_ah_attr(struct ib_device *ibdev, struct rdma_ah_attr *attr) diff --git a/drivers/infiniband/hw/hfi1/trace_dbg.h b/drivers/infiniband/hw/hfi1/trace_dbg.h index 75599d5168db..58304b91380f 100644 --- a/drivers/infiniband/hw/hfi1/trace_dbg.h +++ b/drivers/infiniband/hw/hfi1/trace_dbg.h @@ -33,7 +33,7 @@ DECLARE_EVENT_CLASS(hfi1_trace_template, TP_STRUCT__entry(__string(function, function) __vstring(msg, vaf->fmt, vaf->va) ), - TP_fast_assign(__assign_str(function, function); + TP_fast_assign(__assign_str(function); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("(%s) %s", diff --git a/drivers/infiniband/hw/hfi1/trace_rx.h b/drivers/infiniband/hw/hfi1/trace_rx.h index e6904aa80c00..8d5e12fe88a5 100644 --- a/drivers/infiniband/hw/hfi1/trace_rx.h +++ b/drivers/infiniband/hw/hfi1/trace_rx.h @@ -90,7 +90,7 @@ TRACE_EVENT(hfi1_mmu_invalidate, TP_fast_assign( __entry->ctxt = ctxt; __entry->subctxt = subctxt; - __assign_str(type, type); + __assign_str(type); __entry->start = start; __entry->end = end; ), diff --git a/drivers/infiniband/hw/hfi1/trace_tid.h b/drivers/infiniband/hw/hfi1/trace_tid.h index d129b8195959..e358f5b885fa 100644 --- a/drivers/infiniband/hw/hfi1/trace_tid.h +++ b/drivers/infiniband/hw/hfi1/trace_tid.h @@ -358,7 +358,7 @@ DECLARE_EVENT_CLASS(/* msg */ ), TP_fast_assign(/* assign */ __entry->qpn = qp ? qp->ibqp.qp_num : 0; - __assign_str(msg, msg); + __assign_str(msg); __entry->more = more; ), TP_printk(/* print */ @@ -651,7 +651,7 @@ DECLARE_EVENT_CLASS(/* tid_node */ TP_fast_assign(/* assign */ DD_DEV_ASSIGN(dd_from_ibdev(qp->ibqp.device)); __entry->qpn = qp->ibqp.qp_num; - __assign_str(msg, msg); + __assign_str(msg); __entry->index = index; __entry->base = base; __entry->map = map; diff --git a/drivers/infiniband/hw/hfi1/trace_tx.h b/drivers/infiniband/hw/hfi1/trace_tx.h index c79856d4fdfb..c0ba6b0a2c4e 100644 --- a/drivers/infiniband/hw/hfi1/trace_tx.h +++ b/drivers/infiniband/hw/hfi1/trace_tx.h @@ -740,8 +740,8 @@ TRACE_EVENT(hfi1_sdma_state, __string(newstate, nstate) ), TP_fast_assign(DD_DEV_ASSIGN(sde->dd); - __assign_str(curstate, cstate); - __assign_str(newstate, nstate); + __assign_str(curstate); + __assign_str(newstate); ), TP_printk("[%s] current state %s new state %s", __get_str(dev), diff --git a/drivers/infiniband/sw/rdmavt/trace.h b/drivers/infiniband/sw/rdmavt/trace.h index 4341965a5ea7..bdb6b9326b64 100644 --- a/drivers/infiniband/sw/rdmavt/trace.h +++ b/drivers/infiniband/sw/rdmavt/trace.h @@ -4,7 +4,7 @@ */ #define RDI_DEV_ENTRY(rdi) __string(dev, rvt_get_ibdev_name(rdi)) -#define RDI_DEV_ASSIGN(rdi) __assign_str(dev, rvt_get_ibdev_name(rdi)) +#define RDI_DEV_ASSIGN(rdi) __assign_str(dev) #include "trace_rvt.h" #include "trace_qp.h" diff --git a/drivers/infiniband/sw/rdmavt/trace_rvt.h b/drivers/infiniband/sw/rdmavt/trace_rvt.h index df33c2ca9710..a00489e66ddf 100644 --- a/drivers/infiniband/sw/rdmavt/trace_rvt.h +++ b/drivers/infiniband/sw/rdmavt/trace_rvt.h @@ -24,7 +24,7 @@ TRACE_EVENT(rvt_dbg, ), TP_fast_assign( RDI_DEV_ASSIGN(rdi); - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("[%s]: %s", __get_str(dev), __get_str(msg)) ); diff --git a/drivers/interconnect/trace.h b/drivers/interconnect/trace.h index 3d668ff566bf..206373546528 100644 --- a/drivers/interconnect/trace.h +++ b/drivers/interconnect/trace.h @@ -32,9 +32,9 @@ TRACE_EVENT(icc_set_bw, ), TP_fast_assign( - __assign_str(path_name, p->name); - __assign_str(dev, dev_name(p->reqs[i].dev)); - __assign_str(node_name, n->name); + __assign_str(path_name); + __assign_str(dev); + __assign_str(node_name); __entry->avg_bw = avg_bw; __entry->peak_bw = peak_bw; __entry->node_avg_bw = n->avg_bw; @@ -64,8 +64,8 @@ TRACE_EVENT(icc_set_bw_end, ), TP_fast_assign( - __assign_str(path_name, p->name); - __assign_str(dev, dev_name(p->reqs[0].dev)); + __assign_str(path_name); + __assign_str(dev); __entry->ret = ret; ), diff --git a/drivers/iommu/intel/trace.h b/drivers/iommu/intel/trace.h index 961ac1c1bc21..9defdae6ebae 100644 --- a/drivers/iommu/intel/trace.h +++ b/drivers/iommu/intel/trace.h @@ -32,7 +32,7 @@ TRACE_EVENT(qi_submit, ), TP_fast_assign( - __assign_str(iommu, iommu->name); + __assign_str(iommu); __entry->qw0 = qw0; __entry->qw1 = qw1; __entry->qw2 = qw2; @@ -79,8 +79,8 @@ TRACE_EVENT(prq_report, __entry->dw2 = dw2; __entry->dw3 = dw3; __entry->seq = seq; - __assign_str(iommu, iommu->name); - __assign_str(dev, dev_name(dev)); + __assign_str(iommu); + __assign_str(dev); ), TP_printk("%s/%s seq# %ld: %s", @@ -102,8 +102,8 @@ DECLARE_EVENT_CLASS(cache_tag_log, __field(u32, users) ), TP_fast_assign( - __assign_str(iommu, tag->iommu->name); - __assign_str(dev, dev_name(tag->dev)); + __assign_str(iommu); + __assign_str(dev); __entry->type = tag->type; __entry->domain_id = tag->domain_id; __entry->pasid = tag->pasid; @@ -152,8 +152,8 @@ DECLARE_EVENT_CLASS(cache_tag_flush, __field(unsigned long, mask) ), TP_fast_assign( - __assign_str(iommu, tag->iommu->name); - __assign_str(dev, dev_name(tag->dev)); + __assign_str(iommu); + __assign_str(dev); __entry->type = tag->type; __entry->domain_id = tag->domain_id; __entry->pasid = tag->pasid; diff --git a/drivers/media/platform/nvidia/tegra-vde/trace.h b/drivers/media/platform/nvidia/tegra-vde/trace.h index 7853ab095ca4..e8a75a7bd05d 100644 --- a/drivers/media/platform/nvidia/tegra-vde/trace.h +++ b/drivers/media/platform/nvidia/tegra-vde/trace.h @@ -20,7 +20,7 @@ DECLARE_EVENT_CLASS(register_access, __field(u32, value) ), TP_fast_assign( - __assign_str(hw_name, tegra_vde_reg_base_name(vde, base)); + __assign_str(hw_name); __entry->offset = offset; __entry->value = value; ), diff --git a/drivers/misc/mei/mei-trace.h b/drivers/misc/mei/mei-trace.h index fe46ff2b9d69..5312edbf5190 100644 --- a/drivers/misc/mei/mei-trace.h +++ b/drivers/misc/mei/mei-trace.h @@ -26,7 +26,7 @@ TRACE_EVENT(mei_reg_read, __field(u32, val) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->reg = reg; __entry->offs = offs; __entry->val = val; @@ -45,7 +45,7 @@ TRACE_EVENT(mei_reg_write, __field(u32, val) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->reg = reg; __entry->offs = offs; __entry->val = val; @@ -64,7 +64,7 @@ TRACE_EVENT(mei_pci_cfg_read, __field(u32, val) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->reg = reg; __entry->offs = offs; __entry->val = val; diff --git a/drivers/net/dsa/mv88e6xxx/trace.h b/drivers/net/dsa/mv88e6xxx/trace.h index f59ca04768e7..5bd015b2b97a 100644 --- a/drivers/net/dsa/mv88e6xxx/trace.h +++ b/drivers/net/dsa/mv88e6xxx/trace.h @@ -28,7 +28,7 @@ DECLARE_EVENT_CLASS(mv88e6xxx_atu_violation, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->spid = spid; __entry->portvec = portvec; memcpy(__entry->addr, addr, ETH_ALEN); @@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(mv88e6xxx_vtu_violation, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->spid = spid; __entry->vid = vid; ), diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h index 889f89df9930..6f0e58a2a58a 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h @@ -57,7 +57,7 @@ DECLARE_EVENT_CLASS(dpaa_eth_fd, __entry->fd_offset = qm_fd_get_offset(fd); __entry->fd_length = qm_fd_get_length(fd); __entry->fd_status = fd->status; - __assign_str(name, netdev->name); + __assign_str(name); ), /* This is what gets printed when the trace event is triggered */ diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h index 9b43fadb9b11..956767e0869c 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h @@ -48,7 +48,7 @@ DECLARE_EVENT_CLASS(dpaa2_eth_fd, __entry->fd_addr = dpaa2_fd_get_addr(fd); __entry->fd_len = dpaa2_fd_get_len(fd); __entry->fd_offset = dpaa2_fd_get_offset(fd); - __assign_str(name, netdev->name); + __assign_str(name); ), /* This is what gets printed when the trace event is @@ -144,7 +144,7 @@ DECLARE_EVENT_CLASS(dpaa2_eth_buf, __entry->dma_addr = dma_addr; __entry->map_size = map_size; __entry->bpid = bpid; - __assign_str(name, netdev->name); + __assign_str(name); ), /* This is what gets printed when the trace event is diff --git a/drivers/net/ethernet/fungible/funeth/funeth_trace.h b/drivers/net/ethernet/fungible/funeth/funeth_trace.h index 9e58dfec19d5..b9985900f30b 100644 --- a/drivers/net/ethernet/fungible/funeth/funeth_trace.h +++ b/drivers/net/ethernet/fungible/funeth/funeth_trace.h @@ -32,7 +32,7 @@ TRACE_EVENT(funeth_tx, __entry->len = len; __entry->sqe_idx = sqe_idx; __entry->ngle = ngle; - __assign_str(devname, txq->netdev->name); + __assign_str(devname); ), TP_printk("%s: Txq %u, SQE idx %u, len %u, num GLEs %u", @@ -62,7 +62,7 @@ TRACE_EVENT(funeth_tx_free, __entry->sqe_idx = sqe_idx; __entry->num_sqes = num_sqes; __entry->hw_head = hw_head; - __assign_str(devname, txq->netdev->name); + __assign_str(devname); ), TP_printk("%s: Txq %u, SQE idx %u, SQEs %u, HW head %u", @@ -97,7 +97,7 @@ TRACE_EVENT(funeth_rx, __entry->len = pkt_len; __entry->hash = hash; __entry->cls_vec = cls_vec; - __assign_str(devname, rxq->netdev->name); + __assign_str(devname); ), TP_printk("%s: Rxq %u, CQ head %u, RQEs %u, len %u, hash %u, CV %#x", diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h index b8a1ecb4b8fb..3362b8d14d4f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h @@ -84,7 +84,7 @@ TRACE_EVENT(hns3_tx_desc, __entry->desc_dma = ring->desc_dma_addr, memcpy(__entry->desc, &ring->desc[cur_ntu], sizeof(struct hns3_desc)); - __assign_str(devname, ring->tqp->handle->kinfo.netdev->name); + __assign_str(devname); ), TP_printk( @@ -117,7 +117,7 @@ TRACE_EVENT(hns3_rx_desc, __entry->buf_dma = ring->desc_cb[ring->next_to_clean].dma; memcpy(__entry->desc, &ring->desc[ring->next_to_clean], sizeof(struct hns3_desc)); - __assign_str(devname, ring->tqp->handle->kinfo.netdev->name); + __assign_str(devname); ), TP_printk( diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h index 7e47f0c21d88..7103cf04bffc 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h @@ -33,8 +33,8 @@ TRACE_EVENT(hclge_pf_mbx_get, __entry->vfid = req->mbx_src_vfid; __entry->code = req->msg.code; __entry->subcode = req->msg.subcode; - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->vport[0].nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_vf_to_pf_cmd)); ), @@ -64,8 +64,8 @@ TRACE_EVENT(hclge_pf_mbx_send, TP_fast_assign( __entry->vfid = req->dest_vfid; __entry->code = le16_to_cpu(req->msg.code); - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->vport[0].nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_pf_to_vf_cmd)); ), @@ -101,7 +101,7 @@ DECLARE_EVENT_CLASS(hclge_pf_cmd_template, __entry->rsv = le16_to_cpu(desc->rsv); __entry->index = index; __entry->num = num; - __assign_str(pciname, pci_name(hw->cmq.csq.pdev)); + __assign_str(pciname); for (i = 0; i < HCLGE_DESC_DATA_LEN; i++) __entry->data[i] = le32_to_cpu(desc->data[i]);), @@ -144,7 +144,7 @@ DECLARE_EVENT_CLASS(hclge_pf_special_cmd_template, TP_fast_assign(int i; __entry->index = index; __entry->num = num; - __assign_str(pciname, pci_name(hw->cmq.csq.pdev)); + __assign_str(pciname); for (i = 0; i < PF_DESC_LEN; i++) __entry->data[i] = le32_to_cpu(data[i]); ), diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h index e2e3a2602b6a..66b084309c91 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h @@ -30,8 +30,8 @@ TRACE_EVENT(hclge_vf_mbx_get, TP_fast_assign( __entry->vfid = req->dest_vfid; __entry->code = le16_to_cpu(req->msg.code); - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_pf_to_vf_cmd)); ), @@ -63,8 +63,8 @@ TRACE_EVENT(hclge_vf_mbx_send, __entry->vfid = req->mbx_src_vfid; __entry->code = req->msg.code; __entry->subcode = req->msg.subcode; - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_vf_to_pf_cmd)); ), @@ -101,7 +101,7 @@ DECLARE_EVENT_CLASS(hclge_vf_cmd_template, __entry->rsv = le16_to_cpu(desc->rsv); __entry->index = index; __entry->num = num; - __assign_str(pciname, pci_name(hw->cmq.csq.pdev)); + __assign_str(pciname); for (i = 0; i < HCLGE_DESC_DATA_LEN; i++) __entry->data[i] = le32_to_cpu(desc->data[i]);), diff --git a/drivers/net/ethernet/intel/i40e/i40e_trace.h b/drivers/net/ethernet/intel/i40e/i40e_trace.h index 33b4e30f5e00..759f3d1c4c8f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_trace.h +++ b/drivers/net/ethernet/intel/i40e/i40e_trace.h @@ -89,8 +89,8 @@ TRACE_EVENT(i40e_napi_poll, __entry->tx_clean_complete = tx_clean_complete; __entry->irq_num = q->irq_num; __entry->curr_cpu = get_cpu(); - __assign_str(qname, q->name); - __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); + __assign_str(qname); + __assign_str(dev_name); __assign_bitmask(irq_affinity, cpumask_bits(&q->affinity_mask), nr_cpumask_bits); ), @@ -132,7 +132,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->buf = buf; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -177,7 +177,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->xdp = xdp; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -219,7 +219,7 @@ DECLARE_EVENT_CLASS( TP_fast_assign( __entry->skb = skb; __entry->ring = ring; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( diff --git a/drivers/net/ethernet/intel/iavf/iavf_trace.h b/drivers/net/ethernet/intel/iavf/iavf_trace.h index 82fda6f5abf0..62212011c807 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_trace.h +++ b/drivers/net/ethernet/intel/iavf/iavf_trace.h @@ -83,7 +83,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->buf = buf; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -128,7 +128,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->skb = skb; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -170,7 +170,7 @@ DECLARE_EVENT_CLASS( TP_fast_assign( __entry->skb = skb; __entry->ring = ring; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( diff --git a/drivers/net/ethernet/intel/ice/ice_trace.h b/drivers/net/ethernet/intel/ice/ice_trace.h index b2f5c9fe0149..244cddd2a9ea 100644 --- a/drivers/net/ethernet/intel/ice/ice_trace.h +++ b/drivers/net/ethernet/intel/ice/ice_trace.h @@ -69,7 +69,7 @@ DECLARE_EVENT_CLASS(ice_rx_dim_template, TP_fast_assign(__entry->q_vector = q_vector; __entry->dim = dim; - __assign_str(devname, q_vector->rx.rx_ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s Rx-Q: %d dim-state: %d dim-profile: %d dim-tune: %d dim-st-right: %d dim-st-left: %d dim-tired: %d", __get_str(devname), @@ -96,7 +96,7 @@ DECLARE_EVENT_CLASS(ice_tx_dim_template, TP_fast_assign(__entry->q_vector = q_vector; __entry->dim = dim; - __assign_str(devname, q_vector->tx.tx_ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s Tx-Q: %d dim-state: %d dim-profile: %d dim-tune: %d dim-st-right: %d dim-st-left: %d dim-tired: %d", __get_str(devname), @@ -128,7 +128,7 @@ DECLARE_EVENT_CLASS(ice_tx_template, TP_fast_assign(__entry->ring = ring; __entry->desc = desc; __entry->buf = buf; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s ring: %pK desc: %pK buf %pK", __get_str(devname), __entry->ring, __entry->desc, __entry->buf) @@ -156,7 +156,7 @@ DECLARE_EVENT_CLASS(ice_rx_template, TP_fast_assign(__entry->ring = ring; __entry->desc = desc; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s ring: %pK desc: %pK", __get_str(devname), __entry->ring, __entry->desc) @@ -180,7 +180,7 @@ DECLARE_EVENT_CLASS(ice_rx_indicate_template, TP_fast_assign(__entry->ring = ring; __entry->desc = desc; __entry->skb = skb; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s ring: %pK desc: %pK skb %pK", __get_str(devname), __entry->ring, __entry->desc, __entry->skb) @@ -203,7 +203,7 @@ DECLARE_EVENT_CLASS(ice_xmit_template, TP_fast_assign(__entry->ring = ring; __entry->skb = skb; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s skb: %pK ring: %pK", __get_str(devname), __entry->skb, __entry->ring) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h index 28984d0e848a..5704520f9b02 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h @@ -24,7 +24,7 @@ TRACE_EVENT(otx2_msg_alloc, __field(u16, id) __field(u64, size) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->id = id; __entry->size = size; ), @@ -39,7 +39,7 @@ TRACE_EVENT(otx2_msg_send, __field(u16, num_msgs) __field(u64, msg_size) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->num_msgs = num_msgs; __entry->msg_size = msg_size; ), @@ -55,7 +55,7 @@ TRACE_EVENT(otx2_msg_check, __field(u16, rspid) __field(int, rc) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->reqid = reqid; __entry->rspid = rspid; __entry->rc = rc; @@ -72,8 +72,8 @@ TRACE_EVENT(otx2_msg_interrupt, __string(str, msg) __field(u64, intr) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); - __assign_str(str, msg); + TP_fast_assign(__assign_str(dev); + __assign_str(str); __entry->intr = intr; ), TP_printk("[%s] mbox interrupt %s (0x%llx)\n", __get_str(dev), @@ -87,7 +87,7 @@ TRACE_EVENT(otx2_msg_process, __field(u16, id) __field(int, err) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->id = id; __entry->err = err; ), diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h index 406ebe17405f..b4b3a43e56a0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h @@ -22,10 +22,10 @@ TRACE_EVENT(mlx5_cmd, __field(u32, syndrome) __field(int, err) ), - TP_fast_assign(__assign_str(command_str, command_str); + TP_fast_assign(__assign_str(command_str); __entry->opcode = opcode; __entry->op_mod = op_mod; - __assign_str(status_str, status_str); + __assign_str(status_str); __entry->status = status; __entry->syndrome = syndrome; __entry->err = err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h index f15718db5d0e..78e481b2c015 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h @@ -25,7 +25,7 @@ TRACE_EVENT(mlx5e_rep_neigh_update, struct in6_addr *pin6; __be32 *p32; - __assign_str(devname, nhe->neigh_dev->name); + __assign_str(devname); __entry->neigh_connected = neigh_connected; memcpy(__entry->ha, ha, ETH_ALEN); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h index ac52ef37f38a..4b1ca228012b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h @@ -86,7 +86,7 @@ TRACE_EVENT(mlx5e_tc_update_neigh_used_value, struct in6_addr *pin6; __be32 *p32; - __assign_str(devname, nhe->neigh_dev->name); + __assign_str(devname); __entry->neigh_used = neigh_used; p32 = (__be32 *)__entry->v4; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h index 3038be575923..50f8a7630f86 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h @@ -55,12 +55,11 @@ TRACE_EVENT(mlx5_fw, ), TP_fast_assign( - __assign_str(dev_name, - dev_name(tracer->dev->device)); + __assign_str(dev_name); __entry->trace_timestamp = trace_timestamp; __entry->lost = lost; __entry->event_id = event_id; - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("%s [0x%llx] %d [0x%x] %s", diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h index 458baf0c6415..1ce332f21ebe 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h @@ -17,7 +17,7 @@ TRACE_EVENT(mlx5_esw_vport_qos_destroy, __field(unsigned short, vport_id) __field(unsigned int, tsar_ix) ), - TP_fast_assign(__assign_str(devname, dev_name(vport->dev->device)); + TP_fast_assign(__assign_str(devname); __entry->vport_id = vport->vport; __entry->tsar_ix = vport->qos.esw_tsar_ix; ), @@ -36,7 +36,7 @@ DECLARE_EVENT_CLASS(mlx5_esw_vport_qos_template, __field(unsigned int, max_rate) __field(void *, group) ), - TP_fast_assign(__assign_str(devname, dev_name(vport->dev->device)); + TP_fast_assign(__assign_str(devname); __entry->vport_id = vport->vport; __entry->tsar_ix = vport->qos.esw_tsar_ix; __entry->bw_share = bw_share; @@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(mlx5_esw_group_qos_template, __field(const void *, group) __field(unsigned int, tsar_ix) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->group = group; __entry->tsar_ix = tsar_ix; ), @@ -102,7 +102,7 @@ TRACE_EVENT(mlx5_esw_group_qos_config, __field(unsigned int, bw_share) __field(unsigned int, max_rate) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->group = group; __entry->tsar_ix = tsar_ix; __entry->bw_share = bw_share; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h index 7f7c9af5deed..0537de86f981 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h @@ -22,7 +22,7 @@ DECLARE_EVENT_CLASS(mlx5_sf_dev_template, __field(u16, hw_fn_id) __field(u32, sfnum) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->sfdev = sfdev; __entry->aux_id = aux_id; __entry->hw_fn_id = sfdev->fn_id; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h index 8bf1cd90930d..302ce00da5a9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h @@ -24,7 +24,7 @@ TRACE_EVENT(mlx5_sf_add, __field(u16, hw_fn_id) __field(u32, sfnum) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; @@ -46,7 +46,7 @@ TRACE_EVENT(mlx5_sf_free, __field(u32, controller) __field(u16, hw_fn_id) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; @@ -67,7 +67,7 @@ TRACE_EVENT(mlx5_sf_hwc_alloc, __field(u16, hw_fn_id) __field(u32, sfnum) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; __entry->sfnum = sfnum; @@ -84,7 +84,7 @@ TRACE_EVENT(mlx5_sf_hwc_free, TP_STRUCT__entry(__string(devname, dev_name(dev->device)) __field(u16, hw_fn_id) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->hw_fn_id = hw_fn_id; ), TP_printk("(%s) hw_id=0x%x\n", __get_str(devname), __entry->hw_fn_id) @@ -97,7 +97,7 @@ TRACE_EVENT(mlx5_sf_hwc_deferred_free, TP_STRUCT__entry(__string(devname, dev_name(dev->device)) __field(u16, hw_fn_id) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->hw_fn_id = hw_fn_id; ), TP_printk("(%s) hw_id=0x%x\n", __get_str(devname), __entry->hw_fn_id) @@ -113,7 +113,7 @@ DECLARE_EVENT_CLASS(mlx5_sf_state_template, __field(unsigned int, port_index) __field(u32, controller) __field(u16, hw_fn_id)), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; @@ -152,7 +152,7 @@ TRACE_EVENT(mlx5_sf_update_state, __field(u16, hw_fn_id) __field(u8, state) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h index fd814a190b8b..6352cb004a18 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h @@ -20,7 +20,7 @@ TRACE_EVENT(mlx5_sf_vhca_event, __field(u32, sfnum) __field(u8, vhca_state) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->hw_fn_id = event->function_id; __entry->sfnum = event->sw_function_id; __entry->vhca_state = event->new_vhca_state; diff --git a/drivers/net/fjes/fjes_trace.h b/drivers/net/fjes/fjes_trace.h index 6437ddbd7842..166ef015262b 100644 --- a/drivers/net/fjes/fjes_trace.h +++ b/drivers/net/fjes/fjes_trace.h @@ -85,7 +85,7 @@ TRACE_EVENT(fjes_hw_request_info_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -145,7 +145,7 @@ TRACE_EVENT(fjes_hw_register_buff_addr_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -189,7 +189,7 @@ TRACE_EVENT(fjes_hw_unregister_buff_addr_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -232,7 +232,7 @@ TRACE_EVENT(fjes_hw_start_debug_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -258,7 +258,7 @@ TRACE_EVENT(fjes_hw_stop_debug_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); diff --git a/drivers/net/hyperv/netvsc_trace.h b/drivers/net/hyperv/netvsc_trace.h index f7585563dea5..05e620cbdd29 100644 --- a/drivers/net/hyperv/netvsc_trace.h +++ b/drivers/net/hyperv/netvsc_trace.h @@ -51,7 +51,7 @@ DECLARE_EVENT_CLASS(rndis_msg_class, __field( u32, msg_len ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->queue = q; __entry->req_id = msg->msg.init_req.req_id; __entry->msg_type = msg->ndis_msg_type; @@ -121,7 +121,7 @@ TRACE_EVENT(nvsp_send, __field( u32, msg_type ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->msg_type = msg->hdr.msg_type; ), TP_printk("dev=%s type=%s", @@ -142,7 +142,7 @@ TRACE_EVENT(nvsp_send_pkt, __field( u32, section_size ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->qid = chan->offermsg.offer.sub_channel_index; __entry->channel_type = rpkt->channel_type; __entry->section_index = rpkt->send_buf_section_index; @@ -165,7 +165,7 @@ TRACE_EVENT(nvsp_recv, __field( u32, msg_type ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->qid = chan->offermsg.offer.sub_channel_index; __entry->msg_type = msg->hdr.msg_type; ), diff --git a/drivers/net/wireless/ath/ath10k/trace.h b/drivers/net/wireless/ath/ath10k/trace.h index 64e7a767d963..68b78ca17eaa 100644 --- a/drivers/net/wireless/ath/ath10k/trace.h +++ b/drivers/net/wireless/ath/ath10k/trace.h @@ -55,8 +55,8 @@ DECLARE_EVENT_CLASS(ath10k_log_event, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk( @@ -92,8 +92,8 @@ TRACE_EVENT(ath10k_log_dbg, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->level = level; __assign_vstr(msg, vaf->fmt, vaf->va); ), @@ -121,10 +121,10 @@ TRACE_EVENT(ath10k_log_dbg_dump, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); - __assign_str(msg, msg); - __assign_str(prefix, prefix); + __assign_str(device); + __assign_str(driver); + __assign_str(msg); + __assign_str(prefix); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), @@ -152,8 +152,8 @@ TRACE_EVENT(ath10k_wmi_cmd, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -182,8 +182,8 @@ TRACE_EVENT(ath10k_wmi_event, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -211,8 +211,8 @@ TRACE_EVENT(ath10k_htt_stats, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), @@ -239,8 +239,8 @@ TRACE_EVENT(ath10k_wmi_dbglog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->hw_type = ar->hw_rev; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -269,8 +269,8 @@ TRACE_EVENT(ath10k_htt_pktlog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->hw_type = ar->hw_rev; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(pktlog), buf, buf_len); @@ -301,8 +301,8 @@ TRACE_EVENT(ath10k_htt_tx, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->msdu_id = msdu_id; __entry->msdu_len = msdu_len; __entry->vdev_id = vdev_id; @@ -332,8 +332,8 @@ TRACE_EVENT(ath10k_txrx_tx_unref, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->msdu_id = msdu_id; ), @@ -358,8 +358,8 @@ DECLARE_EVENT_CLASS(ath10k_hdr_event, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = ath10k_frm_hdr_len(data, len); memcpy(__get_dynamic_array(data), data, __entry->len); ), @@ -386,8 +386,8 @@ DECLARE_EVENT_CLASS(ath10k_payload_event, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len - ath10k_frm_hdr_len(data, len); memcpy(__get_dynamic_array(payload), data + ath10k_frm_hdr_len(data, len), __entry->len); @@ -435,8 +435,8 @@ TRACE_EVENT(ath10k_htt_rx_desc, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->hw_type = ar->hw_rev; __entry->len = len; memcpy(__get_dynamic_array(rxdesc), data, len); @@ -472,8 +472,8 @@ TRACE_EVENT(ath10k_wmi_diag_container, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->type = type; __entry->timestamp = timestamp; __entry->code = code; @@ -505,8 +505,8 @@ TRACE_EVENT(ath10k_wmi_diag, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(data), data, len); ), diff --git a/drivers/net/wireless/ath/ath11k/trace.h b/drivers/net/wireless/ath/ath11k/trace.h index 235ab8ea715f..75246b0a82e3 100644 --- a/drivers/net/wireless/ath/ath11k/trace.h +++ b/drivers/net/wireless/ath/ath11k/trace.h @@ -48,8 +48,8 @@ TRACE_EVENT(ath11k_htt_pktlog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->buf_len = buf_len; __entry->pktlog_checksum = pktlog_checksum; memcpy(__get_dynamic_array(pktlog), buf, buf_len); @@ -77,8 +77,8 @@ TRACE_EVENT(ath11k_htt_ppdu_stats, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(ppdu), data, len); ), @@ -105,8 +105,8 @@ TRACE_EVENT(ath11k_htt_rxdesc, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; __entry->log_type = log_type; memcpy(__get_dynamic_array(rxdesc), data, len); @@ -130,8 +130,8 @@ DECLARE_EVENT_CLASS(ath11k_log_event, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk( @@ -171,8 +171,8 @@ TRACE_EVENT(ath11k_wmi_cmd, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -201,8 +201,8 @@ TRACE_EVENT(ath11k_wmi_event, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -230,8 +230,8 @@ TRACE_EVENT(ath11k_log_dbg, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->level = level; WARN_ON_ONCE(vsnprintf(__get_dynamic_array(msg), ATH11K_MSG_MAX, vaf->fmt, @@ -262,10 +262,10 @@ TRACE_EVENT(ath11k_log_dbg_dump, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); - __assign_str(msg, msg); - __assign_str(prefix, prefix); + __assign_str(device); + __assign_str(driver); + __assign_str(msg); + __assign_str(prefix); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), @@ -292,8 +292,8 @@ TRACE_EVENT(ath11k_wmi_diag, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(data), data, len); ), @@ -318,8 +318,8 @@ TRACE_EVENT(ath11k_ps_timekeeper, __field(u32, peer_ps_timestamp) ), - TP_fast_assign(__assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + TP_fast_assign(__assign_str(device); + __assign_str(driver); memcpy(__get_dynamic_array(peer_addr), peer_addr, ETH_ALEN); __entry->peer_ps_state = peer_ps_state; diff --git a/drivers/net/wireless/ath/ath12k/trace.h b/drivers/net/wireless/ath/ath12k/trace.h index 240737e1542d..253c67accb0e 100644 --- a/drivers/net/wireless/ath/ath12k/trace.h +++ b/drivers/net/wireless/ath/ath12k/trace.h @@ -36,8 +36,8 @@ TRACE_EVENT(ath12k_htt_pktlog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->buf_len = buf_len; __entry->pktlog_checksum = pktlog_checksum; memcpy(__get_dynamic_array(pktlog), buf, buf_len); @@ -73,8 +73,8 @@ TRACE_EVENT(ath12k_htt_ppdu_stats, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; __entry->info = ar->pdev->timestamp.info; __entry->sync_tstmp_lo_us = ar->pdev->timestamp.sync_timestamp_hi_us; @@ -117,8 +117,8 @@ TRACE_EVENT(ath12k_htt_rxdesc, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; __entry->type = type; __entry->info = ar->pdev->timestamp.info; @@ -153,8 +153,8 @@ TRACE_EVENT(ath12k_wmi_diag, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(data), data, len); ), diff --git a/drivers/net/wireless/ath/ath6kl/trace.h b/drivers/net/wireless/ath/ath6kl/trace.h index 231a94769ddb..8577aa459c58 100644 --- a/drivers/net/wireless/ath/ath6kl/trace.h +++ b/drivers/net/wireless/ath/ath6kl/trace.h @@ -304,8 +304,8 @@ TRACE_EVENT(ath6kl_log_dbg_dump, ), TP_fast_assign( - __assign_str(msg, msg); - __assign_str(prefix, prefix); + __assign_str(msg); + __assign_str(prefix); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), diff --git a/drivers/net/wireless/ath/trace.h b/drivers/net/wireless/ath/trace.h index 9935cf475b6d..82aac0a4baff 100644 --- a/drivers/net/wireless/ath/trace.h +++ b/drivers/net/wireless/ath/trace.h @@ -44,8 +44,8 @@ TRACE_EVENT(ath_log, ), TP_fast_assign( - __assign_str(device, wiphy_name(wiphy)); - __assign_str(driver, KBUILD_MODNAME); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h index 5d66e94c806d..96032322b165 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h @@ -41,7 +41,7 @@ TRACE_EVENT(brcmf_err, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(func, func); + __assign_str(func); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(func), __get_str(msg)) @@ -57,7 +57,7 @@ TRACE_EVENT(brcmf_dbg, ), TP_fast_assign( __entry->level = level; - __assign_str(func, func); + __assign_str(func); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(func), __get_str(msg)) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h index a0da3248b942..53b3dba50737 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h @@ -81,7 +81,7 @@ TRACE_EVENT(brcms_macintstatus, __field(u32, mask) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->in_isr = in_isr; __entry->macintstatus = macintstatus; __entry->mask = mask; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h index 42b0a91656c4..908ce3c864fe 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h @@ -71,7 +71,7 @@ TRACE_EVENT(brcms_dbg, ), TP_fast_assign( __entry->level = level; - __assign_str(func, func); + __assign_str(func); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(func), __get_str(msg)) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h index cf2cc070f1e5..24ac34fa0207 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h @@ -31,7 +31,7 @@ TRACE_EVENT(brcms_txdesc, __dynamic_array(u8, txh, txh_len) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); memcpy(__get_dynamic_array(txh), txh, txh_len); ), TP_printk("[%s] txdesc", __get_str(dev)) @@ -54,7 +54,7 @@ TRACE_EVENT(brcms_txstatus, __field(u16, ackphyrxsh) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->framelen = framelen; __entry->frameid = frameid; __entry->status = status; @@ -85,7 +85,7 @@ TRACE_EVENT(brcms_ampdu_session, __field(u16, dma_len) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->max_ampdu_len = max_ampdu_len; __entry->max_ampdu_frames = max_ampdu_frames; __entry->ampdu_len = ampdu_len; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h index 1d6c292cf545..0db1fa5477af 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h @@ -57,7 +57,7 @@ TRACE_EVENT(iwlwifi_dbg, ), TP_fast_assign( __entry->level = level; - __assign_str(function, function); + __assign_str(function); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s", __get_str(msg)) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h index c3e09f4fefeb..76166e1b10e5 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h @@ -87,7 +87,7 @@ static inline void trace_ ## name(proto) {} #endif #define DEV_ENTRY __string(dev, dev_name(dev)) -#define DEV_ASSIGN __assign_str(dev, dev_name(dev)) +#define DEV_ASSIGN __assign_str(dev) #include "iwl-devtrace-io.h" #include "iwl-devtrace-ucode.h" diff --git a/drivers/soc/qcom/pmic_pdcharger_ulog.h b/drivers/soc/qcom/pmic_pdcharger_ulog.h index 152e3a6b5480..1cfa58f0e34c 100644 --- a/drivers/soc/qcom/pmic_pdcharger_ulog.h +++ b/drivers/soc/qcom/pmic_pdcharger_ulog.h @@ -18,7 +18,7 @@ TRACE_EVENT(pmic_pdcharger_ulog_msg, __string(msg, msg) ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("%s", __get_str(msg)) ); diff --git a/drivers/soc/qcom/trace-aoss.h b/drivers/soc/qcom/trace-aoss.h index 554029b33b44..fb5b0470c40d 100644 --- a/drivers/soc/qcom/trace-aoss.h +++ b/drivers/soc/qcom/trace-aoss.h @@ -18,7 +18,7 @@ TRACE_EVENT(aoss_send, __string(msg, msg) ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("%s", __get_str(msg)) ); @@ -31,7 +31,7 @@ TRACE_EVENT(aoss_send_done, __field(int, ret) ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); __entry->ret = ret; ), TP_printk("%s: %d", __get_str(msg), __entry->ret) diff --git a/drivers/soc/qcom/trace-rpmh.h b/drivers/soc/qcom/trace-rpmh.h index be6b42ecc1f8..593ec1d4e010 100644 --- a/drivers/soc/qcom/trace-rpmh.h +++ b/drivers/soc/qcom/trace-rpmh.h @@ -26,7 +26,7 @@ TRACE_EVENT(rpmh_tx_done, ), TP_fast_assign( - __assign_str(name, d->name); + __assign_str(name); __entry->m = m; __entry->addr = r->cmds[0].addr; __entry->data = r->cmds[0].data; @@ -55,7 +55,7 @@ TRACE_EVENT(rpmh_send_msg, ), TP_fast_assign( - __assign_str(name, d->name); + __assign_str(name); __entry->m = m; __entry->state = state; __entry->n = n; diff --git a/drivers/thermal/thermal_trace.h b/drivers/thermal/thermal_trace.h index 88a962f560f2..df8f4edd6068 100644 --- a/drivers/thermal/thermal_trace.h +++ b/drivers/thermal/thermal_trace.h @@ -37,7 +37,7 @@ TRACE_EVENT(thermal_temperature, ), TP_fast_assign( - __assign_str(thermal_zone, tz->type); + __assign_str(thermal_zone); __entry->id = tz->id; __entry->temp_prev = tz->last_temperature; __entry->temp = tz->temperature; @@ -60,7 +60,7 @@ TRACE_EVENT(cdev_update, ), TP_fast_assign( - __assign_str(type, cdev->type); + __assign_str(type); __entry->target = target; ), @@ -82,7 +82,7 @@ TRACE_EVENT(thermal_zone_trip, ), TP_fast_assign( - __assign_str(thermal_zone, tz->type); + __assign_str(thermal_zone); __entry->id = tz->id; __entry->trip = trip; __entry->trip_type = trip_type; @@ -156,7 +156,7 @@ TRACE_EVENT(thermal_power_devfreq_get_power, ), TP_fast_assign( - __assign_str(type, cdev->type); + __assign_str(type); __entry->freq = freq; __entry->busy_time = status->busy_time; __entry->total_time = status->total_time; @@ -184,7 +184,7 @@ TRACE_EVENT(thermal_power_devfreq_limit, ), TP_fast_assign( - __assign_str(type, cdev->type); + __assign_str(type); __entry->freq = freq; __entry->cdev_state = cdev_state; __entry->power = power; diff --git a/drivers/usb/cdns3/cdns3-trace.h b/drivers/usb/cdns3/cdns3-trace.h index 40db89e3333c..c4e542f1b9b7 100644 --- a/drivers/usb/cdns3/cdns3-trace.h +++ b/drivers/usb/cdns3/cdns3-trace.h @@ -33,7 +33,7 @@ TRACE_EVENT(cdns3_halt, __field(u8, flush) ), TP_fast_assign( - __assign_str(name, ep_priv->name); + __assign_str(name); __entry->halt = halt; __entry->flush = flush; ), @@ -49,8 +49,8 @@ TRACE_EVENT(cdns3_wa1, __string(msg, msg) ), TP_fast_assign( - __assign_str(ep_name, ep_priv->name); - __assign_str(msg, msg); + __assign_str(ep_name); + __assign_str(msg); ), TP_printk("WA1: %s %s", __get_str(ep_name), __get_str(msg)) ); @@ -63,8 +63,8 @@ TRACE_EVENT(cdns3_wa2, __string(msg, msg) ), TP_fast_assign( - __assign_str(ep_name, ep_priv->name); - __assign_str(msg, msg); + __assign_str(ep_name); + __assign_str(msg); ), TP_printk("WA2: %s %s", __get_str(ep_name), __get_str(msg)) ); @@ -77,7 +77,7 @@ DECLARE_EVENT_CLASS(cdns3_log_doorbell, __field(u32, ep_trbaddr) ), TP_fast_assign( - __assign_str(name, ep_name); + __assign_str(name); __entry->ep_trbaddr = ep_trbaddr; ), TP_printk("%s, ep_trbaddr %08x", __get_str(name), @@ -125,7 +125,7 @@ DECLARE_EVENT_CLASS(cdns3_log_epx_irq, __field(u32, use_streams) ), TP_fast_assign( - __assign_str(ep_name, priv_ep->name); + __assign_str(ep_name); __entry->ep_sts = readl(&priv_dev->regs->ep_sts); __entry->ep_traddr = readl(&priv_dev->regs->ep_traddr); __entry->ep_last_sid = priv_ep->last_stream_id; @@ -214,7 +214,7 @@ DECLARE_EVENT_CLASS(cdns3_log_request, __field(unsigned int, stream_id) ), TP_fast_assign( - __assign_str(name, req->priv_ep->name); + __assign_str(name); __entry->req = req; __entry->buf = req->request.buf; __entry->actual = req->request.actual; @@ -294,7 +294,7 @@ DECLARE_EVENT_CLASS(cdns3_stream_split_transfer_len, __field(unsigned int, stream_id) ), TP_fast_assign( - __assign_str(name, req->priv_ep->name); + __assign_str(name); __entry->req = req; __entry->actual = req->request.length; __entry->length = req->request.actual; @@ -329,7 +329,7 @@ DECLARE_EVENT_CLASS(cdns3_log_aligned_request, __field(u32, aligned_buf_size) ), TP_fast_assign( - __assign_str(name, priv_req->priv_ep->name); + __assign_str(name); __entry->req = &priv_req->request; __entry->buf = priv_req->request.buf; __entry->dma = priv_req->request.dma; @@ -364,7 +364,7 @@ DECLARE_EVENT_CLASS(cdns3_log_map_request, __field(dma_addr_t, dma) ), TP_fast_assign( - __assign_str(name, priv_req->priv_ep->name); + __assign_str(name); __entry->req = &priv_req->request; __entry->buf = priv_req->request.buf; __entry->dma = priv_req->request.dma; @@ -395,7 +395,7 @@ DECLARE_EVENT_CLASS(cdns3_log_trb, __field(unsigned int, last_stream_id) ), TP_fast_assign( - __assign_str(name, priv_ep->name); + __assign_str(name); __entry->trb = trb; __entry->buffer = le32_to_cpu(trb->buffer); __entry->length = le32_to_cpu(trb->length); @@ -467,7 +467,7 @@ DECLARE_EVENT_CLASS(cdns3_log_ep, __field(u8, dequeue) ), TP_fast_assign( - __assign_str(name, priv_ep->name); + __assign_str(name); __entry->maxpacket = priv_ep->endpoint.maxpacket; __entry->maxpacket_limit = priv_ep->endpoint.maxpacket_limit; __entry->max_streams = priv_ep->endpoint.max_streams; diff --git a/drivers/usb/cdns3/cdnsp-trace.h b/drivers/usb/cdns3/cdnsp-trace.h index 4b51011eb00b..f2bcf77a5d0a 100644 --- a/drivers/usb/cdns3/cdnsp-trace.h +++ b/drivers/usb/cdns3/cdnsp-trace.h @@ -48,7 +48,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_ep, __field(u8, drbls_count) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->state = pep->ep_state; __entry->stream_id = stream_id; __entry->enabled = pep->ep_state & EP_HAS_STREAMS; @@ -138,7 +138,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_simple, __string(text, msg) ), TP_fast_assign( - __assign_str(text, msg); + __assign_str(text); ), TP_printk("%s", __get_str(text)) ); @@ -303,7 +303,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_bounce, __field(unsigned int, unalign) ), TP_fast_assign( - __assign_str(name, preq->pep->name); + __assign_str(name); __entry->new_buf_len = new_buf_len; __entry->offset = offset; __entry->dma = dma; @@ -470,7 +470,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_request, ), TP_fast_assign( - __assign_str(name, req->pep->name); + __assign_str(name); __entry->request = &req->request; __entry->preq = req; __entry->buf = req->request.buf; @@ -674,7 +674,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_td_info, __field(dma_addr_t, trb_dma) ), TP_fast_assign( - __assign_str(name, preq->pep->name); + __assign_str(name); __entry->request = &preq->request; __entry->preq = preq; __entry->first_trb = preq->td.first_trb; diff --git a/drivers/usb/chipidea/trace.h b/drivers/usb/chipidea/trace.h index ca0e65b48f0a..1875419cd17f 100644 --- a/drivers/usb/chipidea/trace.h +++ b/drivers/usb/chipidea/trace.h @@ -31,7 +31,7 @@ TRACE_EVENT(ci_log, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(name, dev_name(ci->dev)); + __assign_str(name); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(name), __get_str(msg)) @@ -51,7 +51,7 @@ DECLARE_EVENT_CLASS(ci_log_trb, __field(u32, type) ), TP_fast_assign( - __assign_str(name, hwep->name); + __assign_str(name); __entry->req = &hwreq->req; __entry->td = td; __entry->dma = td->dma; diff --git a/drivers/usb/dwc3/trace.h b/drivers/usb/dwc3/trace.h index d2997d17cfbe..bdeb1aaf65d8 100644 --- a/drivers/usb/dwc3/trace.h +++ b/drivers/usb/dwc3/trace.h @@ -112,7 +112,7 @@ DECLARE_EVENT_CLASS(dwc3_log_request, __field(int, no_interrupt) ), TP_fast_assign( - __assign_str(name, req->dep->name); + __assign_str(name); __entry->req = req; __entry->actual = req->request.actual; __entry->length = req->request.length; @@ -193,7 +193,7 @@ DECLARE_EVENT_CLASS(dwc3_log_gadget_ep_cmd, __field(int, cmd_status) ), TP_fast_assign( - __assign_str(name, dep->name); + __assign_str(name); __entry->cmd = cmd; __entry->param0 = params->param0; __entry->param1 = params->param1; @@ -229,7 +229,7 @@ DECLARE_EVENT_CLASS(dwc3_log_trb, __field(u32, dequeue) ), TP_fast_assign( - __assign_str(name, dep->name); + __assign_str(name); __entry->trb = trb; __entry->bpl = trb->bpl; __entry->bph = trb->bph; @@ -301,7 +301,7 @@ DECLARE_EVENT_CLASS(dwc3_log_ep, __field(u8, trb_dequeue) ), TP_fast_assign( - __assign_str(name, dep->name); + __assign_str(name); __entry->maxpacket = dep->endpoint.maxpacket; __entry->maxpacket_limit = dep->endpoint.maxpacket_limit; __entry->max_streams = dep->endpoint.max_streams; diff --git a/drivers/usb/gadget/udc/cdns2/cdns2-trace.h b/drivers/usb/gadget/udc/cdns2/cdns2-trace.h index 61f241634ea5..ade1752956b1 100644 --- a/drivers/usb/gadget/udc/cdns2/cdns2-trace.h +++ b/drivers/usb/gadget/udc/cdns2/cdns2-trace.h @@ -64,7 +64,7 @@ DECLARE_EVENT_CLASS(cdns2_log_simple, __string(text, msg) ), TP_fast_assign( - __assign_str(text, msg); + __assign_str(text); ), TP_printk("%s", __get_str(text)) ); @@ -103,7 +103,7 @@ TRACE_EVENT(cdns2_ep_halt, __field(u8, flush) ), TP_fast_assign( - __assign_str(name, ep_priv->name); + __assign_str(name); __entry->halt = halt; __entry->flush = flush; ), @@ -119,8 +119,8 @@ TRACE_EVENT(cdns2_wa1, __string(msg, msg) ), TP_fast_assign( - __assign_str(ep_name, ep_priv->name); - __assign_str(msg, msg); + __assign_str(ep_name); + __assign_str(msg); ), TP_printk("WA1: %s %s", __get_str(ep_name), __get_str(msg)) ); @@ -134,7 +134,7 @@ DECLARE_EVENT_CLASS(cdns2_log_doorbell, __field(u32, ep_trbaddr) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->ep_trbaddr = ep_trbaddr; ), TP_printk("%s, ep_trbaddr %08x", __get_str(name), @@ -196,7 +196,7 @@ DECLARE_EVENT_CLASS(cdns2_log_epx_irq, __field(u32, ep_traddr) ), TP_fast_assign( - __assign_str(ep_name, pep->name); + __assign_str(ep_name); __entry->ep_sts = readl(&pdev->adma_regs->ep_sts); __entry->ep_ists = readl(&pdev->adma_regs->ep_ists); __entry->ep_traddr = readl(&pdev->adma_regs->ep_traddr); @@ -288,7 +288,7 @@ DECLARE_EVENT_CLASS(cdns2_log_request, __field(int, end_trb) ), TP_fast_assign( - __assign_str(name, preq->pep->name); + __assign_str(name); __entry->request = &preq->request; __entry->preq = preq; __entry->buf = preq->request.buf; @@ -380,7 +380,7 @@ DECLARE_EVENT_CLASS(cdns2_log_map_request, __field(dma_addr_t, dma) ), TP_fast_assign( - __assign_str(name, priv_req->pep->name); + __assign_str(name); __entry->req = &priv_req->request; __entry->buf = priv_req->request.buf; __entry->dma = priv_req->request.dma; @@ -411,7 +411,7 @@ DECLARE_EVENT_CLASS(cdns2_log_trb, __field(u32, type) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->trb = trb; __entry->buffer = le32_to_cpu(trb->buffer); __entry->length = le32_to_cpu(trb->length); @@ -476,7 +476,7 @@ DECLARE_EVENT_CLASS(cdns2_log_ep, __field(u8, dequeue) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->maxpacket = pep->endpoint.maxpacket; __entry->maxpacket_limit = pep->endpoint.maxpacket_limit; __entry->flags = pep->ep_state; @@ -568,7 +568,7 @@ DECLARE_EVENT_CLASS(cdns2_log_epx_reg_config, __field(u32, ep_cfg_reg) ), TP_fast_assign( - __assign_str(ep_name, pep->name); + __assign_str(ep_name); __entry->burst_size = pep->trb_burst_size; __entry->maxpack_reg = pep->dir ? readw(&pdev->epx_regs->txmaxpack[pep->num - 1]) : readw(&pdev->epx_regs->rxmaxpack[pep->num - 1]); diff --git a/drivers/usb/gadget/udc/trace.h b/drivers/usb/gadget/udc/trace.h index a5ed26fbc2da..4e334298b0e8 100644 --- a/drivers/usb/gadget/udc/trace.h +++ b/drivers/usb/gadget/udc/trace.h @@ -157,7 +157,7 @@ DECLARE_EVENT_CLASS(udc_log_ep, __field(int, ret) ), TP_fast_assign( - __assign_str(name, ep->name); + __assign_str(name); __entry->maxpacket = ep->maxpacket; __entry->maxpacket_limit = ep->maxpacket_limit; __entry->max_streams = ep->max_streams; @@ -233,7 +233,7 @@ DECLARE_EVENT_CLASS(udc_log_req, __field(struct usb_request *, req) ), TP_fast_assign( - __assign_str(name, ep->name); + __assign_str(name); __entry->length = req->length; __entry->actual = req->actual; __entry->num_sgs = req->num_sgs; diff --git a/drivers/usb/mtu3/mtu3_trace.h b/drivers/usb/mtu3/mtu3_trace.h index 03d2a9bac27e..89870175d635 100644 --- a/drivers/usb/mtu3/mtu3_trace.h +++ b/drivers/usb/mtu3/mtu3_trace.h @@ -26,7 +26,7 @@ TRACE_EVENT(mtu3_log, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(name), __get_str(msg)) @@ -127,7 +127,7 @@ DECLARE_EVENT_CLASS(mtu3_log_request, __field(int, no_interrupt) ), TP_fast_assign( - __assign_str(name, mreq->mep->name); + __assign_str(name); __entry->mreq = mreq; __entry->gpd = mreq->gpd; __entry->actual = mreq->request.actual; @@ -182,7 +182,7 @@ DECLARE_EVENT_CLASS(mtu3_log_gpd, __field(u32, dw3) ), TP_fast_assign( - __assign_str(name, mep->name); + __assign_str(name); __entry->gpd = gpd; __entry->dw0 = le32_to_cpu(gpd->dw0_info); __entry->dw1 = le32_to_cpu(gpd->next_gpd); @@ -226,7 +226,7 @@ DECLARE_EVENT_CLASS(mtu3_log_ep, __field(struct mtu3_gpd_ring *, gpd_ring) ), TP_fast_assign( - __assign_str(name, mep->name); + __assign_str(name); __entry->type = mep->type; __entry->slot = mep->slot; __entry->maxp = mep->ep.maxpacket; diff --git a/drivers/usb/musb/musb_trace.h b/drivers/usb/musb/musb_trace.h index f246b14394c4..726e6697d475 100644 --- a/drivers/usb/musb/musb_trace.h +++ b/drivers/usb/musb/musb_trace.h @@ -31,7 +31,7 @@ TRACE_EVENT(musb_log, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(name), __get_str(msg)) @@ -46,9 +46,9 @@ TRACE_EVENT(musb_state, __string(desc, desc) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __entry->devctl = devctl; - __assign_str(desc, desc); + __assign_str(desc); ), TP_printk("%s: devctl: %02x %s", __get_str(name), __entry->devctl, __get_str(desc)) @@ -160,7 +160,7 @@ TRACE_EVENT(musb_isr, __field(u16, int_rx) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __entry->int_usb = musb->int_usb; __entry->int_tx = musb->int_tx; __entry->int_rx = musb->int_rx; @@ -184,7 +184,7 @@ DECLARE_EVENT_CLASS(musb_urb, __field(u32, actual_len) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __entry->urb = urb; __entry->pipe = urb->pipe; __entry->status = urb->status; @@ -325,7 +325,7 @@ DECLARE_EVENT_CLASS(musb_cppi41, ), TP_fast_assign( __entry->ch = ch; - __assign_str(name, dev_name(ch->hw_ep->musb->controller)); + __assign_str(name); __entry->hwep = ch->hw_ep->epnum; __entry->port = ch->port_num; __entry->is_tx = ch->is_tx; diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h index 362e1fc7ef6a..84fcf26e306e 100644 --- a/fs/bcachefs/trace.h +++ b/fs/bcachefs/trace.h @@ -43,7 +43,7 @@ DECLARE_EVENT_CLASS(fs_str, TP_fast_assign( __entry->dev = c->dev; - __assign_str(str, str); + __assign_str(str); ), TP_printk("%d,%d\n%s", MAJOR(__entry->dev), MINOR(__entry->dev), __get_str(str)) @@ -64,7 +64,7 @@ DECLARE_EVENT_CLASS(trans_str, __entry->dev = trans->c->dev; strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); __entry->caller_ip = caller_ip; - __assign_str(str, str); + __assign_str(str); ), TP_printk("%d,%d %s %pS %s", @@ -85,7 +85,7 @@ DECLARE_EVENT_CLASS(trans_str_nocaller, TP_fast_assign( __entry->dev = trans->c->dev; strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); - __assign_str(str, str); + __assign_str(str); ), TP_printk("%d,%d %s %s", diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 10985a4b8259..4de8780a7c48 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -47,7 +47,7 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event, TP_fast_assign( __entry->error = error < 0 ? -error : 0; - __assign_str(dstaddr, clp->cl_hostname); + __assign_str(dstaddr); ), TP_printk( @@ -94,8 +94,8 @@ TRACE_EVENT(nfs4_trunked_exchange_id, TP_fast_assign( __entry->error = error < 0 ? -error : 0; - __assign_str(main_addr, clp->cl_hostname); - __assign_str(trunk_addr, addr); + __assign_str(main_addr); + __assign_str(trunk_addr); ), TP_printk( @@ -365,7 +365,7 @@ TRACE_EVENT(nfs4_state_mgr, TP_fast_assign( __entry->state = clp->cl_state; - __assign_str(hostname, clp->cl_hostname); + __assign_str(hostname); ), TP_printk( @@ -393,8 +393,8 @@ TRACE_EVENT(nfs4_state_mgr_failed, TP_fast_assign( __entry->error = status < 0 ? -status : 0; __entry->state = clp->cl_state; - __assign_str(hostname, clp->cl_hostname); - __assign_str(section, section); + __assign_str(hostname); + __assign_str(section); ), TP_printk( @@ -578,7 +578,7 @@ DECLARE_EVENT_CLASS(nfs4_open_event, __entry->fhandle = 0; } __entry->dir = NFS_FILEID(d_inode(ctx->dentry->d_parent)); - __assign_str(name, ctx->dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -1072,7 +1072,7 @@ DECLARE_EVENT_CLASS(nfs4_lookup_event, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->error = -error; - __assign_str(name, name->name); + __assign_str(name); ), TP_printk( @@ -1156,8 +1156,8 @@ TRACE_EVENT(nfs4_rename, __entry->olddir = NFS_FILEID(olddir); __entry->newdir = NFS_FILEID(newdir); __entry->error = error < 0 ? -error : 0; - __assign_str(oldname, oldname->name); - __assign_str(newname, newname->name); + __assign_str(oldname); + __assign_str(newname); ), TP_printk( @@ -1359,7 +1359,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event, __entry->fileid = 0; __entry->dev = 0; } - __assign_str(dstaddr, clp ? clp->cl_hostname : "unknown"); + __assign_str(dstaddr); ), TP_printk( @@ -1416,7 +1416,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event, __entry->fileid = 0; __entry->dev = 0; } - __assign_str(dstaddr, clp ? clp->cl_hostname : "unknown"); + __assign_str(dstaddr); __entry->stateid_seq = be32_to_cpu(stateid->seqid); __entry->stateid_hash = @@ -1960,7 +1960,7 @@ DECLARE_EVENT_CLASS(nfs4_deviceid_event, ), TP_fast_assign( - __assign_str(dstaddr, clp->cl_hostname); + __assign_str(dstaddr); memcpy(__entry->deviceid, deviceid->data, NFS4_DEVICEID4_SIZE); ), @@ -1998,7 +1998,7 @@ DECLARE_EVENT_CLASS(nfs4_deviceid_status, TP_fast_assign( __entry->dev = server->s_dev; __entry->status = status; - __assign_str(dstaddr, server->nfs_client->cl_hostname); + __assign_str(dstaddr); memcpy(__entry->deviceid, deviceid->data, NFS4_DEVICEID4_SIZE); ), @@ -2036,8 +2036,8 @@ TRACE_EVENT(fl_getdevinfo, ), TP_fast_assign( - __assign_str(mds_addr, server->nfs_client->cl_hostname); - __assign_str(ds_ips, ds_remotestr); + __assign_str(mds_addr); + __assign_str(ds_ips); memcpy(__entry->deviceid, deviceid->data, NFS4_DEVICEID4_SIZE); ), @@ -2083,9 +2083,7 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, be32_to_cpu(hdr->args.stateid.seqid); __entry->stateid_hash = nfs_stateid_hash(&hdr->args.stateid); - __assign_str(dstaddr, hdr->ds_clp ? - rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient, - RPC_DISPLAY_ADDR) : "unknown"); + __assign_str(dstaddr); ), TP_printk( @@ -2139,9 +2137,7 @@ TRACE_EVENT(ff_layout_commit_error, __entry->dev = inode->i_sb->s_dev; __entry->offset = data->args.offset; __entry->count = data->args.count; - __assign_str(dstaddr, data->ds_clp ? - rpc_peeraddr2str(data->ds_clp->cl_rpcclient, - RPC_DISPLAY_ADDR) : "unknown"); + __assign_str(dstaddr); ), TP_printk( @@ -2579,7 +2575,7 @@ DECLARE_EVENT_CLASS(nfs4_xattr_event, __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); - __assign_str(name, name); + __assign_str(name); ), TP_printk( diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index afedb449b54f..1e710654af11 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -409,7 +409,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event, __entry->dir = NFS_FILEID(dir); __entry->flags = flags; __entry->fileid = d_is_negative(dentry) ? 0 : NFS_FILEID(d_inode(dentry)); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -457,7 +457,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event_done, __entry->error = error < 0 ? -error : 0; __entry->flags = flags; __entry->fileid = d_is_negative(dentry) ? 0 : NFS_FILEID(d_inode(dentry)); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -512,7 +512,7 @@ TRACE_EVENT(nfs_atomic_open_enter, __entry->dir = NFS_FILEID(dir); __entry->flags = flags; __entry->fmode = (__force unsigned long)ctx->mode; - __assign_str(name, ctx->dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -551,7 +551,7 @@ TRACE_EVENT(nfs_atomic_open_exit, __entry->dir = NFS_FILEID(dir); __entry->flags = flags; __entry->fmode = (__force unsigned long)ctx->mode; - __assign_str(name, ctx->dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -587,7 +587,7 @@ TRACE_EVENT(nfs_create_enter, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -623,7 +623,7 @@ TRACE_EVENT(nfs_create_exit, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -654,7 +654,7 @@ DECLARE_EVENT_CLASS(nfs_directory_event, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -693,7 +693,7 @@ DECLARE_EVENT_CLASS(nfs_directory_event_done, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->error = error < 0 ? -error : 0; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -747,7 +747,7 @@ TRACE_EVENT(nfs_link_enter, __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->dir = NFS_FILEID(dir); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -783,7 +783,7 @@ TRACE_EVENT(nfs_link_exit, __entry->fileid = NFS_FILEID(inode); __entry->dir = NFS_FILEID(dir); __entry->error = error < 0 ? -error : 0; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -819,8 +819,8 @@ DECLARE_EVENT_CLASS(nfs_rename_event, __entry->dev = old_dir->i_sb->s_dev; __entry->old_dir = NFS_FILEID(old_dir); __entry->new_dir = NFS_FILEID(new_dir); - __assign_str(old_name, old_dentry->d_name.name); - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(old_name); + __assign_str(new_name); ), TP_printk( @@ -868,8 +868,8 @@ DECLARE_EVENT_CLASS(nfs_rename_event_done, __entry->error = -error; __entry->old_dir = NFS_FILEID(old_dir); __entry->new_dir = NFS_FILEID(new_dir); - __assign_str(old_name, old_dentry->d_name.name); - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(old_name); + __assign_str(new_name); ), TP_printk( @@ -1636,8 +1636,8 @@ TRACE_EVENT(nfs_mount_assign, ), TP_fast_assign( - __assign_str(option, option); - __assign_str(value, value); + __assign_str(option); + __assign_str(value); ), TP_printk("option %s=%s", @@ -1657,7 +1657,7 @@ TRACE_EVENT(nfs_mount_option, ), TP_fast_assign( - __assign_str(option, param->key); + __assign_str(option); ), TP_printk("option %s", __get_str(option)) @@ -1675,7 +1675,7 @@ TRACE_EVENT(nfs_mount_path, ), TP_fast_assign( - __assign_str(path, path); + __assign_str(path); ), TP_printk("path='%s'", __get_str(path)) @@ -1710,9 +1710,8 @@ DECLARE_EVENT_CLASS(nfs_xdr_event, __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->version = task->tk_client->cl_vers; __entry->error = error; - __assign_str(program, - task->tk_client->cl_program->name); - __assign_str(procedure, task->tk_msg.rpc_proc->p_name); + __assign_str(program); + __assign_str(procedure); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index b5e48d504062..77bbd23aa150 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -104,7 +104,7 @@ TRACE_EVENT(nfsd_compound, TP_fast_assign( __entry->xid = be32_to_cpu(rqst->rq_xid); __entry->opcnt = opcnt; - __assign_str(tag, tag); + __assign_str(tag); ), TP_printk("xid=0x%08x opcnt=%u tag=%s", __entry->xid, __entry->opcnt, __get_str(tag) @@ -127,7 +127,7 @@ TRACE_EVENT(nfsd_compound_status, __entry->args_opcnt = args_opcnt; __entry->resp_opcnt = resp_opcnt; __entry->status = be32_to_cpu(status); - __assign_str(name, name); + __assign_str(name); ), TP_printk("op=%u/%u %s status=%d", __entry->resp_opcnt, __entry->args_opcnt, @@ -318,7 +318,7 @@ TRACE_EVENT(nfsd_exp_find_key, TP_fast_assign( __entry->fsidtype = key->ek_fsidtype; memcpy(__entry->fsid, key->ek_fsid, 4*6); - __assign_str(auth_domain, key->ek_client->name); + __assign_str(auth_domain); __entry->status = status; ), TP_printk("fsid=%x::%s domain=%s status=%d", @@ -342,8 +342,8 @@ TRACE_EVENT(nfsd_expkey_update, TP_fast_assign( __entry->fsidtype = key->ek_fsidtype; memcpy(__entry->fsid, key->ek_fsid, 4*6); - __assign_str(auth_domain, key->ek_client->name); - __assign_str(path, exp_path); + __assign_str(auth_domain); + __assign_str(path); __entry->cache = !test_bit(CACHE_NEGATIVE, &key->h.flags); ), TP_printk("fsid=%x::%s domain=%s path=%s cache=%s", @@ -365,8 +365,8 @@ TRACE_EVENT(nfsd_exp_get_by_name, __field(int, status) ), TP_fast_assign( - __assign_str(path, key->ex_path.dentry->d_name.name); - __assign_str(auth_domain, key->ex_client->name); + __assign_str(path); + __assign_str(auth_domain); __entry->status = status; ), TP_printk("path=%s domain=%s status=%d", @@ -385,8 +385,8 @@ TRACE_EVENT(nfsd_export_update, __field(bool, cache) ), TP_fast_assign( - __assign_str(path, key->ex_path.dentry->d_name.name); - __assign_str(auth_domain, key->ex_client->name); + __assign_str(path); + __assign_str(auth_domain); __entry->cache = !test_bit(CACHE_NEGATIVE, &key->h.flags); ), TP_printk("path=%s domain=%s cache=%s", @@ -485,7 +485,7 @@ TRACE_EVENT(nfsd_dirent, TP_fast_assign( __entry->fh_hash = fhp ? knfsd_fh_hash(&fhp->fh_handle) : 0; __entry->ino = ino; - __assign_str(name, name); + __assign_str(name); ), TP_printk("fh_hash=0x%08x ino=%llu name=%s", __entry->fh_hash, __entry->ino, __get_str(name) @@ -1000,7 +1000,7 @@ DECLARE_EVENT_CLASS(nfsd_clid_class, __entry->flavor = clp->cl_cred.cr_flavor; memcpy(__entry->verifier, (void *)&clp->cl_verifier, NFS4_VERIFIER_SIZE); - __assign_str(name, clp->cl_name.data); + __assign_str(name); ), TP_printk("addr=%pISpc name='%s' verifier=0x%s flavor=%s client=%08x:%08x", __entry->addr, __get_str(name), @@ -1519,7 +1519,7 @@ TRACE_EVENT(nfsd_cb_setup, TP_fast_assign( __entry->cl_boot = clp->cl_clientid.cl_boot; __entry->cl_id = clp->cl_clientid.cl_id; - __assign_str(netid, netid); + __assign_str(netid); __entry->authflavor = authflavor; __assign_sockaddr(addr, &clp->cl_cb_conn.cb_addr, clp->cl_cb_conn.cb_addrlen) @@ -1864,7 +1864,7 @@ TRACE_EVENT(nfsd_ctl_unlock_ip, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(address, address); + __assign_str(address); ), TP_printk("address=%s", __get_str(address) @@ -1883,7 +1883,7 @@ TRACE_EVENT(nfsd_ctl_unlock_fs, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(path, path); + __assign_str(path); ), TP_printk("path=%s", __get_str(path) @@ -1907,8 +1907,8 @@ TRACE_EVENT(nfsd_ctl_filehandle, TP_fast_assign( __entry->netns_ino = net->ns.inum; __entry->maxsize = maxsize; - __assign_str(domain, domain); - __assign_str(path, path); + __assign_str(domain); + __assign_str(path); ), TP_printk("domain=%s path=%s maxsize=%d", __get_str(domain), __get_str(path), __entry->maxsize @@ -1968,7 +1968,7 @@ TRACE_EVENT(nfsd_ctl_version, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(mesg, mesg); + __assign_str(mesg); ), TP_printk("%s", __get_str(mesg) @@ -2009,7 +2009,7 @@ TRACE_EVENT(nfsd_ctl_ports_addxprt, TP_fast_assign( __entry->netns_ino = net->ns.inum; __entry->port = port; - __assign_str(transport, transport); + __assign_str(transport); ), TP_printk("transport=%s port=%d", __get_str(transport), __entry->port @@ -2070,7 +2070,7 @@ TRACE_EVENT(nfsd_ctl_time, TP_fast_assign( __entry->netns_ino = net->ns.inum; __entry->time = time; - __assign_str(name, name); + __assign_str(name); ), TP_printk("file=%s time=%d", __get_str(name), __entry->time @@ -2089,7 +2089,7 @@ TRACE_EVENT(nfsd_ctl_recoverydir, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(recdir, recdir); + __assign_str(recdir); ), TP_printk("recdir=%s", __get_str(recdir) diff --git a/fs/ocfs2/ocfs2_trace.h b/fs/ocfs2/ocfs2_trace.h index 9898c11bdfa1..60e208b01c8d 100644 --- a/fs/ocfs2/ocfs2_trace.h +++ b/fs/ocfs2/ocfs2_trace.h @@ -82,7 +82,7 @@ DECLARE_EVENT_CLASS(ocfs2__string, __string(name,name) ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) ); @@ -1289,7 +1289,7 @@ DECLARE_EVENT_CLASS(ocfs2__file_ops, __entry->dentry = dentry; __entry->ino = ino; __entry->d_len = d_len; - __assign_str(d_name, d_name); + __assign_str(d_name); __entry->para = para; ), TP_printk("%p %p %p %llu %llu %.*s", __entry->inode, __entry->file, @@ -1425,7 +1425,7 @@ TRACE_EVENT(ocfs2_setattr, __entry->dentry = dentry; __entry->ino = ino; __entry->d_len = d_len; - __assign_str(d_name, d_name); + __assign_str(d_name); __entry->ia_valid = ia_valid; __entry->ia_mode = ia_mode; __entry->ia_uid = ia_uid; @@ -1683,7 +1683,7 @@ TRACE_EVENT(ocfs2_parse_options, ), TP_fast_assign( __entry->is_remount = is_remount; - __assign_str(options, options); + __assign_str(options); ), TP_printk("%d %s", __entry->is_remount, __get_str(options)) ); @@ -1718,8 +1718,8 @@ TRACE_EVENT(ocfs2_initialize_super, __field(int, cluster_bits) ), TP_fast_assign( - __assign_str(label, label); - __assign_str(uuid_str, uuid_str); + __assign_str(label); + __assign_str(uuid_str); __entry->root_dir = root_dir; __entry->system_dir = system_dir; __entry->cluster_bits = cluster_bits; @@ -1746,7 +1746,7 @@ TRACE_EVENT(ocfs2_init_xattr_set_ctxt, __field(int, credits) ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->meta = meta; __entry->clusters = clusters; __entry->credits = credits; @@ -1770,7 +1770,7 @@ DECLARE_EVENT_CLASS(ocfs2__xattr_find, ), TP_fast_assign( __entry->ino = ino; - __assign_str(name, name); + __assign_str(name); __entry->name_index = name_index; __entry->hash = hash; __entry->location = location; @@ -2019,7 +2019,7 @@ TRACE_EVENT(ocfs2_sync_dquot_helper, __entry->dq_id = dq_id; __entry->dq_type = dq_type; __entry->type = type; - __assign_str(s_id, s_id); + __assign_str(s_id); ), TP_printk("%u %u %lu %s", __entry->dq_id, __entry->dq_type, __entry->type, __get_str(s_id)) @@ -2060,7 +2060,7 @@ TRACE_EVENT(ocfs2_dx_dir_search, TP_fast_assign( __entry->ino = ino; __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); __entry->major_hash = major_hash; __entry->minor_hash = minor_hash; __entry->blkno = blkno; @@ -2088,7 +2088,7 @@ TRACE_EVENT(ocfs2_find_files_on_disk, ), TP_fast_assign( __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); __entry->blkno = blkno; __entry->dir = dir; ), @@ -2107,7 +2107,7 @@ TRACE_EVENT(ocfs2_check_dir_for_entry, TP_fast_assign( __entry->dir = dir; __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%llu %.*s", __entry->dir, __entry->namelen, __get_str(name)) @@ -2135,7 +2135,7 @@ TRACE_EVENT(ocfs2_dx_dir_index_root_block, __entry->major_hash = major_hash; __entry->minor_hash = minor_hash; __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); __entry->num_used = num_used; ), TP_printk("%llu %x %x %.*s %u", __entry->dir, @@ -2171,7 +2171,7 @@ DECLARE_EVENT_CLASS(ocfs2__dentry_ops, __entry->dir = dir; __entry->dentry = dentry; __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); __entry->dir_blkno = dir_blkno; __entry->extra = extra; ), @@ -2217,7 +2217,7 @@ TRACE_EVENT(ocfs2_mknod, __entry->dir = dir; __entry->dentry = dentry; __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); __entry->dir_blkno = dir_blkno; __entry->dev = dev; __entry->mode = mode; @@ -2241,9 +2241,9 @@ TRACE_EVENT(ocfs2_link, TP_fast_assign( __entry->ino = ino; __entry->old_len = old_len; - __assign_str(old_name, old_name); + __assign_str(old_name); __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%llu %.*s %.*s", __entry->ino, __entry->old_len, __get_str(old_name), @@ -2279,9 +2279,9 @@ TRACE_EVENT(ocfs2_rename, __entry->new_dir = new_dir; __entry->new_dentry = new_dentry; __entry->old_len = old_len; - __assign_str(old_name, old_name); + __assign_str(old_name); __entry->new_len = new_len; - __assign_str(new_name, new_name); + __assign_str(new_name); ), TP_printk("%p %p %p %p %.*s %.*s", __entry->old_dir, __entry->old_dentry, @@ -2301,7 +2301,7 @@ TRACE_EVENT(ocfs2_rename_target_exists, ), TP_fast_assign( __entry->new_len = new_len; - __assign_str(new_name, new_name); + __assign_str(new_name); ), TP_printk("%.*s", __entry->new_len, __get_str(new_name)) ); @@ -2344,7 +2344,7 @@ TRACE_EVENT(ocfs2_symlink_begin, __entry->dentry = dentry; __entry->symname = symname; __entry->len = len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%p %p %s %.*s", __entry->dir, __entry->dentry, __entry->symname, __entry->len, __get_str(name)) @@ -2360,7 +2360,7 @@ TRACE_EVENT(ocfs2_blkno_stringify, ), TP_fast_assign( __entry->blkno = blkno; - __assign_str(name, name); + __assign_str(name); __entry->namelen = namelen; ), TP_printk("%llu %s %d", __entry->blkno, __get_str(name), @@ -2381,7 +2381,7 @@ TRACE_EVENT(ocfs2_orphan_del, ), TP_fast_assign( __entry->dir = dir; - __assign_str(name, name); + __assign_str(name); __entry->namelen = namelen; ), TP_printk("%llu %s %d", __entry->dir, __get_str(name), @@ -2403,7 +2403,7 @@ TRACE_EVENT(ocfs2_dentry_revalidate, TP_fast_assign( __entry->dentry = dentry; __entry->len = len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%p %.*s", __entry->dentry, __entry->len, __get_str(name)) ); @@ -2420,7 +2420,7 @@ TRACE_EVENT(ocfs2_dentry_revalidate_negative, ), TP_fast_assign( __entry->len = len; - __assign_str(name, name); + __assign_str(name); __entry->pgen = pgen; __entry->gen = gen; ), @@ -2445,7 +2445,7 @@ TRACE_EVENT(ocfs2_find_local_alias, ), TP_fast_assign( __entry->len = len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%.*s", __entry->len, __get_str(name)) ); @@ -2462,7 +2462,7 @@ TRACE_EVENT(ocfs2_dentry_attach_lock, ), TP_fast_assign( __entry->len = len; - __assign_str(name, name); + __assign_str(name); __entry->parent = parent; __entry->fsdata = fsdata; ), @@ -2480,7 +2480,7 @@ TRACE_EVENT(ocfs2_dentry_attach_lock_found, __field(unsigned long long, ino) ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->parent = parent; __entry->ino = ino; ), @@ -2527,7 +2527,7 @@ TRACE_EVENT(ocfs2_get_parent, TP_fast_assign( __entry->child = child; __entry->len = len; - __assign_str(name, name); + __assign_str(name); __entry->ino = ino; ), TP_printk("%p %.*s %llu", __entry->child, __entry->len, @@ -2551,7 +2551,7 @@ TRACE_EVENT(ocfs2_encode_fh_begin, TP_fast_assign( __entry->dentry = dentry; __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); __entry->fh = fh; __entry->len = len; __entry->connectable = connectable; diff --git a/fs/smb/client/trace.h b/fs/smb/client/trace.h index af97389e983e..36d47ce59631 100644 --- a/fs/smb/client/trace.h +++ b/fs/smb/client/trace.h @@ -518,7 +518,7 @@ DECLARE_EVENT_CLASS(smb3_inf_compound_enter_class, __entry->xid = xid; __entry->tid = tid; __entry->sesid = sesid; - __assign_str(path, full_path); + __assign_str(path); ), TP_printk("xid=%u sid=0x%llx tid=0x%x path=%s", __entry->xid, __entry->sesid, __entry->tid, @@ -762,7 +762,7 @@ DECLARE_EVENT_CLASS(smb3_exit_err_class, ), TP_fast_assign( __entry->xid = xid; - __assign_str(func_name, func_name); + __assign_str(func_name); __entry->rc = rc; ), TP_printk("\t%s: xid=%u rc=%d", @@ -815,7 +815,7 @@ DECLARE_EVENT_CLASS(smb3_enter_exit_class, ), TP_fast_assign( __entry->xid = xid; - __assign_str(func_name, func_name); + __assign_str(func_name); ), TP_printk("\t%s: xid=%u", __get_str(func_name), __entry->xid) @@ -852,7 +852,7 @@ DECLARE_EVENT_CLASS(smb3_tcon_class, __entry->xid = xid; __entry->tid = tid; __entry->sesid = sesid; - __assign_str(name, unc_name); + __assign_str(name); __entry->rc = rc; ), TP_printk("xid=%u sid=0x%llx tid=0x%x unc_name=%s rc=%d", @@ -896,7 +896,7 @@ DECLARE_EVENT_CLASS(smb3_open_enter_class, __entry->xid = xid; __entry->tid = tid; __entry->sesid = sesid; - __assign_str(path, full_path); + __assign_str(path); __entry->create_options = create_options; __entry->desired_access = desired_access; ), @@ -1098,7 +1098,7 @@ DECLARE_EVENT_CLASS(smb3_connect_class, __entry->conn_id = conn_id; pss = (struct sockaddr_storage *)__entry->dst_addr; *pss = *dst_addr; - __assign_str(hostname, hostname); + __assign_str(hostname); ), TP_printk("conn_id=0x%llx server=%s addr=%pISpsfc", __entry->conn_id, @@ -1134,7 +1134,7 @@ DECLARE_EVENT_CLASS(smb3_connect_err_class, __entry->rc = rc; pss = (struct sockaddr_storage *)__entry->dst_addr; *pss = *dst_addr; - __assign_str(hostname, hostname); + __assign_str(hostname); ), TP_printk("rc=%d conn_id=0x%llx server=%s addr=%pISpsfc", __entry->rc, @@ -1166,7 +1166,7 @@ DECLARE_EVENT_CLASS(smb3_reconnect_class, TP_fast_assign( __entry->currmid = currmid; __entry->conn_id = conn_id; - __assign_str(hostname, hostname); + __assign_str(hostname); ), TP_printk("conn_id=0x%llx server=%s current_mid=%llu", __entry->conn_id, @@ -1255,7 +1255,7 @@ DECLARE_EVENT_CLASS(smb3_credit_class, TP_fast_assign( __entry->currmid = currmid; __entry->conn_id = conn_id; - __assign_str(hostname, hostname); + __assign_str(hostname); __entry->credits = credits; __entry->credits_to_add = credits_to_add; __entry->in_flight = in_flight; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index e27daa51cab6..92ef4cdc486e 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -561,7 +561,7 @@ TRACE_EVENT(xchk_btree_op_error, __entry->dev = sc->mp->m_super->s_dev; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); @@ -604,7 +604,7 @@ TRACE_EVENT(xchk_ifork_btree_op_error, __entry->ino = sc->ip->i_ino; __entry->whichfork = cur->bc_ino.whichfork; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->ptr = cur->bc_levels[level].ptr; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); @@ -644,7 +644,7 @@ TRACE_EVENT(xchk_btree_error, xfs_fsblock_t fsbno = xchk_btree_cur_fsbno(cur, level); __entry->dev = sc->mp->m_super->s_dev; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); @@ -684,7 +684,7 @@ TRACE_EVENT(xchk_ifork_btree_error, __entry->ino = sc->ip->i_ino; __entry->whichfork = cur->bc_ino.whichfork; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); @@ -723,7 +723,7 @@ DECLARE_EVENT_CLASS(xchk_sbtree_class, __entry->dev = sc->mp->m_super->s_dev; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); __entry->level = level; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 05cb59bd0b80..25ff6fe1eb6c 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -168,7 +168,7 @@ TRACE_EVENT(xlog_intent_recovery_failed, ), TP_fast_assign( __entry->dev = mp->m_super->s_dev; - __assign_str(name, ops->name); + __assign_str(name); __entry->error = error; ), TP_printk("dev %d:%d optype %s error %d", @@ -1913,7 +1913,7 @@ TRACE_EVENT(xfs_alloc_cur_check, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->bno = bno; __entry->len = len; __entry->diff = diff; @@ -2473,7 +2473,7 @@ DECLARE_EVENT_CLASS(xfs_btree_cur_class, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->nlevels = cur->bc_nlevels; __entry->ptr = cur->bc_levels[level].ptr; @@ -2523,7 +2523,7 @@ TRACE_EVENT(xfs_btree_alloc_block, __entry->ino = 0; break; } - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->error = error; if (!error && stat) { if (cur->bc_ops->ptr_len == XFS_BTREE_LONG_PTR_LEN) { @@ -2567,7 +2567,7 @@ TRACE_EVENT(xfs_btree_free_block, __entry->ino = cur->bc_ino.ip->i_ino; else __entry->ino = 0; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agbno = xfs_daddr_to_agbno(cur->bc_mp, xfs_buf_daddr(bp)); ), @@ -2643,7 +2643,7 @@ DECLARE_EVENT_CLASS(xfs_defer_pending_class, ), TP_fast_assign( __entry->dev = mp ? mp->m_super->s_dev : 0; - __assign_str(name, dfp->dfp_ops->name); + __assign_str(name); __entry->intent = dfp->dfp_intent; __entry->flags = dfp->dfp_flags; __entry->committed = dfp->dfp_done != NULL; @@ -2732,7 +2732,7 @@ DECLARE_EVENT_CLASS(xfs_defer_pending_item_class, ), TP_fast_assign( __entry->dev = mp ? mp->m_super->s_dev : 0; - __assign_str(name, dfp->dfp_ops->name); + __assign_str(name); __entry->intent = dfp->dfp_intent; __entry->item = item; __entry->committed = dfp->dfp_done != NULL; @@ -4244,7 +4244,7 @@ TRACE_EVENT(xfs_btree_commit_afakeroot, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agno = cur->bc_ag.pag->pag_agno; __entry->agbno = cur->bc_ag.afake->af_root; __entry->levels = cur->bc_ag.afake->af_levels; @@ -4273,7 +4273,7 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agno = XFS_INO_TO_AGNO(cur->bc_mp, cur->bc_ino.ip->i_ino); __entry->agino = XFS_INO_TO_AGINO(cur->bc_mp, @@ -4312,7 +4312,7 @@ TRACE_EVENT(xfs_btree_bload_level_geometry, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->nlevels = cur->bc_nlevels; __entry->nr_this_level = nr_this_level; @@ -4350,7 +4350,7 @@ TRACE_EVENT(xfs_btree_bload_block, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->block_idx = block_idx; __entry->nr_blocks = nr_blocks; @@ -4573,7 +4573,7 @@ TRACE_EVENT(xfs_force_shutdown, __entry->dev = mp->m_super->s_dev; __entry->ptag = ptag; __entry->flags = flags; - __assign_str(fname, fname); + __assign_str(fname); __entry->line_num = line_num; ), TP_printk("dev %d:%d tag %s flags %s file %s line_num %d", @@ -4755,7 +4755,7 @@ DECLARE_EVENT_CLASS(xfbtree_freesp_class, ), TP_fast_assign( __entry->xfino = file_inode(xfbt->target->bt_file)->i_ino; - __assign_str(btname, cur->bc_ops->name); + __assign_str(btname); __entry->nlevels = cur->bc_nlevels; __entry->fileoff = fileoff; ), @@ -5122,7 +5122,7 @@ DECLARE_EVENT_CLASS(xfs_getparents_rec_class, __entry->bufsize = ppi->gp_bufsize; __entry->parent_ino = pptr->gpr_parent.ha_fid.fid_ino; __entry->parent_gen = pptr->gpr_parent.ha_fid.fid_gen; - __assign_str(name, pptr->gpr_name); + __assign_str(name); ), TP_printk("dev %d:%d ino 0x%llx firstu %u reclen %u bufsize %u parent_ino 0x%llx parent_gen 0x%x name '%s'", MAJOR(__entry->dev), MINOR(__entry->dev), diff --git a/include/ras/ras_event.h b/include/ras/ras_event.h index c011ea236e9b..7c47151d5c72 100644 --- a/include/ras/ras_event.h +++ b/include/ras/ras_event.h @@ -61,7 +61,7 @@ TRACE_EVENT(extlog_mem_event, else __entry->pa_mask_lsb = ~0; __entry->fru_id = *fru_id; - __assign_str(fru_text, fru_text); + __assign_str(fru_text); cper_mem_err_pack(mem, &__entry->data); ), @@ -131,8 +131,8 @@ TRACE_EVENT(mc_event, TP_fast_assign( __entry->error_type = err_type; - __assign_str(msg, error_msg); - __assign_str(label, label); + __assign_str(msg); + __assign_str(label); __entry->error_count = error_count; __entry->mc_index = mc_index; __entry->top_layer = top_layer; @@ -141,7 +141,7 @@ TRACE_EVENT(mc_event, __entry->address = address; __entry->grain_bits = grain_bits; __entry->syndrome = syndrome; - __assign_str(driver_detail, driver_detail); + __assign_str(driver_detail); ), TP_printk("%d %s error%s:%s%s on %s (mc:%d location:%d:%d:%d address:0x%08lx grain:%d syndrome:0x%08lx%s%s)", @@ -239,7 +239,7 @@ TRACE_EVENT(non_standard_event, TP_fast_assign( memcpy(__entry->sec_type, sec_type, UUID_SIZE); memcpy(__entry->fru_id, fru_id, UUID_SIZE); - __assign_str(fru_text, fru_text); + __assign_str(fru_text); __entry->sev = sev; __entry->len = len; memcpy(__get_dynamic_array(buf), err, len); @@ -313,7 +313,7 @@ TRACE_EVENT(aer_event, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->status = status; __entry->severity = severity; __entry->tlp_header_valid = tlp_header_valid; diff --git a/include/trace/events/asoc.h b/include/trace/events/asoc.h index 517015ef36a8..202fc3680c36 100644 --- a/include/trace/events/asoc.h +++ b/include/trace/events/asoc.h @@ -32,8 +32,8 @@ DECLARE_EVENT_CLASS(snd_soc_dapm, ), TP_fast_assign( - __assign_str(card_name, dapm->card->name); - __assign_str(comp_name, dapm->component ? dapm->component->name : "(none)"); + __assign_str(card_name); + __assign_str(comp_name); __entry->val = val; ), @@ -69,7 +69,7 @@ DECLARE_EVENT_CLASS(snd_soc_dapm_basic, ), TP_fast_assign( - __assign_str(name, card->name); + __assign_str(name); __entry->event = event; ), @@ -104,7 +104,7 @@ DECLARE_EVENT_CLASS(snd_soc_dapm_widget, ), TP_fast_assign( - __assign_str(name, w->name); + __assign_str(name); __entry->val = val; ), @@ -150,7 +150,7 @@ TRACE_EVENT(snd_soc_dapm_walk_done, ), TP_fast_assign( - __assign_str(name, card->name); + __assign_str(name); __entry->power_checks = card->dapm_stats.power_checks; __entry->path_checks = card->dapm_stats.path_checks; __entry->neighbour_checks = card->dapm_stats.neighbour_checks; @@ -179,9 +179,9 @@ TRACE_EVENT(snd_soc_dapm_path, ), TP_fast_assign( - __assign_str(wname, widget->name); - __assign_str(pname, path->name ? path->name : DAPM_DIRECT); - __assign_str(pnname, path->node[dir]->name); + __assign_str(wname); + __assign_str(pname); + __assign_str(pnname); __entry->path_connect = path->connect; __entry->path_node = (long)path->node[dir]; __entry->path_dir = dir; @@ -226,7 +226,7 @@ TRACE_EVENT(snd_soc_jack_irq, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -245,7 +245,7 @@ TRACE_EVENT(snd_soc_jack_report, ), TP_fast_assign( - __assign_str(name, jack->jack->id); + __assign_str(name); __entry->mask = mask; __entry->val = val; ), @@ -266,7 +266,7 @@ TRACE_EVENT(snd_soc_jack_notify, ), TP_fast_assign( - __assign_str(name, jack->jack->id); + __assign_str(name); __entry->val = val; ), diff --git a/include/trace/events/avc.h b/include/trace/events/avc.h index b55fda2e0773..fed0f141d5f6 100644 --- a/include/trace/events/avc.h +++ b/include/trace/events/avc.h @@ -36,9 +36,9 @@ TRACE_EVENT(selinux_audited, __entry->denied = sad->denied; __entry->audited = sad->audited; __entry->result = sad->result; - __assign_str(tcontext, tcontext); - __assign_str(scontext, scontext); - __assign_str(tclass, tclass); + __assign_str(tcontext); + __assign_str(scontext); + __assign_str(tclass); ), TP_printk("requested=0x%x denied=0x%x audited=0x%x result=%d scontext=%s tcontext=%s tclass=%s", diff --git a/include/trace/events/bridge.h b/include/trace/events/bridge.h index a6b3a4e409f0..3fe4725c83ff 100644 --- a/include/trace/events/bridge.h +++ b/include/trace/events/bridge.h @@ -25,7 +25,7 @@ TRACE_EVENT(br_fdb_add, ), TP_fast_assign( - __assign_str(dev, dev->name); + __assign_str(dev); memcpy(__entry->addr, addr, ETH_ALEN); __entry->vid = vid; __entry->nlh_flags = nlh_flags; @@ -54,8 +54,8 @@ TRACE_EVENT(br_fdb_external_learn_add, ), TP_fast_assign( - __assign_str(br_dev, br->dev->name); - __assign_str(dev, p ? p->dev->name : "null"); + __assign_str(br_dev); + __assign_str(dev); memcpy(__entry->addr, addr, ETH_ALEN); __entry->vid = vid; ), @@ -80,8 +80,8 @@ TRACE_EVENT(fdb_delete, ), TP_fast_assign( - __assign_str(br_dev, br->dev->name); - __assign_str(dev, f->dst ? f->dst->dev->name : "null"); + __assign_str(br_dev); + __assign_str(dev); memcpy(__entry->addr, f->key.addr.addr, ETH_ALEN); __entry->vid = f->key.vlan_id; ), @@ -108,8 +108,8 @@ TRACE_EVENT(br_fdb_update, ), TP_fast_assign( - __assign_str(br_dev, br->dev->name); - __assign_str(dev, source->dev->name); + __assign_str(br_dev); + __assign_str(dev); memcpy(__entry->addr, addr, ETH_ALEN); __entry->vid = vid; __entry->flags = flags; @@ -141,7 +141,7 @@ TRACE_EVENT(br_mdb_full, TP_fast_assign( struct in6_addr *in6; - __assign_str(dev, dev->name); + __assign_str(dev); __entry->vid = group->vid; if (!group->proto) { diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index d2d94d7c3fb5..fadf406b5260 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -1140,7 +1140,7 @@ TRACE_EVENT(btrfs_space_reservation, ), TP_fast_assign_btrfs(fs_info, - __assign_str(type, type); + __assign_str(type); __entry->val = val; __entry->bytes = bytes; __entry->reserve = reserve; @@ -1169,7 +1169,7 @@ TRACE_EVENT(btrfs_trigger_flush, __entry->flags = flags; __entry->bytes = bytes; __entry->flush = flush; - __assign_str(reason, reason); + __assign_str(reason); ), TP_printk_btrfs("%s: flush=%d(%s) flags=%llu(%s) bytes=%llu", @@ -1622,7 +1622,7 @@ DECLARE_EVENT_CLASS(btrfs_workqueue, TP_fast_assign_btrfs(btrfs_workqueue_owner(wq), __entry->wq = wq; - __assign_str(name, name); + __assign_str(name); ), TP_printk_btrfs("name=%s wq=%p", __get_str(name), diff --git a/include/trace/events/cgroup.h b/include/trace/events/cgroup.h index 0b95865a90f3..af2755bda6eb 100644 --- a/include/trace/events/cgroup.h +++ b/include/trace/events/cgroup.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(cgroup_root, TP_fast_assign( __entry->root = root->hierarchy_id; __entry->ss_mask = root->subsys_mask; - __assign_str(name, root->name); + __assign_str(name); ), TP_printk("root=%d ss_mask=%#x name=%s", @@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(cgroup, __entry->root = cgrp->root->hierarchy_id; __entry->id = cgroup_id(cgrp); __entry->level = cgrp->level; - __assign_str(path, path); + __assign_str(path); ), TP_printk("root=%d id=%llu level=%d path=%s", @@ -137,9 +137,9 @@ DECLARE_EVENT_CLASS(cgroup_migrate, __entry->dst_root = dst_cgrp->root->hierarchy_id; __entry->dst_id = cgroup_id(dst_cgrp); __entry->dst_level = dst_cgrp->level; - __assign_str(dst_path, path); + __assign_str(dst_path); __entry->pid = task->pid; - __assign_str(comm, task->comm); + __assign_str(comm); ), TP_printk("dst_root=%d dst_id=%llu dst_level=%d dst_path=%s pid=%d comm=%s", @@ -181,7 +181,7 @@ DECLARE_EVENT_CLASS(cgroup_event, __entry->root = cgrp->root->hierarchy_id; __entry->id = cgroup_id(cgrp); __entry->level = cgrp->level; - __assign_str(path, path); + __assign_str(path); __entry->val = val; ), diff --git a/include/trace/events/clk.h b/include/trace/events/clk.h index daed3c7a48c1..759f7371a6dc 100644 --- a/include/trace/events/clk.h +++ b/include/trace/events/clk.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(clk, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -97,7 +97,7 @@ DECLARE_EVENT_CLASS(clk_rate, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->rate = rate; ), @@ -145,7 +145,7 @@ DECLARE_EVENT_CLASS(clk_rate_range, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->min = min; __entry->max = max; ), @@ -174,8 +174,8 @@ DECLARE_EVENT_CLASS(clk_parent, ), TP_fast_assign( - __assign_str(name, core->name); - __assign_str(pname, parent ? parent->name : "none"); + __assign_str(name); + __assign_str(pname); ), TP_printk("%s %s", __get_str(name), __get_str(pname)) @@ -207,7 +207,7 @@ DECLARE_EVENT_CLASS(clk_phase, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->phase = phase; ), @@ -241,7 +241,7 @@ DECLARE_EVENT_CLASS(clk_duty_cycle, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->num = duty->num; __entry->den = duty->den; ), @@ -279,8 +279,8 @@ DECLARE_EVENT_CLASS(clk_rate_request, ), TP_fast_assign( - __assign_str(name, req->core ? req->core->name : "none"); - __assign_str(pname, req->best_parent_hw ? clk_hw_get_name(req->best_parent_hw) : "none"); + __assign_str(name); + __assign_str(pname); __entry->min = req->min_rate; __entry->max = req->max_rate; __entry->prate = req->best_parent_rate; diff --git a/include/trace/events/cma.h b/include/trace/events/cma.h index 25103e67737c..383c09f583ac 100644 --- a/include/trace/events/cma.h +++ b/include/trace/events/cma.h @@ -23,7 +23,7 @@ TRACE_EVENT(cma_release, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->pfn = pfn; __entry->page = page; __entry->count = count; @@ -49,7 +49,7 @@ TRACE_EVENT(cma_alloc_start, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->count = count; __entry->align = align; ), @@ -77,7 +77,7 @@ TRACE_EVENT(cma_alloc_finish, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->pfn = pfn; __entry->page = page; __entry->count = count; @@ -110,7 +110,7 @@ TRACE_EVENT(cma_alloc_busy_retry, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->pfn = pfn; __entry->page = page; __entry->count = count; diff --git a/include/trace/events/devfreq.h b/include/trace/events/devfreq.h index 7627c620bbda..6cbc4d59fd96 100644 --- a/include/trace/events/devfreq.h +++ b/include/trace/events/devfreq.h @@ -23,7 +23,7 @@ TRACE_EVENT(devfreq_frequency, ), TP_fast_assign( - __assign_str(dev_name, dev_name(&devfreq->dev)); + __assign_str(dev_name); __entry->freq = freq; __entry->prev_freq = prev_freq; __entry->busy_time = devfreq->last_status.busy_time; @@ -54,7 +54,7 @@ TRACE_EVENT(devfreq_monitor, __entry->busy_time = devfreq->last_status.busy_time; __entry->total_time = devfreq->last_status.total_time; __entry->polling_ms = devfreq->profile->polling_ms; - __assign_str(dev_name, dev_name(&devfreq->dev)); + __assign_str(dev_name); ), TP_printk("dev_name=%-30s freq=%-12lu polling_ms=%-3u load=%-2lu", diff --git a/include/trace/events/devlink.h b/include/trace/events/devlink.h index 77ff7cfc6049..f241e204fe6b 100644 --- a/include/trace/events/devlink.h +++ b/include/trace/events/devlink.h @@ -31,9 +31,9 @@ TRACE_EVENT(devlink_hwmsg, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); __entry->incoming = incoming; __entry->type = type; memcpy(__get_dynamic_array(buf), buf, len); @@ -63,11 +63,11 @@ TRACE_EVENT(devlink_hwerr, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); __entry->err = err; - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("bus_name=%s dev_name=%s driver_name=%s err=%d %s", @@ -93,11 +93,11 @@ TRACE_EVENT(devlink_health_report, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(reporter_name, reporter_name); - __assign_str(msg, msg); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(reporter_name); + __assign_str(msg); ), TP_printk("bus_name=%s dev_name=%s driver_name=%s reporter_name=%s: %s", @@ -125,10 +125,10 @@ TRACE_EVENT(devlink_health_recover_aborted, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(reporter_name, reporter_name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(reporter_name); __entry->health_state = health_state; __entry->time_since_last_recover = time_since_last_recover; ), @@ -158,10 +158,10 @@ TRACE_EVENT(devlink_health_reporter_state_update, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(reporter_name, reporter_name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(reporter_name); __entry->new_state = new_state; ), @@ -192,11 +192,11 @@ TRACE_EVENT(devlink_trap_report, TP_fast_assign( struct net_device *input_dev = metadata->input_dev; - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(trap_name, metadata->trap_name); - __assign_str(trap_group_name, metadata->trap_group_name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(trap_name); + __assign_str(trap_group_name); strscpy(__entry->input_dev_name, input_dev ? input_dev->name : "NULL", IFNAMSIZ); ), diff --git a/include/trace/events/dma_fence.h b/include/trace/events/dma_fence.h index 3963e79ca7b4..a4de3df8500b 100644 --- a/include/trace/events/dma_fence.h +++ b/include/trace/events/dma_fence.h @@ -23,8 +23,8 @@ DECLARE_EVENT_CLASS(dma_fence, ), TP_fast_assign( - __assign_str(driver, fence->ops->get_driver_name(fence)); - __assign_str(timeline, fence->ops->get_timeline_name(fence)); + __assign_str(driver); + __assign_str(timeline); __entry->context = fence->context; __entry->seqno = fence->seqno; ), diff --git a/include/trace/events/erofs.h b/include/trace/events/erofs.h index e18684b02c3d..b9bbfd855f2a 100644 --- a/include/trace/events/erofs.h +++ b/include/trace/events/erofs.h @@ -47,7 +47,7 @@ TRACE_EVENT(erofs_lookup, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->nid = EROFS_I(dir)->nid; - __assign_str(name, dentry->d_name.name); + __assign_str(name); __entry->flags = flags; ), diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 371ba28415f5..ed794b5fefbe 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -354,7 +354,7 @@ TRACE_EVENT(f2fs_unlink_enter, __entry->ino = dir->i_ino; __entry->size = dir->i_size; __entry->blocks = dir->i_blocks; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk("dev = (%d,%d), dir ino = %lu, i_size = %lld, " @@ -843,7 +843,7 @@ TRACE_EVENT(f2fs_lookup_start, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->ino = dir->i_ino; - __assign_str(name, dentry->d_name.name); + __assign_str(name); __entry->flags = flags; ), @@ -871,7 +871,7 @@ TRACE_EVENT(f2fs_lookup_end, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->ino = dir->i_ino; - __assign_str(name, dentry->d_name.name); + __assign_str(name); __entry->cino = ino; __entry->err = err; ), @@ -903,9 +903,9 @@ TRACE_EVENT(f2fs_rename_start, TP_fast_assign( __entry->dev = old_dir->i_sb->s_dev; __entry->ino = old_dir->i_ino; - __assign_str(old_name, old_dentry->d_name.name); + __assign_str(old_name); __entry->new_pino = new_dir->i_ino; - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(new_name); __entry->flags = flags; ), @@ -937,8 +937,8 @@ TRACE_EVENT(f2fs_rename_end, TP_fast_assign( __entry->dev = old_dentry->d_sb->s_dev; __entry->ino = old_dentry->d_inode->i_ino; - __assign_str(old_name, old_dentry->d_name.name); - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(old_name); + __assign_str(new_name); __entry->flags = flags; __entry->ret = ret; ), @@ -1557,7 +1557,7 @@ TRACE_EVENT(f2fs_write_checkpoint, TP_fast_assign( __entry->dev = sb->s_dev; __entry->reason = reason; - __assign_str(dest_msg, msg); + __assign_str(dest_msg); ), TP_printk("dev = (%d,%d), checkpoint for %s, state = %s", @@ -2333,12 +2333,12 @@ DECLARE_EVENT_CLASS(f2fs__rw_start, * because this screws up the tooling that parses * the traces. */ - __assign_str(pathbuf, pathname); + __assign_str(pathbuf); (void)strreplace(__get_str(pathbuf), ' ', '_'); __entry->offset = offset; __entry->bytes = bytes; __entry->i_size = i_size_read(inode); - __assign_str(cmdline, command); + __assign_str(cmdline); (void)strreplace(__get_str(cmdline), ' ', '_'); __entry->pid = pid; __entry->ino = inode->i_ino; diff --git a/include/trace/events/habanalabs.h b/include/trace/events/habanalabs.h index a78d21fa9f29..4a2bb2c896d1 100644 --- a/include/trace/events/habanalabs.h +++ b/include/trace/events/habanalabs.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(habanalabs_mmu_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->virt_addr = virt_addr; __entry->phys_addr = phys_addr; __entry->page_size = page_size; @@ -64,7 +64,7 @@ DECLARE_EVENT_CLASS(habanalabs_dma_alloc_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->cpu_addr = cpu_addr; __entry->dma_addr = dma_addr; __entry->size = size; @@ -103,7 +103,7 @@ DECLARE_EVENT_CLASS(habanalabs_dma_map_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->phys_addr = phys_addr; __entry->dma_addr = dma_addr; __entry->len = len; @@ -141,7 +141,7 @@ DECLARE_EVENT_CLASS(habanalabs_comms_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->op_str = op_str; ), @@ -178,7 +178,7 @@ DECLARE_EVENT_CLASS(habanalabs_reg_access_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->addr = addr; __entry->val = val; ), diff --git a/include/trace/events/huge_memory.h b/include/trace/events/huge_memory.h index ab576898a126..b5f5369b6300 100644 --- a/include/trace/events/huge_memory.h +++ b/include/trace/events/huge_memory.h @@ -191,7 +191,7 @@ TRACE_EVENT(mm_khugepaged_scan_file, TP_fast_assign( __entry->mm = mm; __entry->pfn = folio ? folio_pfn(folio) : -1; - __assign_str(filename, file->f_path.dentry->d_iname); + __assign_str(filename); __entry->present = present; __entry->swap = swap; __entry->result = result; @@ -228,7 +228,7 @@ TRACE_EVENT(mm_khugepaged_collapse_file, __entry->index = index; __entry->addr = addr; __entry->is_shmem = is_shmem; - __assign_str(filename, file->f_path.dentry->d_iname); + __assign_str(filename); __entry->nr = nr; __entry->result = result; ), diff --git a/include/trace/events/hwmon.h b/include/trace/events/hwmon.h index d7a1d0ffb679..d1ff560cd9b5 100644 --- a/include/trace/events/hwmon.h +++ b/include/trace/events/hwmon.h @@ -21,7 +21,7 @@ DECLARE_EVENT_CLASS(hwmon_attr_class, TP_fast_assign( __entry->index = index; - __assign_str(attr_name, attr_name); + __assign_str(attr_name); __entry->val = val; ), @@ -57,8 +57,8 @@ TRACE_EVENT(hwmon_attr_show_string, TP_fast_assign( __entry->index = index; - __assign_str(attr_name, attr_name); - __assign_str(label, s); + __assign_str(attr_name); + __assign_str(label); ), TP_printk("index=%d, attr_name=%s, val=%s", diff --git a/include/trace/events/initcall.h b/include/trace/events/initcall.h index eb903c3f195f..5282afdf3ddf 100644 --- a/include/trace/events/initcall.h +++ b/include/trace/events/initcall.h @@ -18,7 +18,7 @@ TRACE_EVENT(initcall_level, ), TP_fast_assign( - __assign_str(level, level); + __assign_str(level); ), TP_printk("level=%s", __get_str(level)) diff --git a/include/trace/events/intel_ish.h b/include/trace/events/intel_ish.h index e6d7ff55ee8c..64b6612c41bc 100644 --- a/include/trace/events/intel_ish.h +++ b/include/trace/events/intel_ish.h @@ -18,7 +18,7 @@ TRACE_EVENT(ishtp_dump, ), TP_fast_assign( - __assign_str(message, message); + __assign_str(message); ), TP_printk("%s", __get_str(message)) diff --git a/include/trace/events/io_uring.h b/include/trace/events/io_uring.h index e948df7ce625..412c9c210a32 100644 --- a/include/trace/events/io_uring.h +++ b/include/trace/events/io_uring.h @@ -164,7 +164,7 @@ TRACE_EVENT(io_uring_queue_async_work, __entry->work = &req->work; __entry->rw = rw; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, request %p, user_data 0x%llx, opcode %s, flags 0x%llx, %s queue, work %p", @@ -202,7 +202,7 @@ TRACE_EVENT(io_uring_defer, __entry->data = req->cqe.user_data; __entry->opcode = req->opcode; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, request %p, user_data 0x%llx, opcode %s", @@ -303,7 +303,7 @@ TRACE_EVENT(io_uring_fail_link, __entry->opcode = req->opcode; __entry->link = link; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, request %p, user_data 0x%llx, opcode %s, link %p", @@ -392,7 +392,7 @@ TRACE_EVENT(io_uring_submit_req, __entry->flags = (__force unsigned long long) req->flags; __entry->sq_thread = req->ctx->flags & IORING_SETUP_SQPOLL; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, opcode %s, flags 0x%llx, " @@ -436,7 +436,7 @@ TRACE_EVENT(io_uring_poll_arm, __entry->mask = mask; __entry->events = events; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, opcode %s, mask 0x%x, events 0x%x", @@ -475,7 +475,7 @@ TRACE_EVENT(io_uring_task_add, __entry->opcode = req->opcode; __entry->mask = mask; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, opcode %s, mask %x", @@ -538,7 +538,7 @@ TRACE_EVENT(io_uring_req_failed, __entry->addr3 = sqe->addr3; __entry->error = error; - __assign_str(op_str, io_uring_get_opcode(sqe->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, " diff --git a/include/trace/events/iocost.h b/include/trace/events/iocost.h index af8bfed528fc..e772b1bc60d6 100644 --- a/include/trace/events/iocost.h +++ b/include/trace/events/iocost.h @@ -34,8 +34,8 @@ DECLARE_EVENT_CLASS(iocost_iocg_state, ), TP_fast_assign( - __assign_str(devname, ioc_name(iocg->ioc)); - __assign_str(cgroup, path); + __assign_str(devname); + __assign_str(cgroup); __entry->now = now->now; __entry->vnow = now->vnow; __entry->vrate = iocg->ioc->vtime_base_rate; @@ -93,8 +93,8 @@ DECLARE_EVENT_CLASS(iocg_inuse_update, ), TP_fast_assign( - __assign_str(devname, ioc_name(iocg->ioc)); - __assign_str(cgroup, path); + __assign_str(devname); + __assign_str(cgroup); __entry->now = now->now; __entry->old_inuse = old_inuse; __entry->new_inuse = new_inuse; @@ -159,7 +159,7 @@ TRACE_EVENT(iocost_ioc_vrate_adj, ), TP_fast_assign( - __assign_str(devname, ioc_name(ioc)); + __assign_str(devname); __entry->old_vrate = ioc->vtime_base_rate; __entry->new_vrate = new_vrate; __entry->busy_level = ioc->busy_level; @@ -200,8 +200,8 @@ TRACE_EVENT(iocost_iocg_forgive_debt, ), TP_fast_assign( - __assign_str(devname, ioc_name(iocg->ioc)); - __assign_str(cgroup, path); + __assign_str(devname); + __assign_str(cgroup); __entry->now = now->now; __entry->vnow = now->vnow; __entry->usage_pct = usage_pct; diff --git a/include/trace/events/iommu.h b/include/trace/events/iommu.h index 70743db1fb75..373007e567cb 100644 --- a/include/trace/events/iommu.h +++ b/include/trace/events/iommu.h @@ -28,7 +28,7 @@ DECLARE_EVENT_CLASS(iommu_group_event, TP_fast_assign( __entry->gid = group_id; - __assign_str(device, dev_name(dev)); + __assign_str(device); ), TP_printk("IOMMU: groupID=%d device=%s", @@ -62,7 +62,7 @@ DECLARE_EVENT_CLASS(iommu_device_event, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); + __assign_str(device); ), TP_printk("IOMMU: device=%s", __get_str(device) @@ -138,8 +138,8 @@ DECLARE_EVENT_CLASS(iommu_error, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); - __assign_str(driver, dev_driver_string(dev)); + __assign_str(device); + __assign_str(driver); __entry->iova = iova; __entry->flags = flags; ), diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h index a07b4607b663..837c1740d0d0 100644 --- a/include/trace/events/irq.h +++ b/include/trace/events/irq.h @@ -63,7 +63,7 @@ TRACE_EVENT(irq_handler_entry, TP_fast_assign( __entry->irq = irq; - __assign_str(name, action->name); + __assign_str(name); ), TP_printk("irq=%d name=%s", __entry->irq, __get_str(name)) diff --git a/include/trace/events/iscsi.h b/include/trace/events/iscsi.h index 8ff2a3ca5d75..990fd154f586 100644 --- a/include/trace/events/iscsi.h +++ b/include/trace/events/iscsi.h @@ -30,7 +30,7 @@ DECLARE_EVENT_CLASS(iscsi_log_msg, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/include/trace/events/kmem.h b/include/trace/events/kmem.h index 6e62cc64cd92..8a829e0f6e55 100644 --- a/include/trace/events/kmem.h +++ b/include/trace/events/kmem.h @@ -126,7 +126,7 @@ TRACE_EVENT(kmem_cache_free, TP_fast_assign( __entry->call_site = call_site; __entry->ptr = ptr; - __assign_str(name, s->name); + __assign_str(name); ), TP_printk("call_site=%pS ptr=%p name=%s", diff --git a/include/trace/events/lock.h b/include/trace/events/lock.h index 9ebd081e057e..8e89baa3775f 100644 --- a/include/trace/events/lock.h +++ b/include/trace/events/lock.h @@ -37,7 +37,7 @@ TRACE_EVENT(lock_acquire, TP_fast_assign( __entry->flags = (trylock ? 1 : 0) | (read ? 2 : 0); - __assign_str(name, lock->name); + __assign_str(name); __entry->lockdep_addr = lock; ), @@ -59,7 +59,7 @@ DECLARE_EVENT_CLASS(lock, ), TP_fast_assign( - __assign_str(name, lock->name); + __assign_str(name); __entry->lockdep_addr = lock; ), diff --git a/include/trace/events/mmap_lock.h b/include/trace/events/mmap_lock.h index 14db8044c1ff..f2827f98a44f 100644 --- a/include/trace/events/mmap_lock.h +++ b/include/trace/events/mmap_lock.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(mmap_lock, TP_fast_assign( __entry->mm = mm; - __assign_str(memcg_path, memcg_path); + __assign_str(memcg_path); __entry->write = write; ), @@ -65,7 +65,7 @@ TRACE_EVENT_FN(mmap_lock_acquire_returned, TP_fast_assign( __entry->mm = mm; - __assign_str(memcg_path, memcg_path); + __assign_str(memcg_path); __entry->write = write; __entry->success = success; ), diff --git a/include/trace/events/mmc.h b/include/trace/events/mmc.h index 7b706ff21335..f1c2e94f7f68 100644 --- a/include/trace/events/mmc.h +++ b/include/trace/events/mmc.h @@ -68,7 +68,7 @@ TRACE_EVENT(mmc_request_start, __entry->need_retune = host->need_retune; __entry->hold_retune = host->hold_retune; __entry->retune_period = host->retune_period; - __assign_str(name, mmc_hostname(host)); + __assign_str(name); __entry->mrq = mrq; ), @@ -156,7 +156,7 @@ TRACE_EVENT(mmc_request_done, __entry->need_retune = host->need_retune; __entry->hold_retune = host->hold_retune; __entry->retune_period = host->retune_period; - __assign_str(name, mmc_hostname(host)); + __assign_str(name); __entry->mrq = mrq; ), diff --git a/include/trace/events/module.h b/include/trace/events/module.h index 097485c73c01..e5a006be9dc6 100644 --- a/include/trace/events/module.h +++ b/include/trace/events/module.h @@ -41,7 +41,7 @@ TRACE_EVENT(module_load, TP_fast_assign( __entry->taints = mod->taints; - __assign_str(name, mod->name); + __assign_str(name); ), TP_printk("%s %s", __get_str(name), show_module_flags(__entry->taints)) @@ -58,7 +58,7 @@ TRACE_EVENT(module_free, ), TP_fast_assign( - __assign_str(name, mod->name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -82,7 +82,7 @@ DECLARE_EVENT_CLASS(module_refcnt, TP_fast_assign( __entry->ip = ip; __entry->refcnt = atomic_read(&mod->refcnt); - __assign_str(name, mod->name); + __assign_str(name); ), TP_printk("%s call_site=%ps refcnt=%d", @@ -119,7 +119,7 @@ TRACE_EVENT(module_request, TP_fast_assign( __entry->ip = ip; __entry->wait = wait; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%s wait=%d call_site=%ps", diff --git a/include/trace/events/napi.h b/include/trace/events/napi.h index dc03cf8e0369..b567b9ffedc1 100644 --- a/include/trace/events/napi.h +++ b/include/trace/events/napi.h @@ -26,7 +26,7 @@ TRACE_EVENT(napi_poll, TP_fast_assign( __entry->napi = napi; - __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); + __assign_str(dev_name); __entry->work = work; __entry->budget = budget; ), diff --git a/include/trace/events/neigh.h b/include/trace/events/neigh.h index 833143d0992e..12362c35dbc0 100644 --- a/include/trace/events/neigh.h +++ b/include/trace/events/neigh.h @@ -42,7 +42,7 @@ TRACE_EVENT(neigh_create, __be32 *p32; __entry->family = tbl->family; - __assign_str(dev, (dev ? dev->name : "NULL")); + __assign_str(dev); __entry->entries = atomic_read(&tbl->gc_entries); __entry->created = n != NULL; __entry->gc_exempt = exempt_from_gc; @@ -103,7 +103,7 @@ TRACE_EVENT(neigh_update, __be32 *p32; __entry->family = n->tbl->family; - __assign_str(dev, (n->dev ? n->dev->name : "NULL")); + __assign_str(dev); __entry->lladdr_len = lladdr_len; memcpy(__entry->lladdr, n->ha, lladdr_len); __entry->flags = n->flags; @@ -180,7 +180,7 @@ DECLARE_EVENT_CLASS(neigh__update, __be32 *p32; __entry->family = n->tbl->family; - __assign_str(dev, (n->dev ? n->dev->name : "NULL")); + __assign_str(dev); __entry->lladdr_len = lladdr_len; memcpy(__entry->lladdr, n->ha, lladdr_len); __entry->flags = n->flags; diff --git a/include/trace/events/net.h b/include/trace/events/net.h index f667c76a3b02..d55162c12f90 100644 --- a/include/trace/events/net.h +++ b/include/trace/events/net.h @@ -38,7 +38,7 @@ TRACE_EVENT(net_dev_start_xmit, ), TP_fast_assign( - __assign_str(name, dev->name); + __assign_str(name); __entry->queue_mapping = skb->queue_mapping; __entry->skbaddr = skb; __entry->vlan_tagged = skb_vlan_tag_present(skb); @@ -89,7 +89,7 @@ TRACE_EVENT(net_dev_xmit, __entry->skbaddr = skb; __entry->len = skb_len; __entry->rc = rc; - __assign_str(name, dev->name); + __assign_str(name); ), TP_printk("dev=%s skbaddr=%p len=%u rc=%d", @@ -110,8 +110,8 @@ TRACE_EVENT(net_dev_xmit_timeout, ), TP_fast_assign( - __assign_str(name, dev->name); - __assign_str(driver, netdev_drivername(dev)); + __assign_str(name); + __assign_str(driver); __entry->queue_index = queue_index; ), @@ -134,7 +134,7 @@ DECLARE_EVENT_CLASS(net_dev_template, TP_fast_assign( __entry->skbaddr = skb; __entry->len = skb->len; - __assign_str(name, skb->dev->name); + __assign_str(name); ), TP_printk("dev=%s skbaddr=%p len=%u", @@ -191,7 +191,7 @@ DECLARE_EVENT_CLASS(net_dev_rx_verbose_template, ), TP_fast_assign( - __assign_str(name, skb->dev->name); + __assign_str(name); #ifdef CONFIG_NET_RX_BUSY_POLL __entry->napi_id = skb->napi_id; #else diff --git a/include/trace/events/netlink.h b/include/trace/events/netlink.h index 3b7be3b386a4..f036b8a20505 100644 --- a/include/trace/events/netlink.h +++ b/include/trace/events/netlink.h @@ -17,7 +17,7 @@ TRACE_EVENT(netlink_extack, ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("msg=%s", __get_str(msg)) diff --git a/include/trace/events/oom.h b/include/trace/events/oom.h index b799f3bcba82..a42be4c8563b 100644 --- a/include/trace/events/oom.h +++ b/include/trace/events/oom.h @@ -92,7 +92,7 @@ TRACE_EVENT(mark_victim, TP_fast_assign( __entry->pid = task->pid; - __assign_str(comm, task->comm); + __assign_str(comm); __entry->total_vm = PG_COUNT_TO_KB(task->mm->total_vm); __entry->anon_rss = PG_COUNT_TO_KB(get_mm_counter(task->mm, MM_ANONPAGES)); __entry->file_rss = PG_COUNT_TO_KB(get_mm_counter(task->mm, MM_FILEPAGES)); diff --git a/include/trace/events/osnoise.h b/include/trace/events/osnoise.h index 82f741ec0f57..a2379a4f0684 100644 --- a/include/trace/events/osnoise.h +++ b/include/trace/events/osnoise.h @@ -75,7 +75,7 @@ TRACE_EVENT(irq_noise, ), TP_fast_assign( - __assign_str(desc, desc); + __assign_str(desc); __entry->vector = vector; __entry->start = start; __entry->duration = duration; diff --git a/include/trace/events/power.h b/include/trace/events/power.h index 77f14f7a11d4..d2349b6b531a 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -76,7 +76,7 @@ TRACE_EVENT(powernv_throttle, TP_fast_assign( __entry->chip_id = chip_id; - __assign_str(reason, reason); + __assign_str(reason); __entry->pmax = pmax; ), @@ -210,11 +210,10 @@ TRACE_EVENT(device_pm_callback_start, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); - __assign_str(driver, dev_driver_string(dev)); - __assign_str(parent, - dev->parent ? dev_name(dev->parent) : "none"); - __assign_str(pm_ops, pm_ops ? pm_ops : "none "); + __assign_str(device); + __assign_str(driver); + __assign_str(parent); + __assign_str(pm_ops); __entry->event = event; ), @@ -236,8 +235,8 @@ TRACE_EVENT(device_pm_callback_end, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); - __assign_str(driver, dev_driver_string(dev)); + __assign_str(device); + __assign_str(driver); __entry->error = error; ), @@ -279,7 +278,7 @@ DECLARE_EVENT_CLASS(wakeup_source, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->state = state; ), @@ -318,7 +317,7 @@ DECLARE_EVENT_CLASS(clock, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->state = state; __entry->cpu_id = cpu_id; ), @@ -364,7 +363,7 @@ DECLARE_EVENT_CLASS(power_domain, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->state = state; __entry->cpu_id = cpu_id; ), @@ -486,7 +485,7 @@ DECLARE_EVENT_CLASS(dev_pm_qos_request, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->type = type; __entry->new_value = new_value; ), diff --git a/include/trace/events/pwc.h b/include/trace/events/pwc.h index a2da764a3b41..0543702542d9 100644 --- a/include/trace/events/pwc.h +++ b/include/trace/events/pwc.h @@ -26,7 +26,7 @@ TRACE_EVENT(pwc_handler_enter, __entry->urb__actual_length = urb->actual_length; __entry->fbuf__filled = (pdev->fill_buf ? pdev->fill_buf->filled : 0); - __assign_str(name, pdev->v4l2_dev.name); + __assign_str(name); ), TP_printk("dev=%s (fbuf=%p filled=%d) urb=%p (status=%d actual_length=%u)", __get_str(name), @@ -50,7 +50,7 @@ TRACE_EVENT(pwc_handler_exit, __entry->urb = urb; __entry->fbuf = pdev->fill_buf; __entry->fbuf__filled = pdev->fill_buf->filled; - __assign_str(name, pdev->v4l2_dev.name); + __assign_str(name); ), TP_printk(" dev=%s (fbuf=%p filled=%d) urb=%p", __get_str(name), diff --git a/include/trace/events/qdisc.h b/include/trace/events/qdisc.h index 1f4258308b96..f1b5e816e7e5 100644 --- a/include/trace/events/qdisc.h +++ b/include/trace/events/qdisc.h @@ -88,8 +88,8 @@ TRACE_EVENT(qdisc_reset, ), TP_fast_assign( - __assign_str(dev, qdisc_dev(q)->name); - __assign_str(kind, q->ops->id); + __assign_str(dev); + __assign_str(kind); __entry->parent = q->parent; __entry->handle = q->handle; ), @@ -113,8 +113,8 @@ TRACE_EVENT(qdisc_destroy, ), TP_fast_assign( - __assign_str(dev, qdisc_dev(q)->name); - __assign_str(kind, q->ops->id); + __assign_str(dev); + __assign_str(kind); __entry->parent = q->parent; __entry->handle = q->handle; ), @@ -137,8 +137,8 @@ TRACE_EVENT(qdisc_create, ), TP_fast_assign( - __assign_str(dev, dev->name); - __assign_str(kind, ops->id); + __assign_str(dev); + __assign_str(kind); __entry->parent = parent; ), diff --git a/include/trace/events/qla.h b/include/trace/events/qla.h index e7fd55e7dc3d..8800c35525a1 100644 --- a/include/trace/events/qla.h +++ b/include/trace/events/qla.h @@ -25,7 +25,7 @@ DECLARE_EVENT_CLASS(qla_log_event, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(buf, buf); + __assign_str(buf); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/include/trace/events/qrtr.h b/include/trace/events/qrtr.h index 441132c67133..14f822983741 100644 --- a/include/trace/events/qrtr.h +++ b/include/trace/events/qrtr.h @@ -102,7 +102,7 @@ TRACE_EVENT(qrtr_ns_message, ), TP_fast_assign( - __assign_str(ctrl_pkt_str, ctrl_pkt_str); + __assign_str(ctrl_pkt_str); __entry->sq_node = sq_node; __entry->sq_port = sq_port; ), diff --git a/include/trace/events/regulator.h b/include/trace/events/regulator.h index 72b3ba93b0a5..c58481a5d955 100644 --- a/include/trace/events/regulator.h +++ b/include/trace/events/regulator.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(regulator_basic, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); ), TP_printk("name=%s", __get_str(name)) @@ -119,7 +119,7 @@ DECLARE_EVENT_CLASS(regulator_range, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->min = min; __entry->max = max; ), @@ -152,7 +152,7 @@ DECLARE_EVENT_CLASS(regulator_value, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->val = val; ), diff --git a/include/trace/events/rpcgss.h b/include/trace/events/rpcgss.h index f50fcafc69de..7f0c1ceae726 100644 --- a/include/trace/events/rpcgss.h +++ b/include/trace/events/rpcgss.h @@ -154,7 +154,7 @@ DECLARE_EVENT_CLASS(rpcgss_ctx_class, TP_fast_assign( __entry->cred = gc; __entry->service = gc->gc_service; - __assign_str(principal, gc->gc_principal); + __assign_str(principal); ), TP_printk("cred=%p service=%s principal='%s'", @@ -189,7 +189,7 @@ DECLARE_EVENT_CLASS(rpcgss_svc_gssapi_class, TP_fast_assign( __entry->xid = __be32_to_cpu(rqstp->rq_xid); __entry->maj_stat = maj_stat; - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x maj_stat=%s", @@ -225,7 +225,7 @@ TRACE_EVENT(rpcgss_svc_wrap_failed, TP_fast_assign( __entry->xid = be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x", __get_str(addr), __entry->xid) @@ -245,7 +245,7 @@ TRACE_EVENT(rpcgss_svc_unwrap_failed, TP_fast_assign( __entry->xid = be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x", __get_str(addr), __entry->xid) @@ -271,7 +271,7 @@ TRACE_EVENT(rpcgss_svc_seqno_bad, __entry->expected = expected; __entry->received = received; __entry->xid = __be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x expected seqno %u, received seqno %u", @@ -299,7 +299,7 @@ TRACE_EVENT(rpcgss_svc_accept_upcall, __entry->minor_status = minor_status; __entry->major_status = major_status; __entry->xid = be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x major_status=%s (0x%08lx) minor_status=%u", @@ -327,7 +327,7 @@ TRACE_EVENT(rpcgss_svc_authenticate, TP_fast_assign( __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->seqno = gc->gc_seq; - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x seqno=%u", __get_str(addr), @@ -563,7 +563,7 @@ TRACE_EVENT(rpcgss_upcall_msg, ), TP_fast_assign( - __assign_str(msg, buf); + __assign_str(msg); ), TP_printk("msg='%s'", __get_str(msg)) @@ -618,7 +618,7 @@ TRACE_EVENT(rpcgss_context, __entry->timeout = timeout; __entry->window_size = window_size; __entry->len = len; - __assign_str(acceptor, data); + __assign_str(acceptor); ), TP_printk("win_size=%u expiry=%lu now=%lu timeout=%u acceptor=%.*s", @@ -677,7 +677,7 @@ TRACE_EVENT(rpcgss_oid_to_mech, ), TP_fast_assign( - __assign_str(oid, oid); + __assign_str(oid); ), TP_printk("mech for oid %s was not found", __get_str(oid)) diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h index 027ac3ab457d..14392652273a 100644 --- a/include/trace/events/rpcrdma.h +++ b/include/trace/events/rpcrdma.h @@ -304,8 +304,8 @@ DECLARE_EVENT_CLASS(xprtrdma_reply_class, __entry->xid = be32_to_cpu(rep->rr_xid); __entry->version = be32_to_cpu(rep->rr_vers); __entry->proc = be32_to_cpu(rep->rr_proc); - __assign_str(addr, rpcrdma_addrstr(rep->rr_rxprt)); - __assign_str(port, rpcrdma_portstr(rep->rr_rxprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s xid=0x%08x version=%u proc=%u", @@ -335,8 +335,8 @@ DECLARE_EVENT_CLASS(xprtrdma_rxprt, ), TP_fast_assign( - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s", @@ -369,8 +369,8 @@ DECLARE_EVENT_CLASS(xprtrdma_connect_class, TP_fast_assign( __entry->rc = rc; __entry->connect_status = r_xprt->rx_ep->re_connect_status; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s rc=%d connection status=%d", @@ -608,8 +608,8 @@ DECLARE_EVENT_CLASS(xprtrdma_callback_class, TP_fast_assign( __entry->xid = be32_to_cpu(rqst->rq_xid); - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s xid=0x%08x", @@ -687,8 +687,8 @@ TRACE_EVENT(xprtrdma_op_connect, TP_fast_assign( __entry->delay = delay; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s delay=%lu", @@ -716,8 +716,8 @@ TRACE_EVENT(xprtrdma_op_set_cto, TP_fast_assign( __entry->connect = connect; __entry->reconnect = reconnect; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s connect=%lu reconnect=%lu", @@ -746,8 +746,8 @@ TRACE_EVENT(xprtrdma_createmrs, TP_fast_assign( __entry->count = count; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s created %u MRs", @@ -775,8 +775,8 @@ TRACE_EVENT(xprtrdma_nomrs_err, __entry->task_id = rqst->rq_task->tk_pid; __entry->client_id = rqst->rq_task->tk_client->cl_clid; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " peer=[%s]:%s", @@ -1001,8 +1001,8 @@ TRACE_EVENT(xprtrdma_post_recvs, __entry->cq_id = ep->re_attr.recv_cq->res.id; __entry->count = count; __entry->posted = ep->re_receive_count; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s cq.id=%d %u new recvs, %d active", @@ -1031,8 +1031,8 @@ TRACE_EVENT(xprtrdma_post_recvs_err, __entry->cq_id = ep->re_attr.recv_cq->res.id; __entry->status = status; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s cq.id=%d rc=%d", @@ -1445,8 +1445,8 @@ TRACE_EVENT(xprtrdma_cb_setup, TP_fast_assign( __entry->reqs = reqs; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s %u reqs", @@ -1476,7 +1476,7 @@ DECLARE_EVENT_CLASS(svcrdma_accept_class, TP_fast_assign( __entry->status = status; - __assign_str(addr, rdma->sc_xprt.xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s status=%ld", @@ -1962,7 +1962,7 @@ TRACE_EVENT(svcrdma_send_err, TP_fast_assign( __entry->status = status; __entry->xid = __be32_to_cpu(rqst->rq_xid); - __assign_str(addr, rqst->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x status=%d", __get_str(addr), @@ -2025,7 +2025,7 @@ TRACE_EVENT(svcrdma_rq_post_err, TP_fast_assign( __entry->status = status; - __assign_str(addr, rdma->sc_xprt.xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s status=%d", @@ -2138,7 +2138,7 @@ TRACE_EVENT(svcrdma_qp_error, TP_fast_assign( __entry->event = event->event; - __assign_str(device, event->device->name); + __assign_str(device); snprintf(__entry->addr, sizeof(__entry->addr) - 1, "%pISpc", sap); ), diff --git a/include/trace/events/rpm.h b/include/trace/events/rpm.h index bd120e23ce12..2b0b4b6ef862 100644 --- a/include/trace/events/rpm.h +++ b/include/trace/events/rpm.h @@ -33,7 +33,7 @@ DECLARE_EVENT_CLASS(rpm_internal, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->flags = flags; __entry->usage_count = atomic_read( &dev->power.usage_count); @@ -92,7 +92,7 @@ TRACE_EVENT(rpm_return_int, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->ip = ip; __entry->ret = ret; ), @@ -135,7 +135,7 @@ TRACE_EVENT(rpm_status, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->status = status; ), diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h index 68973f650c26..6df2b4685b08 100644 --- a/include/trace/events/sched.h +++ b/include/trace/events/sched.h @@ -411,7 +411,7 @@ TRACE_EVENT(sched_process_exec, ), TP_fast_assign( - __assign_str(filename, bprm->filename); + __assign_str(filename); __entry->pid = p->pid; __entry->old_pid = old_pid; ), @@ -445,10 +445,10 @@ TRACE_EVENT(sched_prepare_exec, ), TP_fast_assign( - __assign_str(interp, bprm->interp); - __assign_str(filename, bprm->filename); + __assign_str(interp); + __assign_str(filename); __entry->pid = task->pid; - __assign_str(comm, task->comm); + __assign_str(comm); ), TP_printk("interp=%s filename=%s pid=%d comm=%s", diff --git a/include/trace/events/sof.h b/include/trace/events/sof.h index cd8e6844cca2..3061423c0667 100644 --- a/include/trace/events/sof.h +++ b/include/trace/events/sof.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(sof_widget_template, __field(int, use_count) ), TP_fast_assign( - __assign_str(name, swidget->widget->name); + __assign_str(name); __entry->use_count = swidget->use_count; ), TP_printk("name=%s use_count=%d", __get_str(name), __entry->use_count) @@ -49,7 +49,7 @@ TRACE_EVENT(sof_ipc3_period_elapsed_position, __field(u64, wallclock) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->host_posn = posn->host_posn; __entry->dai_posn = posn->dai_posn; __entry->wallclock = posn->wallclock; @@ -75,7 +75,7 @@ TRACE_EVENT(sof_pcm_pointer_position, __field(unsigned long, dai_posn) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->pcm_id = le32_to_cpu(spcm->pcm.pcm_id); __entry->stream = substream->stream; __entry->dma_posn = dma_posn; @@ -93,7 +93,7 @@ TRACE_EVENT(sof_stream_position_ipc_rx, __string(device_name, dev_name(dev)) ), TP_fast_assign( - __assign_str(device_name, dev_name(dev)); + __assign_str(device_name); ), TP_printk("device_name=%s", __get_str(device_name)) ); @@ -107,8 +107,8 @@ TRACE_EVENT(sof_ipc4_fw_config, __field(u32, value) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); - __assign_str(key, key); + __assign_str(device_name); + __assign_str(key); __entry->value = value; ), TP_printk("device_name=%s key=%s value=%d", diff --git a/include/trace/events/sof_intel.h b/include/trace/events/sof_intel.h index 4cac5fef99a2..9e579e57b15c 100644 --- a/include/trace/events/sof_intel.h +++ b/include/trace/events/sof_intel.h @@ -22,8 +22,8 @@ TRACE_EVENT(sof_intel_hda_irq, __string(source, source) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); - __assign_str(source, source); + __assign_str(device_name); + __assign_str(source); ), TP_printk("device_name=%s source=%s", __get_str(device_name), __get_str(source)) @@ -38,7 +38,7 @@ DECLARE_EVENT_CLASS(sof_intel_ipc_firmware_template, __field(u32, msg_ext) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->msg = msg; __entry->msg_ext = msg_ext; ), @@ -64,7 +64,7 @@ TRACE_EVENT(sof_intel_D0I3C_updated, __field(u8, reg) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->reg = reg; ), TP_printk("device_name=%s register=%#x", @@ -79,7 +79,7 @@ TRACE_EVENT(sof_intel_hda_irq_ipc_check, __field(u32, irq_status) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->irq_status = irq_status; ), TP_printk("device_name=%s irq_status=%#x", @@ -100,7 +100,7 @@ TRACE_EVENT(sof_intel_hda_dsp_pcm, __field(unsigned long, pos) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->hstream_index = hstream->index; __entry->substream = substream->stream; __entry->pos = pos; @@ -119,7 +119,7 @@ TRACE_EVENT(sof_intel_hda_dsp_stream_status, __field(u32, status) ), TP_fast_assign( - __assign_str(device_name, dev_name(dev)); + __assign_str(device_name); __entry->stream = s->index; __entry->status = status; ), @@ -135,7 +135,7 @@ TRACE_EVENT(sof_intel_hda_dsp_check_stream_irq, __field(u32, status) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->status = status; ), TP_printk("device_name=%s status=%#x", diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index ac05ed06a071..5e8495216689 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -188,10 +188,10 @@ TRACE_EVENT(rpc_clnt_new, __entry->client_id = clnt->cl_clid; __entry->xprtsec = args->xprtsec.policy; __entry->flags = args->flags; - __assign_str(program, clnt->cl_program->name); - __assign_str(server, xprt->servername); - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(program); + __assign_str(server); + __assign_str(addr); + __assign_str(port); ), TP_printk("client=" SUNRPC_TRACE_CLID_SPECIFIER " peer=[%s]:%s" @@ -220,8 +220,8 @@ TRACE_EVENT(rpc_clnt_new_err, TP_fast_assign( __entry->error = error; - __assign_str(program, program); - __assign_str(server, server); + __assign_str(program); + __assign_str(server); ), TP_printk("program=%s server=%s error=%d", @@ -325,8 +325,8 @@ TRACE_EVENT(rpc_request, __entry->client_id = task->tk_client->cl_clid; __entry->version = task->tk_client->cl_vers; __entry->async = RPC_IS_ASYNC(task); - __assign_str(progname, task->tk_client->cl_program->name); - __assign_str(procname, rpc_proc_name(task)); + __assign_str(progname); + __assign_str(procname); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " %sv%d %s (%ssync)", @@ -439,7 +439,7 @@ DECLARE_EVENT_CLASS(rpc_task_queued, __entry->runstate = task->tk_runstate; __entry->status = task->tk_status; __entry->flags = task->tk_flags; - __assign_str(q_name, rpc_qname(q)); + __assign_str(q_name); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -515,10 +515,10 @@ DECLARE_EVENT_CLASS(rpc_reply_event, __entry->task_id = task->tk_pid; __entry->client_id = task->tk_client->cl_clid; __entry->xid = be32_to_cpu(task->tk_rqstp->rq_xid); - __assign_str(progname, task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procname, rpc_proc_name(task)); - __assign_str(servername, task->tk_xprt->servername); + __assign_str(procname); + __assign_str(servername); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -647,8 +647,8 @@ TRACE_EVENT(rpc_stats_latency, __entry->task_id = task->tk_pid; __entry->xid = be32_to_cpu(task->tk_rqstp->rq_xid); __entry->version = task->tk_client->cl_vers; - __assign_str(progname, task->tk_client->cl_program->name); - __assign_str(procname, rpc_proc_name(task)); + __assign_str(progname); + __assign_str(procname); __entry->backlog = ktime_to_us(backlog); __entry->rtt = ktime_to_us(rtt); __entry->execute = ktime_to_us(execute); @@ -697,16 +697,15 @@ TRACE_EVENT(rpc_xdr_overflow, __entry->task_id = task->tk_pid; __entry->client_id = task->tk_client->cl_clid; - __assign_str(progname, - task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procedure, task->tk_msg.rpc_proc->p_name); + __assign_str(procedure); } else { __entry->task_id = -1; __entry->client_id = -1; - __assign_str(progname, "unknown"); + __assign_str(progname); __entry->version = 0; - __assign_str(procedure, "unknown"); + __assign_str(procedure); } __entry->requested = requested; __entry->end = xdr->end; @@ -763,10 +762,9 @@ TRACE_EVENT(rpc_xdr_alignment, __entry->task_id = task->tk_pid; __entry->client_id = task->tk_client->cl_clid; - __assign_str(progname, - task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procedure, task->tk_msg.rpc_proc->p_name); + __assign_str(procedure); __entry->offset = offset; __entry->copied = copied; @@ -1018,8 +1016,8 @@ DECLARE_EVENT_CLASS(rpc_xprt_lifetime_class, TP_fast_assign( __entry->state = xprt->state; - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s state=%s", @@ -1061,8 +1059,8 @@ DECLARE_EVENT_CLASS(rpc_xprt_event, TP_fast_assign( __entry->xid = be32_to_cpu(xid); __entry->status = status; - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s xid=0x%08x status=%d", __get_str(addr), @@ -1140,10 +1138,9 @@ TRACE_EVENT(xprt_retransmit, __entry->xid = be32_to_cpu(rqst->rq_xid); __entry->ntrans = rqst->rq_ntrans; __entry->timeout = task->tk_timeout; - __assign_str(progname, - task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procname, rpc_proc_name(task)); + __assign_str(procname); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -1167,8 +1164,8 @@ TRACE_EVENT(xprt_ping, TP_fast_assign( __entry->status = status; - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s status=%d", @@ -1315,8 +1312,8 @@ TRACE_EVENT(xs_data_ready, ), TP_fast_assign( - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s", __get_str(addr), __get_str(port)) @@ -1339,10 +1336,8 @@ TRACE_EVENT(xs_stream_read_data, TP_fast_assign( __entry->err = err; __entry->total = total; - __assign_str(addr, xprt ? - xprt->address_strings[RPC_DISPLAY_ADDR] : EVENT_NULL_STR); - __assign_str(port, xprt ? - xprt->address_strings[RPC_DISPLAY_PORT] : EVENT_NULL_STR); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s err=%zd total=%zu", __get_str(addr), @@ -1364,8 +1359,8 @@ TRACE_EVENT(xs_stream_read_request, ), TP_fast_assign( - __assign_str(addr, xs->xprt.address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xs->xprt.address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); __entry->xid = be32_to_cpu(xs->recv.xid); __entry->copied = xs->recv.copied; __entry->reclen = xs->recv.len; @@ -1403,7 +1398,7 @@ TRACE_EVENT(rpcb_getport, __entry->version = clnt->cl_vers; __entry->protocol = task->tk_xprt->prot; __entry->bind_version = bind_version; - __assign_str(servername, task->tk_xprt->servername); + __assign_str(servername); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -1493,8 +1488,8 @@ TRACE_EVENT(rpcb_register, TP_fast_assign( __entry->program = program; __entry->version = version; - __assign_str(addr, addr); - __assign_str(netid, netid); + __assign_str(addr); + __assign_str(netid); ), TP_printk("program=%u version=%u addr=%s netid=%s", @@ -1521,7 +1516,7 @@ TRACE_EVENT(rpcb_unregister, TP_fast_assign( __entry->program = program; __entry->version = version; - __assign_str(netid, netid); + __assign_str(netid); ), TP_printk("program=%u version=%u netid=%s", @@ -1551,8 +1546,8 @@ DECLARE_EVENT_CLASS(rpc_tls_class, TP_fast_assign( __entry->requested_policy = clnt->cl_xprtsec.policy; __entry->version = clnt->cl_vers; - __assign_str(servername, xprt->servername); - __assign_str(progname, clnt->cl_program->name) + __assign_str(servername); + __assign_str(progname); ), TP_printk("server=%s %sv%u requested_policy=%s", @@ -1794,10 +1789,9 @@ TRACE_EVENT(svc_process, __entry->xid = be32_to_cpu(rqst->rq_xid); __entry->vers = rqst->rq_vers; __entry->proc = rqst->rq_proc; - __assign_str(service, name); - __assign_str(procedure, svc_proc_name(rqst)); - __assign_str(addr, rqst->rq_xprt ? - rqst->rq_xprt->xpt_remotebuf : EVENT_NULL_STR); + __assign_str(service); + __assign_str(procedure); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x service=%s vers=%u proc=%s", @@ -1915,7 +1909,7 @@ TRACE_EVENT(svc_stats_latency, __entry->execute = ktime_to_us(ktime_sub(ktime_get(), rqst->rq_stime)); - __assign_str(procedure, svc_proc_name(rqst)); + __assign_str(procedure); ), TP_printk(SVC_RQST_ENDPOINT_FORMAT " proc=%s execute-us=%lu", @@ -1980,8 +1974,8 @@ TRACE_EVENT(svc_xprt_create_err, TP_fast_assign( __entry->error = PTR_ERR(xprt); - __assign_str(program, program); - __assign_str(protocol, protocol); + __assign_str(program); + __assign_str(protocol); __assign_sockaddr(addr, sap, salen); ), @@ -2120,8 +2114,8 @@ TRACE_EVENT(svc_xprt_accept, TP_fast_assign( SVC_XPRT_ENDPOINT_ASSIGNMENTS(xprt); - __assign_str(protocol, xprt->xpt_class->xcl_name); - __assign_str(service, service); + __assign_str(protocol); + __assign_str(service); ), TP_printk(SVC_XPRT_ENDPOINT_FORMAT " protocol=%s service=%s", @@ -2260,7 +2254,7 @@ TRACE_EVENT(svcsock_marker, TP_fast_assign( __entry->length = be32_to_cpu(marker) & RPC_FRAGMENT_SIZE_MASK; __entry->last = be32_to_cpu(marker) & RPC_LAST_STREAM_FRAGMENT; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s length=%u%s", __get_str(addr), @@ -2284,7 +2278,7 @@ DECLARE_EVENT_CLASS(svcsock_class, TP_fast_assign( __entry->result = result; __entry->flags = xprt->xpt_flags; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s result=%zd flags=%s", __get_str(addr), @@ -2330,7 +2324,7 @@ TRACE_EVENT(svcsock_tcp_recv_short, __entry->expected = expected; __entry->received = received; __entry->flags = xprt->xpt_flags; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s flags=%s expected=%u received=%u", @@ -2358,7 +2352,7 @@ TRACE_EVENT(svcsock_tcp_state, __entry->socket_state = socket->state; __entry->sock_state = socket->sk->sk_state; __entry->flags = xprt->xpt_flags; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s state=%s sk_state=%s flags=%s", __get_str(addr), @@ -2385,7 +2379,7 @@ DECLARE_EVENT_CLASS(svcsock_accept_class, TP_fast_assign( __entry->status = status; - __assign_str(service, service); + __assign_str(service); __entry->netns_ino = xprt->xpt_net->ns.inum; ), @@ -2421,7 +2415,7 @@ DECLARE_EVENT_CLASS(cache_event, TP_fast_assign( __entry->h = h; - __assign_str(name, cd->name); + __assign_str(name); ), TP_printk("cache=%s entry=%p", __get_str(name), __entry->h) @@ -2466,7 +2460,7 @@ DECLARE_EVENT_CLASS(register_class, __entry->protocol = protocol; __entry->port = port; __entry->error = error; - __assign_str(program, program); + __assign_str(program); ), TP_printk("program=%sv%u proto=%s port=%u family=%s error=%d", @@ -2511,7 +2505,7 @@ TRACE_EVENT(svc_unregister, TP_fast_assign( __entry->version = version; __entry->error = error; - __assign_str(program, program); + __assign_str(program); ), TP_printk("program=%sv%u error=%d", diff --git a/include/trace/events/swiotlb.h b/include/trace/events/swiotlb.h index da05c9ebd224..3b6ddb136e4e 100644 --- a/include/trace/events/swiotlb.h +++ b/include/trace/events/swiotlb.h @@ -20,7 +20,7 @@ TRACE_EVENT(swiotlb_bounced, ), TP_fast_assign( - __assign_str(dev_name, dev_name(dev)); + __assign_str(dev_name); __entry->dma_mask = (dev->dma_mask ? *dev->dma_mask : 0); __entry->dev_addr = dev_addr; __entry->size = size; diff --git a/include/trace/events/target.h b/include/trace/events/target.h index 67fad2677ed5..a13cbf2b3405 100644 --- a/include/trace/events/target.h +++ b/include/trace/events/target.h @@ -154,7 +154,7 @@ TRACE_EVENT(target_sequencer_start, __entry->task_attribute = cmd->sam_task_attr; __entry->control = scsi_command_control(cmd->t_task_cdb); memcpy(__entry->cdb, cmd->t_task_cdb, TCM_MAX_COMMAND_SIZE); - __assign_str(initiator, cmd->se_sess->se_node_acl->initiatorname); + __assign_str(initiator); ), TP_printk("%s -> LUN %03u tag %#llx %s data_length %6u CDB %s (TA:%s C:%02x)", @@ -198,7 +198,7 @@ TRACE_EVENT(target_cmd_complete, min(18, ((u8 *) cmd->sense_buffer)[SPC_ADD_SENSE_LEN_OFFSET] + 8) : 0; memcpy(__entry->cdb, cmd->t_task_cdb, TCM_MAX_COMMAND_SIZE); memcpy(__entry->sense_data, cmd->sense_buffer, __entry->sense_length); - __assign_str(initiator, cmd->se_sess->se_node_acl->initiatorname); + __assign_str(initiator); ), TP_printk("%s <- LUN %03u tag %#llx status %s (sense len %d%s%s) %s data_length %6u CDB %s (TA:%s C:%02x)", diff --git a/include/trace/events/tegra_apb_dma.h b/include/trace/events/tegra_apb_dma.h index 971cd02d2daf..6d9f5075baa3 100644 --- a/include/trace/events/tegra_apb_dma.h +++ b/include/trace/events/tegra_apb_dma.h @@ -16,7 +16,7 @@ TRACE_EVENT(tegra_dma_tx_status, __field(__u32, residue) ), TP_fast_assign( - __assign_str(chan, dev_name(&dc->dev->device)); + __assign_str(chan); __entry->cookie = cookie; __entry->residue = state ? state->residue : (u32)-1; ), @@ -33,7 +33,7 @@ TRACE_EVENT(tegra_dma_complete_cb, __field(void *, ptr) ), TP_fast_assign( - __assign_str(chan, dev_name(&dc->dev->device)); + __assign_str(chan); __entry->count = count; __entry->ptr = ptr; ), @@ -49,7 +49,7 @@ TRACE_EVENT(tegra_dma_isr, __field(int, irq) ), TP_fast_assign( - __assign_str(chan, dev_name(&dc->dev->device)); + __assign_str(chan); __entry->irq = irq; ), TP_printk("%s: irq %d\n", __get_str(chan), __entry->irq) diff --git a/include/trace/events/ufs.h b/include/trace/events/ufs.h index b930669bd1f0..c4e209fbdfbb 100644 --- a/include/trace/events/ufs.h +++ b/include/trace/events/ufs.h @@ -92,7 +92,7 @@ TRACE_EVENT(ufshcd_clk_gating, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->state = state; ), @@ -117,9 +117,9 @@ TRACE_EVENT(ufshcd_clk_scaling, ), TP_fast_assign( - __assign_str(dev_name, dev_name); - __assign_str(state, state); - __assign_str(clk, clk); + __assign_str(dev_name); + __assign_str(state); + __assign_str(clk); __entry->prev_state = prev_state; __entry->curr_state = curr_state; ), @@ -141,8 +141,8 @@ TRACE_EVENT(ufshcd_auto_bkops_state, ), TP_fast_assign( - __assign_str(dev_name, dev_name); - __assign_str(state, state); + __assign_str(dev_name); + __assign_str(state); ), TP_printk("%s: auto bkops - %s", @@ -163,8 +163,8 @@ DECLARE_EVENT_CLASS(ufshcd_profiling_template, ), TP_fast_assign( - __assign_str(dev_name, dev_name); - __assign_str(profile_info, profile_info); + __assign_str(dev_name); + __assign_str(profile_info); __entry->time_us = time_us; __entry->err = err; ), @@ -206,7 +206,7 @@ DECLARE_EVENT_CLASS(ufshcd_template, TP_fast_assign( __entry->usecs = usecs; __entry->err = err; - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->dev_state = dev_state; __entry->link_state = link_state; ), @@ -326,7 +326,7 @@ TRACE_EVENT(ufshcd_uic_command, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->str_t = str_t; __entry->cmd = cmd; __entry->arg1 = arg1; @@ -356,7 +356,7 @@ TRACE_EVENT(ufshcd_upiu, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->str_t = str_t; memcpy(__entry->hdr, hdr, sizeof(__entry->hdr)); memcpy(__entry->tsf, tsf, sizeof(__entry->tsf)); @@ -384,7 +384,7 @@ TRACE_EVENT(ufshcd_exception_event, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->status = status; ), diff --git a/include/trace/events/workqueue.h b/include/trace/events/workqueue.h index 6ef5b7254070..b0de2bc9ed52 100644 --- a/include/trace/events/workqueue.h +++ b/include/trace/events/workqueue.h @@ -38,7 +38,7 @@ TRACE_EVENT(workqueue_queue_work, TP_fast_assign( __entry->work = work; __entry->function = work->func; - __assign_str(workqueue, pwq->wq->name); + __assign_str(workqueue); __entry->req_cpu = req_cpu; __entry->cpu = pwq->pool->cpu; ), diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h index 9adc2bdf2f94..a7e5452b5d21 100644 --- a/include/trace/events/xdp.h +++ b/include/trace/events/xdp.h @@ -416,7 +416,7 @@ TRACE_EVENT(bpf_xdp_link_attach_failed, ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("errmsg=%s", __get_str(msg)) diff --git a/include/trace/stages/stage6_event_callback.h b/include/trace/stages/stage6_event_callback.h index 3690e677263f..1691676fd858 100644 --- a/include/trace/stages/stage6_event_callback.h +++ b/include/trace/stages/stage6_event_callback.h @@ -31,12 +31,10 @@ #define __vstring(item, fmt, ap) __dynamic_array(char, item, -1) #undef __assign_str -#define __assign_str(dst, src) \ +#define __assign_str(dst) \ do { \ char *__str__ = __get_str(dst); \ int __len__ = __get_dynamic_array_len(dst) - 1; \ - WARN_ON_ONCE(!(void *)(src) != !(void *)__data_offsets.dst##_ptr_); \ - WARN_ON_ONCE((src) && strcmp((src), __data_offsets.dst##_ptr_)); \ memcpy(__str__, __data_offsets.dst##_ptr_ ? : \ EVENT_NULL_STR, __len__); \ __str__[__len__] = '\0'; \ diff --git a/kernel/trace/bpf_trace.h b/kernel/trace/bpf_trace.h index 9acbc11ac7bb..c4075b56becc 100644 --- a/kernel/trace/bpf_trace.h +++ b/kernel/trace/bpf_trace.h @@ -19,7 +19,7 @@ TRACE_EVENT(bpf_trace_printk, ), TP_fast_assign( - __assign_str(bpf_string, bpf_string); + __assign_str(bpf_string); ), TP_printk("%s", __get_str(bpf_string)) diff --git a/net/batman-adv/trace.h b/net/batman-adv/trace.h index 5dd52bc5cabb..6b816cf1a953 100644 --- a/net/batman-adv/trace.h +++ b/net/batman-adv/trace.h @@ -40,8 +40,8 @@ TRACE_EVENT(batadv_dbg, ), TP_fast_assign( - __assign_str(device, bat_priv->soft_iface->name); - __assign_str(driver, KBUILD_MODNAME); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/net/dsa/trace.h b/net/dsa/trace.h index 567f29a39707..83f3e5f78491 100644 --- a/net/dsa/trace.h +++ b/net/dsa/trace.h @@ -39,8 +39,8 @@ DECLARE_EVENT_CLASS(dsa_port_addr_op_hw, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; ether_addr_copy(__entry->addr, addr); __entry->vid = vid; @@ -98,8 +98,8 @@ DECLARE_EVENT_CLASS(dsa_port_addr_op_refcount, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; ether_addr_copy(__entry->addr, addr); __entry->vid = vid; @@ -157,8 +157,8 @@ DECLARE_EVENT_CLASS(dsa_port_addr_del_not_found, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; ether_addr_copy(__entry->addr, addr); __entry->vid = vid; @@ -199,7 +199,7 @@ TRACE_EVENT(dsa_lag_fdb_add_hw, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -227,7 +227,7 @@ TRACE_EVENT(dsa_lag_fdb_add_bump, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -255,7 +255,7 @@ TRACE_EVENT(dsa_lag_fdb_del_hw, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -283,7 +283,7 @@ TRACE_EVENT(dsa_lag_fdb_del_drop, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -310,7 +310,7 @@ TRACE_EVENT(dsa_lag_fdb_del_not_found, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -338,8 +338,8 @@ DECLARE_EVENT_CLASS(dsa_vlan_op_hw, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; __entry->vid = vlan->vid; __entry->flags = vlan->flags; @@ -383,8 +383,8 @@ DECLARE_EVENT_CLASS(dsa_vlan_op_refcount, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; __entry->vid = vlan->vid; __entry->flags = vlan->flags; @@ -426,8 +426,8 @@ TRACE_EVENT(dsa_vlan_del_not_found, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; __entry->vid = vlan->vid; ), diff --git a/net/ieee802154/trace.h b/net/ieee802154/trace.h index 62aa6465253a..591ce0a16fc0 100644 --- a/net/ieee802154/trace.h +++ b/net/ieee802154/trace.h @@ -75,7 +75,7 @@ TRACE_EVENT(802154_rdev_add_virtual_intf, ), TP_fast_assign( WPAN_PHY_ASSIGN; - __assign_str(vir_intf_name, name ? name : ""); + __assign_str(vir_intf_name); __entry->type = type; __entry->extended_addr = extended_addr; ), diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h index 8e758b5074bd..b26aacfbc622 100644 --- a/net/mac80211/trace.h +++ b/net/mac80211/trace.h @@ -33,7 +33,7 @@ __string(vif_name, sdata->name) #define VIF_ASSIGN __entry->vif_type = sdata->vif.type; __entry->sdata = sdata; \ __entry->p2p = sdata->vif.p2p; \ - __assign_str(vif_name, sdata->name) + __assign_str(vif_name) #define VIF_PR_FMT " vif:%s(%d%s)" #define VIF_PR_ARG __get_str(vif_name), __entry->vif_type, __entry->p2p ? "/p2p" : "" diff --git a/net/openvswitch/openvswitch_trace.h b/net/openvswitch/openvswitch_trace.h index 3eb35d9eb700..74d75aaebef4 100644 --- a/net/openvswitch/openvswitch_trace.h +++ b/net/openvswitch/openvswitch_trace.h @@ -43,8 +43,8 @@ TRACE_EVENT(ovs_do_execute_action, TP_fast_assign( __entry->dpaddr = dp; - __assign_str(dp_name, ovs_dp_name(dp)); - __assign_str(dev_name, skb->dev->name); + __assign_str(dp_name); + __assign_str(dev_name); __entry->skbaddr = skb; __entry->len = skb->len; __entry->data_len = skb->data_len; @@ -113,8 +113,8 @@ TRACE_EVENT(ovs_dp_upcall, TP_fast_assign( __entry->dpaddr = dp; - __assign_str(dp_name, ovs_dp_name(dp)); - __assign_str(dev_name, skb->dev->name); + __assign_str(dp_name); + __assign_str(dev_name); __entry->skbaddr = skb; __entry->len = skb->len; __entry->data_len = skb->data_len; diff --git a/net/smc/smc_tracepoint.h b/net/smc/smc_tracepoint.h index 9fc5e586d24a..a9a6e3c1113a 100644 --- a/net/smc/smc_tracepoint.h +++ b/net/smc/smc_tracepoint.h @@ -60,7 +60,7 @@ DECLARE_EVENT_CLASS(smc_msg_event, __entry->smc = smc; __entry->net_cookie = sock_net(sk)->net_cookie; __entry->len = len; - __assign_str(name, smc->conn.lnk->ibname); + __assign_str(name); ), TP_printk("smc=%p net=%llu len=%zu dev=%s", @@ -104,7 +104,7 @@ TRACE_EVENT(smcr_link_down, __entry->lgr = lgr; __entry->net_cookie = lgr->net->net_cookie; __entry->state = lnk->state; - __assign_str(name, lnk->ibname); + __assign_str(name); __entry->location = location; ), diff --git a/net/tipc/trace.h b/net/tipc/trace.h index 04af83f0500c..865142ed0ab4 100644 --- a/net/tipc/trace.h +++ b/net/tipc/trace.h @@ -145,7 +145,7 @@ DECLARE_EVENT_CLASS(tipc_skb_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); tipc_skb_dump(skb, more, __get_str(buf)); ), @@ -172,7 +172,7 @@ DECLARE_EVENT_CLASS(tipc_list_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); tipc_list_dump(list, more, __get_str(buf)); ), @@ -200,7 +200,7 @@ DECLARE_EVENT_CLASS(tipc_sk_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); __entry->portid = tipc_sock_get_portid(sk); tipc_sk_dump(sk, dqueues, __get_str(buf)); if (skb) @@ -254,7 +254,7 @@ DECLARE_EVENT_CLASS(tipc_link_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); memcpy(__entry->name, tipc_link_name(l), TIPC_MAX_LINK_NAME); tipc_link_dump(l, dqueues, __get_str(buf)); ), @@ -337,7 +337,7 @@ DECLARE_EVENT_CLASS(tipc_node_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); __entry->addr = tipc_node_get_addr(n); tipc_node_dump(n, more, __get_str(buf)); ), @@ -374,7 +374,7 @@ DECLARE_EVENT_CLASS(tipc_fsm_class, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->os = os; __entry->ns = ns; __entry->evt = evt; @@ -409,8 +409,8 @@ TRACE_EVENT(tipc_l2_device_event, ), TP_fast_assign( - __assign_str(dev_name, dev->name); - __assign_str(b_name, b->name); + __assign_str(dev_name); + __assign_str(b_name); __entry->evt = evt; __entry->b_up = test_bit(0, &b->up); __entry->carrier = netif_carrier_ok(dev); diff --git a/net/wireless/trace.h b/net/wireless/trace.h index 9bf987519811..87986170d1b1 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -372,7 +372,7 @@ TRACE_EVENT(rdev_add_virtual_intf, ), TP_fast_assign( WIPHY_ASSIGN; - __assign_str(vir_intf_name, name ? name : ""); + __assign_str(vir_intf_name); __entry->type = type; ), TP_printk(WIPHY_PR_FMT ", virtual intf name: %s, type: %d", diff --git a/samples/trace_events/trace-events-sample.h b/samples/trace_events/trace-events-sample.h index 500981eca74d..55f9a3da92d5 100644 --- a/samples/trace_events/trace-events-sample.h +++ b/samples/trace_events/trace-events-sample.h @@ -136,10 +136,11 @@ * * To assign a string, use the helper macro __assign_str(). * - * __assign_str(foo, bar); + * __assign_str(foo); * - * In most cases, the __assign_str() macro will take the same - * parameters as the __string() macro had to declare the string. + * The __string() macro saves off the string that is passed into + * the second parameter, and the __assign_str() will store than + * saved string into the "foo" field. * * __vstring: This is similar to __string() but instead of taking a * dynamic length, it takes a variable list va_list 'va' variable. @@ -177,7 +178,7 @@ * The length is saved via the __string_len() and is retrieved in * __assign_str(). * - * __assign_str(foo, bar); + * __assign_str(foo); * * Then len + 1 is allocated to the ring buffer, and a nul terminating * byte is added. This is similar to: @@ -311,8 +312,8 @@ TRACE_EVENT(foo_bar, __entry->bar = bar; memcpy(__get_dynamic_array(list), lst, __length_of(lst) * sizeof(int)); - __assign_str(str, string); - __assign_str(lstr, foo); + __assign_str(str); + __assign_str(lstr); __assign_vstr(vstr, fmt, va); __assign_bitmask(cpus, cpumask_bits(mask), num_possible_cpus()); __assign_cpumask(cpum, cpumask_bits(mask)); @@ -418,7 +419,7 @@ TRACE_EVENT_CONDITION(foo_bar_with_cond, ), TP_fast_assign( - __assign_str(foo, foo); + __assign_str(foo); __entry->bar = bar; ), @@ -459,7 +460,7 @@ TRACE_EVENT_FN(foo_bar_with_fn, ), TP_fast_assign( - __assign_str(foo, foo); + __assign_str(foo); __entry->bar = bar; ), @@ -506,7 +507,7 @@ DECLARE_EVENT_CLASS(foo_template, ), TP_fast_assign( - __assign_str(foo, foo); + __assign_str(foo); __entry->bar = bar; ), diff --git a/sound/core/pcm_trace.h b/sound/core/pcm_trace.h index 350b40b906ca..adb9b1f3bbfa 100644 --- a/sound/core/pcm_trace.h +++ b/sound/core/pcm_trace.h @@ -95,7 +95,7 @@ TRACE_EVENT(hw_ptr_error, __entry->device = (substream)->pcm->device; __entry->number = (substream)->number; __entry->stream = (substream)->stream; - __assign_str(reason, why); + __assign_str(reason); ), TP_printk("pcmC%dD%d%s/sub%d: ERROR: %s", __entry->card, __entry->device, diff --git a/sound/hda/trace.h b/sound/hda/trace.h index 2cc493434a8f..280c42f3eb75 100644 --- a/sound/hda/trace.h +++ b/sound/hda/trace.h @@ -24,7 +24,7 @@ TRACE_EVENT(hda_send_cmd, __field(u32, cmd) ), TP_fast_assign( - __assign_str(name, dev_name((bus)->dev)); + __assign_str(name); __entry->cmd = cmd; ), TP_printk("[%s:%d] val=0x%08x", __get_str(name), __entry->cmd >> 28, __entry->cmd) @@ -39,7 +39,7 @@ TRACE_EVENT(hda_get_response, __field(u32, res) ), TP_fast_assign( - __assign_str(name, dev_name((bus)->dev)); + __assign_str(name); __entry->addr = addr; __entry->res = res; ), @@ -55,7 +55,7 @@ TRACE_EVENT(hda_unsol_event, __field(u32, res_ex) ), TP_fast_assign( - __assign_str(name, dev_name((bus)->dev)); + __assign_str(name); __entry->res = res; __entry->res_ex = res_ex; ), diff --git a/sound/soc/intel/avs/trace.h b/sound/soc/intel/avs/trace.h index 855b06bb14b0..c9eaa5a60ed3 100644 --- a/sound/soc/intel/avs/trace.h +++ b/sound/soc/intel/avs/trace.h @@ -24,7 +24,7 @@ TRACE_EVENT(avs_dsp_core_op, TP_fast_assign( __entry->reg = reg; __entry->mask = mask; - __assign_str(op, op); + __assign_str(op); __entry->flag = flag; ), @@ -135,7 +135,7 @@ TRACE_EVENT(avs_d0ix, ), TP_fast_assign( - __assign_str(op, op); + __assign_str(op); __entry->proceed = proceed; __entry->header = header; ), -- cgit v1.2.3-59-g8ed1b From cd3b31f9d4174cccafd8da615d73f40c1ce48939 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 22 May 2024 20:17:05 -0400 Subject: bcachefs: Ensure we're RW before journalling Reported-by: syzbot+c60cd352aedb109528bf@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/recovery.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 63f990f5c63a..cf513fc79ce4 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -811,9 +811,11 @@ use_clean: clear_bit(BCH_FS_fsck_running, &c->flags); /* fsync if we fixed errors */ - if (test_bit(BCH_FS_errors_fixed, &c->flags)) { + if (test_bit(BCH_FS_errors_fixed, &c->flags) && + bch2_write_ref_tryget(c, BCH_WRITE_REF_fsync)) { bch2_journal_flush_all_pins(&c->journal); bch2_journal_meta(&c->journal); + bch2_write_ref_put(c, BCH_WRITE_REF_fsync); } /* If we fixed errors, verify that fs is actually clean now: */ -- cgit v1.2.3-59-g8ed1b From d93ff5fa40b9db5f505d508336bc171f54db862e Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 22 May 2024 20:36:42 -0400 Subject: bcachefs: Fix race path in bch2_inode_insert() __destroy_new_inode() is appropriate when we have _just_allocated the inode, but not when it's been fully initialized and on i_sb_list. Reported-by: syzbot+a0ddc9873c280a4cb18f@syzkaller.appspotmail.com Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 9b41c6e9deef..96040a95cf46 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -188,8 +188,7 @@ static struct bch_inode_info *bch2_inode_insert(struct bch_fs *c, struct bch_ino BUG_ON(!old); if (unlikely(old != inode)) { - __destroy_inode(&inode->v); - kmem_cache_free(bch2_inode_cache, inode); + discard_new_inode(&inode->v); inode = old; } else { mutex_lock(&c->vfs_inodes_lock); -- cgit v1.2.3-59-g8ed1b From 1d17de9534cb5a4c4c380d174cc2f9281b34f89e Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Mon, 25 Sep 2023 16:25:20 +0800 Subject: ceph: save cap_auths in MDS client when session is opened Save the cap_auths, which have been parsed by the MDS, in the opened session. [ idryomov: use s64 and u32 instead of int64_t and uint32_t, switch to bool for root_squash, readable and writeable ] Link: https://tracker.ceph.com/issues/61333 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Signed-off-by: Ilya Dryomov --- fs/ceph/mds_client.c | 106 ++++++++++++++++++++++++++++++++++++++++++++++++++- fs/ceph/mds_client.h | 21 ++++++++++ 2 files changed, 126 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 360b686c3c67..6a1e68549475 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -4112,10 +4112,13 @@ static void handle_session(struct ceph_mds_session *session, void *p = msg->front.iov_base; void *end = p + msg->front.iov_len; struct ceph_mds_session_head *h; - u32 op; + struct ceph_mds_cap_auth *cap_auths = NULL; + u32 op, cap_auths_num = 0; u64 seq, features = 0; int wake = 0; bool blocklisted = false; + u32 i; + /* decode */ ceph_decode_need(&p, end, sizeof(*h), bad); @@ -4160,7 +4163,101 @@ static void handle_session(struct ceph_mds_session *session, } } + if (msg_version >= 6) { + ceph_decode_32_safe(&p, end, cap_auths_num, bad); + doutc(cl, "cap_auths_num %d\n", cap_auths_num); + + if (cap_auths_num && op != CEPH_SESSION_OPEN) { + WARN_ON_ONCE(op != CEPH_SESSION_OPEN); + goto skip_cap_auths; + } + + cap_auths = kcalloc(cap_auths_num, + sizeof(struct ceph_mds_cap_auth), + GFP_KERNEL); + if (!cap_auths) { + pr_err_client(cl, "No memory for cap_auths\n"); + return; + } + + for (i = 0; i < cap_auths_num; i++) { + u32 _len, j; + + /* struct_v, struct_compat, and struct_len in MDSCapAuth */ + ceph_decode_skip_n(&p, end, 2 + sizeof(u32), bad); + + /* struct_v, struct_compat, and struct_len in MDSCapMatch */ + ceph_decode_skip_n(&p, end, 2 + sizeof(u32), bad); + ceph_decode_64_safe(&p, end, cap_auths[i].match.uid, bad); + ceph_decode_32_safe(&p, end, _len, bad); + if (_len) { + cap_auths[i].match.gids = kcalloc(_len, sizeof(u32), + GFP_KERNEL); + if (!cap_auths[i].match.gids) { + pr_err_client(cl, "No memory for gids\n"); + goto fail; + } + + cap_auths[i].match.num_gids = _len; + for (j = 0; j < _len; j++) + ceph_decode_32_safe(&p, end, + cap_auths[i].match.gids[j], + bad); + } + + ceph_decode_32_safe(&p, end, _len, bad); + if (_len) { + cap_auths[i].match.path = kcalloc(_len + 1, sizeof(char), + GFP_KERNEL); + if (!cap_auths[i].match.path) { + pr_err_client(cl, "No memory for path\n"); + goto fail; + } + ceph_decode_copy(&p, cap_auths[i].match.path, _len); + + /* Remove the tailing '/' */ + while (_len && cap_auths[i].match.path[_len - 1] == '/') { + cap_auths[i].match.path[_len - 1] = '\0'; + _len -= 1; + } + } + + ceph_decode_32_safe(&p, end, _len, bad); + if (_len) { + cap_auths[i].match.fs_name = kcalloc(_len + 1, sizeof(char), + GFP_KERNEL); + if (!cap_auths[i].match.fs_name) { + pr_err_client(cl, "No memory for fs_name\n"); + goto fail; + } + ceph_decode_copy(&p, cap_auths[i].match.fs_name, _len); + } + + ceph_decode_8_safe(&p, end, cap_auths[i].match.root_squash, bad); + ceph_decode_8_safe(&p, end, cap_auths[i].readable, bad); + ceph_decode_8_safe(&p, end, cap_auths[i].writeable, bad); + doutc(cl, "uid %lld, num_gids %u, path %s, fs_name %s, root_squash %d, readable %d, writeable %d\n", + cap_auths[i].match.uid, cap_auths[i].match.num_gids, + cap_auths[i].match.path, cap_auths[i].match.fs_name, + cap_auths[i].match.root_squash, + cap_auths[i].readable, cap_auths[i].writeable); + } + } + +skip_cap_auths: mutex_lock(&mdsc->mutex); + if (op == CEPH_SESSION_OPEN) { + if (mdsc->s_cap_auths) { + for (i = 0; i < mdsc->s_cap_auths_num; i++) { + kfree(mdsc->s_cap_auths[i].match.gids); + kfree(mdsc->s_cap_auths[i].match.path); + kfree(mdsc->s_cap_auths[i].match.fs_name); + } + kfree(mdsc->s_cap_auths); + } + mdsc->s_cap_auths_num = cap_auths_num; + mdsc->s_cap_auths = cap_auths; + } if (op == CEPH_SESSION_CLOSE) { ceph_get_mds_session(session); __unregister_session(mdsc, session); @@ -4290,6 +4387,13 @@ bad: pr_err_client(cl, "corrupt message mds%d len %d\n", mds, (int)msg->front.iov_len); ceph_msg_dump(msg); +fail: + for (i = 0; i < cap_auths_num; i++) { + kfree(cap_auths[i].match.gids); + kfree(cap_auths[i].match.path); + kfree(cap_auths[i].match.fs_name); + } + kfree(cap_auths); return; } diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index b88e80415224..cc106f1aa7d2 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -71,6 +71,24 @@ enum ceph_feature_type { struct ceph_fs_client; struct ceph_cap; +#define MDS_AUTH_UID_ANY -1 + +struct ceph_mds_cap_match { + s64 uid; /* default to MDS_AUTH_UID_ANY */ + u32 num_gids; + u32 *gids; /* use these GIDs */ + char *path; /* require path to be child of this + (may be "" or "/" for any) */ + char *fs_name; + bool root_squash; /* default to false */ +}; + +struct ceph_mds_cap_auth { + struct ceph_mds_cap_match match; + bool readable; + bool writeable; +}; + /* * parsed info about a single inode. pointers are into the encoded * on-wire structures within the mds reply message payload. @@ -513,6 +531,9 @@ struct ceph_mds_client { struct rw_semaphore pool_perm_rwsem; struct rb_root pool_perm_tree; + u32 s_cap_auths_num; + struct ceph_mds_cap_auth *s_cap_auths; + char nodename[__NEW_UTS_LEN + 1]; }; -- cgit v1.2.3-59-g8ed1b From 596afb0b8933ba6ed7227adcc538db26feb25c74 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Tue, 7 Nov 2023 14:55:46 +0800 Subject: ceph: add ceph_mds_check_access() helper This will help check the mds auth access in client side. Always insert the server path in front of the target path when matching the paths. [ idryomov: use u32 instead of uint32_t ] Link: https://tracker.ceph.com/issues/61333 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Signed-off-by: Ilya Dryomov --- fs/ceph/mds_client.c | 164 +++++++++++++++++++++++++++++++++++++++++++++++++++ fs/ceph/mds_client.h | 3 + 2 files changed, 167 insertions(+) (limited to 'fs') diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 6a1e68549475..c2157f6e0c69 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -5603,6 +5603,170 @@ void send_flush_mdlog(struct ceph_mds_session *s) mutex_unlock(&s->s_mutex); } +static int ceph_mds_auth_match(struct ceph_mds_client *mdsc, + struct ceph_mds_cap_auth *auth, + char *tpath) +{ + const struct cred *cred = get_current_cred(); + u32 caller_uid = from_kuid(&init_user_ns, cred->fsuid); + u32 caller_gid = from_kgid(&init_user_ns, cred->fsgid); + struct ceph_client *cl = mdsc->fsc->client; + const char *spath = mdsc->fsc->mount_options->server_path; + bool gid_matched = false; + u32 gid, tlen, len; + int i, j; + + doutc(cl, "match.uid %lld\n", auth->match.uid); + if (auth->match.uid != MDS_AUTH_UID_ANY) { + if (auth->match.uid != caller_uid) + return 0; + if (auth->match.num_gids) { + for (i = 0; i < auth->match.num_gids; i++) { + if (caller_gid == auth->match.gids[i]) + gid_matched = true; + } + if (!gid_matched && cred->group_info->ngroups) { + for (i = 0; i < cred->group_info->ngroups; i++) { + gid = from_kgid(&init_user_ns, + cred->group_info->gid[i]); + for (j = 0; j < auth->match.num_gids; j++) { + if (gid == auth->match.gids[j]) { + gid_matched = true; + break; + } + } + if (gid_matched) + break; + } + } + if (!gid_matched) + return 0; + } + } + + /* path match */ + if (auth->match.path) { + if (!tpath) + return 0; + + tlen = strlen(tpath); + len = strlen(auth->match.path); + if (len) { + char *_tpath = tpath; + bool free_tpath = false; + int m, n; + + doutc(cl, "server path %s, tpath %s, match.path %s\n", + spath, tpath, auth->match.path); + if (spath && (m = strlen(spath)) != 1) { + /* mount path + '/' + tpath + an extra space */ + n = m + 1 + tlen + 1; + _tpath = kmalloc(n, GFP_NOFS); + if (!_tpath) + return -ENOMEM; + /* remove the leading '/' */ + snprintf(_tpath, n, "%s/%s", spath + 1, tpath); + free_tpath = true; + tlen = strlen(_tpath); + } + + /* + * Please note the tailing '/' for match.path has already + * been removed when parsing. + * + * Remove the tailing '/' for the target path. + */ + while (tlen && _tpath[tlen - 1] == '/') { + _tpath[tlen - 1] = '\0'; + tlen -= 1; + } + doutc(cl, "_tpath %s\n", _tpath); + + /* + * In case first == _tpath && tlen == len: + * match.path=/foo --> /foo _path=/foo --> match + * match.path=/foo/ --> /foo _path=/foo --> match + * + * In case first == _tmatch.path && tlen > len: + * match.path=/foo/ --> /foo _path=/foo/ --> match + * match.path=/foo --> /foo _path=/foo/ --> match + * match.path=/foo/ --> /foo _path=/foo/d --> match + * match.path=/foo --> /foo _path=/food --> mismatch + * + * All the other cases --> mismatch + */ + char *first = strstr(_tpath, auth->match.path); + if (first != _tpath) { + if (free_tpath) + kfree(_tpath); + return 0; + } + + if (tlen > len && _tpath[len] != '/') { + if (free_tpath) + kfree(_tpath); + return 0; + } + } + } + + doutc(cl, "matched\n"); + return 1; +} + +int ceph_mds_check_access(struct ceph_mds_client *mdsc, char *tpath, int mask) +{ + const struct cred *cred = get_current_cred(); + u32 caller_uid = from_kuid(&init_user_ns, cred->fsuid); + u32 caller_gid = from_kgid(&init_user_ns, cred->fsgid); + struct ceph_mds_cap_auth *rw_perms_s = NULL; + struct ceph_client *cl = mdsc->fsc->client; + bool root_squash_perms = true; + int i, err; + + doutc(cl, "tpath '%s', mask %d, caller_uid %d, caller_gid %d\n", + tpath, mask, caller_uid, caller_gid); + + for (i = 0; i < mdsc->s_cap_auths_num; i++) { + struct ceph_mds_cap_auth *s = &mdsc->s_cap_auths[i]; + + err = ceph_mds_auth_match(mdsc, s, tpath); + if (err < 0) { + return err; + } else if (err > 0) { + /* always follow the last auth caps' permision */ + root_squash_perms = true; + rw_perms_s = NULL; + if ((mask & MAY_WRITE) && s->writeable && + s->match.root_squash && (!caller_uid || !caller_gid)) + root_squash_perms = false; + + if (((mask & MAY_WRITE) && !s->writeable) || + ((mask & MAY_READ) && !s->readable)) + rw_perms_s = s; + } + } + + doutc(cl, "root_squash_perms %d, rw_perms_s %p\n", root_squash_perms, + rw_perms_s); + if (root_squash_perms && rw_perms_s == NULL) { + doutc(cl, "access allowed\n"); + return 0; + } + + if (!root_squash_perms) { + doutc(cl, "root_squash is enabled and user(%d %d) isn't allowed to write", + caller_uid, caller_gid); + } + if (rw_perms_s) { + doutc(cl, "mds auth caps readable/writeable %d/%d while request r/w %d/%d", + rw_perms_s->readable, rw_perms_s->writeable, + !!(mask & MAY_READ), !!(mask & MAY_WRITE)); + } + doutc(cl, "access denied\n"); + return -EACCES; +} + /* * called before mount is ro, and before dentries are torn down. * (hmm, does this still race with new lookups?) diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index cc106f1aa7d2..dba514fb7fab 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -602,6 +602,9 @@ extern void ceph_queue_cap_unlink_work(struct ceph_mds_client *mdsc); extern int ceph_iterate_session_caps(struct ceph_mds_session *session, int (*cb)(struct inode *, int mds, void *), void *arg); +extern int ceph_mds_check_access(struct ceph_mds_client *mdsc, char *tpath, + int mask); + extern void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc); static inline void ceph_mdsc_free_path(char *path, int len) -- cgit v1.2.3-59-g8ed1b From ded67830403710a60a4bb00136c0d21ca0cd7dc5 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Wed, 8 Nov 2023 10:54:17 +0800 Subject: ceph: check the cephx mds auth access for setattr If we hit any failre just try to force it to do the sync setattr. Link: https://tracker.ceph.com/issues/61333 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Signed-off-by: Ilya Dryomov --- fs/ceph/inode.c | 46 +++++++++++++++++++++++++++++++++++++--------- 1 file changed, 37 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 7b2e77517f23..3f1f048d9206 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -2480,6 +2480,34 @@ int __ceph_setattr(struct mnt_idmap *idmap, struct inode *inode, bool lock_snap_rwsem = false; bool fill_fscrypt; int truncate_retry = 20; /* The RMW will take around 50ms */ + struct dentry *dentry; + char *path; + int pathlen; + u64 pathbase; + bool do_sync = false; + + dentry = d_find_alias(inode); + if (!dentry) { + do_sync = true; + } else { + path = ceph_mdsc_build_path(mdsc, dentry, &pathlen, &pathbase, 0); + if (IS_ERR(path)) { + do_sync = true; + err = 0; + } else { + err = ceph_mds_check_access(mdsc, path, MAY_WRITE); + } + ceph_mdsc_free_path(path, pathlen); + dput(dentry); + + /* For none EACCES cases will let the MDS do the mds auth check */ + if (err == -EACCES) { + return err; + } else if (err < 0) { + do_sync = true; + err = 0; + } + } retry: prealloc_cf = ceph_alloc_cap_flush(); @@ -2526,7 +2554,7 @@ retry: /* It should never be re-set once set */ WARN_ON_ONCE(ci->fscrypt_auth); - if (issued & CEPH_CAP_AUTH_EXCL) { + if (!do_sync && (issued & CEPH_CAP_AUTH_EXCL)) { dirtied |= CEPH_CAP_AUTH_EXCL; kfree(ci->fscrypt_auth); ci->fscrypt_auth = (u8 *)cia->fscrypt_auth; @@ -2555,7 +2583,7 @@ retry: ceph_vinop(inode), from_kuid(&init_user_ns, inode->i_uid), from_kuid(&init_user_ns, attr->ia_uid)); - if (issued & CEPH_CAP_AUTH_EXCL) { + if (!do_sync && (issued & CEPH_CAP_AUTH_EXCL)) { inode->i_uid = fsuid; dirtied |= CEPH_CAP_AUTH_EXCL; } else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 || @@ -2573,7 +2601,7 @@ retry: ceph_vinop(inode), from_kgid(&init_user_ns, inode->i_gid), from_kgid(&init_user_ns, attr->ia_gid)); - if (issued & CEPH_CAP_AUTH_EXCL) { + if (!do_sync && (issued & CEPH_CAP_AUTH_EXCL)) { inode->i_gid = fsgid; dirtied |= CEPH_CAP_AUTH_EXCL; } else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 || @@ -2587,7 +2615,7 @@ retry: if (ia_valid & ATTR_MODE) { doutc(cl, "%p %llx.%llx mode 0%o -> 0%o\n", inode, ceph_vinop(inode), inode->i_mode, attr->ia_mode); - if (issued & CEPH_CAP_AUTH_EXCL) { + if (!do_sync && (issued & CEPH_CAP_AUTH_EXCL)) { inode->i_mode = attr->ia_mode; dirtied |= CEPH_CAP_AUTH_EXCL; } else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 || @@ -2606,11 +2634,11 @@ retry: inode, ceph_vinop(inode), atime.tv_sec, atime.tv_nsec, attr->ia_atime.tv_sec, attr->ia_atime.tv_nsec); - if (issued & CEPH_CAP_FILE_EXCL) { + if (!do_sync && (issued & CEPH_CAP_FILE_EXCL)) { ci->i_time_warp_seq++; inode_set_atime_to_ts(inode, attr->ia_atime); dirtied |= CEPH_CAP_FILE_EXCL; - } else if ((issued & CEPH_CAP_FILE_WR) && + } else if (!do_sync && (issued & CEPH_CAP_FILE_WR) && timespec64_compare(&atime, &attr->ia_atime) < 0) { inode_set_atime_to_ts(inode, attr->ia_atime); @@ -2646,7 +2674,7 @@ retry: CEPH_FSCRYPT_BLOCK_SIZE)); req->r_fscrypt_file = attr->ia_size; fill_fscrypt = true; - } else if ((issued & CEPH_CAP_FILE_EXCL) && attr->ia_size >= isize) { + } else if (!do_sync && (issued & CEPH_CAP_FILE_EXCL) && attr->ia_size >= isize) { if (attr->ia_size > isize) { i_size_write(inode, attr->ia_size); inode->i_blocks = calc_inode_blocks(attr->ia_size); @@ -2683,11 +2711,11 @@ retry: inode, ceph_vinop(inode), mtime.tv_sec, mtime.tv_nsec, attr->ia_mtime.tv_sec, attr->ia_mtime.tv_nsec); - if (issued & CEPH_CAP_FILE_EXCL) { + if (!do_sync && (issued & CEPH_CAP_FILE_EXCL)) { ci->i_time_warp_seq++; inode_set_mtime_to_ts(inode, attr->ia_mtime); dirtied |= CEPH_CAP_FILE_EXCL; - } else if ((issued & CEPH_CAP_FILE_WR) && + } else if (!do_sync && (issued & CEPH_CAP_FILE_WR) && timespec64_compare(&mtime, &attr->ia_mtime) < 0) { inode_set_mtime_to_ts(inode, attr->ia_mtime); dirtied |= CEPH_CAP_FILE_WR; -- cgit v1.2.3-59-g8ed1b From 845ae9d4926fa69d27e0912e4404d848d19c79a0 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Wed, 8 Nov 2023 10:54:35 +0800 Subject: ceph: check the cephx mds auth access for open Before opening the file locally we need to check the cephx access. Link: https://tracker.ceph.com/issues/61333 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Signed-off-by: Ilya Dryomov --- fs/ceph/file.c | 35 +++++++++++++++++++++++++++++++++-- 1 file changed, 33 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 16873d07692f..4de4bdd7949e 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -366,6 +366,12 @@ int ceph_open(struct inode *inode, struct file *file) struct ceph_file_info *fi = file->private_data; int err; int flags, fmode, wanted; + struct dentry *dentry; + char *path; + int pathlen; + u64 pathbase; + bool do_sync = false; + int mask = MAY_READ; if (fi) { doutc(cl, "file %p is already opened\n", file); @@ -387,6 +393,31 @@ int ceph_open(struct inode *inode, struct file *file) fmode = ceph_flags_to_mode(flags); wanted = ceph_caps_for_mode(fmode); + if (fmode & CEPH_FILE_MODE_WR) + mask |= MAY_WRITE; + dentry = d_find_alias(inode); + if (!dentry) { + do_sync = true; + } else { + path = ceph_mdsc_build_path(mdsc, dentry, &pathlen, &pathbase, 0); + if (IS_ERR(path)) { + do_sync = true; + err = 0; + } else { + err = ceph_mds_check_access(mdsc, path, mask); + } + ceph_mdsc_free_path(path, pathlen); + dput(dentry); + + /* For none EACCES cases will let the MDS do the mds auth check */ + if (err == -EACCES) { + return err; + } else if (err < 0) { + do_sync = true; + err = 0; + } + } + /* snapped files are read-only */ if (ceph_snap(inode) != CEPH_NOSNAP && (file->f_mode & FMODE_WRITE)) return -EROFS; @@ -402,7 +433,7 @@ int ceph_open(struct inode *inode, struct file *file) * asynchronously. */ spin_lock(&ci->i_ceph_lock); - if (__ceph_is_any_real_caps(ci) && + if (!do_sync && __ceph_is_any_real_caps(ci) && (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) { int mds_wanted = __ceph_caps_mds_wanted(ci, true); int issued = __ceph_caps_issued(ci, NULL); @@ -420,7 +451,7 @@ int ceph_open(struct inode *inode, struct file *file) ceph_check_caps(ci, 0); return ceph_init_file(inode, file, fmode); - } else if (ceph_snap(inode) != CEPH_NOSNAP && + } else if (!do_sync && ceph_snap(inode) != CEPH_NOSNAP && (ci->i_snap_caps & wanted) == wanted) { __ceph_touch_fmode(ci, mdsc, fmode); spin_unlock(&ci->i_ceph_lock); -- cgit v1.2.3-59-g8ed1b From 2827badaf8162157271027ea6cc13056890f3e93 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Wed, 8 Nov 2023 11:06:05 +0800 Subject: ceph: check the cephx mds auth access for async dirop Before doing the op locally we need to check the cephx access. Link: https://tracker.ceph.com/issues/61333 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Signed-off-by: Ilya Dryomov --- fs/ceph/dir.c | 28 ++++++++++++++++++++++++++++ fs/ceph/file.c | 31 +++++++++++++++++++++++++++++++ 2 files changed, 59 insertions(+) (limited to 'fs') diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 0e9f56eaba1e..82a2e2a06a65 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1336,8 +1336,12 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry) struct inode *inode = d_inode(dentry); struct ceph_mds_request *req; bool try_async = ceph_test_mount_opt(fsc, ASYNC_DIROPS); + struct dentry *dn; int err = -EROFS; int op; + char *path; + int pathlen; + u64 pathbase; if (ceph_snap(dir) == CEPH_SNAPDIR) { /* rmdir .snap/foo is RMSNAP */ @@ -1351,6 +1355,30 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry) CEPH_MDS_OP_RMDIR : CEPH_MDS_OP_UNLINK; } else goto out; + + dn = d_find_alias(dir); + if (!dn) { + try_async = false; + } else { + path = ceph_mdsc_build_path(mdsc, dn, &pathlen, &pathbase, 0); + if (IS_ERR(path)) { + try_async = false; + err = 0; + } else { + err = ceph_mds_check_access(mdsc, path, MAY_WRITE); + } + ceph_mdsc_free_path(path, pathlen); + dput(dn); + + /* For none EACCES cases will let the MDS do the mds auth check */ + if (err == -EACCES) { + return err; + } else if (err < 0) { + try_async = false; + err = 0; + } + } + retry: req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS); if (IS_ERR(req)) { diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 4de4bdd7949e..4b8d59ebda00 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -790,6 +790,9 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, bool try_async = ceph_test_mount_opt(fsc, ASYNC_DIROPS); int mask; int err; + char *path; + int pathlen; + u64 pathbase; doutc(cl, "%p %llx.%llx dentry %p '%pd' %s flags %d mode 0%o\n", dir, ceph_vinop(dir), dentry, dentry, @@ -807,6 +810,34 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, */ flags &= ~O_TRUNC; + dn = d_find_alias(dir); + if (!dn) { + try_async = false; + } else { + path = ceph_mdsc_build_path(mdsc, dn, &pathlen, &pathbase, 0); + if (IS_ERR(path)) { + try_async = false; + err = 0; + } else { + int fmode = ceph_flags_to_mode(flags); + + mask = MAY_READ; + if (fmode & CEPH_FILE_MODE_WR) + mask |= MAY_WRITE; + err = ceph_mds_check_access(mdsc, path, mask); + } + ceph_mdsc_free_path(path, pathlen); + dput(dn); + + /* For none EACCES cases will let the MDS do the mds auth check */ + if (err == -EACCES) { + return err; + } else if (err < 0) { + try_async = false; + err = 0; + } + } + retry: if (flags & O_CREAT) { if (ceph_quota_is_max_files_exceeded(dir)) -- cgit v1.2.3-59-g8ed1b From d8fc89815f67db960d1684e1c89c36292a981250 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Mon, 4 Dec 2023 13:01:13 +0800 Subject: ceph: add CEPHFS_FEATURE_MDS_AUTH_CAPS_CHECK feature bit Since we have support checking the mds auth cap in kclient, just set the feature bit. Link: https://tracker.ceph.com/issues/61333 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Signed-off-by: Ilya Dryomov --- fs/ceph/mds_client.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index dba514fb7fab..cfa18cf915a0 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -35,8 +35,9 @@ enum ceph_feature_type { CEPHFS_FEATURE_32BITS_RETRY_FWD, CEPHFS_FEATURE_NEW_SNAPREALM_INFO, CEPHFS_FEATURE_HAS_OWNER_UIDGID, + CEPHFS_FEATURE_MDS_AUTH_CAPS_CHECK, - CEPHFS_FEATURE_MAX = CEPHFS_FEATURE_HAS_OWNER_UIDGID, + CEPHFS_FEATURE_MAX = CEPHFS_FEATURE_MDS_AUTH_CAPS_CHECK, }; #define CEPHFS_FEATURES_CLIENT_SUPPORTED { \ @@ -52,6 +53,7 @@ enum ceph_feature_type { CEPHFS_FEATURE_OP_GETVXATTR, \ CEPHFS_FEATURE_32BITS_RETRY_FWD, \ CEPHFS_FEATURE_HAS_OWNER_UIDGID, \ + CEPHFS_FEATURE_MDS_AUTH_CAPS_CHECK, \ } /* -- cgit v1.2.3-59-g8ed1b From 8898e7f288c47d450a3cf1511c791a03550c0789 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 23 May 2024 01:14:26 -0400 Subject: eventfs: Keep the directories from having the same inode number as files The directories require unique inode numbers but all the eventfs files have the same inode number. Prevent the directories from having the same inode numbers as the files as that can confuse some tooling. Link: https://lore.kernel.org/linux-trace-kernel/20240523051539.428826685@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Masahiro Yamada Fixes: 834bf76add3e6 ("eventfs: Save directory inodes in the eventfs_inode structure") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 0256afdd4acf..55a40a730b10 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -50,8 +50,12 @@ static struct eventfs_root_inode *get_root_inode(struct eventfs_inode *ei) /* Just try to make something consistent and unique */ static int eventfs_dir_ino(struct eventfs_inode *ei) { - if (!ei->ino) + if (!ei->ino) { ei->ino = get_next_ino(); + /* Must not have the file inode number */ + if (ei->ino == EVENTFS_FILE_INODE_INO) + ei->ino = get_next_ino(); + } return ei->ino; } -- cgit v1.2.3-59-g8ed1b From 27c046484382d78b4abb0a6e9905a20121af9b35 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 23 May 2024 01:14:27 -0400 Subject: tracefs: Update inode permissions on remount When a remount happens, if a gid or uid is specified update the inodes to have the same gid and uid. This will allow the simplification of the permissions logic for the dynamically created files and directories. Link: https://lore.kernel.org/linux-trace-kernel/20240523051539.592429986@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Masahiro Yamada Fixes: baa23a8d4360d ("tracefs: Reset permissions on remount if permissions are options") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 17 +++++++++++++---- fs/tracefs/inode.c | 15 ++++++++++++--- 2 files changed, 25 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 55a40a730b10..5dfb1ccd56ea 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -317,20 +317,29 @@ void eventfs_remount(struct tracefs_inode *ti, bool update_uid, bool update_gid) if (!ei) return; - if (update_uid) + if (update_uid) { ei->attr.mode &= ~EVENTFS_SAVE_UID; + ei->attr.uid = ti->vfs_inode.i_uid; + } + - if (update_gid) + if (update_gid) { ei->attr.mode &= ~EVENTFS_SAVE_GID; + ei->attr.gid = ti->vfs_inode.i_gid; + } if (!ei->entry_attrs) return; for (int i = 0; i < ei->nr_entries; i++) { - if (update_uid) + if (update_uid) { ei->entry_attrs[i].mode &= ~EVENTFS_SAVE_UID; - if (update_gid) + ei->entry_attrs[i].uid = ti->vfs_inode.i_uid; + } + if (update_gid) { ei->entry_attrs[i].mode &= ~EVENTFS_SAVE_GID; + ei->entry_attrs[i].gid = ti->vfs_inode.i_gid; + } } } diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index a827f6a716c4..9252e0d78ea2 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -373,12 +373,21 @@ static int tracefs_apply_options(struct super_block *sb, bool remount) rcu_read_lock(); list_for_each_entry_rcu(ti, &tracefs_inodes, list) { - if (update_uid) + if (update_uid) { ti->flags &= ~TRACEFS_UID_PERM_SET; + ti->vfs_inode.i_uid = fsi->uid; + } - if (update_gid) + if (update_gid) { ti->flags &= ~TRACEFS_GID_PERM_SET; - + ti->vfs_inode.i_gid = fsi->gid; + } + + /* + * Note, the above ti->vfs_inode updates are + * used in eventfs_remount() so they must come + * before calling it. + */ if (ti->flags & TRACEFS_EVENT_INODE) eventfs_remount(ti, update_uid, update_gid); } -- cgit v1.2.3-59-g8ed1b From 340f0c7067a95281ad13734f8225f49c6cf52067 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 23 May 2024 01:14:28 -0400 Subject: eventfs: Update all the eventfs_inodes from the events descriptor The change to update the permissions of the eventfs_inode had the misconception that using the tracefs_inode would find all the eventfs_inodes that have been updated and reset them on remount. The problem with this approach is that the eventfs_inodes are freed when they are no longer used (basically the reason the eventfs system exists). When they are freed, the updated eventfs_inodes are not reset on a remount because their tracefs_inodes have been freed. Instead, since the events directory eventfs_inode always has a tracefs_inode pointing to it (it is not freed when finished), and the events directory has a link to all its children, have the eventfs_remount() function only operate on the events eventfs_inode and have it descend into its children updating their uid and gids. Link: https://lore.kernel.org/all/CAK7LNARXgaWw3kH9JgrnH4vK6fr8LDkNKf3wq8NhMWJrVwJyVQ@mail.gmail.com/ Link: https://lore.kernel.org/linux-trace-kernel/20240523051539.754424703@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: baa23a8d4360d ("tracefs: Reset permissions on remount if permissions are options") Reported-by: Masahiro Yamada Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 44 +++++++++++++++++++++++++++++++------------- 1 file changed, 31 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 5dfb1ccd56ea..129d0f54ba62 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -305,27 +305,27 @@ static const struct file_operations eventfs_file_operations = { .llseek = generic_file_llseek, }; -/* - * On a remount of tracefs, if UID or GID options are set, then - * the mount point inode permissions should be used. - * Reset the saved permission flags appropriately. - */ -void eventfs_remount(struct tracefs_inode *ti, bool update_uid, bool update_gid) +static void eventfs_set_attrs(struct eventfs_inode *ei, bool update_uid, kuid_t uid, + bool update_gid, kgid_t gid, int level) { - struct eventfs_inode *ei = ti->private; + struct eventfs_inode *ei_child; - if (!ei) + /* Update events// */ + if (WARN_ON_ONCE(level > 3)) return; if (update_uid) { ei->attr.mode &= ~EVENTFS_SAVE_UID; - ei->attr.uid = ti->vfs_inode.i_uid; + ei->attr.uid = uid; } - if (update_gid) { ei->attr.mode &= ~EVENTFS_SAVE_GID; - ei->attr.gid = ti->vfs_inode.i_gid; + ei->attr.gid = gid; + } + + list_for_each_entry(ei_child, &ei->children, list) { + eventfs_set_attrs(ei_child, update_uid, uid, update_gid, gid, level + 1); } if (!ei->entry_attrs) @@ -334,13 +334,31 @@ void eventfs_remount(struct tracefs_inode *ti, bool update_uid, bool update_gid) for (int i = 0; i < ei->nr_entries; i++) { if (update_uid) { ei->entry_attrs[i].mode &= ~EVENTFS_SAVE_UID; - ei->entry_attrs[i].uid = ti->vfs_inode.i_uid; + ei->entry_attrs[i].uid = uid; } if (update_gid) { ei->entry_attrs[i].mode &= ~EVENTFS_SAVE_GID; - ei->entry_attrs[i].gid = ti->vfs_inode.i_gid; + ei->entry_attrs[i].gid = gid; } } + +} + +/* + * On a remount of tracefs, if UID or GID options are set, then + * the mount point inode permissions should be used. + * Reset the saved permission flags appropriately. + */ +void eventfs_remount(struct tracefs_inode *ti, bool update_uid, bool update_gid) +{ + struct eventfs_inode *ei = ti->private; + + /* Only the events directory does the updates */ + if (!ei || !ei->is_events || ei->is_freed) + return; + + eventfs_set_attrs(ei, update_uid, ti->vfs_inode.i_uid, + update_gid, ti->vfs_inode.i_gid, 0); } /* Return the evenfs_inode of the "events" directory */ -- cgit v1.2.3-59-g8ed1b From 0bcfd9aa4dafa03b88d68bf66b694df2a3e76cf3 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 23 May 2024 01:14:29 -0400 Subject: tracefs: Clear EVENT_INODE flag in tracefs_drop_inode() When the inode is being dropped from the dentry, the TRACEFS_EVENT_INODE flag needs to be cleared to prevent a remount from calling eventfs_remount() on the tracefs_inode private data. There's a race between the inode is dropped (and the dentry freed) to where the inode is actually freed. If a remount happens between the two, the eventfs_inode could be accessed after it is freed (only the dentry keeps a ref count on it). Currently the TRACEFS_EVENT_INODE flag is cleared from the dentry iput() function. But this is incorrect, as it is possible that the inode has another reference to it. The flag should only be cleared when the inode is really being dropped and has no more references. That happens in the drop_inode callback of the inode, as that gets called when the last reference of the inode is released. Remove the tracefs_d_iput() function and move its logic to the more appropriate tracefs_drop_inode() callback function. Link: https://lore.kernel.org/linux-trace-kernel/20240523051539.908205106@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Masahiro Yamada Fixes: baa23a8d4360d ("tracefs: Reset permissions on remount if permissions are options") Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/inode.c | 33 +++++++++++++++++---------------- 1 file changed, 17 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index 9252e0d78ea2..7c29f4afc23d 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -426,10 +426,26 @@ static int tracefs_show_options(struct seq_file *m, struct dentry *root) return 0; } +static int tracefs_drop_inode(struct inode *inode) +{ + struct tracefs_inode *ti = get_tracefs(inode); + + /* + * This inode is being freed and cannot be used for + * eventfs. Clear the flag so that it doesn't call into + * eventfs during the remount flag updates. The eventfs_inode + * gets freed after an RCU cycle, so the content will still + * be safe if the iteration is going on now. + */ + ti->flags &= ~TRACEFS_EVENT_INODE; + + return 1; +} + static const struct super_operations tracefs_super_operations = { .alloc_inode = tracefs_alloc_inode, .free_inode = tracefs_free_inode, - .drop_inode = generic_delete_inode, + .drop_inode = tracefs_drop_inode, .statfs = simple_statfs, .show_options = tracefs_show_options, }; @@ -455,22 +471,7 @@ static int tracefs_d_revalidate(struct dentry *dentry, unsigned int flags) return !(ei && ei->is_freed); } -static void tracefs_d_iput(struct dentry *dentry, struct inode *inode) -{ - struct tracefs_inode *ti = get_tracefs(inode); - - /* - * This inode is being freed and cannot be used for - * eventfs. Clear the flag so that it doesn't call into - * eventfs during the remount flag updates. The eventfs_inode - * gets freed after an RCU cycle, so the content will still - * be safe if the iteration is going on now. - */ - ti->flags &= ~TRACEFS_EVENT_INODE; -} - static const struct dentry_operations tracefs_dentry_operations = { - .d_iput = tracefs_d_iput, .d_revalidate = tracefs_d_revalidate, .d_release = tracefs_d_release, }; -- cgit v1.2.3-59-g8ed1b From 625acf9d5e56e10b92636003df7f5bddb21a7a34 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 22 May 2024 12:49:43 -0400 Subject: eventfs: Consolidate the eventfs_inode update in eventfs_get_inode() To simplify the code, create a eventfs_get_inode() that is used when an eventfs file or directory is created. Have the internal tracefs_inode updated the appropriate flags in this function and update the inode's mode as well. Link: https://lore.kernel.org/lkml/20240522165031.624864160@goodmis.org Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 42 +++++++++++++++++++++++------------------- 1 file changed, 23 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 129d0f54ba62..92987b5c8d9d 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -412,6 +412,25 @@ static void update_inode_attr(struct dentry *dentry, struct inode *inode, inode->i_gid = attr->gid; } +static struct inode *eventfs_get_inode(struct dentry *dentry, struct eventfs_attr *attr, + umode_t mode, struct eventfs_inode *ei) +{ + struct tracefs_inode *ti; + struct inode *inode; + + inode = tracefs_get_inode(dentry->d_sb); + if (!inode) + return NULL; + + ti = get_tracefs(inode); + ti->private = ei; + ti->flags |= TRACEFS_EVENT_INODE; + + update_inode_attr(dentry, inode, attr, mode); + + return inode; +} + /** * lookup_file - look up a file in the tracefs filesystem * @parent_ei: Pointer to the eventfs_inode that represents parent of the file @@ -432,7 +451,6 @@ static struct dentry *lookup_file(struct eventfs_inode *parent_ei, void *data, const struct file_operations *fop) { - struct tracefs_inode *ti; struct inode *inode; if (!(mode & S_IFMT)) @@ -441,13 +459,11 @@ static struct dentry *lookup_file(struct eventfs_inode *parent_ei, if (WARN_ON_ONCE(!S_ISREG(mode))) return ERR_PTR(-EIO); - inode = tracefs_get_inode(dentry->d_sb); + /* Only directories have ti->private set to an ei, not files */ + inode = eventfs_get_inode(dentry, attr, mode, NULL); if (unlikely(!inode)) return ERR_PTR(-ENOMEM); - /* If the user updated the directory's attributes, use them */ - update_inode_attr(dentry, inode, attr, mode); - inode->i_op = &eventfs_file_inode_operations; inode->i_fop = fop; inode->i_private = data; @@ -455,9 +471,6 @@ static struct dentry *lookup_file(struct eventfs_inode *parent_ei, /* All files will have the same inode number */ inode->i_ino = EVENTFS_FILE_INODE_INO; - ti = get_tracefs(inode); - ti->flags |= TRACEFS_EVENT_INODE; - // Files have their parent's ei as their fsdata dentry->d_fsdata = get_ei(parent_ei); @@ -477,28 +490,19 @@ static struct dentry *lookup_file(struct eventfs_inode *parent_ei, static struct dentry *lookup_dir_entry(struct dentry *dentry, struct eventfs_inode *pei, struct eventfs_inode *ei) { - struct tracefs_inode *ti; struct inode *inode; + umode_t mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO; - inode = tracefs_get_inode(dentry->d_sb); + inode = eventfs_get_inode(dentry, &ei->attr, mode, ei); if (unlikely(!inode)) return ERR_PTR(-ENOMEM); - /* If the user updated the directory's attributes, use them */ - update_inode_attr(dentry, inode, &ei->attr, - S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO); - inode->i_op = &eventfs_dir_inode_operations; inode->i_fop = &eventfs_file_operations; /* All directories will have the same inode number */ inode->i_ino = eventfs_dir_ino(ei); - ti = get_tracefs(inode); - ti->flags |= TRACEFS_EVENT_INODE; - /* Only directories have ti->private set to an ei, not files */ - ti->private = ei; - dentry->d_fsdata = get_ei(ei); d_add(dentry, inode); -- cgit v1.2.3-59-g8ed1b From 37cd0d1266971942f5cda3e4ca1c6fb005635b14 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 22 May 2024 12:49:44 -0400 Subject: eventfs: Remove getattr and permission callbacks Now that inodes have their permissions updated on remount, the only other places to update the inode permissions are when they are created and in the setattr callback. The getattr and permission callbacks are not needed as the inodes should already be set at their proper settings. Remove the callbacks, as it not only simplifies the code, but also allows more flexibility to fix the inconsistencies with various corner cases (like changing the permission of an instance directory). Link: https://lore.kernel.org/lkml/20240522165031.782066021@goodmis.org Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 40 ---------------------------------------- 1 file changed, 40 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 92987b5c8d9d..f32c6f7eb29f 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -250,49 +250,9 @@ static void update_events_attr(struct eventfs_inode *ei, struct super_block *sb) ei->attr.gid = parent->i_gid; } -static void set_top_events_ownership(struct inode *inode) -{ - struct tracefs_inode *ti = get_tracefs(inode); - struct eventfs_inode *ei = ti->private; - - /* The top events directory doesn't get automatically updated */ - if (!ei || !ei->is_events) - return; - - update_events_attr(ei, inode->i_sb); - - if (!(ei->attr.mode & EVENTFS_SAVE_UID)) - inode->i_uid = ei->attr.uid; - - if (!(ei->attr.mode & EVENTFS_SAVE_GID)) - inode->i_gid = ei->attr.gid; -} - -static int eventfs_get_attr(struct mnt_idmap *idmap, - const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) -{ - struct dentry *dentry = path->dentry; - struct inode *inode = d_backing_inode(dentry); - - set_top_events_ownership(inode); - - generic_fillattr(idmap, request_mask, inode, stat); - return 0; -} - -static int eventfs_permission(struct mnt_idmap *idmap, - struct inode *inode, int mask) -{ - set_top_events_ownership(inode); - return generic_permission(idmap, inode, mask); -} - static const struct inode_operations eventfs_dir_inode_operations = { .lookup = eventfs_root_lookup, .setattr = eventfs_set_attr, - .getattr = eventfs_get_attr, - .permission = eventfs_permission, }; static const struct inode_operations eventfs_file_inode_operations = { -- cgit v1.2.3-59-g8ed1b From 6e3d7c903c3e82d626f1bc32c790fc86a179d14c Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 22 May 2024 12:49:45 -0400 Subject: eventfs: Cleanup permissions in creation of inodes The permissions being set during the creation of the inodes was updating eventfs_inode attributes as well. Those attributes should only be touched by the setattr or remount operations, not during the creation of inodes. The eventfs_inode attributes should only be used to set the inodes and should not be modified during the inode creation. Simplify the code and fix the situation by: 1) Removing the eventfs_find_events() and doing a simple lookup for the events descriptor in eventfs_get_inode() 2) Remove update_events_attr() as the attributes should only be used to update the inode and should not be modified here. 3) Add update_inode_attr() that uses the attributes to determine what the inode permissions should be. 4) As the parent_inode of the eventfs_root_inode structure is no longer needed, remove it. Now on creation, the inode gets the proper permissions without causing side effects to the ei->attr field. Link: https://lore.kernel.org/lkml/20240522165031.944088388@goodmis.org Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 90 +++++++++++++----------------------------------- 1 file changed, 23 insertions(+), 67 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index f32c6f7eb29f..320e49056d3a 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -37,7 +37,6 @@ static DEFINE_MUTEX(eventfs_mutex); struct eventfs_root_inode { struct eventfs_inode ei; - struct inode *parent_inode; struct dentry *events_dir; }; @@ -229,27 +228,6 @@ static int eventfs_set_attr(struct mnt_idmap *idmap, struct dentry *dentry, return ret; } -static void update_events_attr(struct eventfs_inode *ei, struct super_block *sb) -{ - struct eventfs_root_inode *rei; - struct inode *parent; - - rei = get_root_inode(ei); - - /* Use the parent inode permissions unless root set its permissions */ - parent = rei->parent_inode; - - if (rei->ei.attr.mode & EVENTFS_SAVE_UID) - ei->attr.uid = rei->ei.attr.uid; - else - ei->attr.uid = parent->i_uid; - - if (rei->ei.attr.mode & EVENTFS_SAVE_GID) - ei->attr.gid = rei->ei.attr.gid; - else - ei->attr.gid = parent->i_gid; -} - static const struct inode_operations eventfs_dir_inode_operations = { .lookup = eventfs_root_lookup, .setattr = eventfs_set_attr, @@ -321,60 +299,30 @@ void eventfs_remount(struct tracefs_inode *ti, bool update_uid, bool update_gid) update_gid, ti->vfs_inode.i_gid, 0); } -/* Return the evenfs_inode of the "events" directory */ -static struct eventfs_inode *eventfs_find_events(struct dentry *dentry) +static void update_inode_attr(struct inode *inode, umode_t mode, + struct eventfs_attr *attr, struct eventfs_root_inode *rei) { - struct eventfs_inode *ei; - - do { - // The parent is stable because we do not do renames - dentry = dentry->d_parent; - // ... and directories always have d_fsdata - ei = dentry->d_fsdata; - - /* - * If the ei is being freed, the ownership of the children - * doesn't matter. - */ - if (ei->is_freed) - return NULL; - - // Walk upwards until you find the events inode - } while (!ei->is_events); - - update_events_attr(ei, dentry->d_sb); - - return ei; -} - -static void update_inode_attr(struct dentry *dentry, struct inode *inode, - struct eventfs_attr *attr, umode_t mode) -{ - struct eventfs_inode *events_ei = eventfs_find_events(dentry); - - if (!events_ei) - return; - - inode->i_mode = mode; - inode->i_uid = events_ei->attr.uid; - inode->i_gid = events_ei->attr.gid; - - if (!attr) - return; - - if (attr->mode & EVENTFS_SAVE_MODE) + if (attr && attr->mode & EVENTFS_SAVE_MODE) inode->i_mode = attr->mode & EVENTFS_MODE_MASK; + else + inode->i_mode = mode; - if (attr->mode & EVENTFS_SAVE_UID) + if (attr && attr->mode & EVENTFS_SAVE_UID) inode->i_uid = attr->uid; + else + inode->i_uid = rei->ei.attr.uid; - if (attr->mode & EVENTFS_SAVE_GID) + if (attr && attr->mode & EVENTFS_SAVE_GID) inode->i_gid = attr->gid; + else + inode->i_gid = rei->ei.attr.gid; } static struct inode *eventfs_get_inode(struct dentry *dentry, struct eventfs_attr *attr, umode_t mode, struct eventfs_inode *ei) { + struct eventfs_root_inode *rei; + struct eventfs_inode *pei; struct tracefs_inode *ti; struct inode *inode; @@ -386,7 +334,16 @@ static struct inode *eventfs_get_inode(struct dentry *dentry, struct eventfs_att ti->private = ei; ti->flags |= TRACEFS_EVENT_INODE; - update_inode_attr(dentry, inode, attr, mode); + /* Find the top dentry that holds the "events" directory */ + do { + dentry = dentry->d_parent; + /* Directories always have d_fsdata */ + pei = dentry->d_fsdata; + } while (!pei->is_events); + + rei = get_root_inode(pei); + + update_inode_attr(inode, mode, attr, rei); return inode; } @@ -823,7 +780,6 @@ struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry // Note: we have a ref to the dentry from tracefs_start_creating() rei = get_root_inode(ei); rei->events_dir = dentry; - rei->parent_inode = d_inode(dentry->d_sb->s_root); ei->entries = entries; ei->nr_entries = size; -- cgit v1.2.3-59-g8ed1b From 2dd00ac1d38afba1b59e439abc300a9b0ce696bf Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 22 May 2024 12:49:46 -0400 Subject: eventfs: Do not use attributes for events directory The top "events" directory has a static inode (it's created when it is and removed when the directory is removed). There's no need to use the events ei->attr to determine its permissions. But it is used for saving the permissions of the "events" directory for when it is created, as that is needed for the default permissions for the files and directories underneath it. For example: # cd /sys/kernel/tracing # mkdir instances/foo # chown 1001 instances/foo/events The files under instances/foo/events should still have the same owner as instances/foo (which the instances/foo/events ei->attr will hold), but the events directory now has owner 1001. Link: https://lore.kernel.org/lkml/20240522165032.104981011@goodmis.org Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Signed-off-by: Steven Rostedt (Google) --- fs/tracefs/event_inode.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/tracefs/event_inode.c b/fs/tracefs/event_inode.c index 320e49056d3a..5d88c184f0fc 100644 --- a/fs/tracefs/event_inode.c +++ b/fs/tracefs/event_inode.c @@ -210,7 +210,9 @@ static int eventfs_set_attr(struct mnt_idmap *idmap, struct dentry *dentry, * determined by the parent directory. */ if (dentry->d_inode->i_mode & S_IFDIR) { - update_attr(&ei->attr, iattr); + /* Just use the inode permissions for the events directory */ + if (!ei->is_events) + update_attr(&ei->attr, iattr); } else { name = dentry->d_name.name; @@ -789,14 +791,12 @@ struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry uid = d_inode(dentry->d_parent)->i_uid; gid = d_inode(dentry->d_parent)->i_gid; - ei->attr.uid = uid; - ei->attr.gid = gid; - /* - * When the "events" directory is created, it takes on the - * permissions of its parent. But can be reset on remount. + * The ei->attr will be used as the default values for the + * files beneath this directory. */ - ei->attr.mode |= EVENTFS_SAVE_UID | EVENTFS_SAVE_GID; + ei->attr.uid = uid; + ei->attr.gid = gid; INIT_LIST_HEAD(&ei->children); INIT_LIST_HEAD(&ei->list); -- cgit v1.2.3-59-g8ed1b From 05afeeebcac850a016ec4fb1f681ceda11963562 Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 23 Apr 2024 15:31:56 +0300 Subject: fs/ntfs3: Fix case when index is reused during tree transformation In most cases when adding a cluster to the directory index, they are placed at the end, and in the bitmap, this cluster corresponds to the last bit. The new directory size is calculated as follows: data_size = (u64)(bit + 1) << indx->index_bits; In the case of reusing a non-final cluster from the index, data_size is calculated incorrectly, resulting in the directory size differing from the actual size. A check for cluster reuse has been added, and the size update is skipped. Fixes: 82cae269cfa95 ("fs/ntfs3: Add initialization of super block") Signed-off-by: Konstantin Komarov Cc: stable@vger.kernel.org --- fs/ntfs3/index.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'fs') diff --git a/fs/ntfs3/index.c b/fs/ntfs3/index.c index daabaad63aaf..14284f0ed46a 100644 --- a/fs/ntfs3/index.c +++ b/fs/ntfs3/index.c @@ -1533,6 +1533,11 @@ static int indx_add_allocate(struct ntfs_index *indx, struct ntfs_inode *ni, goto out1; } + if (data_size <= le64_to_cpu(alloc->nres.data_size)) { + /* Reuse index. */ + goto out; + } + /* Increase allocation. */ err = attr_set_size(ni, ATTR_ALLOC, in->name, in->name_len, &indx->alloc_run, data_size, &data_size, true, @@ -1546,6 +1551,7 @@ static int indx_add_allocate(struct ntfs_index *indx, struct ntfs_inode *ni, if (in->name == I30_NAME) i_size_write(&ni->vfs_inode, data_size); +out: *vbn = bit << indx->idx2vbn_bits; return 0; -- cgit v1.2.3-59-g8ed1b From 302e9dca8428979c9c99f2dbb44dc1783f5011c3 Mon Sep 17 00:00:00 2001 From: Konstantin Komarov Date: Tue, 23 Apr 2024 17:21:58 +0300 Subject: fs/ntfs3: Break dir enumeration if directory contents error If we somehow attempt to read beyond the directory size, an error is supposed to be returned. However, in some cases, read requests do not stop and instead enter into a loop. To avoid this, we set the position in the directory to the end. Signed-off-by: Konstantin Komarov Cc: stable@vger.kernel.org --- fs/ntfs3/dir.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/ntfs3/dir.c b/fs/ntfs3/dir.c index 5cf3d9decf64..45e556fd7c54 100644 --- a/fs/ntfs3/dir.c +++ b/fs/ntfs3/dir.c @@ -475,6 +475,7 @@ static int ntfs_readdir(struct file *file, struct dir_context *ctx) vbo = (u64)bit << index_bits; if (vbo >= i_size) { ntfs_inode_err(dir, "Looks like your dir is corrupt"); + ctx->pos = eod; err = -EINVAL; goto out; } -- cgit v1.2.3-59-g8ed1b From f5d4e04634c9cf68bdf23de08ada0bb92e8befe7 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Mon, 20 May 2024 22:26:19 +0900 Subject: nilfs2: fix use-after-free of timer for log writer thread Patch series "nilfs2: fix log writer related issues". This bug fix series covers three nilfs2 log writer-related issues, including a timer use-after-free issue and potential deadlock issue on unmount, and a potential freeze issue in event synchronization found during their analysis. Details are described in each commit log. This patch (of 3): A use-after-free issue has been reported regarding the timer sc_timer on the nilfs_sc_info structure. The problem is that even though it is used to wake up a sleeping log writer thread, sc_timer is not shut down until the nilfs_sc_info structure is about to be freed, and is used regardless of the thread's lifetime. Fix this issue by limiting the use of sc_timer only while the log writer thread is alive. Link: https://lkml.kernel.org/r/20240520132621.4054-1-konishi.ryusuke@gmail.com Link: https://lkml.kernel.org/r/20240520132621.4054-2-konishi.ryusuke@gmail.com Fixes: fdce895ea5dd ("nilfs2: change sc_timer from a pointer to an embedded one in struct nilfs_sc_info") Signed-off-by: Ryusuke Konishi Reported-by: "Bai, Shuangpeng" Closes: https://groups.google.com/g/syzkaller/c/MK_LYqtt8ko/m/8rgdWeseAwAJ Tested-by: Ryusuke Konishi Cc: Signed-off-by: Andrew Morton --- fs/nilfs2/segment.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index 6be7dd423fbd..7cb34e1c9206 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -2118,8 +2118,10 @@ static void nilfs_segctor_start_timer(struct nilfs_sc_info *sci) { spin_lock(&sci->sc_state_lock); if (!(sci->sc_state & NILFS_SEGCTOR_COMMIT)) { - sci->sc_timer.expires = jiffies + sci->sc_interval; - add_timer(&sci->sc_timer); + if (sci->sc_task) { + sci->sc_timer.expires = jiffies + sci->sc_interval; + add_timer(&sci->sc_timer); + } sci->sc_state |= NILFS_SEGCTOR_COMMIT; } spin_unlock(&sci->sc_state_lock); @@ -2320,10 +2322,21 @@ int nilfs_construct_dsync_segment(struct super_block *sb, struct inode *inode, */ static void nilfs_segctor_accept(struct nilfs_sc_info *sci) { + bool thread_is_alive; + spin_lock(&sci->sc_state_lock); sci->sc_seq_accepted = sci->sc_seq_request; + thread_is_alive = (bool)sci->sc_task; spin_unlock(&sci->sc_state_lock); - del_timer_sync(&sci->sc_timer); + + /* + * This function does not race with the log writer thread's + * termination. Therefore, deleting sc_timer, which should not be + * done after the log writer thread exits, can be done safely outside + * the area protected by sc_state_lock. + */ + if (thread_is_alive) + del_timer_sync(&sci->sc_timer); } /** @@ -2349,7 +2362,7 @@ static void nilfs_segctor_notify(struct nilfs_sc_info *sci, int mode, int err) sci->sc_flush_request &= ~FLUSH_DAT_BIT; /* re-enable timer if checkpoint creation was not done */ - if ((sci->sc_state & NILFS_SEGCTOR_COMMIT) && + if ((sci->sc_state & NILFS_SEGCTOR_COMMIT) && sci->sc_task && time_before(jiffies, sci->sc_timer.expires)) add_timer(&sci->sc_timer); } @@ -2539,6 +2552,7 @@ static int nilfs_segctor_thread(void *arg) int timeout = 0; sci->sc_timer_task = current; + timer_setup(&sci->sc_timer, nilfs_construction_timeout, 0); /* start sync. */ sci->sc_task = current; @@ -2606,6 +2620,7 @@ static int nilfs_segctor_thread(void *arg) end_thread: /* end sync. */ sci->sc_task = NULL; + timer_shutdown_sync(&sci->sc_timer); wake_up(&sci->sc_wait_task); /* for nilfs_segctor_kill_thread() */ spin_unlock(&sci->sc_state_lock); return 0; @@ -2669,7 +2684,6 @@ static struct nilfs_sc_info *nilfs_segctor_new(struct super_block *sb, INIT_LIST_HEAD(&sci->sc_gc_inodes); INIT_LIST_HEAD(&sci->sc_iput_queue); INIT_WORK(&sci->sc_iput_work, nilfs_iput_work_func); - timer_setup(&sci->sc_timer, nilfs_construction_timeout, 0); sci->sc_interval = HZ * NILFS_SC_DEFAULT_TIMEOUT; sci->sc_mjcp_freq = HZ * NILFS_SC_DEFAULT_SR_FREQ; @@ -2748,7 +2762,6 @@ static void nilfs_segctor_destroy(struct nilfs_sc_info *sci) down_write(&nilfs->ns_segctor_sem); - timer_shutdown_sync(&sci->sc_timer); kfree(sci); } -- cgit v1.2.3-59-g8ed1b From 936184eadd82906992ff1f5ab3aada70cce44cee Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Mon, 20 May 2024 22:26:20 +0900 Subject: nilfs2: fix unexpected freezing of nilfs_segctor_sync() A potential and reproducible race issue has been identified where nilfs_segctor_sync() would block even after the log writer thread writes a checkpoint, unless there is an interrupt or other trigger to resume log writing. This turned out to be because, depending on the execution timing of the log writer thread running in parallel, the log writer thread may skip responding to nilfs_segctor_sync(), which causes a call to schedule() waiting for completion within nilfs_segctor_sync() to lose the opportunity to wake up. The reason why waking up the task waiting in nilfs_segctor_sync() may be skipped is that updating the request generation issued using a shared sequence counter and adding an wait queue entry to the request wait queue to the log writer, are not done atomically. There is a possibility that log writing and request completion notification by nilfs_segctor_wakeup() may occur between the two operations, and in that case, the wait queue entry is not yet visible to nilfs_segctor_wakeup() and the wake-up of nilfs_segctor_sync() will be carried over until the next request occurs. Fix this issue by performing these two operations simultaneously within the lock section of sc_state_lock. Also, following the memory barrier guidelines for event waiting loops, move the call to set_current_state() in the same location into the event waiting loop to ensure that a memory barrier is inserted just before the event condition determination. Link: https://lkml.kernel.org/r/20240520132621.4054-3-konishi.ryusuke@gmail.com Fixes: 9ff05123e3bf ("nilfs2: segment constructor") Signed-off-by: Ryusuke Konishi Tested-by: Ryusuke Konishi Cc: Cc: "Bai, Shuangpeng" Signed-off-by: Andrew Morton --- fs/nilfs2/segment.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index 7cb34e1c9206..b6c3c5f8628e 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -2168,19 +2168,28 @@ static int nilfs_segctor_sync(struct nilfs_sc_info *sci) struct nilfs_segctor_wait_request wait_req; int err = 0; - spin_lock(&sci->sc_state_lock); init_wait(&wait_req.wq); wait_req.err = 0; atomic_set(&wait_req.done, 0); + init_waitqueue_entry(&wait_req.wq, current); + + /* + * To prevent a race issue where completion notifications from the + * log writer thread are missed, increment the request sequence count + * "sc_seq_request" and insert a wait queue entry using the current + * sequence number into the "sc_wait_request" queue at the same time + * within the lock section of "sc_state_lock". + */ + spin_lock(&sci->sc_state_lock); wait_req.seq = ++sci->sc_seq_request; + add_wait_queue(&sci->sc_wait_request, &wait_req.wq); spin_unlock(&sci->sc_state_lock); - init_waitqueue_entry(&wait_req.wq, current); - add_wait_queue(&sci->sc_wait_request, &wait_req.wq); - set_current_state(TASK_INTERRUPTIBLE); wake_up(&sci->sc_wait_daemon); for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (atomic_read(&wait_req.done)) { err = wait_req.err; break; -- cgit v1.2.3-59-g8ed1b From eb85dace897c5986bc2f36b3c783c6abb8a4292e Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Mon, 20 May 2024 22:26:21 +0900 Subject: nilfs2: fix potential hang in nilfs_detach_log_writer() Syzbot has reported a potential hang in nilfs_detach_log_writer() called during nilfs2 unmount. Analysis revealed that this is because nilfs_segctor_sync(), which synchronizes with the log writer thread, can be called after nilfs_segctor_destroy() terminates that thread, as shown in the call trace below: nilfs_detach_log_writer nilfs_segctor_destroy nilfs_segctor_kill_thread --> Shut down log writer thread flush_work nilfs_iput_work_func nilfs_dispose_list iput nilfs_evict_inode nilfs_transaction_commit nilfs_construct_segment (if inode needs sync) nilfs_segctor_sync --> Attempt to synchronize with log writer thread *** DEADLOCK *** Fix this issue by changing nilfs_segctor_sync() so that the log writer thread returns normally without synchronizing after it terminates, and by forcing tasks that are already waiting to complete once after the thread terminates. The skipped inode metadata flushout will then be processed together in the subsequent cleanup work in nilfs_segctor_destroy(). Link: https://lkml.kernel.org/r/20240520132621.4054-4-konishi.ryusuke@gmail.com Signed-off-by: Ryusuke Konishi Reported-by: syzbot+e3973c409251e136fdd0@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=e3973c409251e136fdd0 Tested-by: Ryusuke Konishi Cc: Cc: "Bai, Shuangpeng" Signed-off-by: Andrew Morton --- fs/nilfs2/segment.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index b6c3c5f8628e..60d4f59f7665 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -2190,6 +2190,14 @@ static int nilfs_segctor_sync(struct nilfs_sc_info *sci) for (;;) { set_current_state(TASK_INTERRUPTIBLE); + /* + * Synchronize only while the log writer thread is alive. + * Leave flushing out after the log writer thread exits to + * the cleanup work in nilfs_segctor_destroy(). + */ + if (!sci->sc_task) + break; + if (atomic_read(&wait_req.done)) { err = wait_req.err; break; @@ -2205,7 +2213,7 @@ static int nilfs_segctor_sync(struct nilfs_sc_info *sci) return err; } -static void nilfs_segctor_wakeup(struct nilfs_sc_info *sci, int err) +static void nilfs_segctor_wakeup(struct nilfs_sc_info *sci, int err, bool force) { struct nilfs_segctor_wait_request *wrq, *n; unsigned long flags; @@ -2213,7 +2221,7 @@ static void nilfs_segctor_wakeup(struct nilfs_sc_info *sci, int err) spin_lock_irqsave(&sci->sc_wait_request.lock, flags); list_for_each_entry_safe(wrq, n, &sci->sc_wait_request.head, wq.entry) { if (!atomic_read(&wrq->done) && - nilfs_cnt32_ge(sci->sc_seq_done, wrq->seq)) { + (force || nilfs_cnt32_ge(sci->sc_seq_done, wrq->seq))) { wrq->err = err; atomic_set(&wrq->done, 1); } @@ -2362,7 +2370,7 @@ static void nilfs_segctor_notify(struct nilfs_sc_info *sci, int mode, int err) if (mode == SC_LSEG_SR) { sci->sc_state &= ~NILFS_SEGCTOR_COMMIT; sci->sc_seq_done = sci->sc_seq_accepted; - nilfs_segctor_wakeup(sci, err); + nilfs_segctor_wakeup(sci, err, false); sci->sc_flush_request = 0; } else { if (mode == SC_FLUSH_FILE) @@ -2746,6 +2754,13 @@ static void nilfs_segctor_destroy(struct nilfs_sc_info *sci) || sci->sc_seq_request != sci->sc_seq_done); spin_unlock(&sci->sc_state_lock); + /* + * Forcibly wake up tasks waiting in nilfs_segctor_sync(), which can + * be called from delayed iput() via nilfs_evict_inode() and can race + * with the above log writer thread termination. + */ + nilfs_segctor_wakeup(sci, 0, true); + if (flush_work(&sci->sc_iput_work)) flag = true; -- cgit v1.2.3-59-g8ed1b From 6d065f507d82307d6161ac75c025111fb8b08a46 Mon Sep 17 00:00:00 2001 From: Yuanyuan Zhong Date: Thu, 23 May 2024 12:35:31 -0600 Subject: mm: /proc/pid/smaps_rollup: avoid skipping vma after getting mmap_lock again After switching smaps_rollup to use VMA iterator, searching for next entry is part of the condition expression of the do-while loop. So the current VMA needs to be addressed before the continue statement. Otherwise, with some VMAs skipped, userspace observed memory consumption from /proc/pid/smaps_rollup will be smaller than the sum of the corresponding fields from /proc/pid/smaps. Link: https://lkml.kernel.org/r/20240523183531.2535436-1-yzhong@purestorage.com Fixes: c4c84f06285e ("fs/proc/task_mmu: stop using linked list and highest_vm_end") Signed-off-by: Yuanyuan Zhong Reviewed-by: Mohamed Khalfella Cc: David Hildenbrand Cc: Matthew Wilcox (Oracle) Cc: Signed-off-by: Andrew Morton --- fs/proc/task_mmu.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index e5a5f015ff03..f8d35f993fe5 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -970,12 +970,17 @@ static int show_smaps_rollup(struct seq_file *m, void *v) break; /* Case 1 and 2 above */ - if (vma->vm_start >= last_vma_end) + if (vma->vm_start >= last_vma_end) { + smap_gather_stats(vma, &mss, 0); + last_vma_end = vma->vm_end; continue; + } /* Case 4 above */ - if (vma->vm_end > last_vma_end) + if (vma->vm_end > last_vma_end) { smap_gather_stats(vma, &mss, last_vma_end); + last_vma_end = vma->vm_end; + } } } for_each_vma(vmi, vma); -- cgit v1.2.3-59-g8ed1b From 8a1607233566fef7fbe30416e5d3b0567d0d1fba Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 May 2024 09:38:48 +0100 Subject: cifs: Fix smb3_insert_range() to move the zero_point Fix smb3_insert_range() to move the zero_point over to the new EOF. Without this, generic/147 fails as reads of data beyond the old EOF point return zeroes. Fixes: 3ee1a1fc3981 ("cifs: Cut over to using netfslib") Signed-off-by: David Howells cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev Signed-off-by: Steve French --- fs/smb/client/smb2ops.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index ef18cd30f66c..b87b70edd0be 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -3636,6 +3636,7 @@ static long smb3_insert_range(struct file *file, struct cifs_tcon *tcon, rc = smb2_copychunk_range(xid, cfile, cfile, off, count, off + len); if (rc < 0) goto out_2; + cifsi->netfs.zero_point = new_eof; rc = smb3_zero_data(file, tcon, off, len, xid); if (rc < 0) -- cgit v1.2.3-59-g8ed1b From 93a43155127fec0f8cc942d63b76668c2f8f69fa Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 24 May 2024 15:23:36 +0100 Subject: cifs: Fix missing set of remote_i_size Occasionally, the generic/001 xfstest will fail indicating corruption in one of the copy chains when run on cifs against a server that supports FSCTL_DUPLICATE_EXTENTS_TO_FILE (eg. Samba with a share on btrfs). The problem is that the remote_i_size value isn't updated by cifs_setsize() when called by smb2_duplicate_extents(), but i_size *is*. This may cause cifs_remap_file_range() to then skip the bit after calling ->duplicate_extents() that sets sizes. Fix this by calling netfs_resize_file() in smb2_duplicate_extents() before calling cifs_setsize() to set i_size. This means we don't then need to call netfs_resize_file() upon return from ->duplicate_extents(), but we also fix the test to compare against the pre-dup inode size. [Note that this goes back before the addition of remote_i_size with the netfs_inode struct. It should probably have been setting cifsi->server_eof previously.] Fixes: cfc63fc8126a ("smb3: fix cached file size problems in duplicate extents (reflink)") Signed-off-by: David Howells cc: Steve French cc: Paulo Alcantara cc: Shyam Prasad N cc: Rohith Surabattula cc: Jeff Layton cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev Signed-off-by: Steve French --- fs/smb/client/cifsfs.c | 6 +++--- fs/smb/client/smb2ops.c | 1 + 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index ec5b639f421a..a665aac9be9f 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -1226,7 +1226,7 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off, struct cifsFileInfo *smb_file_src = src_file->private_data; struct cifsFileInfo *smb_file_target = dst_file->private_data; struct cifs_tcon *target_tcon, *src_tcon; - unsigned long long destend, fstart, fend, new_size; + unsigned long long destend, fstart, fend, old_size, new_size; unsigned int xid; int rc; @@ -1293,6 +1293,7 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off, goto unlock; if (fend > target_cifsi->netfs.zero_point) target_cifsi->netfs.zero_point = fend + 1; + old_size = target_cifsi->netfs.remote_i_size; /* Discard all the folios that overlap the destination region. */ cifs_dbg(FYI, "about to discard pages %llx-%llx\n", fstart, fend); @@ -1305,9 +1306,8 @@ static loff_t cifs_remap_file_range(struct file *src_file, loff_t off, if (target_tcon->ses->server->ops->duplicate_extents) { rc = target_tcon->ses->server->ops->duplicate_extents(xid, smb_file_src, smb_file_target, off, len, destoff); - if (rc == 0 && new_size > i_size_read(target_inode)) { + if (rc == 0 && new_size > old_size) { truncate_setsize(target_inode, new_size); - netfs_resize_file(&target_cifsi->netfs, new_size, true); fscache_resize_cookie(cifs_inode_cookie(target_inode), new_size); } diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index b87b70edd0be..4ce6c3121a7e 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -2028,6 +2028,7 @@ smb2_duplicate_extents(const unsigned int xid, * size will be queried on next revalidate, but it is important * to make sure that file's cached size is updated immediately */ + netfs_resize_file(netfs_inode(inode), dest_off + len, true); cifs_setsize(inode, dest_off + len); } rc = SMB2_ioctl(xid, tcon, trgtfile->fid.persistent_fid, -- cgit v1.2.3-59-g8ed1b